Red Hat Bugzilla – Bug 260041
Same with 2.6.23-0.169.rc5.git1.fc8
Last modified: 2007-11-30 17:12:14 EST
The old kernel starts without troubles, the new one can not find LVM Installed on VMwareServer running od Fedora 7 See attached Screenshot.... [root@fedora-test ~]# rpm -qa | grep kernel yum-kernel-module-1.1.6-1.fc8 kernel-headers-2.6.23-0.142.rc3.git10.fc8 kernel-devel-2.6.23-0.136.rc3.git7.fc8 kernel-2.6.23-0.142.rc3.git10.fc8 kernel-devel-2.6.23-0.142.rc3.git10.fc8 kernel-2.6.23-0.136.rc3.git7.fc8
Created attachment 176301 [details] Screenshot vom booting VM
Can you give kernels between 0.136 and 0.142 a try to narrow down where the issue first crops up? Available here: http://koji.fedoraproject.org/packages/kernel/2.6.23/
139, 140 and 142 have the same problem hmm its possible that mkinitrd makes the fuck... I have removed kernel and kernel-devel-packages that are not booting and installed the both in all two other version with "yum --nogpgcheck localinstall" ____ If it helps you there is an tar-image available from the vmware-setup My connection has 2Mbit Upload, so it may take a while for 7.3 GB ftp://fedora:fedora@local.rhsoft.net/Fedora.tar System starts in "init 3" and prefers kde for waiting on kde4-snaps There is a normal user "harry" with password "linuxuser" and all sudo-permissions and root has the same password
142 doesn't find LVM on non-VMWare laptop 139 works 100%
OK, seems that the problem with 139 began on vmware and now the same happens on "real-hosts"... I hope VMware-support will be better in many cases - Installing FC7/FC8-test1 on SCSI-Drive under VMware do not work, with IDE-Emulation it hangs without comments if you create your own partitions. Its a bit painful because i have not enough computers for playing with devel-snapshots on real hardware :-)
Hrm... So one report of 139 working fine, one of it not working... I'm certainly inclined to think mkinitrd could well be at least partially responsible... I've got 142 booting on multiple systems of my own right now, but I don't think I updated mkinitrd on them in the past few days.
The timestamps are not the same as on your bugzilla-server but on the second output you see coming the non-botting kernel 142 and mkinitrd in one action [root@fedora-test ~]# cat /var/log/messages | grep mkinitrd Aug 28 14:57:28 fedora-test yum: Updated: mkinitrd - 6.0.12-1.fc8.i386#000 Aug 28 14:57:27 fedora-test yum: Updated: nash - 6.0.12-1.fc8.i386#000 Aug 28 14:57:28 fedora-test yum: Updated: mkinitrd - 6.0.12-1.fc8.i386#000 Aug 28 14:57:41 fedora-test yum: Updated: vim-common - 2:7.1.87-1.fc8.i386#000 Aug 28 14:57:43 fedora-test yum: Updated: perl-DBI - 1.58-2.fc8.i386#000 Aug 28 14:57:44 fedora-test yum: Installed: beecrypt-devel - 4.1.2-13.i386#000 Aug 28 14:57:45 fedora-test yum: Updated: vim-enhanced - 2:7.1.87-1.fc8.i386#000 Aug 28 14:58:29 fedora-test yum: Installed: kernel - 2.6.23-0.142.rc3.git10.fc8.i686#000
With the last kernel-update it works again, but cant say if kernel himself or the new mkinitrd did the trick - please look for better vmware-support on fedora in future especially installing from dvd-image :-) Installing: kernel i686 2.6.23-0.149.rc4.fc8 development 16 M kernel-devel i686 2.6.23-0.149.rc4.fc8 development 4.6 M Updating: kernel-headers i386 2.6.23-0.149.rc4.fc8 development 660 k
If I'm not mistaken, this wasn't a vmware-specific issue, it actually rendered some physical machines unbootable as well.
Hmm 2.6.23-0.169.rc5.git1.fc8 Says booting "ABCDE...", restarts the VM says the same, restarts the VM the whole day :-( 2.6.23-0.169.rc5.git1.fc8 has no problems Same VMware-Installation I hope its ok tu use the same bug-report instead of opening an new....
Created attachment 190861 [details] VMware-Screenshot one second before reboot
That last one's actually a dupe of another bug, which escapes me. Was affecting physical machines as well, and was fixed in a later kernel.