Greg Cowell

I've done some preliminary tests. The following lists the times it took for the EV3 to fall over. Latest Python code running on Stretch: 40 sec, 3 min, 3 min,...

I have found that the reliability has been increasing as my testing proceeds. Now getting 10 to 15 minutes before the EV3 falls over. Perhaps an anomaly or perhaps due...

>Could you also share how you launched the program? Brickman >could you report if it was "sudden"? Yep, quite sudden. >Do you see this error message just before or after...

Thanks Laurens. Yep, when I said “connection being briefly lost” I was referring to the cable being disconnected briefly.

**System info (from `ev3dev-sysinfo`)** ``` Image file: ev3dev-stretch-ev3-generic-2017-09-30 Kernel version: 4.9.49-ev3dev-1.5.1-ev3 Brickman: 0.10.0 ev3devKit: 0.5.2 Board: board0 BOARD_INFO_HW_REV=7 BOARD_INFO_MODEL=LEGO MINDSTORMS EV3 BOARD_INFO_ROM_REV=6 BOARD_INFO_SERIAL_NUM=0016534FB669 BOARD_INFO_TYPE=main ```

I've also seen that behaviour. If I can get it to happen again, I'll do some logging to try to see what is happening. I suspect that the sensor read...

>Sometimes the physical impact can cause this, when the cables briefly loose contact. In that case, the actual cause may occur prior to this. Yes, confirmed. The exception was being...

Just did three tests with the modified kernel. No buffer overrun messages were logged. The first two tests only lasted a couple of minutes. The third test lasted 22 minutes...

>Not sure if this is happening but if I do not hold the EV3 perfectly still during the calibration/gyro offset calculation phase, am I right in assuming that this will...

Thanks Laurens. I’ll focus on data logging next.