vgimport erors

[Puff]

When inporting a VG I am getting the following errors:

vgimport: Physical volume "/dev/dsk/c0t3d0" is not a block special file.
fail in reading_lvmrec, nread = 0, btor = 4096.

c0t3d0 is not a valid disk on any of my system and I can not find where vgimport is getting this disk info from. This is happening on two different servers for two different VGs. one server gets the fail in reading_lvmrec error and one does not. I have tried to create new mapfiles to use for the imports. It does not matter which VG I try to import, I always get this message.

Thanks!

[Mel Burslan]

what is the exact syntax of vgimport command you are using ? Are you using a "-f " option to read the pv_paths from ?

[Junior C.]

Puff,

Try the following

# insf -e
# ioscan -fnC disk
then
#vgimport

-Junior

[Sudeesh]

puff,
can you post the vgimport command line you used?

Sudeesh

[RAC_1]

What exactly commnd syntax are you using?? Are you using map file (exported with -s on vgexport) or are you using -f option on vgimport??

It may happen that between vgexport and vgimport that disk has disappeared.

[Devender Khatana]

Hi,

It is possible the old VG where these disks were used earlier, atleast one of the disk was at this address.

HTH,
Devender

[VEL_1]

Hope the following link will help you:

http://unix.derkeiler.com/Mailing-Lists/HP-UX-Admin/2005-03/0029.html


Thanks.

[vinod_25]

Hi Puff

This is happening because you are trying to import a disk that has missing device files.

Perform the following steps to resolve your errors:

1. # insf -e -C disk

2. # vgimport -v -s -m /tmp/vgnews0.map /dev/vgnews0

this will solve the problem.

Regards

Vinod

[Mark Nieuwboer]

Hi Puff,

what does pvdisplay say about the disk.
Maybe it's an old disk wich is know in the kernel as boot disk.

grtz. Mark

[Puff]

I am using the following for the vgimport:

vgimport -sv -m vgabc

It does not matter which vg on the system I try to do the import on. A pvdisplay says there is no such file or dircetory called /dev/dsk/c0t3d0. I have tried to run the insf -e and ioscans and this is not a valid disk. We use a SAN for all of our storage that is not part of the O/S. I believe that a channel of 0 would be an internal connection not one of our san connections. The root volume group that does use internal disks are on channels 1 and 2. Since this is happening on two different servers and both are showing the same ctd id for the disk, could this be the tape drive or CDROM that it thinks is part of a volume group?