GPY crash within a crash
-
We've been testing boot.py on a GPY by letting the program run with the GPY connected to Atom & looking at where in the program failures occur.
In 1.18.1.r7 our GPY would eventually terminate a program that ended with deepsleep by finding a syntax error in boot.py when it was time to wake up, or it would reformat boot.py to the default null progrm. Programs running in a while 1: that did not deepsleep tended to last longer.
With the recent change to 1.20.1.r1 (with LittleFS) we decided to retest behaviour. Programs ending with a deepsleep seemed better behaved but we just had event with a non deepsleep program with the dreaded Guru Mediation Error. There was another exception during the core dump, the GPY then rebooted, ran the program for a few lines then tried to upload from Atom like what happens if we click the Upload button.
We are puzzled by the un-commanded attempt to re-upload boot.py from Atom. It seems like the GPY stores Atom operations (like mouse clicking the Upload button) then tries to do the operation again when it gets confused? We're wondering if this device is making any headway towards becoming a stable platform?
-
@kjm said in GPY crash within a crash:
I'm gonna stick with the 1.20.1.r1 Robert did for me for a little while, it has adhoc wifi by default plus I'm in the middle of chasing down a bug
@kjm said in GPY crash within a crash:
I don't like to change firmware mid hunt
I understand that. However, we’ve just released another bunch of firmware images with hopefully improved robustness, so [1] might make you happy. If you feel lucky, you might want to try this and we will be happy to hear about the outcome from you.
-
@andreas Tnx Andreas. I'm gonna stick with the 1.20.1.r1 Robert did for me for a little while, it has adhoc wifi by default plus I'm in the middle of chasing down a bug ('if not lte.isattached()' sometimes says true when the modem is, in fact, attached) & I don't like to change firmware mid hunt. Very grateful for the persistence of people such as yourselves.
PS: Had no idea bees where such a big thing in Europe. https://newatlas.com/good-thinking/beep-base-bee-hive-monitor/
-
Dear @kjm,
good to hear that you are using LittleFS now as this will not corrupt your filesystem any more.
Regarding core panics, we still want to help people from the community observing similar things within some conditions so we just built another bunch of firmware images using different fixes from the upstream ESP-IDF.
So, without warranties or preliminary hopes of any kind, you might want to try [1] on your GPy. See also [2] and [3] for further references about it.
With kind regards,
Andreas.[1] https://packages.hiveeyes.org/hiveeyes/foss/pycom/vanilla/GPy-1.20.1.r1-0.5.0-vanilla-butterfly-csfix.tar.gz
[2] https://community.hiveeyes.org/t/investigating-esp32-core-panics/2480
[3] https://community.hiveeyes.org/t/pycom-inofficial-firmware-bakery/2742
-
@kjm Please can you post a code snippet the reproduces the core dump ?