Database Corruption (1 Viewer)

[way2bord]

After auto-repairing and restoring a corrupted database, I discovered an automatically generated access table:

I've never encountered a negative errorcode before.
Anyone familiar with the following...?

table: MSysCompactError
 
[U]Fields[/U] : [U]Value[/U]
ErrorCode: -1017
ErrorDescription: Record is deleted.
ErrorRecid: 爂
ErrorTable: <tablename>

Also...
:banghead: Any reccommendations on how to pinpoint what is causing database corruption? We've gone months with zero-issues, but recently the backend database has corrupted twice in 2 weeks.

I've a gun feeling there's some sort of introduced conflict from new end-users (we've had a large influx of new users - potentially using legacy equipment or home-PCs versus standard configged IT hardware). Could it be different OS or Access versions? (ie. an '07 or '03 opening the DB and creating the first lock file, followed by a '10 logging in?)

Any feedback or insight appeciated.

 

[Galaxiom]

Connection dropout problems often cause corruption. Is your network hardware up to the increased number of users?

 

[way2bord]

Connection dropout problems often cause corruption. Is your network hardware up to the increased number of users?

No problems with the company network; perhaps offsite end-user issues connecting over VPN? I don't have a lot of control over this aspect. Any other recommendations, or suggestions to implement, from a database design/management perspective, that could reduce the above problem?

 

[Galaxiom]

Offsite is definitely an issue with Access. Passing the pages from the Access backend to the front end and back across a WAN will almost always cause corruption.

Ideally, change to using Terminal Services so that the front end is run on the server and the users are remotely controlling the session.

 

[RainLover]

It may not do a lot but worth a go.

Create a new Database, import the Tables and make the new one the current Back End.

Do you have anything else in the back end. Like code.

Is it only a Front end Back end situation. You don't have anything else like a Code Library that you share.

 

[RainLover]

Don't forget a Compact and Repair on your new Back End.

Do heaps of back ups.

I would guess that at the moment you still have some corruption.

If it does not compact and Repair you will have to test every Table until you find the fault.

 

[davegeei]

MS Access MDB and ACCDB database files may get damaged due to many number of reason. To repair such error i would recommend RecoveryFix for Access Database recovery software.


Thanks

 

[RainLover]

MS Access MDB and ACCDB database files may get damaged due to many number of reason. To repair such error i would recommend RecoveryFix for Access Database recovery software.
Thanks

I have never used that program/tool. Does it tell you what was wrong or does it simply fix it.

And does it actually fix a damaged record or does it just delete.

Would be interested in knowing. Thanks.

 

[Simon_MT]

If you are using Acess over a WAN you need to use a Terminal Server or something that allows a long Session Timeout. If there is an outage you can resume the Session.

Simon

 

[way2bord]

Thank you for the feedback.

There's never any problem recovering from the corruption - and plenty of backups have been made.

I would, however, like to minimize repeat occurrences of corruption since any time a recovery is necessary, the end-users are unable to access the data.

The database is split appropriately. There is currently only tables in the backend, while the front end does all processing/forms/etc. and only sends/receives table-data updates.

It sounds like a Terminal Server is one of the best options...but I am very unfamiliar with its use, especially in regards to an Access database. :banghead:

Manual Compact and Repairs are run regularly, but I'd love to automate a compact and repair scenario, given that A) users may be connected, B) the backend is not / front ends are not running continuously.

 

[Galaxiom]

See nobody can tell you the exact cause behind database corruption since there are many.

I said what the cause was in post 4. Access connecting across any network with potential for significant latency will corrupt the back end.

 

[RainLover]

Just for clarification.

Does every user have their own front end. I am thinking more about your remote users.

 

[way2bord]

Just for clarification.

Does every user have their own front end. I am thinking more about your remote users.

A version controlled front-end is installed on all clients.

 

[Simon_MT]

The problem with WAN access is an outage. This has to mitigated.

Simon

 

[dcima]

I had the same problem, try with AccessFIX, I always use this software when I have a corrupt database

 

[RainLover]

I would think that this Question has been answered.

If not create a new Thread.