Replies: 3 comments 2 replies
-
I quite like the idea behind this. Sometimes I'll have a set of jobs fail for the same reason (API is down, whatever) and it would be handy to be able to retry those jobs in one go without having to dig out their ID's or click them in Horizon. Would it be worth a confirmation step that showed what was going to be retried? Just in case of a typo or over-broad 'pattern'? List of jobs and a y/n prompt? |
Beta Was this translation helpful? Give feedback.
-
You could tell the job to retry forever with an exponential backoff. When the API comes back online, the job will be retried and succeed. |
Beta Was this translation helpful? Give feedback.
-
Taylor didn't wanna merge because it's a clutter #49072 so that answers it |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
When there are various failed jobs eg.: email failed to send
it would be nice if I could retry only a certain type of jobs
On the
failed_jobs
there's anexception
column includes the whole stack trace something likeI propose we could include a new argument
--pattern
to which you would normally pass a namespaced error such asSymfony\Component\Mailer\Exception\TransportException
This would run all jobs where the exception is born out this error
Beta Was this translation helpful? Give feedback.
All reactions