lkml.org 
[lkml]   [2005]   [May]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subject[2.6.11]: Bugs in handle striped LVM volumes ?

I just on move from one ATA disk to four SCSI.
During this move I create two RAID1 devices and on top of this devices I
create neccessary LVM volumes.
After create volumes, make fs structures I'm copy all resources from old
ATA disk to new volumes, mounting new volumes in /mnt/new I start
"cp -aux / /mnt/new". Copying was finished _without problems_.
After this I try run "chroot /mnt/new" for adjust configurarion in copied
resources but during this I was kicked with error message about errors in
/bin/bash :>
After this was generated tons of messages to syslog like:

May 18 17:55:26 test1 kernel: attempt to access beyond end of device
May 18 17:55:26 test1 kernel: dm-2: rw=0, want=11331644016, limit=31457280
May 18 17:55:26 test1 kernel: attempt to access beyond end of device
May 18 17:55:26 test1 kernel: dm-2: rw=0, want=14546609024, limit=31457280
May 18 17:55:26 test1 kernel: attempt to access beyond end of device
May 18 17:55:26 test1 kernel: dm-2: rw=0, want=4311810520, limit=31457280
May 18 17:55:26 test1 kernel: attempt to access beyond end of device
[..]
May 18 17:56:31 test1 kernel: EXT3-fs error (device dm-2): ext3_free_blocks: Freeing blocks not in datazone - block = 1416455501, count = 1
May 18 17:56:31 test1 kernel: Aborting journal on device dm-2.
May 18 17:56:31 test1 kernel: EXT3-fs error (device dm-2): ext3_free_blocks: Freeing blocks not in datazone - block = 1818326127, count = 1
May 18 17:56:31 test1 kernel: ext3_abort called.
May 18 17:56:31 test1 kernel: EXT3-fs error (device dm-2): ext3_journal_start_sb: Detected aborted journal
May 18 17:56:31 test1 kernel: Remounting filesystem read-only
May 18 17:56:31 test1 kernel: EXT3-fs error (device dm-2) in start_transaction: Journal has aborted

For above I'm using striped volume and I suspect this as root of problems
so I repeat all but on volumes without stripping (remove stripped dm-2
volume and on the same place was created not stripped)
and in this case all was perfect

My current MD/LVM configuration (without striping):

[root@test1 /]# cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb2[1] sda2[0]
1959808 blocks [2/2] [UU]

md2 : active raid1 sdb3[1] sda3[0]
33776576 blocks [2/2] [UU]

md3 : active raid1 sdd3[1] sdc3[0]
69762176 blocks [2/2] [UU]

md1 : active raid1 sdd2[1] sdc2[0]
1959808 blocks [2/2] [UU]

lvm> pvs
PV VG Fmt Attr PSize PFree
/dev/hda2 VolGroup00 lvm2 a- 74,41G 96,00M
/dev/md2 test1 lvm2 a- 32,21G 0
/dev/md3 test1 lvm2 a- 66,53G 756,00M
lvm> lvdisplay -v test1
Using logical volume(s) on command line
--- Logical volume ---
LV Name /dev/test1/root
VG Name test1
LV UUID 6K8754-j5dv-PAV9-FMkY-LMg6-fjUo-7nIRwf
LV Write Access read/write
LV Status available
# open 1
LV Size 15,00 GB
Current LE 3840
Segments 2
Allocation inherit
Read ahead sectors 0
Block device 253:2

--- Logical volume ---
LV Name /dev/test1/home
VG Name test1
LV UUID dSrvjz-sAnv-zGkW-x0xN-S27U-nlvu-lcAC8t
LV Write Access read/write
LV Status available
# open 1
LV Size 83,00 GB
Current LE 21248
Segments 2
Allocation inherit
Read ahead sectors 0
Block device 253:3

Striped volume test1/root was created using "lvcreate test1 -n root -L 15G -i2 -I8"
(above was created by simple "lvcreate test1 -n root -L 15G").

Kernel version it is 2.6.11 (from curren Fedora devel .. but I don't see
in src.rpm any LVM/DM changes compare to vanilla so probably it will be
repeatable on vanilla).

Anyone is using LVM2 stripping on 2.6.11 ?
Anyone known something about bugs in LVM stripping code ?

kloczek
--
-----------------------------------------------------------
*Ludzie nie mają problemów, tylko sobie sami je stwarzają*
-----------------------------------------------------------
Tomasz Kłoczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek@rudy.mif.pg.gda.pl*
\
 
 \ /
  Last update: 2005-05-18 22:04    [W:0.031 / U:1.376 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site