For several days now we've been seeing corrupted search catalogs on a number of mailbox databases on our Exchange 2010 8 member DAG. Each time this happens I'm seeing the following events in the Application Logs of the affected servers:
MSExchange Search Indexer : Event ID 118
Exchange Search Indexer has temporarily disabled indexing of the mailbox database EX24R12 (GUID = 6644034b-2058-478e-bfde-442423cce41b) due to an error (Microsoft.Mapi.MapiExceptionMdbOffline: MapiExceptionMdbOffline: Unable to read events. (hr=0x80004005, ec=1142)
- and -
ExchangeStoreDB : Event ID 123
At '7/9/2010 9:54:22 AM' the Microsoft Exchange Information Store Database 'EX12S12' copy on this server experienced a corrupted search catalog. Consult the event log on the server for other "ExchangeStoreDb" and "MSExchange Search Indexer" events for more specific information about the failure. Reseeding the catalog is recommended via the 'Update-MailboxDatabaseCopy' task.
Each time this happens I'm able to resolve the issue by running "Update-MailboxDatabaseCopy MDB\Server -CatalogOnly", however this is continuing to happen and I can find to root cause. Today it happened nearly immediately after we performed a database switchover of a number of databases, however not all databases (and not all mailbox servers) involved in the switchover experienced the corruption. Additionally, the databases themselves all report as health, and I'm not seeing any disk or I/O errors in the System Logs of the affected mailbox servers. This doesn't appear to be any kind of physical disk corruption.
This has happened with enough frequency (and has occurred without having been preceded by a database switchover event) that while all databases are reporting their ContentIndexState as healthy at present (after running the Update-MailboxDatabaseCopy -CatalogOnly cmdlet), I am reasonably sure that this condition will occur once again. What can I do to determine the root cause of this?