Stop rollup job APIedit

This functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.

This API stops an existing, started rollup job. If the job does not exist an exception will be thrown. Stopping an already stopped job has no action.


POST _rollup/job/<job_id>/_stop

Path Parametersedit

job_id (required)
(string) Identifier for the job

Query Parametersedit

wait_for_completion (optional)
(boolean) if set to true, causes the API to block until the indexer state completely stops. If set to false, the API returns immediately and the indexer will be stopped asynchronously in the background. Defaults to false.
timeout (optional)
(TimeValue) if wait_for_completion=true, the API will block for (at maximum) the specified duration while waiting for the job to stop. If more than timeout time has passed, the API will throw a timeout exception. Note: even if a timeout exception is thrown, the stop request is still processing and will eventually move the job to STOPPED. The timeout simply means the API call itself timed out while waiting for the status change. Defaults to 30s

Request Bodyedit

There is no request body for the Stop Job API.


You must have manage or manage_rollup cluster privileges to use this API. For more information, see Security privileges.


If we have an already-started rollup job named sensor, it can be stopped with:

POST _rollup/job/sensor/_stop

Which will return the response:

  "stopped": true

If however we try to stop a job which doesn’t exist:

POST _rollup/job/does_not_exist/_stop

A 404 resource_not_found exception will be thrown:

    "error" : {
        "root_cause" : [
                "type" : "resource_not_found_exception",
                "reason" : "Task for Rollup Job [does_not_exist] not found",
                "stack_trace": ...
        "type" : "resource_not_found_exception",
        "reason" : "Task for Rollup Job [does_not_exist] not found",
        "stack_trace": ...
    "status": 404

Waiting for the job to stopedit

Since only a stopped job can be deleted, it can be useful to block the StopJob API until the indexer has fully stopped. This is accomplished with the wait_for_completion query parameter, and optionally a timeout:

POST _rollup/job/sensor/_stop?wait_for_completion=true&timeout=10s

The parameter will block the API call from returning until either the job has moved to STOPPED, or the specified time has elapsed. If the specified time elapses without the job moving to STOPPED, a timeout exception will be thrown.

If wait_for_completion=true is specified without a timeout, a default timeout of 30 seconds is used.