VMware Cloud Community
virtualdude
Contributor
Contributor

Path status showing dead after update to hosts and reboot of vcenter

Hello,

I updated my 2 hosts to the latest build today and also rebooted the vcenter server.

now when i goto the manage paths of the LUNS its showing "dead" under san identifier

when i run esxcfg config to list the paths i do see it as on and fixed/preferred

i am using 2x qlogic iscsi hbas on each host and its only listing DEAD on vmhba for both of the hosts

vmhba2 on both hosts shows fine.

using 3x lefthand san modules with saniq 8.

i also did a rescan and it still shows dead

any ideas?

Thanks in advance

0 Kudos
11 Replies
virtualdude
Contributor
Contributor

attached is a screenshot of one of my luns

0 Kudos
Zwood
Contributor
Contributor

I am seeing the same issue. I have the same hardware - lefthand with qLogic HBA's

Any ideas?

0 Kudos
virtualdude
Contributor
Contributor

It appears to be a "display issue" both paths are working but it just

says dead. Hope its fixed in vsphere 4. I forgot the command from the

console that can verify that both links are up.

Are you running jumbo frames and flow control?

What switches are you using?

Did you set the arp redirection on the qlogic hbas?

0 Kudos
Zwood
Contributor
Contributor

I new it was just a diplay issue, but I was hoping VMware had a fix. The command line is esxcfg -mpath -1, I think.

I am not using jumbo frames or flow control.

Cisco switches. 3120s on the Blade side and Cisco 4900Ms on the SAN switches

I do have arp redirection enabled on the qLogic HBA's.

0 Kudos
virtualdude
Contributor
Contributor

It could also be the arp redirection that causes the problem. I made the

change to the hba and then rebooted. I Didn't notice it until a day

later.

I haven't tried changing it back without arp redirect to see if it makes

a difference.

How's the performance of the SAN? Which NSM are you using?

Danny

0 Kudos
Zwood
Contributor
Contributor

For me, the DEAD setting didn't show up until I installed ESX 3.5 update 2. With Update 1 the managed paths showed the correct settings. The problem is still there with Update 4.

Performance is great. We use NSM 2120's with SAS drives. Very happy with Lefthand, we have been a customer for about 4-5 years now.

You?

0 Kudos
virtualdude
Contributor
Contributor

We been using the (3) 2120's also for about a year now with Vmware.

Still a newbie with the san and virtualization. The San has been working

fine. I enabled flow control and jumbo frames and notice a boost in

performance.

Were about 33% virtualized.

0 Kudos
virtualdude
Contributor
Contributor

Go get your vsphere 4 now!

0 Kudos
Zwood
Contributor
Contributor

I will probably wait for the 4.x version before rolling this out. There's a lot of new functionality VMware built into this release. Pretty impressive.

0 Kudos
Systums_Admin
Contributor
Contributor

Same problem here...My setup 3 NSM 160's, SANIQ 8, ESX 3.5 Update 4., QLogic HBA 4060c .....I thought I was the only one having this issue. I have an open support request with VM on this.......still waiting, will keep you guys posted...

0 Kudos
Iridium130m
Contributor
Contributor

Did you get a response back from VMware on this? I'm working on a similiar issue and am trying to get SRM going, but am not detecting VMFS datastores on my replicated volumes. I have to wonder if this display issue is extending to SRM's ability to detect datastores properly.

0 Kudos