Unplugging A301 does not disable command from HWC2

[rzblue]

Steps to reproduce:

1. Enable robot
2. Run A301 via hardware client
3. While A301 is runnning, unplug motor
   • Hardware client correctly exits device screen and removes device from list
4. Plug A301 back in
   • A301 starts running at previous command
5. Click device to open status screen
   • shows data indicating motor is running

20260616-2344-16.6132049.mp4

[jfabellera]

This is an issue in REVLib, so you will be able to see this on a robot as well (deploying a program that does not command a previously commanded motor).

This should be an easy fix.

Edit: this is a separate issue

[jfabellera]

Actually my previous comment is about a separate issue.

I followed the steps to reproduce and looked at the CAN bus monitor on Systemcore to verify that RHC2 stopped sending the setpoint command (because I was confident it did), and it in fact does stop sending the messages, but the motor still spins on reconnection.

To confirm it is not the motor firmware saving the setpoint, I plugged in a separate motor than the one that was unplugged and that one also started spinning.

The smoking gun was if I reconnect the motor a second time after reproducing the issue, it doesn't happen again. From looking at the CAN bus monitor, you can see that a few setpoint commands are sent after the device is disconnected because the client takes some time to detect the device has disconnected. Since A301 is the only device on the bus, those messages have no where to go when it is disconnected which indicates the Motioncore is queuing the messages up. So when the device is reconnected, it sees those extra setpoint commands (even though the client is not explicitly sending them in that moment), and the motor starts spinning again.

To further confirm this theory, I connected multiple devices to one channel on the Motioncore along with an A301 and followed the steps, and I was not able to reproduce the issue because there were other devices on the bus for the message to go to. This also occurs on Systemcore.

In the video below, you can see the message count start when the motor is running and stops when I stop or disconnect the motor. The message count stays the same throughout unplugging and plugging back in.

@bhjelstrom So the question is, is this something that can and should be handled on the Systemcore? I can also see this issue propagating in user code.

Recording.2026-06-17.171800.mp4

[jfabellera]

This was confirmed by Limelight to be a bug that should be fixed soon.

[rzblue]

Out of curiosity, how was the bug fixed?

[jfabellera]

@bhjelstrom should be able to give a better answer, but I believe the queued up messages I theorized about gets cleared when there is no bus.