mount error on vxfs

[mr. N.V. (Nitin) Mehta]

I get the following mount error on vsfx
HP9000 Kseries HP-UX 11.0

mount - read of super-block on /dev/vgap2/l_app06 failed: I/O error

doing the following command:
fsck -F vxfs -o full /dev/vgap2/l_app06
Gives ERROR:
cannot open device /dev/vgap2/l_app06
file system check failure, aborting ...

Recently I applied the following
XSWGR1100 B.11.00.50.5 HP-UX General Release Patches, September 2000
After this errors are coming up.

What do I do to recover this file system.
Regards,
Nitin Mehta.

[Stefan Farrelly]

You have 2 possiblities;

1. The Sep2000 bundle did indeed introduce a bug with either mount or fsck.

2. The Sep2000 bundle install was just a coincedence, you actually have an error on a disk causing the mount and fsck errors.

Your stuck at the moment as it is, cant fsck or mount it, so before you look into a disk problem/replacement then lets see if the Sep2000 bundle introduced a new version of mount and/or fsck. Do> swlist -l file XSWGR11 (check spelling!) | grep mount
and same except grep for fsck and same for libvxfs (main vxfs library in the kernel). If theres a patch listed which has a new version of either/both then try swremoving those individual patches. Then try mounting your lvol and/or fsck'ing it. if it doesnt work then most likely you have a disk error which will require replacement.

[mr. N.V. (Nitin) Mehta]

I checked using vgdisplay to see the volume group information of the file system involved.
# vgdisplay /dev/vgap2
vgdisplay: Warning: couldn't query physical volume "/dev/dsk/c0t4d0":
The specified path does not correspond to physical volume attached to
this volume group
vgdisplay: Warning: couldn't query all of the physical volumes.
--- Volume groups ---
VG Name /dev/vgap2
VG Write Access read/write
VG Status available
Max LV 255
Cur LV 5
Open LV 5
Max PV 16
Cur PV 3
Act PV 2
Max PE per PV 4341
VGDA 4
PE Size (Mbytes) 4
Total PE 8680
Alloc PE 8122
Free PE 558
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0

This indicates that there is some problem with on of the physical volume involved.
I tried also the pvdisplay of the volume
and got the following error:

# pvdisplay /dev/dsk/c0t4d0
pvdisplay: Warning: couldn't query physical volume "/dev/dsk/c0t4d0":
The specified path does not correspond to physical volume attached to
this volume group
pvdisplay: Warning: couldn't query all of the physical volumes.
pvdisplay: Couldn't retrieve the names of the physical volumes
belonging to volume group "/dev/vgap2".
pvdisplay: Cannot display physical volume "/dev/dsk/c0t4d0".

Is it indeed a Hard Disk problem or something else?

[Vincente Fernandes]

"fsck -F vxfs -o full /dev/vgap2/l_app06
Gives ERROR: cannot open device /dev/vgap2/l_app06
file system check failure, aborting ..."
This error usally occurs when the Volume Group is not activated. Try doing "vgchange -a y /dev/vgap2".
With regards to second error message it can also happen due to your volume group not activated. Do a vgchange first which should solve your problem.

[Vincente Fernandes]

Not a single LVM command will work when your volume group is not acitvated. Is your VG part of MC/Service Guard then i guess you need to activated it by using "-e" option. Do a man vgchane for more details.

[CHRIS_ANORUO]

Run vgscan -a
You can also check the drivers update through SAM -Kernel Configuration -> Drivers
Check for scsi and spt0, their status should be in.

[John Palmer]

What does ioscan report for the disk?

run 'ioscan -fnC disk' and check the 'Driver S/W State' details.

This definitely sounds like a problem with the internal disk.

[Eddie Warren]

I agree with doing an ioscan, just to see if the device is CLAIMED or is displaying NO_HDW.

Try:

ioscan -fnC disk

Also try diskinfo for the suspected failing disk. If it displays all zero's, you probably have a bad disk.

diskinfo /dev/rdsk/c0t4d0

[John Hall]

I got this same error when doing a cold install of HP-UX 11.0 on a 10.20 system. The volume group I was vgimport'ing consisted of 2 physical disks and since the second physical disk was merely a mirrordisk copy of the first I did not think I needed to specify both disks on the vgimport. When I did specify both disks on the vgimport everything was fine and the fsck worked. Does your volume group here consist of more than one physical disk?