Repeated unexpected traps (M1)

[mbenzvi]

This is moving the issue I brought up in another thread to this one. To see original:

Originally posted by @mbenzvi in #138 (comment)

I ran the command:

PATH=/opt/homebrew/Cellar/gnu-getopt/2.37.2/bin/:$PATH ./run-mtgo

I have included the log for the issue:
output-current-version.log

There seems to be a point where things break and I repeatedly get the issue Got unexpected trap -1

I also tried the older version using per recommendation:

`PATH=/opt/homebrew/Cellar/gnu-getopt/2.37.2/bin/:$PATH ./run-mtgo panard/mtgo:2021-07-24'

This got stuck on uncaught target signal 11 (Segmentation fault) - core dumped although also had the same error as the current version multiple times in the running.

output-old-version.log

I have an M1 chip on OS big sur. Let me know if more information is needed.

[pauleve]

Apparently, on M1, docker falls back to qemu to emulate an intel architecture, which might cause troubles.
The only thing I can think of now is trying to change line 9 of the run-mtgo script to

opts="--platform linux/amd64"

Btw, you can download the latest run-mtgo script and you won't need the PATH trick anymore (in theory).

[mbenzvi]

Thank you for the advice on downloading. I downloaded it, edited the command as per your recommendation and am running into something else (and the shorthand issue from before seemed to return). Here is the log file:

output.log

[pauleve]

Sorry, the fix was buggy, please download run-mtgo again, and retry the change.

[mbenzvi]

I downloaded and made the suggested change, but got the following error:

WARNING: image with reference panard/mtgo was found but does not match the specified platform: wanted linux/amd64, actual: linux/386

Log: output-pre-386-fix.log

So i replaced the amd64 with 386 in the opts command of the run-mtgo file and then ran into a similar issue as before:

qemu: uncaught target signal 11 (Segmentation fault) - core dumped and then nothing happens and I need to kill it via docker kill mtgo_running

Log: output-post-386-fix.log

[pauleve]

Issue moved to #188