ClearML/clearml-agent
1
0
Fork 0
mirror of https://github.com/clearml/clearml-agent synced 2025-06-26 18:16:15 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Commit Graph

  • bc85ddf78d Fix pytorch direct resolve replacing wheel link with directly installed version allegroai 2023-05-21 22:53:51 +03:00
  • 5b5fb0b8a6 Add agent.package_manager.pytorch_resolve configuration setting with pip or direct values. pip sets extra index based on cuda and lets pip resolve, direct is the previous parsing algorithm that does the matching and downloading (default pip) allegroai 2023-05-21 22:53:11 +03:00
  • fec0ce1756 Better message for agent init when an existing clearml.conf is found allegroai 2023-05-21 22:51:11 +03:00
  • 1e09b88b7a Add alias CLEARML_AGENT_DOCKER_AGENT_REPO env var for the FORCE_CLEARML_AGENT_REPO env var allegroai 2023-05-21 22:50:01 +03:00
  • b6ca0fa6a5 Print error on resource monitor failure v1.5.3rc2 allegroai 2023-05-11 16:18:11 +03:00
  • 307ec9213e Fix git+ssh:// links inside installed packages not being converted properly to HTTPS authenticated and vice versa v1.5.3rc1 allegroai 2023-05-11 16:16:51 +03:00
  • a78a25d966 Support new Retry.DEFAULT_BACKOFF_MAX in a backwards-compatible way allegroai 2023-05-11 16:16:18 +03:00
  • ebb6231f5a Add CLEARML_AGENT_STANDALONE_CONFIG_BC to support backwards compatibility in standalone mode v1.5.3rc0 allegroai 2023-05-11 16:15:06 +03:00
  • e1d65cb280 Update clearml-agent gif (#137) pollfly 2023-04-10 10:58:10 +03:00
  • 3fe92a92ba Version bump to v1.5.2 v1.5.2 allegroai 2023-03-29 12:49:33 +03:00
  • 154db59ce6 Add agent.package_manager.poetry_install_extra_args configuration option allegroai 2023-03-28 14:37:48 +03:00
  • afffa83063 Fix git+ssh:// links inside installed packages not being converted properly to https authenticated links allegroai 2023-03-28 14:35:51 +03:00
  • 787c7d88bb Fix additional poetry cwd support feature allegroai 2023-03-28 14:35:41 +03:00
  • 667c2ced3d Fix very old pip version support (<20) allegroai 2023-03-28 14:34:19 +03:00
  • 7f5b3c8df4 Fix None config file in session causes k8s agent to raise exception allegroai 2023-03-28 14:33:55 +03:00
  • 46ded2864d Fix restart feature should be tested against agent session allegroai 2023-03-28 14:33:33 +03:00
  • 40456be948 Black formatting Refactor path support allegroai 2023-03-05 18:04:35 +02:00
  • 8d51aed679 Protect against cache folders without permission allegroai 2023-03-05 18:03:39 +02:00
  • bfc4ba38cd Fix torch inside nvidia containers to use preinstalled version (i.e. ==x.y.z.* matching) allegroai 2023-03-05 17:59:57 +02:00
  • 3cedc104df Add poetry cwd support (#142) Niels ten Boom 2023-03-05 13:19:57 +01:00
  • b367c80477 Switch entrypoint shell from sh to bash (#141) Marijan Smetko 2023-02-28 20:55:16 +01:00
  • 262b6d3a00 Update services agent entrypoint allegroai 2023-02-05 10:40:02 +02:00
  • 95e996bfda Reintroduce CLEARML_AGENT_SERVICES_DOCKER_RESTART accidentally reverted by a previous merge allegroai 2023-02-05 10:34:38 +02:00
  • b6d132b226 Fix build fails when target is relative path allegroai 2023-02-05 10:33:32 +02:00
  • 4f17a2c17d Fix K8s glue does not delete pending pods if the tasks they represent were aborted allegroai 2023-02-05 10:32:16 +02:00
  • 00e8e9eb5a Do not allow request exceptions (only on the initial login call) allegroai 2023-02-05 10:30:45 +02:00
  • af6a77918f Fix _ is allowed in k8s label names allegroai 2023-02-05 10:29:48 +02:00
  • 855622fd30 Support custom service on Worker.get() calls allegroai 2023-02-05 10:29:09 +02:00
  • 8cd12810f3 Fix login uses GET with payload which breaks when trying to connect a server running in GCP allegroai 2023-02-05 10:28:41 +02:00
  • ebb955187d Fix agent update version (#132) achaiah 2023-01-08 17:10:26 +00:00
  • 85e1fadf9b Fix typos (#131) pollfly 2022-12-28 19:39:59 +02:00
  • 249b51a31b Version bump v1.5.2rc0 allegroai 2022-12-13 15:29:10 +02:00
  • da19ef26c4 Fix pinging running task (and change default to once a minute) allegroai 2022-12-13 15:26:26 +02:00
  • f69e16ea9d Fix clearml-agent build --docker stuck on certain containers allegroai 2022-12-13 15:24:32 +02:00
  • efa1f71dac Version bump to v1.5.1 v1.5.1 allegroai 2022-12-10 22:18:21 +02:00
  • 692cb8cf13 Update six requirements allegroai 2022-12-10 22:18:10 +02:00
  • ebdc215632 Remove " from pip commands in venv v1.5.1rc2 allegroai 2022-12-10 20:58:30 +02:00
  • b2da639582 Add CLEARML_AGENT_FORCE_SYSTEM_SITE_PACKAGES env var (default true) to allow overriding default "system_site_packages: true" behavior when running tasks in containers (docker mode and k8s-glue) v1.5.1rc1 allegroai 2022-12-10 20:00:46 +02:00
  • 71fdb43f10 Version bump to v1.5.1rc0 v1.5.1rc0 allegroai 2022-12-07 22:09:40 +02:00
  • ca2791c65e Fix pip support allowing multiple pip version constraints (by default, one for <PY3.10 and one for >=PY3.10) allegroai 2022-12-07 22:09:25 +02:00
  • dd75cedaab Upgrade requirements for attrs, jsonschema, pyparsing and pyjwt allegroai 2022-12-07 22:08:15 +02:00
  • 669fb1a6e5 Fix using deprecated types validator argument raises an error (deprecated even before jsonschema 3.0.0 and unsupported since 4.0.0) allegroai 2022-12-07 22:07:53 +02:00
  • 5d517c91b5 Add agent.disable_task_docker_override configuration option to disable docker override specified in executing tasks allegroai 2022-12-07 22:07:11 +02:00
  • 6be75abc86 Add default output URI selection to "clearml-agent init" allegroai 2022-12-07 22:06:10 +02:00
  • 4c777fa2ee Version bump to v1.5.0 v1.5.0 allegroai 2022-12-05 16:42:44 +02:00
  • dc5e0033c8 Remove support for kubectl run Allow customizing pod name prefix and limit pod label Return deleted pods from cleanup Some refactoring allegroai 2022-12-05 11:40:19 +02:00
  • 3dd5973734 Filter by phase when detecting hanging pods More debug print-outs Use task session when possible Push task into k8s scheduler queue only if running from the same tenant Make sure we pass git_user/pass to the task pod Fix cleanup command not issued when no pods exist in a multi-queue setup allegroai 2022-12-05 11:29:59 +02:00
  • 53d379205f Support raise_error in get_bash_output() allegroai 2022-12-05 11:26:40 +02:00
  • 57cde21c48 Send task.ping for executing tasks every 120 seconds (set using the agent.task_ping_interval_sec configuration option) allegroai 2022-12-05 11:22:25 +02:00
  • 396abf13b6 Fix get_task_session() may cause an old copy of the APIClient to be used containing a reference to the previous session allegroai 2022-12-05 11:20:32 +02:00
  • 6e7fb5f331 Fix sending task logs fails when agent is not running in the same tenant allegroai 2022-12-05 11:19:14 +02:00
  • 1d5c118b70 Fix setting CLEARML_API_DEFAULT_REQ_METHOD raises an error allegroai 2022-12-05 11:18:12 +02:00
  • 18612aac4d Improve configuration examples allegroai 2022-12-05 11:17:27 +02:00
  • 76c533a2e8 Fix access to config object allegroai 2022-11-11 13:34:17 +02:00
  • 9eee213683 Add option to crash agent on exception using agent.crash_on_exception configuration setting (#123) Niels ten Boom 2022-11-06 16:15:39 +01:00
  • e4861fc0fb Add missing settings in clearml.conf allegroai 2022-11-06 12:36:01 +02:00
  • 53ef984065 Update README allegroai 2022-11-06 11:53:16 +02:00
  • 26e62da1a8 version bump to 1.5.0rc0 v1.5.0rc0 allegroai 2022-10-23 13:03:04 +03:00
  • d2f3614ab0 Add support for CLEARML_AGENT_DOCKER_ARGS_HIDE_ENV environment variable (see agent.hide_docker_command_env_vars config option) allegroai 2022-10-23 13:02:46 +03:00
  • c6d767bd64 Make venv caching the default behavior allegroai 2022-10-23 12:59:55 +03:00
  • efb06891a8 Add support for PyTorch new extra_index_url repo support. We will find the correct index url based on the cuda version, and let pip do the rest. allegroai 2022-10-23 12:59:29 +03:00
  • 70771b12a9 Remove unused code allegroai 2022-10-23 12:58:49 +03:00
  • 3f7a4840cc Add support for operator != in package version (mostly for pytorch resolving) allegroai 2022-10-23 12:58:25 +03:00
  • e28048dc25 Change default pip version used to "pip<21" for better Python 3.10 support allegroai 2022-10-23 12:58:01 +03:00
  • 2ef5d38b32 Remove future (Python 2 is not supported for clearml-agent) allegroai 2022-10-23 12:56:48 +03:00
  • d216d70cdf Upgrade packages for better Python 3.10 support allegroai 2022-10-23 12:56:17 +03:00
  • 0de10345f7 Moved pyhocon to internal packages allegroai 2022-10-23 12:55:41 +03:00
  • a243fa211f Improve venv cache disabled message allegroai 2022-10-23 12:53:25 +03:00
  • d794b047be Fix system_site_packages is not turned on in k8s glue allegroai 2022-10-23 12:37:35 +03:00
  • f0fd62a28f Fix docker extra args showing up in configuration printout allegroai 2022-10-23 12:37:13 +03:00
  • e8493d3807 Refactor override configuration to a method allegroai 2022-10-23 12:36:06 +03:00
  • 5353e9c44d Update README.md Allegro AI 2022-10-19 02:47:10 +03:00
  • 75f5814f9f Update README.md Allegro AI 2022-10-19 02:44:53 +03:00
  • 94b8b5520d Update README.md Allegro AI 2022-10-19 02:18:56 +03:00
  • 42450dcbc4 Update clearml.conf v1.4.1 allegroai 2022-10-07 15:33:19 +03:00
  • ef47225d41 Version bump to v1.4.1 allegroai 2022-10-07 15:27:49 +03:00
  • e61accefb9 PEP8 + refactor allegroai 2022-10-07 15:26:31 +03:00
  • 5c1543d112 Add agent.disable_ssh_mount configuration option (same as CLEARML_AGENT_DISABLE_SSH_MOUNT env var) allegroai 2022-10-07 15:24:39 +03:00
  • 7ff6aee20c Add warning if venv cache is disabled allegroai 2022-10-07 15:23:10 +03:00
  • 37ea381d98 Add support for docker args filters allegroai 2022-10-07 15:22:42 +03:00
  • 67fc884895 Fix --gpus all not reporting GPU stats on worker machine allegroai 2022-10-07 15:22:13 +03:00
  • 1e3646b57c Fix docker command for monitoring child agents allegroai 2022-10-07 15:21:32 +03:00
  • ba2db4e727 Version bump to v1.4.0 v1.4.0 allegroai 2022-09-29 18:21:04 +03:00
  • 077148be00 version bump allegroai 2022-09-16 17:29:42 +03:00
  • 594ee5842e Allow to pverride pytorch lookup page: "agent.package_manager.torch_page / torch_nightly_page / torch_url_template_prefix" allegroai 2022-09-15 20:16:41 +03:00
  • a69766bd8b Add CLEARML_AGENT_CHILD_AGENTS_COUNT_CMD to allow overriding child agent count command in k8s allegroai 2022-09-15 20:16:01 +03:00
  • 857a750eb1 Fix GCP load balancer not fwd GET request body, allow to change default request Action to Put/Post/Get. see api.http.default_method or CLEARML_API_DEFAULT_REQ_METHOD allegroai 2022-09-15 20:15:42 +03:00
  • 26aa50f1b5 Fix k8s glue extra_bash_init_cmd location in initial bash script v1.4.0rc3 allegroai 2022-09-02 23:50:03 +03:00
  • 8b4f1eefc2 Add more debug printouts in k8s glue v1.4.0rc2 allegroai 2022-09-02 23:49:19 +03:00
  • 97c2e21dcc Fix resolving k8s pending queue may cause a queue with a uuid name to be created allegroai 2022-09-02 23:45:24 +03:00
  • 918dd39b87 Add docker ssh_ro_folder (default: "/.ssh") changed docker ssh_folder (default: "~/.ssh") allegroai 2022-09-02 23:44:36 +03:00
  • 7776e906c4 Fix second .ssh temp mount fails if container changes the files inside allegroai 2022-09-02 23:43:58 +03:00
  • 1bf865ec08 Fix name not escaped as regex (all services "get_all" use regex for name) allegroai 2022-09-02 23:43:42 +03:00
  • d74b9f105b Fix resolving k8s pending queue may cause a queue with a uuid name to be created v1.4.0rc1 allegroai 2022-09-02 23:45:24 +03:00
  • d0b8eeed15 Add docker ssh_ro_folder (default: "/.ssh") changed docker ssh_folder (default: "~/.ssh") allegroai 2022-09-02 23:44:36 +03:00
  • 221db3e175 Fix second .ssh temp mount fails if container changes the files inside v1.4.0rc0 allegroai 2022-09-02 23:43:58 +03:00
  • 2c71f9a821 Fix name not escaped as regex (all services "get_all" use regex for name) allegroai 2022-09-02 23:43:42 +03:00
  • 3f1ce847dc Fixed documentation (#117) Luca Cerone 2022-09-01 16:18:48 +02:00
  • 9006c2d28f Add support for abort callback registration allegroai 2022-08-29 18:06:59 +03:00
  • ec216198a0 Add agent.enable_git_ask_pass to improve passing user/pass to git commands allegroai 2022-08-29 18:06:26 +03:00