topic Re: Bad disk . in Disk Enclosures

Bad disk .

rveri — Wed, 14 Sep 2005 12:26:16 GMT

How do I know , which disk has failed. hp-ux 11i version , connected to xp1024. HP got alert message about the disk failure. How I can know about this.

Thanks.
Veri.

Re: Bad disk .

Florian Heigl (new acc) — Wed, 14 Sep 2005 12:32:03 GMT

start with

vgdisplay -v | grep -e "PV Name" -e "PV Status"
(Look for 'unavailable')

and

ioscan -fnCdisk | grep NO_HW

both commands should report the failed disk.

Re: Bad disk .

Torsten. — Wed, 14 Sep 2005 12:38:50 GMT

It depends where the disk is located. You can't see the disk if it is inside the XP1024. The array itself is monitoring the disks (and other components) and send a message to HP if something is wrong. HP will give you a call immediately regarding the failure.

Re: Bad disk .

melvyn burnard — Wed, 14 Sep 2005 12:42:52 GMT

XP disc arrays usually have a connection to a phone line, and when they discover an internal failure or problme, they "phone home" to report this.
Hence HP will have been informed, and would log a call for the disc inside the XP to be replaced.
You generally will not know there is any issue until you get informed by HP that they need to replace a part in the XP.

Re: Bad disk .

rveri — Wed, 14 Sep 2005 14:36:51 GMT

Yes we got a call from HP for the disk to be replaced. And scheduled for that.

This could be maped to , one of the Superdome partition ,in the superdome ( Total 10 Partition )

As a sysadmin , what information should I need to know , as I have to raise a internal change ticket or helpdesk ticket for that.

Many Thanks,
Veri.

Re: Bad disk .

Torsten. — Wed, 14 Sep 2005 16:19:05 GMT

Ask the HP CE what disk was replaced, just for your documentation. This will sound very technical ;-))

The disk exchange doesn't impact your server, because the array rebuild the data from the bad disk to a spare disk and if the disk is replaced, it copies the data back. It takes a while, but there is no impact to your applications.

Re: Bad disk .

Devender Khatana — Thu, 15 Sep 2005 00:39:16 GMT

Hi Veri,

A disk in a XP array is not directly visible to host. It is part of a raid group which consists of 4 or 7 disks. Also one raid group may be giving you more that 10-15 LUNs to various hosts which will be available even after failed disk. In such case it will not be possible for any of the host to know that disk has failed.

You can infact refer to XP- SVP to get s detailed information regarding the location of the disk. The replacement has to be carried out by a XP trainned engineer only. If SVP is not accesible you can use command view XP to find out the location of failed disk.

HTH,
Devender

Re: Bad disk .

rveri — Thu, 15 Sep 2005 13:23:45 GMT

Thx all,
what is xpinfo is used for. !

Re: Bad disk .

Devender Khatana — Thu, 15 Sep 2005 22:51:04 GMT

Hi,

xpinfo is the command which should be installed on the hosts which access a XP array. The use of xpinfo is to give a detailed information of the various LUNs a host is accessing from one or more XP arrays.

HTH,
Devender

Re: Bad disk .

Devender Khatana — Fri, 16 Sep 2005 09:41:07 GMT

Hi,

On a given host it gives all important information. It looks like

#xpinfo

Scanning disk devices...

Device File : /dev/rdsk/c4t0d0 Model : XP512
Port : CL1C Serial # : 00032380
Host Target : 00 Code Rev : 0118
Array LUN : 00 Subsystem : 0004
CU:LDev : 00:98 CT Group : ---
Type : OPEN-E CA Volume : SMPL
Size : 13893 MB BC0 (MU#0) : SMPL
ALPA : e4 BC1 (MU#1) : SMPL
Loop Id : 02 BC2 (MU#2) : SMPL
SCSI Id : --- RAID Level : RAID5
FC-LUN : 4000000000000000 RAID Group : 1-10
Port WWN : 500060e8027e7c02 ACP Pair : 1
Disk Mechs : R109 R119 R129 R139

Device File : /dev/rdsk/c4t0d1 Model : XP512
Port : CL1C Serial # : 00032380
Host Target : 00 Code Rev : 0118
Array LUN : 01 Subsystem : 0005
CU:LDev : 01:98 CT Group : ---
Type : OPEN-E CA Volume : SMPL
Size : 13893 MB BC0 (MU#0) : SMPL
ALPA : e4 BC1 (MU#1) : SMPL
Loop Id : 02 BC2 (MU#2) : SMPL
SCSI Id : --- RAID Level : RAID5
FC-LUN : 4001000000000000 RAID Group : 2-10
Port WWN : 500060e8027e7c02 ACP Pair : 2
Disk Mechs : R149 R159 R169 R179

Device File : /dev/rdsk/c4t0d2 Model : XP512
Port : CL1C Serial # : 00032380
Host Target : 00 Code Rev : 0118
Array LUN : 02 Subsystem : 0006
CU:LDev : 02:98 CT Group : ---
Type : OPEN-E CA Volume : SMPL
Size : 13893 MB BC0 (MU#0) : SMPL
ALPA : e4 BC1 (MU#1) : SMPL
Loop Id : 02 BC2 (MU#2) : SMPL
SCSI Id : --- RAID Level : RAID5
FC-LUN : 4002000000000000 RAID Group : 5-10
Port WWN : 500060e8027e7c02 ACP Pair : 1
Disk Mechs : R209 R219 R229 R239

Or something like this-

#xpinfo -i

Scanning disk devices...

Device File ALPA Tgt Lun Port CU:LDev Type Serial#
===================================================
/dev/rdsk/c4t0d0 e4 00 00 CL1C 00:98 OPEN-E 00032380
/dev/rdsk/c4t0d1 e4 00 01 CL1C 01:98 OPEN-E 00032380
/dev/rdsk/c4t0d2 e4 00 02 CL1C 02:98 OPEN-E 00032380
/dev/rdsk/c4t0d3 e4 00 03 CL1C 03:98 OPEN-E 00032380
/dev/rdsk/c4t0d4 e4 00 04 CL1C 00:99 OPEN-E 00032380
/dev/rdsk/c4t0d5 e4 00 05 CL1C 01:99 OPEN-E 00032380
/dev/rdsk/c4t0d6 e4 00 06 CL1C 02:99 OPEN-E 00032380
/dev/rdsk/c4t0d7 e4 00 07 CL1C 03:99 OPEN-E 00032380
/dev/rdsk/c4t1d0 e4 01 08 CL1C 00:9a OPEN-E 00032380
/dev/rdsk/c4t1d1 e4 01 09 CL1C 01:9a OPEN-E 00032380

HTH,
Devender