I think this bug needs more attention. I just had a friend (newbie to Linux, FWIW) who was doing a Jaunty -> Karmic upgrade, but unfortunately interrupted the upgrade for some reason.
After that, the kernel packages were upgraded, but libudev0 not, and after regenerating the initrd, the old *broken* udev libs were used, giving just the error mentioned, but while running the "mountall" command (from initrd?). The problem was exactly the same as described in http://duq.ca/duqamuq/?p=331 , but I took another approach to fix the system, which was to: 1) append "init=/bin/bash rw" to the kernel parameters using the grub shell 2) apt-get install libudev0 (the package was already downloaded, so I needed no network connection) 3) reboot 4) go to a command-line prompt, and continue upgrade with: apt-get dist-upgrade Obviously this is non-trivial for newbies, and the solution (I think) would be to have libudev0 upgraded prior to the kernel packages during an upgrade. -- libudev0 needs updated shlibs (unstable API) https://bugs.launchpad.net/bugs/427672 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
