maillog of 2GB - logrotate not working

[ghazlewood]

I have a 9.2.1 CentOS 5 92090422.13 server which has a 2GB maillog file. For some reason logrotate isn't rotating this file.

[root@server log]# ls -lah
-rw-r----- 1 root root 2.0G Mar 3 10:16 maillog
-rw-r--r-- 1 root root 175M Mar 3 09:56 maillog.processed.1.gz
-rw-r--r-- 1 root root 2.1M Oct 1 04:54 maillog.processed.2.gz
-rw-r--r-- 1 root root 2.0M Sep 30 05:04 maillog.processed.3.gz
-rw-r--r-- 1 root root 335K Mar 3 04:53 smtp_parsed.log
-rw-r--r-- 1 root root 386 Oct 1 04:54 smtp_pendings.log
-rw-r--r-- 1 root root 136 Mar 3 04:53 smtp_pendings.tmp
-rw-r--r-- 1 root root 0 Feb 26 04:02 xferlog
-rw-r--r-- 1 root root 5.5M Feb 26 04:02 xferlog.processed
-rw-r--r-- 1 root root 834K May 8 2010 xferlog.processed.1.gz

I've tried re-running the /usr/local/psa/admin/sbin/statistics command without any success

[IgorG]

Try to run

# /usr/local/psa/bin/sw-engine-pleskrun /usr/local/psa/admin/plib/DailyMaintainance/script.php

from /etc/cron.daily/50plesk-daily and check that it was logrotated.

[ghazlewood]

OK running the command you suggest highlights that there are some databases which it cannot read properly.

statistics: Unable to get dir size of /var/lib/mysql/data-base_namewithhypen

Looking at them they are only DBs with hypens in the name. I am renaming these databases, I remember reading a similar thread but did not realise that I too had databases with malformed names.

I also get lots of

[new_unode] Warning: String exceeds storage size (711)

Can I safely ignore these messages?

[ghazlewood]

I still have an ever increasing maillog file even after correcting the database names. Any further suggestions please?