StoreVirtual Storage
cancel
Showing results for 
Search instead for 
Did you mean: 

SANiQ 9 upgrade - connection to node stales

Fred Blum
Regular Advisor

SANiQ 9 upgrade - connection to node stales


The overall upgrade process pauzes at 83%. Hyper-V and SAN Node 1 were updated fine. Now it says copying installation files to machine NodeName2 (0%). It has been saying that for over two hours.

There is still quorum (2 out of 3) and storage access.

There is a F11 option to cancel this step. What to do?
3 REPLIES
Fred Blum
Regular Advisor

Re: SANiQ 9 upgrade - connection to node stales


When I look at the details page of both the SAN nodes and the Hyper-V FOM I see software version 9.0.00.3561.0.

But CMC 9 was unable to connect and login to node2, failing at 83% of the upgrade process and for that reason failed to complete copying installation files to node2 (0%). Furthermore what steps were still to be performed as Node1 still had a timer and not a completion tick as with the Hyper-V FOM. So basically san node1 is a question mark also.

I can still ping node2 but it doesnot allow login with a read timeout error.

My volumes are still empty as this a pre-production test setup. Can I drop nRaid, remove node2 from the cluster and management group and insert the SANiQ 9 restore CD into the node? Reconfigure the node and re-add? And than do the same to node1?

Fred Blum
Regular Advisor

Re: SANiQ 9 upgrade - connection to node stales


I tried removing the node from the management group, but I cannot login to remove it from the management group. It's status is still up and seen by CMC.


Fred Blum
Regular Advisor

Re: SANiQ 9 upgrade - connection to node stales

I rebooted the unresponsive node. After reboot I was able to login again and everything appears status normal. I reran the upgrade and I was notified that patch 10091-00 still needed to be installed.

I believe this patch for the raid controller to be the culprit. After installing on node 1 the update windows stalls again with the message: "waiting for node to reconnect after reboot". Which it did not while node1 was up and running happily after the reboot. A new reboot of node1 did the installation wizard fail.
Restarting the update wizard shows node1 and FOM ticked for this patch.
After applying the patch to node2 the connection was restored and the upgrade wizard finished with installation status succesfull.
When I looked at the node2 there was however a alarm that the storage of this node was unformated!

After a new reboot the alarm was gone and the status is normal.