From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2) Gecko/20040803 Description of problem: Installation of FC2 exits abnormally on a Dual Pentium Pro PR440FX SCSI only system when a FAT32 formatted media is present in the installed Fujitsu MCE 3064SS MO drive. Version-Release number of selected component (if applicable): anaconda-10.0-5 How reproducible: Always Steps to Reproduce: 1. Insert FAT32 formatted media into MO drive. 2. Boot system from FC2 boot CD. 3. Pursue installation until installer exits. Actual Results: Installer issues "install exited abnormally" and terminates the installation procedure. Expected Results: Installation procedure should continue without aborting. Additional info: The boot CD drive is mapped onto /dev/scd0, the MO drive onto /dev/sdb. Application of available install options like "text", "noprobe", etc. and combinations hereof do not change the unfavourable outcome. Switching to /dev/tty4 reveals complaints about a non-ISO9660 media present in /dev/sdb. Removal of the media from the MO drive hereafter allowed to install FC2 without any further problem. No failure of this kind was observed during a prior installation of FC3 test 1 although the MO disk was present then, too.
If you switch to tty2 and run 'python -c "import kudzu; print kudzu.probe(kudzu.CLASS_CDROM, kudzu.BUS_UNSPEC, 0)"' what do you get as output?
Ok, switching between virtual consoles back and forth is still possible after the termination notification and rebooting via ctrl-alt-del, too. But after switching to tty2, the keyboard seems to be stuck. Well, at least, there is no alphnumerical output on the screen. Any clue how to get around this problem?
Does this still happen with fc4 test?
No, because as mentioned in my original bug report, this issue was already gone as of FC3T1.
Oh, even better. Almost certainly an artifact of the early-ish 2.6 kernel shipped in fc2 then.