A/P Cluster failed after shutdown Remote Node

Hi,

my remote node of the A/P cluster had a issue with the BBU module so I had to do a shutdown. First I tried a reboot and checked if the iSCSI volumes was accessible and they did as expected. Then I did a shutdown of the Remote node, no warnings like "your cluster will be stopped" displayed. After the machine was down, the volumes was no longer accessible.

After I started the cluster manually on the primary node (getting a warning that this is not recommended) the volumes was back online.
As soon as the remote node powered up again it joined the cluster automatically without problems.

I did shutdowns may times before without issues... Maybe you have an idea why it crashed this time?!
Something wrong in my configuration?


Network Configuration:
eth0 - iSCSI Path 1
eth1 - WebGUI
eth2 - iSCSI Path 2
eth3 - Volume Replication (bond0)
eth4 - Volume Replication (bond0)
eth5 - iSCSI Path 3

Failover Configuration:
4x Auxilary paths over eth0, eth2, eth5 (the three iSCSI paths) and bond0 (vol.repl.)
4x Ping Nodes (2x VMware hosts, 2x network switches) - all of them was availible all the time

Erros in log:
Cluster: dss12209856 (remote node) could not be reached. (that's ok...)
Cluster: Auxiliary path eth0 is not available. (for all aux.paths.... that's ok...)
Cluster: All ping nodes are not available. (why??? all ping nodes was availible all the time)
Cluster: All auxiliary paths are not available, access to dss39068452 (local node) resources pool will be suspended.



Thanks.

Bye, Manuel

It sound like the Remote node had the volumes in source mode. What version are you running? For the ping nodes not being seen where you able to go into the Console screen and use the Ping tool from the Console tools to see if you can ping them?