Kernel-2.6.2-1.87 locking up computer attempting reiserfs mount

[Jim Cornette]

Hoyt wrote:
> On Wednesday 18 February 2004 11:53 pm, Jim Cornette wrote:
> 
>>/var/log/messages excerpt
>>kernel: found reiserfs format "3.6" with standard journal
>>kernel: reiserfs:warning: CONFIG_REISERFS_CHECK is set ON
>>kernel: reiserfs:warning: - it is slow mode for debugging.
>>kernel: Reiserfs journal params: device hda2, size 8192, journal first
>>block 18, max trans len 1024, max batch 900, max commit age 30, max
>>trans age 30
>>kernel: reiserfs: checking transaction log (hda2) for (hda2)
>>Feb 18 23:22:31 ... kernel: Using r5 hash to sort names
> 
> 
> i have the identical experience. Removing the reiserfs partitions allows the 
> kernel to boot. I even re-compiled and removed CONFIG_REISERFS_CHECK, but no 
> help. 
> 
>>From /var/log/messages:
> 
> Feb 20 08:41:52 titan kernel: found reiserfs format "3.6" with standard 
> journal
> Feb 20 08:41:57 titan kernel: Reiserfs journal params: device hda7, size 8192, 
> journal first block 18, max trans len 1024, max batch 900, max commit age 30, 
> max trans age 30
> Feb 20 08:41:57 titan kernel: reiserfs: checking transaction log (hda7) for 
> (hda7)
> Feb 20 08:41:57 titan kernel: Using r5 hash to sort names
> 
> and then it freezes.
> 
> 

I've tried a few things with reiserfs to get it to mount under 
Kernel-2.6.2-1.87.

I tried turning off selinux, no help. Mounting the reiserfs volume after 
boot. (Result, it loads reiserfs/0 but the computer cannot be brought 
down to a proper halt.

The closest that I have come to a clean shutdown was with telinit 6. 
This brought the system down to a state where I was not asked to check 
the volume integrity on startup.

The strange thing is that the computer seems to act normally after a 
mount of the reiserfs volume was attempted and seems to lock. I could do 
about everything except shutdown the system and could not see any 
indication that the voume was mounted.

When I went back to Kernel-2.6.2-1.81 and tried to manually mount the 
volume from a terminal, with the same command. The reiserfs would mount 
properly and I could access the files on the reiserfs volume.

I don't know what it is that is causing the reiserfs error, but it seems 
to be present in the Kernel-2.6.2-1.85 and above.

I booted up the reiserfs volume that contains my fedora 1 installation 
and it appears to be working as it was before the Kernel-2.6.2-1.85 
release of the kernel. No new file corruption that I noticed.

If I try to mount the reiserfs volume from a gnome-terminal, I get no 
message feedback. If I try mounting the volume from a regular mingetty 
terminal, I get a similar dialog before the freeze printed to the screen.

Is it that mount needs updated to handle later reiserfs volumes, 
something flawed in a recent kernel change or Selinux or some other 
factor will not work for reiser on kernel Kernel-2.6.2-1.81 or later.

Jim