Exchange 2003 mailbox migrations crash target store.exe

Hi Everyone.

I am in the process of migrating many Exchange 2003 servers to some new ones (we will be upgrading to Exchange 2010 in about 18 months, but thats another story).

I have about 1000 users (site has 2000, company has about 5000), and most have moved across without issue. the old servers are at least three years old, some older, so  the new servers with new hardware are working well, except for this issue.

From the oldest server, I have a several mailboxes that will crash the Information store on the target server during attempted migration with the following errors...

event ID 9659
The Microsoft Exchange Information Store encountered an unexpected exception 0xC0000005 at address 8B318966 while processing a request for user {Username}.

event ID 1000
Faulting application store.exe, version 6.5.7653.21, stamp 474db761, faulting module unknown, version 0.0.0.0, stamp 00000000, debug? 0, fault address 0x8b318966.

event ID 1000
Faulting application store.exe, version 6.5.7653.21, stamp 474db761, faulting module msvcrt.dll, version 7.0.3790.3959, stamp 45d70b06, debug? 0, fault address 0x00026f76.

Restarting the information store will get the destination server going again, and apart from that it seems to perform well, and without issue.

I am running Exchange 2003 SP2 on Windows Server 2003 R2 sp2 (32 bit)
I have updated with KB943193, which seemed to describe this error, but it had no effect.

Most mailboxes have got across (I am down to the last 50 or so), the ones left are mainly booking resources, which implies that they are calendar bookings that are out of range (or something). usually retrying the migration will succeed (as if the migration process "jumps" the "poison" record once it encounters it).

I would persevere, however it is very ungainly to have the information store crash constantly on the target(s).

I get the same result using exmerge to migrate the mailboxes.

My Question - has anyone else had this issue? (one of the senior techs here at my work has, so I know it is not uncommon)

Does anyone know of a way I can migrate the mailboxes without this issue?.

I have already done many ESEUTIL defrags, and ISINTEG runs (till there were zero errors) but to no avail, I was considering moving the entire mailstore files (ebd and stm)  across, but I feel that will just move the corruptions along, I would rather try to fix them, as well as that, there are already 16 Mailstores on the target, so I can't.

Just to be clear, this has happened on 3 completely different servers, so I think I can rule out Hardware issues, and I have had different firmware/driver setups causing the same issue.
All Hardware is HP DL380s running FW 8.3 and Smartstart 8.7, another server with same issues was running FW8.2 and SS8.6

any ideas?

thanks
OscargG
OscarIT GuyAsked:
Who is Participating?
 
OscarConnect With a Mentor IT GuyAuthor Commented:
Hi Bryon,
Thanks for your reply :)
Most of the problem users are "Resources" such as meeting rooms with bookings for mettings etc.
To migrate them without their calendar, I may as well just recreate them afresh on the new server.
That would be my plan Z, as losing bookings for a meeting room would be a significant loss, especially since the migrations have gone so well to this point...

Im still looking for a way to either clean them before or during the migration, or possibly a fix to stop  the store crashing at the destination.

Although doing an exmerge from todays date onwards could be  a way... I have a gut feeling it will be to do with incorrectly formatted recurring series - but I only have at best sketchy evidence for that.
0
 
B HCommented:
have you tried to exmerge the problem users, but with custom settings like, only a certain date range... or discluding the calendar?
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.