replace root mirror drive

[Jerry_109]

# uname -a ; model
HP-UX B.11.11 U 9000/800/A500-7X
+++++++++++++++++++++++++++++++++++

Hello HP,

Can you delivery the correct information for replacing a faulty mirror drive ? I received information ( listed below ) indicating drive at H/W path :
disk 2 0/0/2/1.15.0 sdisk CLAIMED DEVICE HP 36.4GMAM3367MC
/dev/dsk/c3t15d0 /dev/rdsk/c3t15d0

needs to be replaced. Lbolt errors were received in the syslog.log

+++++++++++++++++++++++++++++++++++++++++++++

Feb 28 07:05:59 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 76593339, dev: 1f03f000, io_id: 30aa9f7
Feb 28 08:28:41 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 77089539, dev: 1f03f000, io_id: 30addfd
Mar 4 04:49:41 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 110335639, dev: 1f03f000, io_id: 31352d8
Mar 6 21:44:07 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 133702339, dev: 1f03f000, io_id: 31cafc2
Mar 6 23:31:06 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 134344239, dev: 1f03f000, io_id: 31cf059
Mar 7 07:28:33 hohp08 vmunix: SCSI: Request Timeout; Abort Tag -- lbolt: 137208939, dev: 1f03f000, io_id: 31e2607
root@hohp08[/var/adm/syslog]
# /etc/ioscan -nfC disk
Class I H/W Path Driver S/W State H/W Type Description
==========================================================================
disk 0 0/0/1/0.3.0 sdisk CLAIMED DEVICE HP DVD-ROM 305
/dev/dsk/c0t3d0 /dev/rdsk/c0t3d0
disk 1 0/0/1/1.15.0 sdisk CLAIMED DEVICE HP 36.4GMAM3367MC
/dev/dsk/c1t15d0 /dev/rdsk/c1t15d0
disk 2 0/0/2/1.15.0 sdisk CLAIMED DEVICE HP 36.4GMAM3367MC


+++++++++++++++++++++++++++++++++++++++++++
lbot errors on hohp08 is due to disk media error
0/0/2/1.15.0 SCSI Disk (HP36.4GMAM3367 Information Successful

Error log:
Product Id: MAM3367MC Vendor: HP 36.4G
Device Type: SCSI Disk Firmware Rev: HP00
Device Qualifier: HP36.4GMAM3367MC Logical Unit: 0
Serial Number: KP900290
Capacity (M Byte): 34732.89
Block Size: 512
Max Block Address: 71132959
Error Logs
Total Retries: 609951 Buffer Overruns: N/A
Read Reverse Errors: N/A Buffer Underruns: N/A
Write Errors: 0 Non-Medium Errors: 18
Verify Errors: 0

+++++++++++++++++++++++++++++++++++++++++++
# lvlnboot -v
Boot Definitions for Volume Group /dev/vg00:
Physical Volumes belonging in Root Volume Group:
/dev/dsk/c1t15d0 (0/0/1/1.15.0) -- Boot Disk
/dev/dsk/c3t15d0 (0/0/2/1.15.0) -- Boot Disk
Boot: stand on: /dev/dsk/c1t15d0
/dev/dsk/c3t15d0
Root: root on: /dev/dsk/c1t15d0
/dev/dsk/c3t15d0
Swap: swap on: /dev/dsk/c1t15d0
/dev/dsk/c3t15d0
Dump: swap on: /dev/dsk/c1t15d0, 0

Current path "/dev/dsk/c19t0d0" is an alternate link, skip.
+++++++++++++++++++++++++++++++++++++++++++
Replacing bad root mirror drive
1. Shutdown the server , replace the faulty disk.
2. Boot normally.If error because of quorum

reboot
interrupt when "10 sec is allowed"
bo pri isl
interact with isl "y"
ISL> hpux -lm

3. Identify the new disk , with ioscan.
4. vgcfgrestore -n /dev/vg00 /dev/rdsk/c2t6d0
5. vgchange -a y /dev/vg00
6. mkboot -l /dev/rdsk/c2t6d0
7. mkboot -a "hpux -lq (;0) /stand/vmunix" /dev/rdsk/c2t6d0
8. lvlnboot -R /dev/vg00
9. lvlnboot -v /dev/vg00 [ Verify ]
10. vgsync /dev/vg00
[ Wait till synchronizaion completes. ]
11. setboot
[ check , the pri and alt path. ]
12.reboot and check

[Geoff Wild]

You shouldn't have to reboot is it is hot swap!

After failed disk has been replaced by HP (assuming the disk is c1t5d0):

vgcfgrestore -n /dev/vg00 /dev/rdsk/c1t5d0

mkboot /dev/rdsk/c1t5d0

mkboot -a "hpux -lq (;0) /stand/vmunix" /dev/rdsk/c1t5d0

vgchange -a y /dev/vg00

lvlnboot -Rv /dev/vg00

vgsync /dev/vg00

NOTE:
1) If the disk is completely "dead", such as if you run ioscan and status is "no_hw" then you can hot swap the disk online.

2) However in circumstances where the disk has not fully failed please do one of the following to avoid data corruption :

a) reduce mirror before replacing the disk
b) deactivate VG before replacing the disk
c) shutdown system to replace the disk

Rgds...Geoff

[Jerry_109]

are we talking :

lvreduce -m 0 /dev/vg00/lvolX /dev/dsk/cxtxdx

all the lv's for vg00 ?

[Geoff Wild]

You only have to lvreduce if the disk isn't completly dead....

I had a year and a half ago when we replaced a disk that wasn't completing dead - could not re-mirror until after a lot of pain/single user modes/etc....

It is good practice to just lvreduce all the lvols and vgreduce the disk, then swap it out, then recreate the root mirror...

Rgds...Geoff

[Jerry_109]

Can I vgreduce the mirror drive while online?
I mean, I don't have to vgchange -y n /dev/vg00 before vgreduce ?

[Sivakumar TS]

Hi,

No need to deactivate a VG before doing vgreduce.

Regards,

Siva.

[Sivakumar TS]

Hi,

vgreduce cannot be performed if the volume group is activated in shared mode.

Regards,

Siva.

[Tiziano Contorno _]

Remember to use -k option when reducing a bad disk, it avoids hanging the command just in case the disk is not responding.

From man: "-k This option should be used only in the special instance when you want to reduce a mirrored logical volume on a physical volume that is missing or has failed"

I think it's a force mode.

lvreduce -k -m 0

Regards.

[Bill Hassell]

Partially dead disks can be a pain but with hotswap disks, you can pull out the bad one. That gets rid of whatever hardware failure was occurring. Then check syslog to see that LVM has discovered the missing disk (now = dead/missing) so lvreduce -m 0 ... is not needed.