Cannot mount directory after crash

[kenny chia]

Hi all
I have a problem which caused my DS2300 disk to go offline. Not exactly sure what caused it
But I manage to online them and run fsck for the following lvols

/dev/vg01/lvol1
/dev/vg01/lvol2
/dev/vg01/lvol3

which is all in my DS2300 server.

I could mount my directories to lvol1 and lvol2 but when I mount lvol3 I got the following error

# mount /users2
vxfs mount: /dev/vg01/lvol3 no such device

What could be the cause of the problem?

[Luk Vandenbussche]

Hi,

Can you send the output of vgdisplay -v vg01

[Senthil Kumar .A_1]

Hi Kenny,

Please provide us the outputs of the following commands,

vgdisplay -v vg01
lvdisplay /dev/vg01/lvol3
ll /dev/vg01

regards,
Senthil Kumar .A

[kenny chia]

Hi
Attached is the output

[Luk Vandenbussche]

Everything looks fine

Try

fsck -F vxfs /dev/vg01/rlvol3
mount /dev/vg01/lvol1 /users2

[kenny chia]

There seems to be no problem for my attached results.
At the bottom is my dmesg results. /users2 seems to be badly affected and the situation seems hopeless. Should I do mkfs_vxfs instead of trying to recover?

vxfs: mesg 055: vx_metaioerr - /dev/vg01/lvol3 file system meta data write error
vxfs: mesg 056: vx_dataioerr - /dev/vg01/lvol1 file system file data read error
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 008: vx_direrr - /users2 file system inode 22 block 64323 error 22
vxfs: mesg 031: vx_disable - /dev/vg01/lvol3 file system disabled
vxfs: mesg 031: vx_disable - /dev/vg01/lvol1 file system disabled
vxfs: mesg 049: vx_esum_bad - /dev/vg01/lvol2 file system extent allocation unit summary number 5 marked bad

---------------- End of dmesg ------------

If my fstab entry for /users2 is as follows..

/dev/vg01/lvol3 /users2 vxfs rw,suid,largefiles,delaylog,datainlog 0 2

should my newfs command be
# newfs -F vxfs -o largefiles /dev/vg01/lvol3

Thanks

[Luk Vandenbussche]

Hi,

lvol3 is striped on two disk.

Try the following

fsck -F vxfs -o full /dev/vg01/lvol1

or
restart your DS2300 it the fsck doesn't solve your problem.

[Senthil Kumar .A_1]

Hi Kenny,

If you issue newfs, you are at the risk of loosing the data in that volume, why don't you check fsck , once again, before taking such a drastic step.

Regards,
Senthil Kumar .A

[kenny chia]

hmm...
full fsck not much of a help.


# fsck -F vxfs -o full /dev/vg01/rlvol3
log replay in progress
pass0 - checking structural files
pass1 - checking inode sanity and blocks
pass2 - checking directory linkage
warning - no lost+found directory for fileset 999
pass3 - checking reference counts
pass4 - checking resource maps
OK to clear log? (ynq)y
set state to CLEAN? (ynq)y
# pwd
/
# mount /users2
vxfs mount: /dev/vg01/lvol3 no such device
#

[Luk Vandenbussche]

Hi,

Try also mount /dev/vg01/lvol3 /users2 please.

Perhaps here might be an error in your fstab