Ancestors

Written by Andrew Helwer on 2025-01-11 at 21:42

Does anybody know a tutorial for debugging a crash in your desktop environment on Linux? I am mostly interested as an exercise in actually using the “source” part of open source. Like I imagine it requires:

1. identifying the exact DE version being run

2. fetching the source of that DE version from somewhere and building it in debug configuration

3. running that DE with a debugger attached then triggering the crash

The DE aspect makes it weird, like I would have to run it in a VM maybe so the crash doesn’t take down whatever program I’m using to run the debugger too.

[#]AskFedi #FOSS #Linux

=> More informations about this toot | More toots from ahelwer@discuss.systems

Toot

Written by Ken Milmore on 2025-01-11 at 22:12

@ahelwer You could start off by enabling core dumps, which can be used for postmortem debugging with gdb. Install the -dbg package for the relevant application and you should get reasonable information from a backtrace. Also don't ignore log files like .xsession-errors. Building specifically for debug tends to be very specific to the application and any libraries which may be involved, so get as much information as you can before trying to build.

=> More informations about this toot | More toots from kbm0@mastodon.social

Descendants