Time-out issue Always on SQL 2012 Cluster

Question

Currently I'm working on a large deployment of a SharePoint 2013 environment (stretched farm over 2 DC's). We are using a SQL 2012 alwayson multisubnet cluster (each DC has 1 SQL node). During the installation of SharePoint we encountered several connection
time-out errors from the SQL environment. (Exception calling "Open" with "0" argument(s): "Connection Timeout Expired.  The timeout period elapsed during the post-login
phase.  The connection could have timed out while waiting for server to complete the login process and respond;)
Steps we follow for SharePoint installation are:
Steps to configure SharePoint 2013 on an AlwaysOn Availability group:
1>Create SharePoint databases (with the PowerShell script)
  Configure a SQL alias for a consistent SQL instance name.
 Create farm on SQL instance/replica 1 (one of the SQL nodes), creating all databases needed (service-apps & web-apps). -->
we are connected directly to 1 SQL node and not connecting to the cluster. Installation server is in same DC as the SQL node.
   Stop SharePoint so databases remain static during migration to an AlwaysOn cluster.
2>Move database to AlwaysOn high-availability group.
Restore all the DBs onto SQL replica 2 (with NORECOVERY).
Create AlwaysOn availability group, or use existing
Join the replica 2 databases to availability group.
Create listener.
3>Migrate SharePoint onto AlwaysOn on cluster
Make all SharePoint DB MultiSubnetFailover aware with PowerShell Cmdlet
Reconfigure SQL alias for new listener
àat this point the SharePoint farm is connecting to the SQL cluster (listener of AlwaysOn availability group of the SQL Instance), but we never reached this point so far.
 Restart SharePoint services with updated alias.
Event errors and SQL log errors that I found:
Date,Source,Severity,Message,Category,Event,User,Computer
09/18/2014 07:34:59,Microsoft-Windows-FailoverClustering,Error,Cluster network name resource 'SPAG_CU8000001105' failed registration of one or more associated DNS name(s) for the following reason:<nl/>DNS bad key.<nl/>.<nl/><nl/>Ensure
that the network adapters associated with dependent IP address resources are configured with at least one accessible DNS server.,(19),1196,NT AUTHORITY\SYSTEM,CU8000000015
09/18/2014 07:34:59,MSSQL$SPAG,Information,The Service Broker endpoint is in disabled or stopped state.,(2),9666,,CU8000000015
09/18/2014 07:34:52,Microsoft-Windows-FailoverClustering,Error,Cluster network name resource 'SPWMAG_CU8000002105' failed registration of one or more associated DNS name(s) for the following reason:<nl/>DNS bad key.<nl/>.<nl/><nl/>Ensure
that the network adapters associated with dependent IP address resources are configured with at least one accessible DNS server.,(19),1196,NT AUTHORITY\SYSTEM,CU8000000015
09/18/2014 07:34:52,MSSQL$SPWMAG,Information,The Service Broker endpoint is in disabled or stopped state.,(2),9666,,CU8000000015
09/18/2014 07:34:47,Service Control Manager,Information,The WMI Performance Adapter service entered the running state.,(0),7036,,CU8000000015
09/18/2014 07:32:41,Microsoft-Windows-FailoverClustering,Error,Cluster network name resource 'SPSDAG_CU8000003105' failed registration of one or more associated DNS name(s) for the following reason:<nl/>DNS bad key.<nl/>.<nl/><nl/>Ensure
that the network adapters associated with dependent IP address resources are configured with at least one accessible DNS server.,(19),1196,NT AUTHORITY\SYSTEM,CU8000000015
09/18/2014 07:32:41,MSSQL$SPSDAG,Information,The Service Broker endpoint is in disabled or stopped state.,(2),9666,,CU8000000015
09/18/2014 07:31:09,PowerShell,Information,Engine state is changed from Available to Stopped. <nl/><nl/>Details: <nl/> NewEngineState=Stopped<nl/> PreviousEngineState=Available<nl/><nl/> SequenceNumber=61464<nl/><nl/> HostName=OpsMgr
PowerShell Host<nl/> HostVersion=7.0.5000.0<nl/> HostId=32012185-8d9a-41c2-be56-91929c02f1e8<nl/> EngineVersion=4.0<nl/> RunspaceId=af176e01-185d-4574-ab9b-0fd745178d29<nl/> PipelineId=<nl/> CommandName=<nl/> CommandType=<nl/> ScriptName=<nl/> CommandPath=<nl/> CommandLine=,(4),403,,CU8000000015
We are not allow to update/write in the DNS for the multisubnet cluster IP registration, so I think that explains the "failed registration
" error. But can this explains our time-out errors during the SharePoint installation? For the installation we are connection directly to 1 SQL node and not to the SQL
cluster.
Any help is appreciated!
Ronald Bruinsma - Independent SharePoint Consultant - iDocs.info - The Netherlands -- Please don't forget to propose this post as an answer or mark it as helpful if it did help you. Thanks.

Answer 1

Don't just change the connection timeout on your SharePoint farm. This will just hide the real issue. From your error message, it seems that the DNS record for the Availability Group listener name is not being written. Talk to your AD administrators to validate
if dynamic DNS registration is configured for the DNS servers. If it is, AD will create the DNS entry of the virtual computer object for the Availability Group listener name. The WSFC should also have Create Computer Objects permission in the AD Organizational
Unit where your Availability Group listener name will be created.
On a side note, make sure that you configure a separate SharePoint 2013 farm for your DR environment that will use the content databases joined in the Availability Group. For a stretched farm deployments, latency should be less than 1ms one way as
per this article. And while you can add your admin content and config databases on the Availability Group, asynchronous commit is
not supported.
Edwin Sarmiento SQL Server MVP | Microsoft Certified Master
Blog |
Twitter | LinkedIn
SQL Server High Availability and Disaster Recover Deep Dive Course