On Mon, 13 Feb 2006, dann frazier wrote: > Yes, udev postinstall did get executed later. If a udev upgrade is > supposed to trigger initrd regeneration (it looks like it might), then > this is probably not an upgrade path issue. However, it looks like > initrd regeneration failed in my case.
udev is using initramfs-tools hooks so needs to to update the initramfs if initramfs-tools is installed and configured. evms adds also his hooks. needs still to be done for lvm2 and mdadm. as currently you have nice races between initramfs generation and possible later lvm2 installation + root creation that will then miss the dm/md stuff in the initramfs. > Here's the upgrade log: > http://dannf.org/bugs/352705/upgrade.log indeed there is a bug in update-initramfs triggered due to the new kernel upgrade seen in the udev postinst: /boot/initrd.img-2.6.15-1-mckinley was been altered. Cannot update. -- maks -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]