Bug 1494280 - Supercolony seems to be overloaded on a regular basis
Summary: Supercolony seems to be overloaded on a regular basis
Alias: None
Product: GlusterFS
Classification: Community
Component: project-infrastructure
Version: mainline
Hardware: Unspecified
OS: Unspecified
Target Milestone: ---
Assignee: bugs@gluster.org
QA Contact:
Depends On:
TreeView+ depends on / blocked
Reported: 2017-09-21 22:11 UTC by M. Scherer
Modified: 2018-10-08 03:15 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Last Closed: 2018-10-08 03:15:10 UTC
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:

Attachments (Terms of Use)

Description M. Scherer 2017-09-21 22:11:45 UTC
Description of problem:

Not sure why, but supercolony seems to take a long time to let me connect with ssh since a few days. I did look and can't see anything overly wrong.

Munin show some mail activity, but this should result into slowdown for login.

Looking at postfix log, it seems to not do g suspicious (even if we have postgrey message, so I did restart postgrey). I see nothing weird either on mailman log, no specific acitivty on I/O, network, or CPU. 

Will look later

Comment 1 M. Scherer 2017-09-22 13:02:47 UTC
So log also stopped, so it might be related to that. (the syslog server seems to also be full...)

Comment 2 M. Scherer 2017-09-22 13:20:56 UTC
There is also a ton of errors: 
Sep 22 12:59:19 supercolony postfix/smtp[28860]: certificate verification failed for aspmx.l.google.com[]:25: untrusted issuer /C=US/O=Equifax/OU=Equifax Secure Certificate Authority

But the problem seems to be linked to rsyslog, I removed the logs export and now, everything is faster.

Comment 3 M. Scherer 2017-09-22 13:29:07 UTC
Ok so the log server is full, and it seems to slow down all servers sending something there.

Comment 4 M. Scherer 2017-10-09 12:06:00 UTC
So, ansible did reinstall the logs export and login on supercolony was slow again. I have cleaned (or rather) compressed the log on the syslog server. Seems for some reason to have generated lots of logs by itself in the past.

Comment 5 Nigel Babu 2018-10-08 03:15:10 UTC
This is now fixed and I'm going to close the bug.

Note You need to log in before you can comment on or make changes to this bug.