Hi folks,

I am having some problems defining multipathing on servers I am preparing
for
clustering. I am relatively a Linux novice (with a lot of NetWare experience
:-( )
and could really use some help.

I want to set up a two node OES SP2 cluster on two HP BL20P blade servers.
The shared storage is on an EVA 5000, which has two controllers, hence the
need for multipathing - even though only one of the QLogic HBAs is
connected on the servers.

I am attaching the multipath.conf file which I cobbled up. I have no idea if
it is correct.
As I said, I AM a Linux novice.

When I run the hp ultility hp_rescan -a on the first server, I see two raid
controllers and
twice the two LUNs that I have defined to the server:

WIS-eDir1:~ # hp_rescan -a
Sending rescan signal to /proc/scsi/qla2xxx/0...
Sending rescan signal to /proc/scsi/qla2xxx/1...
Scanning /proc/scsi/qla2xxx/0, target 1, LUN 32


scsi0 00 00 00 COMPAQ HSV110 3020 RAID
scsi0 00 00 01 COMPAQ HSV110 3020 Direct-Access
scsi0 00 01 00 COMPAQ HSV110 3020 RAID
scsi0 00 01 01 COMPAQ HSV110 3020 Direct-Access
scsi0 00 00 02 COMPAQ HSV110 3020 Direct-Access
scsi0 00 01 02 COMPAQ HSV110 3020 Direct-Access


When I run multipath -l, I see in the output that obviously something is
wrong:

WIS-eDir1:~ # multipath -l
dm names N
dm table eva0 N
dm table eva0 N
dm status eva0 N
dm info eva0 O
dm table eva1 N
dm table eva1 N
dm status eva1 N
dm info eva1 O
eva0 (3600508b400010fa10001900002410000)
[size=10 GB][features="1 queue_if_no_path"][hwhandler="0"]
\_ round-robin 0 [enabled]
\_ 0:0:0:2 sdc 8:32 [failed][faulty]
\_ round-robin 0 [active]
\_ 0:0:1:2 sdd 8:48 [active][ready]

eva1 (3600508b400010fa100019000023b0000)
[size=1 GB][features="1 queue_if_no_path"][hwhandler="0"]
\_ round-robin 0 [active]
\_ 0:0:0:1 sda 8:0 [active][ready]
\_ round-robin 0 [enabled]
\_ 0:0:1:1 sdb 8:16 [active][faulty]


Thanks in advance! Any help would be greatly appreciated.

Steve Druck
Weizmann Institute of Science