Project

General

Profile

Bug #21874

Device Reading NaN

Added by Beau Harrison 9 months ago. Updated 9 months ago.

Status:
Remission
Priority:
Normal
Assignee:
Start date:
02/08/2019
Due date:
% Done:

0%

Estimated time:
Duration:

Description

U:LNLOSS is reading NaN occasionally. We should avoid this output and present an appropriate error.

History

#1 Updated by Beau Harrison 9 months ago

Dennis pointed out that there is an explicit check that pointsCollected is not 0 to avoid dividing by 0. This is true and I added several logging statements yesterday looking at the variables in the same scope to understand how a NaN is generated.

#2 Updated by Beau Harrison 9 months ago

So it turns out that the numerator, runningSum, is NaN. runningSum is only ever +='d to so this must mean that some value that is NaN is being added to the runningSum. I've added more debugging statements in the receiveData function to understand how to better handle the incoming data.

#3 Updated by Beau Harrison 9 months ago

  • Status changed from New to Remission

There was a hardware issue causing errors in the source devices. The hardware issue has been resolved, making this issue more difficult to troubleshoot. There is an error parameter in the receiveData method that is not being used. If this comes up again this might be a good place to start. We can attempt to push an ACNET error to the TIMEAV device when the source device is in error.

DPE02 logs should show any anomalies if this issue returns.



Also available in: Atom PDF