This is a problem that occurred from time to time to a customer with an Exchange 2013 infrastructure consisting of 1 CAS and 2 Mailbox servers. Sometimes, about 1 time per month but sometimes even more frequently, Clients whose mailbox was present on one of the mailbox servers got disconnected. After a server reboot the clients get reconnected.
Checking in event viewer around the time of disconnections this error was present with Event ID 7031:
“The Microsoft Exchange RPC Client Access service terminated unexpectedly. It has done this 1 time(s). The following corrective action will be taken in 5000 milliseconds: Restart the service.”
Looking’ on the internet I found this technet article that might solve the problem:
FIRST SOLUTION:
The steps to correct the problem are these:
- Open ADSIEdit
- Connect to the Configuration container
- Expand Services
- Expand Microsoft Exchange
- Expand your Organization name
- Expand Administrative Groups
- Highlight Databases
- In the right hand pane you will see a list of databases.
- Right click on the database object and chose Properties
- Scroll down to msExchHomePublicMDB
- If you see reference to the old database delete the value
- Check the rest of the databases to make sure that they are not populated in the same way.
There are also other solutions which, however,, my personal opinion, seem to me little reasonable and I found all of them in unofficial pages, whatever I report them for completeness.
SECOND SOLUTION (mah…. I do not understand the meaning if it's the service that crashes):
- On windows 2008 à Click on start à Run or on windows 2012 open command prompt with run as administrator
- Type regedit and hit enter
-
Go to the following location
HKEY_LOCAL_MACHINESoftwarePoliciesMicrosoftWindows NT
- Right Click and select new Dword Value
- Type MinimumConnectionTimeout with the decimal value 120
- Close regedit and restart the server.
THIRD SOLUTION (I have no idea why it should work):
- Change the log on account of MS Exchange RPC Client Access service from “system account” to “Netwotk Service” account.
FOURTH SOLUTION (unwise):
- Stop and Start the Microsoft Exchange Health Manager service.
UPDATE
After more than a month without disruptions I can say fairly certain that the cause of disconnections is the one described in the first solution.