|
> a failover should be initiated whenever a resource on the primary fails.
> Unfortunately Heartbeat only has the capability to detect a complete
server
> failure, not resource failures. In order to detect those failures you have
> to use another program like mon to monitore your resources (e.g. network
> connectivity) and let mon take appropriate steps in case of a failure (e.g
> shutdown Heartbeat or reboot server).
I am aware that heartbeat cannot monitor resources, but while it is possible
to monitor network connectivity of the active machine with mon, it's
impossible ot monitor the service running on the vip from the same machine
(or so it seems from the docs), thus impossible to check if the lvs is
functioning correctly. the question is, has anyone seen a case where lvs
fails (and I am not talking about the real servers, but about the load
balancer module) but the machine lives on, and if such a case exist, does
anyone has a solution for monitoring and taking action when such failure
occures ?
of course I will have external monitoring on the cluster from external machine,
but if it detects a failure in the cluster the only thing it will be able to do
is to alert me.
|
