GUI show all services down, CLI shows all ok

[ekkas]

Release 6.0.5_GA_2213.RHEL5_64_20100203001950 CentOS5_64 FOSS edition.

This is a problem/complicated setup...

It is 2 servers running HA with DRBD & Heartbeat.
The problem arised after the drive was full due to a backup script that made a copy of everything without enough drive space.
Now I get :
Under server stats, last 48 hours: No data available
GUI server status: All red crosses
CLI zmcontrol status: All running


Now the problem is this machine has 2 hostnames, 2 IPs, depending on the HA status. 'Real' hostname zim1 resolves to 10.10.10.1(allways), but while it is active(HA-primary) it also has mail.mydomain.com which resolves to 10.10.10.3.

Now in the :
zmprov -l gcf zimbraLogHostname
it return the mail.mydomain.com
but hostname -f
returns zim1.mydomain.com
It used to all work before the full-drive.

Only errors I see in /var/log/messages is:
aspell[12068]: segfault at 0000000000000000 rip 0000003d57e4cef5 rsp 00007fffd9cd2f60 error 4

and in /var/log/zimbra.log I don't see anything strange. (Although I might have missed it)

I tried:
/opt/zimbra/bin/zmsyslogsetup
and the zmcontrol stop/start
but still the same.

Anything else I can try to get the GUI to record stats and show services up again? (Apart from a reboot?)

Thanks in advance.

Ekkas

[sirio81]

I have the same problem after an upgrade from zimbra 5.0.16 to 6.0.10 on centos 5 64bit.

[ekkas]

If I remember correctly it was fixed by first making sure there is enough free space, then repaired the SQL tables as showed in the wiki.