Description of problem: Few of us have noticed that there is a long delay before we receive mails (in our inbox) posted to gluster-devel mailing list (unless one is in 'To'/'CC' list). For example the last mail we received (as of today) is from Kaleb posted on 28th Sep titled "Community Gluster Package Matrix, updated". But there are so many other mails sent, post this mail, to gluster-devel mailing list - http://www.gluster.org/pipermail/gluster-devel/2016-September/date.html There is similar issue with gluster-users ML as well. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Seems to be mailman needing a restart. I also see issue on rsyslog side, that i was investigating yesterday, causing delay into logging (like ssh login), so that's likely the problem.
So while I do not think this is the cause, I also rebooted supercolony for various upgrade, and because dmesg did show weird cpu errors messages. Restarting syslog fix the issue, but that's not a good fix, as I do not know what cause it.
Ok so I found a potential issue in the rsyslog config, who is configured to get stuff only from rackspace. I am submitting a fix in the upstream role and hotfix for now on the server.
So here is the fix I pushed upstream: https://github.com/OSAS/ansible-role-rsyslog/pull/6 and while I wait for review to be integrated, I pushed this to at least have the fix: https://github.com/gluster/gluster.org_ansible_configuration/commit/a9ea05b45446dca7363c72130ccbe7dc056db123
This is now fixed.