Hide Forgot
Created attachment 526825 [details] abrt=description The installer from the Fedora 16 x86_64 Live CD fails at the "Examining Storage Devices" stage. The machine currently has a Fedora 14 system installed on a raid5 partition; upgraded in place from earlier installs. This looks similar to bug 731177, and indeed report_Bugzilla wouldn't let me submit this issue because the anaconda backtrace is a duplicate. However, that issue was resolved in selinux-policy -28, and the Live CD has: [liveuser@localhost ~]$ yum info selinux-policy Loaded plugins: langpacks, presto, refresh-packagekit Installed Packages Name : selinux-policy Arch : noarch Version : 3.10.0 Release : 32.fc16 Size : 8.6 M Repo : installed From repo : koji-override-0 And I see no mdadm-related avc denial in /var/log/messages.
Created attachment 526826 [details] abrt-anaconda-log
Created attachment 526829 [details] /var/log/messages
There's something wrong with md127. See the following from the syslog: Oct 6 23:13:25 localhost kernel: [ 43.814105] md/raid:md127: device sdb2 operational as raid disk 1 Oct 6 23:13:25 localhost kernel: [ 43.814503] md/raid:md127: allocated 3228kB Oct 6 23:13:25 localhost kernel: [ 43.814579] md/raid:md127: not enough operational devices (2/3 failed) Oct 6 23:13:25 localhost kernel: [ 43.814899] md/raid:md127: failed to run raid set. Oct 6 23:13:25 localhost kernel: [ 43.814901] md: pers->run() failed ... Anaconda does not support the use of degraded md arrays. Because of the md udev rules broken arrays still appear in the system in spite of being unusable. The only easy way to get around this would be to tell anaconda to ignore sdb completely or to wipe the raid signature from sdb2. I don't know if either of these is feasible for you.
Good catch! I can confirm in that I was able to proceed with the install after resyncing the raid device. Would have been nice if anaconda had said so, instead of crashing through. :/
*** Bug 717082 has been marked as a duplicate of this bug. ***
Installer exceptions aren't bugs?