Greetings, I'm having an issue with mkinitrd during autoyast installation (see attached screenshots for the error message). I'm installing on virtual machines, using an image of a bought 11.2 DVD to boot. The rest of the installation happens over a local HTTP-based repository[0], the autoyast files are on the same host. Weird thing is, the problem only appears during autoyast installation and not when doing a manual install from the same DVD image. (So the image should be fine.) The autoyast profile, on the other hand, is based on one that worked fine with 11.1 and broke the instant we installed 11.2 from it[1]. (So the profile should be fine.) I've been triple-and-more-checking it for errors but have found none so far (I will attach a made-anonymous version around tomorrow, I can't access the machines right now). I haven't found any clue in the log files as to where the problem is. I've been using "confirm", have switched to the second console when the error brings the install to a screeching halt, looked at the end of the log files, nothing. No errors at all. It seems the problem really is in the mkinitrd rpm (or the arguments it is being called with); however, it works fine after installing (IIRC, the version during and after install are the same), and even the autoinstalled files work fine after punching away those error messages. (Still, it's a nasty showstopper for an unattended installation.) I haven't found any bugs resembling the issue and couldn't find anybody on the internet having this problem (except some less-than-useful comments[2]). Is this a known issue? What can I do to isolate the problem? I'm thankful for ANY suggestion at this point. Otherwise my next, desperate attempt might be to create a postpartitioning-script or chroot-script to patch debug output into the mkinitrd-related shellscripts. tty, 686f6c6d 0: IIRC, I assumed it could be a local problem and verified that the versions are the same as in the official repos. I will check that again. 1: Of course, what I mean is we slighty modified our existing profile. After creating a fresh reference profile, we merged our old settings into the new profile. With a different incarnation with no differences in the relevant - bootloader etc. - sections whatsoever (IIRC, will check again), the problem does not occur. There, however, we boot over PXE, a scenario that is not practical with the setup where the problem occurs. 2: http://news.opensuse.org/2009/10/15/opensuse-11-2-on-its-way-to-become-final... (search the page for "illegal option" or "Kazuya")
