04-20-2016 08:26 AM
Hi,
I have the following set up on my cluster:
The main interface
eth0
And 2 vlan tagged interfaces
eth0.100
eth0.200
Each interface has been plumbed with an IP and I have a NIC resource set up to monitor them
Seeing as the server is a HP blade using virtual connect, I cannot use Mii to monitor the NICs so I need to use packet counting.
So the problem is if eth0 fails but eth0.100 and eth0.200 are still OK, the NIC resrouce for eth0 will not go into a faulted state as it is counting all packets that are received including the vlan tagged packets. Is there anything I can do to get around this other than writing my own agent?
All suggestions welcome,
Thanks,
Rossa.