Keep losing RAID device

Chris Mason lists at masonc.com
Thu Aug 12 12:36:43 UTC 2004


I have a redhat 9 server with 6 drives, four of which are 160GB drives in a
RAID 5 config. Each time I reboot I loose one of the raid devices. However,
I can reinsert the drive without problem and it is synced again. However, as
it takes nearly four hours to resync, this is a pain. Any ideas why the
drive is being rejected on boot? I don't see any problems with the drive
otherwise and I run smartd to monitor all the drives.

[root at munster root]# dmesg|grep hdg

    ide3: BM-DMA at 0x7008-0x700f, BIOS settings: hdg:pio, hdh:pio
hdg: WDC WD1600JB-00EVA0, ATA DISK drive
hdg: attached ide-disk driver.
hdg: host protected area => 1
hdg: 312581808 sectors (160042 MB) w/8192KiB Cache, CHS=19457/255/63,
UDMA(100)
 hdg: unknown partition table
md0: former device hdg is unavailable, removing from array!
md0: former device hdg is unavailable, removing from array!
md: trying to hot-add hdg to md0 ... 
md: bind<hdg,4>
md: hdg [events: 00000055]<6>(write) hdg's sb offset: 156290816
md0: resyncing spare disk hdg to replace failed disk


[root at munster root]# mdadm --add /dev/md0 /dev/hdg

mdadm: hot added /dev/hdg

[root at munster root]# cat /proc/mdstat
Personalities : [raid5] 
read_ahead 1024 sectors
md0 : active raid5 hdg[4] hde1[0] hdi1[2] hdk1[3]
      468864768 blocks level 5, 32k chunk, algorithm 2 [4/3] [U_UU]
      [>....................]  recovery =  0.0% (88552/156288256)
finish=235.0min speed=11069K/sec
unused devices: <none>



Chris Mason





More information about the redhat-list mailing list