HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
BladeSystem - General
cancel
Showing results for 
Search instead for 
Did you mean: 

VC FC Errors and ESX 3.5 Servers

 
jcarlson77
Occasional Visitor

VC FC Errors and ESX 3.5 Servers

Hi everyone,

Last week my SAN admin and I were working on reconfiguring some disk in our environment. When the disk work was done, I looked for the LUN on the ESX hosts and noticed that nearly all 8 of my ESX hosts in 2 c7000's are having path issues.

My SAN admin says that there is nothing wrong with the SAN and he can see every WWN.

When I look at the VMware side, I see dead paths on some links, or no paths at all (yet VM's are still running on these ESX hosts).

When I look at the VC logs, I see a lot of strange errors about the FC state being degraded or "uplink ports have different SAN fabric WWNs." However there isn't much information to go on with these warnings.

Before I open a case with HP and/or VMware, I was hoping someone out there could shed some light on these errors in the attached log.

Here is the firmware revision list.

OA - 2.41
iLO - 1.77 (all 4 servers in chassis)
VC Ethernet - 2.10
VC Fiber - 1.32

Thanks in advance.
6 REPLIES
JKytsi
Honored Contributor

Re: VC FC Errors and ESX 3.5 Servers

Hi,

What happened 2009-10-11T22:30:23-05:00 ?

Does Your VCM show all OK ?
Remember to give Kudos to answers! (click the KUDOS star)

You can find me from Twitter @JKytsi
JKytsi
Honored Contributor

Re: VC FC Errors and ESX 3.5 Servers

And this "uplink ports have different SAN fabric WWNs" would be something that in your VC specified SAN the uplinks are not going to same SAN. Ask your SAN admin about that.
Remember to give Kudos to answers! (click the KUDOS star)

You can find me from Twitter @JKytsi
jcarlson77
Occasional Visitor

Re: VC FC Errors and ESX 3.5 Servers

One of my backup ethernet networks has 2 links that are down (I suspect bad cables or not fully seated in the core switch).

Both SAN fabrics show "Degraded" but all ports are green and have an "OK" status.

Everything else looks fine.
jcarlson77
Occasional Visitor

Re: VC FC Errors and ESX 3.5 Servers

And this "uplink ports have different SAN fabric WWNs" would be something that in your VC specified SAN the uplinks are not going to same SAN. Ask your SAN admin about that.

Can you explain a little more about the uplinks not going to the same SAN? Would that be a FC switch configuration?
JKytsi
Honored Contributor

Re: VC FC Errors and ESX 3.5 Servers

yes it would be. As in Virtual Connect You create a SAN and assign uplinks to it. All uplinks inside same VC created SAN must go to same fabric.

If You have two FC modules, You can't connect both to same fabric
Remember to give Kudos to answers! (click the KUDOS star)

You can find me from Twitter @JKytsi
The Brit
Honored Contributor

Re: VC FC Errors and ESX 3.5 Servers

As pointed out by Jarkko, if you create a "SAN" in VC, and assign ports to it, then all of the assigned ports must be physically connected to the same external SAN switch.

for example if you create a single VC SAN, SAN_A for example, and assign all 4 uplink ports to it, then all 4 fibre cables MUST connect to the same external SAN (Switch)/Fabric. You cannot in this case connect ports 1 & 2 to Fabric_A and 3 & 4 to Fabric_B.

If you want to do that you should create 2 VC SANs, SAN_A1 (say) and assign ports 1 & 2 to this SAN, and a second VC SAN, SAN_A2 and assign ports 3 & 4 to it. You can now connect uplinks (ports) 1 & 2 to Fabric_A, and uplinks 3 & 4 to Fabric_B.

If you have a second VC FC module in the adjacent bay, then you can do the same there, by creating VC SANs, SAN_B1 and SAN_B2.

It IS legal to connect (for example) uplinks from both modules to the same external SAN switch, however they must be assigned in VC to different VC SANs.

Remember also, that these new SAN names must be propagated back into your Profiles to ensure that your VM's can continue to talk to their storage.

Just in case we are assuming too much, confirm that your SAN admin is seeing the VC assigned WWID's and not the Physical WWIDs.

Hope this helps.

Dave.