The **Schedule History** page in the **Alerts \& Schedules** section of the **Admin** menu shows the recent history of the Scheduler's activity.

> [!NOTE]
> **Note:** If you have a permission that provides access to only select pages in the Admin panel, such as [`manage_schedules`](https://docs.cloud.google.com/looker/docs/2622/admin-panel-users-roles.md.txt#manage_schedules), [`manage_themes`](https://docs.cloud.google.com/looker/docs/2622/admin-panel-users-roles.md.txt#manage_themes), or [`see_admin`](https://docs.cloud.google.com/looker/docs/2622/admin-panel-users-roles.md.txt#see_admin), but you don't have the [Admin role](https://docs.cloud.google.com/looker/docs/2622/admin-panel-users-roles.md.txt#default_roles), the page or pages that are described here may not be visible to you in the Admin panel.

You can view scheduler plans on the [Admin settings - Schedules](https://docs.cloud.google.com/looker/docs/2622/admin-panel-alerts-and-schedules-schedule.md.txt) page. The schedule plan specifies when content should be sent to specific addresses.

A new schedule job is created each time that content is sent. For example, if a schedule specifies sending specific content to certain addresses every Monday, then Looker will make a new scheduler job each Monday to send that content to the specified addresses.

## Scheduling status

At the top of the Scheduler History page, you can filter the scheduled jobs by status:

- **All**: All recent scheduler jobs
- **Running**: Only scheduler jobs that are running
- **Successful**: Only scheduler jobs that completed successfully
- **Failed**: Only scheduler jobs that failed to complete successfully

A status of **Successful** indicates that the delivery itself was
successful, but it doesn't guarantee the presence of data rows. If a time-based
schedule fetches results from the query cache and a previous query cached an
empty result, the schedule delivers an empty file and still shows as
**Successful** . To ensure that your schedules always pull fresh data from the
database, consider using a
[datagroup trigger](https://docs.cloud.google.com/looker/docs/2622/delivering-looks-explores.md.txt#specifying_the_datagroup_trigger)
rather than a standard time-based schedule.

## Information included

| Column | Definition |
|---|---|
| Time | When the scheduled job began |
| Test | Shows a checkmark if the job was a test |
| Status | Whether the job is current running, complete, or failed |
| Stage | The current stage of a running job |
| User | The user that created the schedule |
| Source | The content that was scheduled, either a look or a dashboard |
| Name | The name of the look or dashboard, which also serves as link to the content |
| Type | The type of data delivery, either Amazon S3, Email, Segment, SFTP, or Webhook |
| Format | The format of the content in the scheduled job, such as CSV |
| Message | For a failed attempt, the error message |
| Runtime | How long the job ran or has been running |

## Using error messages

Use error messages to debug scheduled jobs. For example, if the owner of a schedule is missing a permission or the owner's user ID is not enabled when a schedule was run on their behalf, the job fails.

Another example is scheduled jobs that fail in the scheduled job queue without the job having been closed. The Scheduler monitors the scheduled job queue for jobs that have been enqueued for an hour or more while more recent jobs have been successfully dequeuing. If this occurs, the enqueued scheduled job is closed and reported as **Failed** on the **Schedule History** page, returning one of the following messages:

- Job is no longer processing `@<state>` (where `<state>` is either **execute** or **deliver**)
- Job was expected to dequeue by `@<time>`

If you see one of these messages, your scheduled job will execute the next time it is scheduled to run.


## Getting additional details

Click the **Details** button to the right of any schedule to view additional information, including recipient addresses, whether the row limit was reached, the start and end times, and other details about the scheduled job. For scheduled content that uses a datagroup, the Details window will also show the model and the name of the datagroup that is being used as the schedule trigger.

## Programmatically retrying failed scheduled plans

Looker doesn't automatically retry scheduled plans that fail as a result of intermittent issues (such as temporary database connection timeouts). However, you can use the Looker API to build an automated retry mechanism by querying System Activity logs with [`run_inline_query`](https://docs.cloud.google.com/looker/docs/reference/looker-api/latest/methods/Query/run_inline_query.md.txt) and re-triggering failed deliveries with [`run_scheduled_plan_once`](https://docs.cloud.google.com/looker/docs/reference/looker-api/latest/methods/ScheduledPlan/run_scheduled_plan_once). For detailed steps, see [How do I programmatically retry failed scheduled plans?](https://docs.cloud.google.com/looker/docs/2622/usage-reports-with-system-activity-explores#how_do_i_programmatically_retry_failed_scheduled_plans?).