I’m on a Linux distro called NixOS, and I’ve gone as far as I can towards packaging Everdo:
> everdo Trace/breakpoint trap (core dumped)
Jun 15 09:39:26 whittles-iskandar kernel: traps: ThreadPoolForeg trap int3 ip:55e4e1e0dc03 sp:7f9944ab3fe0 error:0 in everdo[55e4e1578000+60cd000] Jun 15 09:39:26 whittles-iskandar systemd: Started Process Core Dump (PID 15387/UID 0). Jun 15 09:39:26 whittles-iskandar systemd-coredump: Resource limits disable core dumping for process 15360 (everdo). Jun 15 09:39:26 whittles-iskandar systemd-coredump: The core will not be stored: size 18446744073709551615 is greater than 804257792 (the configured maximum) Jun 15 09:39:26 whittles-iskandar systemd-coredump: e]8;;man:core(5)[🡕]e]8;; Process 15360 (everdo) of user 1000 dumped core. Jun 15 09:39:26 whittles-iskandar systemd: email@example.com: Succeeded.
Here’s what I’ve done: Everdo: init at 1.5.14 · WhittlesJr/nixpkgs@dc5a4eb · GitHub
NixOS is different from most Linux distros in that it does not use universal paths for its files, like
/opt/Everdo/everdo. Instead, it stores packages off under unique locations based on the hash of its contents, like
/nix/store/ar3b6abcjsda54j6c16ilzwhvywnp3zx-everdo-1.5.14/lib/everdo. These are linked together on a per-version basis, so when I upgrade, the old stuff remains but new stuff gets added, and pointers essentially switch over to the new version.
All that to say, if your application is trying to call something under
/bin/ or some other hard-coded location, that could be the culprit.