Splitting a volume and creating new VG from that vol

[ibnabbas]

I am new to HP-UX. I have a requirement to split a mirrored volume (abc) from the VG (vgsource) and create a new VG (vgbackup) from that LUN (/dev/disk/disk111) and keep the same volume name (abc) and mount it as a separate filesystem.

[Torsten.]

It is not clear at all what you want to do.

If you split a mirrored *LVOL* (see lvsplit), then this new LVOL is inside the same VG, of course. Once splitted, you can mount it.

Creating a new VG means delete the data.

[Manix]

Looks you are trying to get a free LUN /disk
which is part of a mirror lvol.

You need to split the mirror then reduce the
VG to free up the PV ( lun /dik) and then recreate a new VG from scratch ,you need
pvcreate ,vgcraete ,lvcreate ( same name can be used ).

Hope this help.

Thanks
Manix

[Torsten.]

still not clear - same name for what???

[ibnabbas]

Tortsen,

Let me rephrase.

Situation : The lvol abc is mirrored across 2 disks (disk21 and disk111) in VG vgsource.

Both disks are fine.

I want to break the mirror take disk111 out of vgsource and be able create a new VG (say vgnew) using disk111 and recreate the lvol abc with data well preserved !!!

Thanks.

[ibnabbas]

Oh! you mean I cant take an lvol from one VG and move it to another VG (new or exisitng) ?

[Torsten.]

Exactly.

A LVOL always remains in the same VG.

[ibnabbas]

Any solution how I can clone the current lvol and move that data to another lvol in another VG.

Thx

[Torsten.]

"clone" may be the keyword ...


Is your array able to "clone"?


However, you may consider to create another LUN, VG and LVOL and copy the data.


Why do you need the data in another VG?

[Emil Velez]

I actually demonstrate this capability.

Assumption the lvol that you split off is on a separate disk


Example: VG has LVs 1, 2, and 3 on PVs 0 and 1. LVs 4, 5, and 6 on PVs 2, 3, 4, and 5.
# vgchange -a n VG
# vgexport -m MAP VG
# vgchgid PV2 PV3 PV4 PV5
# mkdir /dev/VG;
mknod /dev/VG/group c 64 0xnn0000
# Where nn is the original volume group number.
# mkdir /dev/new_VG;
mknod /dev/new_VG/group c 64 0xmm0000
# Where mm is the new volume group number.
# vgimport -m MAP /dev/VG PV0 PV1
# vgimport -m MAP /dev/new_VG PV2 PV3 PV4 PV5
# vgchange -a y -q n VG;
# vgchange -a y new_VG
# lvremove -f VG/LV4 VG/LV5 VG/LV6 new_VG/LV1 new_VG/LV2 new_VG/LV3
# vgreduce -f VG;
# vgreduce -f new_VG
# vgcfgbackup VG; vgcfgbackup new_VG

This is not a supported procedure or capability but it is using supported commands

Think it through and try it on crash and burn system.