Corrupt DB

[Carl_R]

We are using a split database (access97). Both the backend(MDB) and frontend(MDE) reside on the network.

We are experiencing corruption on a daily basis.
Some records are ending up #ERROR and some #DELETE

This requires a repair of the backend which rebuilds the index ok but now if you try to update one of the previously damaged records, a message is generated that the record is locked by another user (even though nobody is actually using the application) - even though this message is generated, you can still update the record??

Is it possible for the frontend(MDE) to become corrupted?

It's very frustrating because we cannot isolate where the problem lies.

Any suggestions or ways we can minimise corruption will be a huge help.

 

[jjturner]

Just had my first bout with a corrupt db today, split exactly as yours, with both ends on the network rather than f/e resident on each Client workstation.

Luckily I was able to create a repaired/restored/compacted copy of the b/e and just copied over the corrupt b/e for starters.

This allowed me to run the db. However, the db promptly re-crashed.

From there, I copied over the presumably corrupt f/e (since I was now working with a repaired b/e) with my backup f/e.

Everything has been hunky-dory since (so far). So, to at least partly answer your question, yes, I believe f/e can get corrupted, at least from what I can deduce from my own experience.

Whoops - one big exception - I'm not using an MDE as the f/e, just standard MDB.

HTH,
John

 

[Carl_R]

Hi John

Repaired the BE and now running with the backup version of the FE MDE and all looks good.

Cheers
Carl

 

[Rich]

Just curious, why did you make the FE an MDE?