Bug 160986 - Anaconda cannot handle "LABEL=foo" style fstab entries
Anaconda cannot handle "LABEL=foo" style fstab entries
Product: Fedora
Classification: Fedora
Component: anaconda (Show other bugs)
All Linux
medium Severity high
: ---
: ---
Assigned To: Chris Lumens
Mike McLean
Depends On:
  Show dependency treegraph
Reported: 2005-06-19 09:44 EDT by david.hagood
Modified: 2007-11-30 17:11 EST (History)
0 users

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2005-06-21 17:05:41 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)
The offending fstab entry (856 bytes, text/plain)
2005-06-19 09:44 EDT, david.hagood
no flags Details

  None (edit)
Description david.hagood 2005-06-19 09:44:11 EDT
Description of problem:
Anaconda cannot parse fstabs that contain "LABEL=foo" style mounts

Version-Release number of selected component (if applicable):

How reproducible:
Every time.

Steps to Reproduce:
1. Create a system with "LABEL=foo" style mounts for root file system.
2. Try to do an upgrade with FC4 anaconda

Actual results:
Anacoda barfs with "One of your filesystem is in a consistant state" error, and
offers no choice but to reboot.

Expected results:
Anaconda should figure out the mount, or at least allow me to hand-mount the
filesystem on another console and try again.

Additional info:
Anaconda has had this problem since RH8.
Comment 1 david.hagood 2005-06-19 09:44:11 EDT
Created attachment 115662 [details]
The offending fstab entry
Comment 2 Chris Lumens 2005-06-20 16:17:23 EDT
This appears to be working for me on an FC3 -> FC4 upgrade.  Can you switch over
to tty3 (ctrl-alt-f3) and see if there are any obvious error messages? 
Additionally, can you attach the /tmp/anaconda.log file to this report?
Comment 3 david.hagood 2005-06-20 19:32:08 EDT
OK, the info on VC3 is:
Moving (1) to step welcome
moving (1) to step findrootparts
isys.py:mount()- going to mount /tmp/hda1 on /mnt/sysimage
isys.py:mount()- going to mount /tmp/hda6 on /mnt/sysimage
isys.py:mount()- going to mount /tmp/sda1 on /mnt/sysimage
isys.py:mount()- going to mount /tmp/sda2 on /mnt/sysimage
moving (1) to step findinstall
moving (1) to step partitionobjinit
error munging swap label: [Errno 22] Invalid argument
error reading jfs label on /tmp/disk: [Errno 2] Invalid argument
moving (1) to step upgrademount
going to mount hda6 on /mnt/sysimage as ext3
isys.py:mount()- going to mount /tmp/hda6 on /mnt/sysimage
error mungin swap label: [Errno 22] Invalid argument
error reading jfs label on /tmp/disk: [Errno 22] Invalid argument
trying to mount sda2 on /mnt/work

And from there out there are no unusual looking messages.

Comment 4 Chris Lumens 2005-06-21 11:17:43 EDT
Workaround - change your fstab entries to specify a filesystem instead of saying
"auto" for those labeled entries.
Comment 5 david.hagood 2005-06-21 18:10:43 EDT
OK, so, that's it? Just a work-around, not an effort to actually fix the problem?

Or does this mean "closed, will be fixed in Rawhide"?

The whole idea of having things like label mounts and fstype=auto is to allow
the system to better handle changes without needed to tweak a bunch of files.

If Anaconda cannot deal with the same setups that mount does, then Anaconda is
Comment 6 Chris Lumens 2005-06-22 09:43:05 EDT
No, the workaround was for your immediate benefit so you can do whatever you
need to do while we work on another anaconda release.  The "closed - rawhide"
means I have committed a fix for it.  From the CVS commit mailing list:

 2005-06-21  Chris Lumens  <clumens@redhat.com>

+       * fsset.py (AutoFileSystem):  Add mount and umount methods so entries
+       in /etc/fstab marked as "auto" can still be mounted on upgrade
+       (#160986).

Note You need to log in before you can comment on or make changes to this bug.