Hi Vipin,

When we register a host,it means that we assing the wwpn(port name) of HBA as initiator to the storage?

tnx a lot,

marius

By definition, registration is the process of making the host known to the storage. By registering a host, you will be assigning the hostname and IP address to the host connections/HBAs.

Once the host is connected via multiple paths with storage (same HBA port via different storage ports/ multiple HBA ports etc...), it will display multiple entries in the connectivity status (even though it is from same HBA/host ). You have to register the paths,  ( HBAs- as you mentioned in the comment above). This is the process of host registration.

Hi Vipin,

Please clarify with me things.I seen somewhere that also to register host,we could register HBA of that host in Clariion.Right?So wwpn of each port within HBA?

Is this other way to register a host which is different of registering ip and hostname(of a certain host) in storage?

Hi Vipin,

In order to register a host(wwpn of HBA) with a storage,there is necessary to register HBAs, IP and host name?

tnx a lot,

marius

While registering paths (HBA / host) you have to give the Host name, IP address, failover mode, initiator type etc...

Hope How to manually register a host HBA with a CLARiiON array using Navisphere Manager and CLI 6.4?will help you in understanding the same.

There are a number of documents available on the support.emc.com that will explain the basic requirements for connecting a host to an array. One way to register a host is by using the EMC Server Initialization Tool (EMC Server Utility), or by installing the EMC Host Agent. These two programs are shipped with each array and you install them on your host. They will then aid in registering a host that is connected to an array.

If the array is a Unified VNX, you can go to mydocs.emc.com and create a customer installation guide based on the host and the array (VNX) that you have.

glen

Hi Vipin,

Tnx a lot for everything.

Only a thing now.

If a  certain port of one of the SAN controller fails,than the LUN who is seen through this port to a host,could be mapped to other port in controller,so it will be available to host after zoning.right?

give me an example of how an faulty port or an whole controller faulty is seen in storage log.

tnx a lot,

marius

LUN is available through every port on storage processor A and B, it's just a matter of what port your host is zoned to.  That's why for redundancy reasons you zone HBA to multiple ports on storage processor so that port failure does not cause an outage (you also must run multipathing software PowerPath or native mpio)

I've seen instances where port on Clariion/VNX would go bad but it did not give any indications of that in Navisphere/Unisphere. You do see that all of sudden your hosts lost a path and that specific port is no longer logged-in to the switch.