Created attachment 1761066 [details] journal.log Description of problem: 'dnf update systemd' results in a coredump of systemd and sometimes other processes; this is new behavior since systemd-248. The problem doesn't happen with 'dnf offline-(upgrade/distrosync)' or GNOME Shell pkoffline updates; and sounds like openqa tests don't hit this possibly because they aren't doing a user login before running the update. Version-Release number of selected component (if applicable): systemd-248~rc2 both -1 and -3 How reproducible: 5 for 5 Steps to Reproduce: 1. Workstation, login graphically 2. Launch GNOME Terminal 3. sudo dnf update *rpm -rw-r--r--. 1 chris chris 4560335 Mar 5 14:34 systemd-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 534507 Mar 5 14:34 systemd-container-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 604438 Mar 5 14:34 systemd-libs-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 488253 Mar 5 14:35 systemd-networkd-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 24660 Mar 5 14:35 systemd-oomd-defaults-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 327658 Mar 5 14:35 systemd-pam-248~rc2-3.fc34.x86_64.rpm -rw-r--r--. 1 chris chris 27151 Mar 5 14:35 systemd-rpm-macros-248~rc2-3.fc34.noarch.rpm -rw-r--r--. 1 chris chris 1579712 Mar 5 14:35 systemd-udev-248~rc2-3.fc34.x86_64.rpm Actual results: [18694.838020] systemd[1]: Caught <ABRT>, dumped core as pid 8919. Expected results: Not crash. Given that this is an ill advised update while logged in, maybe it shouldn't restart? I'm not sure if it's possible to just update it and not restart/reload, leave that up to reboot. Additional info: [chris@flap Downloads]$ coredumpctl info 8919 PID: 8919 (systemd) UID: 0 (root) GID: 0 (root) Signal: 6 (ABRT) Timestamp: Fri 2021-03-05 14:35:51 MST (3h 9min ago) Command Line: /usr/lib/systemd/systemd --switched-root --system --deserialize 31 Executable: /usr/lib/systemd/systemd Control Group: /init.scope Unit: init.scope Slice: -.slice Boot ID: e4adfc6dc21a4c558414b0145f0b4761 Machine ID: 4fa861500602414cb8efcf9038143a57 Hostname: flap.local Storage: /var/lib/systemd/coredump/core.systemd.0.e4adfc6dc21a4c558414b0145f0b4761.8919.1614980151000000.zst (inaccessible) Message: Process 8919 (systemd) of user 0 dumped core. Stack trace of thread 8919: #0 0x00007f9aa39fc58b kill (libc.so.6 + 0x3d58b) #1 0x000056205e1f7b68 crash (/usr/lib/systemd/systemd (deleted) + 0x45b68) #2 0x00007f9aa3ba1960 __restore_rt (libpthread.so.0 + 0x13960) #3 0x00007f9aa39fc292 raise (libc.so.6 + 0x3d292) #4 0x00007f9aa39e58a4 abort (libc.so.6 + 0x268a4) #5 0x00007f9aa3a3ecd7 __libc_message (libc.so.6 + 0x7fcd7) #6 0x00007f9aa3a4694c malloc_printerr (libc.so.6 + 0x8794c) #7 0x00007f9aa3a4782c unlink_chunk.constprop.0 (libc.so.6 + 0x8882c) #8 0x00007f9aa3a479d0 malloc_consolidate (libc.so.6 + 0x889d0) #9 0x00007f9aa3a48200 _int_free (libc.so.6 + 0x89200) #10 0x00007f9aa3a4ba48 __libc_free (libc.so.6 + 0x8ca48) #11 0x00007f9aa3e221d9 n/a (/usr/lib/systemd/libsystemd-shared-248.so (deleted) + 0x1521d9) [chris@flap Downloads]$ Fri 2021-03-05 14:35:51 MST 8919 0 0 SIGABRT inaccessible /usr/lib/systemd/systemd n/a Looks like it's missing ACL? # ll total 19212 -rw-r-----+ 1 root root 808691 Mar 5 14:35 core.gsd-datetime.1000.e4adfc6dc21a4c558414b0145f0b4761.1800.1614980147000000.zst -rw-r-----. 1 root root 729942 Mar 5 14:35 core.systemd.0.e4adfc6dc21a4c558414b0145f0b4761.8919.1614980151000000.zst # getfacl * # file: core.gsd-datetime.1000.e4adfc6dc21a4c558414b0145f0b4761.1800.1614980147000000.zst # owner: root # group: root user::rw- user:chris:r-- group::r-- mask::r-- other::--- # file: core.systemd.0.e4adfc6dc21a4c558414b0145f0b4761.8919.1614980151000000.zst # owner: root # group: root user::rw- group::r-- other::--- If I copy it to ~/ $ coredumpctl --file=core.systemd.0.e4adfc6dc21a4c558414b0145f0b4761.8919.1614980151000000.zst Failed to open journal files: Bad message
*** This bug has been marked as a duplicate of bug 1931034 ***