>If you can automate suspension of host monitors, you will be well served.
>That way, when you take down part of a cluster (e.g. z/VM IPL), the central 
>host monitors don't start beeping. 
> I say "suspend" rather than "disable" because if they aren't back online by 
> the end of the service window, you WANT the alarms to kick in IF they see 
> that the VM systems is up.  (I say.  No point in whinging about Linux guests 
> not being up if the VM system is still down, wot?)


Right, you need that too.
You also want to know if you failed over to your backup device (oh, like say 
the backup device on a vswitch :) so you can go fix the primary.   So you must 
know those messages and alarm on them (or alarm on anything that you don't know 
about :).   And backup devices need their health monitored too so that you 
don't end up failing over to something that isn't really there. 

It's all a work in progress forever I think...   Never a shortage of things to 
do!

Marcy 

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Reply via email to