mirror of
https://github.com/clearml/clearml-docs
synced 2025-03-03 02:32:49 +00:00
Add ClearML Agent config options (#669)
This commit is contained in:
parent
d2dbd30bb4
commit
a7954101b4
@ -27,10 +27,14 @@ but can be overridden by command-line arguments.
|
||||
|**CLEARML_AGENT_GIT_HOST** | Sets Git host (only sending login to this host) |
|
||||
|**CLEARML_AGENT_EXEC_USER** | User for Agent executing tasks (root by default) |
|
||||
|**CLEARML_AGENT_EXTRA_DOCKER_ARGS** | Overrides extra docker args configuration |
|
||||
|**CLEARML_AGENT_EXTRA_DOCKER_LABELS** | List of labels to add to docker container. See [Docker documentation](https://docs.docker.com/config/labels-custom-metadata/). |
|
||||
|**CLEARML_EXTRA_PIP_INSTALL_FLAGS**| List of additional flags to use when the agent install packages. For example: `["--use-deprecated=legacy-resolver", ]`|
|
||||
|**CLEARML_AGENT_EXTRA_PYTHON_PATH** | Sets extra python path |
|
||||
|**CLEARML_AGENT_INITIAL_CONNECT_RETRY_OVERRIDE** | Overrides initial server connection behavior (true by default), allows explicit number to specify number of connect retries) |
|
||||
|**CLEARML_AGENT_NO_UPDATE** | Boolean. Set to `true` to skip agent update in the k8s pod container before the agent executes the task |
|
||||
|**CLEARML_AGENT_K8S_HOST_MOUNT / CLEARML_AGENT_DOCKER_HOST_MOUNT** | Specifies Agent's mount point for Docker / K8s |
|
||||
|**CLEARML_K8S_GLUE_START_AGENT_SCRIPT_PATH** | Provide an alternate path to place the agent startup script generated inside a k8s task pod (instead of the default `~/~/__start_agent__.sh`) |
|
||||
|**CLEARML_AGENT_PACKAGE_PYTORCH_RESOLVE**|Sets the PyTorch resolving mode. The options are: <ul><li>`none` - No resolving. Install PyTorch like any other package</li><li>`pip` (default) - Sets extra index based on cuda and lets pip resolve</li><li>`direct` - Resolve a direct link to the PyTorch wheel by parsing the pytorch.org pip repository, and matching the automatically detected cuda version with the required PyTorch wheel. If the exact cuda version is not found for the required PyTorch wheel, it will try a lower cuda version until a match is found</li></ul> |
|
||||
|**CLEARML_AGENT_DEBUG_INFO** | Provide additional debug information for a specific context (currently only the `docker` value is supported) |
|
||||
|**CLEARML_AGENT_CHILD_AGENTS_COUNT_CMD** | Provide an alternate bash command to list child agents while working in services mode |
|
||||
|**CLEARML_AGENT_SKIP_PIP_VENV_INSTALL** | Skips Python virtual env installation on execute and provides a custom venv binary |
|
||||
|
@ -110,6 +110,16 @@ in either case).
|
||||
|
||||
---
|
||||
|
||||
**`agent.docker_allow_host_environ`** (*bool*)
|
||||
|
||||
* Set to `true` to allow passing host environments into docker container with Task's docker container arguments. For example: `"-e HOST_NAME=$HOST_NAME"`.
|
||||
|
||||
:::caution
|
||||
Use with care! This might introduce security risks by allowing access to keys/secret on the host machine.
|
||||
:::
|
||||
|
||||
---
|
||||
|
||||
**`agent.docker_apt_cache`** (*string*)
|
||||
|
||||
* The apt (Linux package tool) cache folder for mapping Ubuntu package caching into Docker.
|
||||
@ -460,6 +470,12 @@ ___
|
||||
|
||||
* A list of URLs for additional artifact repositories when installing Python packages.
|
||||
|
||||
---
|
||||
|
||||
**`agent.package_manager.extra_pip_install_flags`** (*[string]*)
|
||||
|
||||
* A list of additional flags to use when the agent install packages. For example: `["--use-deprecated=legacy-resolver", ]`
|
||||
|
||||
---
|
||||
|
||||
**`agent.package_manager.force_upgrade`** (*bool*)
|
||||
@ -518,6 +534,17 @@ ___
|
||||
|
||||
---
|
||||
|
||||
**`agent.package_manager.pytorch_resolve`** (*str*)
|
||||
|
||||
* Set the PyTorch resolving mode. The options are:
|
||||
* `pip` (default) - Sets extra index based on cuda and lets pip resolve
|
||||
* `none` - No resolving. Install PyTorch like any other package
|
||||
* `direct` - Resolve a direct link to the PyTorch wheel by parsing the pytorch.org pip repository and matching the
|
||||
automatically detected cuda version with the required PyTorch wheel. If the exact cuda version is not found for the
|
||||
required PyTorch wheel, it will try a lower cuda version until a match is found
|
||||
|
||||
---
|
||||
|
||||
<a id="system_site_packages"/>
|
||||
|
||||
**`agent.package_manager.system_site_packages`** (*bool*)
|
||||
|
Loading…
Reference in New Issue
Block a user