Topic: Softare RAID problems - at a loss

[Mahmood Al-Yousif]

Hi, after installing and initializing the raidmonitor (raidmonitor -iv) I get the following report that clearly shows that one disk is either not being utilized or failed:

...................
RAID Monitor Report
...................

Current /proc/mdstat saved in /root/raidmonitor/mdstat:

Personalities : [raid1]
read_ahead 1024 sectors
md2 : active raid1 hda3[0]
      264960 blocks [2/1] [U_]

md1 : active raid1 hda2[0]
      38708544 blocks [2/1] [U_]

md0 : active raid1 hda1[0]
      104320 blocks [2/1] [U_]

unused devices:

Current partition info saved in /root/raidmonitor/sfdisk.out:

# partition table of /dev/hda
unit: sectors

/dev/hda1 : start=       63, size=  208782, Id=fd, bootable
/dev/hda2 : start=   208845, size=77417235, Id=fd
/dev/hda3 : start= 77626080, size=  530145, Id=fd
/dev/hda4 : start=        0, size=       0, Id= 0
# partition table of /dev/hdc
unit: sectors

/dev/hdc1 : start=       63, size=  208782, Id=fd, bootable
/dev/hdc2 : start=   208845, size=77417235, Id=fd
/dev/hdc3 : start= 77626080, size=  530145, Id=fd
/dev/hdc4 : start=        0, size=       0, Id= 0

I have read the how-to, but when I tried to import hdc.out file whose contents are:

# partition table of /dev/hdc
unit: sectors

/dev/hdc1 : start=       63, size=  208782, Id=fd, bootable
/dev/hdc2 : start=   208845, size=77417235, Id=fd
/dev/hdc3 : start= 77626080, size=  530145, Id=fd
/dev/hdc4 : start=        0, size=       0, Id= 0

I get an error that says the disk is in use, hence won't allow me to continue.

I'd apprecaite your comments on how to fix this problem.

regards
mahmood

[Mahmood Al-Yousif]

Sorry this is on a fresh install of 5.6.

[Nathan Fowler]

Try to raidhotadd the devices...

I'm assuming your secondary HDD is /dev/hdb (change commands to suit if this isn't the case)

raidhotadd /dev/md2 /dev/hdb3
raidhotadd /dev/md1 /dev/hdb2
raidhotadd /dev/md0 /dev/hdb1

Note that I may have the "/dev/hdb3" and "/dev/md2" values transposed, if this is the case just switch them around.  I'm not infront of my console at the moment so I can't look.

Then cat /proc/mdstat and see if they start rebuilding.

Hope this helped,
Nathan

[Mahmood Al-Yousif]

Thank you very much for that. the hd2 and 1 worked, however if I:

raidhotadd /dev/md0 /dev/hdc1

it says "/dev/md0: can not hot-add disk: invalid argument."

This is what I now get from /proc/mdstat:

Personalities : [raid1]
read_ahead 1024 sectors
md2 : active raid1 hdc3[1] hda3[0]
      264960 blocks [2/2] [UU]

md1 : active raid1 hdc2[2] hda2[0]
      38708544 blocks [2/1] [U_]
      [=>...................]  recovery =  5.7% (2216404/38708544) finish=61.6min speed=9858K/sec
md0 : active raid1 hda1[0]
      104320 blocks [2/1] [U_]

unused devices:



I wonder if I have to wait for it to finish mirroring then try again to add the md0 mirror.

[Mahmood Al-Yousif]

okay the recovery has now finished, but still when I try to add md0 I get the following in the /var/log/messages:

Feb 22 11:05:58 server kernel: md: trying to hot-add hdc1 to md0 ...
Feb 22 11:05:58 server kernel: md: can not import hdc1, has active inodes!
Feb 22 11:05:58 server kernel: md: error, md_import_device() returned -16

[Mahmood Al-Yousif]

appreciate your help with this one... still getting the dreaded

/dev/md0: can not hot-add disk: disk busy!

when I:  
raidhotadd /dev/md0 /dev/hda1

mahmood