[Novalug] Raid question

[Megan Larko]

Happy Thursday!

I have a machine running CentOS 5 kernel 
2.6.18-53.1.13.el5_lustre.1.6.4.3smp.

The room which houses it had an electrical failure yesterday afternoon. 
  Only one of the two raid units came up.    I am trying to figure out 
how to get the information from the non-existent raid.  The other raid 
array is fine.

The /etc/mdadm.conf file looks like this:
MAILADDR larkoc at iges.org
DEVICES partitions
ARRAY /dev/md0 level=raid1 num-devices=2 
UUID=4d41eaf5:e7835635:7185fe85:85ab60b5
ARRAY /dev/md1 level=raid1 num-devices=2 
UUID=1de037bb:01cca9a5:8c4af28f:05188493

The UUIDs are on one single line with "ARRAY"; my email program wraps.

As mentioned, one array is fine:
[root at mds1 ~]# cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb[0] sdd[1]
       244129728 blocks [2/2] [UU]

unused devices: <none>

The physical disks seem okay on the 3ware controller (from ./tw_cli):
//mds1> focus c0

//mds1/c0> show

Unit  UnitType  Status         %Cmpl  Stripe  Size(GB)  Cache  AVerify 
IgnECC
------------------------------------------------------------------------------
u0    RAID-1    OK             -      -       74.4951   ON     ON       OFF
u1    SINGLE    OK             -      -       232.82    OFF    OFF      -
u2    RAID-1    OK             -      -       232.82    OFF    ON       OFF
u3    SINGLE    OK             -      -       232.82    OFF    OFF      -
u4    SINGLE    OK             -      -       65.1826   OFF    OFF      -
u5    SINGLE    OK             -      -       65.1826   OFF    OFF      -

Port   Status           Unit   Size        Blocks        Serial
---------------------------------------------------------------
p0     OK               u4     69.25 GB    145226112     WD-WMANS1505420
p1     OK               u5     69.25 GB    145226112     WD-WMANS1484929
p2     OK               u1     232.88 GB   488397168     3ND2W1KL
p3     OK               u3     232.88 GB   488397168     3ND2WAZS
p4     OK               u2     232.88 GB   488397168     3ND2VKWP
p5     OK               u2     232.88 GB   488397168     3ND2WASB
p6     OK               u0     74.53 GB    156301488     5PS24ZWJ
p7     OK               u0     74.53 GB    156301488     5PS1Y0A1

The standard mdadm commands are failing because /dev/md1 does not exist.
[root at mds1 ~]# mdadm -A --scan
mdadm: error opening /dev/md1: No such file or directory

[root at mds1 ~]# mdadm -A -R /dev/md1
mdadm: error opening /dev/md1: No such file or directory

Why am I getting /dev/md1: No such file or dir?   The block device is 
not there now but I thought the raid program started it?  As the raid 
was a simple mirror,  how do I find out which device (sdg?  sdf?  other? 
contained the information so I could perhaps mount that device to check 
the data?

Opinion:  I liked the old-style mdadm.conf (raidtab anyone?) that had 
the devices listed explicitly in the config file.

Thanks for you suggestions.

megan