antispam/amavis services keep restarting

[Chris-M]

Hello,

I am noticing that the antispam/amavis services keep stopping and starting (note: they do appear to restart successfully, ie they aren't remaining stopped) often throughout the day.

The following is being logged in zimbra.log:

Code: Select all

Jun  1 00:45:11 mail amavis[13030]: Net::Server: 2018/06/01-00:45:11 Server closing!
Jun  1 00:45:16 mail amavis-mc[12473]: Master process shutting down
Jun  1 00:45:16 mail amavis-mc[12473]: Terminating process [12477] (/opt/zimbra/common/sbin/amavis-services msg-forwarder)
Jun  1 00:45:16 mail amavis-mc[12473]: Terminating process [12478] (/opt/zimbra/common/sbin/amavis-services childproc-minder)
Jun  1 00:45:16 mail amavis-mc[12473]: Terminating process [12479] (/opt/zimbra/common/sbin/amavis-services snmp-responder)
Jun  1 00:45:16 mail amavis-services[12477]: Task 'msg-forwarder' [12477] shutting down
Jun  1 00:45:16 mail amavis-services[12477]: msg-forwarder closing inner socket
Jun  1 00:45:16 mail amavis-services[12478]: Task 'childproc-minder' [12478] shutting down
Jun  1 00:45:16 mail amavis-services[12478]: childproc-minder closing inner socket
Jun  1 00:45:16 mail amavis-services[12477]: msg-forwarder closing outer socket
Jun  1 00:45:16 mail amavis-services[12477]: msg-forwarder closing context
Jun  1 00:45:16 mail amavis-services[12478]: childproc-minder closing outer socket
Jun  1 00:45:16 mail amavis-services[12478]: childproc-minder closing context
Jun  1 00:45:16 mail amavis-services[12478]: Task 'childproc-minder' [12478] exiting: TERM
Jun  1 00:45:16 mail amavis-services[12477]: Task 'msg-forwarder' [12477] exiting: TERM
Jun  1 00:45:16 mail amavis-mc[12473]: Process [12478] exited (/opt/zimbra/common/sbin/amavis-services childproc-minder) after 58729.4 s: exit 0
Jun  1 00:45:16 mail amavis-mc[12473]: Process [12477] exited (/opt/zimbra/common/sbin/amavis-services msg-forwarder) after 58729.4 s: exit 0
Jun  1 00:45:26 mail amavis-mc[12473]: Killing process [12479] (/opt/zimbra/common/sbin/amavis-services snmp-responder)
Jun  1 00:45:26 mail amavis-mc[12473]: Process [12479] exited (/opt/zimbra/common/sbin/amavis-services snmp-responder) after 58739.4 s: KILLED, signal 9 (0009)
Jun  1 00:45:26 mail amavis-mc[12473]: Master process exiting: TERM
Jun  1 00:45:29 mail zmconfigd[457]: Service status change: mail.domain.com amavis changed from running to stopped

Can anyone suggest how we can troubleshoot what is causing this? There are no load issues on the server (<~0.5) and the usage of the Zimbra server itself is very low.

We are running ZCS 8.8.6_GA_1906.FOSS on CentOS 6.9.

Thanks in advance.

[fetal]

I'm having this same issue as well. Were you able to figure it out. I thought I had made some headway when I upgraded memory on the server. It seems to happen less frequently now, but it still happens.

Ubuntu 14. 8..8.12

[Chris-M]

We no longer have this issue, but I don't recall what the solution was unfortunately. I presume we found a solution from searching on Google - would suggest you try the same, although if you've increased memory and this has lessened the problem that would suggest it may be a resource issue.

Perhaps try to reduce Zimbra's memory footprint and/or make it more performant:

https://wiki.zimbra.com/wiki/Making_Zim ... inimal_RAM
http://blog.jeshurun.ca/technology/zimb ... nce-tuning