XP back to NetApps Migration

[Need to migrate XP Back]

Currently on a HP-UX 11.23 server we have a VG with Strict PVG policy enforced.
We need to migrate the data from 13Gb LUNS provided by XP Back to NetApps 50Gb LUNS.
If I've e.g. 10 x 13Gb LUNS could I mirror the LV onto 3 x 50Gb LUNS?
Would I be able to do this without any downtime?

[TTr]

If your server has access to the XP and the NetApp at the same time, yes you can use the mirroring approach. No downtime is needed. The only problem is if your existing Volume groups do not have enough extends (size and number of extends) to handle the 50GB LUNs. You need to check with vgdisplay and look at the "Max PE per PV" and multiply that number with the PE size. If the 50GB LUNs can not fit the VGs look into the "vgmodify" command. It is available via a patch on 11.23.

[Need to migrate XP Back]

Let me rephrase the query:

We have a VG with strict PVG enforced. There are two PVGâ s created called PODA & PODB.

As we have LUNs shared from HP XP Back, the max LUN size is 13Gb, as the database size is in TBâ s, there are a huge no. of disks under one VG.

Currently we are planning to move to NetApp LUNs, which would be able to share a size of around e.g. 50GB.

My query is: If Iâ ve a VG of say 130 Gb ( 10 x 13Gb) LUNs, could I mirror it with say (3 x 50 Gb) LUNs?

I was planning to break off PODB (PVG) from the VG and attached a new PVG e.g. WDR1 of the same size but lesser no. of disks. Would the LVâ s be mirrored even though PVG strict policy is enforced?

If this doesnâ t work out, is it possible for me to remove the strict PVG policy and get this done?

[TTr]

The strict PVG polisy is to allocate mirrored extends on different PVGs. So if you create a new PVG for the NetApp LUNs the strict PVG policy is obeyed.
Why is the allocation polict set right now? Do you have mirroring setup already on the XP array and how is it set up?

Your problem is still going to be whether the NetApp LUNs will "fit" into the existing VGs, or more appropriately if you will be able to vgextend the existing volume gorups over the new the NetApp LUNs.

[Need to migrate XP Back]

We already have mirroring in place for the LV's.
I was planning to reduce the mirror for each LV.
Delete the PVG PODB and create a new one with the reduced no. of disks. I believe PVG strictness applies at the VG level and not on the LV.
Recreate the LV mirror with the new set of disks. Will this work?

I've provided an e.g. of our setup below:

vgdisplay -v vgr3p-dbci
--- Volume groups ---
VG Name /dev/vgr3p-dbci
VG Write Access read/write
VG Status available, exclusive
Max LV 255
Cur LV 13
Open LV 13
Max PV 64
Cur PV 28
Act PV 28
Max PE per PV 3474
VGDA 56
PE Size (Mbytes) 8
Total PE 48600
Alloc PE 47502
Free PE 1098
Total PVG 2
Total Spare PVs 0
Total Spare PVs in use 0

--- Logical volumes ---
LV Name /dev/vgr3p-dbci/lvr3pdveb09
LV Status available/syncd
LV Size (Mbytes) 2048
Current LE 256
Allocated PE 512
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3pascs09
LV Status available/syncd
LV Size (Mbytes) 2048
Current LE 256
Allocated PE 512
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3p920_64
LV Status available/syncd
LV Size (Mbytes) 7168
Current LE 896
Allocated PE 1792
Used PV 4

LV Name /dev/vgr3p-dbci/lvr3poriloga
LV Status available/syncd
LV Size (Mbytes) 208
Current LE 26
Allocated PE 52
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3porilogb
LV Status available/syncd
LV Size (Mbytes) 208
Current LE 26
Allocated PE 52
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3pmirloga
LV Status available/syncd
LV Size (Mbytes) 208
Current LE 26
Allocated PE 52
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3pmirlogb
LV Status available/syncd
LV Size (Mbytes) 208
Current LE 26
Allocated PE 52
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3preorg
LV Status available/syncd
LV Size (Mbytes) 2048
Current LE 256
Allocated PE 512
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3psaparch
LV Status available/syncd
LV Size (Mbytes) 2048
Current LE 256
Allocated PE 512
Used PV 4

LV Name /dev/vgr3p-dbci/lvr3poraarch
LV Status available/syncd
LV Size (Mbytes) 172032
Current LE 21504
Allocated PE 43008
Used PV 26

LV Name /dev/vgr3p-dbci/lvr3porainv
LV Status available/syncd
LV Size (Mbytes) 1024
Current LE 128
Allocated PE 256
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3pbackup
LV Status available/syncd
LV Size (Mbytes) 256
Current LE 32
Allocated PE 64
Used PV 2

LV Name /dev/vgr3p-dbci/lvr3pftp
LV Status available/syncd
LV Size (Mbytes) 504
Current LE 63
Allocated PE 126
Used PV 4


--- Physical volumes ---
PV Name /dev/dsk/c32t3d0
PV Name /dev/dsk/c45t3d0 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c32t4d4
PV Name /dev/dsk/c45t4d4 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c33t11d6
PV Name /dev/dsk/c46t11d6 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c33t11d7
PV Name /dev/dsk/c46t11d7 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c33t12d0
PV Name /dev/dsk/c46t12d0 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c33t12d1
PV Name /dev/dsk/c46t12d1 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c28t3d0
PV Name /dev/dsk/c48t3d0 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c28t4d4
PV Name /dev/dsk/c48t4d4 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c29t11d6
PV Name /dev/dsk/c49t11d6 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c29t11d7
PV Name /dev/dsk/c49t11d7 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c29t12d0
PV Name /dev/dsk/c49t12d0 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c29t12d1
PV Name /dev/dsk/c49t12d1 Alternate Link
PV Status available
Total PE 1735
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t1d7
PV Name /dev/dsk/c33t1d7 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t2d1
PV Name /dev/dsk/c33t2d1 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t2d2
PV Name /dev/dsk/c33t2d2 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t9d3
PV Name /dev/dsk/c33t9d3 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t12d4
PV Name /dev/dsk/c33t12d4 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t12d5
PV Name /dev/dsk/c33t12d5 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t12d6
PV Name /dev/dsk/c33t12d6 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c46t12d7
PV Name /dev/dsk/c33t12d7 Alternate Link
PV Status available
Total PE 1736
Free PE 494
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t1d7
PV Name /dev/dsk/c29t1d7 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t2d1
PV Name /dev/dsk/c29t2d1 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t2d2
PV Name /dev/dsk/c29t2d2 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t9d3
PV Name /dev/dsk/c29t9d3 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t12d4
PV Name /dev/dsk/c29t12d4 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t12d5
PV Name /dev/dsk/c29t12d5 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t12d6
PV Name /dev/dsk/c29t12d6 Alternate Link
PV Status available
Total PE 1736
Free PE 0
Autoswitch On
Proactive Polling On

PV Name /dev/dsk/c49t12d7
PV Name /dev/dsk/c29t12d7 Alternate Link
PV Status available
Total PE 1736
Free PE 604
Autoswitch On
Proactive Polling On


--- Physical volume groups ---
PVG Name PODB
PV Name /dev/dsk/c28t3d0
PV Name /dev/dsk/c28t4d4
PV Name /dev/dsk/c29t11d6
PV Name /dev/dsk/c29t11d7
PV Name /dev/dsk/c29t12d0
PV Name /dev/dsk/c29t12d1
PV Name /dev/dsk/c48t3d0
PV Name /dev/dsk/c48t4d4
PV Name /dev/dsk/c49t11d6
PV Name /dev/dsk/c49t11d7
PV Name /dev/dsk/c49t12d0
PV Name /dev/dsk/c49t12d1
PV Name /dev/dsk/c49t1d7
PV Name /dev/dsk/c49t2d1
PV Name /dev/dsk/c49t2d2
PV Name /dev/dsk/c49t9d3
PV Name /dev/dsk/c49t12d4
PV Name /dev/dsk/c49t12d5
PV Name /dev/dsk/c49t12d6
PV Name /dev/dsk/c49t12d7
PV Name /dev/dsk/c29t1d7
PV Name /dev/dsk/c29t2d1
PV Name /dev/dsk/c29t2d2
PV Name /dev/dsk/c29t9d3
PV Name /dev/dsk/c29t12d4
PV Name /dev/dsk/c29t12d5
PV Name /dev/dsk/c29t12d6
PV Name /dev/dsk/c29t12d7

PVG Name PODA
PV Name /dev/dsk/c32t3d0
PV Name /dev/dsk/c32t4d4
PV Name /dev/dsk/c33t11d6
PV Name /dev/dsk/c33t11d7
PV Name /dev/dsk/c33t12d0
PV Name /dev/dsk/c33t12d1
PV Name /dev/dsk/c45t3d0
PV Name /dev/dsk/c45t4d4
PV Name /dev/dsk/c46t11d6
PV Name /dev/dsk/c46t11d7
PV Name /dev/dsk/c46t12d0
PV Name /dev/dsk/c46t12d1
PV Name /dev/dsk/c46t1d7
PV Name /dev/dsk/c46t2d1
PV Name /dev/dsk/c46t2d2
PV Name /dev/dsk/c46t9d3
PV Name /dev/dsk/c46t12d4
PV Name /dev/dsk/c46t12d5
PV Name /dev/dsk/c46t12d6
PV Name /dev/dsk/c46t12d7
PV Name /dev/dsk/c33t1d7
PV Name /dev/dsk/c33t2d1
PV Name /dev/dsk/c33t2d2
PV Name /dev/dsk/c33t9d3
PV Name /dev/dsk/c33t12d4
PV Name /dev/dsk/c33t12d5
PV Name /dev/dsk/c33t12d6
PV Name /dev/dsk/c33t12d7

[TTr · Accepted solution]

Are the NetApp LUns already in the VGs? The list of PVs shows all LUNs are 13GB in size.

The PVG is setup at the VG level when you create and extend the VG (vgcreate, vgextend).

> Recreate the LV mirror with the new set of disks. Will this work?
It is the right idea but there is a problem with the size of the new LUNs. Your exisitng VGs can not use 50GB LUNs because in The vgdisplay output
> Max PE per PV 3474
> PE Size (Mbytes) 8
Your PVs in this volume group can only have up to 3474 extends. The size of each extend is 8MB so the largest LUN you can put in this VG is 3474 x 8 = 27,792MB, taht is about 26GB.

You need to look into changing the VG parameters with vgmodify. The vgmodify is a new command, in HP-UX 11.23 it is available with a patch, so it may or may not be in your system.

[Need to migrate XP Back]

Thanks, vgmodify works in our setup.

The action taken would probably be in the following way:

Reduce the Logical Volume (LV) mirroring for each LV available in the VG.
Use vgmodify to change the Max PE per PV to support the NetApp LUN size (With the current setup, it would only support 26GB LUNs)
Reduce the VG from the disks present in PODB.
Create a new PVG and add the NetApp LUNs to it.
Recreate the LV mirror.
After the filesystem sync has completed, we would be removing PODA from the VG and continue with Steps 4 & 5.
Run commands vgimport & vgexport to copy the maps.


I am yet to decide on the exact procedure & commands.

[TTr]

You are on the right track and you will be OK on your own...