Description of problem: Updated to the newest version of preupgrade on F13 system, preupgrade downloaded files and packages normally, but after reboot, it failed to find install.img and ks file. In grub.conf: title Upgrade to Fedora 14 (Laughlin) kernel /upgrade/vmlinuz preupgrade repo=hd::/var/cache/yum/preupgrade stage2=hd:UUID=41ed99ac-a112-4429-8c96-d584dff131a8:/upgrade/install.img ks=hd:UUID=41ed99ac-a112-4429-8c96-d584dff131a8:/upgrade/ks.cfg initrd /upgrade/initrd.img In ks.cfg: # ks.cfg generated by preupgrade lang en_US.UTF-8 keyboard us bootloader --upgrade --location=none upgrade --root-device=UUID=ef4e416a-d1ce-493b-81d1-62fcc987c5d5 reboot %post grubby --remove-kernel=/boot/upgrade/vmlinuz rm -rf /boot/upgrade /var/cache/yum/preupgrade* %end Version-Release number of selected component (if applicable): preupgrade-1.1.4-1.fc13 How reproducible: 100% Actual results: Failed to upgrade after reboot Expected results: I've encountered the similar bugs in preupgrade-1.1.4-1.fc12, but they are fixed in preupgrade-1.1.7-1.fc12, so preupgrade-1.1.7-1.fc13 should also be pushed to F13 system?
Mark as beta blocker due to: The installer must be able to successfully complete an upgrade installation from a clean, fully updated default installation of the previous stable Fedora release, either via preupgrade or by booting to the installer manually
Perhaps #627000 could be linked with this blocker bug, and extend the expectation to upgrades of Fedora-13 systems with encrypted file systems.
*** Bug 627377 has been marked as a duplicate of this bug. ***
With preupgrade-1.1.4-1.fc13.noarch even entering /images instead of /upgrade would not work. So I installed preupgrade-1.1.7-1.fc12.noarch from Koji and preformed a preupgrade again (without removing the cached files from preupgrade 1.1.4-1). Still the installer could find the installation images, but hand entering /images worked! At bit more info in Bug 627377 marked as a duplicate above. The rest of the install went fine.
*** Bug 629245 has been marked as a duplicate of this bug. ***
It turns out that anaconda fails to find install.img and ks.cfg. Reproduced it on F-13 by both preupgrade and preupgrade-cli. I further tested this issue, when it failed to find install.img, I manually input /upgrade/install.img instead of /images/install.img, then anaconda started. I referred to anaconda.log on tty2, oddly, it seems that 'stage2=hd:UUID=41ed99ac-a112-4429-8c96-d584dff131a8:/upgrade/install.img' as well as 'ks=hd:UUID=41ed99ac-a112-4429-8c96-d584dff131a8:/upgrade/ks.cfg' is not honored since both of these are not mentioned under "kernel command line:" and anaconda is blaming for "no ‘stage2=’ given" (See attached /tmp/logs). So is it a anaconda bug?
Created attachment 442578 [details] anaconda.log after manually changing install.img path
Interestingly, if I don't use UUID by changing "UUID=41ed99ac-a112-4429-8c96-d584dff131a8" to "/dev/sda1", it works normally.
This time I use pxeboot images to test kernel arguments like stage2/repo/ks=hd:UUID=:... and it turns out that as long as "UUID=" is used, anaconda seems to ignore the command line and there's no such input displayed under "kernel command line:" in anaconda.log. But if "UUID=" is replaced by "/dev/sda1", it works well. So reassign it against anaconda component.
According to bcl, the fix to address this issue is - http://git.fedorahosted.org/git/?p=anaconda.git;a=commit;h=943447b8058375c3d8e04d2a354e3868359532a4
Discussed at today's blocker review meeting, we accepted this bug as a Beta blocker.