Error: -200290, The generation has stopped to prevent the regeneration of old samples.

[whemdan]

I am running many 8 second analog output data generation tests on multiple devices synchronized by a trigger signal. the trigger signal is generated using a counter output pulse generationed on one of the devices. For the most part, the code runs for hours without any problem, but every few hours it would stop with an -200290 error.

I am setting up and running finite data generations, where ALL samples have been predownloaded to the device - so I do not understand how it is even possible to get this error.

I have attached an NI IOTrace file.

Observe the error on line 4750 on the task ending in 60E0. Notice that this task along with the 2 other tasks (3 tasks for the 3 synchronized devices in total) were running peacefully before that and had output a large number of their data.

4 Tasks are started lines 3901, 3905, 3907 and 3911 for the 3 analog and 1 digital output generations. All of these 4 tasks are finit generations and were fed all their sample data prior to starting the finite generation. So it should be impossible to get an -200290 error on any of those operations. I have a 5th Task used to trigger the 4 other tasks, it is simply a counter output pulse generation and it stops after 4 seconds (line 4561).

The code works most of the time, and the issue is observed every 5-6 hours.

I am using NI USB-6351, USB-6218, and an NI-USB-6218.

* The NI USB-6351 is used to generate the trigger signal on 1 task.

* All 3 devices are setup to wait on that start trigger.

* I am setting up analog output tasks with 2 channels on each of the devices as well as digital output lines on the 6351.

* All tasks have their data pre-queued to them and are run in finite mode - so a 200290 error should be impossible.

Any suggestions?

I was not able to attach the NI IO Trace because the file might be too large...

I am running on a Windows 64 bit machine, using NIDAQmx 9.8.0f0

Thanks,

   Wael Hemdan

   MathWorks

[whemdan]

Hi,

I am having difficulties submitting the original NI IO Trace file. I am guessing it is too large.

It seems the same bug is reproducible with only one device... So I am working on reproducing a simpler and smaller example to post soon.

Thanks,

Wael Hemdan

MathWorks.

[Cavarval]

Hello Wael,

Reproducing the problem in a simpler program is definitely a good idea.

In the mean time, you can take a look at these knowledgebase articles:

http://zone.ni.com/reference/en-XX/help/370473H-01/mstudiowebhelp/html/nidaqmxdrivererrors/

http://digital.ni.com/public.nsf/allkb/9548564333A0437F86256FC00060516C?OpenDocument

http://digital.ni.com/public.nsf/allkb/A806A0349FBF30CD8625795A007376F3?OpenDocument

You can find a couple of hints there that might help you with your application.