We had an interesting one come in today from a customer. They reported that on one of their exchange servers (2013 CU 8) the replication service had been constantly restarting for around a week.

We hopped on and immediately took a look at the event logs, we found the below:

(Event 2070 The Microsoft Exchange Replication Service encountered an error while inspecting the logs and database for ….  on startup)

Event 2070

Additionally in the MailboxDatabaseFailureItems log we found:

(Event 1, MailboxDatabaseFailureItems)

 

Event 1

We had tried this fix: https://support.microsoft.com/en-us/kb/3003580 but it didn’t solve our issue.

Going on a hunch we suspended the database the logs were mentioning and voila the service stabilised!

We then removed the copy (and the files) of the dodgy db and re-added it, all working as expected!

 

About the author