Hi All,
I am trying to implement smart zoning in a lab environment before considering it for our production environment.
Switch platform (for the time being) is MDS 9124. Running code for lab is 5.2(8b).
It seems that documentation on SZ is pretty parse besides the basic CLI config guides.
I am wondering what the recommendation is on zone population, IE, it was single init single target, but now should it be multi init single target (IE zone per SP Port), single init multi target (IE zone per host), or multi init multi target (IE zone per cluster).
Here is the example:
1x FC Array with 2 SP's. Each SP has a connection to each fabric...
Odd Fabric
SP A1
SP B1
Even Fabric
SP A0
SP B0
4x ESXi hosts with a single connection to each fabric...
Odd Fabric
ESXi01-A1
ESXi02-A1
ESXi03-A1
ESXi04-A1
Even Fabric
ESXi01-B1
ESXi02-B1
ESXi03-B1
ESXi04-B1
From what I have been reading (specifically this PDF:
http://www.cisco.com/en/US/prod/collateral/ps4159/ps6409/ps5987/at_a_glance_c45-708533.pdf), it sounds like Cisco is recommending the zone per cluster model I mentioned above, with the thought that this will actually act like single init single target.
This would mean my Smart Zones would look like this:
Odd Fabric "SmartZone_Odd_Array1234_vSphereCluster"
SP A1
SP B1
ESXi01-A1
ESXi02-A1
ESXi03-A1
ESXi04-A1
Even Fabric "SmartZone_Even_Array1234_vSphereCluster"
SP A0
SP B0
ESXi01-B1
ESXi02-B1
ESXi03-B1
ESXi04-B1
Anyone have any thoughts on this approach and its supportability / sustainability from Cisco's perspective?
Thanks,
Jon
