Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 8776 - JFS requiring reboot after fsck.jfs
Summary: JFS requiring reboot after fsck.jfs
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: New packages (show other bugs)
Hardware: x86 Linux
: High normal
Assignee: Brad Cowan (RETIRED)
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-10-05 05:20 UTC by Will Reid
Modified: 2003-02-04 19:42 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Will Reid 2002-10-05 05:20:31 UTC
Hard reboot or shut off damages file system.  fsck.jsf during boot doesn't 
clean the JFS partition.  'fsck.jfs -afv /dev/hdb2' does clean the partition, 
but 'mount /dev/hdb2 /stuff_jfs' locks whatever console called mount.  During 
reboot the system freezes at 'unmounting filesystems'.  Another hard reboot is 
require but does not damage any FS.  After this reboot 'fsck.jfs' is not 
required as 'mount /dev/hdb2 /stuff_jfs' works fine as the JFS partition is 
now truly clean.
Comment 1 Brandon Low (RETIRED) gentoo-dev 2002-10-05 19:47:59 UTC
hmm... I wonder what this is all about... when I have the chance and am home 
I'll poke at some jfs FSs to see what I can find out.
Comment 2 csheridan 2002-10-06 22:14:53 UTC
This is also a problem with ext3. hard reboot or shutdown does not cleanly
unmount the file system, fsck can not be properly called to clean the problem
during reboot, citing an input/output error. fsck must be used off a rescue/boot
disk to clean the problem. Using kernel 2.4.19.
Comment 3 Will Reid 2002-10-07 15:57:48 UTC
Here's why the fsck.jfs would work in my case.  I'm thinking your ext3 
problems requiring a rescue disk are on /.  Well, since fsck can't be run on a 
mounted partition and also can't be run w/o having / mounted a rescue disk 
would have to be used.

p3-1000 root # mount
/dev/hda2 on / type ext3 (rw,noatime)
proc on /proc type proc (rw)
none on /dev type devfs (rw)
tmpfs on /mnt/.init.d type tmpfs (rw,mode=0644,size=1024k)
/dev/hda1 on /boot type ext3 (rw,noatime)
/dev/hdb1 on /stuff_xfs type xfs (rw,noatime)
/dev/hdb2 on /stuff_jfs type jfs (rw,noatime)
/dev/hdb4 on /stuff_ext3 type ext3 (rw,noatime)
/dev/hdb3 on /stuff_reiser type reiserfs (rw)

I have that same problem with my / ext3 partition, but my JFS partitions 
should be fixable w/o a reboot.  

bcowan said they should even be fixed automatically, so I'm adding this...  I 
had to manually fsck the reiser partition to mount it yesterday.
Comment 4 Brad Cowan (RETIRED) gentoo-dev 2002-10-07 17:36:08 UTC
I checked into this last night and it seems to be a bug with havinf XFS and JFS
in the lolo sources...I was able to reproduce the error, but I lost data. I
switched to XFS sources and XFS and JFS seem to play nice together.
Comment 5 Brandon Low (RETIRED) gentoo-dev 2002-10-07 17:45:31 UTC
I would like to extend my personal thanks to bcowan for being brave enough to 
test this on his precious data.  There is a reason that XFS is not patched into 
lolo-sources/gentoo-sources by default... it might break things... therefore if 
you plan to use XFS and JFS use xfs-sources where everything is designed around 
XFS...