Disk failure

[k_philemon]

We have a RP2470 HP-Unix Server wheare 11.11 is loaded. It's a database server. database in /dev/vgsg volumegroup.In this vgsg volume group we have 6 HDD configured in raid 1(mirroring) 36*6.
/dev/vgsg
1M-^FM-WM-@>Y^RM-A^F/dev/dsk/c4t0d0
/dev/dsk/c4t1d0
/dev/dsk/c4t2d0
/dev/dsk/c4t3d0
/dev/dsk/c4t4d0
/dev/dsk/c4t5d0.
In which one of the HDD is not sencing now(c4t4d0) 36gb.now i have to replace the hdd of 36gb capacity. what is the procidure to syscronize, what are the steps to be taken. how to do it online.
please help and give me any suggesion

--- Volume groups ---
VG Name /dev/vgsg
VG Write Access read/write
VG Status available, exclusive
Max LV 255
Cur LV 112
Open LV 111
Max PV 16
Cur PV 6
Act PV 5
Max PE per PV 20000
VGDA 10
PE Size (Mbytes) 8
Total PE 21700
Alloc PE 21421
Free PE 279
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0

please help in reverting.

[Raj D.]

Philemon ,

i)What is the status now shows for the disk c4t4d0 in ioscan, does it shows CLAIMED or NO_HW .

ii) If it is a Hotswappable disk.

hth,
Raj.

[k_philemon]

Dear Raj,
Thanks for your reply.

In ioscan disk is not getting senced. and it is a hotswap disk.
root@rp1 /etc>ioscan -fnC disk
Class I H/W Path Driver S/W State H/W Type Description
======================================================================
disk 0 0/0/1/1.15.0 sdisk CLAIMED DEVICE HP 36.4GMAM3367MC
/dev/dsk/c1t15d0 /dev/rdsk/c1t15d0
disk 1 0/0/2/0.1.0 sdisk CLAIMED DEVICE HP DVD-ROM 305
/dev/dsk/c2t1d0 /dev/rdsk/c2t1d0
disk 2 0/0/2/1.15.0 sdisk CLAIMED DEVICE HP 36.4GMAM3367MC
/dev/dsk/c3t15d0 /dev/rdsk/c3t15d0
disk 3 0/4/0/0.0.0 sdisk CLAIMED DEVICE HP 36.4GST336706LC
/dev/dsk/c4t0d0 /dev/rdsk/c4t0d0
disk 4 0/4/0/0.1.0 sdisk CLAIMED DEVICE HP 36.4GST336706LC
/dev/dsk/c4t1d0 /dev/rdsk/c4t1d0
disk 5 0/4/0/0.2.0 sdisk CLAIMED DEVICE HP 36.4GST336706LC
/dev/dsk/c4t2d0 /dev/rdsk/c4t2d0
disk 6 0/4/0/0.3.0 sdisk CLAIMED DEVICE HP 36.4GST336706LC
/dev/dsk/c4t3d0 /dev/rdsk/c4t3d0
disk 8 0/4/0/0.5.0 sdisk CLAIMED DEVICE HP 36.4GST336706LC
/dev/dsk/c4t5d0 /dev/rdsk/c4t5d0
disk 11 0/4/0/0.9.0 sdisk CLAIMED DEVICE HP 73.4GATLAS10K3_73_S
CA
/dev/dsk/c4t9d0 /dev/rdsk/c4t9d0
disk 12 0/4/0/0.10.0 sdisk CLAIMED DEVICE HP 73.4GATLAS10K3_73_S
CA
/dev/dsk/c4t10d0 /dev/rdsk/c4t10d0

[Darrel Louis]

Hi,

Please read the following doc for the steps to follow.
http://docs.hp.com/en/5991-1236/When_Good_Disks_Go_Bad.pdf

Have you already tried to get the disk out and put it back in to let it spin.

Darrel

[Raj D.]

Philemon,

check all the lvm setting , and note down with which lv it is mirroed.

-

# vgreduce -f vgsg
# mv /etc/lvmtab lvmtab.date
# vgscan -v

- Replace the disk ,

After replacing :
# vgcfgrestore -n vgsg /dev/dsk/c4t4d0
# vgchange -a y vgsg

REsync the mirror with the mirror copy .


cheers,
Raj.

[Mridul Shrivastava]

If ioscan is not showing the disk it means disk is dead, Try pulling out the disk and put it again and check if it scans it.

If this doesn't help you have to replace the disk and then follow the steps mentioned by Raj above.

[Adisuria Wangsadinata_1]

Hi Philemon,

If you would like to do it online, you need to check it first whether your harddisk is a hotswappable AND if your system already have the patches that enable LVM online disk replacement. If you dont have both of this condition, better schedule a downtime.

If you have a schedule downtime, just use the steps below :

1. Shutdown the system replace the faulty harddisk

2. Boot the system into multi mode as per usual

3. Use the vgcfgrestore command to restore the LVM configuration data from a configuration backup file to the physical volume:

# vgcfgrestore -n vgsg /dev/rdsk/c4t4d0

4. Activate the volume group for including any physical volumes that were previously listed as missing:

# vgchange -a y vgsg

5. The vgsync command synchronizes the physical extents of each mirrored logical volume in the volume group. Synchronization occurs only on the physical extents that are stale mirrors of the original logical extent.

# vgsync vgsg

NOTE: The synchronization process can be time consuming, depending on the hardware characteristics and the amount of data.

6. Check successful completion by

# pvdisplay -v /dev/dsk/c4t4d0

which should not show up any stale extents and finally by

# vgdisplay -v vgsg

Hope this information can help you.

Cheers,
AW