[SOLVED] After install server status shows errors for all services

[uxbod]

In-fact I had this problem on a vanilla 6.0.4 install; I did compound things though as I had moved SSH onto a non-standard port due to internal security policies. The resolution was re-creating the SSH key and updating; restarting ZCS and then as root running

/opt/zimbra/bin/zmsyslogsetup

. Once that was done switch to the zimbra user and execute

/opt/zimbra/libexec/zmstatuslog

and check that /var/log/zimbra.log was being updated

grep "STATUS" /var/log/zimbra-stats.log | grep stats | head -n10

[phoenix]

Originally Posted by Oded Arbel

On the face of it, the information you requested doesn't seem to be related to the logger (that I now believe to be the problem)

On the face of it you would be correct but incorrect DNS can cause problems with displaying logger data.

Originally Posted by Oded Arbel

.. and none of these things are mentioned in the Wiki page referenced.

That's correct it isn't but you also didn't supply the rest of the information from the wiki page that was asked for. Anyway, now you've confirmed the DNS is correct you need to follow the instructions that uxbod has posted and re-initialise the logger.

[Oded Arbel]

Thanks for the explanation.

After running 'zmsyslogsetup', the syslog-ng.conf file is not valid - if I try to restart syslog-ng, it fails and tells me the syntax is incorrect.

The problem is this line:

source zimbra_src { unix-stream("/dev/log"; keep-alive(yes); max-connections(20); }; # zimbra

Where syslog-ng doesn't understand the syntax where unix-stream() gets additional parameters. I changed it into:

source zimbra_src { unix-stream("/dev/log"); }; # zimbra

And now syslog-ng runs fine.

After that I did 'tail -f /var/log/zimbra*' and ran 'zmstatuslog' but there wasn't any output in the log files.

Can you offer any other things I may try?

[uxbod]

su - zimbra
zmloggerctl stop ; zmloggerctl start

and then run zmstatuslog again.

[Oded Arbel]

A stop and start just gives me:

Stopping logswatch...done.
Starting logswatch...done.

after which 'zmloggerctl status' still give me nothing.

Running '/opt/zimbra/libexec/zmstatuslog' also has no output, and nothing in any log (I also looked in /opt/zimbra/log/*), but its not surprising as from looking at the code zmstatuslog just calls 'zmcontrol status' and parses the output.

zmloggerctl is a bit more intersting - it calls zmlogswatchctl which in turn test if 'logswatch' is running by inspecting its pid and sending a signal, and it does run on my system, and also checks 'zmrrdfetch-server' which apparently isn't running on my system (the pid file does not exist) but zmlogswatchctl doesn't seem to mind because it concludes with the output

zmlogswatch is running


Now here is the most intresting part - as I've mentioned, zmstatuslog calls zm control and parses the status, so I ran it manually and it tells me this:

Host vhost1.heptagon.co.il
antispam Running
antivirus Running
imapproxy Running
ldap Running
logger Running
mailbox Running
memcached Running
mta Running
snmp Running
spell Running
stats Stopped


So I tried:

$ bin/zmstatctl status
Running: zmstat-allprocs
process 1193 in /opt/zimbra/zmstat/pid/zmstat-cpu.pid not running
process 1203 in /opt/zimbra/zmstat/pid/zmstat-fd.pid not running
process 1191 in /opt/zimbra/zmstat/pid/zmstat-proc.pid not running
process 1195 in /opt/zimbra/zmstat/pid/zmstat-vm.pid not running
process 1324 in /opt/zimbra/zmstat/pid/zmstat-mtaqueue.pid not running
process 1326 in /opt/zimbra/zmstat/pid/zmstat-nginx.pid not running
Running: zmstat-df
Running: zmstat-mysql
process 1197 in /opt/zimbra/zmstat/pid/zmstat-io-x.pid not running
process 1201 in /opt/zimbra/zmstat/pid/zmstat-io.pid not running


So I ran 'bin/zmstatctl start' and it "invoked" lots of stuff and then the above status returned all "Running".

Now I ran 'zmstatuslog' log again, which still has no output and nothing in the logs and the 'server status' screen is still showing all the local services as down, and "server statistics" still has the error

Message: system failure: Unable to read logger stats Error code: service.FAILURE Method: GetLoggerStatsRequest Details:soap:Receiver


I'm starting to give up on the thing, but if you have any more ideas, keep them coming :-)

[skuran]

Hi all,

My server status page broke after (probably) changing the ip address of the server. I followed the Final Set-up section of the Quick start guide again.

1. recreated the keys $zmupdateauthkeys --- the key is updated
2. as root ran #/opt/zimbra/bin/zmsyslogsetup
3. restarted rsyslog #/etc/init.d/rsyslog restart
4. restarted zimbra services
$zmcontrol stop
$zmcontrol start
5. Refreshed server status page

The problem is gone!

Cheers

[Oded Arbel]

Thanks for the suggestion. I just completed the process as you described it and I had no errors during the process (except the syslog setup which is broken as I commented previously and I had to fix by hand again).

After completing this, the server status screen is still all red. On the face of it it doesn't appear we have the same problem - I installed a new server, didn't change any IP so there should be no reason to regenerate keys.

[skuran]

this doesn't seem to be a authkey issue. my status page is broken again without changing any settings. Where should I begin?

[Oded Arbel]

You can try to see what zmstatctl status tells you, though restarting it did little to help me.

[davidg]

Originally Posted by Oded Arbel

... stuff ...

I've got the exactly the same problem as you have! Still looking around to find a solution since it bothers me i cannot see that info.

Any other 'official' word on this?

Edit: @Oded, what kind of OS are you using? I'm running SLES 10...



Regards,
D.