We had a client report a weird experience Monday and we wanted to share it for the benefit of others.
They'd just migrated a few metric tons of calendar data into Exchange 2003 sp2.
The data looks fine, but after deleting a resource account that had been registered to the Auto Accept Agent (see the Deployment Guide) they discovered their server CPU usage pegged at 100% (not even a calendar migration usually does this!) and various other wrath-of-your-favorite-deity-type plagues on their Exchange server.
Turns out this is a well-known problem documented in the KnowledgeBase article An Exchange Server 2003 SP2 server becomes unresponsive after you delete a mailbox on which you registered the Auto Accept Agent event sink.
The best way to deal with the problem is to avoid it. (Patient: It hurts when I do this. Doctor: Well don't do that.)
But if you're already in the soup, fixing this problem if it happens to you involves using MfcMAPI. More than that you're going to need to know which registered resource was removed so you can make things right. Since there are no logs to guide you if this should suddenly happen, we recommend the Microsoft Exchange Team Blog article How do you know which mailboxes are registered with the Auto Accept Agent?
Keep track of your resources!
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment