> ## Documentation Index
> Fetch the complete documentation index at: https://wb-21fd5541-style-guide-models-runs-20260604-113608.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

> Resume paused, stopped, or crashed W&B runs using the resume parameter options in wandb.init().

# Resume a run

This page shows you how to configure W\&B to resume a run that was paused, stopped, or crashed, so you can continue logging from where you left off instead of starting over. To specify how W\&B responds if a run stops or crashes, set the `resume` parameter in `wandb.init()`. When you initialize a run, W\&B checks whether the run ID already exists and applies the behavior defined by the `resume` value.

The following table outlines the behavior of W\&B based on the argument passed to the `resume` parameter and whether the run ID exists or not.

| Argument  | Description                                                                                                                   | Run ID exists                                                  | Run ID doesn't exist                              | Use case                                         |
| --------- | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- | ------------------------------------------------- | ------------------------------------------------ |
| `"must"`  | W\&B must resume run specified by the run ID.                                                                                 | W\&B resumes run with the same run ID. Resumes from last step. | W\&B raises an error.                             | Resume a run that must use the same run ID.      |
| `"allow"` | Let W\&B resume run if run ID exists.                                                                                         | W\&B resumes run with the same run ID. Resumes from last step. | W\&B initializes a new run with specified run ID. | Resume a run without overriding an existing run. |
| `"never"` | Never let W\&B resume a run specified by the run ID.                                                                          | Raise an error if a run with the specified ID already exists.  | W\&B initializes a new run with specified run ID. |                                                  |
| `"auto"`  | Let W\&B automatically attempt to resume run if run ID exists. Restart the run from the same directory as the failed process. | W\&B resumes run with the same run ID.                         | W\&B initializes a new run with specified run ID. | Enable runs to automatically resume.             |

<Note>
  When to use `auto` compared to `allow`:

  W\&B recommends that you use `resume="allow"` and specify the specific run ID you want to resume.

  The `resume="auto"` option doesn't require you to specify a run ID, but it can lead to unexpected behavior if you have multiple runs that fail in the same directory or if the file directory structure changes. You must also ensure that you restart the run from the same directory as the failed process when you use `resume="auto"`.
</Note>

For all the following examples, replace values enclosed within `[]` with your own.

<Tip>[View a live demo of a resumed run](https://wandb.ai/wandb/resume-run/workspace?nw=nwuserjuliarose).</Tip>

## Resume a run that must use the same run ID

If a run is stopped, crashes, or fails, you can resume it using the same run ID. To do so, initialize a run and specify the following:

* Set the `resume` parameter to `"must"` (`resume="must"`).
* Provide the run ID of the run that stopped or crashed.

The following code snippet shows how to accomplish this with the W\&B Python SDK:

```python theme={null}
with wandb.init(entity="[ENTITY]", project="[PROJECT]", id="[RUN-ID]", resume="must") as run:
        # Your training code here
```

<Warning>
  Unexpected results occur if multiple processes use the same `id` concurrently.

  For more information about how to manage multiple processes, see [Log distributed training experiments](/models/track/log/distributed-training/).
</Warning>

## Resume a run without overriding the existing run

Resume a run that stopped or crashed without overriding the existing run. This is helpful if your process doesn't exit successfully, because W\&B falls back to starting a new run when the specified run ID doesn't already exist. The next time you start W\&B, W\&B starts logging from the last step.

Set the `resume` parameter to `"allow"` (`resume="allow"`) when you initialize a run with W\&B. Provide the run ID of the run that stopped or crashed. The following code snippet shows how to accomplish this with the W\&B Python SDK:

```python theme={null}
import wandb

with wandb.init(entity="[ENTITY]", project="[PROJECT]", id="[RUN-ID]", resume="allow") as run:
        # Your training code here
```

## Enable runs to automatically resume

Automatic resuming lets W\&B pick up a failed run without requiring you to specify a run ID, which is useful for long-running training jobs that might restart unattended. The following code snippet shows how to enable runs to automatically resume with the Python SDK or with environment variables.

<Tabs>
  <Tab title="W&B Python SDK">
    Pass `auto` as an argument to the `resume` parameter when you initialize a run. Ensure that you restart the run from the same directory as the failed process.

    The following code snippet enables automatic resuming. Replace values enclosed within `[]` with your own:

    ```python theme={null}
    with wandb.init(entity="[ENTITY]", project="[PROJECT]", id="[RUN-ID]", resume="auto") as run:
            # Your training code here
    ```
  </Tab>

  <Tab title="Shell script">
    The following example shows how to specify the W\&B `WANDB_RUN_ID` variable in a bash script:

    ```bash title="run_experiment.sh" theme={null}
    RUN_ID="$1"

    WANDB_RESUME=auto WANDB_RUN_ID="$RUN_ID" python eval.py
    ```

    Within your terminal, you can run the shell script along with the W\&B run ID. The following code snippet passes the run ID `akj172`:

    ```bash theme={null}
    sh run_experiment.sh akj172
    ```
  </Tab>
</Tabs>

<Warning>
  Automatic resuming only works if the process is restarted on top of the same filesystem as the failed process.
</Warning>

To illustrate why the same filesystem matters, suppose you execute a Python script called `train.py` in a directory called `Users/Employee/Desktop/ImageClassify/training/`. Within `train.py`, the script creates a run that enables automatic resuming. Next, suppose the training script stops. To resume this run, you must restart your `train.py` script within `Users/Employee/Desktop/ImageClassify/training/`.

<Note>
  If you can't share a filesystem, specify the `WANDB_RUN_ID` environment variable or pass the run ID with the W\&B Python SDK. See [Create a custom run ID](/models/runs/run-identifiers#create-a-custom-run-id) for more information about run IDs.
</Note>

## Resume preemptible Sweeps runs

Handle preemption signals so W\&B can automatically requeue interrupted [sweep](/models/sweeps/) runs for another agent. This pattern is useful when the sweep agent runs on preemptible compute, such as a SLURM preemptible queue, an Amazon EC2 Spot Instance, or a Google Cloud preemptible VM.

The instructions below apply when you start sweep agents with the [`wandb agent`](/models/ref/cli/wandb-agent) CLI. The CLI starts your training program as a **subprocess**. The instructions do not fully apply when you use only the Python API [`wandb.agent()`](/models/ref/python/functions/agent). The Python API runs the training function in a thread, so OS signal delivery and forwarding differ from the CLI agent behavior.

### Handle a preemption signal

Register a handler for the signal that your scheduler or platform uses to indicate preemption, such as `SIGUSR1` or `SIGTERM`. In the handler:

1. Call [`mark_preempting()`](/models/ref/python/experiments/run#mark_preempting) when a run is active.
2. Perform any required cleanup, such as saving a checkpoint.
3. Exit with a nonzero status code. A common convention for signal termination is `128 + signum`.

Do not call `mark_preempting()` unconditionally immediately after `wandb.init()`. Doing so can mark every failure, including code bugs, as preemption and requeue the run repeatedly.

For runnable examples, `--forward-signals` on the CLI agent, and a full reference table for different uses of `mark_preempting()`, see [Signal handling and sweep runs](/models/sweeps/signal-handling-sweep-runs).

When you follow that pattern, W\&B records run state roughly as follows:

| Scenario                                                                                                                                             | Run state                                               |
| ---------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------- |
| Run completes normally with exit code 0                                                                                                              | FINISHED                                                |
| Run fails with a non-zero exit code                                                                                                                  | FAILED                                                  |
| Run receives an unhandled signal (for example `SIGKILL`)                                                                                             | CRASHED after about 5 minutes                           |
| Run receives a handled preemption signal (for example `SIGTERM` or `SIGUSR1`), the handler calls `mark_preempting()`, and the process exits non-zero | PREEMPTED. The run is queued for the next agent request |

<Info>
  When a sweep agent fetches a preempted run, the training process must call `wandb.init()` within 60 minutes. If initialization does not occur, such as when the process fails after fetching the run but before calling `wandb.init()`, W\&B does not make the run available to another agent until the 60-minute lease expires.
</Info>

Sweep agents process requeued runs before requesting new hyperparameter combinations from the sweep search algorithm. After the queue is empty, the sweep resumes normal scheduling.