active-active ISCSI: Failover takes a long time for moving

[rbg]

hi,

we have 2 x DSS7up12 with unlimited space and active-active failover license. At the moment I have the problem, that I want to move over the targets to the seconds node, it takes very long time and isn't finished yet, after 60min. It is the first time, that is seems to be hang. Yesterday it was moving in 20sec.

What should I do?

cu denny

[To-M]

I would check the RAID health on the other node and see if there is any bad or dropped packets on both systems from the Console Screen enter CTRL ALT N then select the NIC and then select Info.

[rbg]

hi,

sorry for the very late answer :-/

I was able to reboot the node and than starting the cluster.

[rbg]

hi,

just a quick reply:

I added the last night new Ping Nodes after stopping the cluster. I wasn't able to start the cluster after than again. I had to reboot both nodes to get the cluster running again. The log says "remote ... unstable ... killing ... ". This must be a problem since up12.

[To-M]

You might want to send us the log file so if we can see any network issues or you can check the ha-debug log in the log file. Also check the NICs from the Console in CTRL ALT N then select the NIC and select Info to see if there is any RX or TX packets that are bad or dropped.

[rbg]

hi @To-M

sorry, I was a few weeks on holidays ...

I have a similar problem again. I did nearly nothing and clicked only on some ISCSI targets to view the IQN and the Apache doesn't respond anymore. I get a blank page or connection timeout. This happens the second or third time. The last one on my node2 and today on node1 :-( . Last time, the console was extremely slow. In the actual case, I tried also restarting Apache, but after choosing "Apache control" and hitting enter, it throws me back to "Import data / Apache Control".

So, the only way seems to reboot :-/

I will send you the logs from the first case.