Hi Guys,

We currently have Exchange 2013 virtualised on Windows Server 2012 hyper-V. We have two sites and two DAGS. We have 8 Exchange 2013 servers virtualised across two DAGS across two sites.

The issue that I'm having is when I failover a MailboxDatabase across to one specific server, the contentIndexState is in a failed state. However when it fails back to the original server the contentindexstate is healthy.

Checking the event logs under exchange High Availability I see the following warning:

Database **** copy on server **** is to far behind. Copy queue is: 10824 and replay queue is: 0

Error I see is:

Database **** best copy could not be found.

What I did was suspend the copy on the bad server via powershell using the following command:

Suspend-MailboxDatabaseCopy ****

Update-MailboxDatabaseCopy **** -SourceServer ***** -DeleteExistingFiles:$True

Resume-MailboxDatabaseCopy *****

This completes successfully, does take some time however but it completes successfully.

When I try to fail it over I still get the same warnings and errors.

Please help as I'm stumped on this one.

Whenever we perform database swtichover or server failover, we need to ensure that all database copies are health.  if the content index failed, then we need to fix that using below mentioned command.  It will reseed only content index.

https://technet.microsoft.com/en-us/library/ee633475(v=exchg.150).aspx

Update-MailboxDatabaseCopy -Identity DB1\MBX1 -CatalogOnly

Hi,

Did this issue affect only one Exchange server?

You mentioned that when I failover a MailboxDatabase across to one specific server, the contentIndexState is in a failed state. does this specific server locate in the same site where the active database copy resides? Or is this a different site?

And from the warning message, the copy queue length is high. Please use the Test-ReplicationHealth command to check result.

After you reseed database copy, please check if the copy queue length goes down to 0.

Best regards,

If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

Whenever we perform database swtichover or server failover, we need to ensure that all database copies are health.  if the content index failed, then we need to fix that using below mentioned command.  It will reseed only content index.

https://technet.microsoft.com/en-us/library/ee633475(v=exchg.150).aspx

Update-MailboxDatabaseCopy -Identity DB1\MBX1 -CatalogOnly

• Proposed as answer by Rajkumar Nagaraj Thursday, February 12, 2015 1:57 PM

Hi Belinda,

This issue occurs on the Exchange Server in the same site i.e.

DAG-01 will have the following Exchange Servers: EXA01 (Site A), EXA02 (Site A), EXB01 (Site B), EXB02 (Site B)

DAG-02 will have the following Exchange Servers:  EXA03 (Site A), EXA04 (Site A),EXB03 (Site B),EXB04 (Site B)

So our issue occurs in Site A, when failing from EXA01 to EXA02. This puts the content index state into failed across DAG-01. However when the database fails back from EXA02 to EXA01, the content index state on all servers in DAG-01 is healthy. This occurs on all our databases in DAG-01 when failing from EXA01 to EXA02. We have currently 22 databases in each DAG.

Looking forward to your feedback

Hi Belinda,

Tested the replication health and its passed. No errors popped up.

Hi

I tried to update the catalog as suggested. The update was successful, however when failing over it still sets the content index state to failed. Failing back to source server the content index state is healthy again.

Hi All,

Thanks for all your help. Issue is now resolved.

For anyone else having this issue, below is the fix. We raised a case with Microsoft and after two days of capturing logs and analysis we found that the server having the issue had the following set to true DatabaseCopyActivationDisabledAndMoveNow

To fix this we ran the following command 

SetSet-MailboxServer ServerName -DatabaseCopyActivationDisabledAndMoveNow  $false

The