[linux-lvm] Either the superblock or the partition table is likely to be corrupted

[linux-lvm] Either the superblock or the partition table is likely to be corrupted

[James Corteciano]

[-- Attachment #1: Type: text/plain, Size: 2015 bytes --]

Hi All,

I have volume group called "vg-nfs1" and it is controlled by RHCS
(rhcs-nfs1) to mount in rhcs-node1 server. This group has 3 logical volumes
namely lv-wf1 (160G), lv-wf2 (20G), and lv-dev (10G).

Now I have plan to reduce lv-wf1 into 70G and lv-wf2 increase into 30G.
These are the steps I did:
1) Umount /nfs/lv-wf1, /nfs/lv-wf2, and /nfs/lv-dev but I can't unmounted
them because message said "device busy". Since it's running in RHCS, I use
"clusvcadm -s rhcs-nfs1" and the service is stop successfully.

2) I run system-config-lvm and reduce lv-wf1 from 160G to 70G and increased
lv-wf2 from 10G to 30G.

Now, when I run the rhcs-nfs1 service using the command "clusvcadm -e
rhcs-nfs1 -m rhcs-node1", I got failed result. I have checked the
/var/log/messages and this is the output.

--------------------------------------------------------------
clurgmgrd: [7981]: <err> 'fsck -p /dev/vg-nfs1/lv-wf1' failed, error=4;
check /tmp/fs-wf1-fs.fsck.log.e24076 for errors
clurgmgrd[7981]: <warning> #68: Failed to start service:rhcs-nfs1; return
value: 1
multipathd: dm-18: remove map (uevent)
multipathd: dm-19: remove map (uevent)
multipathd: dm-20: remove map (uevent)
clurgmgrd[7981]: <warning> #71: Relocating failed service service:rhcs-nfs1
--------------------------------------------------------------

When I check the /tmp/fs-wf1-fs.fsck.log.e24076, this is the output:

--------------------------------------------------------------
fsck 1.39 (29-May-2006)
/dev/vg-nfs1/lv-wf1: The filesystem size (according to the superblock) is
41943040 blocks
The physical size of the device is 18350080 blocks
Either the superblock or the partition table is likely to be corrupt!

/dev/vg-nfs1/lv-wf1: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY.
        (i.e., without -a or -p options)
--------------------------------------------------------------

# ll -d /dev/vg-nfs1/lv-wf1
ls: /dev/vg-nfs1/lv-wf1: No such file or directory


I hope anyone could help me to rectify this.

Thank you.

Regards,
James

[-- Attachment #2: Type: text/html, Size: 2608 bytes --]

Re: [linux-lvm] Either the superblock or the partition table is likely to be corrupted

[Charles Marcus]

On 2010-09-07 12:45 PM, James Corteciano <james@linux-source.org> wrote:
> 2) I run system-config-lvm and reduce lv-wf1 from 160G to 70G and
> increased lv-wf2 from 10G to 30G. 
> 
> Now, when I run the rhcs-nfs1 service using the command "clusvcadm
> -e rhcs-nfs1 -m rhcs-node1", I got failed result. I have checked the 
> /var/log/messages and this is the output.

Dumb/obvious question... you did reduce the size of the filesystem
*before* reducing the size of the LV, right?

-- 

regards,

Charles

Re: [linux-lvm] Either the superblock or the partition table is likely to be corrupted

[Clyde E. Kunkel]

  On 09/07/2010 02:11 PM, Charles Marcus wrote:
> On 2010-09-07 12:45 PM, James Corteciano<james@linux-source.org>  wrote:
>> 2) I run system-config-lvm and reduce lv-wf1 from 160G to 70G and
>> increased lv-wf2 from 10G to 30G.
>>
>> Now, when I run the rhcs-nfs1 service using the command "clusvcadm
>> -e rhcs-nfs1 -m rhcs-node1", I got failed result. I have checked the
>> /var/log/messages and this is the output.
> Dumb/obvious question... you did reduce the size of the filesystem
> *before* reducing the size of the LV, right?
>

And, if it was done out of sequence, is there a way to still fix it?

TIA

Re: [linux-lvm] Either the superblock or the partition table is likely to be corrupted

[Ray Morris]

>> Dumb/obvious question... you did reduce the size of the filesystem
>> *before* reducing the size of the LV, right?
>> 
> 
> And, if it was done out of sequence, is there a way to still fix it?

   If you did not increase the ize of the lv-wf2 FILESYSTEM yet,
you should be able to undo what was done.  See vgcfgrestore.
--
Ray Morris
support@bettercgi.com

Strongbox - The next generation in site security:
http://www.bettercgi.com/strongbox/

Throttlebox - Intelligent Bandwidth Control
http://www.bettercgi.com/throttlebox/

Strongbox / Throttlebox affiliate program:
http://www.bettercgi.com/affiliates/user/register.php


On 09/10/2010 09:22:26 AM, Clyde E. Kunkel wrote:
>  On 09/07/2010 02:11 PM, Charles Marcus wrote:
>> On 2010-09-07 12:45 PM, James Corteciano<james@linux-source.org>   
>> wrote:
>>> 2) I run system-config-lvm and reduce lv-wf1 from 160G to 70G and
>>> increased lv-wf2 from 10G to 30G.
>>> 
>>> Now, when I run the rhcs-nfs1 service using the command "clusvcadm
>>> -e rhcs-nfs1 -m rhcs-node1", I got failed result. I have checked the
>>> /var/log/messages and this is the output.
>> Dumb/obvious question... you did reduce the size of the filesystem
>> *before* reducing the size of the LV, right?
>> 
> 
> And, if it was done out of sequence, is there a way to still fix it?
> 
> TIA
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
>