Re: iSCSI: Login stage transition; T bit

["Julian Satran" <Julian_Satran@il.ibm.com>]

Comments in text.

Basically we had a a login with very little redundance earlier and it wa deemed cryptic and error prne.
Currently every step is clearly stated and although some infor is redundant it clearly states the intentions of the issuying party.
I suggest we stop improving on this.

Julo

Luben Tuikov <luben@splentec.com> Sent by: luben@ns.splentec.com 26-03-02 22:31 Please respond to Luben Tuikov; Please respond to iSCSI

To:        iSCSI <ips@ece.cmu.edu>, Julian Satran/Haifa/IBM@IBMIL, "Mallikarjun C." <cbm@rose.hp.com>         cc:                 Subject:        iSCSI: Login stage transition; T bit

Hello there,

Q1. It appears to be the case that 
   o  CSG =/= NSG iff T=1, and 
   o  CSG < NSG, from the table on pg. 64, v11.

Thus, T=1 iff CSG < NSG, so the T bit can be retired,
by imposing that there be no garbage in NSG (that is
NSG >= CSG for all login PDU's).


+++ yes it is redundant but intentionally so NSG hast to be 0 if T is not 1 +++
Q2. After a request/response with T=1 and NSG > CSG
(appropriately), what is the RFC verb for
the initiator. That is, MUST the initiator switch
to the next stage, or SHOULD or MAY?
+++ it must switch as the initiator is the one that requested it.
Q3. Same as Q2 but NSG=3?
+++ same answer +++
Thanks,
-- 
Luben