topic Re: Model 20 help in Disk Enclosures

Model 20 help

rgilson — Mon, 13 Oct 2008 13:55:40 GMT

Our G70 server crashed (drive failure). After replacing drives and getting the server operational, we couldn't mount our RAID volumes on the Model 20. Actually two of the three volumes show. If I go into sam I get:

Disk Devices 0 of 7 selected
------------------------------------------------------------------------------
Hardware Number Volume Total
Path of Paths Use Group Mbytes Description
/---------------------------------------------------------------------------\
| 16.0 1 -- -- -- HP Model 20 Disk Array ^
| 16.0.0 1 LVM vg02 16026 HP Model 20 LUN (RAID
| 16.0.1 1 Unused -- 33646 HP Model 20 LUN (RAID
| 16.1 1 -- -- -- HP Model 20 Disk Array
| 52.3.0 1 Unused vg00 8683 SEAGATE ST19171N
| 52.4.0 1 LVM vg00 2033 HP C2490 SCSI Disk Dri
| 52.6.0 1 LVM vg00 4095 Quantum XP34361S

I select the "unused" disk (16.0.1) and try to add the volume group but the response is "volume group already exists, adding will delete all data". Obviously I don't want to lose the data. How do get this lv to show up again?
FYI
Model G70 hpux 10.20
Model 20 w/ 2 storage processors + bpu +
(10) 9gb drives

Re: Model 20 help

Torsten. — Tue, 14 Oct 2008 04:31:02 GMT

Some things to try:

Do a preview of vgscan

# vgscan -p -v

and check the array status either with stm or gridmanager.

Re: Model 20 help

rgilson — Tue, 14 Oct 2008 12:13:10 GMT

Here's the output of vgscan:

vgscan -p -v
vgscan: Warning: couldn't query physical volume "/dev/dsk/c1t3d0":
The specified path does not correspond to physical volume attached to
this volume group
vgscan: Warning: couldn't query all of the physical volumes.
vgscan: The physical volume "/dev/dsk/c0t0d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c0t1d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c1t3d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c1t4d0" is already recorded in the "/etc/l
vmtab" file.
vgscan: The physical volume "/dev/dsk/c1t6d0" is already recorded in the "/etc/l
vmtab" file.
Current path "/dev/dsk/c0t1d0" is an alternate link, skip.
reading LVMREC, please wait....
Couldn't stat physical volume "/dev/dsk/c1t2d0":
Invalid argument

/dev/vg02
/dev/dsk/c0t0d0
/dev/dsk/c0t1d0

reading VGDA, please wait....
reading VGDA, please wait....
filling VGDA, please wait....
Following Physical Volumes belong to one Volume Group.
Unable to match these Physical Volumes to a Volume Group.
Use the vgimport command to complete the process.
/dev/dsk/c0t0d1
/dev/dsk/c0t1d1

Following Physical Volumes belong to one Volume Group.
Unable to match these Physical Volumes to a Volume Group.
Use the vgimport command to complete the process.
/dev/dsk/c1t3d0

/dev/vg00
/dev/dsk/c1t4d0
/dev/dsk/c1t6d0

Gridmgr shows:
SP A:
Prom Revision 1.73
Micro Code Revision 8.60
All 10 drives show ENA

SP B:
Prom Revision 1.73
Micro Code Revision 8.60
All 10 drives show RDY