Dovecot connections dropping at 60s during large UID FETCH
Hi guys,
I'm hitting a wall with a specific user on my server (CloudLinux 9 + cPanel/WHM). The user is getting disconnected frequently, and I'm seeing this in the logs:
Disconnected: Connection closed: read(size=7128) failed: Connection reset by peer (UID FETCH running for 0.055 + waiting input/output for 61.227 secs, 58 B in + 25664970+8192 B out, state=wait-output)
It seems like every time the client tries to fetch a large chunk of data (around 25MB according to the log), the connection dies exactly at the 60-second mark.
I'm planning to disable the Hibernate timeout (set it to 0) and bump the IP connections limit, but I wanted to ask if there's any other timeout I might be missing that would trigger exactly at 60 seconds.
Could the "Hibernate" feature be killing the process while it's still busy pushing data to a slow client? Any other tips on what to look for?
Thanks in advance!
-
Hey there! I'm not finding anything on my end that would trigger at exactly 60 seconds, and even the Hibernate value you mentioned defaults to 30. I'd try that and then possibly submit a ticket if it's something you can reproduce.
0 -
Can I open a ticket if I purchased my License through Licensepal?
0 -
You would open the ticket through your license provider and then they would escalate the issue to us if necessary:
0
Please sign in to leave a comment.
Comments
3 comments