2 arrays sharing same disks, one shows removed other does not?

[FLATcura]

i built a centos 5.4 server for samba using 5 1tb drives i had left over.

i decided to use software raid across all 5 disks

partitioned /boot (md0) and the swap (md1) as raid 1
cloning the 200mb partitions on all 5 disks
partitioned / (md2) as the rest of the space as raid 5
again cloning on all 5 disks

everything seemed perfect until i wanted to see how it would respond to a drive failing, so i shut it down and unplugged /dev/sde

powered it back up, everything looked great
powered off, reattached cable, powered back up and...

md0 and md2 show sda has been removed however md1 showed sda is all fine..

im so mind boggled by the fact that only the swap array sees sda, and that the one i unplugged was sde..

any insight is greatly appreciated

thanks for your time

 

[FLATcura]

ok, so i ran hdparm -I /dev/sd*

and discovered that sda was in fact the one i unplugged not sde

however when i booted the machine just after unplugging the disk, cat /proc/mdstat showed me sde was not in the list..

excuse my greenness

 

[FLATcura]

i just had to re add two of them and allow them to recover, still confused why the swap partition didnt need to be readded, but problem solved

mods remove thread if you like

/i feel shame