Re: RFC1122 urgent data conformance question

kurt@unirsvl.rsvl.unisys.com Wed, 30 September 1992 15:20 UTC

Received: from ietf.nri.reston.va.us by IETF.NRI.Reston.VA.US id aa05940; 30 Sep 92 11:20 EDT
Received: from NRI.RESTON.VA.US by IETF.NRI.Reston.VA.US id aa05936; 30 Sep 92 11:20 EDT
Received: from NNSC.NSF.NET by NRI.Reston.VA.US id aa10771; 30 Sep 92 11:24 EDT
Received: from nnsc.nsf.net by NNSC.NSF.NET id aa11588; 30 Sep 92 11:12 EDT
Received: from relay1.UU.NET by NNSC.NSF.NET id aa11583; 30 Sep 92 11:11 EDT
Received: from uunet.uu.net (via LOCALHOST.UU.NET) by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA15839; Wed, 30 Sep 92 11:11:24 -0400
Received: from s5000.UUCP by uunet.uu.net with UUCP/RMAIL (queueing-rmail) id 111006.28356; Wed, 30 Sep 1992 11:10:06 EDT
Received: by unirsvl.RSVL.UNISYS.COM (smail2.5) id AA20081; 30 Sep 92 09:23:07 CDT (Wed)
Subject: Re: RFC1122 urgent data conformance question
To: Bob Braden <uunet!ISI.EDU!braden@uunet.uu.net>
Date: Wed, 30 Sep 1992 09:22:59 -0500
Cc: 4jde@uunet.uu.net, ietf-hosts@nnsc.nsf.net, kurt@uunet.uu.net, ietf-hosts@nncs.nsf
In-Reply-To: <199209291744.AA26042@zephyr.isi.edu>; from "Bob Braden" at Sep 29, 92 10:44 am
X-Mailer: ELM [version 2.2 PL0]
Message-Id: <9209300923.AA20067@unirsvl.RSVL.UNISYS.COM>
Sender: ietf-archive-request@IETF.NRI.Reston.VA.US
From: kurt@unirsvl.rsvl.unisys.com

> 
> 
> > 
> > Question: If the TCP receives an urgent input and queues the urgent input with
> > an indication that it is urgent on the application's input queue, does this, in
> > your opinion, conform to the above excerpt from RFC1122?
> > 
> 
> In my opinion, no.  The Urgent interface to the application layer is
> intended to solve the problem of the application NOT reading its input,
> eg because it is looping or otherwise mis-behaving.  Urgent is intended
> to activate an asynchronous thread (in practice, in the Telnet under-
> belly of the application layer) to unstick things.

From my understanding I thought that the asynchronous notification was needed
because in the standard UNIX environment there is only one activity/process per
connection. This creates the problem where the activity/process can be 
looping, or just doing output without checking to see if there is input to
process. Since in the case that I described there is an activity dedicated to
processing input, this problem doesn't exist. Therefore I would think that
in my case, the asynchronous notification would be not applicable instead of
a non-conformance. How does that sound?


> 
> From your description of the host operating system environment, I gather
> that this is not a general-purpose host, which is the intended domain of
> RFC-1122. 

No, this is not a special purpose host, it is the standard Unisys 2200      
mainframe. We have been doing things this way for over 10 years. The 2200 
mainframe has been a multiprocessor machine for approx 20 years.
It is possible for a general purpose host to be something other than a single
processor UNIX box. :-)

>                 (But what happens if the application DOES go into a
> loop and stop reading its input?)

I would assume that it would be the same as if the asynchronous thread did the
same thing. You always have this possibility.

Since you cc'd the other WG, I pulled it out of the header and am sending this
there also. Thanks for your reply.

Kurt Matthys
Unisys Corp
kurt@rsvl.unisys.com