November 20, 2017 at 4:31 pm
What is the best way to manage SQL Server rebooting after monthly security patching that is in a 3 node cluster configuration with AG groups in sync and async replicas ?
CAn not reboot manually as it is not an option.
I now have to manually open the config tool to start all services instances manually.
All SQL instances are in stopped mode.
Thank you for the hint.
I saw some errors about multi subnet ip address failure and that it can be ignored.
However , Is there a recommended way to do reboots for this configuration ?
Like do we pause the Synching ?
Then reboot the Secondary and then failover the primary to secondary and make it primary and then the new secondary to reboot etc. ?
November 21, 2017 at 1:25 am
dvprao - Monday, November 20, 2017 4:31 PMWhat is the best way to manage SQL Server rebooting after monthly security patching that is in a 3 node cluster configuration with AG groups in sync and async replicas ?
CAn not reboot manually as it is not an option.
I now have to manually open the config tool to start all services instances manually.
All SQL instances are in stopped mode.
What errors do you get in the eventlog and errorlog?
If the services are set to automatic then after patching and rebooting to finish applying the patches they should come online automatically, so something is not auto starting the services. My guess would be that all 3 nodes are rebooting at the same time and thus loosing quorum and the cluster roles are shutting down, but need to see what the error logs say.
November 22, 2017 at 11:31 am
Based on what you mentioned , I rebooted one by one . Even then the issue persists.
The SQL instances are in stopped state.
I see in the logs :
That since the Database is not started because it is part of the AG Group and so the AG will start it .
I am not seeing anything else.
I am really surprised that no body else is seeing this as an issue. And even HAG resource books do not talk about managing the patching process/UPdate process in detail.
Thanks for any help provided.
November 28, 2017 at 4:22 am
anthony.green - Tuesday, November 21, 2017 1:25 AMdvprao - Monday, November 20, 2017 4:31 PMWhat is the best way to manage SQL Server rebooting after monthly security patching that is in a 3 node cluster configuration with AG groups in sync and async replicas ?
CAn not reboot manually as it is not an option.
I now have to manually open the config tool to start all services instances manually.
All SQL instances are in stopped mode.What errors do you get in the eventlog and errorlog?
If the services are set to automatic then after patching and rebooting to finish applying the patches they should come online automatically, so something is not auto starting the services. My guess would be that all 3 nodes are rebooting at the same time and thus loosing quorum and the cluster roles are shutting down, but need to see what the error logs say.
wont stop the instances starting, only the AG cluster role. Unless of course they're FCIs.
What errors are seen in the logs would be a good start
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
November 28, 2017 at 9:44 am
Based on past experience with A SQL instance on a VM, I changed the Startup type to "Automatic delayed " from "Automatic" and rebooted the 2 nodes one by one.
That seemed to have done the trick. I will be rebooting the primary node as well during the weekend and will post the result.
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply