systemd version the issue has been seen with

254.3

Used distribution

NixOS unstable

Linux kernel version used

6.1.47

CPU architectures issue was seen on

x86_64

Component

systemd-journald

Expected behaviour you didn't see

I would expect this to fail but a better error message would be nice.

Unexpected behaviour you saw

When systemd-journald started the following assertion happened:

Assertion 'path_is_absolute(p)' failed at src/basic/chase.c:528, function chase(). Aborting.

then systemd-coredump triggered

Steps to reproduce the problem

The issue happened because multiple zpools mounted different datasets at /. After removing the duplicated mounts for / the issue went away. We could boot the system by adding single to the grub boot options.

I don't have access to the system right now but zfs list look similar to my mock:

 ➜ zfs list
NAME              USED  AVAIL     REFER  MOUNTPOINT
rpool             800G   900G       96K  /
rpool/nixos       800G   900G      600G  none
rpool/nixos/data  800G   900G      600G  legacy
rpool/nixos/nix   200G   900G      200G  /nix
ppool             800G   900G       96K  /
ppool/nixos       800G   900G      600G  none
ppool/nixos/data  200G   000G      200G  legacy

So I think to reproduce the following condition must be met:
Mount multiple file systems early at boot to / and confuse the chase code which fails to start systemd-journald.

Additional program output to the terminal or log subsystem illustrating the issue