RAID mdadm

[seabird]

Hi everyone,

Fedora 16 keeps reporting one of my drives as bad. I pulled the disk out of my array and ran several test onto it, but all seems fine so I will ignore the warning. The problem I have now is that the disk doesn't want to be re-added into /dev/md0

First off, I failed and removed the drive:

mdadm /dev/md0 --fail /dev/sdd1
mdadm /dev/md0 -r /dev/sdd1

When trying to re-add it I get this:

 mdadm /dev/md0 -a /dev/sdd1
mdadm: /dev/sdd1 reports being an active member for /dev/md0, but a --re-add fails.
mdadm: not performing --add as that would convert /dev/sdd1 in to a spare.
mdadm: To make this a spare, use "mdadm --zero-superblock /dev/sdd1" first.

I do not want it to become a spare but a active drive again.

Where am I going wrong?

[seabird]

ok, seems to re-add after a superblock zero

mdadm --zero-superblock /dev/sdd1

[droidhacker]

Yep, makes perfect sense.
Here's the problem; once the data on the removed disk is no longer synchronized with the rest of the array, it needs to be regenerated anyway. It is, therefore, unwilling to destroy what amounts to a *backup* of your array, or to pollute the still-running array, by readding a disk that is out of sync with the array.

Also... it seems that it told you to zero it out. Maybe next time pay attention to what it is telling you? Those messages aren't just for fun....

[Briago]

droidhacker: To be fair to seabird, the actual message says to zero-superblock if you want to add as a spare (I read this as a hot-spare, not active). It doesn't say that this is what you should do to sucessfully readd it to the active array.

After reading this post, I tried the zero-superblock command and it is now sync'ing the raid again.

Thank you for confirming the use of zero-superblock.