[reiserfs-list] SCSI error, kernel panic ...

[Eoin Verling]

Hi,

I'm running linux kernel 2.4.6, reiserfs version 3.6.25

I've just recently seen the kernel panic, that sends the machine into an unusable 
state, my only option is to reboot. It seems like the reiserfs partition is corrupt. I 
can mount the drive and list all the files, I can even copy some of the files to 
another partition, but when I try to copy some of the files, I get this kernel panic.

When I mount the partition first, there is the following error

reiserfs: checking trans log (dev 08:0a)
scsi0: ERROR on channel 0, id 0, lun 0, CDB: Read (10) 00 01 98 a8 8e 00 00 08 00
Info fld=0x198a893, Current sd08:0a: sense key Medium Error
Additional sense indicates Unrecovered read error
I/O error: dev 08:0a, sector 65680
Using tea hash to sort names
reiserfs: using 3.5.x disk format
ReiserFS version 3.6.25

Below is the kernel panic error

Sep 14 11:20:38 kerry kernel: Unable to handle kernel NULL pointer dereference at 
virtual address 00
Sep 14 11:20:38 kerry kernel:  printing eip:
Sep 14 11:20:38 kerry kernel: c01935f5
Sep 14 11:20:38 kerry kernel: *pde = 
Sep 14 11:20:38 kerry kernel: Oops: 
Sep 14 11:20:38 kerry kernel: CPU:0
Sep 14 11:20:38 kerry kernel: EIP:0010:[c01935f5]
Sep 14 11:20:38 kerry kernel: EFLAGS: 00010202
Sep 14 11:20:38 kerry kernel: eax: e084e000   ebx: dfa79400   ecx: 0001   edx: 

Sep 14 11:20:38 kerry kernel: esi: 000b   edi: 000f   ebp: e084e108   esp: 
c18b3edc
Sep 14 11:20:38 kerry kernel: ds: 0018   es: 0018   ss: 0018
Sep 14 11:20:38 kerry kernel: Process kupdated (pid: 6, stackpage=c18b3000)
Sep 14 11:20:38 kerry kernel: Stack: 0003 0108 0108 c0193b62 dfa79400 
000f 000b 
Sep 14 11:20:38 kerry kernel:0108 0108 0108 e084e148  
0008  c019
Sep 14 11:20:38 kerry kernel:dfa79400 e084e108 0001 c18b3f98 c18b2332 
c028af56 dfa79400 
Sep 14 11:20:38 kerry kernel: Call Trace: [c0193b62] [c019712e] [c0195dff] 
[c0187768] [c01349f3
Sep 14 11:20:38 kerry kernel:[c01054cc]
Sep 14 11:20:38 kerry kernel:
Sep 14 11:20:38 kerry kernel: Code: f6 42 18 04 74 2d 8b 42 18 a8 04 74 09 52 e8 58 d1 
f9 ff 83
Sep 14 11:46:08 kerry -- MARK --


So, my last option now is to run reiserfsck --rebuild-tree ... I tried reiserfsck by 
itself, and reiserfsck --skip-journal ... and both of those die without completing (or 
fixing)

Is --rebuild-tree a safe option? It's like there are bad blocks on the drive, is there 
anyway of moving data around these bad blocks?

cheers
E

[reiserfs-list] undelete files ...

[Eoin Verling]

Hi,

That fatal question ... I've deleted something on a reiserFS partition, and I'm trying 
to recover it.

The details, SuSE 7.1, 2.4.6 kernel (reiserFS that comes with distro suse 7.1)

I have a large partition with the deleted file, I brought the machine to single-user 
and unmounted the partition in question as soon as I spotted my idiotic mistake! I 
copied the partition to another partition (same size, using 'dd'), so I'm trying to 
recover from the copied partition.

I've tried:-

# reiserfsck --rebuild-tree /dev/partition

.. to no avail. It completed, but my deleted files were not there.

I'm currently trying:-

# reiserfsck --rebuild-tree --scan-whole-partition /dev/partition

and I@m waiting to see what that tells me.

If what I'm attempting is possible, I'ld appreciate you letting me know. 

I'm in a fix with this one.
cheers
E

- --  _
Eoin Verling_/ \_   5300 Cork Airport Business Park,
SysAdmin   / \_/ \  Kinsale Rd., Cork, Ireland.
Comnitel Technologies  \_/ \_/  Ph: +353 21 7306008
[EMAIL PROTECTED]\_/Fax: +353 21 7306024