Raid: Mdadm error

velvetbulldozer · 02-27-2012, 12:31 PM

Hi guys,

I have a 1TB raid5 array of 3 disks running on Centos 6.2, and one disk got lost.(this a test bench of 3 usb hd)

When I try to reconnect the missing drive I get the error: ¨Error adding spare: mdadm exited with exit code 2: mdadmption --force not valid in manage mode¨

It seems like mdadm thinks I am about to add a spare, but I am not.

I am using Red Hat disk utility 2.30.1. The array is working with clients writing to it.

Any suggestions how to solve this?

Thanks,

Dan

troop · 02-27-2012, 01:30 PM

You should remove failed disk from array. mdadm /dev/mdX -r /dev/sdX
Then re-add.

velvetbulldozer · 02-27-2012, 01:43 PM

Thanks.
I get the following error:

¨mdadm /dev/md127 -r /dev/sdc1
mdadm: hot remove failed for /dev/sdc1: No such device or address¨

Any ideas?

Quote:

Originally Posted by troop

You should remove failed disk from array. mdadm /dev/mdX -r /dev/sdX
Then re-add.

troop · 02-28-2012, 03:15 AM

post mdadm -D /dev/md127 here

velvetbulldozer · 02-28-2012, 10:02 AM

After reboot i managed to attach a new drive. During the re-sync process it lost the new drive again and the whole array became extremly slow (200Kb read/write). I have a raid one in the lab now with a new error. I make a new post.
Thanks.

Quote:

Originally Posted by troop

post mdadm -D /dev/md127 here