[3.5b1?] CAN from 6HC to Mini 5+ fails after board reset

[dc42]

@Diamondback does the same problem occur if you send M999 to reset the main board explicitly?

[Diamondback]

@dc42 Yes, the same happens when sending M999.
A tiny bit more info, when this occurs, the main board takes a very long time to start up (as if it was waiting for the CAN connection to start up or something), eventually it does boot up but has no actual connection to the expansion board (tools that use heaters there show no temps etc)

[dc42]

@Diamondback thanks. I suspect this is a new issue in RRF 3.5. Let me know if you find it happens using 3.4.5 too.

[Diamondback]

@dc42 Just went back to 3.4.5 so that I can actually print and this issue does NOT happen there, so seems to be new with 3.5b1.

[dc42]

@Diamondback thanks for confirming. Did you run G32 on the main board before the problem occurred? If so then please see https://forum.duet3d.com/post/303514.

[Diamondback]

@dc42 No, this happened after a fresh boot.

[dc42]

@Diamondback thanks, I'll try to reproduce it.

[dc42]

@Diamondback I have reproduced it:

• Doing a software reset of the main board causes the expansion board to lose CAN connection;
• Pressing the reset button on the expansion board restores the connection;
• Pressing the reset button on the main board causes only a temporary loss of CAN connection.

[dc42]

@Diamondback I have fixed this, along with the problem of not being able to update a main board used as an expansion board over CAN. I have put a new Mini5+ binary at https://www.dropbox.com/sh/28zqy8t165nwa2f/AACZvmVSGhWMSpjKX43MsSWSa?dl=0.

[Diamondback]

@dc42 awesome, will test soon! Thanks a lot!