Description of problem: I did reinstall my own server, previously setup with software RAID and lvm (except for /boot). I fully reinstall it, all disk, RAID and lvm were recognized and the install went fine (minimal install). But the reboot failed, /dev/VG/LV didn't exists. I had to reboot from CD in rescue mode, chroot into the install and run "dracut -f". After this, everything went fine. Steps to Reproduce: Install a computer having two partition under RAID and an lvm over it, then reboot failed. cat /proc/mdstat: Personalities : [raid1] md0 : active raid1 sda2[0] sdb2[1] 1919680 blocks [2/2] [UU] md1 : active raid1 sda5[1] sdb5[0] 730515584 blocks [2/2] [UU] md1 is the lvm rootfs on / type rootfs (rw) /dev/mapper/virgo-root on / type ext4 (rw,relatime,user_xattr,barrier=1,data=ordered) /dev/sda1 on /boot type ext4 (rw,relatime,user_xattr,barrier=1,data=ordered) /dev/mapper/temp-tmp on /tmp type ext4 (rw,relatime,user_xattr,barrier=1,data=ordered) /dev/mapper/temp-nettmp on /mnt/nettmp type xfs (rw,relatime,attr2,noquota) /dev/mapper/virgo-var on /var type ext4 (rw,relatime,user_xattr,barrier=1,data=ordered) /dev/mapper/virgo-users on /mnt/users type xfs (rw,relatime,attr2,usrquota) /dev/mapper/virgo-data on /mnt/data type xfs (rw,relatime,attr2,noquota) /dev/mapper/virgo-projects on /mnt/projects type xfs (rw,relatime,attr2,noquota) /dev/mapper/temp-mirrors on /mnt/mirrors type xfs (rw,relatime,attr2,noquota)
CC: (none) => mageia, thierry.vignaud
Can you attach your /root/drakx/bug.gz?
Keywords: (none) => NEEDINFO
Created attachment 2741 [details] Install report
Did you kept your old initrd? In order to compare lsinitrd on both?
This is likely the theme bug where the initrd is created too early in the install process. Can you test and see if the following work around works for you. 1. Do the install as normal. 2. When complete, and on the screen where you enter the root password, flip to tty2 (ctl+alt+f2) and type: rm /mnt/boot/initrd-3*.img 3. Flip back to the installer (alt+f7) and complete the process. Does this produce better results?
Unfortunately, I cannot test again on this computer: it is my own server (mail server, proxy, gateway, dhcpd, storage, etc). It is easy to test using VirtualBox with a VM having two disk.
Can you do that test please?
This affects LVM-only installations. I ran into it on a new clean install. I don't know if non-LVM RAID installations are affected. Next time I install a VM or physical I will test Colin's workaround.
CC: (none) => bgmilne
(In reply to comment #5) > Unfortunately, I cannot test again on this computer: it is my own server (mail > server, proxy, gateway, dhcpd, storage, etc). > > It is easy to test using VirtualBox with a VM having two disk. (In reply to comment #6) > Can you do that test please? (In reply to comment #7) > This affects LVM-only installations. I ran into it on a new clean install. > > I don't know if non-LVM RAID installations are affected. > > Next time I install a VM or physical I will test Colin's workaround. @ Olivier or Buchan did one of you get around to testing it?
CC: (none) => marja11
(In reply to comment #8) > @ Olivier or Buchan > > did one of you get around to testing it? ping?
Can anyone confirm this bug is still valid for Megeia 2 or Cauldron. Roelof
CC: (none) => r.wobben
1. Changing version to cauldron since we can't fix installer for released versions (that's what alphas and betas are for) 2. Is this bug still valid with latest installer?
CC: (none) => stormiVersion: 2 => Cauldron
Unanswered too long. If seen with current installer, please reopen and set whiteboard to '5beta1' (or higher), removing 'needinfo. Thanks.
Status: NEW => RESOLVEDResolution: (none) => OLD