Both DAG indexes for a database failed

Question

Hi all, looking for a suggestion on how I might be able to resolve this indexing issue.
We've got a two node Exchange 2013 DAG. On the primary copy of the database, the index keeps failing and restarting its crawl. On the secondary copy, the node is showing that the index is Failed and Suspended.
We'd stopped the search services and blown away the index folder for the DB on the mounted copy and let the crawl resume, but after a while we'll see it fail and then resume still. Looking at the High Availability > Seeding logs, event 1112 is getting
logged every few minutes or more:
Automatic Reseed Manager has detected a content indexing error for database 'DB06'. The catalog may be corrupted and will be reseeded automatically. contentIndexError: The content index of old schema version 1. Newer version 12 is available.
Indexes on all other databases are healthy.
So I need to figure out the best possible way to recover the indexes for this database. I can't fail over to the secondary node with a failed index, but it looks like I could force it if I have to. I haven't yet tried deleting the index on the secondary
copy, and my suspicion is trying to reseed off the mounted copy will just be bad for the secondary copy. And the schema error concerns me too.
Any suggestions on how best to resolve this?

Answer 1

The only error I'm seeing is event 1009 in the Application log:
Log Name:      Application
Source:        MSExchangeFastSearch
Date:          5/16/2014 1:11:54 PM
Event ID:      1009
Task Category: General
Level:         Warning
Keywords:      Classic
User:          N/A
Computer:      XMB2.domain.COM
Description:
The indexing of mailbox database DB06 encountered an unexpected exception. Error details: Microsoft.Exchange.Search.Core.Abstraction.OperationFailedException: The component operation has failed. ---> Microsoft.Exchange.Search.Core.Abstraction.OperationFailedException:
The component operation has failed. ---> Microsoft.Exchange.Search.Fast.FastDocumentTimeoutException: The Content Submission Service returned failure for the document: Document timeout expired
. ---> System.TimeoutException: Document timeout expired
   --- End of inner exception stack trace ---
   at Microsoft.Exchange.Search.Fast.FastFeeder.EndSubmitDocument(IAsyncResult asyncResult)
   at Microsoft.Exchange.Search.Mdb.CrawlerFeeder.DocumentCompleteCallback(IAsyncResult asyncResult)
   --- End of inner exception stack trace ---
   at Microsoft.Exchange.Search.Core.Common.Executable.EndExecute(IAsyncResult asyncResult)
   at Microsoft.Exchange.Search.Engine.SearchFeedingController.ExecuteComplete(IAsyncResult asyncResult)
   --- End of inner exception stack trace ---
   at Microsoft.Exchange.Search.Core.Common.Executable.EndExecute(IAsyncResult asyncResult)
   at Microsoft.Exchange.Search.Engine.SearchRootController.ExecuteComplete(IAsyncResult asyncResult)
Other than that, as it crawls the index before failing, I see it cycling through tmp files, two or three at a time. Each one changes out and it keeps going until it fails. I don't recall seeing this happen with indexes before, this is pretty new to me.