While I have found /var/lib/systemd/coredump/* to be extremely useful when gdb doesn't give the right answers, I have not been able to figure out the rationale behind it and what existed before it, in the sysvinit world-view.

I came across this recently as I have been getting segmentation faults on some software but running them under gdb didn't get the desired output. Running through /var/lib/systemd/coredump/ found the lost files and was able to run coredumpctl on some of it while the rest was able to unarchive the lz4 files and run them gdb and get the backtraces.

Any history, rationale would be nice.

While man pages of systemd-coredump, coredumpctl and even journalctl provide clues as to how it is to be used, doesn't tell anything about the rationale or reasoning/history behind it.

closed as too broad by Stephen Harris, Romeo Ninov, Rui F Ribeiro, Isaac, schily Nov 6 '18 at 13:59

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    Since it looks like this will be closed, I will partially answer: I don't know why systemd's developers decided to offer their own coredump handler. Before this, some systems provided ABRT or Apport to provide some features beyond just creating a core file. Before that, you could specify a specific directory for all your core dumps. Before that, core dumps were put in the process's working directory, wherever it happened to be. – Mark Plotnick Nov 5 '18 at 20:41
  • @MarkPlotnick it couldn't possibly be embrace and extend, could it? – roaima Nov 5 '18 at 22:52

The interfaces seems focused around using coredumpctl to start gdb on core dump.

Origins seems to be from v39 with a focus to put metadata into journalctl.

Later converged into abrt got the same messages (was merging at the time hard/rejected?).

Its development still seem to be a year after abrt was going and apport had a significant number of releases so it looks like a NIH unfortunately.

  • 1
    NIH = Not Invented Here, got you. – shirish Nov 6 '18 at 5:41
  • I am awaiting for couple of days to see if somebody adds some more info., if not will award you the answer. Thank you for your patience. – shirish Nov 6 '18 at 15:14
  • It was an interesting question. It was a pleasure to look up the timeline of how things developed. – danblack Nov 6 '18 at 21:37

Not the answer you're looking for? Browse other questions tagged or ask your own question.