Database access problem.

[PeterD]

We have users with build 737 through 739 using a shared EAP model on the network. The model is about 25meg. I can open the model but when accessing some of the classes I get the error message: DAO.Recordset[3197] ... The Microsoft Jet database engine stopped the proces because you and another user are attempting to change the same data at the same time.

I have locked and unlocked the objects and I have used the EA utility to repair the database but the problem remains. What does this message mean and what can I do about it?

[sargasso]

Ah the joys of JET databases  

AKAICT you'd be better off directing this question to sparx support - it looks like someone has left the database lock file in place.

hth
Bruce

[dhaynes]

I just encountered this same problem. I think I may know where it came from.

I may have had the .eap file open on my computer at home while I SFTP'ed that file to here at work (using Open SSH secure FTP client on Windows 2K). Currently every time I try to open the diagram that I last had it open to at home I get the error and there doesn't seem to be any way of fixing it at the moment.

When I get home this evening I'll try closing it and transferring the file again. and see if that cleans up the error.

[dhaynes]

Actually what happened in my case is that after I left my unattended sftp file transfer, it hung in mid transfer waiting for me to confirm an overwrite on another file I had queued up to copy.

So my .eap file was open for writing and only partially transferred at the time I was trying to access it from this computer. I finished the transfer last night and all is well now.

[PeterD]

The problem appears to have been triggered when a users computer crashed while accessing the database. I copied the database to a new file and the problem remained.

Using the EA utility to compact the database corrected the problem, although I don't know how.

[sargasso]

Peter,
This is not an uncommon problem with Jet databases.  The dbe uses a separate "lock" file called <filename>.ldb which usually resides in the same directory as the db.  When all users properly exit, the lock file is deleted, when one or more users crash the lock file is left there and generates that and other error messages.
(p.s. The file is not really a record lock file - its a "user is active" file)
Whenever tou get one of these access denied types of message its always a good idea to get everyone out of the database and check that the file has been deleted - if it hasn't then you at least know that you should check/compact the database.

thh
Bruce