In C-Mode, i would like to know what i should use in my zoning : the vserver wwpn or each LIF of node ?
If i should use the LIF, what is the interest of vserver fcp ?
Actually i am using the wwpn of vserver :
> vserver fcp showStatusVserver Target Name Admin---------- ---------------------------- ------svm1-sd 20:05:00:a0:98:54:91:02 up
And on my brocade switch i have this :
zone: SRV_BCK_0120:05:00:a0:98:54:91:02 (netapp vserver)c0:50:76:03:ba:5a:00:28 (host)
But sometimes i see on the web the LIF of each port which use :
zone : SRV_BCK_01
c0:50:76:03:ba:5a:00:28 (host)20:01:00:a0:98:54:91:02 (nodea_3a_lif1)20:02:00:a0:98:54:91:02 (nodea_3b_lif1)20:03:00:a0:98:54:91:02 (nodeb_3a_lif1)20:04:00:a0:98:54:91:02 (nodeb_3b_lif1)
See The Solution
each vserver is its own virtual storage server.
As such, each vserver has a WWNN (world wide node name) in order to identify the storage system and several WWPNs (world wide port names) for the individual LIFs.
cDOT uses NPIV in order to map several WWPNs of different vservers (as you can have multiple) to the same physical port.
In order to match SAN best practice you need to create single initiator zones that contain the LIF WWPNs, not the vserver WWNN.
Just think about a vserver as an array and the LIFs are the arrays's ports.
So this one is right:
View solution in original post
Thank you very much for informations.