RE:iSCSI: Outboard Tunnel Mode

["John Hufferd" <hufferd@us.ibm.com>]

You need to parse the word Installation and Implementation more carefully.
"Use" applies to the Installation locations choices. Implement is what the
vendors must provide.

We do not need to go over this any more!  There is an IETF requirement for
"MUST Implement, and Optional to USE".  Debate is useless.  Lets move on.

.
.
.
John L. Hufferd
Senior Technical Staff Member (STSM)
IBM/SSG San Jose Ca
Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
Home Office (408) 997-6136, Cell: (408) 499-9702
Internet address: hufferd@us.ibm.com


"Lakshmi Ramasubramanian" <nramas@windows.microsoft.com>@ece.cmu.edu on
12/17/2001 12:39:08 PM

Sent by:  owner-ips@ece.cmu.edu


To:   John Hufferd/San Jose/IBM@IBMUS, <VAHUJA@aol.com>
cc:   <ips@ece.cmu.edu>
Subject:  RE:iSCSI: Outboard Tunnel Mode



If it is NOT MUST USE, I guess most implementations would just
choose "None" for security options. What is the point in saying
MUST IMPLEMENT but is NOT MUST USE?

 -lakshmi

-----Original Message-----
From: John Hufferd [mailto:hufferd@us.ibm.com]
Sent: Monday, December 17, 2001 11:13 AM
To: VAHUJA@aol.com
Cc: ips@ece.cmu.edu
Subject: Re: Outboard Tunnel Mode



Installations can do what ever they want.  The Security functions are
must implement, NOT must use.

.
.
.
John L. Hufferd
Senior Technical Staff Member (STSM)
IBM/SSG San Jose Ca
Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688 Home
Office (408) 997-6136, Cell: (408) 499-9702 Internet address:
hufferd@us.ibm.com


VAHUJA@aol.com@ece.cmu.edu on 12/17/2001 10:09:10 AM

Sent by:  owner-ips@ece.cmu.edu


To:   ips@ece.cmu.edu
cc:
Subject:  Outboard Tunnel Mode



Folks,

May be I missed something in SLC meeting.  I can expect several
implementations of iSCSI not include any security.Reason - I can see
that customers would often rely on the company's existing VPNs (outboard
Router
etc) to protect their data (storage or otherwise) over IP networks. From
a CIO's viewpoint, this approach may make more sense than extending yet
another layer of IPSec into its servers just for storage data.

 It is not clear to me from the standard if it will be a non-compliance
of iSCSI standard. If so, we may potentially have many non-compliances.