Rapid disk use with DSC data logging

[Lenvno]

I recently installed Labview 8.2.1 with DSC module.  I have tried a few shared variable projects with data logging and now I find my hard drive space is being consumed very rapidly.  The variable manager indicates that only a few shared variables are left from my projects.  They would produce very little data in the Citadel database.  Certainly not enough to consume 100 meg + of disk per day.  I am obviously missing something that would turn off some hidden logging process.  Any suggestions?

[RavensFan]

How wide or narrow is your logging deadband?  If it is too narrow, such as zero, you might be logging every datapoint unnecessarily.

[Ben]

Ravens Fan made a very good suggestion!

To follow-up on that idea, go to MAX and starting exploring your historiacl database. One or more of the saved traces will have a lot of data if Ravens Fan suspicions are corect.

Ben

[Lenvno]

Thanks for the input.  I did have another look at historical data and there are not large traces.  I used the Variable manager to stop all processes and I stopped the shared variable engine.  The leak continues.  The only way I can stop it is to use the Windows Task Manager to shut down the Citadel5 process.  I have searched the computer and cannot find any large database files that would account for the disk shrink.  I have also uninstalled and re-installed the DSC module.  No success.  

If I knew where the data was located I might be able to determine the source.  The National Instruments directory is not growing even though all the processes had their database pointed to a directory inside the NI directory!

Len

[Matt361]

Len -

Are you using an AMD Processor? There is a problem in LV8.2 when logging data using DSC and the AMD Processor. It's related to the way the Processor goes to sleep. (there's a patch that we can use if this is your problem)

What type of time frame is this happening?

Good Luck,

Matt

[Lenvno]

Thanks Matt but I am using a Pentium running WindowsXP Pro.   I may have isolated the problem which was consuming 100 meg of disk per day.  In one test library I had set up processes logging to a separate citadel5 database instead of the default one in the NI directory.  All the processes going to that directory had been stopped and the shared variables stopped showing up in the Variable manager tree.  In any case I deleted that database using MAX and the disk shrink stopped.  I did not get my disk space back but at least I'm no longer loosing it.

Len