• The Horde webmail has been deprecated. Its complete removal is scheduled for April 2025. For details and recommended actions, see the Feature and Deprecation Plan.
  • We’re working on enhancing the Monitoring feature in Plesk, and we could really use your expertise! If you’re open to sharing your experiences with server and website monitoring or providing feedback, we’d love to have a one-hour online meeting with you.

logrotate ALERT and lastlog really big

M

manarak

Guest
Hi

in my messages log, it tells me:

logrotate ALERT existed abnormally with [1]

plus, my lastlog is really big.

what should I check?
 
Check for any log files which have grown too big (over 2 GB?) in size.
 
I checked the contents of var/log and there are no really large logs, except lastlog with 19MB.
 
What about the other log locations:

/usr/local/psa/var/log
/home/httpd/vhosts/domain.tld/statistics/logs

Also, do you have your /tmp set for noexec? If so, then you may want to check out this info I found by googling:
As a quick fix, concerned users will need to edit
/etc/cron.daily/logrotate and assign TMPDIR to a partition that is not
mounted noexec; e.g. TMPDIR=/var/tmp; export TMPDIR
 
ok, will check the logs.
the tip about noexec could be it, I mounted /tmp as noexec.

will give feddback if this worked in the evening.

thank you
 
hmmmm...

Logs in stats dirs and psa dir are all ok.

I executed
/usr/sbin/logrotate /etc/logrotate.conf
and no error came up??

=> So I am still chasing the error.

It seems the lastlog is not being rotated at all (not even after I executed the above)
Where does one tell the server what logs to rotate?
 
Where does one tell the server what logs to rotate?
/etc/logwatch.conf

/etc/logrotate.d/*

/usr/local/psa/etc/logrotate.d/*
 
hmmm... probably my syslog logrotate job is failing because of /tmp being noexec.

I added export TMPDIR=/var/logtmp to longrotate.conf
and I added lastlog to my syslogs.

I then executed logrotate by hand.

But it seems nothing changed with the logs.

The server seems to be logging all the syslogs to <name of log>.1 (archived log 1) instead of logging to the plain logname file...

i.e.

messages has size 0, date is one week old
messages.1 has a big size and is still being updated.

I got this mess since I have upgraded!
 
Did you restart the syslog daemon/service?

Also check the values set in /etc/syslog.conf for what gets logged to where... something like:

Code:
*.info;mail.none;news.none;authpriv.none;cron.none     /var/log/messages
It may be silly, but have you checked the ownership and permissions on /var/log/messages ?
 
Back
Top