The mailbox and mta dies in FC4 GA version

[meikka]

Hi

I have problem that the mailbox and mta dies. Ihave upgraded the system from M2 , which was working OK. The system in FC4 in Xen virtual machine with 738 MB memory.

I have attached part of the zimbra log file and the only thing what has happened it the reload of postfix configuration.

What may cause the stop of the services? Where can I find more information of the cause?


-- Timo

Release 3.0.0_GA_156.FC4_20060206095649 FC4 FOSS edition

[zimbra@mail root]$ zmcontrol status
Host mail.virtaneva.com
antispam Running
antivirus Running
ldap Running
logger Running
mailbox Stopped
mta Stopped
snmp Running
spell Running


Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: antispam: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: antivirus: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: ldap: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: logger: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: mailbox: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: mta: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: snmp: Running
Feb 9 22:32:10 mail zimbramon[16574]: 16574:info: 2006-02-09 22:32:04, STATUS: mail.virtaneva.com: spell: Running
Feb 9 22:32:16 mail postfix/postfix-script: refreshing the Postfix mail system
Feb 9 22:32:16 mail postfix/master[4454]: reload configuration /opt/zimbra/postfix-2.2.5/conf
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: antispam: Running
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: antivirus: Running
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: ldap: Running
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: logger: Running
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: mailbox: Stopped
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: mta: Stopped
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: snmp: Running
Feb 9 22:33:08 mail zimbramon[17286]: 17286:info: 2006-02-09 22:33:02, STATUS: mail.virtaneva.com: spell: Running

[marcmac]

Xen virtual machines don't seem to work well with zimbra.

That said - search the forums on ways to get details about what exactly is dying, and how to get logs for it, and post the logs.

[meikka]

Hi

I have spent 2 days to find out the cause of the death of those services. I have searched all the forums without finding any solutions .

I does not help much to say that I have a lousy system and I should search .

I was hoping that someone could give a hint what I have missed.

-- Timo

PS. The Zimbra is great, but I need to get it working before rollout.

[marcmac]

What's in /opt/zimbra/log/zimbra.log?
su - zimbra
sh -x bin/zmmtactl status
sh -x bin/zmmailboxctl status

I can't help you if I don't know what's not running.

[meikka]

I will give the results tomorrow.

-- Timo

[drewage]

I am having this same problem since upgrading to GA from Milestone 3 Beta Update 2. It worked fine immediately after the upgrade, and then I went to check on it after a day of not using it and it was unresponsive.

zmcontrol status
Host myhost.com
        antivirus               Running
        ldap                    Running
        logger                  Running
        mailbox                 Stopped
        mta                     Stopped
        snmp                    Running
        spell                   Running

sh -x bin/zmmtactl status
+ BASE=/opt/zimbra
+ SCRIPTS='zmmtaconfigctl postfix zmsaslauthdctl'
+ case "$1" in
+ STATUS=0
+ for i in '$SCRIPTS'
+ /opt/zimbra/bin/zmmtaconfigctl status
+ R=0
+ '[' 0 -ne 0 ']'
+ for i in '$SCRIPTS'
+ /opt/zimbra/bin/postfix status
+ R=0
+ '[' 0 -ne 0 ']'
+ for i in '$SCRIPTS'
+ /opt/zimbra/bin/zmsaslauthdctl status
cat: /opt/zimbra/cyrus-sasl-2.1.21.ZIMBRA/state/saslauthd.pid: No such file or directory
+ R=1
+ '[' 1 -ne 0 ']'
+ STATUS=1
+ exit 1

and

sh -x bin/zmmailboxctl status
+ BASE=/opt/zimbra
+ SCRIPTS='zmmtaconfigctl tomcat mysql.server'
+ STATUS=0
+ case "$1" in
+ for i in '$SCRIPTS'
+ '[' zmmtaconfigctl = mysql.server ']'
+ /opt/zimbra/bin/zmmtaconfigctl status
+ R=0
+ '[' 0 -ne 0 ']'
+ for i in '$SCRIPTS'
+ '[' tomcat = mysql.server ']'
+ /opt/zimbra/bin/tomcat status
+ R=1
+ '[' 1 -ne 0 ']'
+ STATUS=1
+ for i in '$SCRIPTS'
+ '[' mysql.server = mysql.server ']'
+ /opt/zimbra/bin/mysqladmin status
+ R=0
+ '[' 0 -ne 0 ']'
+ exit 1

Any ideas ?

[marcmac]

is zimbraMtaAuthHost set? is zimbraMtaAuthURL set? Is the mode (http/https) in zimbraMtaAuthURL consistent with the port your tomcat instance is listening on? If not, reset zimbraMtaAuthHost to the value it's currently set to - this will reset zimbraMtaAuthURL (and work around a bug).

I believe that saslauthd is probably dying because tomcat is dying.

What happens when you run tomcat start? What shows up in /opt/zimbra/log/zimbra.log? /opt/zimbra/tomcat/logs/catalina.out?

After you start tomcat, try starting saslauthd.

[drewage]

is zimbraMtaAuthHost set? is zimbraMtaAuthURL set? Is the mode (http/https) in zimbraMtaAuthURL consistent with the port your tomcat instance is listening on?

Can you point me in the direction of how to find this information ?

I know the mode is http...everything worked fine the first couple of days after upgrading.

[marcmac]

Originally Posted by drewage

Can you point me in the direction of how to find this information ?

I know the mode is http...everything worked fine the first couple of days after upgrading.

There's a bug that resets the mode on upgrade, which is why I asked.

zmprov gs | grep zimbraMtaAuth
zmprov gs | grep zimbraMailMode
zmprov gs | grep zimbraMailPort
zmprov gs | grep zimbraMailSSLPort

[drewage]

Yes, I encountered that bug after upgrading and changed it to http

zmprov gs | grep zimbraMtaAuth
zmprov gs | grep zimbraMailMode
zmprov gs | grep zimbraMailPort
zmprov gs | grep zimbraMailSSLPort

Does that pipe to a file or something ? It provides no screen output when I run those commands..