SQL Server 2017 Cumulative Update process for Clustered SQL Instance

Question

SQL Server 2017 Cumulative Update process for Clustered SQL Instance

Michael Lukaszewicz 1

I have the following clustered environment:
SQLA node
SQLB node

I have the following SQL instances:
SQL1 (normally active on SQLA)
SQL2 (normally active on SQLB)

The databases for SQL1 are stored on a CSV disk D: which is connected to SQLA when the SQL1 role is on SQLA owner node in the cluster.
The databases for SQL2 are stored on a CSV disk E: which is connected to SQLB when the SQL2 role is on SQLB owner node in the cluster.
SQLB node does not have CSV disk D: connected since SQL1 instance is not active on SQLA until I failover SQL1 to SQLB.
Vice versa for the CSV drive hosting DBs for SQL2.

My question is:
Can I run the SQL Server 2017 cumulative update on SQLB for both SQL1\SQL2 instances after failing over SQL2 instance to SQLA node?
And then once SQLB is finished patching, then failover SQL1 and SQL2 instances to SQLB, and do the updates on SQLA for both instances?
Would this work even though the CSV disk that contains the DBs is no longer connected to the maintenance node after failing over the instances?
Does the update only update the SQL server instance that is on the local C: drive of that node?

Also another question:
When I failover all SQL instances and put the node into pause, I get the following error in the SQL server 2017 update rule check:
"Not clustered or the cluster service is up and online", it fails even when I turn off the Cluster service.
Is it safe to use the following command line to overcome this or is there a reason why it still shows up as an error:
<> Path of patch executable>:\Setup /SkipRules=Cluster_IsOnlineIfClustered /action=patch

CathyJi-MSFT 22,396 Reputation points Microsoft External Staff

2022-03-14T01:41:50.8+00:00

Hi @Michael Lukaszewicz ,

Any update for this thread? Did the reply could help you?

1 answer

Your answer

CathyJi-MSFT 22,396 Reputation points Microsoft External Staff

2022-03-14T01:41:50.8+00:00

Hi @Michael Lukaszewicz ,

Any update for this thread? Did the reply could help you?

Answer 1

CathyJi-MSFT 22,396 Microsoft External Staff

Hi @Michael Lukaszewicz ,

> Can I run the SQL Server 2017 cumulative update on SQLB for both SQL1\SQL2 instances after failing over SQL2 instance to SQLA node?

Yes, you can. When you failover SQLFCI SQL2, then SQL2 on server A node is passive node, then SQL 1 and SQL 2 on server A node are both passive nodes. Then you upgrade both FCI SQL 1 and FCI SQL 2.

To upgrade a SQL Server failover cluster instance, use SQL Server setup to upgrade each node participating in the failover cluster instance, one at a time, starting with the passive nodes. Please refer to MS document Upgrade a failover cluster instance to get detail upgrade steps.

> And then once SQLB is finished patching, then failover SQL1 and SQL2 instances to SQLB, and do the updates on SQLA for both instances?

Yes, you are right. Perform a rolling upgrade in SQL failover cluster instance.

> "Not clustered or the cluster service is up and online", it fails even when I turn off the Cluster service.

To control the failover behavior of cluster nodes during the upgrade process, run the upgrade operation at the command prompt and use the /FAILOVERCLUSTERROLLOWNERSHIP parameter. For more information, see Install SQL Server from the Command Prompt.

Suggest you upgrading SQL FCI from SQL server upgrade setup UI.

If the answer is helpful, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment".

Michael Lukaszewicz 1 Reputation point

2022-03-10T15:14:22.307+00:00

When you failover SQLFCI SQL2, then SQL2 on server A node is passive node, then SQL 1 and SQL 2 on server A node are both passive nodes. Then you upgrade both FCI SQL 1 and FCI SQL 2.

Do you mean after failing over SQL2 to server A then SQL1 and 2 on server B are both passive?

I noticed that when I do NOT put the node that I want to upgrade into Pause mode + failover sql instance in the failover cluster manager, then the Cluster_IsOnlineIfClustered rule passes in the setup

But then I get the warning:
Rule Description:
"Checks if the local computer is an owner of an online cluster group for a failover cluster instance that contains the SQL Server service, Analysis Services service or a generic service."
Message/Corrective Action:
"The local computer is an owner of a cluster group that contains the SQL Server service, Analysis Services service or a generic service for a failover cluster instance. If you continue, the SQL Server instances may be taken offline causing downtime while the patch is being applied."

Does the SQL server update setup automatically failover any SQL instances before beginning the upgrade process?

Also how long does the cumulative update process take?
CathyJi-MSFT 22,396 Reputation points Microsoft External Staff

2022-03-11T06:54:23.213+00:00

Hi @Michael Lukaszewicz ,

> Do you mean after failing over SQL2 to server A then SQL1 and 2 on server B are both passive?

Yes, you are right.

>Does the SQL server update setup automatically failover any SQL instances before beginning the upgrade process?

How did you upgrade SQL server instance ? SQL upgrade UI( by clicking SQL setup file) or command? Did you check all information that I posted in my first reply? If not, please read again. If you using UI to upgrade SQL, please manually failover the SQL FCI before upgrade. In addition, suggest you reading the MS document Upgrade a failover cluster instance that I offered in my first reply.

>Also how long does the cumulative update process take?

It's hard to say. But SQL FCI downtime limited to a single manual failover (or two manual failovers if failing back to the original primary).
Michael Lukaszewicz 1 Reputation point

2022-03-14T13:32:11.32+00:00

Thanks for your information.

I found out how to get all the rules to pass for the SQL upgrade UI.

1) Failover the SQL FCI to the active node (node 1)
2) Do NOT put the passive node (node 2) into Pause in failover cluster manager, or else you will fail the "Not clustered or the cluster service is up and online" rule.
Instead go to the SQL instance properties and make sure the node you are doing updates on does not allow the SQL instances to fail back (not an allowed node).
3) Do updates on passive node (node 2)
4) Reboot passive node (node 2)
5) Change properties of SQL instances to allow passive node (node 2) for failover.
6) Failover instances to node 2.
7) Repeat upgrade process on node 1.
CathyJi-MSFT 22,396 Reputation points Microsoft External Staff

2022-03-15T07:49:54.74+00:00

Hi @Michael Lukaszewicz ,

Thanks for your reply. I am so glad to hear that your issue is resolved.

By the way, If the response helped, do "Accept Answer". By doing so, it will benefit all community members who are having this similar issue. Your contribution is highly appreciated.

Share via

SQL Server 2017 Cumulative Update process for Clustered SQL Instance

1 answer

Your answer