https://bugzilla.novell.com/show_bug.cgi?id=223770 ------- Comment #13 from markus.kossmann@schlundmail.de 2006-11-28 07:10 MST ------- (In reply to comment #8)
What seems to happen is.. 1. before finishing with install cd 1, some command is issued which causes failures on some drives. dunno what yet. as the device is still open, tray is locked, so eject button doesn't work. 2. user clicks the eject button. ioctl command is issued to unlock and eject the tray but the device is still opened by the process which issued the failed command. 3. user changes cd and tell the installer to continue. because the failed process is still holding on to the cdrom device, the device is not revalidated and thus the device's size remains the same as cd1's size which is shorter than cd2. So, the result is, reading beyond end of device and media check failure for cd's larger than cd1. The solution: find who issues the failing command and fix it such that it doesn't issue such command && it closes fd properly even after failure. Would this also explain, that running md5sum from a shell gives good result for CD1 ( which should be still opened) but md5sum on CD2 results with a bad checksum ?
-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.