During the delivery of a IBM SVC Split cluster with ISL solution, unfortunately for me, I came across a limitation of the SAN Health, who has “difficulty” to discover the topology of the SAN configured through the use of logical switch within physical switch. The design that I wanted to get automatically through the use of tools Brocade SAN Health is the following:
I supposed to use SAN Health as used in the past, just configuring the FC switches inside the tool using their Management IP address. The problem is that multiple logical switch are referring to same management IP address, in addition is those are interconnected via ISL to other logical switch, it cause confusion in SAN Health discovery.
Asking to my Brocade friend, I got the following circumvention.
On physical switches with ip add: 10.200.2.122 and ip add: 10.200.3.122 with fid: 22 and 26, that are the two UP in the above picture I have configured the new following userid
userconfig –add baseadmin22 -r admin -l 22 -h 22 -c admin
userconfig –add baseadmin26 -r admin -l 26 -h 26 -c admin
userconfig –add baseadmin23 -r admin -l 23 -h 23 -c admin
userconfig –add baseadmin27 -r admin -l 27 -h 27 -c admin
On physical switches with ip add: 10.200.2.124 and ip add: 10.200.3.123 with fid: 28 and 24, that are the two DOWN in the above picture I have configured the new following userid
userconfig –add baseadmin28 -r admin -l 28 -h 28 -c admin
userconfig –add baseadmin24 -r admin -l 24 -h 24 -c admin
userconfig –add baseadmin29 -r admin -l 29 -h 29 -c admin
userconfig –add baseadmin25 -r admin -l 25 -h 25 -c admin
The idea was to be able to access to each logical switch in the fabric with a different userid to be able to discover its counterpart. In that way I had to run the SAN Health twice in order to get two different report, one for my SAN named Priv and one for my SAN named Pub.
At the moment I asked to Brocade if this is a real SAN Health limitation and when it will be fixed….or just my wrong usage !! I will let you know !