Skip to content

Conversation

@romain-priour-lc
Copy link
Collaborator

Currently, if a job is swept, there is no mechanism to retry it. This PR adds a "retryable" field, which the sweeper then uses to retry jobs instead of aborting them if they are stuck.

This enables jobs to be retried even if the worker goes OOM.

  1. Worker goes OOM, and job is stuck
  2. Sweeper from another worker eventually picks up the stuck job, and sets it to retry
  3. The job is eventually picked up from the queue at a later point in time.

Before these changes, the stuck job was swept, and immediately aborted.

@QuentinBrosse
Copy link

FTR: INF-1032

Copy link

@QuentinBrosse QuentinBrosse left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for that!

@romain-priour-lc romain-priour-lc merged commit d256e30 into main Oct 27, 2025
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants