Server Status: ERROR

[NaughtyusMaximus]

Hi all,
When I go to the Server Status section, I get this message:

"Server status data is not available. To see the server status, logger service must be installed."

The logger shows as running from zmcontrol. If look in the logs, I can see the following:

"*** logswatch version 3.2.3 (pid:23254) started at Sat Nov 29 23:52:33 PST 2008

DBI connect('database=zimbra_logger;mysql_read_default _file=/opt/zimbra/conf/my.logger.cnf;mysql_socket=/opt/zimbra/logger/db/mysql.sock','zimbra',...) faile$
Can't call method "prepare" on unblessed reference at /opt/zimbra/libexec/zmlogger line 85, <> line 26983.
"

Any thoughts? I'm worried this is a symptom of a larger problem.

[phoenix]

This has been covered many times in the forums, a search will find further details. You could, however, start here: Logger - Zimbra :: Wiki

[NaughtyusMaximus]

I searched before I posted, but apparently I don't know what to be searching for. Thank you for the wiki link - unfortunately I was unable to find anything useful there, nothing on that page appeared to be applicable. Most of the problems assume that the logger service isn't running. Mine is running, but I'm unable to connect to MySQL, and get the error 'too many connections'.

Searching for 'too many connections', I did come up with this thread: [SOLVED] Daily reports stopped sending - Mysql Too many connections

The fix seems to be increasing the line:
max_connections = 10

in /opt/zimbra/cong/my.logger.cnf

Is there danger in doing this? And what is a safe number to increase it to?

Thanks again for any assistance

[NaughtyusMaximus]

phoenix, would you mind posting a link to one of the other threads this has been discussed in? I'm still unable to find anything applicable.

Thanks again

[justvodka]

I have the same problem. When looking into this problem I found the there were 10 zmlogger processes connected to the logger mysql daemon via unix sockets.
Those zmlogger processes are spawned by tempory swatch scripts

#ps -fax (only relevant output and truncated the long logswatch line)
16152 ? S 0:09 /usr/bin/perl /tmp/.swatch_script.26133
17037 ? S 0:01 \_ /usr/bin/perl /opt/zimbra/libexec/zmlogger
16251 ? S 0:00 /usr/bin/perl /opt/zimbra/libexec/logswatch [..cut..]
16281 ? S 0:08 \_ /usr/bin/perl /tmp/.swatch_script.16251
16986 ? S 0:01 \_ /usr/bin/perl /opt/zimbra/libexec/zmlogger

But it looks like the zmlogger processes and the .swatch_script.xxxx don't get terminated properly. Every day I get new ones.
after "zmcontrol stop" those .swatch_script.xxxx and zmlogger processes are still there.

So setting max_connections to a higher value isn't the real sollution. If set to 100, after 100 days you would probably get the same problem.

[phoenix]

Download the replacement zmlogswatchctl from this bug report (comment #7), make a copy of the original zmlogswatchctl and then place the downloaded copy into the original directory. Stop Zimbra with:

zmcontrol stop

and check to see if any Zimbra processes are still running, if they are kill them and then start Zimbra again. The replacement script will work correctly and not leave any ghost copies around when it stops/starts the logger.