ClearML/clearml-docs
1
0
Fork 0
mirror of https://github.com/clearml/clearml-docs synced 2025-04-05 05:40:54 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update ClearML Session page (#795)

Browse Source
This commit is contained in:
pollfly 2024-03-11 19:27:47 +02:00 committed by GitHub
parent 12482948d3
commit dce8b12932
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 63 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

89
docs/apps/clearml_session.md
View File

@ -143,6 +143,18 @@ To access a git repository remotely, add a `--git-credentials` flag and set it t
file is sent to the interactive session. This is helpful if working on private git repositories, and it allows for seamless
cloning and tracking of git references, including untracked changes.
#### Uploading Local Files to Remote Session
You can upload local files and directories from your local machine into the remote session by specifying their path with
`--upload-files <file_path>`. The entire content of the directory or file will be copied into your remote
`clearml-session` container under the `~/session-files/` directory.
```commandline
clearml-session --upload-files /mnt/data/stuff
```
You can upload your files in conjunction with the `--store-workspace` option to easily move workloads between local
development machines and remote machines with persistent workspace synchronization. See [Storing and Synchronizing Workspace](#storing-and-synchronizing-workspace).
#### Starting a Debugging Session
You can debug previously executed experiments registered in the ClearML system on a remote interactive session.
Input into `clearml-session` the ID of a Task to debug, then `clearml-session` clones the experiment's git repository and
@ -158,6 +170,31 @@ The Task must be connected to a git repository, since currently single script de
1. Click on the JupyterLab / VS Code link, or connect directly to the SSH session.
1. In JupyterLab / VS Code, access the experiment's repository in the `environment/task_repository` folder.
#### Storing and Synchronizing Workspace
You can store and sync your interactive session workspace with the `--store-workspace` option. `clearml-session`
will automatically create a snapshot of your entire workspace when shutting it down, and later restore in a
new session on any remote machine.
Specify the remote workspace root-folder by adding `--store-workspace <workspace_root_folder>` to the command line. In
the remote session container, put all your code and data under the `<workspace_root_folder>` directory. When your
session is shut down, the workspace folder will be automatically packaged and stored on the ClearML file server.
```commandline
clearml-session --store-workspace ~/workspace --docker python:3.10-bullseye
```
In your next `clearml-session` execution, specify `--store-workspace <workspace_root_folder>` again and `clearml-session`
will grab the previous workspace snapshot and restore it into the new remote container in `<workspace_root_folder>`.
```commandline
clearml-session --store-workspace ~/workspace --docker python:3.10-bullseye
```
To continue a specific session and restore its workspace, specify the session ID with `--continue-session <session_id>`:
```commandline
clearml-session --continue-session <session_id> --store-workspace ~/workspace --docker python:3.10-bullseye
```
### Command Line Options
@ -166,42 +203,42 @@ The Task must be connected to a git repository, since currently single script de
| Command line options | Description | Default value |
|-----|---|---|
| `--attach`| Attach to running interactive session | Previous session|
| `--project`| Set the project name to the interactive session task| `DevOps` |
| `--shutdowm`, `-S`| Shut down an active session | Previous session|
| `--base-task-id` | Pass the ID of a task that will become the base task, so the session will use its configurations | `none` or the previously entered base task |
| `--config-file` | Specify a path to another configuration file for `clearml-session` to store its previous state | `.clearml_session.json` or previously entered configuration file |
| `--continue-session` | Pass the session of a previous session to continue, restoring your workspace (see `--store-workspace`) | `none` |
| `--debugging-session` | Pass existing Task ID, create a copy of the experiment on a remote machine, and launch Jupyter/SSH for interactive access. Example `--debugging-session <task_id>`| `none`|
| `--disable-session-cleanup` | If `True`, previous interactive sessions are not deleted | `false`|
| `--requirements`| Specify `requirements.txt` file to install when setting the interactive session. | `none` or previously used requirements (can be overridden by calling `--packages`)|
| `--packages`| Additional packages to add. Supports version numbers. Example: `--packages torch==1.7 tqdm` | Previously added packages.|
| `--upload-files`| Specify local files/folders to upload to the remote session|`None`|
| `--git-credentials` | If `True`, local `.git-credentials` file is sent to the interactive session.| `false`|
| `--docker`| Select the docker image to use in the interactive session on |`nvidia/cuda:11.6.2-runtime-ubuntu20.04` or previously used docker image|
| `--docker-args` | Add additional arguments for the docker image to use in the interactive session | `none` or the previously used docker-args |
| `--debugging-session` | Pass existing Task ID, create a copy of the experiment on a remote machine, and launch Jupyter/SSH for interactive access. Example `--debugging-session <task_id>`| `none`|
| `--queue`| Select the queue to launch the interactive session on | Previously used queue|
| `--shell` | Open the SSH session directly. Notice, quiting the SSH session will not shut down the remote session|`None`|
| `--jupyter-lab` | Install a JupyterLab on interactive session | `true` |
| `--vscode-server` | Install VSCode on interactive session | `true` |
| `--vscode-version` | Set VSCode server (code-server) version, as well as VSCode python extension version <vscode:python-ext> (example: "3.7.4:2020.10.332292344")| `4.14.1:2023.12.0`|
| `--vscode-extensions` |Install additional VSCode extensions and VSCode python extensions (example: `ms-python.python,ms-python.black-formatter,ms-python.pylint,ms-python.flake8`)|`None`|
| `--public-ip` | If `true`, register the public IP of the remote machine (if you are running the session on a public cloud) | `false` - Session runs on the machine whose agent is executing the session|
| `--remote-ssh-port`| Set the remote SSH server port, running on the agent's machine | 10022|
| `--force_dropbear`| Force using `dropbear` instead of SSHd |`None`|
| `--git-credentials` | If `True`, local `.git-credentials` file is sent to the interactive session.| `false`|
| `--init-script` | Specify a BASH init script file to be executed when the interactive session is being set up | `none` or previously entered BASH script |
| `--user-folder` | Specify the path for the session's remote base folder for the session | Home folder(`~/`) or previously entered user folder path |
| `--config-file` | Specify a path to another configuration file for `clearml-session` to store its previous state | `.clearml_session.json` or previously entered configuration file |
| `--remote-gateway` | Specify a gateway IP to pass to the interactive session, if an external address needs to be accessed | `none`|
| `--base-task-id` | Pass the ID of a task that will become the base task, so the session will use its configurations | `none` or the previously entered base task |
| `--jupyter-lab` | Install a JupyterLab on interactive session | `true` |
| `--keepalive` | If set, enables transparent proxy that keep sockets alive to maintain the connection to the remote resource | `false` - do not use transparent socket for mitigating connection drops |
| `--packages`| Additional packages to add. Supports version numbers. Example: `--packages torch==1.7 tqdm` | Previously added packages.|
| `--password`| Set your own SSH password for the interactive session | A randomly generated password or a previously used one |
| `--project`| Set the project name to the interactive session task| `DevOps` |
| `--public-ip` | If `true`, register the public IP of the remote machine (if you are running the session on a public cloud) | `false` - Session runs on the machine whose agent is executing the session|
| `--queue`| Select the queue to launch the interactive session on | Previously used queue|
| `--queue-excluded-tag` | The queue option list will exclude queues with specified tags. See the `tags` parameter in the [queues.create](../references/api/queues.md#post-queuescreate) API call | `none` |
| `--queue-include-tag` | The queue option list will include only queues with specified tags. See the `tags` parameter in the [queues.create](../references/api/queues.md#post-queuescreate) API call | `none` |
| `--skip-docker-network` | Don't pass the `--network host` flag to the Docker that is launching the remote session. See [Networking using the host network](https://docs.docker.com/network/network-tutorial-host/) | `false`|
| `--username`| Set your own SSH username for the interactive session | `root` or a previously used username |
| `--password`| Set your own SSH password for the interactive session | A randomly generated password or a previously used one |
| `--force_dropbear`| Force using `dropbear` instead of SSHd |`None`|
| `--store-workspace` | Upload/Restore remote workspace folder and extract it into next the session. Use with `--continue-session` to continue your previous work from your exact container state | `none` |
| `--continue-session` | Pass the session of a previous session to continue, restoring your workspace (see `--store-workspace`) | `none` |
| `--remote-gateway` | Specify a gateway IP to pass to the interactive session, if an external address needs to be accessed | `none`|
| `--remote-ssh-port`| Set the remote SSH server port, running on the agent's machine | 10022|
| `--requirements`| Specify `requirements.txt` file to install when setting the interactive session. | `none` or previously used requirements (can be overridden by calling `--packages`)|
| `--session-name` | Set the name of the interactive session Task| `none` |
| `--session-tags` | Add tags to the interactive session for increased visibility | `none` |
| `--version`| Display the clearml-session utility version| N/A|
| `--shell` | Open the SSH session directly. Notice, quiting the SSH session will not shut down the remote session|`None`|
| `--shutdowm`, `-S`| Shut down an active session | Previous session|
| `--skip-docker-network` | Don't pass the `--network host` flag to the Docker that is launching the remote session. See [Networking using the host network](https://docs.docker.com/network/network-tutorial-host/) | `false`|
| `--store-workspace` | Upload/Restore remote workspace folder and extract it into next the session. Use with `--continue-session` to continue your previous work from your exact container state | `none` |
| `--upload-files`| Specify local files/folders to upload to the remote session|`None`|
| `--user-folder` | Specify the path for the session's remote base folder for the session | Home folder(`~/`) or previously entered user folder path |
| `--username`| Set your own SSH username for the interactive session | `root` or a previously used username |
| `--verbose` | Increase verbosity of logging | `none` |
| `--version`| Display the clearml-session utility version| N/A|
| `--vscode-extensions` |Install additional VSCode extensions and VSCode python extensions (example: `ms-python.python,ms-python.black-formatter,ms-python.pylint,ms-python.flake8`)|`None`|
| `--vscode-server` | Install VSCode on interactive session | `true` |
| `--vscode-version` | Set VSCode server (code-server) version, as well as VSCode python extension version <vscode:python-ext> (example: "3.7.4:2020.10.332292344")| `4.14.1:2023.12.0`|
| `--yes`, `-y`| Automatic yes to prompts; assume "yes" as answer to all prompts and run non-interactively |N/A|
</div>