Problem with mounting a FS (file system)

[bunty.....]

hi,
one of my file system got unmounted.

#bdf
..
..
bdf:/oracle/data : I/O error
#cd /oracle
#ll
./data not found
#cd
#umount /opt/oracle
#mount /dev/vg01/oracle_data /oracle/data
vxfs mount: /dev/vg01/oracle_data is corrupted. needs checking
#fsck -y /dev/vg01/oracle_data
super block file system size incorrect, run full fsck
vxfs fsck: cannot initialize aggregate
file system check failure, aborting ...

#lvdisplay /dev/vg01/oracle_data
LV Name /dev/vg01/oracle_data
VG Name /dev/vg01
LV Permission read/write
LV Status available/syncd
Mirror copies 0
Consistency Recovery MWC
Schedule parallel
LV Size (Mbytes) 7000
Current LE 1750
Allocated PE 1750
Stripes 0
Stripe Size (Kbytes) 0
Bad block on
Allocation strict
IO Timeout (Seconds) default


Please suggest

[Ivan Ferreira]

Try with:

fsck -o full -y /dev/vg01/oracle_data

[James R. Ferguson]

Hi Bunty:

Try:

# fsck -F vxfs -y -o full /dev/vg01/oracle_data

Regards!

...JRF...

[Deoncia Grayson_1]

Also, make sure data is still a directory in /oracle. If /oracle/data is not there.....

try:

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

[bunty.....]

I tried it but...
# fsck -o full -y /dev/vg01/oracle_data
super block file system size incorrect, restore super-blocks? (ynq)y
file system is larger than device
vxfs fsck: cannot initialize aggregate
file system check failure, aborting ...

# mount -F vxfs /dev/vg01/oracle_data /oracle/data
vxfs mount: /dev/vg01/oracle_data is corrupted. needs checking

# mount /dev/vg01/oracle_data /oracle/data
vxfs mount: /dev/vg01/oracle_data is corrupted. needs checking

Please suggest

[Sundar_7]

How about nolog ? not sure if that will make any different though

# fsck -F vxfs -o full,nolog /dev/vg01/roracle_data

[Ivan Ferreira]

You may need to re-create the file system and restore from your backup. I hope that you have a recent backup.

[James R. Ferguson]

Hi Bunty:

"...file system is larger than device..."

This seems related to your earlier post/problem today:

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=991142

I think you may be forced to recreate your filesystem and restore your data from a backup.

Regards!

...JRF...

[bunty.....]

# fsck -F vxfs -y -o full /dev/vg01/oracle_data
file system is larger than device
vxfs fsck: cannot initialize aggregate
file system check failure, aborting ...

#fsck -F vxfs -o full /dev/vg01/roracle_data
file system is larger than device
vxfs fsck: cannot initialize aggregate
file system check failure, aborting ...

please suggest

[Deoncia Grayson_1]

I agree with JRF, you're going to have recreate this filesystem and restore from backup

[bunty.....]

hi JRF,
these are different different servers

[Geoff Wild]

WAIT!

Sounds like you extended the file system but not the lvol - put it back.

/dev/vg01/oracle_data is 7000 Mbytes

fsadm -b 7000M /dev/vg01/roracle_data

The fsck it.

Rgds...Geoff

[James R. Ferguson]

Hi (again) Bunty:

OK, Technical Knowledge Base document #KBRC00010646 suggests that it's your lucky day.

# lvextend -l /dev/vg01/oracle_data

Note that we are extending in extents. Add a few more than the number (n) of extents needed to match your filesystem size.

Since your extent size appears to be four (4), if your filesystem size was 8000 MB, then you would need 200 extents (plus a few more) for the value of 'n'.

# extendfs -F vxfs -v /dev/vg01/oracle_data

This makes the filesystem grow to the size of the logical volume.

# fsck -F vxfs -o full -y /dev/vg01/oracle_data

# mount /dev/vg01/oracle_data /opt/oracle

By the way, you should (?) be able to query the actual filesystem size with:

# fstyp -v /dev/vg01/oracle_data

Look at the value for 'f_size'.

Regards!

...JRF...

[Deoncia Grayson_1]

fsadm -F vxfs -b 7168000 /oracle/data

(remember this is in bytes)

[bunty.....]

# fsadm -b 7000M /dev/vg01/oracle_data
vxfs fsadm: /dev/vg01/oracle_data is not the root inode of a vxfs file system

# fsadm -F vxfs -b 7000M /dev/vg01/oracle_data
vxfs fsadm: /dev/vg01/oracle_data is not the root inode of a vxfs file system

# fsadm -F vxfs -b 7168000 /dev/vg01/oracle_data
vxfs fsadm: /dev/vg01/oracle_data is not the root inode of a vxfs file system

# lvextend -L 7168 /dev/vg01/oracle_data
lvextend: Logical volume "/dev/vg01/oracle_data" could not be changed:

now what happen:
all the logical volume / file system getting same problem under this volume group.
please suggest soon.

[Deoncia Grayson_1]

Bunty,

Do you have a backup of this filesystem?

[bunty.....]

No, i dont have backup for these file systems

[James R. Ferguson]

Hi Bunty:

Did you read my last suggestion, above? Did you try it?

...JRF...

[Deoncia Grayson_1]

After reading through all documentation thats posted out there and the error you are now receiving, try JRF suggestion... all documents points to his solution...

you can run on a search on your erros and you will be directed to the same solution

[Devender Khatana]

Hi,

You mentioned that all the LVOL are having this problem in this VG. Can you post

#vgdisplay -v vg01

It could be due to non availability of free space in the VG.

HTH,
Devender