November 13, 2014 at 8:39 am
Hi,
We are running SQL Server 2008 R2 on a two-node Windows Server 2012 R2 failover cluster.
During unexpected network outages, we noticed that the failover cluster attempted to fail over to the other server only to fail (since the network was down). Once the network is back up, is there a way for the failover cluster to automatically come back up?
Any help is greatly appreciated!
November 13, 2014 at 8:41 am
dajonx (11/13/2014)
Hi,We are running SQL Server 2008 R2 on a two-node Windows Server 2012 R2 failover cluster.
During unexpected network outages, we noticed that the failover cluster attempted to fail over to the other server only to fail (since the network was down). Once the network is back up, is there a way for the failover cluster to automatically come back up?
Any help is greatly appreciated!
why does the instance not come online on the partner, have you checked the logs to find out why this is?
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
November 13, 2014 at 8:52 am
It was during a switch firmware update which required a reboot thus the entire network was down. Both nodes were entirely down during the reboot so the failover cluster logs has a lot of "cluster network is down", "health check for IP interface failed", and "cluster resource 'SQL IP Address 1 of type 'IP Address' in clustered roles 'SQL Server' failed" messages.
November 13, 2014 at 9:01 am
when the cluster comes back online the first available node will grab the resources.
If you know you're gioing to have an outage such as that it would be more prudent to take control and stop the cluster group, then bring it back online rather than just let if fall over on its backside.
Efficient change control would promote this too
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
November 13, 2014 at 9:17 am
I wasn't informed that there would be a network outage. I would've failed over the databases to a different site in that case.
So if the cluster is down, it won't come back up automatically when the network is back up? I know the cluster will come back up via human intervention, but I was just curious is the cluster is "smart" enough to come back up on its own.
November 13, 2014 at 2:05 pm
it'll try so many times then comply with the default policy which leaves the resources offline rather than just constantly attempting restarts. After the default of one hour restarts will be attempted
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply