Start free, upgrade when you're ready. No credit card required.
Free
No credit card required
Get started freeFeatures
Integrations
Limits
Build
+ usage above allocation
Start buildingFeatures
Integrations
Limits
Launch
+ usage above allocation
Choose LaunchFeatures
Integrations
Limits
One task created through the API is one task. The Celery integration creates one task per execution, so each time a tracked Celery task runs it counts once. Child tasks count like any other task.
Updating a task after it is created never counts, however often you do it.
Not the first three. Retries are nested under the first attempt, and up to three retries of the same task are free. From the fourth retry on, each one counts as a task, so a task that retries forever still has a cost.
A task that fails and retries once or twice costs one task. Only tasks that retry more than three times are billed for the extra retries.
No. Calls to task.update(), status changes and heartbeats are free. Only creating a
task counts.
Each scheduled run of a monitor is tracked as one task, so it counts once. A job that runs every hour uses about 730 tasks a month.
Creating a new task returns an HTTP 429 error until the next month. Tasks that already exist keep updating, so work in progress is not cut off. The count resets at the start of each month (UTC).
We will warn you by email before you reach the limit.
Tasks that are never tracked are never created, so they do not count. With the Celery integration:
includes: track only tasks whose names match these patterns.excludes: skip tasks whose names match, such as heartbeats and cleanup jobs.taskbadger_track=False: skip a single execution.CelerySystemIntegration(
includes=[r"myapp\..*"],
excludes=[r"myapp\.cleanup\..*"],
)
noisy_job.apply_async(args, taskbadger_track=False)
See the Celery docs for details.
All plans include the full REST API, public task pages, and email notifications.
Need something custom? Get in touch