Daniel

> I'm changing this PR to WIP since Windows CI fails and I need to find out why. It was an easy bug to fix so WIP is removed.

WIP while fixing the issue in https://github.com/JMRI/JMRI/issues/12906#issuecomment-1965777470

> WIP while fixing the issue in [#12906 (comment)](https://github.com/JMRI/JMRI/issues/12906#issuecomment-1965777470) WIP removed since bug is fixed.

I presses the numbers and wait for the platform line to change color. Then I do a new try and then again, ... But I'm unable to get the error....

> I don't get an error since I check for a null result. You will see the messages in the system console. Is it this message you mean? `00:35:02,248 jmri.jmrit.logixng.actions.LogData...

Could you try my branch `danielb987 : LogixNG_SensorGroups`? https://github.com/danielb987/JMRI/tree/LogixNG_SensorGroups Or add the two changes manually to your branch? It's only two logging statements I have added. https://github.com/JMRI/JMRI/compare/master...danielb987:JMRI:LogixNG_SensorGroups It will show...

Note to myself: ``` 1: 15:22:09,323 ixng.actions.ActionListenOnBeansTable WARN - IQDA:AUTO:0008 :: propertyChange: Prop name: KnownState, source: IS104, old: 2, new: 4 [AWT-EventQueue-0] 2: 15:22:09,323 ixng.actions.ActionListenOnBeansTable WARN - IQDA:AUTO:0008 :: propertyChange:...

> Lines 1 and 2 are different beans. Yes, different sensors. But I ment they are the same LogixNG action IQDA:AUTO:0008. The problem is that the action doesn't buffer propertyChange...

> While the possibility of two events at the same time is very small it is not zero. A sensor group is probably the most likely to cause this problem....

@dsand47 PR #12925 fixes this bug. It's difficult code to get right and I don't have much tests for these classes so it would be good if you could test...