Interface file kept on server vs workstations

[321]

Hello all,

I'm currently using the separation model. I have previously kept the interface file on the server in the same folder as the data file. I've recently tried putting the interface file as a separate copy on a couple users’ workstations. This has improved performance. However, I'm a little weary on how record locking (or lack thereof) will work. I was wondering if anyone has had trouble with corrupt data when 2 users are attempting to change the same data. As a test, I went into the same field as another user w/out any locking warning from filemaker. (Both interface files independent, on separate workstations, accessing data from the same file on the server) I have lockout scripts in place that I use prior to editing a record or starting scripts in which a change of data occurs. But, those scripts rely on filemaker recognizing other users having the same record open. I could change those scripts to not use filemaker for the lockout, and do it using fields, but I'd prefer not to....

So that brings me to my question: Has anyone ever had any corruption, or data loss issues due to using the scenario described above?

Thanks,

Eric

[LaRetta]

Hi Eric,

I am surprised and would be very interested in the type of tests you performed to conclude that UI on user stations increased performance since overall this has not been my conclusion. I am very open to changing my thinking here since the only tests I ran involved importing and/or external data manipulation. And now I want to test even further so please share.

As a test, I went into the same field as another user w/out any locking warning from filemaker.

And you are aware that entering same field doesn't cause record lock - only when attempt is made to change data?

So that brings me to my question: Has anyone ever had any corruption, or data loss issues due to using the scenario described above?

No because I've never tried it but I will now add it to my list of tests.

Are you really still on vs. 7? You also haven't specified your OS. Your FM version, OS and even Updater version might all contribute or change test results as each FM version (and Updater) comes with its own set of characteristics and bugs.