Hi Bowling, You seem to be on a similar track as me . I have had my espresso this time so my answer on this thread should make more sense.
I have three 9509s serving two sites: redundant switches at production site, single switch at DR site, and use a similar naming convention as you.
Since the zones are unique to the VSAN and I only really manage the zones from that VSANs perspective, I don't put a numerical value to the zone itself but use the following convention: servername_hba#_arrayname-storageport
example for a cluster node: SQL1-N1_HBA0_CX700aSPA0
The even HBA numbers in a host (0,2, etc.) are in VSAN10, and the odd HBA numbers (1,3, etc.) are in VSAN20. VSAN 10 is on Switch 1 and VSAN20 is on Switch 2, so by looking at the HBA number I know which switch and VSAN the port should be logged onto.
I have one back-end (production) VSAN per switch, and have not found a need yet to segregate each switch into more than one VSAN. ( I have a front-end VSAN in each switch but that is for replication, totally different subject). In the production data center there are three storage arrays and about 50 hosts and so far this naming convention has scaled well.
I am open to hear what other people have found that works for them as well.
i like to use either ZNE_Host_hba1_storageSSN_portnumber, for example for CX i would use ZNE_FILESERVER_hba0_0236_SPA0 or ZNE_FILESERVER_hba0_10a1 for DMX. Also my aliases start with ali ..for example ali_FILESERVER_h0. I tell you why i do it. There are instances where i need to generate a report of how many hosts are connected to a particular port, because i use ZNE in my zone names, i simply dump config show to a text file and then filter it by ZNE. It's just easier for me to distinguish my zones from my aliases.
we use Z_hostname_A for 1 fabric and Z_hostname_B for another fabric, this is the standard that was being followed by ancestors in this company...though i would have liked to see them put array name in the zonename as well.
but then again we do not create diff zone for each array port (we create diff zone for each hba port) so we get by with it. with our scheme, we just need to expand the zone to see where the host is connected.
AranH1
2163 Posts
402
0
Posted May 14th, 2007 13:00
You seem to be on a similar track as me
I have three 9509s serving two sites: redundant switches at production site, single switch at DR site, and use a similar naming convention as you.
Since the zones are unique to the VSAN and I only really manage the zones from that VSANs perspective, I don't put a numerical value to the zone itself but use the following convention:
servername_hba#_arrayname-storageport
example for a cluster node:
SQL1-N1_HBA0_CX700aSPA0
The even HBA numbers in a host (0,2, etc.) are in VSAN10, and the odd HBA numbers (1,3, etc.) are in VSAN20. VSAN 10 is on Switch 1 and VSAN20 is on Switch 2, so by looking at the HBA number I know which switch and VSAN the port should be logged onto.
I have one back-end (production) VSAN per switch, and have not found a need yet to segregate each switch into more than one VSAN. ( I have a front-end VSAN in each switch but that is for replication, totally different subject). In the production data center there are three storage arrays and about 50 hosts and so far this naming convention has scaled well.
I am open to hear what other people have found that works for them as well.
Aran