Link to home
Start Free TrialLog in
Avatar of Eldata
Eldata

asked on

Store.exe crashes, error MSExchangeIS 9659

And there we go again.
After a period of rest the problem is back worse than ever.
Every 10 minutes or so Store.exe crashes and recovers, leaving all Outlook clients frozen for several minutes and creating dumpfiles of over 600 MB(!) in size. These dumpfiles are in the C:/Windows/PCHealth file, and are my main concern because they fill the harddisk rapidly.
The errors in the eventlog are:
MSEXchangeIS 9659 (exception 0xC0000005)
Microsoft Exchange Serv 1000 (exsp.dll and msvcrt.dll)
Micosoft Exchange Serv 1000 (store.exe and msvcrt.dll)
Application error 1000 (Faulting application Store.exe, Faulting module msvcrt.dll)

In the past error 9659 pointed at a certain user, and we solved the problem by shutting down a pc that that particular user was logged on to.
Now the user is "Administrator" which is logged on nowhere.

Please help me: this problem is driving me NUTS!!

Oh, yes... this is Exchange 2003 SP2 on a Windows 2003 server.

And if anyone could tell me how to prevent those ridiculous dumpfiles in the PCHealth folder to begin with, I would be eternally grateful.
Avatar of KevinBall
KevinBall
Flag of United Kingdom of Great Britain and Northern Ireland image

How big is your priv database?  And what kind of disk volume is it running on?
Avatar of Eldata
Eldata

ASKER

The priv.edb is 6,7 GB., it is on a basic NTFS volume
ASKER CERTIFIED SOLUTION
Avatar of abhaigh
abhaigh
Flag of United Kingdom of Great Britain and Northern Ireland image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Disable all 3rd party including AV on the Exchange server and then monitor the the Msexchange IS service
I have seen this kind of an issue before and was caused by a third part monitoring software installed on the server
Avatar of Eldata

ASKER

@abhaigh: thanks, I found that one too. I completely emptied the Administrator mailbox, so there can't be any corrupt messages there. For integrity check the database needs to be offline, I will do this after hours.
@FearNoMore: I have had this problem in the past, and reïnstalled the TrendMicro AntiVirus back then.
That didn't solve the problem.
Is this Standard or Enterprise edition?
Well reinstalling the AV does not solve the problem....
This problem can be caused by the AV threads interfering with the IS
I had the issue issue with one of my clients and we had disabled all 3rd party services including AV and found that the IS was no longer crashing.
Give it a shot and try what happens
Avatar of Eldata

ASKER

@ KevinBall: Standard
Avatar of Eldata

ASKER

@FearNoMore: I forgot, after the troubles before I didn't even bother reinstall TrendMicro. So no AV software here
Eldata,

I would recommend reinstalling the windows service pack and then reinstall exchange service pack :) i know that sounds weird suggesting to reinstall but msvcrt.dll is a windows native dll library file. Either you can re-register it or simply reapply sp for windows and then for exchange as well. I dont think its a problem with information database files or their integrity.

M
Avatar of Eldata

ASKER

OK, last night I ran ISINTEG against the database. It did find errors and fixed them.
After a restart the first thing that happened was another dumpfile, so I was not happy.
The Outlook clients were horribly slow and kept freezing. We were going from bad to worse.
However: today all seems to be fine. No more dumpfiles, and Outlook for all users working like a charm.
New entry in the eventlog though: every couple of minutes warning 1173 MSEchangeIS Mailbox.
Will look into that now.

Anyway, I will leave the question open for a couple of days. If the problem doesn't return I will award the points to aghaigh.
This kb explains the 1173 error, and that is can be safely ignored if the error being referenced is 0x6bb

http://support.microsoft.com/kb/913126
Avatar of Eldata

ASKER

For the last couple of days the Exchange server has been behaving. It seems the Integrity Check did the trick, though I suspect emptying the Administrator mailbox may have had something to do with it as well (as "administrator" was in the error message as the user performing an invalid action).
Anyway, I will award the points because you definitely put me on the right track!