Degraded iSCSI Storage Path Redundancy after update

[CHAC09]

Hey guys,

Just recently deployed 3x r710's running ESXi 4 going back to 2x PS4000 iSCSI storage arrays.

They each have 4x Nics setup on one vSwitch that is vlan off for console and motion. Then I have a seperate PCI dual 1gb nic adapter in each machine going back to it's own vSwitch for iSCSI traffic.

On that vSwitch I am running 4x iSCSI kernels.

All has been running great until I got VUM going and decided to apply some critical updates to the ESX hosts. In particular this update:

Now each of my hosts reports this back via email on a daily basis:

Target: ESX_HOST

Stateless event alarm

Alarm Definition:

( OR OR )

Event details:

Path redundancy to storage device naa.6090a048e0fbd7b7bfab8401000040bf degraded. Path vmhba34:C1:T3:L0 is down. 3 remaining active paths. Affected datastores: "Datastore4".

I have reverted back to a previous version of the Hypervisor on one of my 3 ESX hosts by pressing Shift-R then confirming. Then when I check against critical baselines it shows updates as being required, but the above update is still installed?

Any suggestions welcome. At this stage everything is still ticking over nicely, but I really want to try and get it resolved.

Can post additional logs if required.

Cheers.

[manfriday]

I seem to be encountering the same alarm issue with esxi 4 and my 2 x PS6000E's.

Did you ever find a resolution to this?

[GeorgeMahesh]

Has the server been restarted after the update or the iSCSI sessions logged out and logged back in. The patch that you have attempted to update has got some changes to the vmkernel iSCSI driver. I would suggest to restart the iSCSI or the server itself.

On the other hand, I would like to know if you have configured port binding to get those 4 active paths to the datastores. Could you please provide the esxcfg-vswitch -l, esxcfg-vmknic -l and the port binding information. You could try to re-establish the port binding if it is upset because of the update.

[Kmittal]

-make sure that the path policy is Fixed and no something else.

-to check follow the compatibility guide for the array:

http://www.vmware.com/resources/compatibility/search.php

On storage tab type Eqallogic and look for the model.

[gheywood]

I have the same errors. I am running vSphere u1 (so build 208167) and my Equallogic units are PS5000vx's running 4.3.4 firmware. I am using fixed paths (vmware) though for some volumes, and I am using round-robin (vmware) to connect to the others.

Path redundancy to storage device naa.6090a028b0f1df297fbdb4fa0100305b degraded. Path vmhba33

:C2:T14:L0 is down. 3 remaining active paths. Affected datastores: "VMFS-3".

warning

29/04/2010 07:21:23

I have not applied any patches post U1, but have rebooted the hosts in question.

[ShaneWendel]

This alarm is pretty much useless on EqualLogic arrays.  Whenever the array kicks off a load balance request it'll trigger.