RE: iSCSI: Creation of new LUNs

[Charles Monia <cmonia@NishanSystems.com>]

Hi:

A note on AEN support:

Although all SCSI transports provide a way to deliver an AEN, few
implementations support it in practice.  Since a device can implement both
initiator and target functions (as it would for 3rd party copy requests),
the usual approach is for the initiator to also present itself as a target
that supports the standard SCSI model for "host" devices. The data
associated with an asynchronous event is then transferred to the initiator
(temporarily acting as a target) by means of the SCSI SEND command. I guess
one could characterize this as a temporary role reversal.

Of course, a protocol can define some other mechanism that serves the same
end.

Charles


> -----Original Message-----
> From: Black_David@emc.com [mailto:Black_David@emc.com]
> Sent: Thursday, September 14, 2000 7:05 PM
> To: julian_satran@il.ibm.com; ips@ece.cmu.edu
> Subject: RE: iSCSI: Creation of new LUNs
> 
> 
> This sort of proposal should go to T10.  A potential
> issue is that not all SCSI transports support
> Asynchronous Event Notification (AEN).  iSCSI could
> make AEN mandatory to implement as part of addressing
> this situation, but standardizing the use of AEN to
> report LUN addition needs to be addressed in T10.
> An I-D summarizing the problem and proposed solution
> without reference to iSCSI or any other transport
> might be a good start on this.
> 
> --David
> 

<other text deleted>