Filesystem problem (Cannot find the Vfs value for file system )

[ldedios]

Hi,

I need help again. I encountered two problems on one server.

1) I have one FS that I can mounted once I issue "df -k" but when I checked it using "lsvg -l" command. The output is:
lv_appl jfs2 186 372 2 open/syncd N/A

I was wondering why the mount point is N/A. I tried the fsck command to fix the problem but this is the message i got:
fsck: 0506-040 Cannot find the Vfs value for file system /dev/lv_appl.

2)the other problem is I saw the FS mounted when I issue "df -k" but when I checked it using "lsvg -l" the mount point is /NEW. I tried to unmount the original FS and remove /NEW. I was trying to put the original FS name but /NEW keeps on coming back. AGain I tried fsck but the error is:

he current volume is: /dev/lv_archivelogs
fsck: 0507-020 Invalid magic number in the primary superblock.
Primary superblock is corrupt.
fsck: 0507-013 Invalid magic number in the secondary superblock.
Secondary superblock is corrupt.
fsck: 0507-018 Superblock is corrupt and cannot be repaired.
Both primary and secondary copies are corrupt. Cannot continue.

Both these FS belong to one volume group only. Kindly help me on this. Any information will be valuable.

Thanks.

 

[ogniemi]

check for errors in error log. are there any storage problems reported? check if your disks are still available in the system.

if possible unmount all filesystems residing on the VG, varyoffvg and then try again varyonvg and mount filesystems.

 

[ldedios]

thanks ogniemi for your reply. I have checked the error report and no error related to storage. Is there another way to resolve this for the server is a production server. Might have another way to resolve this.

Also what might have been the cause for this type of error?

thanks.

 

[khalidaaa]

I think there is a problem with your /etc/filesystem. Can you paste the output of the /etc/filesystem showing only the information related to those filesystems?

There might be a problem as well in the ODM but let's first check the /etc/filesystems

Regards,
Khalid

 

[ldedios]

here is the output of /etc/filesystem:

/prod/archivelogs:
dev = /dev/lv_archivelogs
vfs = jfs2
log = /dev/erpjfsloglv
mount = false
check = false
options = cio,rw
account = false

/NEW mount point [ pod/appl]: ===>>> this is suppose to be /prod/appl
dev = /dev/lv_appl
vfs = jfs2
log = /dev/erpjfsloglv
mount = false
check = false
options = rw
account = false

let me know if I can edit this or what can I do to correct the problem.

Thanks a lot for the help.

 

[khalidaaa]

It is not recommended to edit the /etc/filesystems!

So the best thing is to use smit.

It seems that there is a confusion between your mount points! If you look at your first point which is lv_appl as being N/A. It looks that you are already mounting this under /NEW. It seems that you have double mounts over each other!

I advise you to unmount both filesystems first:

umount /dev/lv_appl
umount /dev/lv_archivelogs

Confirm that they were unmounted using df -k

Then go to smit chjfs2 --> Choose /NEW

Go to the NEW mount point field and change it to /prod/appl --> Hit Enter

Then try to mount the filesystems again:

mount /dev/lv_archivelogs
mount /dev/lv_appl

Then confirm that your changes were successful using df -k

Hope this will solve your problem.

Regards,
Khalid

 

[ldedios]

thanks khalid for the reply. I will try to do this and will update all of you regarding this.