Version bump to v1.6

Add queues.peek_task and queues.get_num_entries

.gitignore

@@ -12,7 +12,6 @@ test-reports
 .pytest_cache
 venv
 *.noseids
-build
 *.egg-info
 .cache
 .mypy_cache

README.md

@@ -8,28 +8,43 @@
 [![GitHub license](https://img.shields.io/badge/license-SSPL-green.svg)](https://img.shields.io/badge/license-SSPL-green.svg)
 [![Python versions](https://img.shields.io/badge/python-3.6%20%7C%203.7-blue.svg)](https://img.shields.io/badge/python-3.6%20%7C%203.7-blue.svg)
 [![GitHub version](https://img.shields.io/github/release-pre/allegroai/trains-server.svg)](https://img.shields.io/github/release-pre/allegroai/trains-server.svg)
+[![Artifact Hub](https://img.shields.io/endpoint?url=https://artifacthub.io/badge/repository/allegroai)](https://artifacthub.io/packages/search?repo=allegroai)
 
 </div>
 
 ---
 <div align="center">
 
-**v0.16 Upgrade Notice**
+**Note regarding Apache Log4j2 Remote Code Execution (RCE) Vulnerability - CVE-2021-44228 - ESA-2021-31**
 
 </div>
 
-In v0.16, the Elasticsearch subsystem of ClearML Server has been upgraded from version 5.6 to version 7.6. This change necessitates the migration of the database contents to accommodate the change in index structure across the different versions.  
+According to [ElasticSearch's latest report](https://discuss.elastic.co/t/apache-log4j2-remote-code-execution-rce-vulnerability-cve-2021-44228-esa-2021-31/291476), 
+supported versions of Elasticsearch (6.8.9+, 7.8+) used with recent versions of the JDK (JDK9+) **are not susceptible to either remote code execution or information leakage**
+due to Elasticsearch’s usage of the Java Security Manager.
 
-Follow [this procedure](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_es7_migration.html) to migrate existing data.
+**As the latest version of ClearML Server uses Elasticsearch 7.10+ with JDK15, it is not affected by these vulnerabilities.**
+
+As a precaution, we've upgraded the ES version to 7.16.2 and added the mitigation recommended by ElasticSearch to our latest [docker-compose.yml](https://github.com/allegroai/clearml-server/blob/cfccbe05c158b75e520581f86e9668291da5c70a/docker/docker-compose.yml#L42) file.
+
+While previous Elasticsearch versions (5.6.11+, 6.4.0+ and 7.0.0+) used by older ClearML Server versions are only susceptible to the information leakage vulnerability
+(which in any case **does not permit access to data within the Elasticsearch cluster**), 
+we still recommend upgrading to the latest version of ClearML Server. Alternatively, you can apply the mitigation as implemented in our latest 
+[docker-compose.yml](https://github.com/allegroai/clearml-server/blob/cfccbe05c158b75e520581f86e9668291da5c70a/docker/docker-compose.yml#L42) file.    
+
+**Update 15 December**: A further vulnerability (CVE-2021-45046) was disclosed on December 14th.
+ElasticSearch's guidance for Elasticsearch remains unchanged by this new vulnerability, thus **not affecting ClearML Server**.
+
+**Update 22 December**: To keep with ElasticSearch's recommendations, we've upgraded the ES version to the newly released 7.16.2
 
 ---
 
-### ClearML Server
+## ClearML Server
 #### *Formerly known as Trains Server*
 
 The **ClearML Server** is the backend service infrastructure for [ClearML](https://github.com/allegroai/clearml).
 It allows multiple users to collaborate and manage their experiments.
-By default, **ClearML** is set up to work with the **ClearML** demo server, which is open to anyone and resets periodically.
+**ClearML** offers a [free hosted service](https://app.clear.ml/), which is maintained by **ClearML** and open to anyone.
 In order to host your own server, you will need to launch the **ClearML Server** and point **ClearML** to it.
 
 The **ClearML Server** contains the following components:
@@ -45,7 +60,7 @@ You can quickly [deploy](#launching-the-clearml-server)  your **ClearML Server**
 ## System design
 
 
-![Alt Text](https://allegro.ai/clearml/docs/_images/ClearML_Server_Diagram.png)
+![Alt Text](docs/ClearML_Server_Diagram.png)
 
 The **ClearML Server** has two supported configurations:
 - Single IP (domain) with the following open ports
@@ -78,20 +93,19 @@ For example, to see if port `8080` is in use:
     
 Launch The **ClearML Server** in any of the following formats:
 
-- Pre-built [AWS EC2 AMI](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_aws_ec2_ami.html)
-- Pre-built [GCP Custom Image](hhttps://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_gcp.html)
+- Pre-built [AWS EC2 AMI](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_aws_ec2_ami)
+- Pre-built [GCP Custom Image](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_gcp)
 - Pre-built Docker Image
-    - [Linux](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_linux_mac.html)
-    - [macOS](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_linux_mac.html)
-    - [Windows 10](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_win.html)
+    - [Linux](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_linux_mac)
+    - [macOS](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_linux_mac)
+    - [Windows 10](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_win)
 - Kubernetes    
-    - [Kubernetes Helm](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_kubernetes_helm.html)
-    - Manual [Kubernetes installation](https://allegro.ai/clearml/docs/docs/deploying_clearml/clearml_server_kubernetes.html)
+    - [Kubernetes Helm](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_kubernetes_helm)
+    - Manual [Kubernetes installation](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_kubernetes)
 
 ## Connecting ClearML to your ClearML Server
 
-By default, the **ClearML** client is set up to work with the [**ClearML** demo server](https://demoapp.demo.clear.ml/).  
-To have the **ClearML** client use your **ClearML Server** instead:
+In order to set up the **ClearML** client to work with your **ClearML Server**:
 - Run the `clearml-init` command for an interactive setup.
 - Or manually edit `~/clearml.conf` file, making sure the server settings (`api_server`, `web_server`, `file_server`) are configured correctly, for example:
 
@@ -138,8 +152,8 @@ Do not enqueue training / inference tasks into the `services` queue, as it will
 
 The **ClearML Server** provides a few additional useful features, which can be manually enabled:
  
-* [Web login authentication](https://allegro.ai/clearml/docs/deploying_clearml/clearml_server_config/#web-login-authentication)
-* [Non-responsive experiments watchdog](https://allegro.ai/clearml/docs/deploying_clearml/clearml_server_config/#task_watchdog)  
+* [Web login authentication](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_config#web-login-authentication)
+* [Non-responsive experiments watchdog](https://clear.ml/docs/latest/docs/deploying_clearml/clearml_server_config#non-responsive-task-watchdog)  
 
 ## Restarting ClearML Server
 
@@ -189,14 +203,14 @@ To upgrade your existing **ClearML Server** deployment:
    ```
 
 1. Configure the ClearML-Agent Services (not supported on Windows installation). 
-   If `TRAINS_HOST_IP` is not provided, ClearML-Agent Services will use the external 
-   public address of the **ClearML Server**. If `TRAINS_AGENT_GIT_USER` / `TRAINS_AGENT_GIT_PASS` are not provided, 
+   If `CLEARML_HOST_IP` is not provided, ClearML-Agent Services will use the external 
+   public address of the **ClearML Server**. If `CLEARML_AGENT_GIT_USER` / `CLEARML_AGENT_GIT_PASS` are not provided, 
    the ClearML-Agent Services will not be able to access any private repositories for running service tasks.
    
    ```bash
-   export TRAINS_HOST_IP=server_host_ip_here
-   export TRAINS_AGENT_GIT_USER=git_username_here
-   export TRAINS_AGENT_GIT_PASS=git_password_here
+   export CLEARML_HOST_IP=server_host_ip_here
+   export CLEARML_AGENT_GIT_USER=git_username_here
+   export CLEARML_AGENT_GIT_PASS=git_password_here
    ```
 
 1. Spin up the docker containers, it will automatically pull the latest **ClearML Server** build    
@@ -205,12 +219,12 @@ To upgrade your existing **ClearML Server** deployment:
    docker-compose -f docker-compose.yml up
    ```
 
-**\* If something went wrong along the way, check our FAQ: [Common Docker Upgrade Errors](https://allegro.ai/clearml/docs/docs/faq/faq.html).**
+**\* If something went wrong along the way, check our FAQ: [Common Docker Upgrade Errors](https://clear.ml/docs/latest/docs/faq/).**
 
 
 ## Community & Support
 
-If you have any questions, look to the ClearML [FAQ](https://allegro.ai/clearml/docs/docs/faq/faq.html), or
+If you have any questions, look to the ClearML [FAQ](https://clear.ml/docs/latest/docs/faq), or
 tag your questions on [stackoverflow](https://stackoverflow.com/questions/tagged/clearml) with '**clearml**' tag.
 
 For feature requests or bug reports, please use [GitHub issues](https://github.com/allegroai/clearml-server/issues).

apiserver/apierrors/errors.conf

@@ -26,6 +26,9 @@
     23: ["invalid_domain_name", "malformed domain name"]
     24: ["not_public_object", "object is not public"]
 
+    # Auth / Login
+    75: ["invalid_access_key", "access key not found for user"]
+
     # Tasks
     100: ["task_error", "general task error"]
     101: ["invalid_task_id", "invalid task id"]
@@ -71,6 +74,8 @@
     408: ["cannot_update_project_location", "Cannot update project location. Use projects.move instead"]
     409: ["project_path_exceeds_max", "Project path exceed the maximum allowed depth"]
     410: ["project_source_and_destination_are_the_same", "Project has the same source and destination paths"]
+    411: ["project_cannot_be_moved_under_itself", "Project can not be moved under itself in the projects hierarchy"]
+    412: ["project_cannot_be_merged_into_its_child", "Project can not be merged into its own child"]
 
     # Queues
     701: ["invalid_queue_id", "invalid queue id"]
@@ -84,7 +89,7 @@
 
     # Database
     800: ["data_validation_error", "data validation error"]
-    801: ["expected_unique_data", "value combination already exists"]
+    801: ["expected_unique_data", "value combination already exists (unique field already contains this value)"]
 
     # Workers
     1001: ["invalid_worker_id", "invalid worker id"]

apiserver/apimodels/auth.py

@@ -75,11 +75,17 @@ class CreateUserResponse(Base):
 class Credentials(Base):
     access_key = StringField(required=True)
     secret_key = StringField(required=True)
+    label = StringField()
 
 
 class CredentialsResponse(Credentials):
     secret_key = StringField()
     last_used = DateTimeField(default=None)
+    last_used_from = StringField()
+
+
+class CreateCredentialsRequest(Base):
+    label = StringField()
 
 
 class CreateCredentialsResponse(Base):
@@ -90,6 +96,11 @@ class GetCredentialsResponse(Base):
     credentials = ListField(CredentialsResponse)
 
 
+class EditCredentialsRequest(Base):
+    access_key = StringField(required=True)
+    label = StringField()
+
+
 class RevokeCredentialsRequest(Base):
     access_key = StringField(required=True)
 

apiserver/apimodels/events.py

@@ -2,7 +2,7 @@ from enum import auto
 from typing import Sequence, Optional
 
 from jsonmodels import validators
-from jsonmodels.fields import StringField, BoolField
+from jsonmodels.fields import StringField, BoolField, EmbeddedField
 from jsonmodels.models import Base
 from jsonmodels.validators import Length, Min, Max
 
@@ -14,12 +14,18 @@ from apiserver.utilities.stringenum import StringEnum
 
 
 class HistogramRequestBase(Base):
-    samples: int = IntField(default=6000, validators=[Min(1), Max(6000)])
+    samples: int = IntField(default=2000, validators=[Min(1), Max(6000)])
     key: ScalarKeyEnum = ActualEnumField(ScalarKeyEnum, default=ScalarKeyEnum.iter)
 
 
+class MetricVariants(Base):
+    metric: str = StringField(required=True)
+    variants: Sequence[str] = ListField(items_types=str)
+
+
 class ScalarMetricsIterHistogramRequest(HistogramRequestBase):
     task: str = StringField(required=True)
+    metrics: Sequence[MetricVariants] = ListField(items_types=MetricVariants)
 
 
 class MultiTaskScalarMetricsIterHistogramRequest(HistogramRequestBase):
@@ -39,9 +45,10 @@ class MultiTaskScalarMetricsIterHistogramRequest(HistogramRequestBase):
 class TaskMetric(Base):
     task: str = StringField(required=True)
     metric: str = StringField(default=None)
+    variants: Sequence[str] = ListField(items_types=str)
 
 
-class DebugImagesRequest(Base):
+class MetricEventsRequest(Base):
     metrics: Sequence[TaskMetric] = ListField(
         items_types=TaskMetric, validators=[Length(minimum_value=1)]
     )
@@ -57,13 +64,14 @@ class TaskMetricVariant(Base):
     variant: str = StringField(required=True)
 
 
-class GetDebugImageSampleRequest(TaskMetricVariant):
+class GetHistorySampleRequest(TaskMetricVariant):
     iteration: Optional[int] = IntField()
-    scroll_id: Optional[str] = StringField()
     refresh: bool = BoolField(default=False)
+    scroll_id: Optional[str] = StringField()
+    navigate_current_metric: bool = BoolField(default=True)
 
 
-class NextDebugImageSampleRequest(Base):
+class NextHistorySampleRequest(Base):
     task: str = StringField(required=True)
     scroll_id: Optional[str] = StringField()
     navigate_earlier: bool = BoolField(default=True)
@@ -74,14 +82,34 @@ class LogOrderEnum(StringEnum):
     desc = auto()
 
 
-class LogEventsRequest(Base):
+class TaskEventsRequestBase(Base):
     task: str = StringField(required=True)
     batch_size: int = IntField(default=500)
+
+
+class TaskEventsRequest(TaskEventsRequestBase):
+    metrics: Sequence[MetricVariants] = ListField(items_types=MetricVariants)
+    event_type: EventType = ActualEnumField(EventType, default=EventType.all)
+    order: Optional[str] = ActualEnumField(LogOrderEnum, default=LogOrderEnum.asc)
+    scroll_id: str = StringField()
+    count_total: bool = BoolField(default=True)
+
+
+class LogEventsRequest(TaskEventsRequestBase):
+    batch_size: int = IntField(default=5000)
     navigate_earlier: bool = BoolField(default=True)
     from_timestamp: Optional[int] = IntField()
     order: Optional[str] = ActualEnumField(LogOrderEnum)
 
 
+class ScalarMetricsIterRawRequest(TaskEventsRequestBase):
+    batch_size: int = IntField()
+    key: ScalarKeyEnum = ActualEnumField(ScalarKeyEnum, default=ScalarKeyEnum.iter)
+    metric: MetricVariants = EmbeddedField(MetricVariants, required=True)
+    count_total: bool = BoolField(default=False)
+    scroll_id: str = StringField()
+
+
 class IterationEvents(Base):
     iter: int = IntField()
     events: Sequence[dict] = ListField(items_types=dict)
@@ -92,13 +120,38 @@ class MetricEvents(Base):
     iterations: Sequence[IterationEvents] = ListField(items_types=IterationEvents)
 
 
-class DebugImageResponse(Base):
+class MetricEventsResponse(Base):
     metrics: Sequence[MetricEvents] = ListField(items_types=MetricEvents)
     scroll_id: str = StringField()
 
 
-class TaskMetricsRequest(Base):
+class MultiTasksRequestBase(Base):
     tasks: Sequence[str] = ListField(
         items_types=str, validators=[Length(minimum_value=1)]
     )
+
+
+class SingleValueMetricsRequest(MultiTasksRequestBase):
+    pass
+
+
+class TaskMetricsRequest(MultiTasksRequestBase):
     event_type: EventType = ActualEnumField(EventType, required=True)
+
+
+class TaskPlotsRequest(Base):
+    task: str = StringField(required=True)
+    iters: int = IntField(default=1)
+    scroll_id: str = StringField()
+    no_scroll: bool = BoolField(default=False)
+    metrics: Sequence[MetricVariants] = ListField(items_types=MetricVariants)
+
+
+class ClearScrollRequest(Base):
+    scroll_id: str = StringField()
+
+
+class ClearTaskLogRequest(Base):
+    task: str = StringField(required=True)
+    threshold_sec = IntField()
+    allow_locked = BoolField(default=False)

apiserver/apimodels/metadata.py

@@ -1,7 +1,7 @@
 from typing import Sequence
 
 from jsonmodels import validators
-from jsonmodels.fields import StringField
+from jsonmodels.fields import StringField, BoolField
 from jsonmodels.models import Base
 
 from apiserver.apimodels import ListField
@@ -21,3 +21,4 @@ class AddOrUpdateMetadata(Base):
     metadata: Sequence[MetadataItem] = ListField(
         [MetadataItem], validators=validators.Length(minimum_value=1)
     )
+    replace_metadata = BoolField(default=False)

apiserver/apimodels/models.py

@@ -30,7 +30,7 @@ class CreateModelRequest(models.Base):
     ready = fields.BoolField(default=True)
     ui_cache = DictField()
     task = fields.StringField()
-    metadata = ListField(items_types=[MetadataItem])
+    metadata = DictField(value_types=[MetadataItem])
 
 
 class CreateModelResponse(models.Base):
@@ -75,3 +75,7 @@ class DeleteMetadataRequest(DeleteMetadata):
 
 class AddOrUpdateMetadataRequest(AddOrUpdateMetadata):
     model = fields.StringField(required=True)
+
+
+class ModelsGetRequest(models.Base):
+    include_stats = fields.BoolField(default=False)

apiserver/apimodels/organization.py

@@ -1,5 +1,7 @@
 from jsonmodels import fields, models
 
+from apiserver.apimodels import DictField
+
 
 class Filter(models.Base):
     tags = fields.ListField([str])
@@ -9,3 +11,11 @@ class Filter(models.Base):
 class TagsRequest(models.Base):
     include_system = fields.BoolField(default=False)
     filter = fields.EmbeddedField(Filter)
+
+
+class EntitiesCountRequest(models.Base):
+    projects = DictField()
+    tasks = DictField()
+    models = DictField()
+    pipelines = DictField()
+    datasets = DictField()

apiserver/apimodels/pipelines.py

@@ -0,0 +1,19 @@
+from jsonmodels import models, fields
+
+from apiserver.apimodels import ListField
+
+
+class Arg(models.Base):
+    name = fields.StringField(required=True)
+    value = fields.StringField(required=True)
+
+
+class StartPipelineRequest(models.Base):
+    task = fields.StringField(required=True)
+    queue = fields.StringField(required=True)
+    args = ListField(Arg)
+
+
+class StartPipelineResponse(models.Base):
+    pipeline = fields.StringField(required=True)
+    enqueued = fields.BoolField(required=True)

apiserver/apimodels/projects.py

@@ -1,6 +1,6 @@
 from jsonmodels import models, fields
 
-from apiserver.apimodels import ListField, ActualEnumField
+from apiserver.apimodels import ListField, ActualEnumField, DictField
 from apiserver.apimodels.organization import TagsRequest
 from apiserver.database.model import EntityVisibility
 
@@ -27,7 +27,7 @@ class ProjectOrNoneRequest(models.Base):
     include_subprojects = fields.BoolField(default=True)
 
 
-class GetHyperParamRequest(ProjectOrNoneRequest):
+class GetParamsRequest(ProjectOrNoneRequest):
     page = fields.IntField(default=0)
     page_size = fields.IntField(default=500)
 
@@ -51,10 +51,19 @@ class ProjectHyperparamValuesRequest(MultiProjectRequest):
     allow_public = fields.BoolField(default=True)
 
 
+class ProjectModelMetadataValuesRequest(MultiProjectRequest):
+    key = fields.StringField(required=True)
+    allow_public = fields.BoolField(default=True)
+
+
 class ProjectsGetRequest(models.Base):
+    include_dataset_stats = fields.BoolField(default=False)
     include_stats = fields.BoolField(default=False)
+    include_stats_filter = DictField()
+    stats_with_children = fields.BoolField(default=True)
     stats_for_state = ActualEnumField(EntityVisibility, default=EntityVisibility.active)
     non_public = fields.BoolField(default=False)
     active_users = fields.ListField(str)
     check_own_contents = fields.BoolField(default=False)
     shallow_search = fields.BoolField(default=False)
+    search_hidden = fields.BoolField(default=False)

apiserver/apimodels/queues.py

@@ -2,7 +2,7 @@ from jsonmodels import validators
 from jsonmodels.fields import StringField, IntField, BoolField, FloatField
 from jsonmodels.models import Base
 
-from apiserver.apimodels import ListField
+from apiserver.apimodels import ListField, DictField
 from apiserver.apimodels.metadata import (
     MetadataItem,
     DeleteMetadata,
@@ -19,13 +19,22 @@ class CreateRequest(Base):
     name = StringField(required=True)
     tags = ListField(items_types=[str])
     system_tags = ListField(items_types=[str])
-    metadata = ListField(items_types=[MetadataItem])
+    metadata = DictField(value_types=[MetadataItem])
 
 
 class QueueRequest(Base):
     queue = StringField(required=True)
 
 
+class GetByIdRequest(QueueRequest):
+    max_task_entries = IntField()
+
+
+class GetNextTaskRequest(QueueRequest):
+    queue = StringField(required=True)
+    get_task_info = BoolField(default=False)
+
+
 class DeleteRequest(QueueRequest):
     force = BoolField(default=False)
 
@@ -34,7 +43,7 @@ class UpdateRequest(QueueRequest):
     name = StringField()
     tags = ListField(items_types=[str])
     system_tags = ListField(items_types=[str])
-    metadata = ListField(items_types=[MetadataItem])
+    metadata = DictField(value_types=[MetadataItem])
 
 
 class TaskRequest(QueueRequest):
@@ -54,6 +63,7 @@ class GetMetricsRequest(Base):
     from_date = FloatField(required=True, validators=validators.Min(0))
     to_date = FloatField(required=True, validators=validators.Min(0))
     interval = IntField(required=True, validators=validators.Min(1))
+    refresh = BoolField(default=False)
 
 
 class QueueMetrics(Base):

apiserver/apimodels/tasks.py

@@ -96,6 +96,7 @@ class UpdateRequest(TaskUpdateRequest):
 
 class EnqueueRequest(UpdateRequest):
     queue = StringField()
+    queue_name = StringField()
 
 
 class DeleteRequest(UpdateRequest):
@@ -108,6 +109,14 @@ class SetRequirementsRequest(TaskRequest):
     requirements = DictField(required=True)
 
 
+class CompletedRequest(UpdateRequest):
+    publish = BoolField(default=False)
+
+
+class CompletedResponse(UpdateResponse):
+    published = IntField(default=0)
+
+
 class PublishRequest(UpdateRequest):
     publish_model = BoolField(default=True)
 
@@ -262,6 +271,7 @@ class StopManyRequest(TaskBatchRequest):
 
 class EnqueueManyRequest(TaskBatchRequest):
     queue = StringField()
+    queue_name = StringField()
     validate_tasks = BoolField(default=False)
 
 

apiserver/apimodels/workers.py

@@ -96,6 +96,7 @@ class WorkerResponseEntry(WorkerEntry):
 
 class GetAllRequest(Base):
     last_seen = IntField(default=3600)
+    tags = ListField(str)
 
 
 class GetAllResponse(Base):

apiserver/bll/auth/__init__.py

@@ -2,7 +2,11 @@ from datetime import datetime
 
 from apiserver import database
 from apiserver.apierrors import errors
-from apiserver.apimodels.auth import GetTokenResponse, CreateUserRequest, Credentials as CredModel
+from apiserver.apimodels.auth import (
+    GetTokenResponse,
+    CreateUserRequest,
+    Credentials as CredModel,
+)
 from apiserver.apimodels.users import CreateRequest as Users_CreateRequest
 from apiserver.bll.user import UserBLL
 from apiserver.config_repo import config
@@ -57,9 +61,10 @@ class AuthBLL:
                 api_version=str(ServiceRepo.max_endpoint_version()),
                 server_version=str(get_version()),
                 server_build=str(get_build_number()),
+                feature_set="basic",
             )
 
-            return GetTokenResponse(token=token.decode("ascii"))
+            return GetTokenResponse(token=token)
 
     @staticmethod
     def create_user(request: CreateUserRequest, call: APICall = None) -> str:
@@ -144,7 +149,7 @@ class AuthBLL:
 
     @classmethod
     def create_credentials(
-        cls, user_id: str, company_id: str, role: str = None
+        cls, user_id: str, company_id: str, role: str = None, label: str = None,
     ) -> CredModel:
 
         with translate_errors_context():
@@ -153,9 +158,11 @@ class AuthBLL:
             if not user:
                 raise errors.bad_request.InvalidUserId(**query)
 
-            cred = CredModel(access_key=get_client_id(), secret_key=get_secret_key())
+            cred = CredModel(
+                access_key=get_client_id(), secret_key=get_secret_key(), label=label
+            )
             user.credentials.append(
-                Credentials(key=cred.access_key, secret=cred.secret_key)
+                Credentials(key=cred.access_key, secret=cred.secret_key, label=label)
             )
             user.save()
 

apiserver/bll/event/event_bll.py

@@ -6,30 +6,34 @@ from collections import defaultdict
 from contextlib import closing
 from datetime import datetime
 from operator import attrgetter
-from typing import Sequence, Set, Tuple, Optional, Dict
+from typing import Sequence, Set, Tuple, Optional, List, Mapping, Union
 
-import six
-from elasticsearch import helpers
+import elasticsearch
 from elasticsearch.helpers import BulkIndexError
 from mongoengine import Q
 from nested_dict import nested_dict
 
-from apiserver.bll.event.debug_sample_history import DebugSampleHistory
 from apiserver.bll.event.event_common import (
     EventType,
-    EventSettings,
     get_index_name,
     check_empty_data,
     search_company_events,
     delete_company_events,
+    MetricVariants,
+    get_metric_variants_condition,
+    uncompress_plot,
+    get_max_metric_and_variant_counts,
 )
+from apiserver.bll.event.events_iterator import EventsIterator, TaskEventsResult
+from apiserver.bll.event.history_debug_image_iterator import HistoryDebugImageIterator
+from apiserver.bll.event.history_plot_iterator import HistoryPlotIterator
+from apiserver.bll.event.metric_debug_images_iterator import MetricDebugImagesIterator
+from apiserver.bll.event.metric_plots_iterator import MetricPlotsIterator
 from apiserver.bll.util import parallel_chunked_decorator
 from apiserver.database import utils as dbutils
 from apiserver.es_factory import es_factory
 from apiserver.apierrors import errors
-from apiserver.bll.event.debug_images_iterator import DebugImagesIterator
 from apiserver.bll.event.event_metrics import EventMetrics
-from apiserver.bll.event.log_events_iterator import LogEventsIterator, TaskEventsResult
 from apiserver.bll.task import TaskBLL
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
@@ -37,14 +41,17 @@ from apiserver.database.model.task.task import Task, TaskStatus
 from apiserver.redis_manager import redman
 from apiserver.timing_context import TimingContext
 from apiserver.tools import safe_get
-from apiserver.utilities.dicts import flatten_nested_items
+from apiserver.utilities.dicts import flatten_nested_items, nested_get
 from apiserver.utilities.json import loads
 
 # noinspection PyTypeChecker
 EVENT_TYPES: Set[str] = set(map(attrgetter("value"), EventType))
 LOCKED_TASK_STATUSES = (TaskStatus.publishing, TaskStatus.published)
-MAX_LONG = 2**63 - 1
-MIN_LONG = -2**63
+MAX_LONG = 2 ** 63 - 1
+MIN_LONG = -(2 ** 63)
+
+
+log = config.logger(__file__)
 
 
 class PlotFields:
@@ -65,14 +72,20 @@ class EventBLL(object):
 
     def __init__(self, events_es=None, redis=None):
         self.es = events_es or es_factory.connect("events")
+        self.redis = redis or redman.connection("apiserver")
         self._metrics = EventMetrics(self.es)
         self._skip_iteration_for_metric = set(
             config.get("services.events.ignore_iteration.metrics", [])
         )
-        self.redis = redis or redman.connection("apiserver")
-        self.debug_images_iterator = DebugImagesIterator(es=self.es, redis=self.redis)
-        self.debug_sample_history = DebugSampleHistory(es=self.es, redis=self.redis)
-        self.log_events_iterator = LogEventsIterator(es=self.es)
+        self.debug_images_iterator = MetricDebugImagesIterator(
+            es=self.es, redis=self.redis
+        )
+        self.debug_image_sample_history = HistoryDebugImageIterator(
+            es=self.es, redis=self.redis
+        )
+        self.plots_iterator = MetricPlotsIterator(es=self.es, redis=self.redis)
+        self.plot_sample_history = HistoryPlotIterator(es=self.es, redis=self.redis)
+        self.events_iterator = EventsIterator(es=self.es)
 
     @property
     def metrics(self) -> EventMetrics:
@@ -94,7 +107,7 @@ class EventBLL(object):
     def add_events(
         self, company_id, events, worker, allow_locked_tasks=False
     ) -> Tuple[int, int, dict]:
-        actions = []
+        actions: List[dict] = []
         task_ids = set()
         task_iteration = defaultdict(lambda: 0)
         task_last_scalar_events = nested_dict(
@@ -197,7 +210,6 @@ class EventBLL(object):
 
             actions.append(es_action)
 
-        action: Dict[dict]
         plot_actions = [
             action["_source"]
             for action in actions
@@ -219,7 +231,7 @@ class EventBLL(object):
                 with TimingContext("es", "events_add_batch"):
                     # TODO: replace it with helpers.parallel_bulk in the future once the parallel pool leak is fixed
                     with closing(
-                        helpers.streaming_bulk(
+                        elasticsearch.helpers.streaming_bulk(
                             self.es,
                             actions,
                             chunk_size=chunk_size,
@@ -260,7 +272,8 @@ class EventBLL(object):
             invalid_iterations_count = errors_per_type.get(invalid_iteration_error)
             if invalid_iterations_count:
                 raise BulkIndexError(
-                    f"{invalid_iterations_count} document(s) failed to index.", [invalid_iteration_error]
+                    f"{invalid_iterations_count} document(s) failed to index.",
+                    [invalid_iteration_error],
                 )
 
         if not added:
@@ -303,11 +316,7 @@ class EventBLL(object):
     @parallel_chunked_decorator(chunk_size=10)
     def uncompress_plots(self, plot_events: Sequence[dict]):
         for event in plot_events:
-            plot_data = event.pop(PlotFields.plot_data, None)
-            if plot_data and event.get(PlotFields.plot_str) is None:
-                event[PlotFields.plot_str] = zlib.decompress(
-                    base64.b64decode(plot_data)
-                ).decode()
+            uncompress_plot(event)
 
     @staticmethod
     def _is_valid_json(text: str) -> bool:
@@ -466,24 +475,37 @@ class EventBLL(object):
         task_id: str,
         num_last_iterations: int,
         event_type: EventType,
+        metric_variants: MetricVariants = None,
     ):
         if check_empty_data(self.es, company_id=company_id, event_type=event_type):
             return []
 
+        must = [{"term": {"task": task_id}}]
+        if metric_variants:
+            must.append(get_metric_variants_condition(metric_variants))
+        query = {"bool": {"must": must}}
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args,
+        )
+        max_variants = int(max_variants // num_last_iterations)
+
         es_req: dict = {
             "size": 0,
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
                         "variants": {
                             "terms": {
                                 "field": "variant",
-                                "size": EventSettings.max_variants_count,
+                                "size": max_variants,
                                 "order": {"_key": "asc"},
                             },
                             "aggs": {
@@ -499,15 +521,13 @@ class EventBLL(object):
                     },
                 }
             },
-            "query": {"bool": {"must": [{"term": {"task": task_id}}]}},
+            "query": query,
         }
 
         with translate_errors_context(), TimingContext(
             "es", "task_last_iter_metric_variant"
         ):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req
-            )
+            es_res = search_company_events(body=es_req, **search_args)
 
         if "aggregations" not in es_res:
             return []
@@ -527,6 +547,8 @@ class EventBLL(object):
         sort=None,
         size: int = 500,
         scroll_id: str = None,
+        no_scroll: bool = False,
+        metric_variants: MetricVariants = None,
     ):
         if scroll_id == self.empty_scroll:
             return TaskEventsResult()
@@ -555,6 +577,8 @@ class EventBLL(object):
 
             if last_iterations_per_plot is None:
                 must.append({"terms": {"task": tasks}})
+                if metric_variants:
+                    must.append(get_metric_variants_condition(metric_variants))
             else:
                 should = []
                 for i, task_id in enumerate(tasks):
@@ -563,6 +587,7 @@ class EventBLL(object):
                         task_id=task_id,
                         num_last_iterations=last_iterations_per_plot,
                         event_type=event_type,
+                        metric_variants=metric_variants,
                     )
                     if not last_iters:
                         continue
@@ -600,7 +625,7 @@ class EventBLL(object):
                     event_type=event_type,
                     body=es_req,
                     ignore=404,
-                    scroll="1h",
+                    **({} if no_scroll else {"scroll": "1h"}),
                 )
 
         events, total_events, next_scroll_id = self._get_events_from_es_res(es_res)
@@ -623,6 +648,42 @@ class EventBLL(object):
 
         return events, total_events, next_scroll_id
 
+    def get_debug_image_urls(
+        self, company_id: str, task_id: str, after_key: dict = None
+    ) -> Tuple[Sequence[str], Optional[dict]]:
+        if check_empty_data(self.es, company_id, EventType.metrics_image):
+            return [], None
+
+        es_req = {
+            "size": 0,
+            "aggs": {
+                "debug_images": {
+                    "composite": {
+                        "size": 1000,
+                        **({"after": after_key} if after_key else {}),
+                        "sources": [{"url": {"terms": {"field": "url"}}}],
+                    }
+                }
+            },
+            "query": {
+                "bool": {
+                    "must": [{"term": {"task": task_id}}, {"exists": {"field": "url"}}]
+                }
+            },
+        }
+
+        es_response = search_company_events(
+            self.es,
+            company_id=company_id,
+            event_type=EventType.metrics_image,
+            body=es_req,
+        )
+        res = nested_get(es_response, ("aggregations", "debug_images"))
+        if not res:
+            return [], None
+
+        return [bucket["key"]["url"] for bucket in res["buckets"]], res.get("after_key")
+
     def get_plot_image_urls(
         self, company_id: str, task_id: str, scroll_id: Optional[str]
     ) -> Tuple[Sequence[dict], Optional[str]]:
@@ -669,19 +730,20 @@ class EventBLL(object):
         sort=None,
         size=500,
         scroll_id=None,
-    ):
+        no_scroll=False,
+    ) -> TaskEventsResult:
         if scroll_id == self.empty_scroll:
-            return [], scroll_id, 0
+            return TaskEventsResult()
 
         if scroll_id:
             with translate_errors_context(), TimingContext("es", "get_task_events"):
                 es_res = self.es.scroll(scroll_id=scroll_id, scroll="1h")
         else:
-            task_ids = [task_id] if isinstance(task_id, six.string_types) else task_id
-
             if check_empty_data(self.es, company_id=company_id, event_type=event_type):
                 return TaskEventsResult()
 
+            task_ids = [task_id] if isinstance(task_id, str) else task_id
+
             must = []
             if metric:
                 must.append({"term": {"metric": metric}})
@@ -691,26 +753,24 @@ class EventBLL(object):
             if last_iter_count is None:
                 must.append({"terms": {"task": task_ids}})
             else:
-                should = []
-                for i, task_id in enumerate(task_ids):
-                    last_iters = self.get_last_iters(
-                        company_id=company_id,
-                        event_type=event_type,
-                        task_id=task_id,
-                        iters=last_iter_count,
-                    )
-                    if not last_iters:
-                        continue
-                    should.append(
-                        {
-                            "bool": {
-                                "must": [
-                                    {"term": {"task": task_id}},
-                                    {"terms": {"iter": last_iters}},
-                                ]
-                            }
+                tasks_iters = self.get_last_iters(
+                    company_id=company_id,
+                    event_type=event_type,
+                    task_id=task_ids,
+                    iters=last_iter_count,
+                )
+                should = [
+                    {
+                        "bool": {
+                            "must": [
+                                {"term": {"task": task}},
+                                {"terms": {"iter": last_iters}},
+                            ]
                         }
-                    )
+                    }
+                    for task, last_iters in tasks_iters.items()
+                    if last_iters
+                ]
                 if not should:
                     return TaskEventsResult()
                 must.append({"bool": {"should": should}})
@@ -731,7 +791,7 @@ class EventBLL(object):
                     event_type=event_type,
                     body=es_req,
                     ignore=404,
-                    scroll="1h",
+                    **({} if no_scroll else {"scroll": "1h"}),
                 )
 
         events, total_events, next_scroll_id = self._get_events_from_es_res(es_res)
@@ -748,35 +808,40 @@ class EventBLL(object):
         if check_empty_data(self.es, company_id=company_id, event_type=event_type):
             return {}
 
+        query = {"bool": {"must": [{"term": {"task": task_id}}]}}
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args,
+        )
         es_req = {
             "size": 0,
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
                         "variants": {
                             "terms": {
                                 "field": "variant",
-                                "size": EventSettings.max_variants_count,
+                                "size": max_variants,
                                 "order": {"_key": "asc"},
                             }
                         }
                     },
                 }
             },
-            "query": {"bool": {"must": [{"term": {"task": task_id}}]}},
+            "query": query,
         }
 
         with translate_errors_context(), TimingContext(
             "es", "events_get_metrics_and_variants"
         ):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req
-            )
+            es_res = search_company_events(body=es_req, **search_args)
 
         metrics = {}
         for metric_bucket in es_res["aggregations"]["metrics"].get("buckets"):
@@ -787,33 +852,42 @@ class EventBLL(object):
 
         return metrics
 
-    def get_task_latest_scalar_values(self, company_id: str, task_id: str):
+    def get_task_latest_scalar_values(
+        self, company_id, task_id
+    ) -> Tuple[Sequence[dict], int]:
         event_type = EventType.metrics_scalar
         if check_empty_data(self.es, company_id=company_id, event_type=event_type):
-            return {}
+            return [], 0
 
+        query = {
+            "bool": {
+                "must": [
+                    {"query_string": {"query": "value:>0"}},
+                    {"term": {"task": task_id}},
+                ]
+            }
+        }
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args,
+        )
         es_req = {
             "size": 0,
-            "query": {
-                "bool": {
-                    "must": [
-                        {"query_string": {"query": "value:>0"}},
-                        {"term": {"task": task_id}},
-                    ]
-                }
-            },
+            "query": query,
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
                         "variants": {
                             "terms": {
                                 "field": "variant",
-                                "size": EventSettings.max_variants_count,
+                                "size": max_variants,
                                 "order": {"_key": "asc"},
                             },
                             "aggs": {
@@ -844,9 +918,7 @@ class EventBLL(object):
         with translate_errors_context(), TimingContext(
             "es", "events_get_metrics_and_variants"
         ):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req
-            )
+            es_res = search_company_events(body=es_req, **search_args)
 
         metrics = []
         max_timestamp = 0
@@ -905,47 +977,65 @@ class EventBLL(object):
         return iterations, vectors
 
     def get_last_iters(
-        self, company_id: str, event_type: EventType, task_id: str, iters: int
-    ):
+        self,
+        company_id: str,
+        event_type: EventType,
+        task_id: Union[str, Sequence[str]],
+        iters: int,
+    ) -> Mapping[str, Sequence]:
         if check_empty_data(self.es, company_id=company_id, event_type=event_type):
-            return []
+            return {}
 
+        task_ids = [task_id] if isinstance(task_id, str) else task_id
         es_req: dict = {
             "size": 0,
             "aggs": {
-                "iters": {
-                    "terms": {
-                        "field": "iter",
-                        "size": iters,
-                        "order": {"_key": "desc"},
-                    }
+                "tasks": {
+                    "terms": {"field": "task"},
+                    "aggs": {
+                        "iters": {
+                            "terms": {
+                                "field": "iter",
+                                "size": iters,
+                                "order": {"_key": "desc"},
+                            }
+                        }
+                    },
                 }
             },
-            "query": {"bool": {"must": [{"term": {"task": task_id}}]}},
+            "query": {"bool": {"must": [{"terms": {"task": task_ids}}]}},
         }
 
         with translate_errors_context(), TimingContext("es", "task_last_iter"):
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req
+                self.es, company_id=company_id, event_type=event_type, body=es_req,
             )
 
         if "aggregations" not in es_res:
-            return []
+            return {}
 
-        return [b["key"] for b in es_res["aggregations"]["iters"]["buckets"]]
+        return {
+            tb["key"]: [ib["key"] for ib in tb["iters"]["buckets"]]
+            for tb in es_res["aggregations"]["tasks"]["buckets"]
+        }
+
+    @staticmethod
+    def _validate_task_state(company_id: str, task_id: str, allow_locked: bool = False):
+        extra_msg = None
+        query = Q(id=task_id, company=company_id)
+        if not allow_locked:
+            query &= Q(status__nin=LOCKED_TASK_STATUSES)
+            extra_msg = "or task published"
+        res = Task.objects(query).only("id").first()
+        if not res:
+            raise errors.bad_request.InvalidTaskId(
+                extra_msg, company=company_id, id=task_id
+            )
 
     def delete_task_events(self, company_id, task_id, allow_locked=False):
-        with translate_errors_context():
-            extra_msg = None
-            query = Q(id=task_id, company=company_id)
-            if not allow_locked:
-                query &= Q(status__nin=LOCKED_TASK_STATUSES)
-                extra_msg = "or task published"
-            res = Task.objects(query).only("id").first()
-            if not res:
-                raise errors.bad_request.InvalidTaskId(
-                    extra_msg, company=company_id, id=task_id
-                )
+        self._validate_task_state(
+            company_id=company_id, task_id=task_id, allow_locked=allow_locked
+        )
 
         es_req = {"query": {"term": {"task": task_id}}}
         with translate_errors_context(), TimingContext("es", "delete_task_events"):
@@ -959,13 +1049,58 @@ class EventBLL(object):
 
         return es_res.get("deleted", 0)
 
+    def clear_task_log(
+        self,
+        company_id: str,
+        task_id: str,
+        allow_locked: bool = False,
+        threshold_sec: int = None,
+    ):
+        self._validate_task_state(
+            company_id=company_id, task_id=task_id, allow_locked=allow_locked
+        )
+        if check_empty_data(
+            self.es, company_id=company_id, event_type=EventType.task_log
+        ):
+            return 0
+
+        with translate_errors_context(), TimingContext("es", "clear_task_log"):
+            must = [{"term": {"task": task_id}}]
+            sort = None
+            if threshold_sec:
+                timestamp_ms = int(threshold_sec * 1000)
+                must.append(
+                    {
+                        "range": {
+                            "timestamp": {
+                                "lt": (es_factory.get_timestamp_millis() - timestamp_ms)
+                            }
+                        }
+                    }
+                )
+                sort = {"timestamp": {"order": "desc"}}
+            es_req = {
+                "query": {"bool": {"must": must}},
+                **({"sort": sort} if sort else {}),
+            }
+            es_res = delete_company_events(
+                es=self.es,
+                company_id=company_id,
+                event_type=EventType.task_log,
+                body=es_req,
+                refresh=True,
+            )
+            return es_res.get("deleted", 0)
+
     def delete_multi_task_events(self, company_id: str, task_ids: Sequence[str]):
         """
         Delete mutliple task events. No check is done for tasks write access
         so it should be checked by the calling code
         """
         es_req = {"query": {"terms": {"task": task_ids}}}
-        with translate_errors_context(), TimingContext("es", "delete_multi_tasks_events"):
+        with translate_errors_context(), TimingContext(
+            "es", "delete_multi_tasks_events"
+        ):
             es_res = delete_company_events(
                 es=self.es,
                 company_id=company_id,
@@ -975,3 +1110,16 @@ class EventBLL(object):
             )
 
         return es_res.get("deleted", 0)
+
+    def clear_scroll(self, scroll_id: str):
+        if scroll_id == self.empty_scroll:
+            return
+        # noinspection PyBroadException
+        try:
+            self.es.clear_scroll(scroll_id=scroll_id)
+        except elasticsearch.exceptions.NotFoundError:
+            pass
+        except elasticsearch.exceptions.RequestError:
+            pass
+        except Exception as ex:
+            log.exception("Failed clearing scroll %s", scroll_id)

apiserver/bll/event/event_common.py

@@ -1,10 +1,15 @@
+import base64
+import zlib
 from enum import Enum
-from typing import Union, Sequence
+from typing import Union, Sequence, Mapping, Tuple
 
 from boltons.typeutils import classproperty
 from elasticsearch import Elasticsearch
 
 from apiserver.config_repo import config
+from apiserver.database.errors import translate_errors_context
+from apiserver.timing_context import TimingContext
+from apiserver.tools import safe_get
 
 
 class EventType(Enum):
@@ -16,7 +21,13 @@ class EventType(Enum):
     all = "*"
 
 
+SINGLE_SCALAR_ITERATION = -2**31
+MetricVariants = Mapping[str, Sequence[str]]
+
+
 class EventSettings:
+    _max_es_allowed_aggregation_buckets = 10000
+
     @classproperty
     def max_workers(self):
         return config.get("services.events.events_retrieval.max_metrics_concurrency", 4)
@@ -28,17 +39,23 @@ class EventSettings:
         )
 
     @classproperty
-    def max_metrics_count(self):
-        return config.get("services.events.events_retrieval.max_metrics_count", 100)
-
-    @classproperty
-    def max_variants_count(self):
-        return config.get("services.events.events_retrieval.max_variants_count", 100)
+    def max_es_buckets(self):
+        percentage = (
+            min(
+                100,
+                config.get(
+                    "services.events.events_retrieval.dynamic_metrics_count_threshold",
+                    80,
+                ),
+            )
+            / 100
+        )
+        return int(self._max_es_allowed_aggregation_buckets * percentage)
 
 
 def get_index_name(company_id: str, event_type: str):
     event_type = event_type.lower().replace(" ", "_")
-    return f"events-{event_type}-{company_id}"
+    return f"events-{event_type}-{company_id.lower()}"
 
 
 def check_empty_data(es: Elasticsearch, company_id: str, event_type: EventType) -> bool:
@@ -63,4 +80,87 @@ def delete_company_events(
     es: Elasticsearch, company_id: str, event_type: EventType, body: dict, **kwargs
 ) -> dict:
     es_index = get_index_name(company_id, event_type.value)
-    return es.delete_by_query(index=es_index, body=body, **kwargs)
+    return es.delete_by_query(
+        index=es_index, body=body, conflicts="proceed", **kwargs
+    )
+
+
+def count_company_events(
+    es: Elasticsearch, company_id: str, event_type: EventType, body: dict, **kwargs
+) -> dict:
+    es_index = get_index_name(company_id, event_type.value)
+    return es.count(index=es_index, body=body, **kwargs)
+
+
+def get_max_metric_and_variant_counts(
+    es: Elasticsearch,
+    company_id: Union[str, Sequence[str]],
+    event_type: EventType,
+    query: dict,
+    **kwargs,
+) -> Tuple[int, int]:
+    dynamic = config.get(
+        "services.events.events_retrieval.dynamic_metrics_count", False
+    )
+    max_metrics_count = config.get(
+        "services.events.events_retrieval.max_metrics_count", 100
+    )
+    max_variants_count = config.get(
+        "services.events.events_retrieval.max_variants_count", 100
+    )
+    if not dynamic:
+        return max_metrics_count, max_variants_count
+
+    es_req: dict = {
+        "size": 0,
+        "query": query,
+        "aggs": {"metrics_count": {"cardinality": {"field": "metric"}}},
+    }
+    with translate_errors_context(), TimingContext(
+        "es", "get_max_metric_and_variant_counts"
+    ):
+        es_res = search_company_events(
+            es, company_id=company_id, event_type=event_type, body=es_req, **kwargs,
+        )
+
+    metrics_count = safe_get(
+        es_res, "aggregations/metrics_count/value", max_metrics_count
+    )
+    if not metrics_count:
+        return max_metrics_count, max_variants_count
+
+    return metrics_count, int(EventSettings.max_es_buckets / metrics_count)
+
+
+def get_metric_variants_condition(metric_variants: MetricVariants,) -> Sequence:
+    conditions = [
+        {
+            "bool": {
+                "must": [
+                    {"term": {"metric": metric}},
+                    {"terms": {"variant": variants}},
+                ]
+            }
+        }
+        if variants
+        else {"term": {"metric": metric}}
+        for metric, variants in metric_variants.items()
+    ]
+
+    return {"bool": {"should": conditions}}
+
+
+class PlotFields:
+    valid_plot = "valid_plot"
+    plot_len = "plot_len"
+    plot_str = "plot_str"
+    plot_data = "plot_data"
+    source_urls = "source_urls"
+
+
+def uncompress_plot(event: dict):
+    plot_data = event.pop(PlotFields.plot_data, None)
+    if plot_data and event.get(PlotFields.plot_str) is None:
+        event[PlotFields.plot_str] = zlib.decompress(
+            base64.b64decode(plot_data)
+        ).decode()

apiserver/bll/event/event_metrics.py

@@ -4,8 +4,9 @@ from collections import defaultdict
 from concurrent.futures.thread import ThreadPoolExecutor
 from functools import partial
 from operator import itemgetter
-from typing import Sequence, Tuple
+from typing import Sequence, Tuple, Mapping
 
+from boltons.iterutils import bucketize
 from elasticsearch import Elasticsearch
 from mongoengine import Q
 
@@ -15,6 +16,10 @@ from apiserver.bll.event.event_common import (
     EventSettings,
     search_company_events,
     check_empty_data,
+    MetricVariants,
+    get_metric_variants_condition,
+    get_max_metric_and_variant_counts,
+    SINGLE_SCALAR_ITERATION,
 )
 from apiserver.bll.event.scalar_key import ScalarKey, ScalarKeyEnum
 from apiserver.config_repo import config
@@ -34,7 +39,12 @@ class EventMetrics:
         self.es = es
 
     def get_scalar_metrics_average_per_iter(
-        self, company_id: str, task_id: str, samples: int, key: ScalarKeyEnum
+        self,
+        company_id: str,
+        task_id: str,
+        samples: int,
+        key: ScalarKeyEnum,
+        metric_variants: MetricVariants = None,
     ) -> dict:
         """
         Get scalar metric histogram per metric and variant
@@ -46,7 +56,12 @@ class EventMetrics:
             return {}
 
         return self._get_scalar_average_per_iter_core(
-            task_id, company_id, event_type, samples, ScalarKey.resolve(key)
+            task_id=task_id,
+            company_id=company_id,
+            event_type=event_type,
+            samples=samples,
+            key=ScalarKey.resolve(key),
+            metric_variants=metric_variants,
         )
 
     def _get_scalar_average_per_iter_core(
@@ -57,6 +72,7 @@ class EventMetrics:
         samples: int,
         key: ScalarKey,
         run_parallel: bool = True,
+        metric_variants: MetricVariants = None,
     ) -> dict:
         intervals = self._get_task_metric_intervals(
             company_id=company_id,
@@ -64,6 +80,7 @@ class EventMetrics:
             task_id=task_id,
             samples=samples,
             field=key.field,
+            metric_variants=metric_variants,
         )
         if not intervals:
             return {}
@@ -152,6 +169,57 @@ class EventMetrics:
 
         return res
 
+    def get_task_single_value_metrics(
+        self, company_id: str, task_ids: Sequence[str]
+    ) -> Mapping[str, dict]:
+        """
+        For the requested tasks return all the events delivered for the single iteration (-2**31)
+        """
+        if check_empty_data(
+            self.es, company_id=company_id, event_type=EventType.metrics_scalar
+        ):
+            return {}
+
+        with TimingContext("es", "get_task_single_value_metrics"):
+            task_events = self._get_task_single_value_metrics(company_id, task_ids)
+
+        def _get_value(event: dict):
+            return {
+                field: event.get(field)
+                for field in ("metric", "variant", "value", "timestamp")
+            }
+
+        return {
+            task: [_get_value(e) for e in events]
+            for task, events in bucketize(task_events, itemgetter("task")).items()
+        }
+
+    def _get_task_single_value_metrics(
+        self, company_id: str, task_ids: Sequence[str]
+    ) -> Sequence[dict]:
+        es_req = {
+            "size": 10000,
+            "query": {
+                "bool": {
+                    "must": [
+                        {"terms": {"task": task_ids}},
+                        {"term": {"iter": SINGLE_SCALAR_ITERATION}},
+                    ]
+                }
+            },
+        }
+        with translate_errors_context():
+            es_res = search_company_events(
+                body=es_req,
+                es=self.es,
+                company_id=company_id,
+                event_type=EventType.metrics_scalar,
+            )
+            if not es_res["hits"]["total"]["value"]:
+                return []
+
+        return [hit["_source"] for hit in es_res["hits"]["hits"]]
+
     MetricInterval = Tuple[str, str, int, int]
     MetricIntervalGroup = Tuple[int, Sequence[Tuple[str, str]]]
 
@@ -197,6 +265,7 @@ class EventMetrics:
         task_id: str,
         samples: int,
         field: str = "iter",
+        metric_variants: MetricVariants = None,
     ) -> Sequence[MetricInterval]:
         """
         Calculate interval per task metric variant so that the resulting
@@ -204,21 +273,32 @@ class EventMetrics:
         Return the list og metric variant intervals as the following tuple:
         (metric, variant, interval, samples)
         """
+        must = self._task_conditions(task_id)
+        if metric_variants:
+            must.append(get_metric_variants_condition(metric_variants))
+        query = {"bool": {"must": must}}
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args,
+        )
+        max_variants = int(max_variants // 2)
         es_req = {
             "size": 0,
-            "query": {"term": {"task": task_id}},
+            "query": query,
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
                         "variants": {
                             "terms": {
                                 "field": "variant",
-                                "size": EventSettings.max_variants_count,
+                                "size": max_variants,
                                 "order": {"_key": "asc"},
                             },
                             "aggs": {
@@ -233,9 +313,7 @@ class EventMetrics:
         }
 
         with translate_errors_context(), TimingContext("es", "task_stats_get_interval"):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req,
-            )
+            es_res = search_company_events(body=es_req, **search_args)
 
         aggs_result = es_res.get("aggregations")
         if not aggs_result:
@@ -287,33 +365,42 @@ class EventMetrics:
         """
         interval, metrics = metrics_interval
         aggregation = self._add_aggregation_average(key.get_aggregation(interval))
-        aggs = {
-            "metrics": {
-                "terms": {
-                    "field": "metric",
-                    "size": EventSettings.max_metrics_count,
-                    "order": {"_key": "asc"},
-                },
-                "aggs": {
-                    "variants": {
-                        "terms": {
-                            "field": "variant",
-                            "size": EventSettings.max_variants_count,
-                            "order": {"_key": "asc"},
-                        },
-                        "aggs": aggregation,
-                    }
-                },
-            }
-        }
-        aggs_result = self._query_aggregation_for_task_metrics(
-            company_id=company_id,
-            event_type=event_type,
-            aggs=aggs,
-            task_id=task_id,
-            metrics=metrics,
+        query = self._get_task_metrics_query(task_id=task_id, metrics=metrics)
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=event_type
         )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args,
+        )
+        max_variants = int(max_variants // 2)
+        es_req = {
+            "size": 0,
+            "query": query,
+            "aggs": {
+                "metrics": {
+                    "terms": {
+                        "field": "metric",
+                        "size": max_metrics,
+                        "order": {"_key": "asc"},
+                    },
+                    "aggs": {
+                        "variants": {
+                            "terms": {
+                                "field": "variant",
+                                "size": max_variants,
+                                "order": {"_key": "asc"},
+                            },
+                            "aggs": aggregation,
+                        }
+                    },
+                }
+            },
+        }
 
+        with translate_errors_context():
+            es_res = search_company_events(body=es_req, **search_args)
+
+        aggs_result = es_res.get("aggregations")
         if not aggs_result:
             return {}
 
@@ -340,19 +427,18 @@ class EventMetrics:
             for key, value in aggregation.items()
         }
 
-    def _query_aggregation_for_task_metrics(
-        self,
-        company_id: str,
-        event_type: EventType,
-        aggs: dict,
-        task_id: str,
-        metrics: Sequence[Tuple[str, str]],
-    ) -> dict:
-        """
-        Return the result of elastic search query for the given aggregation filtered
-        by the given task_ids and metrics
-        """
-        must = [{"term": {"task": task_id}}]
+    @staticmethod
+    def _task_conditions(task_id: str) -> list:
+        return [
+            {"term": {"task": task_id}},
+            {"range": {"iter": {"gt": SINGLE_SCALAR_ITERATION}}},
+        ]
+
+    @classmethod
+    def _get_task_metrics_query(
+        cls, task_id: str, metrics: Sequence[Tuple[str, str]],
+    ):
+        must = cls._task_conditions(task_id)
         if metrics:
             should = [
                 {
@@ -367,20 +453,9 @@ class EventMetrics:
             ]
             must.append({"bool": {"should": should}})
 
-        es_req = {
-            "size": 0,
-            "query": {"bool": {"must": must}},
-            "aggs": aggs,
-        }
+        return {"bool": {"must": must}}
 
-        with translate_errors_context(), TimingContext("es", "task_stats_scalar"):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req,
-            )
-
-        return es_res.get("aggregations")
-
-    def get_tasks_metrics(
+    def get_task_metrics(
         self, company_id, task_ids: Sequence, event_type: EventType
     ) -> Sequence:
         """
@@ -406,12 +481,12 @@ class EventMetrics:
     ) -> Sequence:
         es_req = {
             "size": 0,
-            "query": {"bool": {"must": [{"term": {"task": task_id}}]}},
+            "query": {"bool": {"must": self._task_conditions(task_id)}},
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": EventSettings.max_es_buckets,
                         "order": {"_key": "asc"},
                     }
                 }

apiserver/bll/event/events_iterator.py

@@ -0,0 +1,198 @@
+from typing import Optional, Tuple, Sequence, Any
+
+import attr
+import jsonmodels.models
+import jwt
+from elasticsearch import Elasticsearch
+from jwt.algorithms import get_default_algorithms
+
+from apiserver.bll.event.event_common import (
+    check_empty_data,
+    search_company_events,
+    EventType,
+    MetricVariants,
+    get_metric_variants_condition,
+    count_company_events,
+)
+from apiserver.bll.event.scalar_key import ScalarKeyEnum, ScalarKey
+from apiserver.config_repo import config
+from apiserver.database.errors import translate_errors_context
+from apiserver.timing_context import TimingContext
+
+
+@attr.s(auto_attribs=True)
+class TaskEventsResult:
+    total_events: int = 0
+    next_scroll_id: str = None
+    events: list = attr.Factory(list)
+
+
+class EventsIterator:
+    def __init__(self, es: Elasticsearch):
+        self.es = es
+
+    def get_task_events(
+        self,
+        event_type: EventType,
+        company_id: str,
+        task_id: str,
+        batch_size: int,
+        navigate_earlier: bool = True,
+        from_key_value: Optional[Any] = None,
+        metric_variants: MetricVariants = None,
+        key: ScalarKeyEnum = ScalarKeyEnum.timestamp,
+        **kwargs,
+    ) -> TaskEventsResult:
+        if check_empty_data(self.es, company_id, event_type):
+            return TaskEventsResult()
+
+        from_key_value = kwargs.pop("from_timestamp", from_key_value)
+
+        res = TaskEventsResult()
+        res.events, res.total_events = self._get_events(
+            event_type=event_type,
+            company_id=company_id,
+            task_id=task_id,
+            batch_size=batch_size,
+            navigate_earlier=navigate_earlier,
+            from_key_value=from_key_value,
+            metric_variants=metric_variants,
+            key=ScalarKey.resolve(key),
+        )
+        return res
+
+    def count_task_events(
+        self,
+        event_type: EventType,
+        company_id: str,
+        task_id: str,
+        metric_variants: MetricVariants = None,
+    ) -> int:
+        if check_empty_data(self.es, company_id, event_type):
+            return 0
+
+        query, _ = self._get_initial_query_and_must(task_id, metric_variants)
+        es_req = {
+            "query": query,
+        }
+
+        with translate_errors_context(), TimingContext("es", "count_task_events"):
+            es_result = count_company_events(
+                self.es, company_id=company_id, event_type=event_type, body=es_req,
+            )
+
+            return es_result["count"]
+
+    def _get_events(
+        self,
+        event_type: EventType,
+        company_id: str,
+        task_id: str,
+        batch_size: int,
+        navigate_earlier: bool,
+        key: ScalarKey,
+        from_key_value: Optional[Any],
+        metric_variants: MetricVariants = None,
+    ) -> Tuple[Sequence[dict], int]:
+        """
+        Return up to 'batch size' events starting from the previous key-field value (timestamp or iter) either in the
+        direction of earlier events (navigate_earlier=True) or in the direction of later events.
+        If from_key_field is not set then start either from latest or earliest.
+        For the last key-field value all the events are brought (even if the resulting size exceeds batch_size)
+        so that events with this value will not be lost between the calls.
+        """
+        query, must = self._get_initial_query_and_must(task_id, metric_variants)
+
+        # retrieve the next batch of events
+        es_req = {
+            "size": batch_size,
+            "query": query,
+            "sort": {key.field: "desc" if navigate_earlier else "asc"},
+        }
+
+        if from_key_value:
+            es_req["search_after"] = [from_key_value]
+
+        with translate_errors_context(), TimingContext("es", "get_task_events"):
+            es_result = search_company_events(
+                self.es, company_id=company_id, event_type=event_type, body=es_req,
+            )
+            hits = es_result["hits"]["hits"]
+            hits_total = es_result["hits"]["total"]["value"]
+            if not hits:
+                return [], hits_total
+
+            events = [hit["_source"] for hit in hits]
+
+            # retrieve the events that match the last event timestamp
+            # but did not make it into the previous call due to batch_size limitation
+            es_req = {
+                "size": 10000,
+                "query": {
+                    "bool": {
+                        "must": must + [{"term": {key.field: events[-1][key.field]}}]
+                    }
+                },
+            }
+            es_result = search_company_events(
+                self.es, company_id=company_id, event_type=event_type, body=es_req,
+            )
+            last_second_hits = es_result["hits"]["hits"]
+            if not last_second_hits or len(last_second_hits) < 2:
+                # if only one element is returned for the last timestamp
+                # then it is already present in the events
+                return events, hits_total
+
+            already_present_ids = set(hit["_id"] for hit in hits)
+            last_second_events = [
+                hit["_source"]
+                for hit in last_second_hits
+                if hit["_id"] not in already_present_ids
+            ]
+
+            # return the list merged from original query results +
+            # leftovers from the last timestamp
+            return (
+                [*events, *last_second_events],
+                hits_total,
+            )
+
+    @staticmethod
+    def _get_initial_query_and_must(
+        task_id: str, metric_variants: MetricVariants = None
+    ) -> Tuple[dict, list]:
+        if not metric_variants:
+            must = [{"term": {"task": task_id}}]
+            query = {"term": {"task": task_id}}
+        else:
+            must = [
+                {"term": {"task": task_id}},
+                get_metric_variants_condition(metric_variants),
+            ]
+            query = {"bool": {"must": must}}
+        return query, must
+
+
+class Scroll(jsonmodels.models.Base):
+    def get_scroll_id(self) -> str:
+        return jwt.encode(
+            self.to_struct(),
+            key=config.get(
+                "services.events.events_retrieval.scroll_id_key", "1234567890"
+            ),
+        )
+
+    @classmethod
+    def from_scroll_id(cls, scroll_id: str):
+        try:
+            return cls(
+                **jwt.decode(
+                    scroll_id,
+                    key=config.get(
+                        "services.events.events_retrieval.scroll_id_key", "1234567890"
+                    ),
+                    algorithms=get_default_algorithms(),
+                )
+            )
+        except jwt.PyJWTError:
+            raise ValueError("Invalid Scroll ID")

apiserver/bll/event/history_debug_image_iterator.py

@@ -0,0 +1,56 @@
+from typing import Sequence, Tuple, Callable
+
+from elasticsearch import Elasticsearch
+from redis.client import StrictRedis
+
+from apiserver.utilities.dicts import nested_get
+from .event_common import EventType
+from .history_sample_iterator import HistorySampleIterator, VariantState
+
+
+class HistoryDebugImageIterator(HistorySampleIterator):
+    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+        super().__init__(redis, es, EventType.metrics_image)
+
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        return [{"exists": {"field": "url"}}]
+
+    def _get_variants_conditions(self, variants: Sequence[VariantState]) -> dict:
+        variants_conditions = [
+            {
+                "bool": {
+                    "must": [
+                        {"term": {"variant": v.name}},
+                        {"range": {"iter": {"gte": v.min_iteration}}},
+                    ]
+                }
+            }
+            for v in variants
+        ]
+        return {"bool": {"should": variants_conditions}}
+
+    def _process_event(self, event: dict) -> dict:
+        return event
+
+    def _get_min_max_aggs(self) -> Tuple[dict, Callable[[dict], Tuple[int, int]]]:
+        # The min iteration is the lowest iteration that contains non-recycled image url
+        aggs = {
+            "last_iter": {"max": {"field": "iter"}},
+            "urls": {
+                # group by urls and choose the minimal iteration
+                # from all the maximal iterations per url
+                "terms": {"field": "url", "order": {"max_iter": "asc"}, "size": 1},
+                "aggs": {
+                    # find max iteration for each url
+                    "max_iter": {"max": {"field": "iter"}}
+                },
+            },
+        }
+
+        def get_min_max_data(variant_bucket: dict) -> Tuple[int, int]:
+            urls = nested_get(variant_bucket, ("urls", "buckets"))
+            min_iter = int(urls[0]["max_iter"]["value"])
+            max_iter = int(variant_bucket["last_iter"]["value"])
+            return min_iter, max_iter
+
+        return aggs, get_min_max_data

apiserver/bll/event/history_plot_iterator.py

@@ -0,0 +1,36 @@
+from typing import Sequence, Tuple, Callable
+
+from elasticsearch import Elasticsearch
+from redis.client import StrictRedis
+
+from .event_common import EventType, uncompress_plot
+from .history_sample_iterator import HistorySampleIterator, VariantState
+
+
+class HistoryPlotIterator(HistorySampleIterator):
+    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+        super().__init__(redis, es, EventType.metrics_plot)
+
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        return []
+
+    def _get_variants_conditions(self, variants: Sequence[VariantState]) -> dict:
+        return {"terms": {"variant": [v.name for v in variants]}}
+
+    def _process_event(self, event: dict) -> dict:
+        uncompress_plot(event)
+        return event
+
+    def _get_min_max_aggs(self) -> Tuple[dict, Callable[[dict], Tuple[int, int]]]:
+        # The min iteration is the lowest iteration that contains non-recycled image url
+        aggs = {
+            "last_iter": {"max": {"field": "iter"}},
+            "first_iter": {"min": {"field": "iter"}},
+        }
+
+        def get_min_max_data(variant_bucket: dict) -> Tuple[int, int]:
+            min_iter = int(variant_bucket["first_iter"]["value"])
+            max_iter = int(variant_bucket["last_iter"]["value"])
+            return min_iter, max_iter
+
+        return aggs, get_min_max_data

apiserver/bll/event/history_sample_iterator.py

@@ -1,8 +1,10 @@
+import abc
 import operator
-from typing import Sequence, Tuple, Optional
+from operator import attrgetter
+from typing import Sequence, Tuple, Optional, Callable, Mapping
 
 import attr
-from boltons.iterutils import first
+from boltons.iterutils import first, bucketize
 from elasticsearch import Elasticsearch
 from jsonmodels.fields import StringField, ListField, IntField, BoolField
 from jsonmodels.models import Base
@@ -15,6 +17,7 @@ from apiserver.bll.event.event_common import (
     EventType,
     check_empty_data,
     search_company_events,
+    get_max_metric_and_variant_counts,
 )
 from apiserver.bll.redis_cache_manager import RedisCacheManager
 from apiserver.database.errors import translate_errors_context
@@ -24,11 +27,12 @@ from apiserver.utilities.dicts import nested_get
 
 class VariantState(Base):
     name: str = StringField(required=True)
+    metric: str = StringField(default=None)
     min_iteration: int = IntField()
     max_iteration: int = IntField()
 
 
-class DebugSampleHistoryState(Base, JsonSerializableMixin):
+class HistorySampleState(Base, JsonSerializableMixin):
     id: str = StringField(required=True)
     iteration: int = IntField()
     variant: str = StringField()
@@ -38,78 +42,120 @@ class DebugSampleHistoryState(Base, JsonSerializableMixin):
     reached_last: bool = BoolField()
     variant_states: Sequence[VariantState] = ListField([VariantState])
     warning: str = StringField()
+    navigate_current_metric = BoolField(default=True)
 
 
 @attr.s(auto_attribs=True)
-class DebugSampleHistoryResult(object):
+class HistorySampleResult(object):
     scroll_id: str = None
     event: dict = None
     min_iteration: int = None
     max_iteration: int = None
 
 
-class DebugSampleHistory:
-    EVENT_TYPE = EventType.metrics_image
-
-    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+class HistorySampleIterator(abc.ABC):
+    def __init__(self, redis: StrictRedis, es: Elasticsearch, event_type: EventType):
         self.es = es
+        self.event_type = event_type
         self.cache_manager = RedisCacheManager(
-            state_class=DebugSampleHistoryState,
+            state_class=HistorySampleState,
             redis=redis,
             expiration_interval=EventSettings.state_expiration_sec,
         )
 
-    def get_next_debug_image(
+    def get_next_sample(
         self, company_id: str, task: str, state_id: str, navigate_earlier: bool
-    ) -> DebugSampleHistoryResult:
+    ) -> HistorySampleResult:
         """
-        Get the debug image for next/prev variant on the current iteration
-        If does not exist then try getting image for the first/last variant from next/prev iteration
+        Get the sample for next/prev variant on the current iteration
+        If does not exist then try getting sample for the first/last variant from next/prev iteration
         """
-        res = DebugSampleHistoryResult(scroll_id=state_id)
+        res = HistorySampleResult(scroll_id=state_id)
         state = self.cache_manager.get_state(state_id)
         if not state or state.task != task:
             raise errors.bad_request.InvalidScrollId(scroll_id=state_id)
 
-        if check_empty_data(self.es, company_id=company_id, event_type=self.EVENT_TYPE):
+        if check_empty_data(self.es, company_id=company_id, event_type=self.event_type):
             return res
 
-        image = self._get_next_for_current_iteration(
+        event = self._get_next_for_current_iteration(
             company_id=company_id, navigate_earlier=navigate_earlier, state=state
         ) or self._get_next_for_another_iteration(
             company_id=company_id, navigate_earlier=navigate_earlier, state=state
         )
-        if not image:
+        if not event:
             return res
 
-        self._fill_res_and_update_state(image=image, res=res, state=state)
+        self._fill_res_and_update_state(event=event, res=res, state=state)
         self.cache_manager.set_state(state=state)
         return res
 
     def _fill_res_and_update_state(
-        self, image: dict, res: DebugSampleHistoryResult, state: DebugSampleHistoryState
+        self, event: dict, res: HistorySampleResult, state: HistorySampleState
     ):
-        state.variant = image["variant"]
-        state.iteration = image["iter"]
-        res.event = image
-        var_state = first(s for s in state.variant_states if s.name == state.variant)
+        self._process_event(event)
+        state.variant = event["variant"]
+        state.metric = event["metric"]
+        state.iteration = event["iter"]
+        res.event = event
+        var_state = first(
+            vs
+            for vs in state.variant_states
+            if vs.name == state.variant and vs.metric == state.metric
+        )
         if var_state:
             res.min_iteration = var_state.min_iteration
             res.max_iteration = var_state.max_iteration
 
+    @abc.abstractmethod
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        pass
+
+    @abc.abstractmethod
+    def _process_event(self, event: dict) -> dict:
+        pass
+
+    @abc.abstractmethod
+    def _get_variants_conditions(self, variants: Sequence[VariantState]) -> dict:
+        pass
+
+    def _get_metric_variants_condition(self, variants: Sequence[VariantState]) -> dict:
+        metrics = bucketize(variants, key=attrgetter("metric"))
+        metrics_conditions = [
+            {
+                "bool": {
+                    "must": [
+                        {"term": {"metric": metric}},
+                        self._get_variants_conditions(vs),
+                    ]
+                }
+            }
+            for metric, vs in metrics.items()
+        ]
+        return {"bool": {"should": metrics_conditions}}
+
     def _get_next_for_current_iteration(
-        self, company_id: str, navigate_earlier: bool, state: DebugSampleHistoryState
+        self, company_id: str, navigate_earlier: bool, state: HistorySampleState
     ) -> Optional[dict]:
         """
-        Get the image for next (if navigated earlier is False) or previous variant sorted by name for the same iteration
+        Get the sample for next (if navigate_earlier is False) or previous variant sorted by name for the same iteration
         Only variants for which the iteration falls into their valid range are considered
-        Return None if no such variant or image is found
+        Return None if no such variant or sample is found
         """
+        if state.navigate_current_metric:
+            variants = [
+                var_state
+                for var_state in state.variant_states
+                if var_state.metric == state.metric
+            ]
+        else:
+            variants = state.variant_states
+
         cmp = operator.lt if navigate_earlier else operator.gt
         variants = [
             var_state
-            for var_state in state.variant_states
-            if cmp(var_state.name, state.variant)
+            for var_state in variants
+            if cmp((var_state.metric, var_state.name), (state.metric, state.variant))
             and var_state.min_iteration <= state.iteration
         ]
         if not variants:
@@ -117,14 +163,14 @@ class DebugSampleHistory:
 
         must_conditions = [
             {"term": {"task": state.task}},
-            {"term": {"metric": state.metric}},
-            {"terms": {"variant": [v.name for v in variants]}},
             {"term": {"iter": state.iteration}},
-            {"exists": {"field": "url"}},
+            self._get_metric_variants_condition(variants),
+            *self._get_extra_conditions(),
         ]
+        order = "desc" if navigate_earlier else "asc"
         es_req = {
             "size": 1,
-            "sort": {"variant": "desc" if navigate_earlier else "asc"},
+            "sort": [{"metric": order}, {"variant": order}],
             "query": {"bool": {"must": must_conditions}},
         }
 
@@ -132,7 +178,10 @@ class DebugSampleHistory:
             "es", "get_next_for_current_iteration"
         ):
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.EVENT_TYPE, body=es_req
+                self.es,
+                company_id=company_id,
+                event_type=self.event_type,
+                body=es_req,
             )
 
         hits = nested_get(es_res, ("hits", "hits"))
@@ -142,61 +191,58 @@ class DebugSampleHistory:
         return hits[0]["_source"]
 
     def _get_next_for_another_iteration(
-        self, company_id: str, navigate_earlier: bool, state: DebugSampleHistoryState
+        self, company_id: str, navigate_earlier: bool, state: HistorySampleState
     ) -> Optional[dict]:
         """
-        Get the image for the first variant for the next iteration (if navigate_earlier is set to False)
+        Get the sample for the first variant for the next iteration (if navigate_earlier is set to False)
         or from the last variant for the previous iteration (otherwise)
-        The variants for which the image falls in invalid range are discarded
-        If no suitable image is found then None is returned
+        The variants for which the sample falls in invalid range are discarded
+        If no suitable sample is found then None is returned
         """
-
-        must_conditions = [
-            {"term": {"task": state.task}},
-            {"term": {"metric": state.metric}},
-            {"exists": {"field": "url"}},
-        ]
+        if state.navigate_current_metric:
+            variants = [
+                var_state
+                for var_state in state.variant_states
+                if var_state.metric == state.metric
+            ]
+        else:
+            variants = state.variant_states
 
         if navigate_earlier:
             range_operator = "lt"
             order = "desc"
             variants = [
                 var_state
-                for var_state in state.variant_states
+                for var_state in variants
                 if var_state.min_iteration < state.iteration
             ]
         else:
             range_operator = "gt"
             order = "asc"
-            variants = state.variant_states
+            variants = variants
 
         if not variants:
             return
 
-        variants_conditions = [
-            {
-                "bool": {
-                    "must": [
-                        {"term": {"variant": v.name}},
-                        {"range": {"iter": {"gte": v.min_iteration}}},
-                    ]
-                }
-            }
-            for v in variants
+        must_conditions = [
+            {"term": {"task": state.task}},
+            self._get_metric_variants_condition(variants),
+            {"range": {"iter": {range_operator: state.iteration}}},
+            *self._get_extra_conditions(),
         ]
-        must_conditions.append({"bool": {"should": variants_conditions}})
-        must_conditions.append({"range": {"iter": {range_operator: state.iteration}}},)
-
         es_req = {
             "size": 1,
-            "sort": [{"iter": order}, {"variant": order}],
+            "sort": [{"iter": order}, {"metric": order}, {"variant": order}],
             "query": {"bool": {"must": must_conditions}},
         }
         with translate_errors_context(), TimingContext(
             "es", "get_next_for_another_iteration"
         ):
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.EVENT_TYPE, body=es_req
+                self.es,
+                company_id=company_id,
+                event_type=self.event_type,
+                body=es_req,
             )
 
         hits = nested_get(es_res, ("hits", "hits"))
@@ -205,7 +251,7 @@ class DebugSampleHistory:
 
         return hits[0]["_source"]
 
-    def get_debug_image_for_variant(
+    def get_sample_for_variant(
         self,
         company_id: str,
         task: str,
@@ -214,36 +260,50 @@ class DebugSampleHistory:
         iteration: Optional[int] = None,
         refresh: bool = False,
         state_id: str = None,
-    ) -> DebugSampleHistoryResult:
+        navigate_current_metric: bool = True,
+    ) -> HistorySampleResult:
         """
-        Get the debug image for the requested iteration or the latest before it
+        Get the sample for the requested iteration or the latest before it
         If the iteration is not passed then get the latest event
         """
-        res = DebugSampleHistoryResult()
-        if check_empty_data(self.es, company_id=company_id, event_type=self.EVENT_TYPE):
+        res = HistorySampleResult()
+        if check_empty_data(self.es, company_id=company_id, event_type=self.event_type):
             return res
 
-        def init_state(state_: DebugSampleHistoryState):
+        def init_state(state_: HistorySampleState):
             state_.task = task
             state_.metric = metric
+            state_.navigate_current_metric = navigate_current_metric
             self._reset_variant_states(company_id=company_id, state=state_)
 
-        def validate_state(state_: DebugSampleHistoryState):
-            if state_.task != task or state_.metric != metric:
+        def validate_state(state_: HistorySampleState):
+            if (
+                state_.task != task
+                or state_.navigate_current_metric != navigate_current_metric
+                or (state_.navigate_current_metric and state_.metric != metric)
+            ):
                 raise errors.bad_request.InvalidScrollId(
                     "Task and metric stored in the state do not match the passed ones",
                     scroll_id=state_.id,
                 )
+            # fix old variant states:
+            for vs in state_.variant_states:
+                if vs.metric is None:
+                    vs.metric = metric
             if refresh:
                 self._reset_variant_states(company_id=company_id, state=state_)
 
-        state: DebugSampleHistoryState
+        state: HistorySampleState
         with self.cache_manager.get_or_create_state(
             state_id=state_id, init_state=init_state, validate_state=validate_state,
         ) as state:
             res.scroll_id = state.id
 
-            var_state = first(s for s in state.variant_states if s.name == variant)
+            var_state = first(
+                vs
+                for vs in state.variant_states
+                if vs.name == variant and vs.metric == metric
+            )
             if not var_state:
                 return res
 
@@ -254,7 +314,7 @@ class DebugSampleHistory:
                 {"term": {"task": task}},
                 {"term": {"metric": metric}},
                 {"term": {"variant": variant}},
-                {"exists": {"field": "url"}},
+                *self._get_extra_conditions(),
             ]
             if iteration is not None:
                 must_conditions.append(
@@ -276,12 +336,12 @@ class DebugSampleHistory:
             }
 
             with translate_errors_context(), TimingContext(
-                "es", "get_debug_image_for_variant"
+                "es", "get_history_sample_for_variant"
             ):
                 es_res = search_company_events(
                     self.es,
                     company_id=company_id,
-                    event_type=self.EVENT_TYPE,
+                    event_type=self.event_type,
                     body=es_req,
                 )
 
@@ -290,86 +350,93 @@ class DebugSampleHistory:
                 return res
 
             self._fill_res_and_update_state(
-                image=hits[0]["_source"], res=res, state=state
+                event=hits[0]["_source"], res=res, state=state
             )
             return res
 
-    def _reset_variant_states(self, company_id: str, state: DebugSampleHistoryState):
-        variant_iterations = self._get_variant_iterations(
-            company_id=company_id, task=state.task, metric=state.metric
+    def _reset_variant_states(self, company_id: str, state: HistorySampleState):
+        metrics = self._get_metric_variant_iterations(
+            company_id=company_id,
+            task=state.task,
+            metric=state.metric if state.navigate_current_metric else None,
         )
         state.variant_states = [
-            VariantState(name=var_name, min_iteration=min_iter, max_iteration=max_iter)
-            for var_name, min_iter, max_iter in variant_iterations
+            VariantState(
+                metric=metric,
+                name=var_name,
+                min_iteration=min_iter,
+                max_iteration=max_iter,
+            )
+            for metric, variants in metrics.items()
+            for var_name, min_iter, max_iter in variants
         ]
 
-    def _get_variant_iterations(
-        self,
-        company_id: str,
-        task: str,
-        metric: str,
-        variants: Optional[Sequence[str]] = None,
-    ) -> Sequence[Tuple[str, int, int]]:
+    @abc.abstractmethod
+    def _get_min_max_aggs(self) -> Tuple[dict, Callable[[dict], Tuple[int, int]]]:
+        pass
+
+    def _get_metric_variant_iterations(
+        self, company_id: str, task: str, metric: str,
+    ) -> Mapping[str, Tuple[str, str, int, int]]:
         """
-        Return valid min and max iterations that the task reported images
-        The min iteration is the lowest iteration that contains non-recycled image url
+        Return valid min and max iterations that the task reported events of the required type
         """
         must = [
             {"term": {"task": task}},
-            {"term": {"metric": metric}},
-            {"exists": {"field": "url"}},
+            *self._get_extra_conditions(),
         ]
-        if variants:
-            must.append({"terms": {"variant": variants}})
+        if metric is not None:
+            must.append({"term": {"metric": metric}})
+        query = {"bool": {"must": must}}
 
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=self.event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args
+        )
+        max_variants = int(max_variants // 2)
+        min_max_aggs, get_min_max_data = self._get_min_max_aggs()
         es_req: dict = {
             "size": 0,
-            "query": {"bool": {"must": must}},
+            "query": query,
             "aggs": {
-                "variants": {
-                    # all variants that sent debug images
+                "metrics": {
                     "terms": {
-                        "field": "variant",
-                        "size": EventSettings.max_variants_count,
+                        "field": "metric",
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
-                        "last_iter": {"max": {"field": "iter"}},
-                        "urls": {
-                            # group by urls and choose the minimal iteration
-                            # from all the maximal iterations per url
+                        "variants": {
                             "terms": {
-                                "field": "url",
-                                "order": {"max_iter": "asc"},
-                                "size": 1,
+                                "field": "variant",
+                                "size": max_variants,
+                                "order": {"_key": "asc"},
                             },
-                            "aggs": {
-                                # find max iteration for each url
-                                "max_iter": {"max": {"field": "iter"}}
-                            },
-                        },
+                            "aggs": min_max_aggs,
+                        }
                     },
                 }
             },
         }
 
         with translate_errors_context(), TimingContext(
-            "es", "get_debug_image_iterations"
+            "es", "get_history_sample_iterations"
         ):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.EVENT_TYPE, body=es_req
-            )
+            es_res = search_company_events(body=es_req, **search_args,)
 
         def get_variant_data(variant_bucket: dict) -> Tuple[str, int, int]:
             variant = variant_bucket["key"]
-            urls = nested_get(variant_bucket, ("urls", "buckets"))
-            min_iter = int(urls[0]["max_iter"]["value"])
-            max_iter = int(variant_bucket["last_iter"]["value"])
+            min_iter, max_iter = get_min_max_data(variant_bucket)
             return variant, min_iter, max_iter
 
-        return [
-            get_variant_data(variant_bucket)
-            for variant_bucket in nested_get(
-                es_res, ("aggregations", "variants", "buckets")
+        return {
+            metric_bucket["key"]: [
+                get_variant_data(variant_bucket)
+                for variant_bucket in nested_get(metric_bucket, ("variants", "buckets"))
+            ]
+            for metric_bucket in nested_get(
+                es_res, ("aggregations", "metrics", "buckets")
             )
-        ]
+        }

apiserver/bll/event/log_events_iterator.py

@@ -1,127 +0,0 @@
-from typing import Optional, Tuple, Sequence
-
-import attr
-from elasticsearch import Elasticsearch
-
-from apiserver.bll.event.event_common import (
-    check_empty_data,
-    search_company_events,
-    EventType,
-)
-from apiserver.database.errors import translate_errors_context
-from apiserver.timing_context import TimingContext
-
-
-@attr.s(auto_attribs=True)
-class TaskEventsResult:
-    total_events: int = 0
-    next_scroll_id: str = None
-    events: list = attr.Factory(list)
-
-
-class LogEventsIterator:
-    EVENT_TYPE = EventType.task_log
-
-    def __init__(self, es: Elasticsearch):
-        self.es = es
-
-    def get_task_events(
-        self,
-        company_id: str,
-        task_id: str,
-        batch_size: int,
-        navigate_earlier: bool = True,
-        from_timestamp: Optional[int] = None,
-    ) -> TaskEventsResult:
-        if check_empty_data(self.es, company_id, self.EVENT_TYPE):
-            return TaskEventsResult()
-
-        res = TaskEventsResult()
-        res.events, res.total_events = self._get_events(
-            company_id=company_id,
-            task_id=task_id,
-            batch_size=batch_size,
-            navigate_earlier=navigate_earlier,
-            from_timestamp=from_timestamp,
-        )
-        return res
-
-    def _get_events(
-        self,
-        company_id: str,
-        task_id: str,
-        batch_size: int,
-        navigate_earlier: bool,
-        from_timestamp: Optional[int],
-    ) -> Tuple[Sequence[dict], int]:
-        """
-        Return up to 'batch size' events starting from the previous timestamp either in the
-        direction of earlier events (navigate_earlier=True) or in the direction of later events.
-        If last_min_timestamp and last_max_timestamp are not set then start either from latest or earliest.
-        For the last timestamp all the events are brought (even if the resulting size
-        exceeds batch_size) so that this timestamp events will not be lost between the calls.
-        In case any events were received update 'last_min_timestamp' and 'last_max_timestamp'
-        """
-
-        # retrieve the next batch of events
-        es_req = {
-            "size": batch_size,
-            "query": {"term": {"task": task_id}},
-            "sort": {"timestamp": "desc" if navigate_earlier else "asc"},
-        }
-
-        if from_timestamp:
-            es_req["search_after"] = [from_timestamp]
-
-        with translate_errors_context(), TimingContext("es", "get_task_events"):
-            es_result = search_company_events(
-                self.es,
-                company_id=company_id,
-                event_type=self.EVENT_TYPE,
-                body=es_req,
-            )
-            hits = es_result["hits"]["hits"]
-            hits_total = es_result["hits"]["total"]["value"]
-            if not hits:
-                return [], hits_total
-
-            events = [hit["_source"] for hit in hits]
-
-            # retrieve the events that match the last event timestamp
-            # but did not make it into the previous call due to batch_size limitation
-            es_req = {
-                "size": 10000,
-                "query": {
-                    "bool": {
-                        "must": [
-                            {"term": {"task": task_id}},
-                            {"term": {"timestamp": events[-1]["timestamp"]}},
-                        ]
-                    }
-                },
-            }
-            es_result = search_company_events(
-                self.es,
-                company_id=company_id,
-                event_type=self.EVENT_TYPE,
-                body=es_req,
-            )
-            last_second_hits = es_result["hits"]["hits"]
-            if not last_second_hits or len(last_second_hits) < 2:
-                # if only one element is returned for the last timestamp
-                # then it is already present in the events
-                return events, hits_total
-
-            already_present_ids = set(hit["_id"] for hit in hits)
-            last_second_events = [
-                hit["_source"]
-                for hit in last_second_hits
-                if hit["_id"] not in already_present_ids
-            ]
-
-            # return the list merged from original query results +
-            # leftovers from the last timestamp
-            return (
-                [*events, *last_second_events],
-                hits_total,
-            )

apiserver/bll/event/metric_debug_images_iterator.py

@@ -0,0 +1,53 @@
+from typing import Sequence, Tuple, Callable
+
+from elasticsearch import Elasticsearch
+from redis.client import StrictRedis
+
+from apiserver.utilities.dicts import nested_get
+from .event_common import EventType
+from .metric_events_iterator import MetricEventsIterator, VariantState
+
+
+class MetricDebugImagesIterator(MetricEventsIterator):
+    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+        super().__init__(redis, es, EventType.metrics_image)
+
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        return [{"exists": {"field": "url"}}]
+
+    def _get_variant_state_aggs(self) -> Tuple[dict, Callable[[dict, VariantState], None]]:
+        aggs = {
+            "urls": {
+                "terms": {
+                    "field": "url",
+                    "order": {"max_iter": "desc"},
+                    "size": 1,  # we need only one url from the most recent iteration
+                },
+                "aggs": {
+                    "max_iter": {"max": {"field": "iter"}},
+                    "iters": {
+                        "top_hits": {
+                            "sort": {"iter": {"order": "desc"}},
+                            "size": 2,  # need two last iterations so that we can take
+                            # the second one as invalid
+                            "_source": "iter",
+                        }
+                    },
+                },
+            }
+        }
+
+        def fill_variant_state_data(variant_bucket: dict,  state: VariantState):
+            """If the image urls get recycled then fill the last_invalid_iteration field"""
+            top_iter_url = nested_get(variant_bucket, ("urls", "buckets"))[0]
+            iters = nested_get(top_iter_url, ("iters", "hits", "hits"))
+            if len(iters) > 1:
+                state.last_invalid_iteration = nested_get(iters[1], ("_source", "iter"))
+
+        return aggs, fill_variant_state_data
+
+    def _process_event(self, event: dict) -> dict:
+        return event
+
+    def _get_same_variant_events_order(self) -> dict:
+        return {"url": {"order": "desc"}}

apiserver/bll/event/metric_events_iterator.py

@@ -1,8 +1,9 @@
+import abc
 from concurrent.futures.thread import ThreadPoolExecutor
 from datetime import datetime
 from functools import partial
 from operator import itemgetter
-from typing import Sequence, Tuple, Optional, Mapping, Set
+from typing import Sequence, Tuple, Optional, Mapping, Callable
 
 import attr
 import dpath
@@ -18,8 +19,10 @@ from apiserver.bll.event.event_common import (
     check_empty_data,
     search_company_events,
     EventType,
+    get_metric_variants_condition, get_max_metric_and_variant_counts,
 )
 from apiserver.bll.redis_cache_manager import RedisCacheManager
+from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model.task.metrics import MetricEventStats
 from apiserver.database.model.task.task import Task
@@ -48,25 +51,24 @@ class TaskScrollState(Base):
         self.last_min_iter = self.last_max_iter = None
 
 
-class DebugImageEventsScrollState(Base, JsonSerializableMixin):
+class MetricEventsScrollState(Base, JsonSerializableMixin):
     id: str = StringField(required=True)
     tasks: Sequence[TaskScrollState] = ListField([TaskScrollState])
     warning: str = StringField()
 
 
 @attr.s(auto_attribs=True)
-class DebugImagesResult(object):
+class MetricEventsResult(object):
     metric_events: Sequence[tuple] = []
     next_scroll_id: str = None
 
 
-class DebugImagesIterator:
-    EVENT_TYPE = EventType.metrics_image
-
-    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+class MetricEventsIterator:
+    def __init__(self, redis: StrictRedis, es: Elasticsearch, event_type: EventType):
         self.es = es
+        self.event_type = event_type
         self.cache_manager = RedisCacheManager(
-            state_class=DebugImageEventsScrollState,
+            state_class=MetricEventsScrollState,
             redis=redis,
             expiration_interval=EventSettings.state_expiration_sec,
         )
@@ -74,19 +76,19 @@ class DebugImagesIterator:
     def get_task_events(
         self,
         company_id: str,
-        task_metrics: Mapping[str, Set[str]],
+        task_metrics: Mapping[str, dict],
         iter_count: int,
         navigate_earlier: bool = True,
         refresh: bool = False,
         state_id: str = None,
-    ) -> DebugImagesResult:
-        if check_empty_data(self.es, company_id, self.EVENT_TYPE):
-            return DebugImagesResult()
+    ) -> MetricEventsResult:
+        if check_empty_data(self.es, company_id, self.event_type):
+            return MetricEventsResult()
 
-        def init_state(state_: DebugImageEventsScrollState):
+        def init_state(state_: MetricEventsScrollState):
             state_.tasks = self._init_task_states(company_id, task_metrics)
 
-        def validate_state(state_: DebugImageEventsScrollState):
+        def validate_state(state_: MetricEventsScrollState):
             """
             Validate that the metrics stored in the state are the same
             as requested in the current call.
@@ -98,7 +100,13 @@ class DebugImagesIterator:
         with self.cache_manager.get_or_create_state(
             state_id=state_id, init_state=init_state, validate_state=validate_state
         ) as state:
-            res = DebugImagesResult(next_scroll_id=state.id)
+            res = MetricEventsResult(next_scroll_id=state.id)
+            specific_variants_requested = any(
+                variants
+                for t, metrics in task_metrics.items()
+                if metrics
+                for m, variants in metrics.items()
+            )
             with ThreadPoolExecutor(EventSettings.max_workers) as pool:
                 res.metric_events = list(
                     pool.map(
@@ -107,6 +115,7 @@ class DebugImagesIterator:
                             company_id=company_id,
                             iter_count=iter_count,
                             navigate_earlier=navigate_earlier,
+                            specific_variants_requested=specific_variants_requested,
                         ),
                         state.tasks,
                     )
@@ -117,11 +126,11 @@ class DebugImagesIterator:
     def _reinit_outdated_task_states(
         self,
         company_id,
-        state: DebugImageEventsScrollState,
-        task_metrics: Mapping[str, Set[str]],
+        state: MetricEventsScrollState,
+        task_metrics: Mapping[str, dict],
     ):
         """
-        Determine the metrics for which new debug image events were added
+        Determine the metrics for which new event_type events were added
         since their states were initialized and re-init these states
         """
         tasks = Task.objects(id__in=list(task_metrics), company=company_id).only(
@@ -131,7 +140,7 @@ class DebugImagesIterator:
         def get_last_update_times_for_task_metrics(
             task: Task,
         ) -> Mapping[str, datetime]:
-            """For metrics that reported debug image events get mapping of the metric name to the last update times"""
+            """For metrics that reported event_type events get mapping of the metric name to the last update times"""
             metric_stats: Mapping[str, MetricEventStats] = task.metric_stats
             if not metric_stats:
                 return {}
@@ -139,10 +148,10 @@ class DebugImagesIterator:
             requested_metrics = task_metrics[task.id]
             return {
                 stats.metric: stats.event_stats_by_type[
-                    self.EVENT_TYPE.value
+                    self.event_type.value
                 ].last_update
                 for stats in metric_stats.values()
-                if self.EVENT_TYPE.value in stats.event_stats_by_type
+                if self.event_type.value in stats.event_stats_by_type
                 and (not requested_metrics or stats.metric in requested_metrics)
             }
 
@@ -158,11 +167,11 @@ class DebugImagesIterator:
         task_metrics_to_recalc = {}
         for task, metrics_times in update_times.items():
             old_metric_states = task_metric_states[task]
-            metrics_to_recalc = set(
-                m
+            metrics_to_recalc = {
+                m: task_metrics[task].get(m)
                 for m, t in metrics_times.items()
                 if m not in old_metric_states or old_metric_states[m].timestamp < t
-            )
+            }
             if metrics_to_recalc:
                 task_metrics_to_recalc[task] = metrics_to_recalc
 
@@ -196,7 +205,7 @@ class DebugImagesIterator:
         ]
 
     def _init_task_states(
-        self, company_id: str, task_metrics: Mapping[str, Set[str]]
+        self, company_id: str, task_metrics: Mapping[str, dict]
     ) -> Sequence[TaskScrollState]:
         """
         Returned initialized metric scroll stated for the requested task metrics
@@ -212,25 +221,43 @@ class DebugImagesIterator:
             for task, metric_states in zip(task_metrics, task_metric_states)
         ]
 
+    @abc.abstractmethod
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        pass
+
+    @abc.abstractmethod
+    def _get_variant_state_aggs(self) -> Tuple[dict, Callable[[dict, VariantState], None]]:
+        pass
+
     def _init_metric_states_for_task(
-        self, task_metrics: Tuple[str, Set[str]], company_id: str
+        self, task_metrics: Tuple[str, dict], company_id: str
     ) -> Sequence[MetricState]:
         """
         Return metric scroll states for the task filled with the variant states
-        for the variants that reported any debug images
+        for the variants that reported any event_type events
         """
         task, metrics = task_metrics
-        must = [{"term": {"task": task}}, {"exists": {"field": "url"}}]
+        must = [{"term": {"task": task}}, *self._get_extra_conditions()]
         if metrics:
-            must.append({"terms": {"metric": list(metrics)}})
+            must.append(get_metric_variants_condition(metrics))
+        query = {"bool": {"must": must}}
+
+        search_args = dict(
+            es=self.es, company_id=company_id, event_type=self.event_type
+        )
+        max_metrics, max_variants = get_max_metric_and_variant_counts(
+            query=query, **search_args
+        )
+        max_variants = int(max_variants // 2)
+        variant_state_aggs, fill_variant_state_data = self._get_variant_state_aggs()
         es_req: dict = {
             "size": 0,
-            "query": {"bool": {"must": must}},
+            "query": query,
             "aggs": {
                 "metrics": {
                     "terms": {
                         "field": "metric",
-                        "size": EventSettings.max_metrics_count,
+                        "size": max_metrics,
                         "order": {"_key": "asc"},
                     },
                     "aggs": {
@@ -238,29 +265,10 @@ class DebugImagesIterator:
                         "variants": {
                             "terms": {
                                 "field": "variant",
-                                "size": EventSettings.max_variants_count,
+                                "size": max_variants,
                                 "order": {"_key": "asc"},
                             },
-                            "aggs": {
-                                "urls": {
-                                    "terms": {
-                                        "field": "url",
-                                        "order": {"max_iter": "desc"},
-                                        "size": 1,  # we need only one url from the most recent iteration
-                                    },
-                                    "aggs": {
-                                        "max_iter": {"max": {"field": "iter"}},
-                                        "iters": {
-                                            "top_hits": {
-                                                "sort": {"iter": {"order": "desc"}},
-                                                "size": 2,  # need two last iterations so that we can take
-                                                # the second one as invalid
-                                                "_source": "iter",
-                                            }
-                                        },
-                                    },
-                                }
-                            },
+                            **({"aggs": variant_state_aggs} if variant_state_aggs else {}),
                         },
                     },
                 }
@@ -268,22 +276,18 @@ class DebugImagesIterator:
         }
 
         with translate_errors_context(), TimingContext("es", "_init_metric_states"):
-            es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.EVENT_TYPE, body=es_req,
-            )
+            es_res = search_company_events(body=es_req, **search_args)
         if "aggregations" not in es_res:
             return []
 
         def init_variant_state(variant: dict):
             """
             Return new variant state for the passed variant bucket
-            If the image urls get recycled then fill the last_invalid_iteration field
             """
             state = VariantState(variant=variant["key"])
-            top_iter_url = dpath.get(variant, "urls/buckets")[0]
-            iters = dpath.get(top_iter_url, "iters/hits/hits")
-            if len(iters) > 1:
-                state.last_invalid_iteration = dpath.get(iters[1], "_source/iter")
+            if fill_variant_state_data:
+                fill_variant_state_data(variant, state)
+
             return state
 
         return [
@@ -298,12 +302,21 @@ class DebugImagesIterator:
             for metric in dpath.get(es_res, "aggregations/metrics/buckets")
         ]
 
+    @abc.abstractmethod
+    def _process_event(self, event: dict) -> dict:
+        pass
+
+    @abc.abstractmethod
+    def _get_same_variant_events_order(self) -> dict:
+        pass
+
     def _get_task_metric_events(
         self,
         task_state: TaskScrollState,
         company_id: str,
         iter_count: int,
         navigate_earlier: bool,
+        specific_variants_requested: bool,
     ) -> Tuple:
         """
         Return task metric events grouped by iterations
@@ -319,7 +332,7 @@ class DebugImagesIterator:
         must_conditions = [
             {"term": {"task": task_state.task}},
             {"terms": {"metric": [m.metric for m in task_state.metrics]}},
-            {"exists": {"field": "url"}},
+            *self._get_extra_conditions(),
         ]
 
         range_condition = None
@@ -330,6 +343,8 @@ class DebugImagesIterator:
         if range_condition:
             must_conditions.append({"range": {"iter": range_condition}})
 
+        metrics_count = len(task_state.metrics)
+        max_variants = int(EventSettings.max_es_buckets / (metrics_count * iter_count))
         es_req = {
             "size": 0,
             "query": {"bool": {"must": must_conditions}},
@@ -344,20 +359,20 @@ class DebugImagesIterator:
                         "metrics": {
                             "terms": {
                                 "field": "metric",
-                                "size": EventSettings.max_metrics_count,
+                                "size": metrics_count,
                                 "order": {"_key": "asc"},
                             },
                             "aggs": {
                                 "variants": {
                                     "terms": {
                                         "field": "variant",
-                                        "size": EventSettings.max_variants_count,
+                                        "size": max_variants,
                                         "order": {"_key": "asc"},
                                     },
                                     "aggs": {
                                         "events": {
                                             "top_hits": {
-                                                "sort": {"url": {"order": "desc"}}
+                                                "sort": self._get_same_variant_events_order()
                                             }
                                         }
                                     },
@@ -368,9 +383,12 @@ class DebugImagesIterator:
                 }
             },
         }
-        with translate_errors_context(), TimingContext("es", "get_debug_image_events"):
+        with translate_errors_context(), TimingContext("es", "_get_task_metric_events"):
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.EVENT_TYPE, body=es_req,
+                self.es,
+                company_id=company_id,
+                event_type=self.event_type,
+                body=es_req,
             )
         if "aggregations" not in es_res:
             return task_state.task, []
@@ -380,18 +398,26 @@ class DebugImagesIterator:
             for m in task_state.metrics
             for v in m.variants
         }
+        allow_uninitialized = (
+            False
+            if specific_variants_requested
+            else config.get(
+                "services.events.events_retrieval.debug_images.allow_uninitialized_variants",
+                False,
+            )
+        )
 
         def is_valid_event(event: dict) -> bool:
             key = event.get("metric"), event.get("variant")
             if key not in invalid_iterations:
-                return False
+                return allow_uninitialized
 
             max_invalid = invalid_iterations[key]
             return max_invalid is None or event.get("iter") > max_invalid
 
         def get_iteration_events(it_: dict) -> Sequence:
             return [
-                ev["_source"]
+                self._process_event(ev["_source"])
                 for m in dpath.get(it_, "metrics/buckets")
                 for v in dpath.get(m, "variants/buckets")
                 for ev in dpath.get(v, "events/hits/hits")

apiserver/bll/event/metric_plots_iterator.py

@@ -0,0 +1,25 @@
+from typing import Sequence
+
+from elasticsearch import Elasticsearch
+from redis.client import StrictRedis
+
+from .event_common import EventType, uncompress_plot
+from .metric_events_iterator import MetricEventsIterator
+
+
+class MetricPlotsIterator(MetricEventsIterator):
+    def __init__(self, redis: StrictRedis, es: Elasticsearch):
+        super().__init__(redis, es, EventType.metrics_plot)
+
+    def _get_extra_conditions(self) -> Sequence[dict]:
+        return []
+
+    def _get_variant_state_aggs(self):
+        return None, None
+
+    def _process_event(self, event: dict) -> dict:
+        uncompress_plot(event)
+        return event
+
+    def _get_same_variant_events_order(self) -> dict:
+        return {"timestamp": {"order": "desc"}}

apiserver/bll/event/scalar_key.py

@@ -4,6 +4,8 @@ Module for polymorphism over different types of X axes in scalar aggregations
 from abc import ABC, abstractmethod
 from enum import auto
 
+from typing import Any
+
 from apiserver.utilities import extract_properties_to_lists
 from apiserver.utilities.stringenum import StringEnum
 from apiserver.config_repo import config
@@ -96,6 +98,10 @@ class ScalarKey(ABC):
         """
         return int(iter_data[self.bucket_key_key]), iter_data["avg_val"]["value"]
 
+    def cast_value(self, value: Any) -> Any:
+        """Cast value to appropriate type"""
+        return value
+
 
 class TimestampKey(ScalarKey):
     """
@@ -117,6 +123,9 @@ class TimestampKey(ScalarKey):
             }
         }
 
+    def cast_value(self, value: Any) -> int:
+        return int(value)
+
 
 class IterKey(ScalarKey):
     """
@@ -134,6 +143,9 @@ class IterKey(ScalarKey):
             }
         }
 
+    def cast_value(self, value: Any) -> int:
+        return int(value)
+
 
 class ISOTimeKey(ScalarKey):
     """

apiserver/bll/model/__init__.py

@@ -1,5 +1,5 @@
 from datetime import datetime
-from typing import Callable, Tuple
+from typing import Callable, Tuple, Sequence, Dict
 
 from apiserver.apierrors import errors
 from apiserver.apimodels.models import ModelTaskPublishResponse
@@ -7,6 +7,7 @@ from apiserver.bll.task.utils import deleted_prefix
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task, TaskStatus
+from .metadata import Metadata
 
 
 class ModelBLL:
@@ -127,3 +128,33 @@ class ModelBLL:
         )
 
         return unarchived
+
+    @classmethod
+    def get_model_stats(
+        cls, company: str, model_ids: Sequence[str],
+    ) -> Dict[str, dict]:
+        if not model_ids:
+            return {}
+
+        result = Model.aggregate(
+            [
+                {
+                    "$match": {
+                        "company": {"$in": [None, "", company]},
+                        "_id": {"$in": model_ids},
+                    }
+                },
+                {
+                    "$addFields": {
+                        "labels_count": {"$size": {"$objectToArray": "$labels"}}
+                    }
+                },
+                {
+                    "$project": {"labels_count": 1},
+                },
+            ]
+        )
+        return {
+            r.pop("_id"): r
+            for r in result
+        }

apiserver/bll/model/metadata.py

@@ -0,0 +1,111 @@
+from typing import Sequence, Union, Mapping
+
+from mongoengine import Document
+
+from apiserver.apierrors import errors
+from apiserver.apimodels.metadata import MetadataItem
+from apiserver.database.model.base import GetMixin
+from apiserver.service_repo import APICall
+from apiserver.utilities.parameter_key_escaper import (
+    ParameterKeyEscaper,
+    mongoengine_safe,
+)
+from apiserver.config_repo import config
+from apiserver.timing_context import TimingContext
+
+log = config.logger(__file__)
+
+
+class Metadata:
+    @staticmethod
+    def metadata_from_api(
+        api_data: Union[Mapping[str, MetadataItem], Sequence[MetadataItem]]
+    ) -> dict:
+        if not api_data:
+            return {}
+
+        if isinstance(api_data, dict):
+            return {
+                ParameterKeyEscaper.escape(k): v.to_struct()
+                for k, v in api_data.items()
+            }
+
+        return {
+            ParameterKeyEscaper.escape(item.key): item.to_struct() for item in api_data
+        }
+
+    @classmethod
+    def edit_metadata(
+        cls,
+        obj: Document,
+        items: Sequence[MetadataItem],
+        replace_metadata: bool,
+        **more_updates,
+    ) -> int:
+        with TimingContext("mongo", "edit_metadata"):
+            update_cmds = dict()
+            metadata = cls.metadata_from_api(items)
+            if replace_metadata:
+                update_cmds["set__metadata"] = metadata
+            else:
+                for key, value in metadata.items():
+                    update_cmds[f"set__metadata__{mongoengine_safe(key)}"] = value
+
+            return obj.update(**update_cmds, **more_updates)
+
+    @classmethod
+    def delete_metadata(cls, obj: Document, keys: Sequence[str], **more_updates) -> int:
+        with TimingContext("mongo", "delete_metadata"):
+            return obj.update(
+                **{
+                    f"unset__metadata__{ParameterKeyEscaper.escape(key)}": 1
+                    for key in set(keys)
+                },
+                **more_updates,
+            )
+
+    @staticmethod
+    def _process_path(path: str):
+        """
+        Frontend does a partial escaping on the path so the all '.' in key names are escaped
+        Need to unescape and apply a full mongo escaping
+        """
+        parts = path.split(".")
+        if len(parts) < 2 or len(parts) > 3:
+            raise errors.bad_request.ValidationError("invalid field", path=path)
+        return ".".join(
+            ParameterKeyEscaper.escape(ParameterKeyEscaper.unescape(p)) for p in parts
+        )
+
+    @classmethod
+    def escape_paths(cls, paths: Sequence[str]) -> Sequence[str]:
+        for prefix in (
+            "metadata.",
+            "-metadata.",
+        ):
+            paths = [
+                cls._process_path(path) if path.startswith(prefix) else path
+                for path in paths
+            ]
+        return paths
+
+    @classmethod
+    def escape_query_parameters(cls, call: APICall) -> dict:
+        if not call.data:
+            return call.data
+
+        keys = list(call.data)
+        call_data = {
+            safe_key: call.data[key]
+            for key, safe_key in zip(keys, Metadata.escape_paths(keys))
+        }
+
+        projection = GetMixin.get_projection(call_data)
+        if projection:
+            GetMixin.set_projection(call_data, Metadata.escape_paths(projection))
+
+        ordering = GetMixin.get_ordering(call_data)
+        if ordering:
+            GetMixin.set_ordering(call_data, Metadata.escape_paths(ordering))
+
+        return call_data

apiserver/bll/organization/tags_cache.py

@@ -6,6 +6,7 @@ from redis import Redis
 
 from apiserver.config_repo import config
 from apiserver.bll.project import project_ids_with_children
+from apiserver.database.model import EntityVisibility
 from apiserver.database.model.base import GetMixin
 from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task
@@ -42,6 +43,8 @@ class _TagsCache:
                     query &= GetMixin.get_list_field_query(name, vals)
         if project:
             query &= Q(project__in=project_ids_with_children([project]))
+        else:
+            query &= Q(system_tags__nin=[EntityVisibility.hidden.value])
 
         return self.db_cls.objects(query).distinct(field)
 

apiserver/bll/project/__init__.py

@@ -1,2 +1,3 @@
 from .project_bll import ProjectBLL
+from .project_queries import ProjectQueries
 from .sub_projects import _ids_with_children as project_ids_with_children

apiserver/bll/project/project_bll.py

@@ -1,6 +1,6 @@
 import itertools
 from collections import defaultdict
-from datetime import datetime
+from datetime import datetime, timedelta
 from functools import reduce
 from itertools import groupby
 from operator import itemgetter
@@ -14,14 +14,17 @@ from typing import (
     TypeVar,
     Callable,
     Mapping,
+    Any,
 )
 
+from boltons.iterutils import partition
 from mongoengine import Q, Document
 
 from apiserver import database
 from apiserver.apierrors import errors
 from apiserver.config_repo import config
 from apiserver.database.model import EntityVisibility, AttributedDocument
+from apiserver.database.model.base import GetMixin
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
 from apiserver.database.model.task.task import Task, TaskStatus, external_task_types
@@ -57,19 +60,27 @@ class ProjectBLL:
         with TimingContext("mongo", "move_project"):
             if source_id == destination_id:
                 raise errors.bad_request.ProjectSourceAndDestinationAreTheSame(
-                    parent=source_id
+                    source=source_id
                 )
             source = Project.get(company, source_id)
-            destination = Project.get(company, destination_id)
+            if destination_id:
+                destination = Project.get(company, destination_id)
+                if source_id in destination.path:
+                    raise errors.bad_request.ProjectCannotBeMergedIntoItsChild(
+                        source=source_id, destination=destination_id
+                    )
+            else:
+                destination = None
 
             children = _get_sub_projects(
                 [source.id], _only=("id", "name", "parent", "path")
             )[source.id]
-            cls.validate_projects_depth(
-                projects=children,
-                old_parent_depth=len(source.path) + 1,
-                new_parent_depth=len(destination.path) + 1,
-            )
+            if destination:
+                cls.validate_projects_depth(
+                    projects=children,
+                    old_parent_depth=len(source.path) + 1,
+                    new_parent_depth=len(destination.path) + 1,
+                )
 
             moved_entities = 0
             for entity_type in (Task, Model):
@@ -140,7 +151,12 @@ class ProjectBLL:
                 raise errors.bad_request.ProjectSourceAndDestinationAreTheSame(
                     location=new_parent.name if new_parent else ""
                 )
-
+            if new_parent and (
+                project_id == new_parent.id or project_id in new_parent.path
+            ):
+                raise errors.bad_request.ProjectCannotBeMovedUnderItself(
+                    project=project_id, parent=new_parent.id
+                )
             moved = _reposition_project_with_children(
                 project, children=children, parent=new_parent
             )
@@ -167,6 +183,7 @@ class ProjectBLL:
                 if new_location != old_location:
                     raise errors.bad_request.CannotUpdateProjectLocation(name=new_name)
                 fields["name"] = new_name
+                fields["basename"] = new_name.split("/")[-1]
 
             fields["last_update"] = datetime.utcnow()
             updated = project.update(upsert=False, **fields)
@@ -193,6 +210,7 @@ class ProjectBLL:
         tags: Sequence[str] = None,
         system_tags: Sequence[str] = None,
         default_output_destination: str = None,
+        parent_creation_params: dict = None,
     ) -> str:
         """
         Create a new project.
@@ -208,6 +226,7 @@ class ProjectBLL:
             user=user,
             company=company,
             name=name,
+            basename=name.split("/")[-1],
             description=description,
             tags=tags,
             system_tags=system_tags,
@@ -215,7 +234,12 @@ class ProjectBLL:
             created=now,
             last_update=now,
         )
-        parent = _ensure_project(company=company, user=user, name=location)
+        parent = _ensure_project(
+            company=company,
+            user=user,
+            name=location,
+            creation_params=parent_creation_params,
+        )
         _save_under_parent(project=project, parent=parent)
         if parent:
             parent.update(last_update=now)
@@ -233,13 +257,14 @@ class ProjectBLL:
         tags: Sequence[str] = None,
         system_tags: Sequence[str] = None,
         default_output_destination: str = None,
+        parent_creation_params: dict = None,
     ) -> str:
         """
         Find a project named `project_name` or create a new one.
         Returns project ID
         """
         if not project_id and not project_name:
-            raise ValueError("project id or name required")
+            raise errors.bad_request.ValidationError("project id or name required")
 
         if project_id:
             project = Project.objects(company=company, id=project_id).only("id").first()
@@ -260,6 +285,7 @@ class ProjectBLL:
             tags=tags,
             system_tags=system_tags,
             default_output_destination=default_output_destination,
+            parent_creation_params=parent_creation_params,
         )
 
     @classmethod
@@ -295,6 +321,7 @@ class ProjectBLL:
             return project
 
     archived_tasks_cond = {"$in": [EntityVisibility.archived.value, "$system_tags"]}
+    visibility_states = [EntityVisibility.archived, EntityVisibility.active]
 
     @classmethod
     def make_projects_get_all_pipelines(
@@ -302,6 +329,8 @@ class ProjectBLL:
         company_id: str,
         project_ids: Sequence[str],
         specific_state: Optional[EntityVisibility] = None,
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
     ) -> Tuple[Sequence, Sequence]:
         archived = EntityVisibility.archived.value
 
@@ -325,10 +354,12 @@ class ProjectBLL:
         status_count_pipeline = [
             # count tasks per project per status
             {
-                "$match": {
-                    "company": {"$in": [None, "", company_id]},
-                    "project": {"$in": project_ids},
-                }
+                "$match": cls.get_match_conditions(
+                    company=company_id,
+                    project_ids=project_ids,
+                    filter_=filter_,
+                    users=users,
+                )
             },
             ensure_valid_fields(),
             {
@@ -356,6 +387,37 @@ class ProjectBLL:
             },
         ]
 
+        def completed_after_subquery(additional_cond, time_thresh: datetime):
+            return {
+                # the sum of
+                "$sum": {
+                    # for each task
+                    "$cond": {
+                        # if completed after the time_thresh
+                        "if": {
+                            "$and": [
+                                "$completed",
+                                {"$gt": ["$completed", time_thresh]},
+                                additional_cond,
+                            ]
+                        },
+                        "then": 1,
+                        "else": 0,
+                    }
+                }
+            }
+
+        def max_started_subquery(condition):
+            return {
+                "$max": {
+                    "$cond": {
+                        "if": condition,
+                        "then": "$started",
+                        "else": datetime.min,
+                    }
+                }
+            }
+
         def runtime_subquery(additional_cond):
             return {
                 # the sum of
@@ -386,25 +448,54 @@ class ProjectBLL:
             }
 
         group_step = {"_id": "$project"}
-
-        for state in EntityVisibility:
+        time_thresh = datetime.utcnow() - timedelta(hours=24)
+        for state in cls.visibility_states:
             if specific_state and state != specific_state:
                 continue
-            if state == EntityVisibility.active:
-                group_step[state.value] = runtime_subquery(
-                    {"$not": cls.archived_tasks_cond}
+            cond = (
+                cls.archived_tasks_cond
+                if state == EntityVisibility.archived
+                else {"$not": cls.archived_tasks_cond}
+            )
+            group_step[state.value] = runtime_subquery(cond)
+            group_step[f"{state.value}_recently_completed"] = completed_after_subquery(
+                cond, time_thresh=time_thresh
+            )
+            group_step[f"{state.value}_max_task_started"] = max_started_subquery(cond)
+
+        def add_state_to_filter(f: Mapping[str, Any]) -> Mapping[str, Any]:
+            if not specific_state:
+                return f
+
+            f = f or {}
+            new_f = {k: v for k, v in f.items() if k != "system_tags"}
+            system_tags = [
+                tag
+                for tag in f.get("system_tags", [])
+                if tag
+                not in (
+                    EntityVisibility.archived.value,
+                    f"-{EntityVisibility.archived.value}",
                 )
-            elif state == EntityVisibility.archived:
-                group_step[state.value] = runtime_subquery(cls.archived_tasks_cond)
+            ]
+
+            if specific_state == EntityVisibility.archived:
+                system_tags.append(EntityVisibility.archived.value)
+            else:
+                system_tags.append(f"-{EntityVisibility.archived.value}")
+            new_f["system_tags"] = system_tags
+
+            return new_f
 
         runtime_pipeline = [
             # only count run time for these types of tasks
             {
-                "$match": {
-                    "company": {"$in": [None, "", company_id]},
-                    "type": {"$in": ["training", "testing", "annotation"]},
-                    "project": {"$in": project_ids},
-                }
+                "$match": cls.get_match_conditions(
+                    company=company_id,
+                    project_ids=project_ids,
+                    filter_=add_state_to_filter(filter_),
+                    users=users,
+                )
             },
             ensure_valid_fields(),
             {
@@ -439,17 +530,65 @@ class ProjectBLL:
             aggregated[pid] = reduce(func, relevant_data)
         return aggregated
 
+    @classmethod
+    def get_dataset_stats(
+        cls, company: str, project_ids: Sequence[str], users: Sequence[str] = None,
+    ) -> Dict[str, dict]:
+        if not project_ids:
+            return {}
+
+        task_runtime_pipeline = [
+            {
+                "$match": {
+                    **cls.get_match_conditions(
+                        company=company,
+                        project_ids=project_ids,
+                        users=users,
+                        filter_={
+                            "system_tags": [f"-{EntityVisibility.archived.value}"]
+                        },
+                    ),
+                    "runtime": {"$exists": True, "$gt": {}},
+                }
+            },
+            {"$project": {"project": 1, "runtime": 1, "last_update": 1}},
+            {"$sort": {"project": 1, "last_update": 1}},
+            {"$group": {"_id": "$project", "runtime": {"$last": "$runtime"}}},
+        ]
+
+        return {
+            r["_id"]: {
+                "file_count": r["runtime"].get("ds_file_count", 0),
+                "total_size": r["runtime"].get("ds_total_size", 0),
+            }
+            for r in Task.aggregate(task_runtime_pipeline)
+        }
+
     @classmethod
     def get_project_stats(
         cls,
         company: str,
         project_ids: Sequence[str],
         specific_state: Optional[EntityVisibility] = None,
+        include_children: bool = True,
+        search_hidden: bool = False,
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+        user_active_project_ids: Sequence[str] = None,
     ) -> Tuple[Dict[str, dict], Dict[str, dict]]:
         if not project_ids:
             return {}, {}
 
-        child_projects = _get_sub_projects(project_ids, _only=("id", "name"))
+        child_projects = (
+            _get_sub_projects(
+                project_ids,
+                _only=("id", "name"),
+                search_hidden=search_hidden,
+                allowed_ids=user_active_project_ids,
+            )
+            if include_children
+            else {}
+        )
         project_ids_with_children = set(project_ids) | {
             c.id for c in itertools.chain.from_iterable(child_projects.values())
         }
@@ -457,6 +596,8 @@ class ProjectBLL:
             company,
             project_ids=list(project_ids_with_children),
             specific_state=specific_state,
+            filter_=filter_,
+            users=users,
         )
 
         default_counts = dict.fromkeys(get_options(TaskStatus), 0)
@@ -483,8 +624,8 @@ class ProjectBLL:
         ) -> Dict[str, dict]:
             return {
                 section: {
-                    status: nested_get(a, (section, status), 0)
-                    + nested_get(b, (section, status), 0)
+                    status: nested_get(a, (section, status), default=0)
+                    + nested_get(b, (section, status), default=0)
                     for status in set(a.get(section, {})) | set(b.get(section, {}))
                 }
                 for section in set(a) | set(b)
@@ -507,6 +648,8 @@ class ProjectBLL:
         ) -> Dict[str, dict]:
             return {
                 section: a.get(section, 0) + b.get(section, 0)
+                if not section.endswith("max_task_started")
+                else max(a.get(section) or datetime.min, b.get(section) or datetime.min)
                 for section in set(a) | set(b)
             }
 
@@ -518,15 +661,30 @@ class ProjectBLL:
         )
 
         def get_status_counts(project_id, section):
+            project_runtime = runtime.get(project_id, {})
+            project_section_statuses = nested_get(
+                status_count, (project_id, section), default=default_counts
+            )
+
+            def get_time_or_none(value):
+                return value if value != datetime.min else None
+
             return {
-                "total_runtime": nested_get(runtime, (project_id, section), 0),
-                "status_count": nested_get(
-                    status_count, (project_id, section), default_counts
+                "status_count": project_section_statuses,
+                "total_tasks": sum(project_section_statuses.values()),
+                "total_runtime": project_runtime.get(section, 0),
+                "completed_tasks_24h": project_runtime.get(
+                    f"{section}_recently_completed", 0
+                ),
+                "last_task_run": get_time_or_none(
+                    project_runtime.get(f"{section}_max_task_started", datetime.min)
                 ),
             }
 
         report_for_states = [
-            s for s in EntityVisibility if not specific_state or specific_state == s
+            s
+            for s in cls.visibility_states
+            if not specific_state or specific_state == s
         ]
 
         stats = {
@@ -554,7 +712,7 @@ class ProjectBLL:
         user_ids: Optional[Sequence[str]] = None,
     ) -> Set[str]:
         """
-        Get the set of user ids that created tasks/models/dataviews in the given projects
+        Get the set of user ids that created tasks/models in the given projects
         If project_ids is empty then all projects are examined
         If user_ids are passed then only subset of these users is returned
         """
@@ -575,6 +733,30 @@ class ProjectBLL:
 
             return res
 
+    @classmethod
+    def get_project_tags(
+        cls,
+        company_id: str,
+        include_system: bool,
+        projects: Sequence[str] = None,
+        filter_: Dict[str, Sequence[str]] = None,
+    ) -> Tuple[Sequence[str], Sequence[str]]:
+        with TimingContext("mongo", "get_tags_from_db"):
+            query = Q(company=company_id)
+            if filter_:
+                for name, vals in filter_.items():
+                    if vals:
+                        query &= GetMixin.get_list_field_query(name, vals)
+
+            if projects:
+                query &= Q(id__in=_ids_with_children(projects))
+
+            tags = Project.objects(query).distinct("tags")
+            system_tags = (
+                Project.objects(query).distinct("system_tags") if include_system else []
+            )
+            return tags, system_tags
+
     @classmethod
     def get_projects_with_active_user(
         cls,
@@ -582,7 +764,7 @@ class ProjectBLL:
         users: Sequence[str],
         project_ids: Optional[Sequence[str]] = None,
         allow_public: bool = True,
-    ) -> Sequence[str]:
+    ) -> Tuple[Sequence[str], Sequence[str]]:
         """
         Get the projects ids where user created any tasks including all the parents of these projects
         If project ids are specified then filter the results by these project ids
@@ -606,13 +788,16 @@ class ProjectBLL:
 
         res = list(res)
         if not res:
-            return res
+            return res, res
 
-        ids_with_parents = _ids_with_parents(res)
-        if project_ids:
-            return [pid for pid in ids_with_parents if pid in project_ids]
+        user_active_project_ids = _ids_with_parents(res)
+        filtered_ids = (
+            list(set(user_active_project_ids) & set(project_ids))
+            if project_ids
+            else list(user_active_project_ids)
+        )
 
-        return ids_with_parents
+        return filtered_ids, user_active_project_ids
 
     @classmethod
     def get_task_parents(
@@ -627,10 +812,14 @@ class ProjectBLL:
         If projects is None or empty then get parents for all the company tasks
         """
         query = Q(company=company_id)
+
         if projects:
             if include_subprojects:
                 projects = _ids_with_children(projects)
             query &= Q(project__in=projects)
+        else:
+            query &= Q(system_tags__nin=[EntityVisibility.hidden.value])
+
         if state == EntityVisibility.archived:
             query &= Q(system_tags__in=[EntityVisibility.archived.value])
         elif state == EntityVisibility.active:
@@ -658,6 +847,8 @@ class ProjectBLL:
         if project_ids:
             project_ids = _ids_with_children(project_ids)
             query &= Q(project__in=project_ids)
+        else:
+            query &= Q(system_tags__nin=[EntityVisibility.hidden.value])
         res = Task.objects(query).distinct(field="type")
         return set(res).intersection(external_task_types)
 
@@ -673,11 +864,51 @@ class ProjectBLL:
             query &= Q(project__in=project_ids)
         return Model.objects(query).distinct(field="framework")
 
+    @staticmethod
+    def get_match_conditions(
+        company: str,
+        project_ids: Sequence[str],
+        filter_: Mapping[str, Any],
+        users: Sequence[str],
+    ):
+        conditions = {
+            "company": {"$in": [None, "", company]},
+            "project": {"$in": project_ids},
+        }
+        if users:
+            conditions["user"] = {"$in": users}
+
+        if not filter_:
+            return conditions
+
+        for field, field_filter in filter_.items():
+            if not (
+                field_filter
+                and isinstance(field_filter, list)
+                and all(isinstance(t, str) for t in field_filter)
+            ):
+                raise errors.bad_request.ValidationError(
+                    f"List of strings expected for the field: {field}"
+                )
+            exclude, include = partition(field_filter, lambda x: x.startswith("-"))
+            conditions[field] = {
+                **({"$in": include} if include else {}),
+                **({"$nin": [e[1:] for e in exclude]} if exclude else {}),
+            }
+
+        return conditions
+
     @classmethod
-    def calc_own_contents(cls, company: str, project_ids: Sequence[str]) -> Dict[str, dict]:
+    def calc_own_contents(
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+    ) -> Dict[str, dict]:
         """
-        Returns the amount of task/dataviews/models per requested project
-        Use separate aggregation calls on Task/Dataview/Model instead of lookup
+        Returns the amount of task/models per requested project
+        Use separate aggregation calls on Task/Model instead of lookup
         aggregation on projects in order not to hit memory limits on large tasks
         """
         if not project_ids:
@@ -685,35 +916,24 @@ class ProjectBLL:
 
         pipeline = [
             {
-                "$match": {
-                    "company": {"$in": [None, "", company]},
-                    "project": {"$in": project_ids},
-                }
+                "$match": cls.get_match_conditions(
+                    company=company,
+                    project_ids=project_ids,
+                    filter_=filter_,
+                    users=users,
+                )
             },
-            {
-                "$project": {"project": 1}
-            },
-            {
-                "$group": {
-                    "_id": "$project",
-                    "count": {"$sum": 1},
-                }
-            }
+            {"$project": {"project": 1}},
+            {"$group": {"_id": "$project", "count": {"$sum": 1}}},
         ]
 
         def get_agrregate_res(cls_: Type[AttributedDocument]) -> dict:
-            return {
-                data["_id"]: data["count"]
-                for data in cls_.aggregate(pipeline)
-            }
+            return {data["_id"]: data["count"] for data in cls_.aggregate(pipeline)}
 
         with TimingContext("mongo", "get_security_groups"):
             tasks = get_agrregate_res(Task)
             models = get_agrregate_res(Model)
             return {
-                pid: {
-                    "own_tasks": tasks.get(pid, 0),
-                    "own_models": models.get(pid, 0),
-                }
+                pid: {"own_tasks": tasks.get(pid, 0), "own_models": models.get(pid, 0)}
                 for pid in project_ids
             }

apiserver/bll/project/project_cleanup.py

@@ -13,7 +13,7 @@ from apiserver.config_repo import config
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
-from apiserver.database.model.task.task import Task, ArtifactModes
+from apiserver.database.model.task.task import Task, ArtifactModes, TaskType
 from apiserver.timing_context import TimingContext
 from .sub_projects import _ids_with_children
 
@@ -30,27 +30,62 @@ class DeleteProjectResult:
     urls: TaskUrls = None
 
 
+def validate_project_delete(company: str, project_id: str):
+    project = Project.get_for_writing(
+        company=company, id=project_id, _only=("id", "path", "system_tags")
+    )
+    if not project:
+        raise errors.bad_request.InvalidProjectId(id=project_id)
+    is_pipeline = "pipeline" in (project.system_tags or [])
+    project_ids = _ids_with_children([project_id])
+    ret = {}
+    for cls in (Task, Model):
+        ret[f"{cls.__name__.lower()}s"] = cls.objects(project__in=project_ids).count()
+    for cls in (Task, Model):
+        query = dict(
+            project__in=project_ids, system_tags__nin=[EntityVisibility.archived.value]
+        )
+        name = f"non_archived_{cls.__name__.lower()}s"
+        if not is_pipeline:
+            ret[name] = cls.objects(**query).count()
+        else:
+            ret[name] = (
+                cls.objects(**query, type=TaskType.controller).count()
+                if cls == Task
+                else 0
+            )
+
+    return ret
+
+
 def delete_project(
     company: str, project_id: str, force: bool, delete_contents: bool
 ) -> Tuple[DeleteProjectResult, Set[str]]:
     project = Project.get_for_writing(
-        company=company, id=project_id, _only=("id", "path")
+        company=company, id=project_id, _only=("id", "path", "system_tags")
     )
     if not project:
         raise errors.bad_request.InvalidProjectId(id=project_id)
-
+    is_pipeline = "pipeline" in (project.system_tags or [])
     project_ids = _ids_with_children([project_id])
     if not force:
-        for cls, error in (
-            (Task, errors.bad_request.ProjectHasTasks),
-            (Model, errors.bad_request.ProjectHasModels),
-        ):
-            non_archived = cls.objects(
-                project__in=project_ids,
-                system_tags__nin=[EntityVisibility.archived.value],
-            ).only("id")
+        query = dict(
+            project__in=project_ids, system_tags__nin=[EntityVisibility.archived.value]
+        )
+        if not is_pipeline:
+            for cls, error in (
+                (Task, errors.bad_request.ProjectHasTasks),
+                (Model, errors.bad_request.ProjectHasModels),
+            ):
+                non_archived = cls.objects(**query).only("id")
+                if non_archived:
+                    raise error("use force=true to delete", id=project_id)
+        else:
+            non_archived = Task.objects(**query, type=TaskType.controller).only("id")
             if non_archived:
-                raise error("use force=true to delete", id=project_id)
+                raise errors.bad_request.ProjectHasTasks(
+                    "please archive all the runs inside the project", id=project_id
+                )
 
     if not delete_contents:
         with TimingContext("mongo", "update_children"):

apiserver/bll/project/project_queries.py

@@ -0,0 +1,370 @@
+import json
+from collections import OrderedDict
+from datetime import datetime
+from typing import (
+    Sequence,
+    Optional,
+    Tuple,
+)
+
+from redis import StrictRedis
+
+from apiserver.config_repo import config
+from apiserver.database.model.model import Model
+from apiserver.database.model.task.task import Task
+from apiserver.redis_manager import redman
+from apiserver.utilities.dicts import nested_get
+from apiserver.utilities.parameter_key_escaper import ParameterKeyEscaper
+from .sub_projects import _ids_with_children
+
+log = config.logger(__file__)
+
+
+class ProjectQueries:
+    def __init__(self, redis=None):
+        self.redis: StrictRedis = redis or redman.connection("apiserver")
+
+    @staticmethod
+    def _get_project_constraint(
+        project_ids: Sequence[str], include_subprojects: bool
+    ) -> dict:
+        """
+        If passed projects is None means top level projects
+        If passed projects is empty means no project filtering
+        """
+        if include_subprojects:
+            if not project_ids:
+                return {}
+            project_ids = _ids_with_children(project_ids)
+
+        if project_ids is None:
+            project_ids = [None]
+        if not project_ids:
+            return {}
+
+        return {"project": {"$in": project_ids}}
+
+    @staticmethod
+    def _get_company_constraint(company_id: str, allow_public: bool = True) -> dict:
+        if allow_public:
+            return {"company": {"$in": [None, "", company_id]}}
+
+        return {"company": company_id}
+
+    @classmethod
+    def get_aggregated_project_parameters(
+        cls,
+        company_id,
+        project_ids: Sequence[str],
+        include_subprojects: bool,
+        page: int = 0,
+        page_size: int = 500,
+    ) -> Tuple[int, int, Sequence[dict]]:
+        page = max(0, page)
+        page_size = max(1, page_size)
+        pipeline = [
+            {
+                "$match": {
+                    **cls._get_company_constraint(company_id),
+                    **cls._get_project_constraint(project_ids, include_subprojects),
+                    "hyperparams": {"$exists": True, "$gt": {}},
+                }
+            },
+            {"$project": {"sections": {"$objectToArray": "$hyperparams"}}},
+            {"$unwind": "$sections"},
+            {
+                "$project": {
+                    "section": "$sections.k",
+                    "names": {"$objectToArray": "$sections.v"},
+                }
+            },
+            {"$unwind": "$names"},
+            {"$group": {"_id": {"section": "$section", "name": "$names.k"}}},
+            {"$sort": OrderedDict({"_id.section": 1, "_id.name": 1})},
+            {"$skip": page * page_size},
+            {"$limit": page_size},
+            {
+                "$group": {
+                    "_id": 1,
+                    "total": {"$sum": 1},
+                    "results": {"$push": "$$ROOT"},
+                }
+            },
+        ]
+
+        result = next(Task.aggregate(pipeline), None)
+
+        total = 0
+        remaining = 0
+        results = []
+
+        if result:
+            total = int(result.get("total", -1))
+            results = [
+                {
+                    "section": ParameterKeyEscaper.unescape(
+                        nested_get(r, ("_id", "section"))
+                    ),
+                    "name": ParameterKeyEscaper.unescape(
+                        nested_get(r, ("_id", "name"))
+                    ),
+                }
+                for r in result.get("results", [])
+            ]
+            remaining = max(0, total - (len(results) + page * page_size))
+
+        return total, remaining, results
+
+    ParamValues = Tuple[int, Sequence[str]]
+
+    def _get_cached_param_values(
+        self, key: str, last_update: datetime, allowed_delta_sec=0
+    ) -> Optional[ParamValues]:
+        try:
+            cached = self.redis.get(key)
+            if not cached:
+                return
+
+            data = json.loads(cached)
+            cached_last_update = datetime.fromtimestamp(data["last_update"])
+            if (last_update - cached_last_update).total_seconds() <= allowed_delta_sec:
+                return data["total"], data["values"]
+        except Exception as ex:
+            log.error(f"Error retrieving params cached values: {str(ex)}")
+
+    def get_task_hyperparam_distinct_values(
+        self,
+        company_id: str,
+        project_ids: Sequence[str],
+        section: str,
+        name: str,
+        include_subprojects: bool,
+        allow_public: bool = True,
+    ) -> ParamValues:
+        company_constraint = self._get_company_constraint(company_id, allow_public)
+        project_constraint = self._get_project_constraint(
+            project_ids, include_subprojects
+        )
+        key_path = f"hyperparams.{ParameterKeyEscaper.escape(section)}.{ParameterKeyEscaper.escape(name)}"
+        last_updated_task = (
+            Task.objects(
+                **company_constraint,
+                **project_constraint,
+                **{f"{key_path.replace('.', '__')}__exists": True},
+            )
+            .only("last_update")
+            .order_by("-last_update")
+            .limit(1)
+            .first()
+        )
+        if not last_updated_task:
+            return 0, []
+
+        redis_key = f"hyperparam_values_{company_id}_{'_'.join(project_ids)}_{section}_{name}_{allow_public}"
+        last_update = last_updated_task.last_update or datetime.utcnow()
+        cached_res = self._get_cached_param_values(
+            key=redis_key,
+            last_update=last_update,
+            allowed_delta_sec=config.get(
+                "services.tasks.hyperparam_values.cache_allowed_outdate_sec", 60
+            ),
+        )
+        if cached_res:
+            return cached_res
+
+        max_values = config.get("services.tasks.hyperparam_values.max_count", 100)
+        pipeline = [
+            {
+                "$match": {
+                    **company_constraint,
+                    **project_constraint,
+                    key_path: {"$exists": True},
+                }
+            },
+            {"$project": {"value": f"${key_path}.value"}},
+            {"$group": {"_id": "$value"}},
+            {"$sort": {"_id": 1}},
+            {"$limit": max_values},
+            {
+                "$group": {
+                    "_id": 1,
+                    "total": {"$sum": 1},
+                    "results": {"$push": "$$ROOT._id"},
+                }
+            },
+        ]
+
+        result = next(Task.aggregate(pipeline, collation=Task._numeric_locale), None)
+        if not result:
+            return 0, []
+
+        total = int(result.get("total", 0))
+        values = result.get("results", [])
+
+        ttl = config.get("services.tasks.hyperparam_values.cache_ttl_sec", 86400)
+        cached = dict(last_update=last_update.timestamp(), total=total, values=values)
+        self.redis.setex(redis_key, ttl, json.dumps(cached))
+
+        return total, values
+
+    @classmethod
+    def get_unique_metric_variants(
+        cls, company_id, project_ids: Sequence[str], include_subprojects: bool
+    ):
+        pipeline = [
+            {
+                "$match": {
+                    **cls._get_company_constraint(company_id),
+                    **cls._get_project_constraint(project_ids, include_subprojects),
+                }
+            },
+            {"$project": {"metrics": {"$objectToArray": "$last_metrics"}}},
+            {"$unwind": "$metrics"},
+            {
+                "$project": {
+                    "metric": "$metrics.k",
+                    "variants": {"$objectToArray": "$metrics.v"},
+                }
+            },
+            {"$unwind": "$variants"},
+            {
+                "$group": {
+                    "_id": {
+                        "metric": "$variants.v.metric",
+                        "variant": "$variants.v.variant",
+                    },
+                    "metrics": {
+                        "$addToSet": {
+                            "metric": "$variants.v.metric",
+                            "metric_hash": "$metric",
+                            "variant": "$variants.v.variant",
+                            "variant_hash": "$variants.k",
+                        }
+                    },
+                }
+            },
+            {"$sort": OrderedDict({"_id.metric": 1, "_id.variant": 1})},
+        ]
+
+        result = Task.aggregate(pipeline)
+        return [r["metrics"][0] for r in result]
+
+    @classmethod
+    def get_model_metadata_keys(
+        cls,
+        company_id,
+        project_ids: Sequence[str],
+        include_subprojects: bool,
+        page: int = 0,
+        page_size: int = 500,
+    ) -> Tuple[int, int, Sequence[dict]]:
+        page = max(0, page)
+        page_size = max(1, page_size)
+        pipeline = [
+            {
+                "$match": {
+                    **cls._get_company_constraint(company_id),
+                    **cls._get_project_constraint(project_ids, include_subprojects),
+                    "metadata": {"$exists": True, "$gt": {}},
+                }
+            },
+            {"$project": {"metadata": {"$objectToArray": "$metadata"}}},
+            {"$unwind": "$metadata"},
+            {"$group": {"_id": "$metadata.k"}},
+            {"$sort": {"_id": 1}},
+            {"$skip": page * page_size},
+            {"$limit": page_size},
+            {
+                "$group": {
+                    "_id": 1,
+                    "total": {"$sum": 1},
+                    "results": {"$push": "$$ROOT"},
+                }
+            },
+        ]
+
+        result = next(Model.aggregate(pipeline), None)
+
+        total = 0
+        remaining = 0
+        results = []
+
+        if result:
+            total = int(result.get("total", -1))
+            results = [
+                ParameterKeyEscaper.unescape(r.get("_id"))
+                for r in result.get("results", [])
+            ]
+            remaining = max(0, total - (len(results) + page * page_size))
+
+        return total, remaining, results
+
+    def get_model_metadata_distinct_values(
+        self,
+        company_id: str,
+        project_ids: Sequence[str],
+        key: str,
+        include_subprojects: bool,
+        allow_public: bool = True,
+    ) -> ParamValues:
+        company_constraint = self._get_company_constraint(company_id, allow_public)
+        project_constraint = self._get_project_constraint(
+            project_ids, include_subprojects
+        )
+        key_path = f"metadata.{ParameterKeyEscaper.escape(key)}"
+        last_updated_model = (
+            Model.objects(
+                **company_constraint,
+                **project_constraint,
+                **{f"{key_path.replace('.', '__')}__exists": True},
+            )
+            .only("last_update")
+            .order_by("-last_update")
+            .limit(1)
+            .first()
+        )
+        if not last_updated_model:
+            return 0, []
+
+        redis_key = f"modelmetadata_values_{company_id}_{'_'.join(project_ids)}_{key}_{allow_public}"
+        last_update = last_updated_model.last_update or datetime.utcnow()
+        cached_res = self._get_cached_param_values(
+            key=redis_key, last_update=last_update
+        )
+        if cached_res:
+            return cached_res
+
+        max_values = config.get("services.models.metadata_values.max_count", 100)
+        pipeline = [
+            {
+                "$match": {
+                    **company_constraint,
+                    **project_constraint,
+                    key_path: {"$exists": True},
+                }
+            },
+            {"$project": {"value": f"${key_path}.value"}},
+            {"$group": {"_id": "$value"}},
+            {"$sort": {"_id": 1}},
+            {"$limit": max_values},
+            {
+                "$group": {
+                    "_id": 1,
+                    "total": {"$sum": 1},
+                    "results": {"$push": "$$ROOT._id"},
+                }
+            },
+        ]
+
+        result = next(Model.aggregate(pipeline, collation=Model._numeric_locale), None)
+        if not result:
+            return 0, []
+
+        total = int(result.get("total", 0))
+        values = result.get("results", [])
+
+        ttl = config.get("services.models.metadata_values.cache_ttl_sec", 86400)
+        cached = dict(last_update=last_update.timestamp(), total=total, values=values)
+        self.redis.setex(redis_key, ttl, json.dumps(cached))
+
+        return total, values

apiserver/bll/project/sub_projects.py

@@ -4,6 +4,7 @@ from typing import Tuple, Optional, Sequence, Mapping
 
 from apiserver import database
 from apiserver.apierrors import errors
+from apiserver.database.model import EntityVisibility
 from apiserver.database.model.project import Project
 
 name_separator = "/"
@@ -25,7 +26,9 @@ def _validate_project_name(project_name: str) -> Tuple[str, str]:
     return name_separator.join(name_parts), name_separator.join(name_parts[:-1])
 
 
-def _ensure_project(company: str, user: str, name: str) -> Optional[Project]:
+def _ensure_project(
+    company: str, user: str, name: str, creation_params: dict = None
+) -> Optional[Project]:
     """
     Makes sure that the project with the given name exists
     If needed auto-create the project and all the missing projects in the path to it
@@ -48,9 +51,10 @@ def _ensure_project(company: str, user: str, name: str) -> Optional[Project]:
         created=now,
         last_update=now,
         name=name,
-        description="",
+        basename=name.split("/")[-1],
+        **(creation_params or dict(description="")),
     )
-    parent = _ensure_project(company, user, location)
+    parent = _ensure_project(company, user, location, creation_params=creation_params)
     _save_under_parent(project=project, parent=parent)
     if parent:
         parent.update(last_update=now)
@@ -98,12 +102,21 @@ def _get_writable_project_from_name(
 
 
 def _get_sub_projects(
-    project_ids: Sequence[str], _only: Sequence[str] = ("id", "path")
+    project_ids: Sequence[str],
+    _only: Sequence[str] = ("id", "path"),
+    search_hidden=True,
+    allowed_ids: Sequence[str] = None,
 ) -> Mapping[str, Sequence[Project]]:
     """
     Return the list of child projects of all the levels for the parent project ids
     """
-    qs = Project.objects(path__in=project_ids)
+    query = dict(path__in=project_ids)
+    if not search_hidden:
+        query["system_tags__nin"] = [EntityVisibility.hidden.value]
+    if allowed_ids:
+        query["id__in"] = allowed_ids
+
+    qs = Project.objects(**query)
     if _only:
         _only = set(_only) | {"path"}
         qs = qs.only(*_only)

apiserver/bll/queue/queue_bll.py

@@ -7,7 +7,7 @@ from elasticsearch import Elasticsearch
 from apiserver import database
 from apiserver.es_factory import es_factory
 from apiserver.apierrors import errors
-from apiserver.bll.queue.queue_metrics import QueueMetrics
+from apiserver.bll.queue.queue_metrics import QueueMetrics, MetricsRefresher
 from apiserver.bll.workers import WorkerBLL
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
@@ -32,7 +32,7 @@ class QueueBLL(object):
         name: str,
         tags: Optional[Sequence[str]] = None,
         system_tags: Optional[Sequence[str]] = None,
-        metadata: Optional[Sequence[dict]] = None,
+        metadata: Optional[dict] = None,
     ) -> Queue:
         """Creates a queue"""
         with translate_errors_context():
@@ -50,8 +50,25 @@ class QueueBLL(object):
             queue.save()
             return queue
 
+    def get_by_name(
+        self, company_id: str, queue_name: str, only: Optional[Sequence[str]] = None,
+    ) -> Queue:
+        qs = Queue.objects(name=queue_name, company=company_id)
+        if only:
+            qs = qs.only(*only)
+
+        return qs.first()
+
+    @staticmethod
+    def _get_task_entries_projection(max_task_entries: int) -> dict:
+        return dict(slice__entries=max_task_entries)
+
     def get_by_id(
-        self, company_id: str, queue_id: str, only: Optional[Sequence[str]] = None
+        self,
+        company_id: str,
+        queue_id: str,
+        only: Optional[Sequence[str]] = None,
+        max_task_entries: int = None,
     ) -> Queue:
         """
         Get queue by id
@@ -62,6 +79,8 @@ class QueueBLL(object):
             qs = Queue.objects(**query)
             if only:
                 qs = qs.only(*only)
+            if max_task_entries:
+                qs = qs.fields(**self._get_task_entries_projection(max_task_entries))
             queue = qs.first()
             if not queue:
                 raise errors.bad_request.InvalidQueueId(**query)
@@ -126,14 +145,32 @@ class QueueBLL(object):
                 )
             queue.delete()
 
-    def get_all(self, company_id: str, query_dict: dict) -> Sequence[dict]:
+    def get_all(
+        self,
+        company_id: str,
+        query_dict: dict,
+        max_task_entries: int = None,
+        ret_params: dict = None,
+    ) -> Sequence[dict]:
         """Get all the queues according to the query"""
         with translate_errors_context():
             return Queue.get_many(
-                company=company_id, parameters=query_dict, query_dict=query_dict
+                company=company_id,
+                parameters=query_dict,
+                query_dict=query_dict,
+                projection_fields=self._get_task_entries_projection(max_task_entries)
+                if max_task_entries
+                else None,
+                ret_params=ret_params,
             )
 
-    def get_queue_infos(self, company_id: str, query_dict: dict) -> Sequence[dict]:
+    def get_queue_infos(
+        self,
+        company_id: str,
+        query_dict: dict,
+        max_task_entries: int = None,
+        ret_params: dict = None,
+    ) -> Sequence[dict]:
         """
         Get infos on all the company queues, including queue tasks and workers
         """
@@ -143,6 +180,10 @@ class QueueBLL(object):
                 company=company_id,
                 query_dict=query_dict,
                 override_projection=projection,
+                projection_fields=self._get_task_entries_projection(max_task_entries)
+                if max_task_entries
+                else None,
+                ret_params=ret_params,
             )
 
             queue_workers = defaultdict(list)
@@ -173,13 +214,15 @@ class QueueBLL(object):
             if any(e.task == task_id for e in queue.entries):
                 raise errors.bad_request.TaskAlreadyQueued(task=task_id)
 
-            self.metrics.log_queue_metrics_to_es(company_id=company_id, queues=[queue])
-
             entry = Entry(added=datetime.utcnow(), task=task_id)
             query = dict(id=queue_id, company=company_id)
             res = Queue.objects(entries__task__ne=task_id, **query).update_one(
                 push__entries=entry, last_update=datetime.utcnow(), upsert=False
             )
+
+            queue.reload()
+            self.metrics.log_queue_metrics_to_es(company_id=company_id, queues=[queue])
+
             if not res:
                 raise errors.bad_request.InvalidQueueOrTaskNotQueued(
                     task=task_id, **query
@@ -219,7 +262,6 @@ class QueueBLL(object):
             queue = self.get_queue_with_task(
                 company_id=company_id, queue_id=queue_id, task_id=task_id
             )
-            self.metrics.log_queue_metrics_to_es(company_id, queues=[queue])
 
             entries_to_remove = [e for e in queue.entries if e.task == task_id]
             query = dict(id=queue_id, company=company_id)
@@ -227,6 +269,9 @@ class QueueBLL(object):
                 pull_all__entries=entries_to_remove, last_update=datetime.utcnow()
             )
 
+            queue.reload()
+            self.metrics.log_queue_metrics_to_es(company_id=company_id, queues=[queue])
+
             return len(entries_to_remove) if res else 0
 
     def reposition_task(
@@ -270,3 +315,25 @@ class QueueBLL(object):
                     )
 
             return new_position
+
+    def count_entries(self, company: str, queue_id: str) -> Optional[int]:
+        res = next(
+            Queue.aggregate(
+                [
+                    {
+                        "$match": {
+                            "company": {"$in": [None, "", company]},
+                            "_id": queue_id,
+                        }
+                    },
+                    {"$project": {"count": {"$size": "$entries"}}},
+                ]
+            ),
+            None,
+        )
+        if res is None:
+            raise errors.bad_request.InvalidQueueId(queue_id=queue_id)
+        return int(res.get("count"))
+
+
+MetricsRefresher.start(queue_metrics=QueueBLL().metrics)

apiserver/bll/queue/queue_metrics.py

@@ -1,8 +1,10 @@
+import json
 from collections import defaultdict
 from datetime import datetime
+from time import sleep
 from typing import Sequence
 
-import elasticsearch.helpers
+from boltons.typeutils import classproperty
 from elasticsearch import Elasticsearch
 
 from apiserver.es_factory import es_factory
@@ -11,25 +13,31 @@ from apiserver.bll.query import Builder as QueryBuilder
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model.queue import Queue, Entry
+from apiserver.redis_manager import redman
 from apiserver.timing_context import TimingContext
+from apiserver.utilities.threads_manager import ThreadsManager
 
 log = config.logger(__file__)
+_conf = config.get("services.queues")
+_queue_metrics_key_pattern = "queue_metrics_{queue}"
+redis = redman.connection("apiserver")
+
+
+class EsKeys:
+    WAITING_TIME_FIELD = "average_waiting_time"
+    QUEUE_LENGTH_FIELD = "queue_length"
+    TIMESTAMP_FIELD = "timestamp"
+    QUEUE_FIELD = "queue"
 
 
 class QueueMetrics:
-    class EsKeys:
-        WAITING_TIME_FIELD = "average_waiting_time"
-        QUEUE_LENGTH_FIELD = "queue_length"
-        TIMESTAMP_FIELD = "timestamp"
-        QUEUE_FIELD = "queue"
-
     def __init__(self, es: Elasticsearch):
         self.es = es
 
     @staticmethod
     def _queue_metrics_prefix_for_company(company_id: str) -> str:
         """Returns the es index prefix for the company"""
-        return f"queue_metrics_{company_id}_"
+        return f"queue_metrics_{company_id.lower()}_"
 
     @staticmethod
     def _get_es_index_suffix():
@@ -49,7 +57,7 @@ class QueueMetrics:
         total_waiting_in_secs = sum((now - e.added).total_seconds() for e in entries)
         return total_waiting_in_secs / len(entries)
 
-    def log_queue_metrics_to_es(self, company_id: str, queues: Sequence[Queue]) -> bool:
+    def log_queue_metrics_to_es(self, company_id: str, queues: Sequence[Queue]) -> int:
         """
         Calculate and write queue statistics (avg waiting time and queue length) to Elastic
         :return: True if the write to es was successful, false otherwise
@@ -63,23 +71,22 @@ class QueueMetrics:
 
         def make_doc(queue: Queue) -> dict:
             entries = [e for e in queue.entries if e.added]
-            return dict(
-                _index=es_index,
-                _source={
-                    self.EsKeys.TIMESTAMP_FIELD: timestamp,
-                    self.EsKeys.QUEUE_FIELD: queue.id,
-                    self.EsKeys.WAITING_TIME_FIELD: self._calc_avg_waiting_time(
-                        entries
-                    ),
-                    self.EsKeys.QUEUE_LENGTH_FIELD: len(entries),
-                },
-            )
+            return {
+                EsKeys.TIMESTAMP_FIELD: timestamp,
+                EsKeys.QUEUE_FIELD: queue.id,
+                EsKeys.WAITING_TIME_FIELD: self._calc_avg_waiting_time(entries),
+                EsKeys.QUEUE_LENGTH_FIELD: len(entries),
+            }
 
-        actions = list(map(make_doc, queues))
+        logged = 0
+        for q in queues:
+            queue_doc = make_doc(q)
+            self.es.index(index=es_index, body=queue_doc)
+            redis_key = _queue_metrics_key_pattern.format(queue=q.id)
+            redis.set(redis_key, json.dumps(queue_doc))
+            logged += 1
 
-        es_res = elasticsearch.helpers.bulk(self.es, actions)
-        added, errors = es_res[:2]
-        return (added == len(actions)) and not errors
+        return logged
 
     def _log_current_metrics(self, company_id: str, queue_ids=Sequence[str]):
         query = dict(company=company_id)
@@ -90,8 +97,7 @@ class QueueMetrics:
 
     def _search_company_metrics(self, company_id: str, es_req: dict) -> dict:
         return self.es.search(
-            index=f"{self._queue_metrics_prefix_for_company(company_id)}*",
-            body=es_req,
+            index=f"{self._queue_metrics_prefix_for_company(company_id)}*", body=es_req,
         )
 
     @classmethod
@@ -105,13 +111,13 @@ class QueueMetrics:
         return {
             "dates": {
                 "date_histogram": {
-                    "field": cls.EsKeys.TIMESTAMP_FIELD,
+                    "field": EsKeys.TIMESTAMP_FIELD,
                     "fixed_interval": f"{interval}s",
                     "min_doc_count": 1,
                 },
                 "aggs": {
                     "queues": {
-                        "terms": {"field": cls.EsKeys.QUEUE_FIELD},
+                        "terms": {"field": EsKeys.QUEUE_FIELD},
                         "aggs": cls._get_top_waiting_agg(),
                     }
                 },
@@ -128,13 +134,13 @@ class QueueMetrics:
             "top_avg_waiting": {
                 "top_hits": {
                     "sort": [
-                        {cls.EsKeys.WAITING_TIME_FIELD: {"order": "desc"}},
-                        {cls.EsKeys.QUEUE_LENGTH_FIELD: {"order": "desc"}},
+                        {EsKeys.WAITING_TIME_FIELD: {"order": "desc"}},
+                        {EsKeys.QUEUE_LENGTH_FIELD: {"order": "desc"}},
                     ],
                     "_source": {
                         "includes": [
-                            cls.EsKeys.WAITING_TIME_FIELD,
-                            cls.EsKeys.QUEUE_LENGTH_FIELD,
+                            EsKeys.WAITING_TIME_FIELD,
+                            EsKeys.QUEUE_LENGTH_FIELD,
                         ]
                     },
                     "size": 1,
@@ -149,6 +155,7 @@ class QueueMetrics:
         to_date: float,
         interval: int,
         queue_ids: Sequence[str],
+        refresh: bool = False,
     ) -> dict:
         """
         Get the company queue metrics in the specified time range.
@@ -158,7 +165,8 @@ class QueueMetrics:
         In case no queue ids are specified the avg across all the
         company queues is calculated for each metric
         """
-        # self._log_current_metrics(company, queue_ids=queue_ids)
+        if refresh:
+            self._log_current_metrics(company_id, queue_ids=queue_ids)
 
         if from_date >= to_date:
             raise bad_request.FieldsValueError("from_date must be less than to_date")
@@ -256,7 +264,47 @@ class QueueMetrics:
                 continue
             res = queue_data["top_avg_waiting"]["hits"]["hits"][0]["_source"]
             queue_metrics[queue_data["key"]] = {
-                "queue_length": res[cls.EsKeys.QUEUE_LENGTH_FIELD],
-                "avg_waiting_time": res[cls.EsKeys.WAITING_TIME_FIELD],
+                "queue_length": res[EsKeys.QUEUE_LENGTH_FIELD],
+                "avg_waiting_time": res[EsKeys.WAITING_TIME_FIELD],
             }
         return queue_metrics
+
+
+class MetricsRefresher:
+    threads = ThreadsManager()
+
+    @classproperty
+    def watch_interval_sec(self):
+        return _conf.get("metrics_refresh_interval_sec", 300)
+
+    @classmethod
+    @threads.register("queue_metrics_refresh_watchdog", daemon=True)
+    def start(cls, queue_metrics: QueueMetrics):
+        if not cls.watch_interval_sec:
+            return
+
+        sleep(10)
+        while not ThreadsManager.terminating:
+            try:
+                for queue in Queue.objects():
+                    timestamp = es_factory.get_timestamp_millis()
+                    doc_time = 0
+                    try:
+                        redis_key = _queue_metrics_key_pattern.format(queue=queue.id)
+                        data = redis.get(redis_key)
+                        if data:
+                            queue_doc = json.loads(data)
+                            doc_time = int(queue_doc.get(EsKeys.TIMESTAMP_FIELD))
+                    except Exception as ex:
+                        log.exception(
+                            f"Error reading queue metrics data for queue {queue.id}: {str(ex)}"
+                        )
+
+                    if (
+                        not doc_time
+                        or (timestamp - doc_time) > cls.watch_interval_sec * 1000
+                    ):
+                        queue_metrics.log_queue_metrics_to_es(queue.company, [queue])
+            except Exception as ex:
+                log.exception(f"Failed collecting queue metrics: {str(ex)}")
+            sleep(60)

apiserver/bll/redis_cache_manager.py

@@ -49,6 +49,21 @@ class RedisCacheManager(Generic[T]):
     def _get_redis_key(self, state_id):
         return f"{self.state_class}/{state_id}"
 
+    def get_or_create_state_core(
+        self,
+        state_id=None,
+        init_state: Callable[[T], None] = _do_nothing,
+        validate_state: Callable[[T], None] = _do_nothing,
+    ) -> T:
+        state = self.get_state(state_id) if state_id else None
+        if state:
+            validate_state(state)
+        else:
+            state = self.state_class(id=database.utils.id())
+            init_state(state)
+
+        return state
+
     @contextmanager
     def get_or_create_state(
         self,
@@ -66,12 +81,9 @@ class RedisCacheManager(Generic[T]):
         :param validate_state: user callback to validate the state if retrieved from cache
         Should throw an exception if the state is not valid. If not passed then no validation is done
         """
-        state = self.get_state(state_id) if state_id else None
-        if state:
-            validate_state(state)
-        else:
-            state = self.state_class(id=database.utils.id())
-            init_state(state)
+        state = self.get_or_create_state_core(
+            state_id=state_id, init_state=init_state, validate_state=validate_state
+        )
 
         try:
             yield state

apiserver/bll/task/param_utils.py

@@ -1,11 +1,10 @@
 import itertools
-from typing import Sequence, Tuple
+from typing import Sequence, Tuple, Optional
 
-import dpath
 
 from apiserver.apierrors import errors
 from apiserver.database.model.task.task import Task
-from apiserver.tools import safe_get
+from apiserver.utilities.dicts import nested_get, nested_delete, nested_set
 from apiserver.utilities.parameter_key_escaper import ParameterKeyEscaper
 
 
@@ -14,7 +13,7 @@ hyperparams_legacy_type = "legacy"
 tf_define_section = "TF_DEFINE"
 
 
-def split_param_name(full_name: str, default_section: str) -> Tuple[str, str]:
+def split_param_name(full_name: str, default_section: str) -> Tuple[Optional[str], str]:
     """
     Return parameter section and name. The section is either TF_DEFINE or the default one
     """
@@ -62,7 +61,7 @@ def _remove_legacy_params(data: dict, with_sections: bool = False) -> int:
     return removed
 
 
-def _get_legacy_params(data: dict, with_sections: bool = False) -> Sequence[str]:
+def _get_legacy_params(data: dict, with_sections: bool = False) -> Sequence[dict]:
     """
     Remove the legacy params from the data dict and return the number of removed params
     If the path not found then return 0
@@ -71,8 +70,10 @@ def _get_legacy_params(data: dict, with_sections: bool = False) -> Sequence[str]
         return []
 
     if with_sections:
-        return itertools.chain.from_iterable(
-            _get_legacy_params(section_data) for section_data in data.values()
+        return list(
+            itertools.chain.from_iterable(
+                _get_legacy_params(section_data) for section_data in data.values()
+            )
         )
 
     return [
@@ -86,15 +87,15 @@ def params_prepare_for_save(fields: dict, previous_task: Task = None):
     Escape all the section and param names for hyper params and configuration to make it mongo sage
     """
     for old_params_field, new_params_field, default_section in (
-        ("execution/parameters", "hyperparams", hyperparams_default_section),
-        ("execution/model_desc", "configuration", None),
+        (("execution", "parameters"), "hyperparams", hyperparams_default_section),
+        (("execution", "model_desc"), "configuration", None),
     ):
-        legacy_params = safe_get(fields, old_params_field)
+        legacy_params = nested_get(fields, old_params_field)
         if legacy_params is None:
             continue
 
         if (
-            not safe_get(fields, new_params_field)
+            not fields.get(new_params_field)
             and previous_task
             and previous_task[new_params_field]
         ):
@@ -117,21 +118,34 @@ def params_prepare_for_save(fields: dict, previous_task: Task = None):
             new_param = dict(name=name, type=hyperparams_legacy_type, value=str(value))
             if section is not None:
                 new_param["section"] = section
-            dpath.new(fields, new_path, new_param)
-        dpath.delete(fields, old_params_field)
+            nested_set(fields, new_path, new_param)
+        nested_delete(fields, old_params_field)
 
-    for param_field in ("hyperparams", "configuration"):
-        params = safe_get(fields, param_field)
-        if params:
-            escaped_params = {
-                ParameterKeyEscaper.escape(key): {
-                    ParameterKeyEscaper.escape(k): v for k, v in value.items()
-                }
-                if isinstance(value, dict)
-                else value
-                for key, value in params.items()
+    def ensure_non_empty(k: str, desc: str) -> str:
+        if not k:
+            raise errors.bad_request.ValidationError(
+                f"Empty {desc} name is not allowed"
+            )
+        return k
+
+    params = fields.get("hyperparams")
+    if params:
+        escaped_params = {
+            ParameterKeyEscaper.escape(ensure_non_empty(key, "section")): {
+                ParameterKeyEscaper.escape(ensure_non_empty(k, "parameter")): v
+                for k, v in value.items()
             }
-            dpath.set(fields, param_field, escaped_params)
+            for key, value in params.items()
+        }
+        fields["hyperparams"] = escaped_params
+
+    params = fields.get("configuration")
+    if params:
+        escaped_params = {
+            ParameterKeyEscaper.escape(ensure_non_empty(key, "configuration")): value
+            for key, value in params.items()
+        }
+        fields["configuration"] = escaped_params
 
 
 def params_unprepare_from_saved(fields, copy_to_legacy=False):
@@ -140,7 +154,7 @@ def params_unprepare_from_saved(fields, copy_to_legacy=False):
     If copy_to_legacy is set then copy hyperparams and configuration data to the legacy location for the old clients
     """
     for param_field in ("hyperparams", "configuration"):
-        params = safe_get(fields, param_field)
+        params = fields.get(param_field)
         if params:
             unescaped_params = {
                 ParameterKeyEscaper.unescape(key): {
@@ -150,18 +164,18 @@ def params_unprepare_from_saved(fields, copy_to_legacy=False):
                 else value
                 for key, value in params.items()
             }
-            dpath.set(fields, param_field, unescaped_params)
+            fields[param_field] = unescaped_params
 
     if copy_to_legacy:
         for new_params_field, old_params_field, use_sections in (
-            (f"hyperparams", "execution/parameters", True),
-            (f"configuration", "execution/model_desc", False),
+            ("hyperparams", ("execution", "parameters"), True),
+            ("configuration", ("execution", "model_desc"), False),
         ):
             legacy_params = _get_legacy_params(
-                safe_get(fields, new_params_field), with_sections=use_sections
+                fields.get(new_params_field), with_sections=use_sections
             )
             if legacy_params:
-                dpath.new(
+                nested_set(
                     fields,
                     old_params_field,
                     {_get_full_param_name(p): p["value"] for p in legacy_params},
@@ -174,7 +188,7 @@ def _process_path(path: str):
     Need to unescape and apply a full mongo escaping
     """
     parts = path.split(".")
-    if len(parts) < 2 or len(parts) > 3:
+    if len(parts) < 2 or len(parts) > 4:
         raise errors.bad_request.ValidationError("invalid task field", path=path)
     return ".".join(
         ParameterKeyEscaper.escape(ParameterKeyEscaper.unescape(p)) for p in parts
@@ -184,8 +198,8 @@ def _process_path(path: str):
 def escape_paths(paths: Sequence[str]) -> Sequence[str]:
     for old_prefix, new_prefix in (
         ("execution.parameters", f"hyperparams.{hyperparams_default_section}"),
-        ("execution.model_desc", f"configuration"),
-        ("execution.docker_cmd", "container")
+        ("execution.model_desc", "configuration"),
+        ("execution.docker_cmd", "container"),
     ):
         path: str
         paths = [path.replace(old_prefix, new_prefix) for path in paths]

apiserver/bll/task/task_bll.py

@@ -1,9 +1,6 @@
-import json
-from collections import OrderedDict
-from datetime import datetime, timedelta
+from datetime import datetime
 from typing import Collection, Sequence, Tuple, Any, Optional, Dict
 
-import dpath
 import six
 from mongoengine import Q
 from redis import StrictRedis
@@ -14,7 +11,7 @@ from apiserver.apierrors import errors
 from apiserver.apimodels.tasks import TaskInputModel
 from apiserver.bll.queue import QueueBLL
 from apiserver.bll.organization import OrgBLL, Tags
-from apiserver.bll.project import ProjectBLL, project_ids_with_children
+from apiserver.bll.project import ProjectBLL
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model.model import Model
@@ -39,7 +36,6 @@ from apiserver.redis_manager import redman
 from apiserver.service_repo import APICall
 from apiserver.services.utils import validate_tags, escape_dict_field, escape_dict
 from apiserver.timing_context import TimingContext
-from apiserver.utilities.parameter_key_escaper import ParameterKeyEscaper
 from .artifacts import artifacts_prepare_for_save
 from .param_utils import params_prepare_for_save
 from .utils import (
@@ -350,54 +346,6 @@ class TaskBLL:
         if validate_models:
             cls.validate_input_models(task)
 
-    @staticmethod
-    def get_unique_metric_variants(
-        company_id, project_ids: Sequence[str], include_subprojects: bool
-    ):
-        if project_ids:
-            if include_subprojects:
-                project_ids = project_ids_with_children(project_ids)
-            project_constraint = {"project": {"$in": project_ids}}
-        else:
-            project_constraint = {}
-        pipeline = [
-            {
-                "$match": dict(
-                    company={"$in": [None, "", company_id]}, **project_constraint,
-                )
-            },
-            {"$project": {"metrics": {"$objectToArray": "$last_metrics"}}},
-            {"$unwind": "$metrics"},
-            {
-                "$project": {
-                    "metric": "$metrics.k",
-                    "variants": {"$objectToArray": "$metrics.v"},
-                }
-            },
-            {"$unwind": "$variants"},
-            {
-                "$group": {
-                    "_id": {
-                        "metric": "$variants.v.metric",
-                        "variant": "$variants.v.variant",
-                    },
-                    "metrics": {
-                        "$addToSet": {
-                            "metric": "$variants.v.metric",
-                            "metric_hash": "$metric",
-                            "variant": "$variants.v.variant",
-                            "variant_hash": "$variants.k",
-                        }
-                    },
-                }
-            },
-            {"$sort": OrderedDict({"_id.metric": 1, "_id.variant": 1})},
-        ]
-
-        with translate_errors_context():
-            result = Task.aggregate(pipeline)
-            return [r["metrics"][0] for r in result]
-
     @staticmethod
     def set_last_update(
         task_ids: Collection[str],
@@ -494,173 +442,6 @@ class TaskBLL:
             **extra_updates,
         )
 
-    @staticmethod
-    def get_aggregated_project_parameters(
-        company_id,
-        project_ids: Sequence[str],
-        include_subprojects: bool,
-        page: int = 0,
-        page_size: int = 500,
-    ) -> Tuple[int, int, Sequence[dict]]:
-        if project_ids:
-            if include_subprojects:
-                project_ids = project_ids_with_children(project_ids)
-            project_constraint = {"project": {"$in": project_ids}}
-        else:
-            project_constraint = {}
-        page = max(0, page)
-        page_size = max(1, page_size)
-        pipeline = [
-            {
-                "$match": {
-                    "company": {"$in": [None, "", company_id]},
-                    "hyperparams": {"$exists": True, "$gt": {}},
-                    **project_constraint,
-                }
-            },
-            {"$project": {"sections": {"$objectToArray": "$hyperparams"}}},
-            {"$unwind": "$sections"},
-            {
-                "$project": {
-                    "section": "$sections.k",
-                    "names": {"$objectToArray": "$sections.v"},
-                }
-            },
-            {"$unwind": "$names"},
-            {"$group": {"_id": {"section": "$section", "name": "$names.k"}}},
-            {"$sort": OrderedDict({"_id.section": 1, "_id.name": 1})},
-            {"$skip": page * page_size},
-            {"$limit": page_size},
-            {
-                "$group": {
-                    "_id": 1,
-                    "total": {"$sum": 1},
-                    "results": {"$push": "$$ROOT"},
-                }
-            },
-        ]
-
-        result = next(Task.aggregate(pipeline), None)
-
-        total = 0
-        remaining = 0
-        results = []
-
-        if result:
-            total = int(result.get("total", -1))
-            results = [
-                {
-                    "section": ParameterKeyEscaper.unescape(
-                        dpath.get(r, "_id/section")
-                    ),
-                    "name": ParameterKeyEscaper.unescape(dpath.get(r, "_id/name")),
-                }
-                for r in result.get("results", [])
-            ]
-            remaining = max(0, total - (len(results) + page * page_size))
-
-        return total, remaining, results
-
-    HyperParamValues = Tuple[int, Sequence[str]]
-
-    def _get_cached_hyperparam_values(
-        self, key: str, last_update: datetime
-    ) -> Optional[HyperParamValues]:
-        allowed_delta = timedelta(
-            seconds=config.get(
-                "services.tasks.hyperparam_values.cache_allowed_outdate_sec", 60
-            )
-        )
-        try:
-            cached = self.redis.get(key)
-            if not cached:
-                return
-
-            data = json.loads(cached)
-            cached_last_update = datetime.fromtimestamp(data["last_update"])
-            if (last_update - cached_last_update) < allowed_delta:
-                return data["total"], data["values"]
-        except Exception as ex:
-            log.error(f"Error retrieving hyperparam cached values: {str(ex)}")
-
-    def get_hyperparam_distinct_values(
-        self,
-        company_id: str,
-        project_ids: Sequence[str],
-        section: str,
-        name: str,
-        include_subprojects: bool,
-        allow_public: bool = True,
-    ) -> HyperParamValues:
-        if allow_public:
-            company_constraint = {"company": {"$in": [None, "", company_id]}}
-        else:
-            company_constraint = {"company": company_id}
-        if project_ids:
-            if include_subprojects:
-                project_ids = project_ids_with_children(project_ids)
-            project_constraint = {"project": {"$in": project_ids}}
-        else:
-            project_constraint = {}
-
-        key_path = f"hyperparams.{ParameterKeyEscaper.escape(section)}.{ParameterKeyEscaper.escape(name)}"
-        last_updated_task = (
-            Task.objects(
-                **company_constraint,
-                **project_constraint,
-                **{f"{key_path.replace('.', '__')}__exists": True},
-            )
-            .only("last_update")
-            .order_by("-last_update")
-            .limit(1)
-            .first()
-        )
-        if not last_updated_task:
-            return 0, []
-
-        redis_key = f"hyperparam_values_{company_id}_{'_'.join(project_ids)}_{section}_{name}_{allow_public}"
-        last_update = last_updated_task.last_update or datetime.utcnow()
-        cached_res = self._get_cached_hyperparam_values(
-            key=redis_key, last_update=last_update
-        )
-        if cached_res:
-            return cached_res
-
-        max_values = config.get("services.tasks.hyperparam_values.max_count", 100)
-        pipeline = [
-            {
-                "$match": {
-                    **company_constraint,
-                    **project_constraint,
-                    key_path: {"$exists": True},
-                }
-            },
-            {"$project": {"value": f"${key_path}.value"}},
-            {"$group": {"_id": "$value"}},
-            {"$sort": {"_id": 1}},
-            {"$limit": max_values},
-            {
-                "$group": {
-                    "_id": 1,
-                    "total": {"$sum": 1},
-                    "results": {"$push": "$$ROOT._id"},
-                }
-            },
-        ]
-
-        result = next(Task.aggregate(pipeline, collation=Task._numeric_locale), None)
-        if not result:
-            return 0, []
-
-        total = int(result.get("total", 0))
-        values = result.get("results", [])
-
-        ttl = config.get("services.tasks.hyperparam_values.cache_ttl_sec", 86400)
-        cached = dict(last_update=last_update.timestamp(), total=total, values=values)
-        self.redis.setex(redis_key, ttl, json.dumps(cached))
-
-        return total, values
-
     @classmethod
     def dequeue_and_change_status(
         cls, task: Task, company_id: str, status_message: str, status_reason: str,

apiserver/bll/task/task_cleanup.py

@@ -1,64 +1,19 @@
 from itertools import chain
 from operator import attrgetter
-from typing import Sequence, Generic, Callable, Type, Iterable, TypeVar, List, Set
+from typing import Sequence, Set, Tuple
 
 import attr
 from boltons.iterutils import partition
-from mongoengine import QuerySet, Document
 
 from apiserver.apierrors import errors
 from apiserver.bll.event import EventBLL
 from apiserver.bll.event.event_bll import PlotFields
-from apiserver.bll.event.event_common import EventType
 from apiserver.bll.task.utils import deleted_prefix
 from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task, TaskStatus, ArtifactModes
 from apiserver.timing_context import TimingContext
 
 event_bll = EventBLL()
-T = TypeVar("T", bound=Document)
-
-
-class DocumentGroup(List[T]):
-    """
-    Operate on a list of documents as if they were a query result
-    """
-
-    def __init__(self, document_type: Type[T], documents: Iterable[T]):
-        super(DocumentGroup, self).__init__(documents)
-        self.type = document_type
-
-    @property
-    def ids(self) -> Set[str]:
-        return {obj.id for obj in self}
-
-    def objects(self, *args, **kwargs) -> QuerySet:
-        return self.type.objects(id__in=self.ids, *args, **kwargs)
-
-
-class TaskOutputs(Generic[T]):
-    """
-    Split task outputs of the same type by the ready state
-    """
-
-    published: DocumentGroup[T]
-    draft: DocumentGroup[T]
-
-    def __init__(
-        self,
-        is_published: Callable[[T], bool],
-        document_type: Type[T],
-        children: Iterable[T],
-    ):
-        """
-        :param is_published: predicate returning whether items is considered published
-        :param document_type: type of output
-        :param children: output documents
-        """
-        self.published, self.draft = map(
-            lambda x: DocumentGroup(document_type, x),
-            partition(children, key=is_published),
-        )
 
 
 @attr.s(auto_attribs=True)
@@ -124,32 +79,18 @@ def collect_debug_image_urls(company: str, task: str) -> Set[str]:
     Return the set of unique image urls
     Uses DebugImagesIterator to make sure that we do not retrieve recycled urls
     """
-    metrics = event_bll.get_metrics_and_variants(
-        company_id=company, task_id=task, event_type=EventType.metrics_image
-    )
-    if not metrics:
-        return set()
-
-    task_metrics = {task: set(metrics)}
-    scroll_id = None
+    after_key = None
     urls = set()
     while True:
-        res = event_bll.debug_images_iterator.get_task_events(
+        res, after_key = event_bll.get_debug_image_urls(
             company_id=company,
-            task_metrics=task_metrics,
-            iter_count=100,
-            state_id=scroll_id,
+            task_id=task,
+            after_key=after_key,
         )
-        if not res.metric_events or not any(
-            iterations for _, iterations in res.metric_events
-        ):
+        urls.update(res)
+        if not after_key:
             break
 
-        scroll_id = res.next_scroll_id
-        for task, iterations in res.metric_events:
-            urls.update(ev.get("url") for it in iterations for ev in it["events"])
-
-    urls.discard({None})
     return urls
 
 
@@ -166,7 +107,9 @@ def cleanup_task(
     :param force: whether to delete task with published outputs
     :return: count of delete and modified items
     """
-    models = verify_task_children_and_ouptuts(task, force)
+    published_models, draft_models, in_use_model_ids = verify_task_children_and_ouptuts(
+        task, force
+    )
 
     event_urls, artifact_urls, model_urls = set(), set(), set()
     if return_file_urls:
@@ -178,28 +121,37 @@ def cleanup_task(
                 for a in task.execution.artifacts.values()
                 if a.mode == ArtifactModes.output and a.uri
             }
-        model_urls = {m.uri for m in models.draft.objects().only("uri") if m.uri}
+        model_urls = {
+            m.uri for m in draft_models if m.uri and m.id not in in_use_model_ids
+        }
 
     deleted_task_id = f"{deleted_prefix}{task.id}"
+    updated_children = 0
     if update_children:
         with TimingContext("mongo", "update_task_children"):
             updated_children = Task.objects(parent=task.id).update(
                 parent=deleted_task_id
             )
-    else:
-        updated_children = 0
 
-    if models.draft and delete_output_models:
-        with TimingContext("mongo", "delete_models"):
-            deleted_models = models.draft.objects().delete()
-    else:
-        deleted_models = 0
+    deleted_models = 0
+    updated_models = 0
+    for models, allow_delete in ((draft_models, True), (published_models, False)):
+        if not models:
+            continue
+        if delete_output_models and allow_delete:
+            deleted_models += Model.objects(
+                id__in=[m.id for m in models if m.id not in in_use_model_ids]
+            ).delete()
+            if in_use_model_ids:
+                Model.objects(id__in=list(in_use_model_ids)).update(unset__task=1)
+            continue
 
-    if models.published and update_children:
-        with TimingContext("mongo", "update_task_models"):
-            updated_models = models.published.objects().update(task=deleted_task_id)
-    else:
-        updated_models = 0
+        if update_children:
+            updated_models += Model.objects(id__in=[m.id for m in models]).update(
+                task=deleted_task_id
+            )
+        else:
+            Model.objects(id__in=[m.id for m in models]).update(unset__task=1)
 
     event_bll.delete_task_events(task.company, task.id, allow_locked=force)
 
@@ -217,7 +169,9 @@ def cleanup_task(
     )
 
 
-def verify_task_children_and_ouptuts(task: Task, force: bool) -> TaskOutputs[Model]:
+def verify_task_children_and_ouptuts(
+    task, force: bool
+) -> Tuple[Sequence[Model], Sequence[Model], Set[str]]:
     if not force:
         with TimingContext("mongo", "count_published_children"):
             published_children_count = Task.objects(
@@ -230,49 +184,42 @@ def verify_task_children_and_ouptuts(task: Task, force: bool) -> TaskOutputs[Mod
                     children=published_children_count,
                 )
 
-    with TimingContext("mongo", "get_task_models"):
-        models = TaskOutputs(
-            attrgetter("ready"),
-            Model,
-            Model.objects(task=task.id).only("id", "task", "ready"),
+    model_fields = ["id", "ready", "uri"]
+    published_models, draft_models = partition(
+        Model.objects(task=task.id).only(*model_fields), key=attrgetter("ready"),
+    )
+    if not force and published_models:
+        raise errors.bad_request.TaskCannotBeDeleted(
+            "has output models, use force=True",
+            task=task.id,
+            models=len(published_models),
         )
-        if not force and models.published:
-            raise errors.bad_request.TaskCannotBeDeleted(
-                "has output models, use force=True",
-                task=task.id,
-                models=len(models.published),
-            )
 
     if task.models and task.models.output:
-        with TimingContext("mongo", "get_task_output_model"):
-            model_ids = [m.model for m in task.models.output]
-            for output_model in Model.objects(id__in=model_ids):
-                if output_model.ready:
-                    if not force:
-                        raise errors.bad_request.TaskCannotBeDeleted(
-                            "has output model, use force=True",
-                            task=task.id,
-                            model=output_model.id,
-                        )
-                    models.published.append(output_model)
-                else:
-                    models.draft.append(output_model)
+        model_ids = [m.model for m in task.models.output]
+        for output_model in Model.objects(id__in=model_ids).only(*model_fields):
+            if output_model.ready:
+                if not force:
+                    raise errors.bad_request.TaskCannotBeDeleted(
+                        "has output model, use force=True",
+                        task=task.id,
+                        model=output_model.id,
+                    )
+                published_models.append(output_model)
+            else:
+                draft_models.append(output_model)
 
-    if models.draft:
-        with TimingContext("mongo", "get_execution_models"):
-            model_ids = models.draft.ids
-            dependent_tasks = Task.objects(models__input__model__in=model_ids).only(
-                "id", "models"
+    in_use_model_ids = {}
+    if draft_models:
+        model_ids = {m.id for m in draft_models}
+        dependent_tasks = Task.objects(models__input__model__in=list(model_ids)).only(
+            "id", "models"
+        )
+        in_use_model_ids = model_ids & {
+            m.model
+            for m in chain.from_iterable(
+                t.models.input for t in dependent_tasks if t.models
             )
-            input_models = {
-                m.model
-                for m in chain.from_iterable(
-                    t.models.input for t in dependent_tasks if t.models
-                )
-            }
-            if input_models:
-                models.draft = DocumentGroup(
-                    Model, (m for m in models.draft if m.id not in input_models)
-                )
+        }
 
-    return models
+    return published_models, draft_models, in_use_model_ids

apiserver/bll/task/task_operations.py

@@ -1,5 +1,5 @@
 from datetime import datetime
-from typing import Callable, Any, Tuple, Union
+from typing import Callable, Any, Tuple, Union, Sequence
 
 from apiserver.apierrors import errors, APIError
 from apiserver.bll.queue import QueueBLL
@@ -25,6 +25,7 @@ from apiserver.database.model.task.task import (
 )
 from apiserver.utilities.dicts import nested_set
 
+log = config.logger(__file__)
 queue_bll = QueueBLL()
 
 
@@ -83,10 +84,7 @@ def unarchive_task(
 
 
 def dequeue_task(
-    task_id: str,
-    company_id: str,
-    status_message: str,
-    status_reason: str,
+    task_id: str, company_id: str, status_message: str, status_reason: str,
 ) -> Tuple[int, dict]:
     query = dict(id=task_id, company=company_id)
     task = Task.get_for_writing(**query)
@@ -94,10 +92,7 @@ def dequeue_task(
         raise errors.bad_request.InvalidTaskId(**query)
 
     res = TaskBLL.dequeue_and_change_status(
-        task,
-        company_id,
-        status_message=status_message,
-        status_reason=status_reason,
+        task, company_id, status_message=status_message, status_reason=status_reason,
     )
     return 1, res
 
@@ -108,8 +103,23 @@ def enqueue_task(
     queue_id: str,
     status_message: str,
     status_reason: str,
+    queue_name: str = None,
     validate: bool = False,
+    force: bool = False,
 ) -> Tuple[int, dict]:
+    if queue_id and queue_name:
+        raise errors.bad_request.ValidationError(
+            "Either queue id or queue name should be provided"
+        )
+
+    if queue_name:
+        queue = queue_bll.get_by_name(
+            company_id=company_id, queue_name=queue_name, only=("id",)
+        )
+        if not queue:
+            queue = queue_bll.create(company_id=company_id, name=queue_name)
+        queue_id = queue.id
+
     if not queue_id:
         # try to get default queue
         queue_id = queue_bll.get_default(company_id).id
@@ -128,6 +138,7 @@ def enqueue_task(
         status_reason=status_reason,
         status_message=status_message,
         allow_same_state_transition=False,
+        force=force,
     ).execute(enqueue_status=task.status)
 
     try:
@@ -153,6 +164,30 @@ def enqueue_task(
     return 1, res
 
 
+def move_tasks_to_trash(tasks: Sequence[str]) -> int:
+    try:
+        collection_name = Task._get_collection_name()
+        trash_collection_name = f"{collection_name}__trash"
+        Task.aggregate(
+            [
+                {"$match": {"_id": {"$in": tasks}}},
+                {
+                    "$merge": {
+                        "into": trash_collection_name,
+                        "on": "_id",
+                        "whenMatched": "replace",
+                        "whenNotMatched": "insert",
+                    }
+                },
+            ],
+            allow_disk_use=True,
+        )
+    except Exception as ex:
+        log.error(f"Error copying tasks to trash {str(ex)}")
+
+    return Task.objects(id__in=tasks).delete()
+
+
 def delete_task(
     task_id: str,
     company_id: str,
@@ -160,6 +195,8 @@ def delete_task(
     force: bool,
     return_file_urls: bool,
     delete_output_models: bool,
+    status_message: str,
+    status_reason: str,
 ) -> Tuple[int, Task, CleanupResult]:
     task = TaskBLL.get_task_with_access(
         task_id, company_id=company_id, requires_write_access=True
@@ -177,6 +214,17 @@ def delete_task(
             current=task.status,
         )
 
+    try:
+        TaskBLL.dequeue_and_change_status(
+            task,
+            company_id=company_id,
+            status_message=status_message,
+            status_reason=status_reason,
+        )
+    except APIError:
+        # dequeue may fail if the task was not enqueued
+        pass
+
     cleanup_res = cleanup_task(
         task,
         force=force,
@@ -185,18 +233,12 @@ def delete_task(
     )
 
     if move_to_trash:
-        collection_name = task._get_collection_name()
-        archived_collection = "{}__trash".format(collection_name)
-        task.switch_collection(archived_collection)
-        try:
-            # A simple save() won't do due to mongoengine caching (nothing will be saved), so we have to force
-            # an insert. However, if for some reason such an ID exists, let's make sure we'll keep going.
-            task.save(force_insert=True)
-        except Exception:
-            pass
-        task.switch_collection(collection_name)
+        # make sure that whatever changes were done to the task are saved
+        # the task itself will be deleted later in the move_tasks_to_trash operation
+        task.save()
+    else:
+        task.delete()
 
-    task.delete()
     update_project_time(task.project)
     return 1, task, cleanup_res
 
@@ -238,6 +280,7 @@ def reset_task(
         set__last_metrics={},
         set__metric_stats={},
         set__models__output=[],
+        set__runtime={},
         unset__output__result=1,
         unset__output__error=1,
         unset__last_worker=1,
@@ -351,6 +394,7 @@ def stop_task(
             "system_tags",
             "last_worker",
             "last_update",
+            "execution.queue",
         ),
         requires_write_access=True,
     )
@@ -364,7 +408,21 @@ def stop_task(
             and (datetime.utcnow() - t.last_update).total_seconds() < update_timeout
         )
 
-    if TaskSystemTags.development in task.system_tags or not is_run_by_worker(task):
+    is_queued = task.status == TaskStatus.queued
+    set_stopped = (
+        is_queued
+        or TaskSystemTags.development in task.system_tags
+        or not is_run_by_worker(task)
+    )
+
+    if set_stopped:
+        if is_queued:
+            try:
+                TaskBLL.dequeue(task, company_id=company_id, silent_fail=True)
+            except APIError:
+                # dequeue may fail if the task was not enqueued
+                pass
+
         new_status = TaskStatus.stopped
         status_message = f"Stopped by {user_name}"
     else:

apiserver/bll/workers/__init__.py

@@ -76,7 +76,7 @@ class WorkerBLL:
                 raise bad_request.InvalidUserId(**query)
             company = Company.objects(id=company_id).only("id", "name").first()
             if not company:
-                raise server_error.InternalError("invalid company", company=company_id)
+                raise bad_request.InvalidId("invalid company", company=company_id)
 
             queue_objs = Queue.objects(company=company_id, id__in=queues).only("id")
             if len(queue_objs) < len(queues):
@@ -113,7 +113,7 @@ class WorkerBLL:
             res = self.redis.delete(
                 company_id, self._get_worker_key(company_id, user_id, worker)
             )
-        if not res:
+        if not res and not config.get("apiserver.workers.auto_unregister", False):
             raise bad_request.WorkerNotRegistered(worker=worker)
 
     def status_report(
@@ -143,7 +143,7 @@ class WorkerBLL:
                 self._log_stats_to_es(
                     company_id=company_id,
                     company_name=entry.company.name,
-                    worker=report.worker,
+                    worker=entry.key,
                     timestamp=report.timestamp,
                     task=report.task,
                     machine_stats=report.machine_stats,
@@ -189,7 +189,10 @@ class WorkerBLL:
             self._save_worker(entry)
 
     def get_all(
-        self, company_id: str, last_seen: Optional[int] = None
+        self,
+        company_id: str,
+        last_seen: Optional[int] = None,
+        tags: Sequence[str] = None,
     ) -> Sequence[WorkerEntry]:
         """
         Get all the company workers that were active during the last_seen period
@@ -210,16 +213,26 @@ class WorkerBLL:
                 if w.last_activity_time.replace(tzinfo=None) >= ref_time
             ]
 
+        if tags:
+            include = {t for t in tags if not t.startswith("-")}
+            exclude = {t[1:] for t in tags if t.startswith("-")}
+            workers = [
+                w
+                for w in workers
+                if (not include or any(t in include for t in w.tags))
+                and (not exclude or all(t not in exclude for t in w.tags))
+            ]
+
         return workers
 
     def get_all_with_projection(
-        self, company_id: str, last_seen: int
+        self, company_id: str, last_seen: int, tags: Sequence[str] = None
     ) -> Sequence[WorkerResponseEntry]:
 
         helpers = list(
             map(
                 WorkerConversionHelper.from_worker_entry,
-                self.get_all(company_id=company_id, last_seen=last_seen),
+                self.get_all(company_id=company_id, last_seen=last_seen, tags=tags),
             )
         )
 
@@ -258,7 +271,7 @@ class WorkerBLL:
             tasks_info = {
                 task.id: task
                 for task in Task.objects(id__in=task_ids).only(
-                    "name", "started", "last_iteration"
+                    "name", "started", "last_iteration", "active_duration"
                 )
             }
 
@@ -283,11 +296,7 @@ class WorkerBLL:
             if helper.task_id:
                 task = tasks_info.get(helper.task_id, None)
                 if task:
-                    worker.task.running_time = (
-                        int((datetime.utcnow() - task.started).total_seconds() * 1000)
-                        if task.started
-                        else 0
-                    )
+                    worker.task.running_time = (task.active_duration or 0) * 1000
                     worker.task.last_iteration = task.last_iteration
 
             update_queue_entries(worker.queue)
@@ -356,10 +365,15 @@ class WorkerBLL:
         self, company: str, user: str = "*", worker_id: str = "*"
     ) -> Sequence[WorkerEntry]:
         """Get worker entries matching the company and user, worker patterns"""
+        entries = []
         match = self._get_worker_key(company, user, worker_id)
         with TimingContext("redis", "workers_get_all"):
-            res = self.redis.scan_iter(match)
-        return [WorkerEntry.from_json(self.redis.get(r)) for r in res]
+            for r in self.redis.scan_iter(match):
+                data = self.redis.get(r)
+                if data:
+                    entries.append(WorkerEntry.from_json(data))
+
+        return entries
 
     @staticmethod
     def _get_es_index_suffix():

apiserver/bll/workers/stats.py

@@ -20,7 +20,7 @@ class WorkerStats:
     @staticmethod
     def worker_stats_prefix_for_company(company_id: str) -> str:
         """Returns the es index prefix for the company"""
-        return f"worker_stats_{company_id}_"
+        return f"worker_stats_{company_id.lower()}_"
 
     def _search_company_stats(self, company_id: str, es_req: dict) -> dict:
         return self.es.search(

apiserver/config/basic.py

@@ -19,7 +19,7 @@ from pyparsing import (
 
 from apiserver.utilities import json
 
-EXTRA_CONFIG_PATHS = ("/opt/clearml/config",)
+EXTRA_CONFIG_PATHS = ("/opt/trains/config", "/opt/clearml/config")
 DEFAULT_PREFIXES = ("clearml", "trains")
 EXTRA_CONFIG_PATH_SEP = ":" if platform.system() != "Windows" else ";"
 
@@ -79,6 +79,8 @@ class BasicConfig:
     def logger(self, name: str) -> logging.Logger:
         if Path(name).is_file():
             name = Path(name).stem
+            if name == "__init__" and Path(name).parent.stem:
+                name = Path(name).parent.stem
         path = ".".join((self.prefix, name))
         return logging.getLogger(path)
 

apiserver/config/default/apiserver.conf

@@ -3,7 +3,7 @@
     debug: false            # Debug mode
     pretty_json: false      # prettify json response
     return_stack: true      # return stack trace on error
-    log_calls: true         # Log API Calls
+    return_stack_to_caller: true # top-level control on whether to return stack trace in an API response
 
     # if 'return_stack' is true and error contains a status code, return stack trace only for these status codes
     # valid values are:
@@ -79,6 +79,11 @@
             max_age: 99999999999
         }
 
+        # provide a cookie domain override per company
+#        cookies_domain_override {
+#            <company-id>: <domain>
+#        }
+
 #        # A list of fixed users
 #        # Note: password may be bcrypt-hashed (generate using `python -c 'import bcrypt; print(bcrypt.hashpw("password", bcrypt.gensalt()))'`)
 #        fixed_users {
@@ -107,6 +112,8 @@
     workers {
         # Auto-register unknown workers on status reports and other calls
         auto_register: true
+        # Assume unknow workers have unregistered (i.e. do not raise unregistered error)
+        auto_unregister: true
         # Timeout in seconds on task status update. If exceeded
         # then task can be stopped without communicating to the worker
         task_update_timeout: 600
@@ -139,4 +146,11 @@
         max_backoff_sec: 5
     }
 
+    getting_started_info {
+        "agentName": "clearml",
+        "configure": "clearml-init",
+        "install": "pip install clearml",
+        "packageName": "clearml"
+    }
+
 }

apiserver/config/default/services/_mongo.conf

@@ -0,0 +1,4 @@
+max_page_size: 500
+
+# expiration time in seconds for the redis scroll states in get_many family of apis
+scroll_state_expiration_seconds: 600

apiserver/config/default/services/events.conf

@@ -12,11 +12,26 @@ events_retrieval {
     # should not exceed the amount of concurrent connections set in the ES driver
     max_metrics_concurrency: 4
 
+    # If set then max_metrics_count and max_variants_count are calculated dynamically on user data
+    dynamic_metrics_count: true
+
+    # The percentage from the ES aggs limit (10000) to use for the max_metrics and max_variants calculation
+    dynamic_metrics_count_threshold: 80
+
     # the max amount of metrics to aggregate on
     max_metrics_count: 100
 
     # the max amount of variants to aggregate on
     max_variants_count: 100
+
+    debug_images {
+        # Allow to return the debug images for the variants with uninitialized valid iterations border
+        allow_uninitialized_variants: true
+    }
+
+    max_raw_scalars_size: 200000
+
+    scroll_id_key: "cTN5VEtWEC6QrHvUl0FTx9kNyO0CcCK1p57akxma"
 }
 
 # if set then plot str will be checked for the valid json on plot add

apiserver/config/default/services/models.conf

@@ -0,0 +1,7 @@
+metadata_values {
+    # maximal amount of distinct model values to retrieve
+    max_count: 100
+
+    # cache ttl sec
+    cache_ttl_sec: 86400
+}

apiserver/config/default/services/queues.conf

@@ -0,0 +1,5 @@
+{
+    metrics_before_from_date: 3600
+    # interval in seconds to update queue metrics. Put 0 to disable
+    metrics_refresh_interval_sec: 300
+}

apiserver/database/__init__.py

@@ -28,6 +28,8 @@ OVERRIDE_PORT_ENV_KEY = (
     "MONGODB_SERVICE_PORT",
 )
 
+OVERRIDE_CONNECTION_STRING_ENV_KEY = "CLEARML_MONGODB_SERVICE_CONNECTION_STRING"
+
 
 class DatabaseEntry(models.Base):
     host = StringField(required=True)
@@ -47,13 +49,17 @@ class DatabaseFactory:
         missing = []
         log.info("Initializing database connections")
 
+        override_connection_string = getenv(OVERRIDE_CONNECTION_STRING_ENV_KEY)
         override_hostname = first(map(getenv, OVERRIDE_HOST_ENV_KEY), None)
-        if override_hostname:
-            log.info(f"Using override mongodb host {override_hostname}")
-
         override_port = first(map(getenv, OVERRIDE_PORT_ENV_KEY), None)
-        if override_port:
-            log.info(f"Using override mongodb port {override_port}")
+
+        if override_connection_string:
+            log.info(f"Using override mongodb connection string {override_connection_string}")
+        else:
+            if override_hostname:
+                log.info(f"Using override mongodb host {override_hostname}")
+            if override_port:
+                log.info(f"Using override mongodb port {override_port}")
 
         for key, alias in get_items(Database).items():
             if key not in db_entries:
@@ -62,11 +68,13 @@ class DatabaseFactory:
 
             entry = cls._create_db_entry(alias=alias, settings=db_entries.get(key))
 
-            if override_hostname:
-                entry.host = furl(entry.host).set(host=override_hostname).url
-
-            if override_port:
-                entry.host = furl(entry.host).set(port=override_port).url
+            if override_connection_string:
+                entry.host = override_connection_string
+            else:
+                if override_hostname:
+                    entry.host = furl(entry.host).set(host=override_hostname).url
+                if override_port:
+                    entry.host = furl(entry.host).set(port=override_port).url
 
             try:
                 entry.validate()

apiserver/database/errors.py

@@ -166,7 +166,10 @@ class MongoEngineErrorsHandler(object):
     @classmethod
     @throws_default_error(errors.server_error.InternalError)
     def invalid_query_error(cls, e, message, **_):
-        pass
+        if e.args:
+            inner = e.args[0]
+            if isinstance(inner, LookUpError):
+                cls.lookup_error(inner, message)
 
 
 @contextmanager

apiserver/database/fields.py

@@ -176,6 +176,13 @@ class SafeMapField(MapField, DictValidationMixin):
             self.error("Empty keys are not allowed in a MapField")
 
 
+class NullableStringField(StringField):
+    def validate(self, value):
+        if value is None:
+            return
+        super(NullableStringField, self).validate(value)
+
+
 class SafeDictField(DictField, DictValidationMixin):
     def validate(self, value):
         self._safe_validate(value)

apiserver/database/model/__init__.py

@@ -60,3 +60,4 @@ def validate_id(cls, company, **kwargs):
 class EntityVisibility(Enum):
     active = "active"
     archived = "archived"
+    hidden = "hidden"

apiserver/database/model/auth.py

@@ -48,7 +48,9 @@ class Credentials(EmbeddedDocument):
     meta = {"strict": False}
     key = StringField(required=True)
     secret = StringField(required=True)
+    label = StringField()
     last_used = DateTimeField()
+    last_used_from = StringField()
 
 
 class User(DbModelMixin, AuthDocument):

apiserver/database/model/base.py

@@ -1,26 +1,41 @@
 import re
 from collections import namedtuple
-from functools import reduce
-from typing import Collection, Sequence, Union, Optional, Type, Tuple, Mapping, Any
+from functools import reduce, partial
+from typing import (
+    Collection,
+    Sequence,
+    Union,
+    Optional,
+    Type,
+    Tuple,
+    Mapping,
+    Any,
+    Callable,
+    Dict,
+    List,
+)
 
-from boltons.iterutils import first, bucketize, partition
+from boltons.iterutils import first, partition
 from dateutil.parser import parse as parse_datetime
-from mongoengine import Q, Document, ListField, StringField
+from mongoengine import Q, Document, ListField, StringField, IntField
 from pymongo.command_cursor import CommandCursor
 
 from apiserver.apierrors import errors
 from apiserver.apierrors.base import BaseError
+from apiserver.bll.redis_cache_manager import RedisCacheManager
 from apiserver.config_repo import config
+from apiserver.database import Database
 from apiserver.database.errors import MakeGetAllQueryError
 from apiserver.database.projection import project_dict, ProjectionHelper
 from apiserver.database.props import PropsMixin
-from apiserver.database.query import RegexQ, RegexWrapper
+from apiserver.database.query import RegexQ, RegexWrapper, RegexQCombination
 from apiserver.database.utils import (
     get_company_or_none_constraint,
     get_fields_choices,
     field_does_not_exist,
     field_exists,
 )
+from apiserver.redis_manager import redman
 
 log = config.logger("dbmodel")
 
@@ -70,6 +85,9 @@ class GetMixin(PropsMixin):
     _ordering_key = "order_by"
     _search_text_key = "search_text"
 
+    _start_key = "start"
+    _size_key = "size"
+
     _multi_field_param_sep = "__"
     _multi_field_param_prefix = {
         ("_any_", "_or_"): lambda a, b: a | b,
@@ -77,6 +95,7 @@ class GetMixin(PropsMixin):
     }
     MultiFieldParameters = namedtuple("MultiFieldParameters", "pattern fields")
 
+    _numeric_locale = {"locale": "en_US", "numericOrdering": True}
     _field_collation_overrides = {}
 
     class QueryParameterOptions(object):
@@ -103,45 +122,106 @@ class GetMixin(PropsMixin):
 
     class ListFieldBucketHelper:
         op_prefix = "__$"
-        legacy_exclude_prefix = "-"
+        _legacy_exclude_prefix = "-"
+        _legacy_exclude_mongo_op = "nin"
 
-        _default = "in"
+        default_mongo_op = "in"
         _ops = {
+            # op -> (mongo_op, sticky)
             "not": ("nin", False),
+            "nop": (default_mongo_op, False),
             "all": ("all", True),
             "and": ("all", True),
+            "any": (default_mongo_op, True),
+            "or": (default_mongo_op, True),
         }
-        _next = _default
-        _sticky = False
 
         def __init__(self, legacy=False):
-            self._legacy = legacy
+            self._current_op = None
+            self._sticky = False
+            self._support_legacy = legacy
+            self.allow_empty = False
 
-        def key(self, v):
+        def _get_op(self, v: str, translate: bool = False) -> Optional[str]:
+            op = (
+                v[len(self.op_prefix) :] if v and v.startswith(self.op_prefix) else None
+            )
+            if translate:
+                tup = self._ops.get(op, None)
+                return tup[0] if tup else None
+            return op
+
+        def _key(self, v) -> Optional[Union[str, bool]]:
             if v is None:
-                self._next = self._default
-                return self._default
-            elif self._legacy and v.startswith(self.legacy_exclude_prefix):
-                self._next = self._default
-                return self._ops["not"][0]
-            elif v.startswith(self.op_prefix):
-                self._next, self._sticky = self._ops.get(
-                    v[len(self.op_prefix) :], (self._default, self._sticky)
-                )
+                self.allow_empty = True
                 return None
 
-            next_ = self._next
-            if not self._sticky:
-                self._next = self._default
-            return next_
+            op = self._get_op(v)
+            if op is not None:
+                # operator - set state and return None
+                self._current_op, self._sticky = self._ops.get(
+                    op, (self.default_mongo_op, self._sticky)
+                )
+                return None
+            elif self._current_op:
+                current_op = self._current_op
+                if not self._sticky:
+                    self._current_op = None
+                return current_op
+            elif self._support_legacy and v.startswith(self._legacy_exclude_prefix):
+                self._current_op = None
+                return False
 
-        def value_transform(self, v):
-            if self._legacy and v and v.startswith(self.legacy_exclude_prefix):
-                return v[len(self.legacy_exclude_prefix) :]
-            return v
+            return self.default_mongo_op
+
+        def get_global_op(self, data: Sequence[str]) -> int:
+            op_to_res = {
+                "in": Q.OR,
+                "all": Q.AND,
+            }
+            data = (x for x in data if x is not None)
+            first_op = (
+                self._get_op(next(data, ""), translate=True) or self.default_mongo_op
+            )
+            return op_to_res.get(first_op, self.default_mongo_op)
+
+        def get_actions(self, data: Sequence[str]) -> Dict[str, List[Union[str, None]]]:
+            actions = {}
+
+            for val in data:
+                key = self._key(val)
+                if key is None:
+                    continue
+                elif self._support_legacy and key is False:
+                    key = self._legacy_exclude_mongo_op
+                    val = val[len(self._legacy_exclude_prefix) :]
+                actions.setdefault(key, []).append(val)
+
+            return actions
 
     get_all_query_options = QueryParameterOptions()
 
+    class GetManyScrollState(ProperDictMixin, Document):
+        meta = {"db_alias": Database.backend, "strict": False}
+
+        id = StringField(primary_key=True)
+        position = IntField(default=0)
+
+    _cache_manager = None
+
+    @classmethod
+    def get_cache_manager(cls):
+        if not cls._cache_manager:
+            cls._cache_manager = RedisCacheManager(
+                state_class=cls.GetManyScrollState,
+                redis=redman.connection("apiserver"),
+                expiration_interval=config.get(
+                    "services._mongo.scroll_state_expiration_seconds", 600
+                ),
+            )
+
+        return cls._cache_manager
+
     @classmethod
     def get(
         cls: Union["GetMixin", Document],
@@ -240,7 +320,9 @@ class GetMixin(PropsMixin):
         Prepare a query object based on the provided query dictionary and various fields.
 
         NOTE: BE VERY CAREFUL WITH THIS CALL, as it allows creating queries that span across companies.
-
+        IMPLEMENTATION NOTE: Make sure that inside this function or the functions it depends on RegexQ is always
+        used instead of Q. Otherwise we can and up with some combination that is not processed according to
+        RegexQ rules
         :param parameters_options: Specifies options for parsing the parameters (see ParametersOptions)
         :param parameters: Query dictionary (relevant keys are these specified by the various field names parameters).
             Supported parameters:
@@ -273,10 +355,13 @@ class GetMixin(PropsMixin):
             ).items():
                 query &= cls.get_range_field_query(field, data)
 
-            for field in opts.fields or []:
-                data = parameters.pop(field, None)
-                if data is not None:
-                    dict_query[field] = data
+            for field, data in cls._pop_matching_params(
+                patterns=opts.fields or [], parameters=parameters
+            ).items():
+                if "._" in field or "_." in field:
+                    query &= RegexQ(__raw__={field: data})
+                else:
+                    dict_query[field.replace(".", "__")] = data
 
             for field in opts.datetime_fields or []:
                 data = parameters.pop(field, None)
@@ -308,22 +393,31 @@ class GetMixin(PropsMixin):
                         break
                     if any("._" in f for f in data.fields):
                         q = reduce(
-                            lambda a, x: func(a, Q(__raw__={x: {"$regex": data.pattern, "$options": "i"}})),
+                            lambda a, x: func(
+                                a,
+                                RegexQ(
+                                    __raw__={
+                                        x: {"$regex": data.pattern, "$options": "i"}
+                                    }
+                                ),
+                            ),
                             data.fields,
-                            Q()
+                            RegexQ(),
                         )
                     else:
                         regex = RegexWrapper(data.pattern, flags=re.IGNORECASE)
                         sep_fields = [f.replace(".", "__") for f in data.fields]
                         q = reduce(
-                            lambda a, x: func(a, RegexQ(**{x: regex})), sep_fields, RegexQ()
+                            lambda a, x: func(a, RegexQ(**{x: regex})),
+                            sep_fields,
+                            RegexQ(),
                         )
                     query = query & q
 
         return query & RegexQ(**dict_query)
 
     @classmethod
-    def get_range_field_query(cls, field: str, data: Sequence[Optional[str]]) -> Q:
+    def get_range_field_query(cls, field: str, data: Sequence[Optional[str]]) -> RegexQ:
         """
         Return a range query for the provided field. The data should contain min and max values
         Both intervals are included. For open range queries either min or max can be None
@@ -347,14 +441,14 @@ class GetMixin(PropsMixin):
         if max_val is not None:
             query[f"{mongoengine_field}__lte"] = max_val
 
-        q = Q(**query)
+        q = RegexQ(**query)
         if min_val is None:
-            q |= Q(**{mongoengine_field: None})
+            q |= RegexQ(**{mongoengine_field: None})
 
         return q
 
     @classmethod
-    def get_list_field_query(cls, field: str, data: Sequence[Optional[str]]) -> Q:
+    def get_list_field_query(cls, field: str, data: Sequence[Optional[str]]) -> RegexQ:
         """
         Get a proper mongoengine Q object that represents an "or" query for the provided values
         with respect to the given list field, with support for "none of empty" in case a None value
@@ -366,30 +460,31 @@ class GetMixin(PropsMixin):
         """
         if not isinstance(data, (list, tuple)):
             data = [data]
-            # raise MakeGetAllQueryError("expected list", field)
 
-        # TODO: backwards compatibility only for older API versions
         helper = cls.ListFieldBucketHelper(legacy=True)
-        actions = bucketize(
-            data, key=helper.key, value_transform=helper.value_transform
-        )
+        global_op = helper.get_global_op(data)
+        actions = helper.get_actions(data)
 
-        allow_empty = None in actions.get("in", {})
         mongoengine_field = field.replace(".", "__")
 
-        q = RegexQ()
-        for action in filter(None, actions):
-            q &= RegexQ(
-                **{f"{mongoengine_field}__{action}": list(set(actions[action]))}
-            )
+        queries = [
+            RegexQ(**{f"{mongoengine_field}__{action}": list(set(actions[action]))})
+            for action in filter(None, actions)
+        ]
 
-        if not allow_empty:
+        if not queries:
+            q = RegexQ()
+        else:
+            q = RegexQCombination(operation=global_op, children=queries)
+
+        if not helper.allow_empty:
             return q
 
         return (
             q
-            | Q(**{f"{mongoengine_field}__exists": False})
-            | Q(**{mongoengine_field: []})
+            | RegexQ(**{f"{mongoengine_field}__exists": False})
+            | RegexQ(**{mongoengine_field: []})
+            | RegexQ(**{mongoengine_field: None})
         )
 
     @classmethod
@@ -417,27 +512,41 @@ class GetMixin(PropsMixin):
         return order_by
 
     @classmethod
-    def validate_paging(
-        cls, parameters=None, default_page=None, default_page_size=None
-    ):
-        """ Validate and extract paging info from from the provided dictionary. Supports default values. """
-        if parameters is None:
-            parameters = {}
-        default_page = parameters.get("page", default_page)
-        if default_page is None:
-            return None, None
-        default_page_size = parameters.get("page_size", default_page_size)
-        if not default_page_size:
-            raise errors.bad_request.MissingRequiredFields(
-                "page_size is required when page is requested", field="page_size"
-            )
-        elif default_page < 0:
+    def validate_paging(cls, parameters=None, default_page=0, default_page_size=None):
+        """
+        Validate and extract paging info from from the provided dictionary. Supports default values.
+        If page is specified then it should be non-negative, if page size is specified then it should be positive
+        If page size is specified and page is not then 0 page is assumed
+        If page is specified then page size should be specified too
+        """
+        parameters = parameters or {}
+
+        start = parameters.get(cls._start_key)
+        if start is not None:
+            return start, cls.validate_scroll_size(parameters)
+
+        max_page_size = config.get("services._mongo.max_page_size", 500)
+        page = parameters.get("page", default_page)
+        if page is not None and page < 0:
             raise errors.bad_request.ValidationError("page must be >=0", field="page")
-        elif default_page_size < 1:
+
+        page_size = parameters.get("page_size", default_page_size or max_page_size)
+        if page_size is not None and page_size < 1:
             raise errors.bad_request.ValidationError(
                 "page_size must be >0", field="page_size"
             )
-        return default_page, default_page_size
+
+        if page_size is not None:
+            page = page or 0
+            page_size = min(page_size, max_page_size)
+            return page * page_size, page_size
+
+        if page is not None:
+            raise errors.bad_request.MissingRequiredFields(
+                "page_size is required when page is requested", field="page_size"
+            )
+
+        return None, None
 
     @classmethod
     def get_projection(cls, parameters, override_projection=None, **__):
@@ -481,6 +590,54 @@ class GetMixin(PropsMixin):
     def set_default_ordering(cls, parameters: dict, value: Sequence[str]) -> None:
         cls.set_ordering(parameters, cls.get_ordering(parameters) or value)
 
+    @classmethod
+    def validate_scroll_size(cls, query_dict: dict) -> int:
+        size = query_dict.get(cls._size_key)
+        if not size or not isinstance(size, int) or size < 1:
+            raise errors.bad_request.ValidationError(
+                "Integer size parameter greater than 1 should be provided when working with scroll"
+            )
+        return size
+
+    @classmethod
+    def get_data_with_scroll_support(
+        cls,
+        query_dict: dict,
+        data_getter: Callable[[], Sequence[dict]],
+        ret_params: dict,
+    ) -> Sequence[dict]:
+        """
+        Retrieves the data by calling the provided data_getter api
+        If scroll parameters are specified then put the query_dict 'start' parameter to the last
+        scroll position and continue retrievals from that position
+        If refresh_scroll is requested then bring once more the data from the beginning
+        till the current scroll position
+        In the end the scroll position is updated and accumulated frames are returned
+        """
+        query_dict = query_dict or {}
+        state: Optional[cls.GetManyScrollState] = None
+        if "scroll_id" in query_dict:
+            size = cls.validate_scroll_size(query_dict)
+            state = cls.get_cache_manager().get_or_create_state_core(
+                query_dict.get("scroll_id")
+            )
+            if query_dict.get("refresh_scroll"):
+                query_dict[cls._size_key] = max(state.position, size)
+                state.position = 0
+            query_dict[cls._start_key] = state.position
+
+        data = data_getter()
+        if cls._start_key in query_dict:
+            query_dict[cls._start_key] = query_dict[cls._start_key] + len(data)
+
+        if state:
+            state.position = query_dict[cls._start_key]
+            cls.get_cache_manager().set_state(state)
+            if ret_params is not None:
+                ret_params["scroll_id"] = state.id
+
+        return data
+
     @classmethod
     def get_many_with_join(
         cls,
@@ -491,6 +648,8 @@ class GetMixin(PropsMixin):
         allow_public=False,
         override_projection=None,
         expand_reference_ids=True,
+        projection_fields: dict = None,
+        ret_params: dict = None,
     ):
         """
         Fetch all documents matching a provided query with support for joining referenced documents according to the
@@ -526,6 +685,8 @@ class GetMixin(PropsMixin):
             query=query,
             query_options=query_options,
             allow_public=allow_public,
+            projection_fields=projection_fields,
+            ret_params=ret_params,
         )
 
         def projection_func(doc_type, projection, ids):
@@ -545,6 +706,45 @@ class GetMixin(PropsMixin):
             v for k, v in cls._field_collation_overrides.items() if field.startswith(k)
         )
 
+    @classmethod
+    def get_count(
+        cls: Union["GetMixin", Document],
+        company,
+        query_dict: dict = None,
+        query_options: QueryParameterOptions = None,
+        query: Q = None,
+        allow_public=False,
+    ) -> int:
+        _query = cls._get_combined_query(
+            company=company,
+            query_dict=query_dict,
+            query_options=query_options,
+            query=query,
+            allow_public=allow_public,
+        )
+        return cls.objects(_query).count()
+
+    @classmethod
+    def _get_combined_query(
+        cls,
+        company,
+        query_dict: dict = None,
+        query_options: QueryParameterOptions = None,
+        query: Q = None,
+        allow_public=False,
+    ) -> Q:
+        if query_dict is not None:
+            q = cls.prepare_query(
+                parameters=query_dict,
+                company=company,
+                parameters_options=query_options,
+                allow_public=allow_public,
+            )
+        else:
+            q = cls._prepare_perm_query(company, allow_public=allow_public)
+
+        return (q & query) if query else q
+
     @classmethod
     def get_many(
         cls,
@@ -556,6 +756,8 @@ class GetMixin(PropsMixin):
         allow_public=False,
         override_projection: Collection[str] = None,
         return_dicts=True,
+        projection_fields: dict = None,
+        ret_params: dict = None,
     ):
         """
         Fetch all documents matching a provided query. Supported several built-in options
@@ -589,23 +791,25 @@ class GetMixin(PropsMixin):
                 if override_collation:
                     break
 
-        if query_dict is not None:
-            q = cls.prepare_query(
-                parameters=query_dict,
-                company=company,
-                parameters_options=query_options,
-                allow_public=allow_public,
-            )
-        else:
-            q = cls._prepare_perm_query(company, allow_public=allow_public)
-        _query = (q & query) if query else q
+        _query = cls._get_combined_query(
+            company=company,
+            query_dict=query_dict,
+            query_options=query_options,
+            query=query,
+            allow_public=allow_public,
+        )
 
         if return_dicts:
-            return cls._get_many_override_none_ordering(
+            data_getter = partial(
+                cls._get_many_override_none_ordering,
                 query=_query,
                 parameters=parameters,
                 override_projection=override_projection,
                 override_collation=override_collation,
+                projection_fields=projection_fields,
+            )
+            return cls.get_data_with_scroll_support(
+                query_dict=query_dict, data_getter=data_getter, ret_params=ret_params,
             )
 
         return cls._get_many_no_company(
@@ -613,6 +817,7 @@ class GetMixin(PropsMixin):
             parameters=parameters,
             override_projection=override_projection,
             override_collation=override_collation,
+            projection_fields=projection_fields,
         )
 
     @classmethod
@@ -637,6 +842,7 @@ class GetMixin(PropsMixin):
         parameters=None,
         override_projection=None,
         override_collation=None,
+        projection_fields: dict = None,
     ):
         """
         Fetch all documents matching a provided query.
@@ -658,7 +864,7 @@ class GetMixin(PropsMixin):
         order_by = cls.validate_order_by(parameters=parameters, search_text=search_text)
         if order_by and not override_collation:
             override_collation = cls._get_collation_override(order_by[0])
-        page, page_size = cls.validate_paging(parameters=parameters)
+        start, size = cls.validate_paging(parameters=parameters)
         include, exclude = cls.split_projection(
             cls.get_projection(parameters, override_projection)
         )
@@ -679,9 +885,12 @@ class GetMixin(PropsMixin):
         if exclude:
             qs = qs.exclude(*exclude)
 
-        if page is not None and page_size:
+        if projection_fields:
+            qs = qs.fields(**projection_fields)
+
+        if start is not None and size:
             # add paging
-            qs = qs.skip(page * page_size).limit(page_size)
+            qs = qs.skip(start).limit(size)
 
         return qs
 
@@ -720,6 +929,7 @@ class GetMixin(PropsMixin):
         parameters: dict = None,
         override_projection: Collection[str] = None,
         override_collation: dict = None,
+        projection_fields: dict = None,
     ) -> Sequence[dict]:
         """
         Fetch all documents matching a provided query. For the first order by field
@@ -742,7 +952,10 @@ class GetMixin(PropsMixin):
         parameters = parameters or {}
         search_text = parameters.get(cls._search_text_key)
         order_by = cls.validate_order_by(parameters=parameters, search_text=search_text)
-        page, page_size = cls.validate_paging(parameters=parameters)
+        start, size = cls.validate_paging(parameters=parameters)
+        if size is not None and size <= 0:
+            return []
+
         include, exclude = cls.split_projection(
             cls.get_projection(parameters, override_projection)
         )
@@ -774,25 +987,31 @@ class GetMixin(PropsMixin):
         if exclude:
             query_sets = [qs.exclude(*exclude) for qs in query_sets]
 
-        if page is None or not page_size:
+        if projection_fields:
+            query_sets = [qs.fields(**projection_fields) for qs in query_sets]
+
+        if start is None or not size:
             return [obj.to_proper_dict(only=include) for qs in query_sets for obj in qs]
 
         # add paging
         ret = []
-        start = page * page_size
-        for qs in query_sets:
-            qs_size = qs.count()
-            if qs_size < start:
-                start -= qs_size
-                continue
+        last_set = len(query_sets) - 1
+        for i, qs in enumerate(query_sets):
+            last_size = len(ret)
             ret.extend(
                 obj.to_proper_dict(only=include)
-                for obj in qs.skip(start).limit(page_size)
+                for obj in (qs.skip(start) if start else qs).limit(size)
             )
-            if len(ret) >= page_size:
+            added = len(ret) - last_size
+
+            if added > 0:
+                start = 0
+                size = max(0, size - added)
+            elif i != last_set:
+                start -= min(start, qs.count())
+
+            if size <= 0:
                 break
-            start = 0
-            page_size -= len(ret)
 
         return ret
 

apiserver/database/model/model.py

@@ -1,11 +1,8 @@
-from typing import Sequence
-
 from mongoengine import (
-    Document,
     StringField,
     DateTimeField,
     BooleanField,
-    EmbeddedDocumentListField,
+    EmbeddedDocumentField,
 )
 
 from apiserver.database import Database, strict
@@ -13,18 +10,21 @@ from apiserver.database.fields import (
     StrippedStringField,
     SafeDictField,
     SafeSortedListField,
+    SafeMapField,
 )
-from apiserver.database.model import DbModelMixin
+from apiserver.database.model import AttributedDocument
 from apiserver.database.model.base import GetMixin
 from apiserver.database.model.metadata import MetadataItem
 from apiserver.database.model.model_labels import ModelLabels
-from apiserver.database.model.company import Company
 from apiserver.database.model.project import Project
 from apiserver.database.model.task.task import Task
-from apiserver.database.model.user import User
 
 
-class Model(DbModelMixin, Document):
+class Model(AttributedDocument):
+    _field_collation_overrides = {
+        "metadata.": AttributedDocument._numeric_locale,
+    }
+
     meta = {
         "db_alias": Database.backend,
         "strict": strict,
@@ -33,8 +33,6 @@ class Model(DbModelMixin, Document):
             "project",
             "task",
             "last_update",
-            "metadata.key",
-            "metadata.type",
             ("company", "framework"),
             ("company", "name"),
             ("company", "user"),
@@ -66,6 +64,7 @@ class Model(DbModelMixin, Document):
             "project",
             "task",
             "parent",
+            "metadata.*",
         ),
         datetime_fields=("last_update",),
     )
@@ -73,8 +72,6 @@ class Model(DbModelMixin, Document):
     id = StringField(primary_key=True)
     name = StrippedStringField(user_set_allowed=True, min_length=3)
     parent = StringField(reference_field="Model", required=False)
-    user = StringField(required=True, reference_field=User)
-    company = StringField(required=True, reference_field=Company)
     project = StringField(reference_field=Project, user_set_allowed=True)
     created = DateTimeField(required=True, user_set_allowed=True)
     task = StringField(reference_field=Task)
@@ -91,6 +88,6 @@ class Model(DbModelMixin, Document):
         default=dict, user_set_allowed=True, exclude_by_default=True
     )
     company_origin = StringField(exclude_by_default=True)
-    metadata: Sequence[MetadataItem] = EmbeddedDocumentListField(
-        MetadataItem, default=list, user_set_allowed=True
+    metadata = SafeMapField(
+        field=EmbeddedDocumentField(MetadataItem), user_set_allowed=True
     )

apiserver/database/model/project.py

@@ -9,8 +9,9 @@ from apiserver.database.model.base import GetMixin
 class Project(AttributedDocument):
 
     get_all_query_options = GetMixin.QueryParameterOptions(
-        pattern_fields=("name", "description"),
+        pattern_fields=("name", "basename", "description"),
         list_fields=("tags", "system_tags", "id", "parent", "path"),
+        range_fields=("last_update",),
     )
 
     meta = {
@@ -20,6 +21,7 @@ class Project(AttributedDocument):
             "parent",
             "path",
             ("company", "name"),
+            ("company", "basename"),
             {
                 "name": "%s.project.main_text_index" % Database.backend,
                 "fields": ["$name", "$id", "$description"],
@@ -36,6 +38,7 @@ class Project(AttributedDocument):
         min_length=3,
         sparse=True,
     )
+    basename = StrippedStringField(required=True)
     description = StringField()
     created = DateTimeField(required=True)
     tags = SafeSortedListField(StringField(required=True))

apiserver/database/model/queue.py

@@ -1,16 +1,19 @@
-from typing import Sequence
-
 from mongoengine import (
     Document,
     EmbeddedDocument,
     StringField,
     DateTimeField,
     EmbeddedDocumentListField,
+    EmbeddedDocumentField,
 )
 
 from apiserver.database import Database, strict
-from apiserver.database.fields import StrippedStringField, SafeSortedListField
-from apiserver.database.model import DbModelMixin
+from apiserver.database.fields import (
+    StrippedStringField,
+    SafeSortedListField,
+    SafeMapField,
+)
+from apiserver.database.model import DbModelMixin, AttributedDocument
 from apiserver.database.model.base import ProperDictMixin, GetMixin
 from apiserver.database.model.company import Company
 from apiserver.database.model.metadata import MetadataItem
@@ -19,23 +22,25 @@ from apiserver.database.model.task.task import Task
 
 class Entry(EmbeddedDocument, ProperDictMixin):
     """ Entry representing a task waiting in the queue """
+
     task = StringField(required=True, reference_field=Task)
-    ''' Task ID '''
+    """ Task ID """
     added = DateTimeField(required=True)
-    ''' Added to the queue '''
+    """ Added to the queue """
 
 
 class Queue(DbModelMixin, Document):
+    _field_collation_overrides = {
+        "metadata.": AttributedDocument._numeric_locale,
+    }
 
     get_all_query_options = GetMixin.QueryParameterOptions(
-        pattern_fields=("name",),
-        list_fields=("tags", "system_tags", "id"),
+        pattern_fields=("name",), list_fields=("tags", "system_tags", "id", "metadata.*"),
     )
 
     meta = {
-        'db_alias': Database.backend,
-        'strict': strict,
-        "indexes": ["metadata.key", "metadata.type"],
+        "db_alias": Database.backend,
+        "strict": strict,
     }
 
     id = StringField(primary_key=True)
@@ -44,10 +49,12 @@ class Queue(DbModelMixin, Document):
     )
     company = StringField(required=True, reference_field=Company)
     created = DateTimeField(required=True)
-    tags = SafeSortedListField(StringField(required=True), default=list, user_set_allowed=True)
+    tags = SafeSortedListField(
+        StringField(required=True), default=list, user_set_allowed=True
+    )
     system_tags = SafeSortedListField(StringField(required=True), user_set_allowed=True)
     entries = EmbeddedDocumentListField(Entry, default=list)
     last_update = DateTimeField()
-    metadata: Sequence[MetadataItem] = EmbeddedDocumentListField(
-        MetadataItem, default=list, user_set_allowed=True
+    metadata = SafeMapField(
+        field=EmbeddedDocumentField(MetadataItem), user_set_allowed=True
     )

apiserver/database/model/task/task.py

@@ -18,6 +18,7 @@ from apiserver.database.fields import (
     UnionField,
     SafeSortedListField,
     EmbeddedDocumentListField,
+    NullableStringField,
 )
 from apiserver.database.model import AttributedDocument
 from apiserver.database.model.base import ProperDictMixin, GetMixin
@@ -158,11 +159,10 @@ external_task_types = set(get_options(TaskType))
 
 
 class Task(AttributedDocument):
-    _numeric_locale = {"locale": "en_US", "numericOrdering": True}
     _field_collation_overrides = {
-        "execution.parameters.": _numeric_locale,
-        "last_metrics.": _numeric_locale,
-        "hyperparams.": _numeric_locale,
+        "execution.parameters.": AttributedDocument._numeric_locale,
+        "last_metrics.": AttributedDocument._numeric_locale,
+        "hyperparams.": AttributedDocument._numeric_locale,
     }
 
     meta = {
@@ -175,6 +175,8 @@ class Task(AttributedDocument):
             "active_duration",
             "parent",
             "project",
+            "last_update",
+            "status_changed",
             "models.input.model",
             ("company", "name"),
             ("company", "user"),
@@ -183,7 +185,10 @@ class Task(AttributedDocument):
             ("company", "type", "system_tags", "status"),
             ("company", "project", "type", "system_tags", "status"),
             ("status", "last_update"),  # for maintenance tasks
-            {"fields": ["company", "project"], "collation": _numeric_locale},
+            {
+                "fields": ["company", "project"],
+                "collation": AttributedDocument._numeric_locale,
+            },
             {
                 "name": "%s.task.main_text_index" % Database.backend,
                 "fields": [
@@ -218,6 +223,7 @@ class Task(AttributedDocument):
             "status",
             "project",
             "parent",
+            "hyperparams.*",
         ),
         range_fields=("started", "active_duration", "last_metrics.*", "last_iteration"),
         datetime_fields=("status_changed", "last_update"),
@@ -232,7 +238,7 @@ class Task(AttributedDocument):
     type = StringField(required=True, choices=get_options(TaskType))
     status = StringField(default=TaskStatus.created, choices=get_options(TaskStatus))
     status_reason = StringField()
-    status_message = StringField()
+    status_message = StringField(user_set_allowed=True)
     status_changed = DateTimeField()
     comment = StringField(user_set_allowed=True)
     created = DateTimeField(required=True, user_set_allowed=True)
@@ -260,7 +266,7 @@ class Task(AttributedDocument):
     configuration = SafeMapField(field=EmbeddedDocumentField(ConfigurationItem))
     runtime = SafeDictField(default=dict)
     models: Models = EmbeddedDocumentField(Models, default=Models)
-    container = SafeMapField(field=StringField(default=""))
+    container = SafeMapField(field=NullableStringField())
     enqueue_status = StringField(
         choices=get_options(TaskStatus), exclude_by_default=True
     )

apiserver/database/projection.py

@@ -3,8 +3,6 @@ from concurrent.futures import ThreadPoolExecutor
 from itertools import groupby, chain
 from typing import Sequence, Dict, Callable, Tuple, Any, Type
 
-import dpath.path
-
 from apiserver.apierrors import errors
 from apiserver.database.props import PropsMixin
 
@@ -275,25 +273,26 @@ class ProjectionHelper(object):
         norm_path = doc_cls.get_dpath_translated_path(path)
         globlist = norm_path.strip(SEP).split(SEP)
 
-        obj_paths = self._cached_results_paths.get(id(obj))
-        if obj_paths is None:
-            obj_paths = self._cached_results_paths[id(obj)] = list(
-                dpath.path.paths(obj, dirs=True, skip=True)
-            )
-
-        paths = [p for p in obj_paths if dpath.path.match(p, globlist)]
-
-        def search_and_replace(p: Sequence[Tuple[str, Type]]) -> Any:
+        def _search_and_replace(target: dict, p: Sequence[str]) -> Sequence[str]:
             parent = None
-            target = obj
-            for part in p:
-                parent = target
-                target = target[part[0]]
-            if parent and factory:
-                parent[p[-1][0]] = factory(target)
-            return target
+            for idx, part in enumerate(p):
+                if isinstance(target, dict) and part in target:
+                    parent = target
+                    target = target[part]
+                elif isinstance(target, list) and part == "*":
+                    return list(
+                        chain.from_iterable(
+                            _search_and_replace(t, p[idx + 1 :]) for t in target
+                        )
+                    )
+                else:
+                    return []
 
-        return [search_and_replace(p) for p in paths]
+            if parent and factory:
+                parent[p[-1]] = factory(target)
+            return [target]
+
+        return _search_and_replace(obj, globlist)
 
     def project(self, results, projection_func):
         """

apiserver/database/props.py

@@ -4,7 +4,12 @@ from threading import Lock
 from typing import Sequence
 
 import six
-from mongoengine import EmbeddedDocumentField, EmbeddedDocumentListField
+from mongoengine import (
+    EmbeddedDocumentField,
+    EmbeddedDocumentListField,
+    EmbeddedDocument,
+    Document,
+)
 from mongoengine.base import get_document
 
 from apiserver.database.fields import (
@@ -25,6 +30,13 @@ class PropsMixin(object):
     __cached_dpath_computed_fields_lock = Lock()
     __cached_dpath_computed_fields = None
 
+    _document_classes = {}
+
+    def __init_subclass__(cls, **kwargs):
+        super().__init_subclass__(**kwargs)
+        if issubclass(cls, (Document, EmbeddedDocument)):
+            PropsMixin._document_classes[cls._class_name] = cls
+
     @classmethod
     def get_fields(cls):
         if cls.__cached_fields is None:
@@ -57,8 +69,14 @@ class PropsMixin(object):
         def resolve_doc(v):
             if not isinstance(v, six.string_types):
                 return v
-            if v == 'self':
+
+            if v == "self":
                 return cls_.owner_document
+
+            doc_cls = PropsMixin._document_classes.get(v)
+            if doc_cls:
+                return doc_cls
+
             return get_document(v)
 
         fields = {k: resolve_doc(v) for k, v in res.items()}
@@ -72,7 +90,7 @@ class PropsMixin(object):
                 ).document_type
                 fields.update(
                     {
-                        '.'.join((field, subfield)): doc
+                        ".".join((field, subfield)): doc
                         for subfield, doc in PropsMixin._get_fields_with_attr(
                             embedded_doc_cls, attr
                         ).items()
@@ -80,10 +98,10 @@ class PropsMixin(object):
                 )
 
         collect_embedded_docs(EmbeddedDocumentField, lambda x: x)
-        collect_embedded_docs(EmbeddedDocumentListField, attrgetter('field'))
-        collect_embedded_docs(LengthRangeEmbeddedDocumentListField, attrgetter('field'))
-        collect_embedded_docs(UniqueEmbeddedDocumentListField, attrgetter('field'))
-        collect_embedded_docs(EmbeddedDocumentSortedListField, attrgetter('field'))
+        collect_embedded_docs(EmbeddedDocumentListField, attrgetter("field"))
+        collect_embedded_docs(LengthRangeEmbeddedDocumentListField, attrgetter("field"))
+        collect_embedded_docs(UniqueEmbeddedDocumentListField, attrgetter("field"))
+        collect_embedded_docs(EmbeddedDocumentSortedListField, attrgetter("field"))
 
         return fields
 
@@ -94,7 +112,7 @@ class PropsMixin(object):
         for depth, part in enumerate(parts):
             if current_cls is None:
                 raise ValueError(
-                    'Invalid path (non-document encountered at %s)' % parts[: depth - 1]
+                    "Invalid path (non-document encountered at %s)" % parts[: depth - 1]
                 )
             try:
                 field_name, field = next(
@@ -103,7 +121,7 @@ class PropsMixin(object):
                     if k == part
                 )
             except StopIteration:
-                raise ValueError('Invalid field path %s' % parts[:depth])
+                raise ValueError("Invalid field path %s" % parts[:depth])
 
             translated_parts.append(part)
 
@@ -119,7 +137,7 @@ class PropsMixin(object):
                 ),
             ):
                 current_cls = field.field.document_type
-                translated_parts.append('*')
+                translated_parts.append("*")
             else:
                 current_cls = None
 
@@ -128,7 +146,7 @@ class PropsMixin(object):
     @classmethod
     def get_reference_fields(cls):
         if cls.__cached_reference_fields is None:
-            fields = cls._get_fields_with_attr(cls, 'reference_field')
+            fields = cls._get_fields_with_attr(cls, "reference_field")
             cls.__cached_reference_fields = OrderedDict(sorted(fields.items()))
         return cls.__cached_reference_fields
 
@@ -143,12 +161,12 @@ class PropsMixin(object):
     @classmethod
     def get_exclude_fields(cls):
         if cls.__cached_exclude_fields is None:
-            fields = cls._get_fields_with_attr(cls, 'exclude_by_default')
+            fields = cls._get_fields_with_attr(cls, "exclude_by_default")
             cls.__cached_exclude_fields = OrderedDict(sorted(fields.items()))
         return cls.__cached_exclude_fields
 
     @classmethod
-    def get_dpath_translated_path(cls, path, separator='.'):
+    def get_dpath_translated_path(cls, path, separator="."):
         if cls.__cached_dpath_computed_fields is None:
             cls.__cached_dpath_computed_fields = {}
         if path not in cls.__cached_dpath_computed_fields:

apiserver/elastic/apply_mappings.py

@@ -5,7 +5,7 @@ Apply elasticsearch mappings to given hosts.
 import argparse
 import json
 from pathlib import Path
-from typing import Optional, Sequence
+from typing import Optional, Sequence, Tuple
 
 from elasticsearch import Elasticsearch
 
@@ -13,7 +13,7 @@ HERE = Path(__file__).resolve().parent
 
 
 def apply_mappings_to_cluster(
-    hosts: Sequence, key: Optional[str] = None, es_args: dict = None
+    hosts: Sequence, key: Optional[str] = None, es_args: dict = None, http_auth: Tuple = None
 ):
     """Hosts maybe a sequence of strings or dicts in the form {"host": <host>, "port": <port>}"""
 
@@ -30,7 +30,7 @@ def apply_mappings_to_cluster(
     else:
         files = p.glob("**/*.json")
 
-    es = Elasticsearch(hosts=hosts, **(es_args or {}))
+    es = Elasticsearch(hosts=hosts, http_auth=http_auth, **(es_args or {}))
     return [_send_template(f) for f in files]
 
 

apiserver/elastic/initialize.py

@@ -82,7 +82,11 @@ def check_elastic_empty() -> bool:
         es_logger.addFilter(log_filter)
         for retry in range(max_retries):
             try:
-                es = Elasticsearch(hosts=cluster_conf.get("hosts"))
+                es = Elasticsearch(
+                    hosts=cluster_conf.get("hosts", None),
+                    http_auth=es_factory.get_credentials("events", cluster_conf),
+                    **cluster_conf.get("args", {})
+                )
                 return not es.indices.get_template(name="events*")
             except exceptions.NotFoundError as ex:
                 log.error(ex)
@@ -109,5 +113,7 @@ def init_es_data():
 
         log.info(f"Applying mappings to ES host: {hosts_config}")
         args = cluster_conf.get("args", {})
-        res = apply_mappings_to_cluster(hosts_config, name, es_args=args)
+        http_auth = es_factory.get_credentials(name)
+
+        res = apply_mappings_to_cluster(hosts_config, name, es_args=args, http_auth=http_auth)
         log.info(res)

apiserver/elastic/mappings/events/events.json

@@ -1,7 +1,8 @@
 {
   "index_patterns": "events-*",
   "settings": {
-    "number_of_shards": 1
+    "number_of_shards": 1,
+    "number_of_replicas": 0
   },
   "mappings": {
     "_source": {

apiserver/elastic/mappings/workers/queue_metrics.json

@@ -1,7 +1,8 @@
 {
   "index_patterns": "queue_metrics_*",
   "settings": {
-    "number_of_shards": 1
+    "number_of_shards": 1,
+    "number_of_replicas": 0
   },
   "mappings": {
     "_source": {

apiserver/elastic/mappings/workers/worker_stats.json

@@ -1,7 +1,8 @@
 {
   "index_patterns": "worker_stats_*",
   "settings": {
-    "number_of_shards": 1
+    "number_of_shards": 1,
+    "number_of_replicas": 0
   },
   "mappings": {
     "_source": {

apiserver/es_factory.py

@@ -1,9 +1,10 @@
 from datetime import datetime
+from functools import lru_cache
 from os import getenv
-from typing import Tuple
+from typing import Tuple, Optional
 
 from boltons.iterutils import first
-from elasticsearch import Elasticsearch, Transport
+from elasticsearch import Elasticsearch
 
 from apiserver.config_repo import config
 
@@ -21,6 +22,10 @@ OVERRIDE_PORT_ENV_KEY = (
     "ELASTIC_SERVICE_PORT",
 )
 
+OVERRIDE_USERNAME_ENV_KEY = ("CLEARML_ELASTIC_SERVICE_USERNAME",)
+
+OVERRIDE_PASSWORD_ENV_KEY = ("CLEARML_ELASTIC_SERVICE_PASSWORD",)
+
 OVERRIDE_HOST = first(filter(None, map(getenv, OVERRIDE_HOST_ENV_KEY)))
 if OVERRIDE_HOST:
     log.info(f"Using override elastic host {OVERRIDE_HOST}")
@@ -29,6 +34,14 @@ OVERRIDE_PORT = first(filter(None, map(getenv, OVERRIDE_PORT_ENV_KEY)))
 if OVERRIDE_PORT:
     log.info(f"Using override elastic port {OVERRIDE_PORT}")
 
+OVERRIDE_USERNAME = first(filter(None, map(getenv, OVERRIDE_USERNAME_ENV_KEY)))
+if OVERRIDE_USERNAME:
+    log.info(f"Using override elastic username {OVERRIDE_USERNAME}")
+
+OVERRIDE_PASSWORD = first(filter(None, map(getenv, OVERRIDE_PASSWORD_ENV_KEY)))
+if OVERRIDE_PASSWORD:
+    log.info("Using override elastic password ********")
+
 _instances = {}
 
 
@@ -48,6 +61,10 @@ class InvalidClusterConfiguration(Exception):
     pass
 
 
+class MissingPasswordForElasticUser(Exception):
+    pass
+
+
 class ESFactory:
     @classmethod
     def connect(cls, cluster_name):
@@ -65,22 +82,45 @@ class ESFactory:
             if not hosts:
                 raise InvalidClusterConfiguration(cluster_name)
 
+            http_auth = cls.get_credentials(cluster_name)
+
             args = cluster_config.get("args", {})
             _instances[cluster_name] = Elasticsearch(
-                hosts=hosts, transport_class=Transport, **args
+                hosts=hosts, http_auth=http_auth, **args
             )
 
         return _instances[cluster_name]
 
+    @classmethod
+    def get_credentials(cls, cluster_name: str, cluster_config: dict = None) -> Optional[Tuple[str, str]]:
+        cluster_config = cluster_config or cls.get_cluster_config(cluster_name)
+        if not cluster_config.get("secure", True):
+            return None
+
+        elastic_user = OVERRIDE_USERNAME or config.get("secure.elastic.user", None)
+        if not elastic_user:
+            return None
+
+        elastic_password = OVERRIDE_PASSWORD or config.get(
+            "secure.elastic.password", None
+        )
+        if not elastic_password:
+            raise MissingPasswordForElasticUser(
+                f"cluster={cluster_name}, username={elastic_user}"
+            )
+
+        return elastic_user, elastic_password
+
     @classmethod
     def get_all_cluster_names(cls):
         return list(config.get("hosts.elastic"))
 
     @classmethod
-    def get_override(cls, cluster_name: str) -> Tuple[str, str]:
+    def get_override_host(cls, cluster_name: str) -> Tuple[str, str]:
         return OVERRIDE_HOST, OVERRIDE_PORT
 
     @classmethod
+    @lru_cache()
     def get_cluster_config(cls, cluster_name):
         """
         Returns cluster config for the specified cluster path
@@ -97,7 +137,7 @@ class ESFactory:
             for entry in cluster_config.get("hosts", []):
                 entry[key] = value
 
-        host, port = cls.get_override(cluster_name)
+        host, port = cls.get_override_host(cluster_name)
 
         if host:
             set_host_prop("host", host)

apiserver/mongo/initialize/pre_populate.py

@@ -21,8 +21,10 @@ from typing import (
     Union,
     Mapping,
     IO,
+    Callable,
 )
 from urllib.parse import unquote, urlparse
+from uuid import uuid4
 from zipfile import ZipFile, ZIP_BZIP2
 
 import mongoengine
@@ -54,6 +56,7 @@ from apiserver.database.model.task.task import (
 from apiserver.database.utils import get_options
 from apiserver.utilities import json
 from apiserver.utilities.dicts import nested_get, nested_set, nested_delete
+from apiserver.utilities.parameter_key_escaper import ParameterKeyEscaper
 
 
 class PrePopulate:
@@ -298,8 +301,9 @@ class PrePopulate:
             if company_id is None:
                 company_id = ""
 
-            # Always use a public user for pre-populated data
-            cls.user_cls(id=user_id, name=user_name, company="").save()
+            existing_user = cls.user_cls.objects(id=user_id).only("id").first()
+            if not existing_user:
+                cls.user_cls(id=user_id, name=user_name, company=company_id).save()
 
             cls._import(zfile, company_id, user_id, metadata)
 
@@ -687,6 +691,19 @@ class PrePopulate:
                 continue
             yield clean
 
+    @classmethod
+    def _generate_new_ids(
+        cls, reader: ZipFile, entity_files: Sequence
+    ) -> Mapping[str, str]:
+        ids = {}
+        for entity_file in entity_files:
+            with reader.open(entity_file) as f:
+                for item in cls.json_lines(f):
+                    orig_id = json.loads(item).get("_id")
+                    if orig_id:
+                        ids[orig_id] = str(uuid4()).replace("-", "")
+        return ids
+
     @classmethod
     def _import(
         cls,
@@ -701,37 +718,46 @@ class PrePopulate:
         Start from entities since event import will require the tasks already in DB
         """
         event_file_ending = cls.events_file_suffix + ".json"
-        entity_files = (
+        entity_files = [
             fi
             for fi in reader.filelist
             if not fi.orig_filename.endswith(event_file_ending)
             and fi.orig_filename != cls.metadata_filename
-        )
+        ]
         metadata = metadata or {}
+        old_to_new_ids = (
+            cls._generate_new_ids(reader, entity_files)
+            if metadata.get("new_ids")
+            else {}
+        )
         tasks = []
         for entity_file in entity_files:
             with reader.open(entity_file) as f:
                 full_name = splitext(entity_file.orig_filename)[0]
                 print(f"Reading {reader.filename}:{full_name}...")
-                res = cls._import_entity(f, full_name, company_id, user_id, metadata)
+                res = cls._import_entity(
+                    f, full_name, company_id, user_id, metadata, old_to_new_ids
+                )
                 if res:
                     tasks = res
 
         if sort_tasks_by_last_updated:
             tasks = sorted(tasks, key=attrgetter("last_update"))
 
+        new_to_old_ids = {v: k for k, v in old_to_new_ids.items()}
         for task in tasks:
+            old_task_id = new_to_old_ids.get(task.id, task.id)
             events_file = first(
                 fi
                 for fi in reader.filelist
-                if fi.orig_filename.endswith(task.id + event_file_ending)
+                if fi.orig_filename.endswith(old_task_id + event_file_ending)
             )
             if not events_file:
                 continue
             with reader.open(events_file) as f:
                 full_name = splitext(events_file.orig_filename)[0]
                 print(f"Reading {reader.filename}:{full_name}...")
-                cls._import_events(f, full_name, company_id, user_id)
+                cls._import_events(f, company_id, user_id, task.id)
 
     @classmethod
     def _get_entity_type(cls, full_name) -> Type[mongoengine.Document]:
@@ -743,6 +769,28 @@ class PrePopulate:
         module = importlib.import_module(module_name)
         return getattr(module, class_name)
 
+    @staticmethod
+    def _upgrade_project_data(project_data: dict) -> dict:
+        if not project_data.get("basename"):
+            name: str = project_data["name"]
+            _, _, basename = name.rpartition("/")
+            project_data["basename"] = basename
+
+        return project_data
+
+    @staticmethod
+    def _upgrade_model_data(model_data: dict) -> dict:
+        metadata_key = "metadata"
+        metadata = model_data.get(metadata_key)
+        if isinstance(metadata, list):
+            metadata = {
+                ParameterKeyEscaper.escape(item["key"]): item
+                for item in metadata
+                if isinstance(item, dict) and "key" in item
+            }
+            model_data[metadata_key] = metadata
+        return model_data
+
     @staticmethod
     def _upgrade_task_data(task_data: dict) -> dict:
         """
@@ -822,14 +870,26 @@ class PrePopulate:
         company_id: str,
         user_id: str,
         metadata: Mapping[str, Any],
+        old_to_new_ids: Mapping[str, str] = None,
     ) -> Optional[Sequence[Task]]:
         cls_ = cls._get_entity_type(full_name)
         print(f"Writing {cls_.__name__.lower()}s into database")
         tasks = []
         override_project_count = 0
+        data_upgrade_funcs: Mapping[Type, Callable] = {
+            cls.task_cls: cls._upgrade_task_data,
+            cls.model_cls: cls._upgrade_model_data,
+            cls.project_cls: cls._upgrade_project_data,
+        }
         for item in cls.json_lines(f):
-            if cls_ == cls.task_cls:
-                item = json.dumps(cls._upgrade_task_data(task_data=json.loads(item)))
+            if old_to_new_ids:
+                for old_id, new_id in old_to_new_ids.items():
+                    # replace ids only when they are standalone strings
+                    # otherwise artifacts uris that contain old ids may get damaged
+                    item = item.replace(f'"{old_id}"', f'"{new_id}"')
+            upgrade_func = data_upgrade_funcs.get(cls_)
+            if upgrade_func:
+                item = json.dumps(upgrade_func(json.loads(item)))
 
             doc = cls_.from_json(item, created=True)
             if hasattr(doc, "user"):
@@ -863,11 +923,15 @@ class PrePopulate:
             return tasks
 
     @classmethod
-    def _import_events(cls, f: IO[bytes], full_name: str, company_id: str, _):
-        _, _, task_id = full_name[0 : -len(cls.events_file_suffix)].rpartition("_")
+    def _import_events(
+        cls, f: IO[bytes], company_id: str, _, task_id: str
+    ):
         print(f"Writing events for task {task_id} into database")
         for events_chunk in chunked_iter(cls.json_lines(f), 1000):
             events = [json.loads(item) for item in events_chunk]
+            for ev in events:
+                ev["task"] = task_id
+                ev["company_id"] = company_id
             cls.event_bll.add_events(
                 company_id, events=events, worker="", allow_locked_tasks=True
             )

apiserver/mongo/migrations/1_0_0.py

@@ -97,22 +97,6 @@ def _migrate_model_labels(db: Database):
             tasks.update_one({"_id": doc["_id"]}, {"$set": set_commands})
 
 
-def _migrate_project_description(db: Database):
-    projects: Collection = db["project"]
-    filter = {
-        "$or": [
-            {
-                "$expr": {"$lt": [{"$strLenCP": "$description"}, 100]},
-                "description": {"$regex": "^Auto-generated at ", "$options": "i"},
-            },
-            {"description": {"$regex": "^Auto-generated during move$", "$options": "i"}},
-            {"description": {"$regex": "^Auto-generated while cloning$", "$options": "i"}},
-        ]
-    }
-    for doc in projects.find(filter=filter):
-        projects.update_one({"_id": doc["_id"]}, {"$unset": {"description": 1}})
-
-
 def _migrate_project_names(db: Database):
     projects: Collection = db["project"]
 
@@ -141,5 +125,4 @@ def migrate_backend(db: Database):
     _migrate_docker_cmd(db)
     _migrate_model_labels(db)
     _migrate_project_names(db)
-    _migrate_project_description(db)
     _drop_all_indices_from_collections(db, ["task*"])

apiserver/mongo/migrations/1_0_2.py

@@ -0,0 +1,22 @@
+from pymongo.collection import Collection
+from pymongo.database import Database
+
+
+def _migrate_project_description(db: Database):
+    projects: Collection = db["project"]
+    filter = {
+        "$or": [
+            {
+                "$expr": {"$lt": [{"$strLenCP": "$description"}, 100]},
+                "description": {"$regex": "^Auto-generated at ", "$options": "i"},
+            },
+            {"description": {"$regex": "^Auto-generated during move$", "$options": "i"}},
+            {"description": {"$regex": "^Auto-generated while cloning$", "$options": "i"}},
+        ]
+    }
+    for doc in projects.find(filter=filter):
+        projects.update_one({"_id": doc["_id"]}, {"$unset": {"description": 1}})
+
+
+def migrate_backend(db: Database):
+    _migrate_project_description(db)

apiserver/mongo/migrations/1_3_0.py

@@ -0,0 +1,29 @@
+from pymongo.collection import Collection
+from pymongo.database import Database
+
+from apiserver.utilities.parameter_key_escaper import ParameterKeyEscaper
+from .utils import _drop_all_indices_from_collections
+
+
+def _convert_metadata(db: Database, name):
+    collection: Collection = db[name]
+
+    metadata_field = "metadata"
+    query = {metadata_field: {"$exists": True, "$type": 4}}
+    for doc in collection.find(filter=query, projection=(metadata_field,)):
+        metadata = {
+            ParameterKeyEscaper.escape(item["key"]): item
+            for item in doc.get(metadata_field, [])
+            if isinstance(item, dict) and "key" in item
+        }
+        collection.update_one(
+            {"_id": doc["_id"]}, {"$set": {"metadata": metadata}},
+        )
+
+
+def migrate_backend(db: Database):
+    collections = ["model", "queue"]
+    for name in collections:
+        _convert_metadata(db, name)
+
+    _drop_all_indices_from_collections(db, collections)

apiserver/mongo/migrations/1_6_0.py

@@ -0,0 +1,12 @@
+from pymongo.collection import Collection
+from pymongo.database import Database
+
+
+def migrate_backend(db: Database):
+    projects: Collection = db["project"]
+    for doc in projects.find({"basename": None}):
+        name: str = doc["name"]
+        _, _, basename = name.rpartition("/")
+        projects.update_one(
+            {"_id": doc["_id"]}, {"$set": {"basename": basename}},
+        )

apiserver/redis_manager.py

@@ -1,10 +1,8 @@
-import threading
 from os import getenv
-from time import sleep
 
 from boltons.iterutils import first
 from redis import StrictRedis
-from redis.sentinel import Sentinel, SentinelConnectionPool
+from rediscluster import RedisCluster
 
 from apiserver.apierrors.errors.server_error import ConfigError, GeneralError
 from apiserver.config_repo import config
@@ -21,6 +19,11 @@ OVERRIDE_PORT_ENV_KEY = (
     "TRAINS_REDIS_SERVICE_PORT",
     "REDIS_SERVICE_PORT",
 )
+OVERRIDE_PASSWORD_ENV_KEY = (
+    "CLEARML_REDIS_SERVICE_PASSWORD",
+    "TRAINS_REDIS_SERVICE_PASSWORD",
+    "REDIS_SERVICE_PASSWORD",
+)
 
 OVERRIDE_HOST = first(filter(None, map(getenv, OVERRIDE_HOST_ENV_KEY)))
 if OVERRIDE_HOST:
@@ -30,99 +33,7 @@ OVERRIDE_PORT = first(filter(None, map(getenv, OVERRIDE_PORT_ENV_KEY)))
 if OVERRIDE_PORT:
     log.info(f"Using override redis port {OVERRIDE_PORT}")
 
-
-class MyPubSubWorkerThread(threading.Thread):
-    def __init__(self, sentinel, on_new_master, msg_sleep_time, daemon=True):
-        super(MyPubSubWorkerThread, self).__init__()
-        self.daemon = daemon
-        self.sentinel = sentinel
-        self.on_new_master = on_new_master
-        self.sentinel_host = sentinel.connection_pool.connection_kwargs["host"]
-        self.msg_sleep_time = msg_sleep_time
-        self._running = False
-        self.pubsub = None
-
-    def subscribe(self):
-        if self.pubsub:
-            try:
-                self.pubsub.unsubscribe()
-                self.pubsub.punsubscribe()
-            except Exception:
-                pass
-            finally:
-                self.pubsub = None
-
-        subscriptions = {"+switch-master": self.on_new_master}
-
-        while not self.pubsub or not self.pubsub.subscribed:
-            try:
-                self.pubsub = self.sentinel.pubsub()
-                self.pubsub.subscribe(**subscriptions)
-            except Exception as ex:
-                log.warn(
-                    f"Error while subscribing to sentinel at {self.sentinel_host} ({ex.args[0]}) Sleeping and retrying"
-                )
-                sleep(3)
-        log.info(f"Subscribed to sentinel {self.sentinel_host}")
-
-    def run(self):
-        if self._running:
-            return
-        self._running = True
-
-        self.subscribe()
-
-        while self.pubsub.subscribed:
-            try:
-                self.pubsub.get_message(
-                    ignore_subscribe_messages=True, timeout=self.msg_sleep_time
-                )
-            except Exception as ex:
-                log.warn(
-                    f"Error while getting message from sentinel {self.sentinel_host} ({ex.args[0]}) Resubscribing"
-                )
-                self.subscribe()
-
-        self.pubsub.close()
-        self._running = False
-
-    def stop(self):
-        # stopping simply unsubscribes from all channels and patterns.
-        # the unsubscribe responses that are generated will short circuit
-        # the loop in run(), calling pubsub.close() to clean up the connection
-        self.pubsub.unsubscribe()
-        self.pubsub.punsubscribe()
-
-
-# todo,future - multi master clusters?
-class RedisCluster(object):
-    def __init__(self, sentinel_hosts, service_name, **connection_kwargs):
-        self.service_name = service_name
-        self.sentinel = Sentinel(sentinel_hosts, **connection_kwargs)
-        self.master = None
-        self.master_host_port = None
-        self.reconfigure()
-        self.sentinel_threads = {}
-        self.listen()
-
-    def reconfigure(self):
-        try:
-            self.master_host_port = self.sentinel.discover_master(self.service_name)
-            self.master = self.sentinel.master_for(self.service_name)
-            log.info(f"Reconfigured master to {self.master_host_port}")
-        except Exception as ex:
-            log.error(f"Error while reconfiguring. {ex.args[0]}")
-
-    def listen(self):
-        def on_new_master(workerThread):
-            self.reconfigure()
-
-        for sentinel in self.sentinel.sentinels:
-            sentinel_host = sentinel.connection_pool.connection_kwargs["host"]
-            self.sentinel_threads[sentinel_host] = MyPubSubWorkerThread(
-                sentinel, on_new_master, msg_sleep_time=0.001, daemon=True
-            )
-            self.sentinel_threads[sentinel_host].start()
+OVERRIDE_PASSWORD = first(filter(None, map(getenv, OVERRIDE_PASSWORD_ENV_KEY)))
 
 
 class RedisManager(object):
@@ -131,6 +42,9 @@ class RedisManager(object):
         for alias, alias_config in redis_config_dict.items():
 
             alias_config = alias_config.as_plain_ordered_dict()
+            alias_config["password"] = config.get(
+                f"secure.redis.{alias}.password", None
+            )
 
             is_cluster = alias_config.get("cluster", False)
 
@@ -142,34 +56,19 @@ class RedisManager(object):
             if port:
                 alias_config["port"] = port
 
-            db = alias_config.get("db", 0)
+            password = OVERRIDE_PASSWORD or alias_config.get("password", None)
+            if password:
+                alias_config["password"] = password
 
-            sentinels = alias_config.get("sentinels", None)
-            service_name = alias_config.get("service_name", None)
-
-            if not is_cluster and sentinels:
-                raise ConfigError(
-                    "Redis configuration is invalid. mixed regular and cluster mode",
-                    alias=alias,
-                )
-            if is_cluster and (not sentinels or not service_name):
-                raise ConfigError(
-                    "Redis configuration is invalid. missing sentinels or service_name",
-                    alias=alias,
-                )
-            if not is_cluster and (not port or not host):
+            if not port or not host:
                 raise ConfigError(
                     "Redis configuration is invalid. missing port or host", alias=alias
                 )
 
             if is_cluster:
-                # todo support all redis connection args via sentinel's connection_kwargs
-                del alias_config["sentinels"]
                 del alias_config["cluster"]
-                del alias_config["service_name"]
-                self.aliases[alias] = RedisCluster(
-                    sentinels, service_name, **alias_config
-                )
+                del alias_config["db"]
+                self.aliases[alias] = RedisCluster(**alias_config)
             else:
                 self.aliases[alias] = StrictRedis(**alias_config)
 
@@ -177,27 +76,21 @@ class RedisManager(object):
         obj = self.aliases.get(alias)
         if not obj:
             raise GeneralError(f"Invalid Redis alias {alias}")
-        if isinstance(obj, RedisCluster):
-            obj.master.get("health")
-            return obj.master
-        else:
-            obj.get("health")
-            return obj
+
+        obj.get("health")
+        return obj
 
     def host(self, alias):
         r = self.connection(alias)
-        pool = r.connection_pool
-        if isinstance(pool, SentinelConnectionPool):
-            connections = pool.connection_kwargs[
-                "connection_pool"
-            ]._available_connections
+        if isinstance(r, RedisCluster):
+            connections = first(r.connection_pool._available_connections.values())
         else:
-            connections = pool._available_connections
+            connections = r.connection_pool._available_connections
 
-        if len(connections) > 0:
-            return connections[0].host
-        else:
+        if not connections:
             return None
 
+        return connections[0].host
+
 
 redman = RedisManager(config.get("hosts.redis"))

apiserver/requirements.txt

@@ -3,7 +3,7 @@ bcrypt>=3.1.4
 boltons>=19.1.0
 boto3==1.14.13
 dpath>=1.4.2,<2.0
-elasticsearch>=7.0.0,<8.0.0
+elasticsearch==7.13.3
 fastjsonschema>=2.8
 flask-compress>=1.4.0
 flask-cors>=3.0.5
@@ -12,19 +12,20 @@ funcsigs==1.0.2
 furl>=2.0.0
 gunicorn>=19.7.1
 humanfriendly==4.18
-jinja2==2.10.1
+jinja2==2.11.3
 jsonmodels>=2.3
 jsonschema>=2.6.0
 luqum>=0.10.0
-mongoengine==0.19.1
+mongoengine==0.23.1
 nested_dict>=1.61
 packaging==20.3
 psutil>=5.6.5
 pyhocon>=0.3.35
-pyjwt<2.0.0
-pymongo==3.10.1
+pyjwt>=2.4.0
+pymongo[srv]==3.12.0
 python-rapidjson>=0.6.3
-redis>=2.10.5
+redis==3.5.3
+redis-py-cluster>=2.1.3
 related>=0.7.2
 requests>=2.13.0
 semantic_version>=2.8.3,<3

apiserver/schema/services/_common.conf

@@ -26,6 +26,10 @@ credentials {
             type: string
             description: Credentials secret key
         }
+        label {
+            type: string
+            description: Optional credentials label
+        }
     }
 }
 batch_operation {

apiserver/schema/services/auth.conf

@@ -15,11 +15,19 @@ _definitions {
                 type: string
                 description: ""
             }
+            label {
+                type: string
+                description: Optional credentials label
+            }
             last_used {
                 type: string
                 description: ""
                 format: "date-time"
             }
+            last_used_from {
+                type: string
+                description: ""
+            }
         }
     }
     role {
@@ -222,6 +230,12 @@ create_credentials {
             }
         }
     }
+    "2.17": ${create_credentials."2.1"} {
+        request.properties.label {
+            type: string
+            description: Optional credentials label
+        }
+    }
 }
 
 get_credentials {
@@ -248,6 +262,38 @@ get_credentials {
     }
 }
 
+edit_credentials {
+    allow_roles = [ "*" ]
+    internal: false
+    "2.19" {
+        description: """Updates the label of the existing credentials for the authenticated user."""
+        request {
+            type: object
+            required: [ access_key ]
+            properties {
+                access_key {
+                    type: string
+                    description: Existing credentials key
+                }
+                label {
+                    type: string
+                    description: New credentials label
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                updated {
+                    description: "Number of credentials updated"
+                    type: integer
+                    enum: [0, 1]
+                }
+            }
+        }
+    }
+}
+
 revoke_credentials {
     allow_roles = [ "*" ]
     internal: false

apiserver/schema/services/login.conf

@@ -93,3 +93,19 @@ supported_modes {
         }
     }
 }
+
+logout {
+    authorize: false
+    allow_roles = [ "*" ]
+    "2.13" {
+        description: """ Logout (including SSO, if used)) """
+        request {
+            type: object
+            additionalProperties: false
+        }
+        response {
+            type: object
+            additionalProperties: false
+        }
+    }
+}

apiserver/schema/services/models.conf

@@ -61,14 +61,14 @@ _definitions {
                 type: string
             }
             tags {
-                description: "User-defined tags list"
                 type: array
+                description: "User-defined tags"
                 items { type: string }
             }
             system_tags {
-                description: "System tags list. This field is reserved for system use, please don't use it."
                 type: array
-                items {type: string}
+                description: "System tags. This field is reserved for system use, please don't use it."
+                items { type: string }
             }
             framework {
                 description: "Framework on which the model is based. Should be identical to the framework of the task which created the model"
@@ -98,9 +98,21 @@ _definitions {
                 additionalProperties: true
             }
             metadata {
-                type: array
                 description: "Model metadata"
-                items {"$ref": "#/definitions/metadata_item"}
+                type: object
+                additionalProperties {
+                    "$ref": "#/definitions/metadata_item"
+                }
+            }
+            stats {
+                description: "Model statistics"
+                type: object
+                properties {
+                    labels_count {
+                        description: Number of the model labels
+                        type: integer
+                    }
+                }
             }
         }
     }
@@ -199,6 +211,36 @@ get_all_ex {
             }
         }
     }
+    "2.15": ${get_all_ex."2.13"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all_ex"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of models to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all_ex to retrieve more data"
+        }
+    }
+    "2.20": ${get_all_ex."2.15"} {
+        request.properties.include_stats {
+            description: "If true, include models statistic in response"
+            type: boolean
+            default: false
+        }
+    }
 }
 get_all {
     "2.1" {
@@ -302,6 +344,29 @@ get_all {
             }
         }
     }
+    "2.15": ${get_all."2.1"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of models to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
+        }
+    }
 }
 get_frameworks {
     "2.8" {
@@ -361,7 +426,7 @@ update_for_task {
                 system_tags {
                     description: "System tags list. This field is reserved for system use, please don't use it."
                     type: array
-                    items {type: string}
+                    items { type: string }
                 }
                 override_model_id {
                     description: "Override model ID. If provided, this model is updated in the task. Exactly one of override_model_id or uri is required."
@@ -427,7 +492,7 @@ create {
                 system_tags {
                     description: "System tags list. This field is reserved for system use, please don't use it."
                     type: array
-                    items {type: string}
+                    items { type: string }
                 }
                 framework {
                     description: "Framework on which the model is based. Case insensitive. Should be identical to the framework of the task which created the model."
@@ -483,9 +548,11 @@ create {
     }
     "2.13": ${create."2.1"} {
         metadata {
-            type: array
             description: "Model metadata"
-            items {"$ref": "#/definitions/metadata_item"}
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/metadata_item"
+            }
         }
     }
 }
@@ -522,7 +589,7 @@ edit {
                 system_tags {
                     description: "System tags list. This field is reserved for system use, please don't use it."
                     type: array
-                    items {type: string}
+                    items { type: string }
                 }
                 framework {
                     description: "Framework on which the model is based. Case insensitive. Should be identical to the framework of the task which created the model."
@@ -578,9 +645,11 @@ edit {
     }
     "2.13": ${edit."2.1"} {
         metadata {
-            type: array
             description: "Model metadata"
-            items {"$ref": "#/definitions/metadata_item"}
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/metadata_item"
+            }
         }
     }
 }
@@ -611,7 +680,7 @@ update {
                 system_tags {
                     description: "System tags list. This field is reserved for system use, please don't use it."
                     type: array
-                    items {type: string}
+                    items { type: string }
                 }
                 ready {
                     description: "Indication if the model is final and can be used by other tasks Default is false."
@@ -661,9 +730,11 @@ update {
     }
     "2.13": ${update."2.1"} {
         metadata {
-            type: array
             description: "Model metadata"
-            items {"$ref": "#/definitions/metadata_item"}
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/metadata_item"
+            }
         }
     }
 }
@@ -672,7 +743,7 @@ publish_many {
         description: Publish models
         request {
             properties {
-                ids.description: "IDs of models to publish"
+                ids.description: "IDs of the models to publish"
                 force_publish_task {
                     description: "Publish the associated tasks (if exist) even if they are not in the 'stopped' state. Optional, the default value is False."
                     type: boolean
@@ -733,7 +804,7 @@ archive_many {
         description: Archive models
         request {
             properties {
-                ids.description: "IDs of models to archive"
+                ids.description: "IDs of the models to archive"
             }
         }
         response {
@@ -769,10 +840,9 @@ delete_many {
         description: Delete models
         request {
             properties {
-                ids.description: "IDs of models to delete"
+                ids.description: "IDs of the models to delete"
                 force {
-                    description: """Force. Required if there are tasks that use the model as an execution model, or if the model's creating task is published.
-                """
+                    description: "Force. Required if there are tasks that use the model as an execution model, or if the model's creating task is published."
                     type: boolean
                 }
             }
@@ -929,6 +999,11 @@ add_or_update_metadata {
                     description: "Metadata items to add or update"
                     items {"$ref": "#/definitions/metadata_item"}
                 }
+                replace_metadata {
+                    description: "If set then the all the metadata items will be replaced with the provided ones. Otherwise only the provided metadata items will be updated or added"
+                    type: boolean
+                    default: false
+                }
             }
         }
         response {

apiserver/schema/services/organization.conf

@@ -102,4 +102,64 @@ get_user_companies {
             }
         }
     }
-}
+}
+get_entities_count {
+    "2.20": {
+        description: "Get counts for the company entities according to the passed search criteria"
+        request {
+            type: object
+            properties {
+                projects {
+                    type: object
+                    additionalProperties: true
+                    description: Search criteria for projects
+                }
+                tasks {
+                    type: object
+                    additionalProperties: true
+                    description: Search criteria for experiments
+                }
+                models {
+                    type: object
+                    additionalProperties: true
+                    description: Search criteria for models
+                }
+                pipelines {
+                    type: object
+                    additionalProperties: true
+                    description: Search criteria for pipelines
+                }
+                datasets {
+                    type: object
+                    additionalProperties: true
+                    description: Search criteria for datasets
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                projects {
+                    type: integer
+                    description: The number of projects matching the criteria
+                }
+                tasks {
+                    type: integer
+                    description: The number of experiments matching the criteria
+                }
+                models {
+                    type: integer
+                    description: The number of models matching the criteria
+                }
+                pipelines {
+                    type: integer
+                    description: The number of pipelines matching the criteria
+                }
+                datasets {
+                    type: integer
+                    description: The number of datasets matching the criteria
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/pipelines.conf

@@ -0,0 +1,47 @@
+_description: "Provides a management API for pipelines in the system."
+_definitions {
+}
+
+start_pipeline {
+    "2.17" {
+        description: "Start a pipeline"
+        request {
+            type: object
+            required: [ task ]
+            properties {
+                task {
+                    description: "ID of the task on which the pipeline will be based"
+                    type: string
+                }
+                queue {
+                    description: "Queue ID in which the created pipeline task will be enqueued"
+                    type: string
+                }
+                args {
+                    description: "Task arguments, name/value to be placed in the hyperparameters Args section"
+                    type: array
+                    items {
+                        type: object
+                        properties {
+                            name: { type: string }
+                            value: { type: [string, null] }
+                        }
+                    }
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                pipeline {
+                    description: "ID of the new pipeline task"
+                    type: string
+                }
+                enqueued {
+                    description: "True if the task was successfuly enqueued"
+                    type: boolean
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/projects.conf

@@ -25,6 +25,10 @@ _definitions {
                 description: "Project name"
                 type: string
             }
+            basename {
+                description: "Project base name"
+                type: string
+            }
             description {
                 description: "Project description"
                 type: string
@@ -42,15 +46,20 @@ _definitions {
                 type: string
                 format: "date-time"
             }
+            last_update {
+                description: "Last update time"
+                type: string
+                format: "date-time"
+            }
             tags {
-                type: array
                 description: "User-defined tags"
+                type: array
                 items { type: string }
             }
             system_tags {
-                type: array
                 description: "System tags. This field is reserved for system use, please don't use it."
-                items {type: string}
+                type: array
+                items { type: string }
             }
             default_output_destination {
                 description: "The default output destination URL for new tasks under this project"
@@ -70,6 +79,18 @@ _definitions {
                 description: "Total run time of all tasks in project (in seconds)"
                 type: integer
             }
+            total_tasks {
+                description: "Number of tasks"
+                type: integer
+            }
+            completed_tasks_24h {
+                description: "Number of tasks completed in the last 24 hours"
+                type: integer
+            }
+            last_task_run {
+                description: "The most recent started time of a task"
+                type: integer
+            }
             status_count {
                 description: "Status counts"
                 type: object
@@ -78,6 +99,10 @@ _definitions {
                         description: "Number of 'created' tasks in project"
                         type: integer
                     }
+                    completed {
+                        description: "Number of 'completed' tasks in project"
+                        type: integer
+                    }
                     queued {
                         description: "Number of 'queued' tasks in project"
                         type: integer
@@ -135,6 +160,10 @@ _definitions {
                 description: "Project name"
                 type: string
             }
+            basename {
+                description: "Project base name"
+                type: string
+            }
             description {
                 description: "Project description"
                 type: string
@@ -152,15 +181,20 @@ _definitions {
                 type: string
                 format: "date-time"
             }
+            last_update {
+                description: "Last update time"
+                type: string
+                format: "date-time"
+            }
             tags {
-                type: array
                 description: "User-defined tags"
+                type: array
                 items { type: string }
             }
             system_tags {
-                type: array
                 description: "System tags. This field is reserved for system use, please don't use it."
-                items {type: string}
+                type: array
+                items { type: string }
             }
             default_output_destination {
                 description: "The default output destination URL for new tasks under this project"
@@ -188,6 +222,28 @@ _definitions {
                     }
                 }
             }
+            own_tasks {
+                description: "The amount of tasks under this project (without children projects). Returned if 'check_own_contents' flag is set in the request"
+                type: integer
+            }
+            own_models {
+                description: "The amount of models under this project (without children projects). Returned if 'check_own_contents' flag is set in the request"
+                type: integer
+            }
+            dataset_stats {
+                description: Project dataset statistics
+                type: object
+                properties {
+                    file_count {
+                        type: integer
+                        description: The number of files stored in the dataset
+                    }
+                    total_size {
+                        type: integer
+                        description: The total dataset size in bytes
+                    }
+                }
+            }
         }
     }
     metric_variant_result {
@@ -294,14 +350,14 @@ create {
                     type: string
                 }
                 tags {
-                    type: array
                     description: "User-defined tags"
+                    type: array
                     items { type: string }
                 }
                 system_tags {
-                    type: array
                     description: "System tags. This field is reserved for system use, please don't use it."
-                    items {type: string}
+                    type: array
+                    items { type: string }
                 }
                 default_output_destination  {
                     description: "The default output destination URL for new tasks under this project"
@@ -359,6 +415,10 @@ get_all {
                     description: "Get only projects whose name matches this pattern (python regular expression syntax)"
                     type: string
                 }
+                basename {
+                    description: "Project base name"
+                    type: string
+                }
                 description {
                     description: "Get only projects whose description matches this pattern (python regular expression syntax)"
                     type: string
@@ -379,7 +439,7 @@ get_all {
                     items { type: string }
                 }
                 page {
-                    description: "Page number, returns a specific page out of the resulting list of dataviews"
+                    description: "Page number, returns a specific page out of the resulting list of projects"
                     type: integer
                     minimum: 0
                 }
@@ -414,7 +474,6 @@ get_all {
                     description: "Projects list"
                     type: array
                     items { "$ref": "#/definitions/projects_get_all_response_single" }
-
                 }
             }
         }
@@ -430,6 +489,36 @@ get_all {
             }
         }
     }
+    "2.14": ${get_all."2.13"} {
+        request.properties.search_hidden {
+            description: "If set to 'true' then hidden projects are included in the search results"
+            type: boolean
+            default: false
+        }
+    }
+    "2.15": ${get_all."2.14"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all_ex"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of projects to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all_ex to retrieve more data"
+        }
+    }
 }
 get_all_ex {
     internal: true
@@ -469,24 +558,63 @@ get_all_ex {
                     default: false
                 }
                 check_own_contents {
-                    description: "If set to 'true' and project ids are passed to the query then for these projects their own tasks, models and dataviews are counted"
+                    description: "If set to 'true' and project ids are passed to the query then for these projects their own tasks and models are counted"
                     type: boolean
                     default: false
                 }
             }
         }
-        response {
+    }
+    "2.14": ${get_all_ex."2.13"} {
+        request.properties.search_hidden {
+            description: "If set to 'true' then hidden projects are included in the search results"
+            type: boolean
+            default: false
+        }
+    }
+    "2.15": ${get_all_ex."2.14"} {
+        request {
             properties {
-                own_tasks {
-                    description: "The amount of tasks under this project (without children projects). Returned if 'check_own_contents' flag is set in the request"
-                    type: integer
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all"
                 }
-                own_models {
-                    description: "The amount of models under this project (without children projects). Returned if 'check_own_contents' flag is set in the request"
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
                     type: integer
+                    minimum: 1
+                    description: "The number of projects to retrieve"
                 }
             }
         }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
+        }
+    }
+    "2.16": ${get_all_ex."2.15"} {
+        request.properties.stats_with_children {
+            description: "If include_stats flag is set then this flag contols whether the child projects tasks are taken into statistics or not"
+            type: boolean
+            default: true
+        }
+    }
+    "2.17": ${get_all_ex."2.16"} {
+        request.properties.include_stats_filter {
+            description: The filter for selecting entities that participate in statistics calculation. For each task field that you want to filter on pass the list of allowed values. Prepend the value with '-' to exclude
+            type: object
+            additionalProperties: true
+        }
+    }
+    "2.20": ${get_all_ex."2.17"} {
+        request.properties.include_dataset_stats {
+            description: "If true, include project dataset statistic in response"
+            type: boolean
+            default: false
+        }
     }
 }
 update {
@@ -504,23 +632,19 @@ update {
                     description: "Project name. Unique within the company."
                     type: string
                 }
-                description {
-                    description: "Project description. "
-                    type: string
-                }
                 description {
                     description: "Project description"
                     type: string
                 }
                 tags {
+                    description: "User-defined tags list"
                     type: array
-                    description: "User-defined tags"
                     items { type: string }
                 }
                 system_tags {
+                    description: "System tags list. This field is reserved for system use, please don't use it."
                     type: array
-                    description: "System tags. This field is reserved for system use, please don't use it."
-                    items {type: string}
+                    items { type: string }
                 }
                 default_output_destination {
                     description: "The default output destination URL for new tasks under this project"
@@ -594,7 +718,7 @@ merge {
             type: object
             properties {
                 moved_entities {
-                    description: "The number of tasks, models and dataviews moved from the merged project into the destination"
+                    description: "The number of tasks and models moved from the merged project into the destination"
                     type: integer
                 }
                 moved_projects {
@@ -605,6 +729,42 @@ merge {
         }
     }
 }
+validate_delete {
+    "2.14" {
+        description: "Validates that the project existis and can be deleted"
+        request {
+            type: object
+            required: [ project ]
+            properties {
+                project {
+                    description: "Project ID"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                tasks {
+                    description: "The total number of tasks under the project and all its children"
+                    type: integer
+                }
+                non_archived_tasks {
+                    description: "The total number of non-archived tasks under the project and all its children"
+                    type: integer
+                }
+                models {
+                    description: "The total number of models under the project and all its children"
+                    type: integer
+                }
+                non_archived_models {
+                    description: "The total number of non-archived models under the project and all its children"
+                    type: integer
+                }
+            }
+        }
+    }
+}
 delete {
     "2.1" {
         description: "Deletes a project"
@@ -613,7 +773,7 @@ delete {
             required: [ project ]
             properties {
                 project {
-                    description: "Project id"
+                    description: "Project ID"
                     type: string
                 }
                 force {
@@ -622,7 +782,6 @@ delete {
                     type: boolean
                     default: false
                 }
-
             }
         }
         response {
@@ -755,6 +914,7 @@ get_hyper_parameters {
         description: """Get a list of all hyper parameter sections and names used in tasks within the given project."""
         request {
             type: object
+            required: [project]
             properties {
                 project {
                     description: "Project ID"
@@ -803,7 +963,105 @@ get_hyper_parameters {
         }
     }
 }
+get_model_metadata_values {
+    "2.17" {
+        description: """Get a list of distinct values for the chosen model metadata key"""
+        request {
+            type: object
+            required: [key]
+            properties {
+                projects {
+                    description: "Project IDs"
+                    type: array
+                    items {type: string}
+                }
+                key {
+                    description: "Metadata key"
+                    type: string
+                }
+                allow_public {
+                    description: "If set to 'true' then collect values from both company and public models otherwise company modeels only. The default is 'true'"
+                    type: boolean
+                }
+                include_subprojects {
+                    description: "If set to 'true' and the project field is set then the result includes metadata values from the subproject models"
+                    type: boolean
+                    default: true
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                total {
+                    description: "Total number of distinct values"
+                    type: integer
+                }
+                values {
+                    description: "The list of the unique values"
+                    type: array
+                    items {type: string}
+                }
+            }
+        }
+    }
+}
+get_model_metadata_keys {
+    "2.17" {
+        description: """Get a list of all metadata keys used in models within the given project."""
+        request {
+            type: object
+            required: [project]
+            properties {
+                project {
+                    description: "Project ID"
+                    type: string
+                }
+                include_subprojects {
+                    description: "If set to 'true' and the project field is set then the result includes metadate keys from the subproject models"
+                    type: boolean
+                    default: true
+                }
 
+                page {
+                    description: "Page number"
+                    default: 0
+                    type: integer
+                }
+                page_size {
+                    description: "Page size"
+                    default: 500
+                    type: integer
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                keys {
+                    description: "A list of model keys"
+                    type: array
+                    items {type: string}
+                }
+                remaining {
+                    description: "Remaining results"
+                    type: integer
+                }
+                total {
+                    description: "Total number of results"
+                    type: integer
+                }
+            }
+        }
+    }
+}
+get_project_tags {
+    "2.17" {
+        description: "Get user and system tags used for the specified projects and their children"
+        request = ${_definitions.tags_request}
+        response = ${_definitions.tags_response}
+    }
+}
 get_task_tags {
     "2.8" {
         description: "Get user and system tags used for the tasks under the specified projects"
@@ -811,7 +1069,6 @@ get_task_tags {
         response = ${_definitions.tags_response}
     }
 }
-
 get_model_tags {
     "2.8" {
         description: "Get user and system tags used for the models under the specified projects"
@@ -933,4 +1190,4 @@ get_task_parents {
             }
         }
     }
-}
+}

apiserver/schema/services/queues.conf

@@ -79,9 +79,11 @@ _definitions {
                 items { "$ref": "#/definitions/entry" }
             }
             metadata {
-                type: array
                 description: "Queue metadata"
-                items {"$ref": "#/definitions/metadata_item"}
+                type: object
+                additionalProperties {
+                    "$ref": "#/definitions/metadata_item"
+                }
             }
         }
     }
@@ -110,11 +112,46 @@ get_by_id {
             }
         }
     }
+    "2.20": ${get_by_id."2.4"} {
+        request.properties.max_task_entries {
+            description: Max number of queue task entries to return
+            type: integer
+        }
+    }
 }
 // typescript generation hack
 get_all_ex {
     internal: true
     "2.4": ${get_all."2.4"}
+    "2.15": ${get_all_ex."2.4"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all_ex"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of queues to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all_ex to retrieve more data"
+        }
+    }
+    "2.20": ${get_all_ex."2.15"} {
+        request.properties.max_task_entries {
+            description: Max number of queue task entries to return
+            type: integer
+        }
+    }
 }
 get_all {
     "2.4" {
@@ -178,6 +215,35 @@ get_all {
             }
         }
     }
+    "2.15": ${get_all."2.4"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of queues to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
+        }
+    }
+    "2.20": ${get_all."2.15"} {
+        request.properties.max_task_entries {
+            description: Max number of queue task entries to return
+            type: integer
+        }
+    }
 }
 get_default {
     "2.4" {
@@ -235,6 +301,15 @@ create {
             }
         }
     }
+    "2.13": ${create."2.4"} {
+        metadata {
+            description: "Queue metadata"
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/metadata_item"
+            }
+        }
+    }
 }
 update {
     "2.4" {
@@ -276,7 +351,15 @@ update {
                     type: object
                     additionalProperties: true
                 }
-
+            }
+        }
+    }
+    "2.13": ${update."2.4"} {
+        metadata {
+            description: "Queue metadata"
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/metadata_item"
             }
         }
     }
@@ -569,6 +652,13 @@ get_queue_metrics : {
             }
         }
     }
+    "2.20": ${get_queue_metrics."2.4"} {
+        request.properties.refresh {
+            type: boolean
+            default: false
+            description: If set then the new queue metrics is taken
+        }
+    }
 }
 add_or_update_metadata {
     "2.13" {
@@ -586,6 +676,11 @@ add_or_update_metadata {
                     description: "Metadata items to add or update"
                     items {"$ref": "#/definitions/metadata_item"}
                 }
+                replace_metadata {
+                    description: "If set then the all the metadata items will be replaced with the provided ones. Otherwise only the provided metadata items will be updated or added"
+                    type: boolean
+                    default: false
+                }
             }
         }
         response {
@@ -630,3 +725,51 @@ delete_metadata {
         }
     }
 }
+peek_task {
+    "2.15" {
+        description: "Peek the next task from a given queue"
+        request {
+            type: object
+            required: [ queue ]
+            properties {
+                queue {
+                    description: "ID of the queue"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                task {
+                    description: "Task ID"
+                    type: string
+                }
+            }
+        }
+    }
+}
+get_num_entries {
+    "2.15" {
+        description: "Get the number of task entries in the given queue"
+        request {
+            type: object
+            required: [ queue ]
+            properties {
+                queue {
+                    description: "ID of the queue"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                num {
+                    description: "Number of entries"
+                    type: integer
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/tasks.conf

@@ -534,7 +534,7 @@ _definitions {
             container {
                 description: "Docker container parameters"
                 type: object
-                additionalProperties { type: string }
+                additionalProperties { type: [string, null] }
             }
             models {
                 description: "Task models"
@@ -685,6 +685,36 @@ get_all_ex {
             }
         }
     }
+    "2.14": ${get_all_ex."2.13"} {
+        request.properties.search_hidden {
+            description: "If set to 'true' then hidden tasks are included in the search results"
+            type: boolean
+            default: false
+        }
+    }
+    "2.15": ${get_all_ex."2.14"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all_ex"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of tasks to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all_ex to retrieve more data"
+        }
+    }
 }
 get_all {
     "2.1" {
@@ -799,6 +829,36 @@ get_all {
             }
         }
     }
+    "2.14": ${get_all."2.1"} {
+        request.properties.search_hidden {
+            description: "If set to 'true' then hidden tasks are included in the search results"
+            type: boolean
+            default: false
+        }
+    }
+    "2.15": ${get_all."2.14"} {
+        request {
+            properties {
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of tasks to retrieve"
+                }
+            }
+        }
+        response.properties.scroll_id {
+            type: string
+            description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
+        }
+    }
 }
 get_types {
     "2.8" {
@@ -935,7 +995,7 @@ clone {
                 new_task_container {
                     description: "The docker container properties for the new task. If not provided then taken from the original task"
                     type: object
-                    additionalProperties { type: string }
+                    additionalProperties { type: [string, null] }
                 }
             }
         }
@@ -1113,7 +1173,7 @@ create {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: string }
+                    additionalProperties { type: [string, null] }
                 }
             }
         }
@@ -1202,7 +1262,7 @@ validate {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: string }
+                    additionalProperties { type: [string, null] }
                 }
             }
         }
@@ -1364,7 +1424,7 @@ edit {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: string }
+                    additionalProperties { type: [string, null] }
                 }
                 runtime {
                     description: "Task runtime mapping"
@@ -1838,7 +1898,7 @@ Fails if the following parameters in the task were not filled:
             ]
             properties {
                 queue {
-                    description: "Queue id. If not provided, task is added to the default queue."
+                    description: "Queue id. If not provided and no queue name is passed then task is added to the default queue."
                     type: string
                 }
             }
@@ -1854,6 +1914,12 @@ Fails if the following parameters in the task were not filled:
             }
         }
     }
+    "2.19": ${enqueue."1.5"} {
+        request.properties.queue_name {
+            description: The name of the queue. If the queue does not exist then it is auto-created. Cannot be used together with the queue id
+            type: string
+        }
+    }
 }
 enqueue_many {
     "2.13": ${_definitions.change_many_request} {
@@ -1862,7 +1928,7 @@ enqueue_many {
             properties {
                 ids.description: "IDs of the tasks to enqueue"
                 queue {
-                    description: "Queue id. If not provided, tasks are added to the default queue."
+                    description: "Queue id. If not provided and no queue name is passed then tasks are added to the default queue."
                     type: string
                 }
                 validate_tasks {
@@ -1881,6 +1947,12 @@ enqueue_many {
             }
         }
     }
+    "2.19": ${enqueue_many."2.13"} {
+        request.properties.queue_name {
+            description: The name of the queue. If the queue does not exist then it is auto-created. Cannot be used together with the queue id
+            type: string
+        }
+    }
 }
 dequeue {
     "1.5" {
@@ -1959,6 +2031,18 @@ completed {
         } ${_references.status_change_request}
         response: ${_definitions.update_response}
     }
+    "2.20": ${completed."2.2"} {
+        request.properties.publish {
+            type: boolean
+            default: false
+            description: If set and the task is completed successfully then it is published
+        }
+        response.properties.published {
+            description: "Number of tasks published (0 or 1)"
+            type: integer
+            enum: [0, 1]
+        }
+    }
 }
 
 ping {

apiserver/schema/services/users.conf

@@ -139,6 +139,22 @@ get_current_user {
             }
         }
     }
+    "2.20": ${get_current_user."2.1"} {
+        response {
+            properties {
+                getting_started {
+                    type: object
+                    description: Getting stated info
+                    additionalProperties: true
+                }
+                created {
+                    type: string
+                    description: User creation time
+                    format: date-time
+                }
+            }
+        }
+    }
 }
 
 get_all_ex {

apiserver/schema/services/workers.conf

@@ -1,152 +1,320 @@
-{
-    _description: "Provides an API for worker machines, allowing workers to report status and get tasks for execution"
-    _definitions {
-        metrics_category {
+_description: "Provides an API for worker machines, allowing workers to report status and get tasks for execution"
+_definitions {
+    metrics_category {
+        type: object
+        properties {
+            name {
+                type: string
+                description: "Name of the metrics category."
+            }
+            metric_keys {
+                type: array
+                items { type: string }
+                description: "The names of the metrics in the category."
+            }
+        }
+    }
+    aggregation_type {
+        type: string
+        enum: [ avg, min, max ]
+        description: "Metric aggregation type"
+    }
+    stat_item {
+        type: object
+        properties {
+            key {
+                type: string
+                description: "Name of a metric"
+            }
+            category {
+                "$ref": "#/definitions/aggregation_type"
+            }
+        }
+    }
+    aggregation_stats {
+        type: object
+        properties {
+            aggregation {
+                "$ref": "#/definitions/aggregation_type"
+            }
+            values {
+                type: array
+                description: "List of values corresponding to the dates in metric statistics"
+                items { type: number }
+            }
+        }
+    }
+    metric_stats {
+        type: object
+        properties {
+            metric {
+                type: string
+                description: "Name of the metric ("cpu_usage", "memory_used" etc.)"
+            }
+            variant {
+                type: string
+                description: "Name of the metric component. Set only if 'split_by_variant' was set in the request"
+            }
+            dates {
+                type: array
+                description: "List of timestamps (in seconds from epoch) in the acceding order. The timestamps are separated by the requested interval. Timestamps where no workers activity was recorded are omitted."
+                items { type: integer }
+            }
+            stats {
+                type: array
+                description: "Statistics data by type"
+                items { "$ref": "#/definitions/aggregation_stats" }
+            }
+        }
+    }
+    worker_stats {
+        type: object
+        properties {
+            worker {
+                type: string
+                description: "ID of the worker"
+            }
+            metrics {
+                type: array
+                description: "List of the metrics statistics for the worker"
+                items { "$ref": "#/definitions/metric_stats" }
+            }
+        }
+    }
+    activity_series {
+        type: object
+        properties {
+            dates {
+                type: array
+                description: "List of timestamps (in seconds from epoch) in the acceding order. The timestamps are separated by the requested interval."
+                items {type: integer}
+            }
+            counts {
+                type: array
+                description: "List of worker counts corresponding to the timestamps in the dates list. None values are returned for the dates with no workers."
+                items {type: integer}
+            }
+        }
+    }
+    worker {
+        type: object
+        properties {
+            id {
+                description: "Worker ID"
+                type: string
+            }
+            user {
+                description: "Associated user (under whose credentials are used by the worker daemon)"
+                "$ref": "#/definitions/id_name_entry"
+            }
+            company {
+                description: "Associated company"
+                "$ref": "#/definitions/id_name_entry"
+            }
+            ip {
+                description: "IP of the worker"
+                type: string
+            }
+            register_time {
+                description: "Registration time"
+                type: string
+                format: "date-time"
+            }
+            last_activity_time {
+                description: "Last activity time (even if an error occurred)"
+                type: string
+                format: "date-time"
+            }
+            last_report_time {
+                description: "Last successful report time"
+                type: string
+                format: "date-time"
+            }
+            task {
+                description: "Task currently being run by the worker"
+                "$ref": "#/definitions/current_task_entry"
+            }
+            project {
+                description: "Project in which currently executing task resides"
+                "$ref": "#/definitions/id_name_entry"
+            }
+            queue {
+                description: "Queue from which running task was taken"
+                "$ref": "#/definitions/queue_entry"
+            }
+            queues {
+                description: "List of queues on which the worker is listening"
+                type: array
+                items { "$ref": "#/definitions/queue_entry" }
+            }
+            tags {
+                description: "User tags for the worker"
+                type: array
+                items: { type: string }
+            }
+        }
+    }
+
+    id_name_entry  {
+        type: object
+        properties {
+            id {
+                description: "ID"
+                type: string
+            }
+            name {
+                description: "Name"
+                type: string
+            }
+        }
+    }
+
+    current_task_entry = ${_definitions.id_name_entry} {
+        properties {
+            running_time {
+                description: "Task running time"
+                type: integer
+            }
+            last_iteration {
+                description: "Last task iteration"
+                type: integer
+            }
+        }
+    }
+
+    queue_entry = ${_definitions.id_name_entry} {
+        properties {
+            next_task {
+                description: "Next task in the queue"
+                "$ref": "#/definitions/id_name_entry"
+            }
+            num_tasks {
+                description: "Number of task entries in the queue"
+                type: integer
+            }
+        }
+    }
+
+    machine_stats  {
+        type: object
+        properties {
+            cpu_usage {
+                description: "Average CPU usage per core"
+                type: array
+                items { type: number }
+            }
+            gpu_usage {
+                description: "Average GPU usage per GPU card"
+                type: array
+                items { type: number }
+            }
+            memory_used {
+                description: "Used memory MBs"
+                type: integer
+            }
+            memory_free {
+                description: "Free memory MBs"
+                type: integer
+            }
+            gpu_memory_free {
+                description: "GPU free memory MBs"
+                type: array
+                items { type: integer }
+            }
+            gpu_memory_used {
+                description: "GPU used memory MBs"
+                type: array
+                items { type: integer }
+            }
+            network_tx {
+                description: "Mbytes per second"
+                type: integer
+            }
+            network_rx {
+                description: "Mbytes per second"
+                type: integer
+            }
+            disk_free_home {
+                description: "Mbytes free space of /home drive"
+                type: integer
+            }
+            disk_free_temp {
+                description: "Mbytes free space of /tmp drive"
+                type: integer
+            }
+            disk_read {
+                description: "Mbytes read per second"
+                type: integer
+            }
+            disk_write {
+                description: "Mbytes write per second"
+                type: integer
+            }
+            cpu_temperature {
+                description: "CPU temperature"
+                type: array
+                items { type: number }
+            }
+            gpu_temperature {
+                description: "GPU temperature"
+                type: array
+                items { type: number }
+            }
+        }
+    }
+}
+get_all {
+    "2.4" {
+        description: "Returns information on all registered workers."
+        request {
             type: object
             properties {
-                name {
-                    type: string
-                    description: "Name of the metrics category."
-                }
-                metric_keys {
-                    type: array
-                    items { type: string }
-                    description: "The names of the metrics in the category."
+                last_seen {
+                    description: """Filter out workers not active for more than last_seen seconds.
+                        A value or 0 or 'none' will disable the filter."""
+                    type: integer
+                    default: 3600
                 }
             }
         }
-        aggregation_type {
-            type: string
-            enum: [ avg, min, max ]
-            description: "Metric aggregation type"
-        }
-        stat_item {
+        response {
             type: object
             properties {
-                key {
-                    type: string
-                    description: "Name of a metric"
-                }
-                category {
-                    "$ref": "#/definitions/aggregation_type"
+                workers {
+                    type: array
+                    items { "$ref": "#/definitions/worker" }
                 }
             }
         }
-        aggregation_stats {
-            type: object
-            properties {
-                aggregation {
-                    "$ref": "#/definitions/aggregation_type"
-                }
-                values {
-                    type: array
-                    description: "List of values corresponding to the dates in metric statistics"
-                    items { type: number }
-                }
-            }
+    }
+    "2.20": ${get_all."2.4"} {
+        request.properties.tags {
+            description: The list of allowed worker tags. Prepend tag value with '-' in order to exclude
+            type: array
+            items { type: string }
         }
-        metric_stats {
-            type: object
-            properties {
-                metric {
-                    type: string
-                    description: "Name of the metric ("cpu_usage", "memory_used" etc.)"
-                }
-                variant {
-                    type: string
-                    description: "Name of the metric component. Set only if 'split_by_variant' was set in the request"
-                }
-                dates {
-                    type: array
-                    description: "List of timestamps (in seconds from epoch) in the acceding order. The timestamps are separated by the requested interval. Timestamps where no workers activity was recorded are omitted."
-                    items { type: integer }
-                }
-                stats {
-                    type: array
-                    description: "Statistics data by type"
-                    items { "$ref": "#/definitions/aggregation_stats" }
-                }
-            }
-        }
-        worker_stats {
+    }
+}
+register {
+    "2.4" {
+        description: "Register a worker in the system. Called by the Worker Daemon."
+        request {
+            required: [ worker ]
             type: object
             properties {
                 worker {
-                    type: string
-                    description: "ID of the worker"
-                }
-                metrics {
-                    type: array
-                    description: "List of the metrics statistics for the worker"
-                    items { "$ref": "#/definitions/metric_stats" }
-                }
-            }
-        }
-        activity_series {
-            type: object
-            properties {
-                dates {
-                    type: array
-                    description: "List of timestamps (in seconds from epoch) in the acceding order. The timestamps are separated by the requested interval."
-                    items {type: integer}
-                }
-                counts {
-                    type: array
-                    description: "List of worker counts corresponding to the timestamps in the dates list. None values are returned for the dates with no workers."
-                    items {type: integer}
-                }
-            }
-        }
-        worker {
-            type: object
-            properties {
-                id {
-                    description: "Worker ID"
+                    description: "Worker id. Must be unique in company."
                     type: string
                 }
-                user {
-                    description: "Associated user (under whose credentials are used by the worker daemon)"
-                    "$ref": "#/definitions/id_name_entry"
-                }
-                company {
-                    description: "Associated company"
-                    "$ref": "#/definitions/id_name_entry"
-                }
-                ip {
-                    description: "IP of the worker"
-                    type: string
-                }
-                register_time {
-                    description: "Registration time"
-                    type: string
-                    format: "date-time"
-                }
-                last_activity_time {
-                    description: "Last activity time (even if an error occurred)"
-                    type: string
-                    format: "date-time"
-                }
-                last_report_time {
-                    description: "Last successful report time"
-                    type: string
-                    format: "date-time"
-                }
-                task {
-                    description: "Task currently being run by the worker"
-                    "$ref": "#/definitions/current_task_entry"
-                }
-                project {
-                    description: "Project in which currently executing task resides"
-                    "$ref": "#/definitions/id_name_entry"
-                }
-                queue {
-                    description: "Queue from which running task was taken"
-                    "$ref": "#/definitions/queue_entry"
+                timeout {
+                    description: "Registration timeout in seconds. If timeout seconds have passed since the worker's last call to register or status_report, the worker is automatically removed from the list of registered workers."
+                    type: integer
+                    default: 600
                 }
                 queues {
-                    description: "List of queues on which the worker is listening"
+                    description: "List of queue IDs on which the worker is listening."
                     type: array
-                    items { "$ref": "#/definitions/queue_entry" }
+                    items { type: string }
                 }
                 tags {
                     description: "User tags for the worker"
@@ -155,348 +323,185 @@
                 }
             }
         }
-
-        id_name_entry  {
+        response {
+            type: object
+            properties {}
+        }
+    }
+}
+unregister {
+    "2.4" {
+        description: "Unregister a worker in the system. Called by the Worker Daemon."
+        request {
+            required: [ worker ]
             type: object
             properties {
-                id {
-                    description: "ID"
-                    type: string
-                }
-                name {
-                    description: "Name"
+                worker {
+                    description: "Worker id. Must be unique in company."
                     type: string
                 }
             }
         }
-
-        current_task_entry = ${_definitions.id_name_entry} {
-            properties {
-                running_time {
-                    description: "Task running time"
-                    type: integer
-                }
-                last_iteration {
-                    description: "Last task iteration"
-                    type: integer
-                }
-            }
+        response {
+            type: object
+            properties {}
         }
-
-        queue_entry = ${_definitions.id_name_entry} {
-            properties {
-                next_task {
-                    description: "Next task in the queue"
-                    "$ref": "#/definitions/id_name_entry"
-                }
-                num_tasks {
-                    description: "Number of task entries in the queue"
-                    type: integer
-                }
-            }
-        }
-
-        machine_stats  {
+    }
+}
+status_report {
+    "2.4" {
+        description: "Called periodically by the worker daemon to report machine status"
+        request {
+            required: [
+                worker
+                timestamp
+            ]
             type: object
             properties {
-                cpu_usage {
-                    description: "Average CPU usage per core"
-                    type: array
-                    items { type: number }
+                worker {
+                    description: "Worker id."
+                    type: string
                 }
-                gpu_usage {
-                    description: "Average GPU usage per GPU card"
-                    type: array
-                    items { type: number }
+                task {
+                    description: "ID of a task currently being run by the worker. If no task is sent, the worker's task field will be cleared."
+                    type: string
                 }
-                memory_used {
-                    description: "Used memory MBs"
+                queue {
+                    description: "ID of the queue from which task was received. If no queue is sent, the worker's queue field will be cleared."
+                    type: string
+                }
+                queues {
+                    description: "List of queue IDs on which the worker is listening. If null, the worker's queues list will not be updated."
+                    type: array
+                    items { type: string }
+                }
+                timestamp {
+                    description: "UNIX time in seconds since epoch."
                     type: integer
                 }
-                memory_free {
-                    description: "Free memory MBs"
-                    type: integer
+                machine_stats {
+                    description: "The machine statistics."
+                    "$ref": "#/definitions/machine_stats"
                 }
-                gpu_memory_free {
-                    description: "GPU free memory MBs"
+                tags {
+                    description: "New user tags for the worker"
                     type: array
-                    items { type: integer }
+                    items: { type: string }
                 }
-                gpu_memory_used {
-                    description: "GPU used memory MBs"
+            }
+        }
+        response {
+            type: object
+            properties {}
+        }
+    }
+}
+get_metric_keys {
+    "2.4" {
+        description: "Returns worker statistics metric keys grouped by categories."
+        request {
+            type: object
+            properties {
+                worker_ids {
+                    description: "List of worker ids to collect metrics for. If not provided or empty then all the company workers metrics are analyzed."
                     type: array
-                    items { type: integer }
+                    items { type: string }
                 }
-                network_tx {
-                    description: "Mbytes per second"
-                    type: integer
-                }
-                network_rx {
-                    description: "Mbytes per second"
-                    type: integer
-                }
-                disk_free_home {
-                    description: "Mbytes free space of /home drive"
-                    type: integer 
-                }
-                disk_free_temp {
-                    description: "Mbytes free space of /tmp drive"
-                    type: integer 
-                }
-                disk_read {
-                    description: "Mbytes read per second"
-                    type: integer
-                }
-                disk_write {
-                    description: "Mbytes write per second"
-                    type: integer
-                }
-                cpu_temperature {
-                    description: "CPU temperature"
+            }
+        }
+        response {
+            type: object
+            properties {
+                categories {
                     type: array
-                    items { type: number }
+                    description: "List of unique metric categories found in the statistics of the requested workers."
+                    items { "$ref": "#/definitions/metrics_category" }
                 }
-                gpu_temperature {
-                    description: "GPU temperature"
+            }
+        }
+    }
+}
+get_stats {
+    "2.4" {
+        description: "Returns statistics for the selected workers and time range aggregated by date intervals."
+        request {
+            type: object
+            required: [ from_date, to_date, interval, items ]
+            properties {
+                worker_ids {
+                    description: "List of worker ids to collect metrics for. If not provided or empty then all the company workers metrics are analyzed."
                     type: array
-                    items { type: number }
+                    items { type: string }
+                }
+                from_date {
+                    description: "Starting time (in seconds from epoch) for collecting statistics"
+                    type: number
+                }
+                to_date {
+                    description: "Ending time (in seconds from epoch) for collecting statistics"
+                    type: number
+                }
+                interval {
+                    description: "Time interval in seconds for a single statistics point. The minimal value is 1"
+                    type: integer
+                }
+                items {
+                    description: "List of metric keys and requested statistics"
+                    type: array
+                    items { "$ref": "#/definitions/stat_item" }
+                }
+                split_by_variant {
+                    description: "If true then break statistics by hardware sub types"
+                    type: boolean
+                    default: false
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                workers {
+                    type: array
+                    description: "List of the requested workers with their statistics"
+                    items { "$ref": "#/definitions/worker_stats" }
                 }
             }
         }
     }
-    get_all {
-        "2.4" {
-            description: "Returns information on all registered workers."
-            request {
-                type: object
-                properties {
-                    last_seen {
-                        description: """Filter out workers not active for more than last_seen seconds.
-                            A value or 0 or 'none' will disable the filter."""
-                        type: integer
-                        default: 3600
-                    }
+}
+get_activity_report {
+    "2.4" {
+        description: "Returns count of active company workers in the selected time range."
+        request {
+            type: object
+            required: [ from_date, to_date, interval ]
+            properties {
+                from_date {
+                    description: "Starting time (in seconds from epoch) for collecting statistics"
+                    type: number
+                }
+                to_date {
+                    description: "Ending time (in seconds from epoch) for collecting statistics"
+                    type: number
+                }
+                interval {
+                    description: "Time interval in seconds for a single statistics point. The minimal value is 1"
+                    type: integer
                 }
             }
-            response {
-                type: object
-                properties {
-                    workers {
-                        type: array
-                        items { "$ref": "#/definitions/worker" }
-                    }
+        }
+        response {
+            type: object
+            properties {
+                total {
+                    description: "Activity series that include all the workers that sent reports in the given time interval."
+                    "$ref": "#/definitions/activity_series"
+                }
+                active {
+                    description: "Activity series that include only workers that worked on a task in the given time interval."
+                    "$ref": "#/definitions/activity_series"
                 }
             }
         }
     }
-    register {
-        "2.4" {
-            description: "Register a worker in the system. Called by the Worker Daemon."
-            request {
-                required: [ worker ]
-                type: object
-                properties {
-                    worker {
-                        description: "Worker id. Must be unique in company."
-                        type: string
-                    }
-                    timeout {
-                        description: "Registration timeout in seconds. If timeout seconds have passed since the worker's last call to register or status_report, the worker is automatically removed from the list of registered workers."
-                        type: integer
-                        default: 600
-                    }
-                    queues {
-                        description: "List of queue IDs on which the worker is listening."
-                        type: array
-                        items { type: string }
-                    }
-                    tags {
-                        description: "User tags for the worker"
-                        type: array
-                        items: { type: string }
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {}
-            }
-        }
-    }
-    unregister {
-        "2.4" {
-            description: "Unregister a worker in the system. Called by the Worker Daemon."
-            request {
-                required: [ worker ]
-                type: object
-                properties {
-                    worker {
-                        description: "Worker id. Must be unique in company."
-                        type: string
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {}
-            }
-        }
-    }
-    status_report {
-        "2.4" {
-            description: "Called periodically by the worker daemon to report machine status"
-            request {
-                required: [ 
-                    worker
-                    timestamp
-                ]
-                type: object
-                properties {
-                    worker {
-                        description: "Worker id."
-                        type: string
-                    }
-                    task {
-                        description: "ID of a task currently being run by the worker. If no task is sent, the worker's task field will be cleared."
-                        type: string
-                    }
-                    queue {
-                        description: "ID of the queue from which task was received. If no queue is sent, the worker's queue field will be cleared."
-                        type: string
-                    }
-                    queues {
-                        description: "List of queue IDs on which the worker is listening. If null, the worker's queues list will not be updated."
-                        type: array
-                        items { type: string }
-                    }
-                    timestamp {
-                        description: "UNIX time in seconds since epoch."
-                        type: integer
-                    }
-                    machine_stats {
-                        description: "The machine statistics."
-                        "$ref": "#/definitions/machine_stats"
-                    }
-                    tags {
-                        description: "New user tags for the worker"
-                        type: array
-                        items: { type: string }
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {}
-            }
-        }
-    }
-    get_metric_keys {
-        "2.4" {
-            description: "Returns worker statistics metric keys grouped by categories."
-            request {
-                type: object
-                properties {
-                    worker_ids {
-                        description: "List of worker ids to collect metrics for. If not provided or empty then all the company workers metrics are analyzed."
-                        type: array
-                        items { type: string }
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {
-                    categories {
-                        type: array
-                        description: "List of unique metric categories found in the statistics of the requested workers."
-                        items { "$ref": "#/definitions/metrics_category" }
-                    }
-                }
-            }
-        }
-    }
-    get_stats {
-        "2.4" {
-            description: "Returns statistics for the selected workers and time range aggregated by date intervals."
-            request {
-                type: object
-                required: [ from_date, to_date, interval, items ]
-                properties {
-                    worker_ids {
-                        description: "List of worker ids to collect metrics for. If not provided or empty then all the company workers metrics are analyzed."
-                        type: array
-                        items { type: string }
-                    }
-                    from_date {
-                        description: "Starting time (in seconds from epoch) for collecting statistics"
-                        type: number
-                    }
-                    to_date {
-                        description: "Ending time (in seconds from epoch) for collecting statistics"
-                        type: number
-                    }
-                    interval {
-                        description: "Time interval in seconds for a single statistics point. The minimal value is 1"
-                        type: integer
-                    }
-                    items {
-                        description: "List of metric keys and requested statistics"
-                        type: array
-                        items { "$ref": "#/definitions/stat_item" }
-                    }
-                    split_by_variant {
-                        description: "If true then break statistics by hardware sub types"
-                        type: boolean
-                        default: false
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {
-                    workers {
-                        type: array
-                        description: "List of the requested workers with their statistics"
-                        items { "$ref": "#/definitions/worker_stats" }
-                    }
-                }
-            }
-        }
-    }
-    get_activity_report {
-        "2.4" {
-            description: "Returns count of active company workers in the selected time range."
-            request {
-                type: object
-                required: [ from_date, to_date, interval ]
-                properties {
-                    from_date {
-                        description: "Starting time (in seconds from epoch) for collecting statistics"
-                        type: number
-                    }
-                    to_date {
-                        description: "Ending time (in seconds from epoch) for collecting statistics"
-                        type: number
-                    }
-                    interval {
-                        description: "Time interval in seconds for a single statistics point. The minimal value is 1"
-                        type: integer
-                    }
-                }
-            }
-            response {
-                type: object
-                properties {
-                    total {
-                        description: "Activity series that include all the workers that sent reports in the given time interval."
-                        "$ref": "#/definitions/activity_series"
-                    }
-                    active {
-                        description: "Activity series that include only workers that worked on a task in the given time interval."
-                        "$ref": "#/definitions/activity_series"
-                    }
-                }
-            }
-        }
-    }
-}
+}

apiserver/server_init/app_sequence.py

@@ -4,7 +4,7 @@ from hashlib import md5
 from flask import Flask
 from flask_compress import Compress
 from flask_cors import CORS
-from semantic_version import Version
+from packaging.version import Version
 
 from apiserver.database import db
 from apiserver.bll.statistics.stats_reporter import StatisticsReporter
@@ -25,6 +25,7 @@ from apiserver.server_init.request_handlers import RequestHandlers
 from apiserver.service_repo import ServiceRepo
 from apiserver.sync import distributed_lock
 from apiserver.updates import check_updates_thread
+from apiserver.utilities.env import get_bool
 from apiserver.utilities.threads_manager import ThreadsManager
 
 log = config.logger(__file__)
@@ -46,10 +47,13 @@ class AppSequence:
     def _attach_request_handlers(self, request_handlers: RequestHandlers):
         self.app.before_first_request(request_handlers.before_app_first_request)
         self.app.before_request(request_handlers.before_request)
+        self.app.after_request(request_handlers.after_request)
 
     def _configure(self):
         CORS(self.app, **config.get("apiserver.cors"))
-        Compress(self.app)
+
+        if get_bool("CLEARML_COMPRESS_RESP", default=True):
+            Compress(self.app)
 
         self.app.config["SECRET_KEY"] = config.get(
             "secure.http.session_secret.apiserver"

apiserver/server_init/request_handlers.py

@@ -1,21 +1,24 @@
 from functools import partial
 
 from flask import request, Response, redirect
+from werkzeug.datastructures import ImmutableMultiDict
 from werkzeug.exceptions import BadRequest
 
 from apiserver.apierrors import APIError
 from apiserver.apierrors.base import BaseError
 from apiserver.config_repo import config
 from apiserver.service_repo import ServiceRepo, APICall
-from apiserver.service_repo.auth import AuthType
+from apiserver.service_repo.auth import AuthType, Token
 from apiserver.service_repo.errors import PathParsingError
 from apiserver.utilities import json
+from apiserver.utilities.dicts import nested_set
 
 log = config.logger(__file__)
 
 
 class RequestHandlers:
     _request_strip_prefix = config.get("apiserver.request.strip_prefix", None)
+    _server_header = config.get("apiserver.response.headers.server", "clearml")
 
     def before_app_first_request(self):
         pass
@@ -26,10 +29,13 @@ class RequestHandlers:
         if "/static/" in request.path:
             return
 
+        if request.content_encoding:
+            return f"Content encoding is not supported ({request.content_encoding})", 415
+
         try:
             call = self._create_api_call(request)
             load_data_callback = partial(self._load_call_data, req=request)
-            content, content_type = ServiceRepo.handle_call(
+            content, content_type, company = ServiceRepo.handle_call(
                 call, load_data_callback=load_data_callback
             )
 
@@ -51,20 +57,53 @@ class RequestHandlers:
 
             if call.result.cookies:
                 for key, value in call.result.cookies.items():
-                    kwargs = config.get("apiserver.auth.cookies")
+                    kwargs = config.get("apiserver.auth.cookies").copy()
                     if value is None:
-                        kwargs = kwargs.copy()
+                        # Removing a cookie
                         kwargs["max_age"] = 0
                         kwargs["expires"] = 0
-                        response.set_cookie(key, "", **kwargs)
-                    else:
-                        response.set_cookie(key, value, **kwargs)
+                        value = ""
+                    elif not company:
+                        # Setting a cookie, let's try to figure out the company
+                        # noinspection PyBroadException
+                        try:
+                            company = Token.decode_identity(value).company
+                        except Exception:
+                            pass
+
+                    if company:
+                        try:
+                            # use no default value to allow setting a null domain as well
+                            kwargs["domain"] = config.get(f"apiserver.auth.cookies_domain_override.{company}")
+                        except KeyError:
+                            pass
+
+                    response.set_cookie(key, value, **kwargs)
 
             return response
         except Exception as ex:
             log.exception(f"Failed processing request {request.url}: {ex}")
             return f"Failed processing request {request.url}", 500
 
+    def after_request(self, response):
+        response.headers["server"] = self._server_header
+        return response
+
+    @staticmethod
+    def _apply_multi_dict(body: dict, md: ImmutableMultiDict):
+        def convert_value(v: str):
+            if v.replace(".", "", 1).isdigit():
+                return float(v) if "." in v else int(v)
+            if v in ("true", "True", "TRUE"):
+                return True
+            if v in ("false", "False", "FALSE"):
+                return False
+            return v
+
+        for k, v in md.lists():
+            v = [convert_value(x) for x in v] if (len(v) > 1 or k.endswith("[]")) else convert_value(v[0])
+            nested_set(body, k.rstrip("[]").split("."), v)
+
     def _update_call_data(self, call, req):
         """ Use request payload/form to fill call data or batched data """
         if req.content_type == "application/json-lines":
@@ -82,23 +121,12 @@ class RequestHandlers:
                     req.on_json_loading_failed(msg)
             call.batched_data = items
         else:
-            json_body = req.get_json(force=True, silent=False) if req.data else None
-            # merge form and args
-            form = req.form.copy()
-            form.update(req.args)
-            form = form.to_dict()
-            # convert string numbers to floats
-            for key in form:
-                if form[key].replace(".", "", 1).isdigit():
-                    if "." in form[key]:
-                        form[key] = float(form[key])
-                    else:
-                        form[key] = int(form[key])
-                elif form[key].lower() == "true":
-                    form[key] = True
-                elif form[key].lower() == "false":
-                    form[key] = False
-            call.data = json_body or form or {}
+            body = (req.get_json(force=True, silent=False) if req.data else None) or {}
+            if req.args:
+                self._apply_multi_dict(body, req.args)
+            if req.form:
+                self._apply_multi_dict(body, req.form)
+            call.data = body
 
     def _call_or_empty_with_error(self, call, req, msg, code=500, subcode=0):
         call = call or APICall(

apiserver/service_repo/apicall.py

@@ -95,8 +95,8 @@ class DataContainer(object):
     @raw_data.setter
     def raw_data(self, value):
         assert isinstance(
-            value, string_types + (types.GeneratorType,)
-        ), "Raw data must be a string type or generator"
+            value, string_types + (types.GeneratorType, bytes)
+        ), "Raw data must be a string type or bytes or generator"
         self._raw_data = value
 
     @property
@@ -310,6 +310,13 @@ class APICall(DataContainer):
     _transaction_headers = _get_headers("Trx")
     """ Transaction ID """
 
+    _redacted_headers = {
+        HEADER_AUTHORIZATION: " ",
+        "Cookie": "=",
+        "X-Jwt-Payload": "",
+    }
+    """ Headers whose value should be redacted. Maps header name to partition char """
+
     @property
     def HEADER_TRANSACTION(self):
         return self._transaction_headers[0]
@@ -389,6 +396,10 @@ class APICall(DataContainer):
         self._auth_cookie = auth_cookie
         self._json_flags = {}
 
+    @property
+    def files(self):
+        return self._files
+
     @property
     def id(self):
         return self._id
@@ -584,17 +595,26 @@ class APICall(DataContainer):
     def json_flags(self):
         return self._json_flags
 
+    @property
+    def extra_meta_fields(self):
+        return {}
+
     def mark_end(self):
         self._end_ts = time.time()
         self._duration = int((self._end_ts - self._start_ts) * 1000)
 
-    def get_response(self, include_stack: bool = False) -> Tuple[Union[dict, str], str]:
+    def get_response(self, include_stack: bool = None) -> Tuple[Union[dict, str], str]:
         """
         Get the response for this call.
         :param include_stack: If True, stack trace stored in this call's result should
-        be included in the response (default is False)
+        be included in the response (default follows configuration)
         :return: Response data (encoded according to self.content_type) and the data's content type
         """
+        include_stack = (
+            include_stack
+            if include_stack is not None
+            else config.get("apiserver.return_stack_to_caller", False)
+        )
 
         def make_version_number(version: PartialVersion) -> Union[None, float, str]:
             """
@@ -629,6 +649,7 @@ class APICall(DataContainer):
                     "result_msg": self.result.msg,
                     "error_stack": self.result.traceback if include_stack else None,
                     "error_data": self.result.error_data,
+                    **self.extra_meta_fields,
                 },
                 "data": self.result.data,
             }
@@ -663,3 +684,19 @@ class APICall(DataContainer):
             error_data=error_data,
             cookies=self._result.cookies,
         )
+
+    def get_redacted_headers(self):
+        headers = self.headers.copy()
+        if not self.requires_authorization or self.auth:
+            # We won't log the authorization header if call shouldn't be authorized, or if it was successfully
+            #  authorized. This means we'll only log authorization header for calls that failed to authorize (hopefully
+            #  this will allow us to debug authorization errors).
+            for header, sep in self._redacted_headers.items():
+                if header in headers:
+                    if sep:
+                        prefix, _, redact = headers[header].partition(sep)
+                    else:
+                        prefix = sep = ""
+                        redact = headers[header]
+                    headers[header] = prefix + sep + f"<{len(redact)} bytes redacted>"
+        return headers

apiserver/service_repo/auth/auth.py

@@ -51,7 +51,7 @@ def authorize_token(jwt_token, *_, **__):
         )
 
 
-def authorize_credentials(auth_data, service, action, call_data_items):
+def authorize_credentials(auth_data, service, action, call):
     """Validate credentials against service/action and request data (dicts).
     Returns a new basic object (auth payload)
     """
@@ -100,7 +100,12 @@ def authorize_credentials(auth_data, service, action, call_data_items):
         if not fixed_user:
             # In case these are proper credentials, update last used time
             User.objects(id=user.id, credentials__key=access_key).update(
-                **{"set__credentials__$__last_used": datetime.utcnow()}
+                **{
+                    "set__credentials__$__last_used": datetime.utcnow(),
+                    "set__credentials__$__last_used_from": call.get_worker(
+                        default=call.real_ip
+                    ),
+                }
             )
 
     with TimingContext("mongo", "company_by_id"):

apiserver/service_repo/auth/payload/token.py

@@ -2,6 +2,8 @@ import jwt
 
 from datetime import datetime, timedelta
 
+from jwt.algorithms import get_default_algorithms
+
 from apiserver.apierrors import errors
 from apiserver.config_repo import config
 from apiserver.database.model.auth import Role
@@ -9,19 +11,24 @@ from apiserver.database.model.auth import Role
 from .auth_type import AuthType
 from .payload import Payload
 
-token_secret = config.get('secure.auth.token_secret')
+token_secret = config.get("secure.auth.token_secret")
+
+log = config.logger(__file__)
 
 
 class Token(Payload):
-    default_expiration_sec = config.get('apiserver.auth.default_expiration_sec')
+    default_expiration_sec = config.get("apiserver.auth.default_expiration_sec")
 
-    def __init__(self, exp=None, iat=None, nbf=None, env=None, identity=None, entities=None, **_):
+    def __init__(
+        self, exp=None, iat=None, nbf=None, env=None, identity=None, entities=None, **_
+    ):
         super(Token, self).__init__(
-            AuthType.bearer_token, identity=identity, entities=entities)
+            AuthType.bearer_token, identity=identity, entities=entities
+        )
         self.exp = exp
         self.iat = iat
         self.nbf = nbf
-        self._env = env or config.get('env', '<unknown>')
+        self._env = env or config.get("env", "<unknown>")
 
     @property
     def env(self):
@@ -62,7 +69,15 @@ class Token(Payload):
 
     @classmethod
     def decode(cls, encoded_token, verify=True):
-        return jwt.decode(encoded_token, token_secret, verify=verify)
+        options = (
+            {"verify_signature": False, "verify_exp": True} if not verify else None
+        )
+        return jwt.decode(
+            encoded_token,
+            token_secret,
+            algorithms=get_default_algorithms(),
+            options=options,
+        )
 
     @classmethod
     def from_encoded_token(cls, encoded_token, verify=True):
@@ -71,26 +86,38 @@ class Token(Payload):
             token = Token.from_dict(decoded)
             assert isinstance(token, Token)
             if not token.identity:
-                raise errors.unauthorized.InvalidToken('token missing identity')
+                raise errors.unauthorized.InvalidToken("token missing identity")
             return token
         except Exception as e:
-            raise errors.unauthorized.InvalidToken('failed parsing token, %s' % e.args[0])
+            raise errors.unauthorized.InvalidToken(
+                "failed parsing token, %s" % e.args[0]
+            )
 
     @classmethod
-    def create_encoded_token(cls, identity, expiration_sec=None, entities=None, **extra_payload):
+    def create_encoded_token(
+        cls, identity, expiration_sec=None, entities=None, **extra_payload
+    ):
         if identity.role not in (Role.system,):
             # limit expiration time for all roles but an internal service
             expiration_sec = expiration_sec or cls.default_expiration_sec
 
         now = datetime.utcnow()
 
-        token = cls(
-            identity=identity,
-            entities=entities,
-            iat=now)
+        token = cls(identity=identity, entities=entities, iat=now)
 
         if expiration_sec:
             # add 'expiration' claim
             token.exp = now + timedelta(seconds=expiration_sec)
 
         return token.encode(**extra_payload)
+
+    @classmethod
+    def decode_identity(cls, encoded_token):
+        # noinspection PyBroadException
+        try:
+            from ..auth import Identity
+
+            decoded = cls.decode(encoded_token, verify=False)
+            return Identity.from_dict(decoded.get("identity", {}))
+        except Exception as ex:
+            log.error(f"Failed parsing identity from encoded token: {ex}")

apiserver/service_repo/auth/utils.py

@@ -1,9 +1,12 @@
 import random
+import string
+
 sys_random = random.SystemRandom()
 
 
-def get_random_string(length=12, allowed_chars='abcdefghijklmnopqrstuvwxyz'
-                                               'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'):
+def get_random_string(
+    length: int = 12, allowed_chars: str = string.ascii_letters + string.digits
+) -> str:
     """
     Returns a securely generated random string.
 
@@ -12,20 +15,20 @@ def get_random_string(length=12, allowed_chars='abcdefghijklmnopqrstuvwxyz'
 
     Taken from the django.utils.crypto module.
     """
-    return ''.join(sys_random.choice(allowed_chars) for _ in range(length))
+    return "".join(sys_random.choice(allowed_chars) for _ in range(length))
 
 
-def get_client_id(length=20):
+def get_client_id(length: int = 20) -> str:
     """
     Create a random secret key.
 
     Taken from the Django project.
     """
-    chars = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'
+    chars = string.ascii_uppercase + string.digits
     return get_random_string(length, chars)
 
 
-def get_secret_key(length=50):
+def get_secret_key(length: int = 50) -> str:
     """
     Create a random secret key.
 
@@ -33,5 +36,5 @@ def get_secret_key(length=50):
     NOTE: asterisk is not supported due to issues with environment variables containing
      asterisks (in case the secret key is stored in an environment variable)
     """
-    chars = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!@#$%^&(-_=+)'
+    chars = string.ascii_letters + string.digits
     return get_random_string(length, chars)

apiserver/service_repo/service_repo.py

@@ -8,8 +8,10 @@ import jsonmodels.models
 
 from apiserver.apierrors import APIError, errors
 from apiserver.config_repo import config
+from apiserver.database.errors import translate_errors_context
 from apiserver.utilities.partial_version import PartialVersion
 from .apicall import APICall
+from .auth import Identity
 from .endpoint import Endpoint
 from .errors import MalformedPathError, InvalidVersionError, CallFailedError
 from .util import parse_return_stack_on_code
@@ -37,7 +39,7 @@ class ServiceRepo(object):
     """If the check is set, parsing will fail for endpoint request with the version that is grater than the current 
     maximum """
 
-    _max_version = PartialVersion("2.13")
+    _max_version = PartialVersion("2.20")
     """ Maximum version number (the highest min_version value across all endpoints) """
 
     _endpoint_exp = (
@@ -233,19 +235,27 @@ class ServiceRepo(object):
         return subcode in subcode_list
 
     @classmethod
-    def _get_company(
+    def _get_identity(
         cls, call: APICall, endpoint: Endpoint = None, ignore_error: bool = False
-    ) -> Optional[str]:
+    ) -> Optional[Identity]:
         authorize = endpoint and endpoint.authorize
         if ignore_error or not authorize:
             try:
-                return call.identity.company
+                return call.identity
             except Exception:
                 return None
-        return call.identity.company
+        return call.identity
+
+    @classmethod
+    def _get_company(
+        cls, call: APICall, endpoint: Endpoint = None, ignore_error: bool = False
+    ) -> Optional[str]:
+        identity = cls._get_identity(call, endpoint=endpoint, ignore_error=ignore_error)
+        return None if identity is None else identity.company
 
     @classmethod
     def handle_call(cls, call: APICall, load_data_callback: Callable = None):
+        company = None
         try:
             if call.failed:
                 raise CallFailedError()
@@ -274,7 +284,8 @@ class ServiceRepo(object):
             # In case call does not require authorization, parsing the identity.company might raise an exception
             company = cls._get_company(call, endpoint)
 
-            ret = endpoint.func(call, company, call.data_model)
+            with translate_errors_context():
+                ret = endpoint.func(call, company, call.data_model)
 
             # allow endpoints to return dict or model (instead of setting them explicitly on the call)
             if ret is not None:
@@ -316,4 +327,4 @@ class ServiceRepo(object):
                 else:
                     log.error(console_msg)
 
-        return content, content_type
+        return content, content_type, company

apiserver/service_repo/validators.py

@@ -69,7 +69,7 @@ def validate_auth(endpoint, call):
         auth = call.authorization or ""
         auth_type, _, auth_data = auth.partition(" ")
         authorize_func = get_auth_func(auth_type)
-        call.auth = authorize_func(auth_data, service, action, call.batched_data)
+        call.auth = authorize_func(auth_data, service, action, call)
     except Exception:
         if endpoint.authorize:
             # if endpoint requires authorization, re-raise exception

apiserver/services/auth.py

@@ -13,6 +13,8 @@ from apiserver.apimodels.auth import (
     CredentialsResponse,
     RevokeCredentialsRequest,
     EditUserReq,
+    CreateCredentialsRequest,
+    EditCredentialsRequest,
 )
 from apiserver.apimodels.base import UpdateResponse
 from apiserver.bll.auth import AuthBLL
@@ -41,14 +43,12 @@ def login(call: APICall, *_, **__):
     )
 
     # Add authorization cookie
-    call.result.cookies[
-        config.get("apiserver.auth.session_auth_cookie_name")
-    ] = call.result.data_model.token
+    call.result.set_auth_cookie(call.result.data_model.token)
 
 
 @endpoint("auth.logout", min_version="2.2")
 def logout(call: APICall, *_, **__):
-    call.result.cookies[config.get("apiserver.auth.session_auth_cookie_name")] = None
+    call.result.set_auth_cookie(None)
 
 
 @endpoint(
@@ -60,9 +60,13 @@ def get_token_for_user(call: APICall, _: str, request: GetTokenForUserRequest):
     """ Generates a token based on a requested user and company. INTERNAL. """
     if call.identity.role not in Role.get_system_roles():
         if call.identity.role != Role.admin and call.identity.user != request.user:
-            raise errors.bad_request.InvalidUserId("cannot generate token for another user")
+            raise errors.bad_request.InvalidUserId(
+                "cannot generate token for another user"
+            )
         if call.identity.company != request.company:
-            raise errors.bad_request.InvalidId("cannot generate token in another company")
+            raise errors.bad_request.InvalidId(
+                "cannot generate token in another company"
+            )
 
     call.result.data_model = AuthBLL.get_token_for_user(
         user_id=request.user,
@@ -95,7 +99,10 @@ def validate_token_endpoint(call: APICall, _, __):
 )
 def create_user(call: APICall, _, request: CreateUserRequest):
     """ Create a user from. INTERNAL. """
-    if call.identity.role not in Role.get_system_roles() and request.company != call.identity.company:
+    if (
+        call.identity.role not in Role.get_system_roles()
+        and request.company != call.identity.company
+    ):
         raise errors.bad_request.InvalidId("cannot create user in another company")
 
     user_id = AuthBLL.create_user(request=request, call=call)
@@ -103,7 +110,7 @@ def create_user(call: APICall, _, request: CreateUserRequest):
 
 
 @endpoint("auth.create_credentials", response_data_model=CreateCredentialsResponse)
-def create_credentials(call: APICall, _, __):
+def create_credentials(call: APICall, _, request: CreateCredentialsRequest):
     if _is_protected_user(call.identity.user):
         raise errors.bad_request.InvalidUserId("protected identity")
 
@@ -111,10 +118,32 @@ def create_credentials(call: APICall, _, __):
         user_id=call.identity.user,
         company_id=call.identity.company,
         role=call.identity.role,
+        label=request.label,
     )
     call.result.data_model = CreateCredentialsResponse(credentials=credentials)
 
 
+@endpoint("auth.edit_credentials")
+def edit_credentials(call: APICall, company_id: str, request: EditCredentialsRequest):
+    identity = call.identity
+    access_key = request.access_key
+
+    updated = User.objects(
+        id=identity.user,
+        company=company_id,
+        credentials__match={"key": access_key},
+    ).update_one(set__credentials__S__label=request.label)
+    if not updated:
+        raise errors.bad_request.InvalidAccessKey(
+            "invalid user or invalid access key",
+            user=identity.user,
+            access_key=access_key,
+            company=company_id,
+        )
+
+    call.result.data = {"updated": updated}
+
+
 @endpoint(
     "auth.revoke_credentials",
     request_data_model=RevokeCredentialsRequest,
@@ -153,7 +182,12 @@ def get_credentials(call: APICall, _, __):
         # we return ONLY the key IDs, never the secrets (want a secret? create new credentials)
         call.result.data_model = GetCredentialsResponse(
             credentials=[
-                CredentialsResponse(access_key=c.key, last_used=c.last_used)
+                CredentialsResponse(
+                    access_key=c.key,
+                    last_used=c.last_used,
+                    label=c.label,
+                    last_used_from=c.last_used_from,
+                )
                 for c in user.credentials
             ]
         )

apiserver/services/events.py

@@ -1,28 +1,42 @@
 import itertools
+import math
 from collections import defaultdict
 from operator import itemgetter
+from typing import Sequence, Optional
 
 import attr
+import jsonmodels.fields
+from boltons.iterutils import bucketize
 
 from apiserver.apierrors import errors
 from apiserver.apimodels.events import (
     MultiTaskScalarMetricsIterHistogramRequest,
     ScalarMetricsIterHistogramRequest,
-    DebugImagesRequest,
-    DebugImageResponse,
+    MetricEventsRequest,
+    MetricEventsResponse,
     MetricEvents,
     IterationEvents,
     TaskMetricsRequest,
     LogEventsRequest,
     LogOrderEnum,
-    GetDebugImageSampleRequest,
-    NextDebugImageSampleRequest,
+    GetHistorySampleRequest,
+    NextHistorySampleRequest,
+    MetricVariants as ApiMetrics,
+    TaskPlotsRequest,
+    TaskEventsRequest,
+    ScalarMetricsIterRawRequest,
+    ClearScrollRequest,
+    ClearTaskLogRequest,
+    SingleValueMetricsRequest,
 )
 from apiserver.bll.event import EventBLL
-from apiserver.bll.event.event_common import EventType
+from apiserver.bll.event.event_common import EventType, MetricVariants
+from apiserver.bll.event.events_iterator import Scroll
+from apiserver.bll.event.scalar_key import ScalarKeyEnum, ScalarKey
 from apiserver.bll.task import TaskBLL
+from apiserver.config_repo import config
 from apiserver.service_repo import APICall, endpoint
-from apiserver.utilities import json
+from apiserver.utilities import json, extract_properties_to_lists
 
 task_bll = TaskBLL()
 event_bll = EventBLL()
@@ -36,7 +50,6 @@ def add(call: APICall, company_id, _):
         company_id, [data], call.worker, allow_locked_tasks=allow_locked
     )
     call.result.data = dict(added=added, errors=err_count, errors_info=err_info)
-    call.kpis["events"] = 1
 
 
 @endpoint("events.add_batch")
@@ -47,7 +60,6 @@ def add_batch(call: APICall, company_id, _):
 
     added, err_count, err_info = event_bll.add_events(company_id, events, call.worker)
     call.result.data = dict(added=added, errors=err_count, errors_info=err_info)
-    call.kpis["events"] = len(events)
 
 
 @endpoint("events.get_task_log", required_fields=["task"])
@@ -110,7 +122,8 @@ def get_task_log(call, company_id, request: LogEventsRequest):
         company_id, task_id, allow_public=True, only=("company", "company_origin")
     )[0]
 
-    res = event_bll.log_events_iterator.get_task_events(
+    res = event_bll.events_iterator.get_task_events(
+        event_type=EventType.task_log,
         company_id=task.get_index_company(),
         task_id=task_id,
         batch_size=request.batch_size,
@@ -255,31 +268,94 @@ def vector_metrics_iter_histogram(call, company_id, _):
     )
 
 
-@endpoint("events.get_task_events", required_fields=["task"])
-def get_task_events(call, company_id, _):
-    task_id = call.data["task"]
-    batch_size = call.data.get("batch_size", 500)
-    event_type = call.data.get("event_type")
-    scroll_id = call.data.get("scroll_id")
-    order = call.data.get("order") or "asc"
+class GetTaskEventsScroll(Scroll):
+    from_key_value = jsonmodels.fields.StringField()
+    total = jsonmodels.fields.IntField()
+    request: TaskEventsRequest = jsonmodels.fields.EmbeddedField(TaskEventsRequest)
+
+
+def make_response(
+    total: int, returned: int = 0, scroll_id: str = None, **kwargs
+) -> dict:
+    return {
+        "returned": returned,
+        "total": total,
+        "scroll_id": scroll_id,
+        **kwargs,
+    }
+
+
+@endpoint("events.get_task_events", request_data_model=TaskEventsRequest)
+def get_task_events(call, company_id, request: TaskEventsRequest):
+    task_id = request.task
 
     task = task_bll.assert_exists(
-        company_id, task_id, allow_public=True, only=("company", "company_origin")
+        company_id, task_id, allow_public=True, only=("company",),
     )[0]
-    result = event_bll.get_task_events(
-        task.get_index_company(),
-        task_id,
-        sort=[{"timestamp": {"order": order}}],
-        event_type=EventType(event_type) if event_type else EventType.all,
-        scroll_id=scroll_id,
-        size=batch_size,
+
+    key = ScalarKeyEnum.iter
+    scalar_key = ScalarKey.resolve(key)
+
+    if not request.scroll_id:
+        from_key_value = None if (request.order == LogOrderEnum.desc) else 0
+        total = None
+    else:
+        try:
+            scroll = GetTaskEventsScroll.from_scroll_id(request.scroll_id)
+        except ValueError:
+            raise errors.bad_request.InvalidScrollId(scroll_id=request.scroll_id)
+
+        if scroll.from_key_value is None:
+            return make_response(
+                scroll_id=request.scroll_id, total=scroll.total, events=[]
+            )
+
+        from_key_value = scalar_key.cast_value(scroll.from_key_value)
+        total = scroll.total
+
+        scroll.request.batch_size = request.batch_size or scroll.request.batch_size
+        request = scroll.request
+
+    navigate_earlier = request.order == LogOrderEnum.desc
+    metric_variants = _get_metric_variants_from_request(request.metrics)
+
+    if request.count_total and total is None:
+        total = event_bll.events_iterator.count_task_events(
+            event_type=request.event_type,
+            company_id=task.company,
+            task_id=task_id,
+            metric_variants=metric_variants,
+        )
+
+    batch_size = min(
+        request.batch_size,
+        int(
+            config.get("services.events.events_retrieval.max_raw_scalars_size", 10_000)
+        ),
     )
 
-    call.result.data = dict(
-        events=result.events,
-        returned=len(result.events),
-        total=result.total_events,
-        scroll_id=result.next_scroll_id,
+    res = event_bll.events_iterator.get_task_events(
+        event_type=request.event_type,
+        company_id=task.company,
+        task_id=task_id,
+        batch_size=batch_size,
+        key=ScalarKeyEnum.iter,
+        navigate_earlier=navigate_earlier,
+        from_key_value=from_key_value,
+        metric_variants=metric_variants,
+    )
+
+    scroll = GetTaskEventsScroll(
+        from_key_value=str(res.events[-1][scalar_key.field]) if res.events else None,
+        total=total,
+        request=request,
+    )
+
+    return make_response(
+        returned=len(res.events),
+        total=total,
+        scroll_id=scroll.get_scroll_id(),
+        events=res.events,
     )
 
 
@@ -288,6 +364,7 @@ def get_scalar_metric_data(call, company_id, _):
     task_id = call.data["task"]
     metric = call.data["metric"]
     scroll_id = call.data.get("scroll_id")
+    no_scroll = call.data.get("no_scroll", False)
 
     task = task_bll.assert_exists(
         company_id, task_id, allow_public=True, only=("company", "company_origin")
@@ -299,6 +376,7 @@ def get_scalar_metric_data(call, company_id, _):
         sort=[{"iter": {"order": "desc"}}],
         metric=metric,
         scroll_id=scroll_id,
+        no_scroll=no_scroll,
     )
 
     call.result.data = dict(
@@ -321,7 +399,7 @@ def get_task_latest_scalar_values(call, company_id, _):
     )
     last_iters = event_bll.get_last_iters(
         company_id=company_id, event_type=EventType.all, task_id=task_id, iters=1
-    )
+    ).get(task_id)
     call.result.data = dict(
         metrics=metrics,
         last_iter=last_iters[0] if last_iters else 0,
@@ -373,6 +451,30 @@ def multi_task_scalar_metrics_iter_histogram(
     )
 
 
+@endpoint("events.get_task_single_value_metrics")
+def get_task_single_value_metrics(
+    call, company_id: str, request: SingleValueMetricsRequest
+):
+    task_ids = call.data["tasks"]
+    tasks = task_bll.assert_exists(
+        company_id=call.identity.company,
+        only=("id", "name", "company", "company_origin"),
+        task_ids=task_ids,
+        allow_public=True,
+    )
+
+    companies = {t.get_index_company() for t in tasks}
+    if len(companies) > 1:
+        raise errors.bad_request.InvalidTaskId(
+            "only tasks from the same company are supported"
+        )
+
+    res = event_bll.metrics.get_task_single_value_metrics(company_id, task_ids)
+    call.result.data = dict(
+        tasks=[{"task": task, "values": values} for task, values in res.items()]
+    )
+
+
 @endpoint("events.get_multi_task_plots", required_fields=["tasks"])
 def get_multi_task_plots_v1_7(call, company_id, _):
     task_ids = call.data["tasks"]
@@ -421,6 +523,7 @@ def get_multi_task_plots(call, company_id, req_model):
     task_ids = call.data["tasks"]
     iters = call.data.get("iters", 1)
     scroll_id = call.data.get("scroll_id")
+    no_scroll = call.data.get("no_scroll", False)
 
     tasks = task_bll.assert_exists(
         company_id=call.identity.company,
@@ -442,6 +545,7 @@ def get_multi_task_plots(call, company_id, req_model):
         sort=[{"iter": {"order": "desc"}}],
         last_iter_count=iters,
         scroll_id=scroll_id,
+        no_scroll=no_scroll,
     )
 
     tasks = {t.id: t.name for t in tasks}
@@ -494,11 +598,22 @@ def get_task_plots_v1_7(call, company_id, _):
     )
 
 
-@endpoint("events.get_task_plots", min_version="1.8", required_fields=["task"])
-def get_task_plots(call, company_id, _):
-    task_id = call.data["task"]
-    iters = call.data.get("iters", 1)
-    scroll_id = call.data.get("scroll_id")
+def _get_metric_variants_from_request(
+    req_metrics: Sequence[ApiMetrics],
+) -> Optional[MetricVariants]:
+    if not req_metrics:
+        return None
+
+    return {m.metric: m.variants for m in req_metrics}
+
+
+@endpoint(
+    "events.get_task_plots", min_version="1.8", request_data_model=TaskPlotsRequest
+)
+def get_task_plots(call, company_id, request: TaskPlotsRequest):
+    task_id = request.task
+    iters = request.iters
+    scroll_id = request.scroll_id
 
     task = task_bll.assert_exists(
         company_id, task_id, allow_public=True, only=("company", "company_origin")
@@ -509,6 +624,8 @@ def get_task_plots(call, company_id, _):
         sort=[{"iter": {"order": "desc"}}],
         last_iterations_per_plot=iters,
         scroll_id=scroll_id,
+        no_scroll=request.no_scroll,
+        metric_variants=_get_metric_variants_from_request(request.metrics),
     )
 
     return_events = result.events
@@ -521,6 +638,56 @@ def get_task_plots(call, company_id, _):
     )
 
 
+@endpoint(
+    "events.plots",
+    request_data_model=MetricEventsRequest,
+    response_data_model=MetricEventsResponse,
+)
+def task_plots(call, company_id, request: MetricEventsRequest):
+    task_metrics = defaultdict(dict)
+    for tm in request.metrics:
+        task_metrics[tm.task][tm.metric] = tm.variants
+    for metrics in task_metrics.values():
+        if None in metrics:
+            metrics.clear()
+
+    tasks = task_bll.assert_exists(
+        company_id,
+        task_ids=list(task_metrics),
+        allow_public=True,
+        only=("company", "company_origin"),
+    )
+
+    companies = {t.get_index_company() for t in tasks}
+    if len(companies) > 1:
+        raise errors.bad_request.InvalidTaskId(
+            "only tasks from the same company are supported"
+        )
+
+    result = event_bll.plots_iterator.get_task_events(
+        company_id=next(iter(companies)),
+        task_metrics=task_metrics,
+        iter_count=request.iters,
+        navigate_earlier=request.navigate_earlier,
+        refresh=request.refresh,
+        state_id=request.scroll_id,
+    )
+
+    call.result.data_model = MetricEventsResponse(
+        scroll_id=result.next_scroll_id,
+        metrics=[
+            MetricEvents(
+                task=task,
+                iterations=[
+                    IterationEvents(iter=iteration["iter"], events=iteration["events"])
+                    for iteration in iterations
+                ],
+            )
+            for (task, iterations) in result.metric_events
+        ],
+    )
+
+
 @endpoint("events.debug_images", required_fields=["task"])
 def get_debug_images_v1_7(call, company_id, _):
     task_id = call.data["task"]
@@ -590,13 +757,13 @@ def get_debug_images_v1_8(call, company_id, _):
 @endpoint(
     "events.debug_images",
     min_version="2.7",
-    request_data_model=DebugImagesRequest,
-    response_data_model=DebugImageResponse,
+    request_data_model=MetricEventsRequest,
+    response_data_model=MetricEventsResponse,
 )
-def get_debug_images(call, company_id, request: DebugImagesRequest):
-    task_metrics = defaultdict(set)
+def get_debug_images(call, company_id, request: MetricEventsRequest):
+    task_metrics = defaultdict(dict)
     for tm in request.metrics:
-        task_metrics[tm.task].add(tm.metric)
+        task_metrics[tm.task][tm.metric] = tm.variants
     for metrics in task_metrics.values():
         if None in metrics:
             metrics.clear()
@@ -623,7 +790,7 @@ def get_debug_images(call, company_id, request: DebugImagesRequest):
         state_id=request.scroll_id,
     )
 
-    call.result.data_model = DebugImageResponse(
+    call.result.data_model = MetricEventsResponse(
         scroll_id=result.next_scroll_id,
         metrics=[
             MetricEvents(
@@ -641,13 +808,13 @@ def get_debug_images(call, company_id, request: DebugImagesRequest):
 @endpoint(
     "events.get_debug_image_sample",
     min_version="2.12",
-    request_data_model=GetDebugImageSampleRequest,
+    request_data_model=GetHistorySampleRequest,
 )
-def get_debug_image_sample(call, company_id, request: GetDebugImageSampleRequest):
+def get_debug_image_sample(call, company_id, request: GetHistorySampleRequest):
     task = task_bll.assert_exists(
         company_id, task_ids=[request.task], allow_public=True, only=("company",)
     )[0]
-    res = event_bll.debug_sample_history.get_debug_image_for_variant(
+    res = event_bll.debug_image_sample_history.get_sample_for_variant(
         company_id=task.company,
         task=request.task,
         metric=request.metric,
@@ -655,6 +822,7 @@ def get_debug_image_sample(call, company_id, request: GetDebugImageSampleRequest
         iteration=request.iteration,
         refresh=request.refresh,
         state_id=request.scroll_id,
+        navigate_current_metric=request.navigate_current_metric,
     )
     call.result.data = attr.asdict(res, recurse=False)
 
@@ -662,13 +830,49 @@ def get_debug_image_sample(call, company_id, request: GetDebugImageSampleRequest
 @endpoint(
     "events.next_debug_image_sample",
     min_version="2.12",
-    request_data_model=NextDebugImageSampleRequest,
+    request_data_model=NextHistorySampleRequest,
 )
-def next_debug_image_sample(call, company_id, request: NextDebugImageSampleRequest):
+def next_debug_image_sample(call, company_id, request: NextHistorySampleRequest):
     task = task_bll.assert_exists(
         company_id, task_ids=[request.task], allow_public=True, only=("company",)
     )[0]
-    res = event_bll.debug_sample_history.get_next_debug_image(
+    res = event_bll.debug_image_sample_history.get_next_sample(
+        company_id=task.company,
+        task=request.task,
+        state_id=request.scroll_id,
+        navigate_earlier=request.navigate_earlier,
+    )
+    call.result.data = attr.asdict(res, recurse=False)
+
+
+@endpoint(
+    "events.get_plot_sample", request_data_model=GetHistorySampleRequest,
+)
+def get_plot_sample(call, company_id, request: GetHistorySampleRequest):
+    task = task_bll.assert_exists(
+        company_id, task_ids=[request.task], allow_public=True, only=("company",)
+    )[0]
+    res = event_bll.plot_sample_history.get_sample_for_variant(
+        company_id=task.company,
+        task=request.task,
+        metric=request.metric,
+        variant=request.variant,
+        iteration=request.iteration,
+        refresh=request.refresh,
+        state_id=request.scroll_id,
+        navigate_current_metric=request.navigate_current_metric,
+    )
+    call.result.data = attr.asdict(res, recurse=False)
+
+
+@endpoint(
+    "events.next_plot_sample", request_data_model=NextHistorySampleRequest,
+)
+def next_plot_sample(call, company_id, request: NextHistorySampleRequest):
+    task = task_bll.assert_exists(
+        company_id, task_ids=[request.task], allow_public=True, only=("company",)
+    )[0]
+    res = event_bll.plot_sample_history.get_next_sample(
         company_id=task.company,
         task=request.task,
         state_id=request.scroll_id,
@@ -678,14 +882,14 @@ def next_debug_image_sample(call, company_id, request: NextDebugImageSampleReque
 
 
 @endpoint("events.get_task_metrics", request_data_model=TaskMetricsRequest)
-def get_tasks_metrics(call: APICall, company_id, request: TaskMetricsRequest):
+def get_task_metrics(call: APICall, company_id, request: TaskMetricsRequest):
     task = task_bll.assert_exists(
         company_id,
         task_ids=request.tasks,
         allow_public=True,
         only=("company", "company_origin"),
     )[0]
-    res = event_bll.metrics.get_tasks_metrics(
+    res = event_bll.metrics.get_task_metrics(
         task.get_index_company(), task_ids=request.tasks, event_type=request.event_type
     )
     call.result.data = {
@@ -706,6 +910,21 @@ def delete_for_task(call, company_id, req_model):
     )
 
 
+@endpoint("events.clear_task_log")
+def clear_task_log(call: APICall, company_id: str, request: ClearTaskLogRequest):
+    task_id = request.task
+
+    task_bll.assert_exists(company_id, task_id, return_tasks=False)
+    call.result.data = dict(
+        deleted=event_bll.clear_task_log(
+            company_id=company_id,
+            task_id=task_id,
+            allow_locked=request.allow_locked,
+            threshold_sec=request.threshold_sec,
+        )
+    )
+
+
 def _get_top_iter_unique_events_per_task(events, max_iters, tasks):
     key = itemgetter("metric", "variant", "task", "iter")
 
@@ -734,13 +953,121 @@ def _get_top_iter_unique_events_per_task(events, max_iters, tasks):
 
 def _get_top_iter_unique_events(events, max_iters):
     top_unique_events = defaultdict(lambda: [])
-    for e in events:
-        key = e.get("metric", "") + e.get("variant", "")
+    for ev in events:
+        key = ev.get("metric", "") + ev.get("variant", "")
         evs = top_unique_events[key]
         if len(evs) < max_iters:
-            evs.append(e)
+            evs.append(ev)
     unique_events = list(
         itertools.chain.from_iterable(list(top_unique_events.values()))
     )
     unique_events.sort(key=lambda e: e["iter"], reverse=True)
     return unique_events
+
+
+class ScalarMetricsIterRawScroll(Scroll):
+    from_key_value = jsonmodels.fields.StringField()
+    total = jsonmodels.fields.IntField()
+    request: ScalarMetricsIterRawRequest = jsonmodels.fields.EmbeddedField(
+        ScalarMetricsIterRawRequest
+    )
+
+
+@endpoint("events.scalar_metrics_iter_raw", min_version="2.16")
+def scalar_metrics_iter_raw(
+    call: APICall, company_id: str, request: ScalarMetricsIterRawRequest
+):
+    key = request.key or ScalarKeyEnum.iter
+    scalar_key = ScalarKey.resolve(key)
+    if request.batch_size and request.batch_size < 0:
+        raise errors.bad_request.ValidationError(
+            "batch_size should be non negative number"
+        )
+
+    if not request.scroll_id:
+        from_key_value = None
+        total = None
+        request.batch_size = request.batch_size or 10_000
+    else:
+        try:
+            scroll = ScalarMetricsIterRawScroll.from_scroll_id(request.scroll_id)
+        except ValueError:
+            raise errors.bad_request.InvalidScrollId(scroll_id=request.scroll_id)
+
+        if scroll.from_key_value is None:
+            return make_response(
+                scroll_id=request.scroll_id, total=scroll.total, variants={}
+            )
+
+        from_key_value = scalar_key.cast_value(scroll.from_key_value)
+        total = scroll.total
+        request.batch_size = request.batch_size or scroll.request.batch_size
+
+    task_id = request.task
+
+    task = task_bll.assert_exists(
+        company_id, task_id, allow_public=True, only=("company",),
+    )[0]
+
+    metric_variants = _get_metric_variants_from_request([request.metric])
+
+    if request.count_total and total is None:
+        total = event_bll.events_iterator.count_task_events(
+            event_type=EventType.metrics_scalar,
+            company_id=task.company,
+            task_id=task_id,
+            metric_variants=metric_variants,
+        )
+
+    batch_size = min(
+        request.batch_size,
+        int(
+            config.get("services.events.events_retrieval.max_raw_scalars_size", 200_000)
+        ),
+    )
+
+    events = []
+    for iteration in range(0, math.ceil(batch_size / 10_000)):
+        res = event_bll.events_iterator.get_task_events(
+            event_type=EventType.metrics_scalar,
+            company_id=task.company,
+            task_id=task_id,
+            batch_size=min(batch_size, 10_000),
+            navigate_earlier=False,
+            from_key_value=from_key_value,
+            metric_variants=metric_variants,
+            key=key,
+        )
+        if not res.events:
+            break
+        events.extend(res.events)
+        from_key_value = str(events[-1][scalar_key.field])
+
+    key = str(key)
+    variants = {
+        variant: extract_properties_to_lists(
+            ["value", scalar_key.field], events, target_keys=["y", key]
+        )
+        for variant, events in bucketize(events, key=itemgetter("variant")).items()
+    }
+
+    call.kpis["events"] = len(events)
+
+    scroll = ScalarMetricsIterRawScroll(
+        from_key_value=str(events[-1][scalar_key.field]) if events else None,
+        total=total,
+        request=request,
+    )
+
+    return make_response(
+        returned=len(events),
+        total=total,
+        scroll_id=scroll.get_scroll_id(),
+        variants=variants,
+    )
+
+
+@endpoint("events.clear_scroll", min_version="2.18")
+def clear_scroll(_, __, request: ClearScrollRequest):
+    if request.scroll_id:
+        event_bll.clear_scroll(request.scroll_id)

apiserver/services/login/__init__.py

@@ -1,5 +1,3 @@
-from jsonmodels.fields import BoolField
-
 from apiserver.apimodels.login import (
     GetSupportedModesRequest,
     GetSupportedModesResponse,
@@ -35,3 +33,8 @@ def supported_modes(call: APICall, _, __: GetSupportedModesRequest):
         ),
         authenticated=call.auth is not None,
     )
+
+
+@endpoint("login.logout", min_version="2.13")
+def logout(call: APICall, _, __):
+    call.result.set_auth_cookie(None)

apiserver/services/models.py

@@ -20,17 +20,16 @@ from apiserver.apimodels.models import (
     AddOrUpdateMetadataRequest,
     ModelsPublishManyRequest,
     ModelsDeleteManyRequest,
+    ModelsGetRequest,
 )
-from apiserver.bll.model import ModelBLL
+from apiserver.bll.model import ModelBLL, Metadata
 from apiserver.bll.organization import OrgBLL, Tags
 from apiserver.bll.project import ProjectBLL, project_ids_with_children
 from apiserver.bll.task import TaskBLL
 from apiserver.bll.task.task_operations import publish_task
 from apiserver.bll.util import run_batch_operation
 from apiserver.config_repo import config
-from apiserver.database.errors import translate_errors_context
 from apiserver.database.model import validate_id
-from apiserver.database.model.metadata import metadata_add_or_update, metadata_delete
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
 from apiserver.database.model.task.task import (
@@ -50,8 +49,8 @@ from apiserver.services.utils import (
     conform_tag_fields,
     conform_output_tags,
     ModelsBackwardsCompatibility,
-    validate_metadata,
-    get_metadata_from_api,
+    unescape_metadata,
+    escape_metadata,
 )
 from apiserver.timing_context import TimingContext
 
@@ -64,19 +63,20 @@ project_bll = ProjectBLL()
 def get_by_id(call: APICall, company_id, _):
     model_id = call.data["model"]
 
-    with translate_errors_context():
-        models = Model.get_many(
-            company=company_id,
-            query_dict=call.data,
-            query=Q(id=model_id),
-            allow_public=True,
+    Metadata.escape_query_parameters(call)
+    models = Model.get_many(
+        company=company_id,
+        query_dict=call.data,
+        query=Q(id=model_id),
+        allow_public=True,
+    )
+    if not models:
+        raise errors.bad_request.InvalidModelId(
+            "no such public or company model", id=model_id, company=company_id,
         )
-        if not models:
-            raise errors.bad_request.InvalidModelId(
-                "no such public or company model", id=model_id, company=company_id,
-            )
-        conform_output_tags(call, models[0])
-        call.result.data = {"model": models[0]}
+    conform_output_tags(call, models[0])
+    unescape_metadata(call, models[0])
+    call.result.data = {"model": models[0]}
 
 
 @endpoint("models.get_by_task_id", required_fields=["task"])
@@ -86,25 +86,25 @@ def get_by_task_id(call: APICall, company_id, _):
 
     task_id = call.data["task"]
 
-    with translate_errors_context():
-        query = dict(id=task_id, company=company_id)
-        task = Task.get(_only=["models"], **query)
-        if not task:
-            raise errors.bad_request.InvalidTaskId(**query)
-        if not task.models or not task.models.output:
-            raise errors.bad_request.MissingTaskFields(field="models.output")
+    query = dict(id=task_id, company=company_id)
+    task = Task.get(_only=["models"], **query)
+    if not task:
+        raise errors.bad_request.InvalidTaskId(**query)
+    if not task.models or not task.models.output:
+        raise errors.bad_request.MissingTaskFields(field="models.output")
 
-        model_id = task.models.output[-1].model
-        model = Model.objects(
-            Q(id=model_id) & get_company_or_none_constraint(company_id)
-        ).first()
-        if not model:
-            raise errors.bad_request.InvalidModelId(
-                "no such public or company model", id=model_id, company=company_id,
-            )
-        model_dict = model.to_proper_dict()
-        conform_output_tags(call, model_dict)
-        call.result.data = {"model": model_dict}
+    model_id = task.models.output[-1].model
+    model = Model.objects(
+        Q(id=model_id) & get_company_or_none_constraint(company_id)
+    ).first()
+    if not model:
+        raise errors.bad_request.InvalidModelId(
+            "no such public or company model", id=model_id, company=company_id,
+        )
+    model_dict = model.to_proper_dict()
+    conform_output_tags(call, model_dict)
+    unescape_metadata(call, model_dict)
+    call.result.data = {"model": model_dict}
 
 
 def _process_include_subprojects(call_data: dict):
@@ -118,44 +118,67 @@ def _process_include_subprojects(call_data: dict):
     call_data["project"] = project_ids_with_children(project_ids)
 
 
-@endpoint("models.get_all_ex", required_fields=[])
-def get_all_ex(call: APICall, company_id, _):
+@endpoint("models.get_all_ex", request_data_model=ModelsGetRequest)
+def get_all_ex(call: APICall, company_id, request: ModelsGetRequest):
     conform_tag_fields(call, call.data)
-    with translate_errors_context():
-        _process_include_subprojects(call.data)
-        with TimingContext("mongo", "models_get_all_ex"):
-            models = Model.get_many_with_join(
-                company=company_id, query_dict=call.data, allow_public=True
-            )
-        conform_output_tags(call, models)
-        call.result.data = {"models": models}
+    _process_include_subprojects(call.data)
+    Metadata.escape_query_parameters(call)
+    with TimingContext("mongo", "models_get_all_ex"):
+        ret_params = {}
+        models = Model.get_many_with_join(
+            company=company_id,
+            query_dict=call.data,
+            allow_public=True,
+            ret_params=ret_params,
+        )
+    conform_output_tags(call, models)
+    unescape_metadata(call, models)
+
+    if not request.include_stats:
+        call.result.data = {"models": models, **ret_params}
+        return
+
+    model_ids = {model["id"] for model in models}
+    stats = ModelBLL.get_model_stats(
+        company=company_id,
+        model_ids=list(model_ids),
+    )
+
+    for model in models:
+        model["stats"] = stats.get(model["id"])
+
+    call.result.data = {"models": models, **ret_params}
 
 
 @endpoint("models.get_by_id_ex", required_fields=["id"])
 def get_by_id_ex(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-    with translate_errors_context():
-        with TimingContext("mongo", "models_get_by_id_ex"):
-            models = Model.get_many_with_join(
-                company=company_id, query_dict=call.data, allow_public=True
-            )
-        conform_output_tags(call, models)
-        call.result.data = {"models": models}
+    Metadata.escape_query_parameters(call)
+    with TimingContext("mongo", "models_get_by_id_ex"):
+        models = Model.get_many_with_join(
+            company=company_id, query_dict=call.data, allow_public=True
+        )
+    conform_output_tags(call, models)
+    unescape_metadata(call, models)
+    call.result.data = {"models": models}
 
 
 @endpoint("models.get_all", required_fields=[])
 def get_all(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-    with translate_errors_context():
-        with TimingContext("mongo", "models_get_all"):
-            models = Model.get_many(
-                company=company_id,
-                parameters=call.data,
-                query_dict=call.data,
-                allow_public=True,
-            )
-        conform_output_tags(call, models)
-        call.result.data = {"models": models}
+    Metadata.escape_query_parameters(call)
+    with TimingContext("mongo", "models_get_all"):
+        ret_params = {}
+        models = Model.get_many(
+            company=company_id,
+            parameters=call.data,
+            query_dict=call.data,
+            allow_public=True,
+            ret_params=ret_params,
+        )
+    conform_output_tags(call, models)
+    unescape_metadata(call, models)
+    call.result.data = {"models": models, **ret_params}
 
 
 @endpoint("models.get_frameworks", request_data_model=GetFrameworksRequest)
@@ -183,15 +206,22 @@ create_fields = {
     "metadata": list,
 }
 
-last_update_fields = ("uri", "framework", "design", "labels", "ready", "metadata")
+last_update_fields = (
+    "uri",
+    "framework",
+    "design",
+    "labels",
+    "ready",
+    "metadata",
+    "system_tags",
+    "tags",
+)
 
 
 def parse_model_fields(call, valid_fields):
     fields = parse_from_call(call.data, valid_fields, Model.get_fields())
     conform_tag_fields(call, fields, validate=True)
-    metadata = fields.get("metadata")
-    if metadata:
-        validate_metadata(metadata)
+    escape_metadata(fields)
     return fields
 
 
@@ -225,82 +255,80 @@ def update_for_task(call: APICall, company_id, _):
             "exactly one field is required", fields=("uri", "override_model_id")
         )
 
-    with translate_errors_context():
+    query = dict(id=task_id, company=company_id)
+    task = Task.get_for_writing(
+        id=task_id,
+        company=company_id,
+        _only=["models", "execution", "name", "status", "project"],
+    )
+    if not task:
+        raise errors.bad_request.InvalidTaskId(**query)
 
-        query = dict(id=task_id, company=company_id)
-        task = Task.get_for_writing(
-            id=task_id,
+    allowed_states = [TaskStatus.created, TaskStatus.in_progress]
+    if task.status not in allowed_states:
+        raise errors.bad_request.InvalidTaskStatus(
+            f"model can only be updated for tasks in the {allowed_states} states",
+            **query,
+        )
+
+    if override_model_id:
+        model = ModelBLL.get_company_model_by_id(
+            company_id=company_id, model_id=override_model_id
+        )
+    else:
+        if "name" not in call.data:
+            # use task name if name not provided
+            call.data["name"] = task.name
+
+        if "comment" not in call.data:
+            call.data["comment"] = f"Created by task `{task.name}` ({task.id})"
+
+        if task.models and task.models.output:
+            # model exists, update
+            model_id = task.models.output[-1].model
+            res = _update_model(call, company_id, model_id=model_id).to_struct()
+            res.update({"id": model_id, "created": False})
+            call.result.data = res
+            return
+
+        # new model, create
+        fields = parse_model_fields(call, create_fields)
+
+        # create and save model
+        now = datetime.utcnow()
+        model = Model(
+            id=database.utils.id(),
+            created=now,
+            last_update=now,
+            user=call.identity.user,
             company=company_id,
-            _only=["models", "execution", "name", "status", "project"],
+            project=task.project,
+            framework=task.execution.framework,
+            parent=task.models.input[0].model
+            if task.models and task.models.input
+            else None,
+            design=task.execution.model_desc,
+            labels=task.execution.model_labels,
+            ready=(task.status == TaskStatus.published),
+            **fields,
         )
-        if not task:
-            raise errors.bad_request.InvalidTaskId(**query)
+        model.save()
+        _update_cached_tags(company_id, project=model.project, fields=fields)
 
-        allowed_states = [TaskStatus.created, TaskStatus.in_progress]
-        if task.status not in allowed_states:
-            raise errors.bad_request.InvalidTaskStatus(
-                f"model can only be updated for tasks in the {allowed_states} states",
-                **query,
+    TaskBLL.update_statistics(
+        task_id=task_id,
+        company_id=company_id,
+        last_iteration_max=iteration,
+        models__output=[
+            ModelItem(
+                model=model.id,
+                name=TaskModelNames[TaskModelTypes.output],
+                updated=datetime.utcnow(),
             )
+        ],
+    )
 
-        if override_model_id:
-            model = ModelBLL.get_company_model_by_id(
-                company_id=company_id, model_id=override_model_id
-            )
-        else:
-            if "name" not in call.data:
-                # use task name if name not provided
-                call.data["name"] = task.name
-
-            if "comment" not in call.data:
-                call.data["comment"] = f"Created by task `{task.name}` ({task.id})"
-
-            if task.models and task.models.output:
-                # model exists, update
-                model_id = task.models.output[-1].model
-                res = _update_model(call, company_id, model_id=model_id).to_struct()
-                res.update({"id": model_id, "created": False})
-                call.result.data = res
-                return
-
-            # new model, create
-            fields = parse_model_fields(call, create_fields)
-
-            # create and save model
-            now = datetime.utcnow()
-            model = Model(
-                id=database.utils.id(),
-                created=now,
-                last_update=now,
-                user=call.identity.user,
-                company=company_id,
-                project=task.project,
-                framework=task.execution.framework,
-                parent=task.models.input[0].model
-                if task.models and task.models.input
-                else None,
-                design=task.execution.model_desc,
-                labels=task.execution.model_labels,
-                ready=(task.status == TaskStatus.published),
-                **fields,
-            )
-            model.save()
-            _update_cached_tags(company_id, project=model.project, fields=fields)
-
-        TaskBLL.update_statistics(
-            task_id=task_id,
-            company_id=company_id,
-            last_iteration_max=iteration,
-            models__output=[
-                ModelItem(
-                    model=model.id,
-                    name=TaskModelNames[TaskModelTypes.output],
-                    updated=datetime.utcnow(),
-                )
-            ],
-        )
-
-        call.result.data = {"id": model.id, "created": True}
+    call.result.data = {"id": model.id, "created": True}
 
 
 @endpoint(
@@ -313,36 +341,33 @@ def create(call: APICall, company_id, req_model: CreateModelRequest):
     if req_model.public:
         company_id = ""
 
-    with translate_errors_context():
+    project = req_model.project
+    if project:
+        validate_id(Project, company=company_id, project=project)
 
-        project = req_model.project
-        if project:
-            validate_id(Project, company=company_id, project=project)
+    task = req_model.task
+    req_data = req_model.to_struct()
+    if task:
+        validate_task(company_id, req_data)
 
-        task = req_model.task
-        req_data = req_model.to_struct()
-        if task:
-            validate_task(company_id, req_data)
+    fields = filter_fields(Model, req_data)
+    conform_tag_fields(call, fields, validate=True)
+    escape_metadata(fields)
 
-        fields = filter_fields(Model, req_data)
-        conform_tag_fields(call, fields, validate=True)
+    # create and save model
+    now = datetime.utcnow()
+    model = Model(
+        id=database.utils.id(),
+        user=call.identity.user,
+        company=company_id,
+        created=now,
+        last_update=now,
+        **fields,
+    )
+    model.save()
+    _update_cached_tags(company_id, project=model.project, fields=fields)
 
-        validate_metadata(fields.get("metadata"))
-
-        # create and save model
-        now = datetime.utcnow()
-        model = Model(
-            id=database.utils.id(),
-            user=call.identity.user,
-            company=company_id,
-            created=now,
-            last_update=now,
-            **fields,
-        )
-        model.save()
-        _update_cached_tags(company_id, project=model.project, fields=fields)
-
-        call.result.data_model = CreateModelResponse(id=model.id, created=True)
+    call.result.data_model = CreateModelResponse(id=model.id, created=True)
 
 
 def prepare_update_fields(call, company_id, fields: dict):
@@ -377,6 +402,7 @@ def prepare_update_fields(call, company_id, fields: dict):
             )
 
     conform_tag_fields(call, fields, validate=True)
+    escape_metadata(fields)
     return fields
 
 
@@ -388,89 +414,85 @@ def validate_task(company_id, fields: dict):
 def edit(call: APICall, company_id, _):
     model_id = call.data["model"]
 
-    with translate_errors_context():
-        model = ModelBLL.get_company_model_by_id(
-            company_id=company_id, model_id=model_id
+    model = ModelBLL.get_company_model_by_id(
+        company_id=company_id, model_id=model_id
+    )
+
+    fields = parse_model_fields(call, create_fields)
+    fields = prepare_update_fields(call, company_id, fields)
+
+    for key in fields:
+        field = getattr(model, key, None)
+        value = fields[key]
+        if (
+            field
+            and isinstance(value, dict)
+            and isinstance(field, EmbeddedDocument)
+        ):
+            d = field.to_mongo(use_db_field=False).to_dict()
+            d.update(value)
+            fields[key] = d
+
+    iteration = call.data.get("iteration")
+    task_id = model.task or fields.get("task")
+    if task_id and iteration is not None:
+        TaskBLL.update_statistics(
+            task_id=task_id, company_id=company_id, last_iteration_max=iteration,
         )
 
-        fields = parse_model_fields(call, create_fields)
-        fields = prepare_update_fields(call, company_id, fields)
+    if fields:
+        if any(uf in fields for uf in last_update_fields):
+            fields.update(last_update=datetime.utcnow())
 
-        for key in fields:
-            field = getattr(model, key, None)
-            value = fields[key]
-            if (
-                field
-                and isinstance(value, dict)
-                and isinstance(field, EmbeddedDocument)
-            ):
-                d = field.to_mongo(use_db_field=False).to_dict()
-                d.update(value)
-                fields[key] = d
-
-        iteration = call.data.get("iteration")
-        task_id = model.task or fields.get("task")
-        if task_id and iteration is not None:
-            TaskBLL.update_statistics(
-                task_id=task_id, company_id=company_id, last_iteration_max=iteration,
-            )
-
-        if fields:
-            if any(uf in fields for uf in last_update_fields):
-                fields.update(last_update=datetime.utcnow())
-
-            updated = model.update(upsert=False, **fields)
-            if updated:
-                new_project = fields.get("project", model.project)
-                if new_project != model.project:
-                    _reset_cached_tags(
-                        company_id, projects=[new_project, model.project]
-                    )
-                else:
-                    _update_cached_tags(
-                        company_id, project=model.project, fields=fields
-                    )
-            conform_output_tags(call, fields)
-            call.result.data_model = UpdateResponse(updated=updated, fields=fields)
-        else:
-            call.result.data_model = UpdateResponse(updated=0)
+        updated = model.update(upsert=False, **fields)
+        if updated:
+            new_project = fields.get("project", model.project)
+            if new_project != model.project:
+                _reset_cached_tags(
+                    company_id, projects=[new_project, model.project]
+                )
+            else:
+                _update_cached_tags(
+                    company_id, project=model.project, fields=fields
+                )
+        conform_output_tags(call, fields)
+        unescape_metadata(call, fields)
+        call.result.data_model = UpdateResponse(updated=updated, fields=fields)
+    else:
+        call.result.data_model = UpdateResponse(updated=0)
 
 
 def _update_model(call: APICall, company_id, model_id=None):
     model_id = model_id or call.data["model"]
 
-    with translate_errors_context():
-        model = ModelBLL.get_company_model_by_id(
-            company_id=company_id, model_id=model_id
+    model = ModelBLL.get_company_model_by_id(
+        company_id=company_id, model_id=model_id
+    )
+
+    data = prepare_update_fields(call, company_id, call.data)
+
+    task_id = data.get("task")
+    iteration = data.get("iteration")
+    if task_id and iteration is not None:
+        TaskBLL.update_statistics(
+            task_id=task_id, company_id=company_id, last_iteration_max=iteration,
         )
 
-        data = prepare_update_fields(call, company_id, call.data)
+    updated_count, updated_fields = Model.safe_update(company_id, model.id, data)
+    if updated_count:
+        if any(uf in updated_fields for uf in last_update_fields):
+            model.update(upsert=False, last_update=datetime.utcnow())
 
-        task_id = data.get("task")
-        iteration = data.get("iteration")
-        if task_id and iteration is not None:
-            TaskBLL.update_statistics(
-                task_id=task_id, company_id=company_id, last_iteration_max=iteration,
+        new_project = updated_fields.get("project", model.project)
+        if new_project != model.project:
+            _reset_cached_tags(company_id, projects=[new_project, model.project])
+        else:
+            _update_cached_tags(
+                company_id, project=model.project, fields=updated_fields
             )
-
-        metadata = data.get("metadata")
-        if metadata:
-            validate_metadata(metadata)
-
-        updated_count, updated_fields = Model.safe_update(company_id, model.id, data)
-        if updated_count:
-            if any(uf in updated_fields for uf in last_update_fields):
-                model.update(upsert=False, last_update=datetime.utcnow())
-
-            new_project = updated_fields.get("project", model.project)
-            if new_project != model.project:
-                _reset_cached_tags(company_id, projects=[new_project, model.project])
-            else:
-                _update_cached_tags(
-                    company_id, project=model.project, fields=updated_fields
-                )
-        conform_output_tags(call, updated_fields)
-        return UpdateResponse(updated=updated_count, fields=updated_fields)
+    conform_output_tags(call, updated_fields)
+    unescape_metadata(call, updated_fields)
+    return UpdateResponse(updated=updated_count, fields=updated_fields)
 
 
 @endpoint(
@@ -635,26 +657,25 @@ def add_or_update_metadata(
     _: APICall, company_id: str, request: AddOrUpdateMetadataRequest
 ):
     model_id = request.model
-    ModelBLL.get_company_model_by_id(company_id=company_id, model_id=model_id)
-
-    updated = metadata_add_or_update(
-        cls=Model, _id=model_id, items=get_metadata_from_api(request.metadata),
-    )
-    if updated:
-        Model.objects(id=model_id).update_one(last_update=datetime.utcnow())
-
-    return {"updated": updated}
+    model = ModelBLL.get_company_model_by_id(company_id=company_id, model_id=model_id)
+    return {
+        "updated": Metadata.edit_metadata(
+            model,
+            items=request.metadata,
+            replace_metadata=request.replace_metadata,
+            last_update=datetime.utcnow(),
+        )
+    }
 
 
 @endpoint("models.delete_metadata", min_version="2.13")
 def delete_metadata(_: APICall, company_id: str, request: DeleteMetadataRequest):
     model_id = request.model
-    ModelBLL.get_company_model_by_id(
+    model = ModelBLL.get_company_model_by_id(
         company_id=company_id, model_id=model_id, only_fields=("id",)
     )
-
-    updated = metadata_delete(cls=Model, _id=model_id, keys=request.keys)
-    if updated:
-        Model.objects(id=model_id).update_one(last_update=datetime.utcnow())
-
-    return {"updated": updated}
+    return {
+        "updated": Metadata.delete_metadata(
+            model, keys=request.keys, last_update=datetime.utcnow()
+        )
+    }

apiserver/services/organization.py

@@ -1,11 +1,17 @@
 from collections import defaultdict
 from operator import itemgetter
+from typing import Mapping, Type
 
-from apiserver.apimodels.organization import TagsRequest
+from mongoengine import Q
+
+from apiserver.apimodels.organization import TagsRequest, EntitiesCountRequest
 from apiserver.bll.organization import OrgBLL, Tags
-from apiserver.database.model import User
+from apiserver.database.model import User, AttributedDocument, EntityVisibility
+from apiserver.database.model.model import Model
+from apiserver.database.model.project import Project
+from apiserver.database.model.task.task import Task
 from apiserver.service_repo import endpoint, APICall
-from apiserver.services.utils import get_tags_filter_dictionary, get_tags_response
+from apiserver.services.utils import get_tags_filter_dictionary, sort_tags_response
 
 org_bll = OrgBLL()
 
@@ -21,17 +27,13 @@ def get_tags(call: APICall, company, request: TagsRequest):
         for field, vals in tags.items():
             ret[field] |= vals
 
-    call.result.data = get_tags_response(ret)
+    call.result.data = sort_tags_response(ret)
 
 
 @endpoint("organization.get_user_companies")
 def get_user_companies(call: APICall, company_id: str, _):
     users = [
-        {
-            "id": u.id,
-            "name": u.name,
-            "avatar": u.avatar,
-        }
+        {"id": u.id, "name": u.name, "avatar": u.avatar}
         for u in User.objects(company=company_id).only("avatar", "name", "company")
     ]
 
@@ -45,3 +47,29 @@ def get_user_companies(call: APICall, company_id: str, _):
             }
         ]
     }
+
+
+@endpoint("organization.get_entities_count", request_data_model=EntitiesCountRequest)
+def get_entities_count(call: APICall, company, _):
+    entity_classes: Mapping[str, Type[AttributedDocument]] = {
+        "projects": Project,
+        "tasks": Task,
+        "models": Model,
+        "pipelines": Project,
+        "datasets": Project,
+    }
+    ret = {}
+    for field, entity_cls in entity_classes.items():
+        data = call.data.get(field)
+        if data is None:
+            continue
+
+        query = Q()
+        if entity_cls in (Project, Task) and not data.get("search_hidden"):
+            query &= Q(system_tags__ne=EntityVisibility.hidden.value)
+
+        ret[field] = entity_cls.get_count(
+            company=company, query_dict=data, query=query, allow_public=True,
+        )
+
+    call.result.data = ret

apiserver/services/pipelines.py

@@ -0,0 +1,68 @@
+import re
+
+from apiserver.apimodels.pipelines import StartPipelineResponse, StartPipelineRequest
+from apiserver.bll.organization import OrgBLL
+from apiserver.bll.project import ProjectBLL
+from apiserver.bll.task import TaskBLL
+from apiserver.bll.task.task_operations import enqueue_task
+from apiserver.database.model.project import Project
+from apiserver.database.model.task.task import Task
+from apiserver.service_repo import APICall, endpoint
+
+org_bll = OrgBLL()
+project_bll = ProjectBLL()
+task_bll = TaskBLL()
+
+
+def _update_task_name(task: Task):
+    if not task or not task.project:
+        return
+
+    project = Project.objects(id=task.project).only("name").first()
+    if not project:
+        return
+
+    _, _, name_prefix = project.name.rpartition("/")
+    name_mask = re.compile(rf"{re.escape(name_prefix)}( #\d+)?$")
+    count = Task.objects(
+        project=task.project, system_tags__in=["pipeline"], name=name_mask
+    ).count()
+    new_name = f"{name_prefix} #{count}" if count > 0 else name_prefix
+    task.update(name=new_name)
+
+
+@endpoint(
+    "pipelines.start_pipeline", response_data_model=StartPipelineResponse,
+)
+def start_pipeline(call: APICall, company_id: str, request: StartPipelineRequest):
+    hyperparams = None
+    if request.args:
+        hyperparams = {
+            "Args": {
+                str(arg.name): {
+                    "section": "Args",
+                    "name": str(arg.name),
+                    "value": str(arg.value),
+                }
+                for arg in request.args or []
+            }
+        }
+
+    task, _ = task_bll.clone_task(
+        company_id=company_id,
+        user_id=call.identity.user,
+        task_id=request.task,
+        hyperparams=hyperparams,
+    )
+
+    _update_task_name(task)
+
+    queued, res = enqueue_task(
+        task_id=task.id,
+        company_id=company_id,
+        queue_id=request.queue,
+        status_message="Starting pipeline",
+        status_reason="",
+    )
+
+    return StartPipelineResponse(pipeline=task.id, enqueued=bool(queued))