Fix glue namespace handling (#63)

* Changed: namespace handling for glue

* Changed: set glue as default agent system

* Changed: bump up version

charts/clearml/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: clearml
 description: MLOps platform
 type: application
-version: "3.9.0"
+version: "3.10.0"
 appVersion: "1.4.0"
 home: https://clear.ml
 icon: https://raw.githubusercontent.com/allegroai/clearml/master/docs/clearml-logo.svg

charts/clearml/README.md

@@ -1,6 +1,6 @@
 # ClearML Ecosystem for Kubernetes
 
-![Version: 3.9.0](https://img.shields.io/badge/Version-3.9.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.4.0](https://img.shields.io/badge/AppVersion-1.4.0-informational?style=flat-square)
+![Version: 3.10.0](https://img.shields.io/badge/Version-3.10.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 1.4.0](https://img.shields.io/badge/AppVersion-1.4.0-informational?style=flat-square)
 
 MLOps platform
 
@@ -141,7 +141,7 @@ For detailed instructions, see the [Optional Configuration](https://github.com/a
 | agentGroups.agent-group-cpu.clearmlGitPassword | string | `nil` |  |
 | agentGroups.agent-group-cpu.clearmlGitUser | string | `nil` |  |
 | agentGroups.agent-group-cpu.clearmlSecretKey | string | `nil` |  |
-| agentGroups.agent-group-cpu.enabled | bool | `true` |  |
+| agentGroups.agent-group-cpu.enabled | bool | `false` |  |
 | agentGroups.agent-group-cpu.image.pullPolicy | string | `"IfNotPresent"` |  |
 | agentGroups.agent-group-cpu.image.repository | string | `"ubuntu"` |  |
 | agentGroups.agent-group-cpu.image.tag | string | `"18.04"` |  |
@@ -165,7 +165,7 @@ For detailed instructions, see the [Optional Configuration](https://github.com/a
 | agentGroups.agent-group-gpu.clearmlGitPassword | string | `nil` |  |
 | agentGroups.agent-group-gpu.clearmlGitUser | string | `nil` |  |
 | agentGroups.agent-group-gpu.clearmlSecretKey | string | `nil` |  |
-| agentGroups.agent-group-gpu.enabled | bool | `true` |  |
+| agentGroups.agent-group-gpu.enabled | bool | `false` |  |
 | agentGroups.agent-group-gpu.image.pullPolicy | string | `"IfNotPresent"` |  |
 | agentGroups.agent-group-gpu.image.repository | string | `"nvidia/cuda"` |  |
 | agentGroups.agent-group-gpu.image.tag | string | `"11.0-base-ubuntu18.04"` |  |
@@ -178,17 +178,17 @@ For detailed instructions, see the [Optional Configuration](https://github.com/a
 | agentGroups.agent-group-gpu.tolerations | list | `[]` |  |
 | agentGroups.agent-group-gpu.updateStrategy | string | `"Recreate"` |  |
 | agentk8sglue.defaultDockerImage | string | `"nvidia/cuda:11.3.1-cudnn8-runtime-ubuntu20.04"` |  |
-| agentk8sglue.enabled | bool | `false` |  |
+| agentk8sglue.enabled | bool | `true` |  |
 | agentk8sglue.id | string | `"k8s-agent"` |  |
 | agentk8sglue.image.repository | string | `"allegroai/clearml-agent-k8s"` |  |
-| agentk8sglue.image.tag | string | `"aws-latest-1.21"` |  |
+| agentk8sglue.image.tag | string | `"latest"` |  |
 | agentk8sglue.maxPods | int | `10` |  |
 | agentk8sglue.podTemplate.env | list | `[]` |  |
 | agentk8sglue.podTemplate.nodeSelector | object | `{}` |  |
 | agentk8sglue.podTemplate.resources | object | `{}` |  |
 | agentk8sglue.podTemplate.tolerations | list | `[]` |  |
 | agentk8sglue.podTemplate.volumes | list | `[]` |  |
-| agentk8sglue.queue | string | `"aws-instances"` |  |
+| agentk8sglue.queue | string | `"default"` |  |
 | agentk8sglue.serviceAccountName | string | `"default"` |  |
 | agentservices.affinity | object | `{}` |  |
 | agentservices.agentVersion | string | `""` |  |

charts/clearml/templates/configmap-agentk8s-template.yaml

@@ -7,7 +7,7 @@ data:
   template.yaml: |
     apiVersion: v1
     metadata:
-      namespace: {{ .Release.namespace }}
+      namespace: {{ .Release.Namespace }}
     spec:
       serviceAccountName: {{ .Values.agentk8sglue.serviceAccountName }}
       volumes:

charts/clearml/templates/deployment-agentk8s.yaml

@@ -35,9 +35,9 @@ spec:
           - name: K8S_GLUE_QUEUE
             value: "{{.Values.agentk8sglue.queue}}"
           - name: K8S_GLUE_EXTRA_ARGS
-            value: "--template-yaml /root/template/template.yaml"
+            value: "--namespace {{ .Release.Namespace }} --template-yaml /root/template/template.yaml"
           - name: K8S_DEFAULT_NAMESPACE
-            value: "{{ .Release.namespace }}"
+            value: "{{ .Release.Namespace }}"
           - name: CLEARML_API_ACCESS_KEY
             valueFrom:
               secretKeyRef:

charts/clearml/templates/rbac-agentk8s.yaml

@@ -17,7 +17,7 @@ metadata:
 subjects:
   - kind: ServiceAccount
     name: default
-    namespace: {{ .Release.namespace }}
+    namespace: {{ .Release.Namespace }}
 roleRef:
   apiGroup: rbac.authorization.k8s.io
   kind: Role

charts/clearml/values.yaml

@@ -257,7 +257,7 @@ agentservices:
 
 agentGroups:
   agent-group-cpu:
-    enabled: true
+    enabled: false
     name: agent-group-cpu
     replicaCount: 1
     updateStrategy: Recreate
@@ -291,7 +291,7 @@ agentGroups:
     affinity: {}
 
   agent-group-gpu:
-    enabled: true
+    enabled: false
     name: agent-group-gpu
     replicaCount: 0
     updateStrategy: Recreate
@@ -328,14 +328,14 @@ agentGroups:
 # GPU autoscaling nodes.
 # https://github.com/allegroai/clearml-agent/tree/master/docker/k8s-glue
 agentk8sglue:
-  enabled: false
+  enabled: true
   image:
     repository: "allegroai/clearml-agent-k8s"
-    tag: "aws-latest-1.21"
+    tag: "latest"
   serviceAccountName: default
   maxPods: 10
   defaultDockerImage: nvidia/cuda:11.3.1-cudnn8-runtime-ubuntu20.04   # default docker image that is spawned as new pod
-  queue: aws-instances    # create this queue manually in the UI first for it to work
+  queue: default
   id: k8s-agent
   podTemplate:
     volumes: []