Areca test 2

• first see areca_test_1

Abstract

check if the controller repair the drive while reading bad sectors

HDD prepare

• use the same setup like in Areca test 1
• create a file with partition size

before:

/dev/sda8 2.0G 36M 1.9G 2% /opt

 dd if=/dev/urandom of=r.data bs=1M

1979+0 records in

1978+0 records out

2074959872 bytes (2.1 GB) copied, 325.844 seconds, 6.4 MB/s

after

/dev/sda8 2.0G 2.0G 0 100% /opt

• copy it to another node
• find out which secotrs are used

 fdisk -lu /dev/sda 

/dev/sda8 52494368 56688639 2097136 83 Linux

=> start sector: 52494368 => end sector: 56688639

• damage the whole drive!!!

do_all_bad.sh

 #!/bin/bash
 # confiure the script here THINK TWICE
 HARD_DRIVE=/dev/sdb
 LBA_SECTORS=1465149168
 # DO NOT TOUCH THE LINE BELOW
 until [ $LBA_SECTORS -lt 5 ]; do
 make_bad_sector $HARD_DRIVE $LBA_SECTORS
 let LBA_SECTORS-=2;
 done

• • this was too slow,so a new loop is running with step 10000 (aprox. 5MB) using damage tool ultrafast_make_bad_sector

HDD end

• the drive is back in the storage server

Volume Set Information

Volume Set Name SYSTEM

Raid Set Name RAID

Volume Capacity 1500.0GB

SCSI Ch/Id/Lun 0/0/0

Raid Level Raid 6

Stripe Size 16KBytes

Block Size 512Bytes

Member Disks 4

Cache Mode Write Back

Tagged Queuing Enabled

Volume State Degraded

• the red front led of the HDD case is lighting and there is a strange sound

from the event table

2008-1-31 14:22:4 IDE Channel 3 Device Failed(SMART)

• copy the r.data file to the other node again and compare

 cmp --verbose r.data r_after_mbs.data 

no output => same data

• reboot -> the problem with the drive remains

• try to check the volume

 cli64-1.72 vsf check vol=1

but

GuiErrMsg<0x43>: VolumeSet Not Normal.

• the hard drive is working again, but i needed to write some special values from /dev/zero to the drive using dd, while connected through USB, so that the SMART system cannot turn the drive off in the boot procedure