[SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

jpk · 2016-11-14 07:57:49

Trying to upgrade system to the latest version in testing. My systemd process 1 (PID 1) coredumps, after this the system cannot be restarted or shutdown. No *ctl tool provided by the systemd package can connect to the bus. By power cycling the system reboots, but hangs during boot. Reverting systemd to version 231 fixes.

Has anyone experienced something similar or now how to solve this?

Last edited by jpk (2016-11-24 19:34:42)

ewaller · 2016-11-14 14:26:29

PID 1 crashing is probably considered a bad thing I have not seen it, but the first thing that comes to my mind is -- do you have microcode updates for your processor installed?
Second, what does your disk space in your root and temp file systems look like?

WorMzy · 2016-11-14 14:51:42

Mod note: Moving to Testing.

EDIT: Are you on 32-bit? https://bbs.archlinux.org/viewtopic.php?id=219142

Last edited by WorMzy (2016-11-14 14:52:56)

jpk · 2016-11-15 07:19:26

tl:dr: still crashing

I installed the "intel-ucode" package and added the image to my in EFI loader configuration and verified it is loaded properly. I than tried upgrading "systemd" to version 232-2 again. Btw, the system architecture is x86-64.

I also tried "mkinitcpio" afterwards, and without it, and did the whole update progress again using "rescue.target", result stays the same. System won't boot and hangs before mounting the drives.

Log entries while installing systemd:

Nov 15 07:58:29 thor dbus-daemon[791]: Reloaded configuration
Nov 15 07:58:29 thor dbus[438]: [system] Reloaded configuration
Nov 15 07:58:29 thor dbus[438]: [system] Reloaded configuration
Nov 15 07:58:29 thor dbus[438]: [system] Reloaded configuration
Nov 15 07:58:30 thor dbus[438]: [system] Reloaded configuration
Nov 15 07:58:30 thor dbus[438]: [system] Reloaded configuration
Nov 15 07:58:31 thor systemd[1]: Reexecuting.
Nov 15 07:58:31 thor systemd[1]: systemd 232 running in system mode. (+PAM -AUDIT -SELINUX -IMA -APPARMOR +SMACK -SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN)
Nov 15 07:58:31 thor systemd[1]: Detected architecture x86-64.
Nov 15 07:58:31 thor kernel: systemd: 36 output lines suppressed due to ratelimiting
Nov 15 07:58:31 thor systemd[1]: Failed to set up device unit: Invalid argument
Nov 15 07:58:31 thor systemd[1]: Assertion 'u' failed at src/core/unit.c:521, function unit_free(). Aborting.
Nov 15 07:58:31 thor systemd[1]: Caught <ABRT>, dumped core as pid 1441.
Nov 15 07:58:31 thor systemd[1]: Freezing execution.
Nov 15 07:58:31 thor systemd-coredump[1442]: Due to PID 1 having crashed coredump collection will now be turned off.
Nov 15 07:58:33 thor systemd-coredump[1442]: Detected coredump of the journal daemon or PID 1, diverted to /var/lib/systemd/coredump/core.systemd.0.372948aec9c040868c8754001c0dd0b1.1441.1479193111000000.lz4.
Nov 15 07:58:40 thor sudo[1220]: pam_unix(sudo:session): session closed for user root

Executing "systemctl" afterwards:

$ systemctl
Failed to list units: Connection timed out

And my log receives this line:

Nov 15 08:00:07 thor dbus[438]: [system] Failed to activate service 'org.freedesktop.systemd1': timed out

My /tmp and / shouldn't be any short of storage IMHO:

Filesystem                Size  Used Avail Use% Mounted on
dev                       7.9G     0  7.9G   0% /dev
run                       7.9G  1.1M  7.9G   1% /run
/dev/sdc2                 3.7T  2.8T  909G  76% /
tmpfs                     7.9G   68K  7.9G   1% /dev/shm
tmpfs                     7.9G     0  7.9G   0% /sys/fs/cgroup
/dev/sda1                 196M   56M  141M  29% /boot
tmpfs                     7.9G  4.0K  7.9G   1% /tmp
tmpfs                     1.6G  8.0K  1.6G   1% /run/user/1000

jpk · 2016-11-24 18:34:59

Still happening with systemd-232-3.

Nov 24 19:28:25 thor systemd-coredump[30634]: Due to PID 1 having crashed coredump collection will now be turned off.
Nov 24 19:28:26 thor systemd-coredump[30634]: Detected coredump of the journal daemon or PID 1, diverted to /var/lib/systemd/coredump/core.systemd.0.bb26ea685f95430f906e0b38e8728894.30633.1480012105000000.lz4.

System is not bootable!

Also:

$  /usr/lib/systemd/systemd --test --system
Aborted (core dumped)

Should I open a bug report? Any other advice?

Last edited by jpk (2016-11-24 18:41:42)

jpk · 2016-11-24 19:34:08

Okay I found the problem. So I have a multi boot setup with Windows. Somehow the partion names were messed up. One of the symlinks in /dev/disk/by-partlabel/ was named "†††††††††††††††††††††††††††††††††††". This caused systemd to crash.

I fixed this using gdisk/cgdisk and renamed these, using only the standard alphabet a-zA-Z.

Output of blkid:

...
/dev/sda6: LABEL="Work" UUID="E63E05143E04E003" TYPE="ntfs" PARTLABEL="M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- M-bM-^@M- " PARTUUID="6b316bcc-450c-11e5-985b-806e6f6e6963"
...

Alad · 2016-11-24 20:13:49

A bad label shouldn't cause PID1 to crash though... please file a bug upstream:

https://github.com/systemd/systemd/issues

Arch Linux

#1 2016-11-14 07:57:49

[SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#2 2016-11-14 14:26:29

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#3 2016-11-14 14:51:42

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#4 2016-11-15 07:19:26

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#5 2016-11-24 18:34:59

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#6 2016-11-24 19:34:08

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

#7 2016-11-24 20:13:49

Re: [SOLVED] When I try to upgrade to systemd-232 systemd PID=1 crashes

Board footer