Revert "relax validation on the presence of the 'shouldRetry' field w…#18
Revert "relax validation on the presence of the 'shouldRetry' field w…#18bbaldino wants to merge 2 commits intojitsi:masterfrom
Conversation
This reverts commit 009420d.
9a12d81 to
d5ddb78
Compare
|
@bbaldino: What do you think about your PR now? |
|
I was just going through some old emails I hadn't gotten to and saw this, sorry for dropping the ball. I'm afraid I've forgotten most of the context, but do remember we did some enhancements on when to retry vs not for Jibri. @aaronkvanmeerten or @bgrozev |
|
I don't remember the reasons for this. The jibri's in use have definitely been updated by now, but it's hard to verify that no code ever sets failureReason without setting shouldRetry (but I trust you did that when you initially opened the PR) |
probably? :) but i've got the same concern. if things are working as-is, maybe it's best to leave it (i.e. not merge this)--at least not without someone doing some investigation to confirm. |
…hen there (#17)"
This reverts commit 725dcef.
DO NOT MERGE
This PR shouldn't be merged until we've cycled out all Jibris deployed before jitsi/jibri#232 is shipped. Once that is the case, we can merge this and enforce the presence of
shouldRetrywhen a failure is reported.