1V91 - Seems less stable than 1v90+ interim versions

Just that a program I have been running for awhile (days) off and on to toggle a puck from another puck and gather statistics, now just dies on both ends after a few hours. Before I was getting a timout 7 error or something like that on the transmitter after many hours. But a button press would reset the timer in code so it would start up again. On the receiver the connections would continue to succeed but the action, the toggle, would stop. But with 1v91 it all worked well enough at first but then just died and non-responsive. Just sharing experience. Nothing concrete. I think I was on the 1v90.12 on both ends before.

If you look at http://www.espruino.com/ChangeLog, 1v90.12 was released at around the micro:bit line.

Most of the stuff since then should really be pretty harmless (or limited to one small area that's not what you're using), but I guess increasing the custom UUID count could have impacted stability.

I've had one running here for days though and it's been ok - albeit not actively connecting.

Have you checked the battery status? The firmware update takes quite a lot of power, and could potentially have pushed batteries over the edge?

But if you can reliably make it crash after an hour that's great (even better if you can do it in less time). If you could post up your code then it's more likely I'll be able to find the bug :)

Yes I will try that (new batteries). My hacking activity has slowed since the holidays and going back to work. But I'll get to it!