Replies: 2 comments 3 replies
-
This is a file descriptor leak, and considering I haven't seen any report of it yet, nor have I encountered it myself, it is likely to be configuration-dependent. It could be in the OpenStack SWIFT handling code indeed, since it's relatively infrequently used. It may help if you could post the output of something like |
Beta Was this translation helpful? Give feedback.
1 reply
-
Same Issue here.
|
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi there, i started having issues with the queue after the recent update and i am not sure if the problem lays on my end (eg. configuration mistake).
Sidekiq logs a number of
exceptions while puma shows a number of "too many requests" for the OVH object storage around that time. I have noticed the hint about S3 single request in the update notes, but would it have effect for the SWIFT module?
I am not sure how to read the sidekiq error bo be honest. It sounds unusual and confusing for me. I mean i can raise the hard/soft-limit, but isn't the problem laying somewhere else here? 🤔
Update: according to users, there are more instances with similar symptoms (timeline not properly updating in particular). My current workaround is a restart cron for every 3h.
Beta Was this translation helpful? Give feedback.
All reactions