RE: Unacceptable ACK in SYN_RCVD state handling

[Robin Uyeshiro <[e-mail]>]

I'm confused.  The stuff on page 68 (69?) seems to contradict the following
from page 36, under the Reset Generation heading, at least for SYN-RECEIVED:

    2.  If the connection is in any non-synchronized state (LISTEN,
    SYN-SENT, SYN-RECEIVED), and the incoming segment acknowledges
    something not yet sent (the segment carries an unacceptable ACK), or
    if an incoming segment has a security level or compartment which
    does not exactly match the level and compartment requested for the
    connection, a reset is sent.

Page 68/69 appears to count SYN-RECEIVED as a synchronized state.

-----Original Message-----
From: Bob Braden [mailto:[e-mail]]
Sent: Monday, April 08, 2002 12:14 PM
To: [e-mail]; [e-mail]
Cc: [e-mail]
Subject: Re: Unacceptable ACK in SYN_RCVD state handling


  *> 
  *> No, the scenario that applies is described on page 68 (sequence number
check),

Yes, sorry, I misinterpreted what you were asking (ACK field as opposed
to sequence field of ACK segment (But the man DID say "receive window"...
mumble).

  *> 
  *> The point u mentioned above is the conflict i am having as well, it
seems to me
  *> that retransmitting
  *> a SYN-ACK seems the right thing to do. But the BSD stack doesn't seem
to be
  *> doing that.

I thought you said earlier that the BSD DOES retransmit the SYN.  In
any case, I would agree that the text of RFC 793 seems to say
otherwise.  This particular test is one that has caused a great deal of
confusion over the years.  From the underlying logic of TCP,  I believe
that retransmitting the SYN is OK.  So it seems you can choose either
interpretation you want (unless there is some  subtle security
implication.)

Bob Braden

  *> 
  *> Murali
  *> 
  *> 
  *> 
  *> 
  *> >
  *> >
  *> > Bob Braden
  *>