[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [fedora-list] Uptate: F11 x86_64 problems -- my disks are getting locked out

> > > On Mon, 06 Jul 2009 12:15:58 -0400
> > > Rich Mahn <rich lat com> wrote:
> > > 
> > > I did a clean install of F11 x86_64 on a system that also has F10
> > > i686, F8, and others for a multi-boot situation.  F11 is working
> > > fairly well except for one problem.  I can't mount one of my disks.

> > snip

> > > Does anybody have any ideas what might be causing this and how to fix
> > > it?
> > > 

> > Run blkid as root and make sure all your disks have the correct UUID
> > in /etc/fstab.  You should be mounting them using UUID instead of
> > device name or label.

> Thanks for telling me about blkid.  A real useful tool.

> > Here is a sample from my fstab:

> > #/dev/sda1
> > UUID=c6e94dd5-0cc3-4940-ac15-cb0ea856fd9a /boot ext3 defaults  1 2 
> > #/dev/sda5 
> > UUID=c94b5547-e098-45ca-8776-cce0e0cd6bac / ext4 defaults      1 1

> > Maybe you are already doing this. 

> > Compare the F10 and F11 /etc/fstab for these disks.  Make sure the
> > mount points are unique.

> > You could check for bugs like this in bugzilla.

> > http://bugzilla.redhat.com

> Since I mount these drives manually, it's not the fstab--at least
> I don't think it is.

> I suspect udev or something like that.  The kernel recognizes
> the disks and paritions okay, but the device nodes for the partitions
> are never created.  They can be forced by appropriate fdisk/hdparm
> or whatever that re-reads the label.  But even then you can't mount
> them.  You can, however, dumpe2fs, dd, and many other useful things.

> I think I started to run parted on the disk and aborted when I noticed
> I had used the wrong disk.  In the current state, F10 is fine and F11
> isn't.  I tried to install F11 i686 to see if it was an X86_64 problem,
> but now I can't install F11.  It crashes when it tries to find disks.
> It looks like maybe it thinks the disk in question is part of a raid
> or something.

> I can get rid of the problem by zeroing the disk.  I'm going to see if
> I can create the problem with parted.

I just found something that probably indicates the problem to those who
can understand these things.  The command

 udevadm info --query=env --name=/dev/sdc

has these extra line that a similar query on any other disk does not have:


Anybody have any idea what this means?


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]