RE: question on reserve and release !

[Eddy Quicksall <eddy_quicksall@ivivity.com>]

From the spec:

"The I_T nexus can be identified by the conjunction of the SCSI port names;
that is, the I_T nexus identifier is the tuple (iSCSI Initiator Name +
',i,'+ ISID, iSCSI Target Name + ',t,'+ Portal Group Tag)."

The "scsi id" you mention would have to be the InitiatorName+ISID.

This would be no different than parallel SCSI when the initiator has two
HBA's on the same bus. So whatever problems you mention already exist.

Eddy

-----Original Message-----
From: Ramesh Ramakrishnan [mailto:ramesh@global.com]
Sent: Friday, October 18, 2002 5:14 PM
To: ips@ece.cmu.edu
Subject: question on reserve and release !


> Hi All,
> 
> I have a question on how iscsi will handle the scsi reserve and release
commands.
> 
> When multiple iscsi initiators make calls to the same iscsi target, the
actual scsi
> device only sees one single scsi id (that of the scsi HBA on the iscsi
target
> machine).  This would mean that different iscsi initiators sending scsi
reserve
> command will be interpreted by the scsi target as commands from the same
> scsi initiator.  This defeats the purpose of implementing the scsi
reserve/release
> commands.  And in a way, makes iscsi not very useful in environments where
> iscsi initiator requires exclusive access to a iscsi target. (Imagine a
iscsi initiator
> doing a back up to a iscsi tape device when another iscsi initiator starts
rewinding
> the same iscsi tape device.)
> 
>  Is this something that should be implemented in the iscsi target
> driver?  I found that Cisco's iscsi driver (1.8.1) for AIX has a  feature
called 
> "reserve/release proxy".  Is this cisco's implementation for getting
around this
> problem?  Has anyone any experience with that driver? Another interesting
> thing:  I only find reference to this reserve/release proxy mentioned in
the AIX driver.
> 
> thanks
> ramesh