SQL Server 2012 SP4 Patch Failed on 1 node in AG Cluster.

Zack
Zack used Ask the Experts™
on
Hi EE,

An upgrade to SQL server 2012 SP4 has failed, context:

I'm patching a 2 node AG cluster the first patch has been applied to secondary node successfully:

I then failed over the secondary node.  Then I began patching primary node:

The primary node patch failed I have got no idea why it failed and the SQL instance is no longer accessible.

I restored from a VM snapshot, tried it again and got the same error.  

Looking at these error log files I can't make sense of what went wrong.

Any assistance is appreciated.

Thank you
ERRORLOG.txt
ERRORLOG1.txt
ERRORLOG2.txt
ERRORLOG3.txt
ERRORLOG4.txt
ERRORLOG5.txt
ERRORLOG6.txt
Comment
Watch Question

Do more with

Expert Office
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
Raja Jegan RSQL Server DBA & Architect, EE Solution Guide
Awarded 2009
Distinguished Expert 2018

Commented:
>> The primary node patch failed I have got no idea why it failed and the SQL instance is no longer accessible.

Have you tried restarting the server to see whether it helped or not..

>> I restored from a VM snapshot, tried it again and got the same error.

Instead, can you try re configuring that node and make it as a valid secondary node..
ZackGeneral IT Goto Guy

Author

Commented:
Hi Raja,

Have you tried restarting the server to see whether it helped or not..

- I did the SQL server was unresponsive according to the data logs the master database was corrupt.

Instead, can you try reconfiguring that node and make it as a valid secondary node.

- I made that node the secondary node before I commenced the update.

Any other ideas?

Thank you.
Raja Jegan RSQL Server DBA & Architect, EE Solution Guide
Awarded 2009
Distinguished Expert 2018

Commented:
>> I did the SQL server was unresponsive according to the data logs the master database was corrupt

Have you tried fixing corrupted master database by rebuilding it or restoring from a backup..

Instead, can you try reconfiguring that node and make it as a valid secondary node.

I mentioned to rebuild the SQL Server instance completely and add it as a new node to your AG group..
Ensure you’re charging the right price for your IT

Do you wonder if your IT business is truly profitable or if you should raise your prices? Learn how to calculate your overhead burden using our free interactive tool and use it to determine the right price for your IT services. Start calculating Now!

ZackGeneral IT Goto Guy

Author

Commented:
Hi Raja,

I did try fixing the corrupted master database by restoring from the backup, but the SQL server was still showing as the same version.

Second query:  I mentioned to rebuild the SQL Server instance completely and add it as a new node to your AG group..

I could try de-attach the node completely:

https://www.mssqltips.com/sqlservertip/4231/how-to-remove-node-from-a-sql-server-failover-cluster/

Then try patch the server.

The question I would have is would the cluster work with only 1 node on it.

Just to reiterate the cluster consists of 2 nodes and 1 witness server.

Thank you.
ZackGeneral IT Goto Guy

Author

Commented:
Hi Raja.

Correcting technical details

I did try fixing the corrupted master database by restoring from the backup, but the SQL server was still showing as the same version.

Second query:  I mentioned to rebuild the SQL Server instance entirely and add it as a new node to your AG group.

I could try de-attach the node completely:

https://www.mssqltips.com/sqlservertip/4619/removing-a-secondary-replica-from-a-sql-server-alwayson-availability-group/

Then try patch the server.

The question I have is would the AG work with only one node on it.

Just to reiterate the cluster consists of 2 nodes and 1 witness server.

Thank you.
SQL Server DBA & Architect, EE Solution Guide
Awarded 2009
Distinguished Expert 2018
Commented:
>> The question I have is would the AG work with only one node on it.

Yes, technically out of the 2 nodes, if Primary goes down due to any reason, then secondary will take over and act as Primary..
But the problem working with only one node is if that node gets any failure, then your applications can't connect at all as no nodes are active..

So, you have time now to fix the secondary node completely and add it correctly..
Hope this clarifies..
ZackGeneral IT Goto Guy

Author

Commented:
Thank you Raja my apologies for the delay in the acceptance of the answer.
Raja Jegan RSQL Server DBA & Architect, EE Solution Guide
Awarded 2009
Distinguished Expert 2018

Commented:
No issues, glad to assist..

Do more with

Expert Office
Submit tech questions to Ask the Experts™ at any time to receive solutions, advice, and new ideas from leading industry professionals.

Start 7-Day Free Trial