Database Corruption Puzzle

[rcbaird]

I have an issue with a corrupted database that is puzzling me, and I’d like to know if anyone has encountered this problem and found a solution.

A few weeks ago a user was connected remotely (by phone) to an Access 2003 database that resides on a Novell server, and she was running an update query on a table containing approximately 24K records, when her computer locked up. The database was corrupted at that point and can no longer be opened, displaying the message “The Microsoft Jet database engine stopped the process because you and another user are attempting to change the same data at the same time.”

I’ve tried fixing the problem with the native Access repair utility as well as Jetcomp, but have had no success. I also purchased a utility that was recommended, AccessFIX, which was able to repair the database so that it could be opened. However, the primary table has lost roughly 20K records.

One user enters records daily, and another user runs updates and reports at the end of each month.The puzzling part of this problem is the fact that up until the database crashed in July, it had been working without issues. Now, however, every copy of the database that the LAN manager has been able to restore from backups is displaying the same message. Even a copy of the file that the user made and placed on the server in another folder as far back as February is displaying the message. Likewise, the copy they sent me displays the message, and I am not connected to the server.

I've had the LAN manager check and make sure there are no stray .ldb files residing in the folder with the database. Does anyone have a suggestion for how to recover the data fully?

 

[jrbarnett]

Is this database split frontend/backend? if so, did the restore replace both the frontend and the backend or only the backend data at the time?

John

 

[rcbaird]

Sorry, I had meant to mention that this is not a split database. It's not my design, but because I had developed another database for them (which does utilize a fe/be), they were hoping I could help them resolve this problem.

 

[MajP]

My first guess is the primary table never existed in the database and was a linked table to another database? That would explain why all old versions show similar corruption and why it appears to have disappeared.

 

[jsteph]

Have you tried opening any of the backups on your client machine, taking Novell issues out of the picture?
--Jim

 

[rcbaird]

Thanks to both the members who responded most recently.

The table which is missing records after being recovered is not a linked table. As for backups, the user backed up once back in February (when the database was still working), but that copy now exhibits the same corruption as the original, even though it was made months ago and the corruption seemingly occurred recently. The same goes for the network backups.

I'm beginning to think that the corruption may have been present for some time, perhaps only rearing its head when the file was copied or backed up on the server, and that that it finally more or less "exploded." I know of no other explanation for why all the backup copies would display the same message, so I'm just guessing.

A person that performs recovery as a service(Wayne at

http://www.everythingaccess.com--very

nice and helpful btw), referred by someone on another forum, has told me the corruption occurred at the file system level and that a portion of the file is simply irrecoverable.

I guess that's the end of the story with this file. Thanks again to all of you on the forum who took the time to offer suggestions.