HP-UX

This has got to be the system from hell. Once again, on the RP4440 (after the supplier replaced the entire box due to the bad RTC battery), finally have it all reloaded with the packages the developers need. The last thing is to add the secondary disk to the OS.

BCH sees both OS and secondary disk, IOSCAN sees the disk, ioscan -funC, the disk shows up, diskinfo reads the disk ok, HOWEVER, when I do a vgcreate I get VGRA appears corrupted at the end of the create. pvcreate -f /dev/rdsk/c3t0d0, works, vgcreate works and Vgdisplay actually shows the vg online and available and good, but the VGRA appears corrupted still occurs. lvmtab gets updated. vgcfgbackup vg01 fails, lvcreate WORKS, but mkfs fails.

If I vgchange -a n vg01, then vgchange -a y vg01 it cannot import the vg because it cannot associate the vg to a physical volume.

I have replaced the disk, Patched to Dec '09 release of OS patches from ITRC. Did searches for LVM, SCSI, DISK, SAM, VGRA and downloaded and installed the patches. Still fails. Any Ideas?

As of now, DD is able to write just fine to the disk (dd if=/dev/zero of=/dev/rdsk/c3t0d0 bs=1024k count=300000)

Moderator's Comments:

Please use code tags

I would try to start all from scratch, that means remove using rmdev the device then

Code:

ioscan -insf

then see if it shows up correctly, ah vg01 was already present? so there might be a mismatch between what it knows of the group and so is looking for a disk that doesn't exist...
My idea would be to create a new vg just to see the disk works... then you could start meditating on how to re-create vg01... as I cannot help more, not having any HP for quite some time know, when here it was decided to drop PA-RISC... ( not my choice...) but remembered having sometimes issues of the sort with clusters...

Do not directly write to the disk device after a pvcreate!

It is +5 years ago that I had access to an HP-UX box...

You might need vgscan to update some meta information in the volume group.