lun not visible-urgent

[krsukuma]

The following servers are cluster FRDKOCL04N2 and FRDKOCL04N1. The issue is in FRDKOCL04N2 the luns are not visible.

Zone info:

Fabric A:

(FRDKOCL04N2_FC1)+(LT2G11_EVA_10_cA_Fc1)+(LT2G11_EVA_10_cB_Fc2)
1000-0000-c93b-870e + 50001FE1 50030529 + 50001FE1 5003052C


(FRDKOCL04N2_FC1)+(LT101_EVA_10_CA_FC1)+(LT101_EVA_10_CB_FC2)

1000-0000-c93b-870e + 50001FE1 50027B19 + 50001FE1 50027B1C



Fabric B:

(FRDKOCL04N2_FC2)+(LT2G11_EVA_10_CA_FC2)+ (LT2G11_EVA_10_CB_FC1)

1000-0000-c93a-7ca1 + 50001FE1 50030528 + 50001FE1 5003052D


(FRDKOCL04N2_FC2)+(LT101_EVA_10_CA_FC2) + (LT101_EVA_10_CB_FC1)

1000-0000-c93a-7ca1 + 50001FE1 50027B18 + 50001FE1 50027B1D

The above two servers are connected to two storage box and these storage box are replicating between each other.
The FRDKOCL04N2 and FRDKOCL04N1 are mapped the same luns.

I checked in broacde switch, zoning is fine, ports are online, health check is good.

I am not sure why in server end the disk is not visible, experts pls suggest whetehr I need to check any thing more.

The server is connected to HP eva 5000

[krsukuma]

Pls find the attachment

[Víctor Cespón · Accepted solution]

EVA connections are correct
Zoning is correct
That server has access to the 4 ports of both EVAs

This LUN is presented to the server?
If it's part of a DR_Group, is in the source or on the destination EVA?

Remember that the destination vdisk is usually not presented to the host, until there's a failover and the roles are reversed. Then the LUN is unpresented from server 1 and presented to server 2.

[krsukuma]

There are 10 disk presented to this cluster N1 and N2, as cluster both are same luns and same Lun id given.There are active-active cluster.
As I checked in server FRDKOCL04N1- from 10 Luns 7 luns are visible and 3 or not.
In FRDKOCL04N2(the reverse) 3 luns are visible and 7 or not.

I think this is due to the active-active configuration in the server.

Please suggest.

[TTr]

Check the Vdisk presentation within the EVA. The Vdisks don't seem to be assigned to both members of the cluster.

[Víctor Cespón]

Is not this the definition of a cluster? Only one member of the cluster can access a resource at the same time.

Forget the OS and the software and check in Command View if the disks are presented to both hosts.

[krsukuma]

From storage End all the disks are presented to the server.

[sirfire]

Check the basics of course....

Can both hosts see other LUNs from the EVA?

Link lights on the eva and hosts?

zoning is enabled?

If disabled make sure default zone is set to "allaccess"

Has the clusters failed?

What version of cluster software?

Does DSM software support clustering?

If so did you enable the cluster option when you loaded it?

In Command view EVA you can see the World Wide Lun Name.

You should be able to see these WWLN's inside your in the nodes.

If you have DSM drivers installed correctly with DSM manager you can also view the WWLN's under "My Computer" "Manage" then "storage" "DSM Manager"

If what you say is correct both hosts with all WWPN's are presented to the hosts then it is most likely a cluster issue or as others have already mentioned one cluster is actively accessing the lun while the other is not?

Outside of that without more information on when it was last available and what actions/events had occurred prior to LUNs going offline i cannot help you.

[Sivakumar MJ._1]

Hi,

Check the VDISK is presented to the both the host, Host mode is same as the OS.

[Sivakumar MJ._1]

Check the wwn is correctly presented in the EVA, Zone is activated in the Switch side...