Thu, 14 May 2009
From the chapter Things, that are so obvious... once they failed
Configuring your monitoring system via DHCP, and wondering why the malfunction of one single server can cause so dramatic alerts in your entire infrastructure.
postet at 09:23 into [Debian] permanent link
Wed, 18 Mar 2009
Slides from CLT talk available
The slides from my talk at the Chemnitzer Linux-Tage are now available. As promised I added additional screenshots of the parts I showed directly.
Again many thanks for showing up in my talk and the feedback I got so far!
Update: I should mention, that the topic was Projektmanagement
mit Subversion und Trac
and the talk as well as the slides are in
German.
postet at 23:18 into [Debian] permanent link
Fri, 13 Mar 2009
Dear Lazyweb,
Since a recent update of one of my systems, amavisd-new keeps on spamming me. I get mails containing stuff like bayes: synced databases from journal in 0 seconds: 674 unique entries (742 total entries). The reasons seems to be a cronjob from amavisd-new (which can be deactivated via debconf):
grep amavisd-new-cron /etc/cron.{d,daily}/* /etc/cron.d/amavisd-new:18 */3 * * * amavis test -e /usr/sbin/amavisd-new-cronjob && /usr/sbin/amavisd-new-cronjob sa-sync /etc/cron.daily/amavisd-new:test -e /usr/sbin/amavisd-new-cronjob && exec /usr/sbin/amavisd-new-cronjob sa-clean
However, the called script contains:
egrep "sa-sync\)|sa-clean\)" -A 2 /usr/sbin/amavisd-new-cronjob sa-sync) do_amavis_cmd "/usr/bin/sa-learn --sync >/dev/null" ;; sa-clean) do_amavis_cmd "/usr/bin/sa-learn --sync --force-expire >/dev/null" ;;
Note that output is redirected to /dev/null, so I should only mails if an error occurs, shouldn't I? I'm really baffled... To the best of my knowledge I shouldn't get these mails. Where am I wrong, and how to formulate that in a bug report?
postet at 15:42 into [Debian] permanent link
Sun, 15 Feb 2009
Greetings...
... from Samoa!
Have fun with Lenny!
PS: Next time, if we release on Valentines, let's choose a single to fly around the globe just to send some mails out...
postet at 21:51 into [Debian] permanent link
Tue, 27 Jan 2009
localhost nagios checks via nrpe
I'll write it down, so maybe I won't forget it: If you want to check
localhost via Nagios nrpe (e.g. because you are lazy and want to reuse all
the nrpe configuration you already have for other hosts), but it fails with
CHECK_NRPE: Error - Could not complete SSL handshake.
, then make
sure, that you allowed 127.0.1.1 in your allowed_hosts.
postet at 15:37 into [Debian] permanent link