Very odd CLX watchdog error.

Earlier today I was online with a Compact-Logix that is being comissioned and I noticed that it was generating a minor watchdog fault about once every second. There was no indication of which task was generating the watchdog error. None of the tasks had a period closely matching that so I took a look at each of the task monitors (there are 6 tasks in this controller). All of the tasks were all executing properly, taking from less than 1 to around 6 milliseconds for the longest task. All prioriites were correct, and there was no task overlap occuring. So next I inhibited each of the tasks one at a time. The only observable change is that when I inhibited two of the longer tasks (one at a time) that the rate at which the watchdog minor fault occured changed to about once ever 2-3 seconds, but with all taks running, the error rate went back to about 1/second. So then I inhibited two tasks at once and the watchdog errors stopped. Then I uninibited the first task, and when the error did not occur, I unihibited the second task (all tasks now running) and the error again did not occur. I've been watching for several hours now and not once again has the watchdog error occured again, even though nothing else has changed.

I have no idea what the cause was. Has anyone ever seen anything like this?

Are you showing any task overlaps when you monitor the task? Other than that, no, I have not seen something like that before.

Do you have any message statements? This is where I would expect to see a potential issue.

Is this version 15?

Have you tried using the Logix5000 Task Monitor Tool software (extra install from the tools folder)? It can give you much more detail about the communications, available processor time, detailed info on tasks, etc? I have found it helpful in optimizing programs, it should help track down your problem.

Darren