# CancelJob
<a name="cancel-job-api"></a>

The `CancelJob` operation stops a running or pending job. Only jobs in **Pending** or **Running** states can be cancelled. Cancellation is asynchronous, so you should monitor the job status to confirm completion. Cancelled jobs may have partial results in the output location depending on when cancellation occurs during processing.

For more information, see [CancelJob](https://docs.aws.amazon.com/location/latest/APIReference/API_geojobs_CancelJob.html) in the *Amazon Location Service API Reference*.

For example requests, responses, and CLI commands for this API, see [How to cancel a job](https://docs.aws.amazon.com/location/latest/developerguide/canceling-job.html).

## Use cases
<a name="cancel-job-use-cases"></a>
+ **Stop unnecessary processing:** Cancel jobs that were submitted with incorrect configuration, are no longer needed due to changed business requirements, or were started by mistake. This can prevent wasted resources and unnecessary charges.
+ **Resource and cost optimization:** Free up processing capacity by cancelling lower-priority jobs when higher-priority tasks need to be processed.

## Understand the request
<a name="cancel-job-request-parameters"></a>

The `CancelJob` request requires only a single parameter to identify which job to cancel. This simple request structure makes it easy to stop processing for jobs that are no longer needed.

The request includes the following parameter:

**Job identification**  
Required parameter that specifies which job to cancel.  
+ `JobId`: The unique identifier of the job to cancel. Must be 1-100 characters matching pattern `[-._\w]+`. This is the same identifier returned by the `StartJob` operation when the job was created. Only jobs in `Pending` or `Running` status can be cancelled.

## Understand the response
<a name="cancel-job-response-details"></a>

The `CancelJob` response confirms the cancellation request and provides the job's updated status. Because cancellation is asynchronous, you should use the `GetJob` operation to monitor the job until it reaches the `Cancelled` status.

The response includes the following fields:

**Job identification**  
Unique identifiers for the job being cancelled.  
+ `JobId`: The unique job identifier.
+ `JobArn`: Amazon Resource Name (ARN) of the job being cancelled.

**Cancellation status**  
Current job status after the cancellation request.  
+ `Status`: Job status immediately after the cancellation request. Typically `Cancelling` to indicate cancellation is in progress. The status transitions to `Cancelled` when cancellation completes. Use `GetJob` to monitor this transition.

**Note**  
Cancelled jobs may have partial results in the output location depending on when cancellation occurs during processing. Check the output location to determine if any partial data was written before cancellation completed. You are billed for the number of records that were processed and written to your output bucket before the job was cancelled.