mailbox randomly stops

Discuss your pilot or production implementation with other Zimbra admins or our engineers.
BloodyIron
Advanced member
Advanced member
Posts: 67
Joined: Sat Sep 13, 2014 2:58 am
Contact:

mailbox randomly stops

Postby BloodyIron » Mon Nov 05, 2012 3:37 pm

Hey Folks,
Setup a new ZCS OSE 8.0.0 a few weeks ago on a VM with 2GB of RAM. We tried it on 1GB previously but it seemed more hassle than it's worth with that much RAM.
Currently it mostly runs properly. We have setup ZPush successfully.
The problem is the mailbox seems to randomly crash after a few days or so. We don't have many accounts or much traffic going through it just yet, so I don't suspect it's a load issue.
The closest thing I can come to something tangible is the /opt/zimbra/log/zmconfigd.log reporting:
/opt/zimbra/bin/zmconfigdctl: fork: Cannot allocate memory


Additionally, earlier today when troubleshooting it I noticed there was a pid issue for mailbox, preventing it from starting. I verified the pid didn't actually exist by trying to kill it, then I moved the pid file aside (instead of deleting), and the service restarted properly. I then deleted the old pid file, but this odd crashing seems to persist.
Unfortunately I do not have monitoring currently going so it's a pain that I have to manually determine this, so it makes it hard to determine when it _actually_ goes down.
Any ideas?


phoenix
Ambassador
Ambassador
Posts: 26522
Joined: Fri Sep 12, 2014 9:56 pm
Location: Liverpool, England

mailbox randomly stops

Postby phoenix » Tue Nov 06, 2012 5:06 am

You don't actually give any details of the load on this server (number of users etc.) but it is at the bottom end of the RAM requirements, I'd suggest you increase it and see how it goes. There's also no details of which operating system you're using nor anything about which hypervisor.
Regards

Bill

Rspamd: A high performance spamassassin replacement

Per ardua ad astra
BloodyIron
Advanced member
Advanced member
Posts: 67
Joined: Sat Sep 13, 2014 2:58 am
Contact:

mailbox randomly stops

Postby BloodyIron » Tue Nov 06, 2012 9:13 am

Whoops!
1) Load reports 0.07 / 0.20 / 0.28 as of this writing
2) Number of users, about 7 right now
3) Ubuntu Server 12.04
4) I don't know the hypervisor, I rent this VM.
5) 2GB is listed as sufficient RAM, why would this not be accurate? (adding RAM isn't the end of the world, but it adds up to our costs :/ )
phoenix
Ambassador
Ambassador
Posts: 26522
Joined: Fri Sep 12, 2014 9:56 pm
Location: Liverpool, England

mailbox randomly stops

Postby phoenix » Tue Nov 06, 2012 12:13 pm

[quote user="BloodyIron"]2GB is listed as sufficient RAM, why would this not be accurate?[/QUOTE]It is accurate but that's a minimum and the recommended is 4GB and would depend on the number of users and their usage profile.
[quote user="BloodyIron"](adding RAM isn't the end of the world, but it adds up to our costs :/ )[/QUOTE]Knowing the hypervisor might be helpful. Do you know if this is real RAM or burst RAM?
Regards

Bill

Rspamd: A high performance spamassassin replacement

Per ardua ad astra
cjm51213
Posts: 43
Joined: Sat Sep 13, 2014 2:43 am

mailbox randomly stops

Postby cjm51213 » Tue Nov 06, 2012 3:41 pm

If it helps you to know, I have Zimbra 8 (ocs) running on Centros 6.3, on 2GB RAM and 2GB swap, and that seems to be just barely enough... I have an exceptionally low volume installation with half a dozen users and less then 1,000 messages per day.
BloodyIron
Advanced member
Advanced member
Posts: 67
Joined: Sat Sep 13, 2014 2:58 am
Contact:

mailbox randomly stops

Postby BloodyIron » Fri Nov 09, 2012 10:20 am

I guess I'll have to look into maybe adding 2GB of SWAP, since we have plenty of space. Why exactly is 2GB listed as a minimum when it really isnt?
The RAM is dedicated, so far as I can tell it is not burst, and it is sold to me as dedicated. I suspect they will not tell me the hypervisor, nfoservers.com if anyone knows which one they're using.
I have been monitoring the RAM usage for the last few days and it doesn't _seem_ to be reaching capacity, but I can't conclusively rule this out just yet.


[quote user="10330phoenix"]It is accurate but that's a minimum and the recommended is 4GB and would depend on the number of users and their usage profile.
Knowing the hypervisor might be helpful. Do you know if this is real RAM or burst RAM?[/QUOTE]
BloodyIron
Advanced member
Advanced member
Posts: 67
Joined: Sat Sep 13, 2014 2:58 am
Contact:

mailbox randomly stops

Postby BloodyIron » Fri Nov 09, 2012 10:26 am

zimbra.log seems to have a fair bit of this, not sure if it's relevant:
Nov  9 10:19:34 (DELETED) zmconfigd[13488]: Fetching All configs

Nov 9 10:19:34 (DELETED) zmconfigd[13488]: All configs fetched in 0.04 seconds

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15862]: stale pid 20436 found in /opt/zimbra/log/zmmailboxd_manager.pid: No such process

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15862]: assuming no other instance is running

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15862]: file /opt/zimbra/log/zmmailboxd.pid does not exist

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15862]: assuming no other instance is running

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15862]: no manager process is running

Nov 9 10:19:34 (DELETED) zmconfigd[13488]: Executed: /opt/zimbra/bin/zmstorectl status norewrite returned 1 (30 - 13) (0.03 sec)

Nov 9 10:19:34 (DELETED) zmconfigd[13488]: Tracking service mailbox

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15873]: stale pid 20436 found in /opt/zimbra/log/zmmailboxd_manager.pid: No such process

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15873]: assuming no other instance is running

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15873]: file /opt/zimbra/log/zmmailboxd.pid does not exist

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15873]: assuming no other instance is running

Nov 9 10:19:34 (DELETED) zmmailboxdmgr[15873]: no manager process is running

Return to “Administrators”

Who is online

Users browsing this forum: No registered users and 12 guests