Monitoring clustered hosts

Roy Kidder nagios at jehster.net
Fri Sep 30 13:56:22 CEST 2005


Depending on which blade server you're using (my company has a couple of
first generation IBM blade boxes), there may not only be IPs assigned to
the individual boxes, but also to the chassis itself, and possibly to the
network interface (the IBMs have an integrated switch). If so, you might
want to think about those variables as well. Again, depending on the
manufacturer, SNMP and tcp/80,442 might be options.

As for the cluster of servers, if they each have a seperate IP, I would
approach it as 10 different boxes, each with their own services (mysql,
oracle, ssh, etc) and then wrap them all in a single service group.

Hope that helps,
Roy

Deborah Martin said:
> Hi,
>
> What would be the best way approach the following :
>
> 10  Blades running Linux (so each blade has its own IP address)
> Together they are one database server
>
> a) I want to monitor each blade to ensure they are up. If down I need
Nagios
> to tell me specifically which blade is down.
>
> I've played with the hosts.cfg file and put in multiple IP addresses but if
> one goes down, the info returned to Nagios isn't telling me
> anything specific (or do I need to tweak this myself ?)
>
> I'm using check_ssh rather than ping as there is a firewall between
Nagios and the blades which doesn't allow ping through.
>
> Any help would be appreciated!
>
>
> regards,
> deborah
>
>
> __________________________________________________
> Virus checked by MessageLabs Virus Control Centre.






-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null





More information about the Users mailing list