On 13.4.2023 21.15, Carlos E. R. wrote:
On 2023-04-13 17:39, Patrick Shanahan wrote:
* Carlos E. R. <> [04-13-23 09:21]:
Hi,
Using 15.5, I just had a coredump, yet systemd claims there is none:
Beta:~ # hw-probe -all WARNING: 'edid-decode' package is not installed Probe for hardware ... Ok Reading logs ... sh: line 1: 9228 Segmentation fault (core dumped) LC_ALL=C vdpauinfo 2>&1 Ok Local probe path: /root/HW_PROBE/LATEST/hw.info Beta:~ # coredumpctl No coredumps found. Beta:~ #
How do I enable them? (Search finds obsolete info for SLE 12)
google suggests modifying /etc/systemd/system.conf
And does it say how? Because my duckduckgo search did not find that.
If it refers to:
#DumpCore=yes
then "yes" is the default setting.
Beta:~ # systemd-analyze cat-config systemd/system.conf | grep -i core #DumpCore=yes #DefaultLimitCORE= Beta:~ #
Do all of these hold true: - kernel.core_pattern sysctl contains path to systemd-coredump binary antique:~ # sysctl kernel.core_pattern kernel.core_pattern = |/usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %h - systemd sockets.target has systemd-coredump.socket defined antique:~ # systemctl list-dependencies sockets.target sockets.target ● ├─avahi-daemon.socket ● ├─dbus.socket ● ├─dm-event.socket ● ├─iscsid.socket ● ├─rpcbind.socket ● ├─systemd-coredump.socket ● ├─systemd-initctl.socket ● ├─systemd-journald-dev-log.socket ● ├─systemd-journald.socket ● ├─systemd-udevd-control.socket ● └─systemd-udevd-kernel.socket - systemd-coredump.socket is active antique:~ # systemctl status systemd-coredump.socket ● systemd-coredump.socket - Process Core Dump Socket Loaded: loaded (/usr/lib/systemd/system/systemd-coredump.socket; static) Active: active (listening) since Fri 2023-04-14 18:04:47 EEST; 13min ago Docs: man:systemd-coredump(8) Listen: /run/systemd/coredump (SequentialPacket) Accepted: 0; Connected: 0; CGroup: /system.slice/systemd-coredump.socket The above is from 15.4 with package default config. I am assuming that 15.5 has not changed that much wrt. how core dumps are handled. If the socket is not active you should see errors in journal: antique:~ # journalctl -t systemd-coredump Apr 14 17:36:05 antique systemd-coredump[17051]: Failed to connect to coredump service: No such file or directory Apr 14 17:39:46 antique systemd-coredump[17126]: Failed to connect to coredump service: No such file or directory Apr 14 17:40:51 antique systemd-coredump[17499]: [🡕] Process 17170 (sleep) of user 1000 dumped core. Found module linux-vdso.so.1 with build-id: bd7bc2f270a364722e0dd33841> Found module ld-linux-x86-64.so.2 with build-id: 306fa1f1f4692920c5a65> Found module libc.so.6 with build-id: 171a59c1c43a8f7b93c3dff765aae0b6> Found module sleep with build-id: 84c8dc3090f596611b5062255c2b96886569> Stack trace of thread 17170: #0 0x00007f7920d75b4a clock_nanosleep@@GLIBC_2.17 (libc.so.6 + 0xdbb4> #1 0x00007f7920d7b8a3 __nanosleep (libc.so.6 + 0xe18a3) #2 0x000055f53ba04b47 n/a (sleep + 0x4b47) #3 0x000055f53ba04909 n/a (sleep + 0x4909) #4 0x000055f53ba01980 n/a (sleep + 0x1980) #5 0x00007f7920ccf24d __libc_start_main (libc.so.6 + 0x3524d) #6 0x000055f53ba01a7a n/a (sleep + 0x1a7a) antique:~ # coredumpctl TIME PID UID GID SIG COREFILE EXE SIZE Fri 2023-04-14 17:40:51 EEST 17170 1000 100 SIGQUIT present /usr/bin/sleep 28.2K