Symptoms
Consider the following scenario:
-
You create a database availability group (DAG) that contains two or more member servers in a Microsoft Exchange Server 2010 environment.
-
You configure the replication service between the member servers of the DAG.
-
Some users perform a Multi-Mailbox Search function against the DAG.
In this scenario, you may encounter mailbox database replication latency in the DAG.
Note When this issue occurs, if the CheckDatabaseRedundancy.ps1 script is running to monitor the redundancy of replicated mailbox databases, you will receive an event that only a single healthy copy of a replicated database exists.
Cause
This issue occurs because the Multi-Mailbox Search function is running during the database replication process. This behavior causes a heavy disk I/O load. Therefore, the latency of replication increases.
Resolution
To resolve this issue, install the following update rollup:
2645995 Description of Update Rollup 1 for Exchange Server 2010 Service Pack 2
Status
Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.
More Information
For more information about the Multi-Mailbox Search function, visit the following Microsoft website:
General information about the Multi-Mailbox Search functionFor more information about how to manage DAGs, visit the following Microsoft website:
How to manage DAGs in an Exchange Server 2010 environmentFor more information about high availability and site resilience, visit the following Microsoft website:
General information about high availability and site resilienceFor more information about how to monitor high availability and site resilience, visit the following Microsoft website:
How to monitor high availability and site resilience in an Exchange Server 2010 environment
