[Fwd: Re: iSCSI: X-bit in Login]

["Mallikarjun C." <cbm@rose.hp.com>]

Copying to the list....


-------- Original Message --------
Subject: Re: iSCSI: X-bit in Login
   Date: Thu, 14 Feb 2002 09:31:33 +0200
   From: "Julian Satran" <Julian_Satran@il.ibm.com>
     To: "Mallikarjun C." <cbm@rose.hp.com>



Mallikarjun,

I agree with this.

Julo


  "Mallikarjun C." <cbm@rose.hp.com>
                                             To:        "ips"
  Sent by: owner-ips@ece.cmu.edu     <ips@ece.cmu.edu>
                                             cc:
  14-02-02 01:58                             Subject:        iSCSI:
                                     X-bit in Login



All,

It seems to me that we do not need the X-bit even in the Login PDU.

"Connection reinstatement"/"implicit connection logout" can be
implicitly
signaled by using an active ISID-TSID-CID combination to request the
target to implicitly logout the connection state machine for the CID.
This is
conceptually nothing but the "option A" choice we made for implicit
session logout sometime ago (remember the "reusing ISID for recovery"
thread?), albeit at the connection level.  As far as I can tell, the
current
X-bit in Login PDU is similar to the C-bit that we considered and
rejected
for "implicit" session logout.  I suggest we apply the same design
principle
here.

I propose that the draft should continue to formally define the notion
of
"connection reinstatement" with the existing rules on its usage, but
retire
the X-bit to make that bit position reserved.  Leaving the X-bit in
would only
lead to needless error cases.

Comments?
--
Mallikarjun

Mallikarjun Chadalapaka
Networked Storage Architecture
Network Storage Solutions Organization
Hewlett-Packard MS 5668
Roseville CA 95747