On 15/01/07, Rui Miguel Silva Seabra rms@1407.org wrote:
I've got a couple of machines acting as directors with service=lvs
Is there something similar to 'clustat' for pulse+lvsd?
I need this in order to make our monitorization scripts "cluster-aware", or else one node will have critical status when there are less nannies than it expected because the services ran over to the other node.
I'm not entirely sure which combination of packages you have installed from your description.
If you're using the heartbeat, heartbeat-ldirectord and ipvsadm packages from the CentOS Extras repository (or the Ultramonkey packages for example) then perhaps...
# ipvsadm -Ln
or
# service ldirectord status # service heartbeat status
... will give you what you need? Or you could write something that regularly polls /var/log/ha-log for information?
Will.