Implement Sentry’s Celery integration in SDK #57800
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Follow-up to #57032.
An executor should declare the integration it wants to support. That declaration (an import path) is sent to a task when the executor runs it as a part of the workload. The SDK loads that string to RuntimeTaskInstance, and the Sentry integration code uses it to import the relevant callable to run.
Since the integration information is only known when the executor passes it to the task runner, I changed the Sentry integration to be initialised lazily when
enrich_erroris called (instead of__init__i.e. initialised when the task runner process starts).