Accessing OMD unter CentOS and Selinux

If you've got the error like this

OMD: Site Not Started

You need to start this site in order to access the web interface.


in your fresh configured and started OMD site, Selinux can cause the issue. Consider to configure Selinux properly (have a lot of fun!) or disable it in /etc/syssconfig/selinux.

Fortigate firewall check package for Check_MK updated

I've updated the package to the version 1.6. This version contains an additional check of IPSec tunnel status of a Fortigate firewall. It works for FGT-300C and FGT-60C units. Other models can work as well, but I could not test.

You can download the new package at the known place from the Check MK Exchange repository.

OMD fails to start after CentOS upgrade to 6.4

I faced a problem starting OMD 0.54 after upgrading CentOS to the version 6.4. The start script returns a bunch of error messages:


[root@omd ~]# omd restart mysite
Doing 'restart' on site mysite:
Creating temporary filesystem /omd/sites/mysite/tmp...mount: can't find /omd/sites/                         mysite/tmp in /etc/fstab or /etc/mtab
ERROR
Removing Crontab...
Stopping xinetd...OK
Stopping icinga...Not running. OK
npcd was not running... could not stop
Stopping rrdcached...waiting for termination...OK
Stopping dedicated Apache for site mysite....OK
Starting dedicated Apache for site mysite...OK
Starting rrdcached...OK
Starting npcd...touch: cannot touch `/omd/sites/mysite/tmp/pnp4nagios/run/npcd.pid'                         : No such file or directory
chown: cannot access `/omd/sites/mysite/tmp/pnp4nagios/run/npcd.pid': No such file or directory
An Error occured while reading your config on line 197
Message was: "Could not open pidfile '/omd/sites/mysite/tmp/pnp4nagios/run/npcd.pid                         ': No such file or directory"
OK
/omd/sites/mysite/etc/rc.d/80-icinga: line 54: /omd/sites/mysite/tmp/nagios/nagios.cfg                         : No such file or directory
Icinga configuration file /omd/sites/mysite/tmp/nagios/nagios.cfg not found. Terminating...
Starting xinetd...OK
Initializing Crontab...OK

I found a couple of reports about the same problem. The root cause seems to be a bug (or a feature?) in the package util-linux-ng-2.17.2-12.9.el6.x86_64 that prevents non root users to mount partitions from /etc/fstab, whose path name conatains a symlink (symbolic link). In case of OMD /omd is a symlink pointing to /opt/omd.

A workaround that helped me to solve the problem was adding /opt to the appropriate path in /etc/fstab:

tmpfs                   /opt/omd/sites/nysite/tmp  tmpfs   noauto,user,mode=755,uid=mysite,gid=mysite 0 0

Check_MK check for Fortigate firewalls

I've got a Fortigate 60C and wanted to monitor it in Check_MK. I could find a plugin written by Justin McMillen, but it didn't work in the version 1.1.2 of Check_MK. Also they seemed to use SNMP OIDs, which my Fortigate 60C didn't have. So I've updated his checks and created a new version of the plugin. The checks use SNMP to get CPU and RAM usage as ell as the session count. You can download the package from Check_MK Exchange site.

Version 1.3.2 of Ex2010_MBDB_Info released

Again I've released a new version of the check. I've changed only the Windows plugin. Now the spaces in a mailbox database name are replaced by underscores. I hope this change would help all of you, who experienced problems with the plugin.

To update it's enough to download the new check package, install it in your check_mk instance by a command like this:

cmk -vP install Ex2010_MBDB_Info-1.3.2.mkp

and substitute the cmd file on your Exchange server by the Ex2010_MBDB_Info.ps1 file from the new package.

You can find the new version of the package at the known place at the check_mk plugin exchange site.

Version 1.3.1 of Ex2010_MBDB_Info released

I've released a new version of the check. I've changed the Windows plugin. It is now a pure Powershell script, not a command script anymore. It should run smoothly to avoid problems, which some of you had in the past running the plugin.
To update it's enough to download the new check package, install it in your check_mk instance by a command like this:

cmk -vP install Ex2010_MBDB_Info-1.3.1.mkp

and substitute the cmd file on your Exchange server by the Ex2010_MBDB_Info.ps1 file from the new package.

You can find the new version of the package at the known place at the check_mk plugin exchange site.

Check_MK check to show some data about Ex2010 mailbox databases

A few days ago I finished to test a new check for check_mk and published it as package at Check_MK Exchange. I named the check "Ex2010_MBDB_Info". In this article I'd like to share a couple of details about the check and its functionality.

The purpose of the check is to get the size of mailbox databases and the count of the mailboxes in each of them. The check package consists of the check itself, the plugin for check_mk agent, the perf-o-meter helper file and the pnp-template.

Check_MK plugin to get firmware and service tag of a DELL Powerconnect switch

check_mk is a great monitoring application! I've been trying it for a couple of days first and I am convinced that check_mk has a big future due to its nice conception, super built-in capabilities and the ability to create own check plugins with less effort.

I also tried to create a new check plugin for DELL Powerconnect switches, because the check plugins for such switches provided by check_mk distribution cannot show the firmware version and the service tag of the switch. I copied a provided check plugin (to be more accurate: snmp_info)and modified it a little bit. I'd like to share my result below, please enjoy it!