Zoning between a 48 port blade to any other blade

[the_san_man]

So I have this friend that has a couple of 9506s. Each switch has a 16-port, 2 x 32-port and 48-port blades. Everything works fine until he attempts to zone an initiator or target between the 48-port blades and anything else. Does anyone have any ideas. I can get the code level for you.

Any gut reactions? BB credit or anything like that?

[AranH1]

Mike,
Couple of questions for you...

Do they have Gen1 or Gen2 supervisors?

What is the SAN-OS rev?

I have Cisco directors (Gen1 and Gen2 supervisors) with new 24 and 48 port, and older 16 and 32 port all in the same chassis and have not seen any issues creating zones with ports from the newer (24 or 48 port) line cards and the older (16 and 32 port) line cards.

Aran

[bodnarg]

My initial guess is that you have the 48-port blade setup in different VSANson the switch.

Not sure what tools you use - but if you use Device Manager you can quickly check this by clicking the drop-down box that says "VSAN ALL" and selecting each defined VSAN in turn to show how your switch is logically carved-up.

If that is the case you need to collapes your VSANs (if that is what you want) which is a different discussion

If not then we need to dig a bit more...

[Smiley6]

I would suggest a couple of things. Can you verify what SAN OS code you are running on your 9506? For any Gen2 line cards, you will need SAN OS 3.x or higher for the line card to be recognized by the switch. Secondly, you can run a couple of commands on the switch to see what VSANs the ports are in for the 48-port line card. By default, they will be in VSAN 1. If you have your current environment to use a different VSAN, then you will need to move the ports into that VSAN for your production. You can use the Device Manager to see what VSANs the ports are in but I like CLI. The CLI command you want to run is:

Switch-Prompt# show VSAN membership <enter>

I would also suggest doing a "show module" to verify the line card is active. Just a couple of suggestions. If it is something else, we can dig deeper as well.

[the_san_man]

To answer all of the questions:

They gen1 sups
SAN OS 3.x (can't remember exactly, I think 3.1.2b maybe I'll call him and check).
The ports were preconfigured from VSAN1 to the proper VSAN when they were first installed.

If the intiators and targets are on the same 48port blade, everything is fine. They can "see" each other and access storage. It's only when they try to zone to/from anything in those 48port blades. I don't recall if I mentioned that there are two 9506s and they both have the same symptoms. Identical in fact.

[Smiley6]

If you could send the output of the "show vsan membership" on the switches and if you could let us know what devices (or ports) that you are trying to communicate with the devices in the 48-port line card. Because what you are saying doesn't make sense on why it's not communicating.

Also, just curious, are there other switches in this network or is it just a single switch SAN?

[RickBrunner1]

Get the following ouput from the switch, that why you do not have to go back and ask for other output.
terminal length 0
sh tech-support

Provide the output of the show flogi, show zoneset, show fcns or post the sh tech-support to an internal EMC location we can access, that way if we need to look at logs we can.

[ConnectrixHelpe]

I woud agree that the most likely situation would be different VSANs.

There are performance related configuraiton constraints on the 48 port cards but that shouldn't affect you in this situation. These constraints are more related to ISLs.

If logs need to be reviewed to isolate what the problem might be, I would recommend opening a support case with EMC to further investigate it.

Thank you.

[the_san_man]

First, I've asked them to take a look at this thread so they can provide first hand information.

It would also appear logical to be a VSAN assignment, however, I configured all of the ports to be on the same VSAN as the rest of the ports in the switch before I left the company. so I know that is not what it is. In fact, I had attended training with my friend that has these switches before I left and I had asked the instructor at that time regarding this issue, and she didn't have any clues either. If memory serves me, she thought it was a BB credit issue. This has been over 6 months ago though, but I even recall out of servicing ports/disabling ports down to just a few thinking it might be a bandwidth issue.

I will let them know that they may need to go ahead and open a case (like I suggested before I left the company). If I find out the outcome, I will post it here.

[the_san_man]

Ok, let me sstart by saying "don't shoot the messenger".

I asked my friends to send me the requested info, and they replied "we got it fixed." They told me that the only thing they did was to failover the sups multiple times and it just miraculously started working. Both 9506s did the same thing. it really makes me wonder about that, but the issue is resolved and they seem to be satisfied. I don't relieve they opened a case with EMC and how they got to that point, I have no idea.

Anyway, thanks to everyone for providing their input.