Path degraded on VMware esx 4.0

[srinut31]

Dear Friends,

"[VMware vCenter - Alarm Cannot connect to storage] Path redundancy to storage device naa.6001438005de97ed00005000008a0000 degraded. Path vmhba3:C0:T1:L54 is down. 3 remaining active paths. Affected datastores: "DataStore1""

We have total 4paths aval in that one path is degraded.
I checked "esx-mpath -b" it's showing all four path's are active.
This alarm only from one ESX HOST. We have four ESX Server this all in cluster Env.

[alefestaedist]

Do you have any other warning/error on logs?

how it is configured the path to the lun? fixed path, round robin?

Alex

[srinut31]

Dear alex,

1)Do you have any other warning/error on logs?

There is no other error's, This is  4Node Cluster Setup. 

2) how it is configured the path to the lun? fixed path, round robin?

This Setup configurd as "round robin"

[alefestaedist]

Okay..the waring should be related to a specific host, what kind of storage you are using ? is it FC, SAS..model..I would check if the error is presented on all of the esx host of the cluster or just on one.

Do you have the same warning at the san level, you should see a path degradation even there..

Alex

[srinut31]

Dear Alex,

Thank for your Quick Replay.

1) the warring should be related to a specific host, what kind of storage you are using ?

This warning generated on only one HOST, Your Right.,   Its HP EVA 8400 Storage. This Setup in on FC Modle

2) I would check if the error is presented on all of the esx host of the cluster or just on one

     This  Present Host is there in Cluster, 

3)  Do you have the same warning at the SAN level, you should see a path degradation even there

    We have 8ptahs those all there in Active mode from SAN Side We checked using "cvutil".  We didn't find any error's on SAN Side.

Even we checked "esx-mptah" on all hosts there its showing all active.

[alefestaedist]

If the error is on only one node I would check on that node if you got one of the following:

1) configuration of the path (is it in round robin as the others?)

2) did you try a rescan on the specific host to see if it generate an error?

3) I would try to check the HBA on that specifc host.

Alex

[srinut31]

Dear Alex,

1) configuration of the path (is it in round robin as the others?)

###############################################LUN 33 got Issue###########

[root@node1 ~]# esxcfg-mpath  -b   | grep L33

   vmhba3:C0:T0:L33 LUN:33 state:active fc Adapter: WWNN: 50:01:43:80:04:26:68:cf WWPN: 50:01:43:80:04:26:68:ce  Target: WWNN: 50:00:1f:e1:50:25:47:50 WWPN: 50:00:1f:e1:50:25:47:59

   vmhba0:C0:T1:L33 LUN:33 state:active fc Adapter: WWNN: 50:01:43:80:04:26:63:f5 WWPN: 50:01:43:80:04:26:63:f4  Target: WWNN: 50:00:1f:e1:50:25:47:50 WWPN: 50:00:1f:e1:50:25:47:5c

   vmhba0:C0:T0:L33 LUN:33 state:active fc Adapter: WWNN: 50:01:43:80:04:26:63:f5 WWPN: 50:01:43:80:04:26:63:f4  Target: WWNN: 50:00:1f:e1:50:25:47:50 WWPN: 50:00:1f:e1:50:25:47:58

   vmhba3:C0:T1:L33 LUN:33 state:active fc Adapter: WWNN: 50:01:43:80:04:26:68:cf WWPN: 50:01:43:80:04:26:68:ce  Target: WWNN: 50:00:1f:e1:50:25:47:50 WWPN: 50:00:1f:e1:50:25:47:5d

2)  did you try a rescan on the specific host to see if it generate an error?

When we found this first thing we did Rescan LUN, After Rescaning also we have same issue.

3) I would try to check the HBA on that specifc host.

We didn't find Any HBA Issue's.

[alefestaedist]

Quite weird, from the info you provided I would re-check the zoning on the san.

I woul Double check the WWN in the zone config of the FC switches.

Alex

[srinut31]

Dear Alex,

We have 2  HBA Cards for Every card there is a Redundancy HBA Card is there for  communication.

Can you specify which out out your askin about, I am not familiar with Zoning and all.

[alefestaedist]

Hi there, sorry for the delay. Basically you should have a zone that allow all your host to access to the lun's in your storage, so if in your "zone config"  one or more WWN are missing that particular host will not have access to the lun in a correct way, IMO you should have a correct zone since your HBA is showed up as active but I suggest you to check anyway.

On the other hand could be that you have a phisical problem on the FC channel/cable but in this case you would have errors on the channel from the san monitor.

Alex

[bulletprooffool]

If you previously did not have this problem and you do now, then you need to find out if any changes were made to your storage switches (fabrics)

If all paths are via the same fiber connections then if it were faulty cabling, then in theory all paths shol dbe degraded (depedning of course on how your redundancy is set up)

Confirm that all cabling is sound and all zoning is sound.

[srinut31]

Dear Alex,

Issue has been resolved.

We restarted Vcenter Server. After restarting Vcenter server we didn't receive any warnings from host.

Thank's for your quick Inputs.

[alefestaedist]

Glad you solved.. sound quite odd but could be that the warning was just a "fake" warning.

Alex

[srinut31]

Dear Alex,

We logged a call for this issue,  HP-Vmware Told to restart the Vcenter Server. I think this Know issue with vcenter 4.0.0

May be your correct in this case Alex.Any way Thanks all your support.