Comment 4 for bug 371338

I've got a similar problem trying to break down a RAID1 array to use the individual partitions for swap. From what I understand the array was created during installation and marked as swap, then had a single partition removed from it (currently activated as swap).

$ uname -a
Linux adlai 2.6.24-23-server #1 SMP Mon Jan 26 00:55:21 UTC 2009 i686 GNU/Linux

$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=8.04
DISTRIB_CODENAME=hardy
DISTRIB_DESCRIPTION="Ubuntu 8.04.2"

$ mdadm --version
mdadm - v2.6.3 - 20th August 2007

$ sudo fdisk -l

Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x9e169e16

   Device Boot Start End Blocks Id System
/dev/sda1 * 1 973 7815591 fd Linux raid autodetect
/dev/sda2 974 31367 244139805 fd Linux raid autodetect
/dev/sda3 60668 60801 1076355 82 Linux swap / Solaris
/dev/sda4 31368 60667 235352250 fd Linux raid autodetect

Partition table entries are not in disk order

Disk /dev/sdb: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x07a6b49b

   Device Boot Start End Blocks Id System
/dev/sdb1 * 1 973 7815591 fd Linux raid autodetect
/dev/sdb2 974 31367 244139805 fd Linux raid autodetect
/dev/sdb3 60668 60801 1076355 82 Linux swap / Solaris
/dev/sdb4 31368 60667 235352250 fd Linux raid autodetect

Partition table entries are not in disk order

(snip)

$ cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md3 : active raid1 sda2[0] sdb2[1]
      244139712 blocks [2/2] [UU]

md4 : active raid1 sda4[0] sdb4[1]
      235352128 blocks [2/2] [UU]

md2 : active raid1 sdb3[1]
      1076288 blocks [2/1] [_U]

md0 : active raid1 sda1[0] sdb1[1]
      7815488 blocks [2/2] [UU]

unused devices: <none>

$ swapon -s
Filename Type Size Used Priority
/dev/sda3 partition 1076344 112 -1

$ sudo mdadm --stop /dev/md2
mdadm: fail to stop array /dev/md2: Device or resource busy

$ sudo mdadm /dev/md2 --fail --verbose /dev/sdb3
mdadm: set /dev/sdb3 faulty in /dev/md2

$ sudo mdadm /dev/md2 --remove --verbose /dev/sdb3
mdadm: hot remove failed for /dev/sdb3: Device or resource busy

The only other thing I can think of is booting from a livecd and zeroing the superblock / running mkswap, but that will probably take a while to schedule downtime for.