ClearML/clearml-agent
1
0
Fork 0
mirror of https://github.com/clearml/clearml-agent synced 2025-04-02 11:59:36 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Commit Graph

  • ee21944f6b Fix UV pip freeze fails master v2.0.0rc2 clearml 2025-04-02 14:42:02 +0300
  • 55790b3c3a Add support for ${CLEARML_TASK.xx} as docker args arguments parsed based on Tasks value clearml 2025-04-02 14:41:35 +0300
  • bb9ad6b213 PEP8 clearml 2025-04-02 14:40:19 +0300
  • 94fc0138b5 Update base images to resolve several vulnerabilities clearml 2025-04-02 14:39:34 +0300
  • f2df45cad6 Refactor clearml 2025-04-02 14:39:14 +0300
  • 326ba81105 Refactor clearml 2025-04-02 14:38:58 +0300
  • 6dedb4ab41
         Merge f640eb6707 into 1926673951 mads-oestergaard 2025-03-10 22:27:14 +0100
  • 1926673951 Fix Argv stores env/cwd if passed in ctor v2.0.0rc1 clearml 2025-03-03 19:23:51 +0200
  • 3b70e1c4a0 Fix uv_replace_pip feature Fix UV cache based on sync/pip-replacement Fix if UV fails but lock file is missing, revert to UV as pip drop in replacement Fix use UV bin instead of UV python package to avoid nested VIRTUAL_ENV issues clearml 2025-03-03 19:23:29 +0200
  • eee261685f Fix bash on_abort callback in services mode clearml 2025-03-03 19:19:57 +0200
  • 66494c598d Fix on_abort bash callback, if main processes leave while on_abort callback is running, wait for the on_abort to complete clearml 2025-02-24 13:46:55 +0200
  • d30d4e7e61 Fix session should retry on Any error if send fails clearml 2025-02-24 13:46:29 +0200
  • 0e2657421f Added CLEARML_AGENT_ABORT_CALLBACK_CMD and CLEARML_AGENT_ABORT_CALLBACK_TIMEOUT (default 180 sec) to define callback command to be called on abort status change clearml 2025-02-24 13:46:00 +0200
  • ee286e2fb7 Fix container default arguments should never be a list clearml 2025-02-24 13:44:52 +0200
  • d87521c36c Add support for container rulebook overrides ('force_container_rules: true') and container rulebook task update ('update_back_task: true'). This addition allows users to override container arguments forcefully based on the tasks properties (repo, tags, project, user etc.), as well as offer additional defaults based on python required packages or python versions clearml 2025-02-24 13:44:26 +0200
  • 8887453328 Cleanup error prints on bash startup script clearml 2025-02-24 13:42:37 +0200
  • 8d3cb34390 Add default support for dns i.e. rocky/centos/fedora containers clearml 2025-02-24 13:41:32 +0200
  • 528bf314ef Update GpuFractionsHandler GPU name to mem size clearml 2025-02-24 13:30:30 +0200
  • 35f820586d Bump urllib3 to "<3" to support new urllib3 v2 clearml 2025-02-24 13:30:08 +0200
  • 4f91c45d38 Fix untitled file based on binary is now py/sh based on requested binary clearml 2025-02-24 13:29:56 +0200
  • 0a13fd79fc Make sure that if we fail to kill a child processes we continue to try the rest v2.0.0rc0 clearml 2025-02-24 13:26:49 +0200
  • c9f5b3d19a Force the stop command to avoid a potential race clearml 2025-02-24 13:26:26 +0200
  • 4219835aa1 Fix pip requirements print dump should be sorted clearml 2025-02-24 13:25:23 +0200
  • d32b82cb01 Integrate docker port mapping, to control non network=host port mapping, including port reassigning for multiple running agents on the same machine clearml 2025-02-24 13:24:58 +0200
  • 97cb47d48e Add docker port mapping parsing and reassigning feature support Add initial component import from clearml-sdk for easier integration clearml 2025-02-24 13:24:35 +0200
  • 8f28d2882a Fix pip freeze dump to comply with yaml fancy print clearml 2025-02-24 13:23:50 +0200
  • 546ffff95d Fix cached venv tried to reinstall priority packages even through they are preinstalled clearml 2025-02-24 13:23:00 +0200
  • a6ae6b2095 Add initial support for --break-system-packages version detection, but the reality is that we do not need it, because by the time we are running it is too late, so we do rm /usr/lib/python3.*/EXTERNALLY-MANAGED clearml 2025-01-26 23:07:09 +0200
  • 369b440b96 Reduce required packages clearml 2025-01-26 23:05:57 +0200
  • 28e9280a4f Reduce required packages clearml 2025-01-26 23:03:16 +0200
  • 7e9e3ad08b Add printout when using custom configuration file clearml 2025-01-26 22:51:09 +0200
  • 44709673f4 Add CLEARML_AGENT_CONFIG_VERBOSE for verbose configuration file loading clearml 2025-01-26 22:50:49 +0200
  • 823c92ba0e Add SECURITY.md clearml 2025-01-26 22:49:14 +0200
  • 4158146420 Version bump to v1.9.3 v1.9.3 clearml 2025-01-19 16:17:56 +0200
  • b9ef1a55cd Fix dependency on windows clearml 2025-01-19 16:16:54 +0200
  • 9fa8d72640 Update github repo link clearml 2025-01-13 18:36:16 +0200
  • e535390815 Add win32file dependency on windows clearml 2025-01-13 18:34:48 +0200
  • 72482ca394
         Merge f7d4e577ac into 91dfa09466 Emir İşman 2025-01-08 15:35:31 -0500
  • 91dfa09466 Fix Python 3.13 support v1.9.3rc2 clearml 2025-01-05 12:14:24 +0200
  • f110bbf5b4 Remove Python 3.5 support clearml 2025-01-05 12:13:57 +0200
  • 070919973b Fix python 3.6 compatibility, no := operator clearml 2025-01-05 12:13:21 +0200
  • 47d35ef48f Fix managed python environment inside container (PEP 668) remove usr/lib/python3.*/EXTERNALLY-MANAGED v1.9.3rc1 clearml 2024-12-26 18:59:42 +0200
  • 54ed234fca Add agent.docker_args_filters to configuration docs clearml 2024-12-26 18:58:58 +0200
  • a26860e79f Fix default value handling in merge_dicts() clearml 2024-12-26 18:58:24 +0200
  • fc1abbab0b Refactor k8s glue clearml 2024-12-26 18:58:00 +0200
  • 4fa61dde1f Support ignoring kubectl errors clearml 2024-12-12 23:41:31 +0200
  • 26d748a4d8 Support creating queue with tags clearml 2024-12-12 23:40:57 +0200
  • 5419fd84ae Add support for Python 3.13 clearml 2024-12-12 23:39:11 +0200
  • d8366dedc6 Fix UV priority Fix UV cache is disabled, UV handles its own cache Fix UV freeze Fix make sure we do not use pip cache if poetry/uv is used (even if we reverted to pip we can't know if someone changed the repository and now in a new version, a lock file exists) clearml 2024-12-12 23:38:42 +0200
  • cc656e2969
         Add support for uv as package manager (#218) mads-oestergaard 2024-11-27 12:44:55 +0100
  • b65e5fed94 Scan more Python 3 versions clearml 2024-11-17 13:55:51 +0200
  • 16a6701d2a rename to sync_extra_args and set UV_CACHE_DIR Mads Oestergaard 2024-11-04 10:05:56 +0100
  • 3d06b410c4 Update configs to highlight sync command Mads Oestergaard 2024-11-04 10:05:39 +0100
  • fb7011fe13 update environ Mads Oestergaard 2024-10-31 11:21:35 +0100
  • dbea22d725 update worker and defs Mads Oestergaard 2024-10-31 11:20:35 +0100
  • 536911b3a8 update configs Mads Oestergaard 2024-10-31 11:20:27 +0100
  • 11bf855547 add uv as a package manager Mads Oestergaard 2024-10-31 11:20:09 +0100
  • 3273f76b46 Version bump to v1.9.2 v1.9.2 clearml 2024-10-28 18:33:04 +0200
  • 9af0f9fe41 Fix reload method is found in the config object clearml 2024-10-28 18:12:22 +0200
  • 205cd47cb9 Fix use req_token_expiration_sec when creating a task session and not the default value clearml 2024-10-28 18:11:42 +0200
  • 0ff428bb96 Fix report index not advancing in resource monitoring causes more than one GPU not to be reported clearml 2024-10-28 18:11:00 +0200
  • f7d4e577ac
         Update requests range to allow <2.33.0 Emir Isman 2024-10-21 15:55:12 +0200
  • bf8d9c96e9
         Handle OSError when checking for is_file (#215) Matteo Destro 2024-10-13 09:08:03 +0200
  • 0044514f72
         Handle OSError when checking for is_file Matteo Destro 2024-10-11 10:40:59 +0200
  • a88487ff25 Add support for pip legacy resolver for versions specified in the agent.package_manager.pip_legacy_resolver configuration option Add skip existing packages allegroai 2024-09-22 22:35:56 +0300
  • 785e22dc87
         Version bump to v1.9.1 v1.9.1 Jake Henning 2024-09-02 01:04:49 +0300
  • 6a2b778d53
         Add default pip version support for Python 3.12 Jake Henning 2024-09-02 01:03:52 +0300
  • b2c3702830 Version bump to v1.9.0 v1.9.0 allegroai 2024-08-28 23:18:26 +0300
  • 6302d43990 Add support for skipping container apt installs using CLEARML_AGENT_SKIP_CONTAINER_APT env var in k8s Add runtime callback support for setting runtime properties per task in k8s Fix remove task from pending queue and set to failed when kubectl apply fails allegroai 2024-08-27 23:01:27 +0300
  • 760bbca74e Fix failed Task in services mode logged "User aborted" instead of failed, add Task reason string allegroai 2024-08-27 22:56:37 +0300
  • e63fd31420 Fix string format allegroai 2024-08-27 22:55:49 +0300
  • 2ff9985db7 Add user ID to the vault loading print allegroai 2024-08-27 22:55:32 +0300
  • b8c762401b Fix use same state transition if supported by the server (instead of stopping the task before re-enqueue) allegroai 2024-08-27 22:54:45 +0300
  • 99e1e54f94 Add support for tasks containing only bash script or python module command allegroai 2024-08-27 22:53:14 +0300
  • a4d3b5bad6 Fix only set Task started status on node rank 0 allegroai 2024-08-27 22:52:31 +0300
  • b21665ed6e Fix do not cache venv cache if venv/python skip env var was set allegroai 2024-08-27 22:52:01 +0300
  • f877aa96e2
         Update Docker base image to Ubuntu 22.04 and Kubectl to 1.29.3 (#201) Surya Teja 2024-07-29 16:41:50 +0100
  • f99344d194
         Add queue priority info to CLI help (#211) pollfly 2024-07-29 18:40:38 +0300
  • d9f2a1999a Fix Only send pip freeze update on RANK 0, only update task status on exit on RANK 0 allegroai 2024-07-29 17:40:24 +0300
  • f866d43693 Add --order-fairness info revital 2024-07-28 09:22:44 +0300
  • 79d0abe707
         Add NO_DOCKER flag to clearml-agent-services entrypoint (#206) Valentin Schabschneider 2024-07-26 18:09:54 +0200
  • 6213ef4c02 Add /bin/bash -c "command" support. Task binary should be set to /bin/bash and entry_point should be set to -c command allegroai 2024-07-24 18:00:13 +0300
  • aef6aa9fc8 Fix a race condition where in rare conditions popping a Task from a queue that was aborted did not set it to started before the watchdog killed it. Does not happen in k8s/slurm allegroai 2024-07-24 17:59:46 +0300
  • 0bb267115b Add venvs_cache.path mount override for non-root containers (use: agent.docker_internal_mounts.venvs_cache) allegroai 2024-07-24 17:59:18 +0300
  • f89a92556f Fix check logger is not None allegroai 2024-07-24 17:55:02 +0300
  • 8ba4d75e80 Add CLEARML_TASK_ID and auth token to pod env vars in original entrypoint flow allegroai 2024-07-24 17:47:48 +0300
  • edc333ba5f Add K8S_GLUE_POD_USE_IMAGE_ENTRYPOINT to allow running images without overriding the entrypoint (useful for agents using prebuilt images in k8s) allegroai 2024-07-24 17:46:27 +0300
  • 2f0553b873 Fix CLEARML_MULTI_NODE_SINGLE_TASK should be read once not every reported line allegroai 2024-07-24 17:45:02 +0300
  • b2a4bf08ac Fix pass --docker only (i.e. no default container image) for --dynamic-gpus feature allegroai 2024-07-24 17:44:35 +0300
  • f18c6b809f Fix slurm multi-node rank detection allegroai 2024-07-24 17:44:05 +0300
  • cd5b4d2186 Add "-m module args" in script entry now supports standalone script, standalone script is converted to "untitled.py" by default or if specified in working_dir such as <dir>:<target_file> for example ".:standalone.py" allegroai 2024-07-24 17:43:21 +0300
  • 5f1bab6711 Add default docker match_rules for enterprise users, NOTICE: matching_rules are ignored if --docker container is passed in command line allegroai 2024-07-24 17:42:55 +0300
  • ab9b9db0c9 Add CLEARML_MULTI_NODE_SINGLE_TASK (values -1, 0, 1, 2) for easier multi-node singe Task workloads allegroai 2024-07-24 17:42:25 +0300
  • 93df021108 Add support for .ipynb script entry files (install nbconvert in runtime, copnvert to python and execute the python script), including CLEARML_AGENT_FORCE_TASK_INIT patching of ipynb files (post python conversion) allegroai 2024-07-24 17:41:59 +0300
  • 700ae85de0 Fix file mode should be optional in configuration files section allegroai 2024-07-24 17:41:06 +0300
  • f367c5a571 Fix git fetch did not update new tags #209 allegroai 2024-07-24 17:39:53 +0300
  • ebc5944b44 Fix setting tasks that someone just marked as aborted to started - only force Task to started after dequeuing it otherwise lease it as is allegroai 2024-07-24 17:39:26 +0300
  • 8f41002845 Add task.script.binary /bin/bash support Fix -m module $env to support parsing the $env before launching allegroai 2024-07-24 17:37:26 +0300
  • 7e8670d57f Find the correct python version when using a pre-installed python environment allegroai 2024-07-21 14:10:38 +0300
  • 77de343863 Use "venv" module if virtualenv is not supported allegroai 2024-07-19 13:18:07 +0300