topic Re: Harddisk status in Operating System - HP-UX

Harddisk status

Mohammad Mostafa — Wed, 08 Jul 2009 06:48:28 GMT

Hi all,
i got the following in the syslog.log

# ----------------
Jun 26 15:50:51 unixdb vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 183112
891, dev: 1f021000, io_id: 272afbd
Jun 26 15:50:54 unixdb EMS [2650]: ------ EMS Event Notification ------ Value:
"CRITICAL (5)" for Resource: "/storage/events/disks/default/0_1_1_0.1.0" (T
hreshold: >= " 3") Execute the following command to obtain event details:
/opt/resmon/bin/resdata -R 173670405 -r /storage/events/disks/default/0_1_1_0.1.
0 -n 173670401 -a
Jun 26 15:52:49 unixdb EMS [2650]: ------ EMS Event Notification ------ Value:
"CRITICAL (5)" for Resource: "/storage/events/disks/default/0_1_1_0.1.0" (T
hreshold: >= " 3") Execute the following command to obtain event details:
/opt/resmon/bin/resdata -R 173670405 -r /storage/events/disks/default/0_1_1_0.1.
0 -n 173670402 -a
Jun 26 15:58:46 unixdb vmunix:
Jun 26 15:58:46 unixdb vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 183160
396, dev: 1f021000, io_id: 272bc8d
Jun 26 16:11:45 unixdb vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 183238
214, dev: 1f021000, io_id: 272d05a

# -------------------

i'm not sure whether the disk has failed or not

# ioscan --> s/w state CLAIMED
# pvdisplay --> PV State available

but the Event Monitoring Service is saying that the resource is down

I checked physically the disk leds and it looks normal.

NB: the disk has a mirrored copy

can you please help?

Re: Harddisk status

SUDHAKAR_18 — Wed, 08 Jul 2009 06:59:05 GMT

Also check the event log file
/var/opt/resmon/log/event.log

have you added or removed disks from the system recently ?

Re: Harddisk status

DeafFrog — Wed, 08 Jul 2009 07:00:31 GMT

Grep for stale extends , if any ,
#!/bin/sh
##This script created for testing any stale extents in Lvol
##If in the output any stale extents are found,pls report directly to Unix admin
####################################################
##Uncomment the next line if debug mode needed
#set -x
touch ~/if_any_stale
for i in `bdf | awk '{ print $1}'| grep -v Filesystem`
do
lvdisplay -v $i > ~techsup1/if_any_stale
grep -i stale ~/if_any_stale
val=`echo $?`
###echo $val
if [ $val -ne 1 ];then
echo "Logical Volume '$i' had stale volumes"
else
echo "Logical Volume $i OK"
fi
done
rm ~/if_any_stale

check in syslog of message "lbolt", do u find "recovered " in also there.

Re: Harddisk status

Steven E. Protter — Wed, 08 Jul 2009 07:09:29 GMT

Shalom,

The EMS message indicates one of two things:
1) The disk has actually failed. This can be confirmed by dd if=/dev/dsk/c1t1d0 of=/dev/null bs=1024 count=10000 (use a real disk device) or cstm mstm xstm disk exercise function.
2) A hot swap disk has been removed from the system and put back in or switched, which results in an lbolt warning which stays around until either a reboot or a dmesg - or dmesg -c. One of those two commands is HP-UX, the other is Linux. I mix them up and usually end up trying both. They are harmless if done wrong.

SEP

Re: Harddisk status

Mohammad Mostafa — Wed, 08 Jul 2009 07:13:12 GMT

Attached the related events.

No, nothing have been added

Re: Harddisk status

SUDHAKAR_18 — Wed, 08 Jul 2009 07:22:22 GMT

An 146Gb Hard disk at path 0/1/1/0.1.0 showing errors.

"Disk at hardware path 0/1/1/0.1.0 : Media failure"

please observe the eventlog.

Re: Harddisk status

Suraj K Sankari — Wed, 08 Jul 2009 07:25:08 GMT

Hi,
see this error at your output
Disk at hardware path 0/1/1/0.1.0 : I/O request failed.
Disk at hardware path 0/1/1/0.1.0 : Media failure

I think there is a problem with disk
did you check with this things

1.ioscan
2.diskinfo
3.dd
Post output or this command
suppose your disk is /dev/dsk/c1t3d0

#ioscan -fnC disk
#diskinfo /dev/rdsk/c1t3d0
#dd if=/dev/rdsk/c1t3d0 of=/dev/null bs=1024k

Suraj

Re: Harddisk status

Mohammad Mostafa — Wed, 08 Jul 2009 08:05:45 GMT

# dd if=/dev/rdsk/c2t1d0 of=/dev/null bs=1024k count=64 &

output:
# 64+0 records in
64+0 records out

------------------------------------

i have this in the syslog.log

Jun 26 20:29:25 unixdb vmunix: LVM: VG 64 0x000000: PVLink 31 0x021000 Failed! The PV is not accessible.
Jun 26 18:23:24 unixdb vmunix: LVM: VG 64 0x000000: PVLink 31 0x021000 Recovered.

------------------------------------

how to check for stale extends?

Re: Harddisk status

whiz_kid911 — Wed, 08 Jul 2009 09:09:27 GMT

look for stale xtents...

lvdisplay -v | grep -i stale

Re: Harddisk status

SoorajCleris — Wed, 08 Jul 2009 09:30:40 GMT

Please use

#ioscan -fnC disk

and see the status . Please avoid the switch "u" and see the output.

Regards,
sooraj U

Re: Harddisk status

Mohammad Mostafa — Wed, 08 Jul 2009 10:48:25 GMT

# ioscan -fnC disk

disk 1 0/1/1/0.0.0 sdisk CLAIMED DEVICE HP 146 GST3146807L
C
/dev/dsk/c2t0d0 /dev/rdsk/c2t0d0
disk 2 0/1/1/0.1.0 sdisk CLAIMED DEVICE HP 146 GST3146807L
C
/dev/dsk/c2t1d0 /dev/rdsk/c2t1d0
disk 3 0/1/1/1.2.0 sdisk CLAIMED DEVICE HP 146 GST3146807L
C
/dev/dsk/c3t2d0 /dev/rdsk/c3t2d0

-- XXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Also, there are no stale extends

Re: Harddisk status

Shiv007 — Thu, 09 Jul 2009 07:24:20 GMT

Hi,

Please try running the information tool from cstm/xstm/mstm

try the below sequence:
cstm>sel dev
cstm> info
cstm> ial

or try
ioscan -funC disk

else

diskinfo obtained from ioscan -funC disk.

Thanks
Shivaprasad