October 4, 2016 at 5:17 am
I have installed yesterday SQL server 2012 SP1 with update enabled for SP2 on a 4 node cluster.
Now i want to add a node so my instance can fail-over.
this is the command:
Setup.exe /qs /ACTION=AddNode /UpdateEnabled="True" /UpdateSource="\\89.99.96.53\softwarestore$\Applications\microsoft\Servers\SQL Server 2012\mergedupdate" /INSTANCENAME="SQL25" /SQLSVCACCOUNT="______" /SQLSVCPASSWORD="______" /AGTSVCACCOUNT="______" /AGTSVCPASSWORD="______" /IACCEPTSQLSERVERLICENSETERMS /PID=______ /FTSVCACCOUNT="______" /FTSVCPASSWORD="______" /INDICATEPROGRESS="True" /FAILOVERCLUSTERGROUP="99-001-003-S205" /FAILOVERCLUSTERNETWORKNAME="99-001-003-S205" /SkipRules=Cluster_VerifyForErrors ServerCoreBlockUnsupportedSxSCheck Cluster_IsWMIServiceOperational
I have installed SQL before, this is an extra instance.
I receive the error message:
Overall summary:
Final result: Failed: see details below
Exit code (Decimal): -2067919938
Exit facility code: 1214
Exit error code: 3006
Exit message: The SQL Server failover cluster services is not online, or the cluster cannot be accessed from one of its nodes. To continue, determine why the cluster is not online and rerun Setup. Do not rerun the rule because the rule cannot detect a cluster environment.
Start time: 2016-10-04 12:09:51
End time: 2016-10-04 12:10:06
Requested action: AddNode
I don't understand how that is possible, since it is not the first time i install a sql on that cluster. The commands are templated as i always keep everything for every instance.
October 4, 2016 at 11:32 am
Check the Failover Cluster Management tool. See if the server you're on is indeed part of the cluster. Also, are you sure you set up the SQL instance as a new cluster SQL instance instead of just a plain vanilla standalone SQL install?
October 5, 2016 at 12:32 am
Thanks for the reply. The server is a member of my cluster. The setup command, see action (addnode) is to add a fail-over after you installed the first with Setup.exe /qs /ACTION=InstallFailoverCluster . You install on the other nodes with: Setup.exe /qs /ACTION=AddNode .
I have installed many other instances this way, now for some dark reason i have this issue. I have done a cluster Validation without storage of course and all seems fine.
October 5, 2016 at 5:19 am
The server may be a member of the cluster, but if SQL isn't installed as a clustered instance, then it doesn't matter. The instance won't fail over and you can't add failover nodes to it.
Assuming you did install a clustered instance of SQL, I don't know what your issue is. But my gut is telling me SQL is installed as a standalone instance instead of a failover instance.
October 5, 2016 at 5:27 am
a standalone is installed with /action=install.
Going to reboot the servers and see what it does. You are totally right and i would say the same.
I keep this updated.
October 8, 2016 at 6:30 am
after reboot the issue is fixed. but that doesn't explain the issue.
October 10, 2016 at 4:32 am
Let me ask you this. Did you reboot the server with SQL on it after installing SQL Server at all?
SQL Server services require a reboot after the initial install and after the service pack installs. If you don't reboot, then some background stuff for install doesn't complete and SQL Server will not function properly.
October 11, 2016 at 12:58 am
Here is the complete flow:
- Tried to install a new SQL instance on one of the nodes (new SQL failover cluster instalaltion) : failed with the cluster error.
- Tried to install a new SQL instance on the node where the cluster service is active (new SQL failover cluster instalaltion) : OK.
- setup returned error code for pending reboot. I always check the end result, normally 0 but another number when install is Ok but requires a reboot.
- Did no reboot and took the next node to install copy of that instance (add node). failed with the cluster error.
So your note about pending restart is a possibility since normally when you install an extra SQL instance, a reboot is not required. So it is possible that the one required a reboot before i tried to install SQL.
in the weekend:
restarted all 4 nodes
installed the instance on the other 3.
installed after that 2 more new instances on all 4.
restarted all 4 again to be sure.
October 11, 2016 at 3:57 am
joeridecock (10/11/2016)
So your note about pending restart is a possibility since normally when you install an extra SQL instance, a reboot is not required.
I have to call you out on this one. I just finished installing a SQL 2012 cluster set up multiple times this year. Each and every time, SQL required a reboot after the install.
Now it won't tell you it requires a reboot, but you literally cannot do anything else (install new service packs, setup certain features) until you do.
So the lack of reboot is your issue. Add that to your list of items to do and you should be fine (unless something else goes wrong).
October 11, 2016 at 7:58 am
I have modfied the procedure and added reboot server(s) at the end.
This will also be discussed in our next DBA meeting.
Thank you for your support, is much appreciated.
October 11, 2016 at 11:30 am
Glad I could help.
Viewing 11 posts - 1 through 10 (of 10 total)
You must be logged in to reply to this topic. Login to reply