From owner-tcp-impl@lerc.nasa.gov Tue Dec 1 01:09:55 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id BAA10497 for ; Tue, 1 Dec 1998 01:09:54 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id XAA02314; Mon, 30 Nov 1998 23:20:03 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from smtprich (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id XAA02310; Mon, 30 Nov 1998 23:20:01 -0500 (EST) Received: from zmers013 by smtprich; Mon, 30 Nov 1998 22:19:55 -0600 Received: from zcard00m.ca.nortel.com by zmers013; Mon, 30 Nov 1998 23:19:45 -0500 Received: by zcard00m.ca.nortel.com with Internet Mail Service (5.0.1460.8) id ; Mon, 30 Nov 1998 23:19:43 -0500 Message-ID: <13E2EF604DE5D111B2E50000F80824E897FBE0@zwdld001.ca.nortel.com> From: "Dabin Wang" To: "'tcp-impl@lerc.nasa.gov'" Subject: send Window and pipe capacity Date: Mon, 30 Nov 1998 23:19:41 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.0.1460.8) Content-Type: text/plain Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Is there any relationship between Send window size and pipe capacity? Thank you Dabin From owner-tcp-impl@lerc.nasa.gov Tue Dec 1 06:17:30 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id GAA13167 for ; Tue, 1 Dec 1998 06:17:30 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id EAA26328; Tue, 1 Dec 1998 04:34:11 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from alpha.xerox.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with SMTP (NASA LeRC 8.7.4.1/2.01-main) id EAA26324; Tue, 1 Dec 1998 04:34:09 -0500 (EST) Received: from parc.xerox.com ([13.0.209.39]) by alpha.xerox.com with SMTP id <54034(2)>; Tue, 1 Dec 1998 01:34:02 PST Message-ID: <3663B82F.BD18B90F@parc.xerox.com> Date: Tue, 1 Dec 1998 01:34:39 PST From: Mike Spreitzer X-Mailer: Mozilla 4.06 [en] (WinNT; U) MIME-Version: 1.0 To: tcp-impl@lerc.nasa.gov, end2end-interest@isi.edu Subject: FYI: HTTP-NG BOF at IETF-43 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk At IETF-43 we will try again to form a WG for HTTP-NG; last time we lacked proposed charter and chair, and now we have these. HTTP-NG has some relation to the work you are doing. In particular, the proposed work includes looking at issues in how HTTP (1.1 and NG) use the transport layer, and what (if anything) can be done to improve the situation. If you are interested, we would welcome your participation. You can find a home page for the proposed WG at . The BOF is currently scheduled for 13:00--15:00 on Dec 7. Mike From owner-tcp-impl@lerc.nasa.gov Tue Dec 1 10:53:23 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id KAA17783 for ; Tue, 1 Dec 1998 10:53:23 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id JAA14759; Tue, 1 Dec 1998 09:04:18 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from guns.lerc.nasa.gov (guns.lerc.nasa.gov [139.88.44.160]) by assateague.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id JAA14749; Tue, 1 Dec 1998 09:04:16 -0500 (EST) Received: from guns.lerc.nasa.gov by guns.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-local) id JAA15519; Tue, 1 Dec 1998 09:04:15 -0500 (EST) Message-Id: <199812011404.JAA15519@guns.lerc.nasa.gov> To: tcp-impl@lerc.nasa.gov Cc: "Vern Paxson" , "Sally Floyd" , tomh@cs.berkeley.edu From: Mark Allman Reply-To: mallman@lerc.nasa.gov Subject: WG Last Call for NewReno Draft Organization: Late Night Hackers, NASA Lewis, Cleveland, Ohio Song-of-the-Day: Who You Are Date: Tue, 01 Dec 1998 09:04:15 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Vern and I feel that the NewReno I-D is ready for publication as an experimental RFC. So, we're holding a 3 week last call (to end on 12/22/98) for any final comments. The draft is available at: ftp://ftp.ietf.org/internet-drafts/draft-ietf-tcpimpl-newreno-00.txt allman From owner-tcp-impl@lerc.nasa.gov Wed Dec 2 14:56:03 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA27912 for ; Wed, 2 Dec 1998 14:56:02 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id MAA20271; Wed, 2 Dec 1998 12:59:34 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from guns.lerc.nasa.gov (guns.lerc.nasa.gov [139.88.44.160]) by assateague.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id MAA20253; Wed, 2 Dec 1998 12:59:31 -0500 (EST) Received: from guns.lerc.nasa.gov by guns.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-local) id MAA06313; Wed, 2 Dec 1998 12:59:30 -0500 (EST) Message-Id: <199812021759.MAA06313@guns.lerc.nasa.gov> To: tcp-impl@lerc.nasa.gov Cc: "Vern Paxson" From: Mark Allman Reply-To: mallman@lerc.nasa.gov Subject: notes for upcoming tcp-impl meeting Organization: Late Night Hackers, NASA Lewis, Cleveland, Ohio Song-of-the-Day: TV Dinner Date: Wed, 02 Dec 1998 12:59:30 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Vern and I are looking for someone to take the official notes at the tcpimpl meeting next week. We would be especially interested in someone who is somewhat knowledgable about PMTU discovery, since we may end up discussing that in some detail. If you can help out, please drop Vern and I a note. Thanks, allman From owner-tcp-impl@lerc.nasa.gov Fri Dec 4 04:56:39 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA21918 for ; Fri, 4 Dec 1998 04:56:39 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id CAA26095; Fri, 4 Dec 1998 02:44:00 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from daffy.ee.lbl.gov (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id CAA26091; Fri, 4 Dec 1998 02:43:58 -0500 (EST) Received: (from vern@localhost) by daffy.ee.lbl.gov (8.9.1/8.9.1) id XAA01294; Thu, 3 Dec 1998 23:43:57 -0800 (PST) Message-Id: <199812040743.XAA01294@daffy.ee.lbl.gov> To: tcp-impl@lerc.nasa.gov Subject: put on your PMTU discovery thinking caps Cc: mallman@lerc.nasa.gov Date: Thu, 03 Dec 1998 23:43:57 PST From: Vern Paxson Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk We've heard a number of people mention that it would be good to document some of the pitfalls & problems of implementing PMTU discovery (something with a bit more meat than RFC 1435). For example, one is that if an ATM switch doesn't have enough buffer to reassemble a bunch of cells into a really large packet, then it will necessarily discard the packet, and if the packet had DF set, it *won't* generate the necessary ICMP response (because it couldn't process the full packet at all), leading to a PMTU black hole. (Matt Mathis pointed this one out a while back.) Even though some of these problems aren't related to TCP implementation, some are, and it's not clear there's a better home for discussing them than tcpimpl. So we've put on the Orlando agenda a discussion of PMTU discovery problems, to try to get a feel for the scope. Neither of your esteemed WG chairs, though, is particularly familiar with these problems. We'd therefore like to encourage discussion on the list over the next few days, so we can get things rolling in advance and figure out some of the common ground in order to cover it quickly at the meeting. We'd also like to encourage tcpimpl'ers to come to the meeting prepared to discuss their favorite PMTU problems. See you next week, Vern From owner-tcp-impl@lerc.nasa.gov Fri Dec 4 11:13:20 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25555 for ; Fri, 4 Dec 1998 11:13:19 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id JAA05523; Fri, 4 Dec 1998 09:15:29 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from Twig.Rodents.Montreal.QC.CA (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id JAA05515; Fri, 4 Dec 1998 09:15:25 -0500 (EST) Received: (from mouse@localhost) by Twig.Rodents.Montreal.QC.CA (8.8.5/8.8.5) id JAA28478; Fri, 4 Dec 1998 09:15:16 -0500 (EST) Date: Fri, 4 Dec 1998 09:15:16 -0500 (EST) From: der Mouse Message-Id: <199812041415.JAA28478@Twig.Rodents.Montreal.QC.CA> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit To: tcp-impl@lerc.nasa.gov Subject: Re: put on your PMTU discovery thinking caps Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Content-Transfer-Encoding: 8bit > [...] pitfalls & problems of implementing PMTU discovery [...] > Even though some of these problems aren't related to TCP > implementation, some are, and it's not clear there's a better home > for discussing them than tcpimpl. ...especially since discussing how to make a PMTU implementation deal with them *is* squarely in tcpimpl's charter (at least as I understand it). The difference between a black hole due to an overloaded ATM switch and a black hole due to firewalled ICMP may be interesting in some respects, but the end systems really don't care about details of *why* their packets aren't getting through; all they care is *that* they aren't, and they're not hearing back about it. Besides, how to do IP-over-ATM in a way that doesn't suffer from such problems is arguably on-charter anyway. :-) If enough ATM cells make it through to reconstruct the packet's headers, you can send back a meaningful ICMP without needing to get all of the payload. I'm not sure which ICMP you should send, though. Need-frag doesn't really apply. Source quench, maybe, since it amounts to "I'm overloaded"? der Mouse mouse@rodents.montreal.qc.ca 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B From owner-tcp-impl@lerc.nasa.gov Fri Dec 4 20:14:24 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA05527 for ; Fri, 4 Dec 1998 20:14:23 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id SAA18562; Fri, 4 Dec 1998 18:43:13 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from gecko.nas.nasa.gov (fw04.lerc.nasa.gov [139.88.145.19]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id SAA18558; Fri, 4 Dec 1998 18:43:11 -0500 (EST) Received: from gecko.nas.nasa.gov (kml@localhost) by gecko.nas.nasa.gov (8.9.1a/NAS8.8.7n) with ESMTP id NAA20029; Fri, 4 Dec 1998 13:10:22 -0800 (PST) Message-Id: <199812042110.NAA20029@gecko.nas.nasa.gov> To: Vern Paxson cc: tcp-impl@lerc.nasa.gov, mallman@lerc.nasa.gov Subject: Re: put on your PMTU discovery thinking caps In-reply-to: Your message of "Thu, 03 Dec 1998 23:43:57 PST." <199812040743.XAA01294@daffy.ee.lbl.gov> Date: Fri, 04 Dec 1998 13:10:21 -0800 From: "Kevin M. Lahey" Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk In message <199812040743.XAA01294@daffy.ee.lbl.gov>Vern Paxson writes >We've heard a number of people mention that it would be good to document >some of the pitfalls & problems of implementing PMTU discovery (something >with a bit more meat than RFC 1435). >So we've put on the Orlando agenda a discussion of PMTU discovery problems, >to try to get a feel for the scope. Neither of your esteemed WG chairs, >though, is particularly familiar with these problems. We'd therefore like >to encourage discussion on the list over the next few days, so we can get >things rolling in advance and figure out some of the common ground in order >to cover it quickly at the meeting. We'd also like to encourage tcpimpl'ers >to come to the meeting prepared to discuss their favorite PMTU problems. I listed two small problems and one large (black hole discovery) problem a few weeks ago. Other than that, I guess I worry about the oft-discussed issue of when to ACK -- if MSS really doesn't mean anything anymore, do we ACK every two, or try to determine on-the-fly the segment size in use by the sender. An additional problem is MSS determination for multi-homed hosts. If we are responding to an SYN, we can use the MSS for the interface on which we received the SYN. If we are originating a SYN, we can use the MSS of the interface out which we expect the packet to be sent. What about routing changes? If we belatedly find out that a connection that can use our HIPPI interface (65280 byte MTU) instead of our Ethernet interface, the MSS would prevent us from taking advantage of the new PMTU. We could use the MSS of the interface with the largest MTU, but what happens we we plug in our HIPPI PCMCIA card :-) half-way through a very long-term connection? Kevin kml@nas.nasa.gov From owner-tcp-impl@lerc.nasa.gov Sat Dec 5 05:42:09 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA18275 for ; Sat, 5 Dec 1998 05:42:09 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id EAA06784; Sat, 5 Dec 1998 04:20:29 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from orm-mail20.provo.novell.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with SMTP (NASA LeRC 8.7.4.1/2.01-main) id EAA06779; Sat, 5 Dec 1998 04:20:27 -0500 (EST) Received: from INET-ORM-Message_Server by orm-mail20.provo.novell.com with Novell_GroupWise; Sat, 05 Dec 1998 02:20:19 -0700 Message-Id: X-Mailer: Novell GroupWise 5.5 Date: Sat, 05 Dec 1998 02:19:44 -0700 From: "Nagampali Narasimharao" To: Subject: Multi Threaded Implementations of TCPIP Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by ietf.org id FAA18275 Hi All, I am a TCP/IP developer and am looking for any kind of information - right from specific issues involved to code - (Multi Threading) MT Enabled TCP/IP stack Impementations. Thanking You in Anticipation Narasimha Rao From owner-tcp-impl@lerc.nasa.gov Tue Dec 8 01:37:54 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA02345 for ; Tue, 8 Dec 1998 01:37:54 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id XAA06376; Mon, 7 Dec 1998 23:48:50 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from mail5.microsoft.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id XAA06372; Mon, 7 Dec 1998 23:48:48 -0500 (EST) Received: by INET-IMC-05 with Internet Mail Service (5.5.2232.9) id ; Mon, 7 Dec 1998 20:48:47 -0800 Message-ID: <416245726D64D211808100805F3198EE0132F732@RED-MSG-46> From: Peter Ford To: "'Vern Paxson'" , mallman@lerc.nasa.gov, tcp-impl@lerc.nasa.gov Cc: sob@harvard.edu Subject: RE: draft agenda for Orlando tcpimpl meeting Date: Mon, 7 Dec 1998 20:48:39 -0800 X-Mailer: Internet Mail Service (5.5.2232.9) Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Vern and Mark, I would like to take the opportunity of the working group meeting in Orlando to discuss the draft for 2001.bis and the value of the initial congestion window. I would like to discuss is modifying the current draft so that ICWND with values of 1 or 2 is allowed. From the discussion on the tcp-impl list I believe we had consensus in the working group that values for ICWND of 1 or 2 would be okay. Values greater than 2 (e.g. 4) are considered by some to be problematic so we decided to put larger values into RFC 2414 as experimental. I would appreciate a short time slot in Orlando to discuss this subject with the working group. thanks, peter > -----Original Message----- > From: Vern Paxson [SMTP:vern@ee.lbl.gov] > Sent: Saturday, November 28, 1998 12:30 AM > To: tcp-impl@lerc.nasa.gov > Cc: mallman@lerc.nasa.gov; sob@harvard.edu > Subject: draft agenda for Orlando tcpimpl meeting > > Here's the draft agenda Mark & I have put together. Comments/additions > welcome, please send them as soon as possible. (A revised agenda will > be sent to the IETF agenda meister on Monday.) > > 1. Agenda bashing (5 minutes) > 2. Status of WG drafts: > - Known Problems, TCP Congestion Control > - NewReno I-D > 3. Report on Usefulness of Documents -- Jamshid Mahdavi > 4. PMTU Issues > > This last is meant to be an effort to sketch out different issues with > implementing PMTU discovery, as we're wondering if there are enough of > these (and sufficient WG energy) to merit writing a document for them. > > Also, note that this is likely to be the last tcpimpl WG meeting, with > the group going on hiatus (though the mailing list still active) once we > get the pending documents out. > > Vern From owner-tcp-impl@cthulhu.engr.sgi.com Wed Dec 9 05:24:36 1998 Received: from sgi.sgi.com (SGI.COM [192.48.153.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA24014 for ; Wed, 9 Dec 1998 05:24:35 -0500 (EST) Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id CAA03796; Wed, 9 Dec 1998 02:19:47 -0800 (PST) mail_from (owner-tcp-impl@cthulhu.engr.sgi.com) Received: (from majordomo-owner@localhost) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id CAA09444 for tcp-impl-list; Wed, 9 Dec 1998 02:09:05 -0800 (PST) mail_from (owner-tcp-impl@relay.engr.sgi.com) Received: from sgi.sgi.com (sgi.engr.sgi.com [192.26.80.37]) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id CAA48121 for ; Wed, 9 Dec 1998 02:09:04 -0800 (PST) mail_from (mahesh@lotus.iitm.ernet.in) Received: from lotus.iitm.ernet.in ([206.103.12.236]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id CAA03366 for ; Wed, 9 Dec 1998 02:08:52 -0800 (PST) mail_from (mahesh@lotus.iitm.ernet.in) Received: from localhost (mahesh@localhost) by lotus.iitm.ernet.in (8.8.5/8.8.5) with SMTP id PAA31265 for ; Wed, 9 Dec 1998 15:50:05 +0530 Date: Wed, 9 Dec 1998 15:50:05 +0530 (IST) From: UMAMAHESH To: tcp-impl@cthulhu.engr.sgi.com Subject: subscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-tcp-impl@cthulhu.engr.sgi.com Precedence: bulk subscribe From owner-tcp-impl@lerc.nasa.gov Fri Dec 11 19:03:10 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA13498 for ; Fri, 11 Dec 1998 19:03:09 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id RAA08326; Fri, 11 Dec 1998 17:25:12 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from hotmail.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with SMTP (NASA LeRC 8.7.4.1/2.01-main) id RAA07074; Fri, 11 Dec 1998 17:13:07 -0500 (EST) Received: (qmail 5255 invoked by uid 0); 11 Dec 1998 22:04:53 -0000 Message-ID: <19981211220453.5254.qmail@hotmail.com> Received: from 192.9.25.22 by www.hotmail.com with HTTP; Fri, 11 Dec 1998 14:04:52 PST X-Originating-IP: [192.9.25.22] From: "Padmapriya Parthasarathy" To: tcp-impl@lerc.nasa.gov Subject: Window scale factor ? MIME-Version: 1.0 Content-Type: text/plain Date: Fri, 11 Dec 1998 14:04:52 PST Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk According to RFC1323, the maximum value for ws is 14. If TCP did not use signed arithmetic (all SEQ_XX macros are signed) for detecting old/new data and used unsigned arithmetic with proper wraparound checks, would the value for ws be 15 ? Thanks Priya ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From owner-tcp-impl@lerc.nasa.gov Fri Dec 11 20:03:39 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA13915 for ; Fri, 11 Dec 1998 20:03:39 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id SAA15273; Fri, 11 Dec 1998 18:33:57 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from frantic.bsdi.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id SAA15269; Fri, 11 Dec 1998 18:33:54 -0500 (EST) Received: (from dab@localhost) by frantic.bsdi.com (8.9.0/8.9.0) id RAA08003; Fri, 11 Dec 1998 17:33:52 -0600 (CST) Date: Fri, 11 Dec 1998 17:33:52 -0600 (CST) From: David Borman Message-Id: <199812112333.RAA08003@frantic.bsdi.com> To: padmapriyap@hotmail.com, tcp-impl@lerc.nasa.gov Subject: Re: Window scale factor ? Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk > From: "Padmapriya Parthasarathy" > Subject: Window scale factor ? > Date: Fri, 11 Dec 1998 14:04:52 PST > > According to RFC1323, the maximum value for ws is 14. If TCP did > not use signed arithmetic (all SEQ_XX macros are signed) > for detecting old/new data and used unsigned arithmetic with > proper wraparound checks, would the value for ws be 15 ? No. Think of the sequence space as a circle. The 1/2 of the circle ahead of any given point is the upcoming data, the 1/2 of the circle behind it is the old data. You don't want to advertise a window of all the new data, because then you are butting right up agains the old data. So, you limit the window to 1/2 of the upcoming data, or 1/4 of the total sequence space. That give you a maximum window shift of 14. -David Borman, dab@bsdi.com From owner-tcp-impl@lerc.nasa.gov Tue Dec 15 11:19:22 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18329 for ; Tue, 15 Dec 1998 11:19:21 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id IAA17364; Tue, 15 Dec 1998 08:55:36 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from berserkly.bsdi.com (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id IAA17350; Tue, 15 Dec 1998 08:55:32 -0500 (EST) Received: from berserkly.bsdi.com (localhost.pe1hzg.ampr.org [127.0.0.1]) by berserkly.bsdi.com (8.9.0/8.9.0) with ESMTP id OAA19267 for ; Tue, 15 Dec 1998 14:55:16 +0100 (CET) Message-Id: <199812151355.OAA19267@berserkly.bsdi.com> To: tcp-impl@lerc.nasa.gov From: Geert Jan de Groot Subject: Re: put on your PMTU discovery thinking caps In-reply-to: Your message of "Thu, 03 Dec 1998 23:43:57 PST." <199812040743.XAA01294@daffy.ee.lbl.gov> Date: Tue, 15 Dec 1998 14:55:05 +0100 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk On Thu, 03 Dec 1998 23:43:57 PST Vern Paxson wrote: > We've heard a number of people mention that it would be good to document > some of the pitfalls & problems of implementing PMTU discovery (something > with a bit more meat than RFC 1435). I have seen a particular failure mode a number of times; while it's not an implementation fault, the failure mode is arcane enough that it has lead to a number of escalations and it would be good to document it. For example, see the following scenario: ------------- ---------- | | | | (1) | | (2) | | (3) <========| ALG/proxy |----------| Router |--------> ISP | | | | | | | | ------------- ---------- (1) is a FDDI ring to a customer's internal network (MTU=4470) (2) is a DMZ existing of a regular ethernet (3) is a T1 serial link to an ISP. There is a configuration error in this network: the MTU on the serial line of the router is 1500 bytes (default); the ISP's end of the serial line is 4470 bytes. The rest of the path to a remote web server has MTU=4470 all the way, so that packets of this size can be sent by the web server to the ISP end of serial line (3) without being fragmented along the path. Without PMTU discovery, this MTU mismatch would go undetected as the proxy would send a small MSS (1460). Because of this, the web server will send 1460-byte packets which will pass Router without a problem. With PMTU however, the host MUST send a large MSS even over ethernet (so that it can take advantage if a route flap causes the traffic to go over FDDI, we could use FDDI's MTU). So, we'll send MSS=4430 and as a result, the web server will send 4470-bytes packets, which will be sent through the ISP's router onto the serial line. However, because Router's serial interface is configured for MTU=1500, the packet will get dropped because it's oversized. No ICMP error message is sent because Router can't receive the whole packet and hence can't generate an ICMP message. Because no ICMP message is sent, packet retransmissions from the web server will have the same size and will also be silently dropped. This leads to the following scenario: - One is able to ping remote site - When sending a HTTP request (in this example), the TCP handshake happens, one is able to send the HTTP request (and have the request data ACKed by TCP), and then total silence till things time out, leading to responses like "the web site doesn't respond", or worse "this problem started with your new release with PMTU discovery, so your product has a problem with PMTU discovery" Typically I detect this kind of problem by sending large-size ping packets to the remote end of a suspect link. If the link is misconfigured, then the ICMP echo response won't make it back to us (but it will work for small-size ICMP echo packets). Traceroute won't be able to detect this because the response packet is always small. I think that documenting these types of scenarios (which is similar to Matt's) would be very useful. While it's indeed not in the direct scope of the tcp-IMPL wg, this venue does have the people with the right experience so that a good document can be made. I have done a little testing and found this type of problem more common than I expected, so writing it up would be good. Geert Jan PS: I just saw a note in the is-is WG where they discussed making the HELLO packet max size, just to detect his problem. I'm not sure if putting this kind of test in a routing protocol is a good idea, but having a widespread test would be good. From owner-tcp-impl@lerc.nasa.gov Tue Dec 15 20:06:23 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27785 for ; Tue, 15 Dec 1998 20:06:23 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id SAA01469; Tue, 15 Dec 1998 18:30:50 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from ice.sni.net (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id SAA01465; Tue, 15 Dec 1998 18:30:48 -0500 (EST) Received: from ice.sni.net (localhost [127.0.0.1]) by ice.sni.net (8.9.1/8.9.1) with ESMTP id QAA00325 for ; Tue, 15 Dec 1998 16:30:54 -0700 (MST) Message-Id: <199812152330.QAA00325@ice.sni.net> X-Mailer: exmh version 2.0.2 2/24/98 To: tcp-impl@lerc.nasa.gov From: Danny McPherson Reply-To: danny@qwest.net Subject: Re: put on your PMTU discovery thinking caps Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 15 Dec 1998 16:30:54 -0700 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Note that ISIS-WG is not "putting this kind of test in a routing protocol", it's been there for quite some time. We're merely leaving it until something better comes along. It is quite useful, though. -danny > PS: I just saw a note in the is-is WG where they discussed making the > HELLO packet max size, just to detect his problem. I'm not sure if > putting this kind of test in a routing protocol is a good idea, > but having a widespread test would be good. From owner-tcp-impl@lerc.nasa.gov Fri Dec 18 14:39:52 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07946 for ; Fri, 18 Dec 1998 14:39:51 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id MAA01664; Fri, 18 Dec 1998 12:51:24 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from guns.lerc.nasa.gov (guns.lerc.nasa.gov [139.88.44.160]) by assateague.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id MAA01659; Fri, 18 Dec 1998 12:51:22 -0500 (EST) Received: from guns.lerc.nasa.gov by guns.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-local) id MAA07588; Fri, 18 Dec 1998 12:51:21 -0500 (EST) Message-Id: <199812181751.MAA07588@guns.lerc.nasa.gov> To: tcp-impl@lerc.nasa.gov Cc: "Vern Paxson" From: Mark Allman Reply-To: mallman@lerc.nasa.gov Subject: new version of TCP Congestion Control Organization: Late Night Hackers, NASA Lewis, Cleveland, Ohio Song-of-the-Day: Friend Of The Devil Date: Fri, 18 Dec 1998 12:51:21 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk Folks- We just sent in a new version of the ``TCP Congestion Control'' I-D (which will be draft-ietf-tcpimpl-cong-control-02.txt). It should be available from the archives soon. You can grab a copy of the I-D, as well as context and word diffs between this version and the last version at: http://gigahertz.lerc.nasa.gov/~mallman/share/tcpimpl/ The following changes were either minor editting nits or in response to things raised in the last call period... -Minor wording tweaks -Changed the allowed initial window from 1*MSS to up to 2*MSS, per the short discussion in Orlando. -We deleted the MSS definition and replaced it with two new definitions. SMSS, which is the segment size the sender uses (per the MTU or Path MTU Discovery, or whatever). And, RMSS is the segment size announced by the receiver in the MSS option (or the default, if the option is not used). Hopefully, by using more specific terms, we have made the text more clear. We believe there is strong concensus for this document now and plan is to send this document to the IESG soon (probably next week with the newreno draft, assuming no large objections to that document appear before the WG last-call ends next Tuesday). allman From owner-tcp-impl@lerc.nasa.gov Sat Dec 19 05:04:13 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA23898 for ; Sat, 19 Dec 1998 05:04:13 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id DAA27338; Sat, 19 Dec 1998 03:03:23 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from daffy.ee.lbl.gov (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id DAA27334; Sat, 19 Dec 1998 03:03:21 -0500 (EST) Received: (from vern@localhost) by daffy.ee.lbl.gov (8.9.1/8.9.1) id AAA07813; Sat, 19 Dec 1998 00:03:20 -0800 (PST) Message-Id: <199812190803.AAA07813@daffy.ee.lbl.gov> To: tcp-impl@lerc.nasa.gov Cc: mallman@lerc.nasa.gov Subject: Re: new version of TCP Congestion Control In-reply-to: Your message of Fri, 18 Dec 1998 12:51:21 PST. Date: Sat, 19 Dec 1998 00:03:20 PST From: Vern Paxson Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk > -Changed the allowed initial window from 1*MSS to up to 2*MSS, per > the short discussion in Orlando. For those of you who weren't at the meeting, this change was approved by strong acclamation. Vern From owner-tcp-impl@cthulhu.engr.sgi.com Sun Dec 20 08:02:59 1998 Received: from sgi.sgi.com (SGI.COM [192.48.153.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02140 for ; Sun, 20 Dec 1998 08:02:58 -0500 (EST) Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id EAA03128; Sun, 20 Dec 1998 04:58:46 -0800 (PST) mail_from (owner-tcp-impl@cthulhu.engr.sgi.com) Received: (from majordomo-owner@localhost) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id EAA52174 for tcp-impl-list; Sun, 20 Dec 1998 04:53:40 -0800 (PST) mail_from (owner-tcp-impl@relay.engr.sgi.com) Received: from sgi.sgi.com (sgi.engr.sgi.com [192.26.80.37]) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id EAA68477 for ; Sun, 20 Dec 1998 04:53:39 -0800 (PST) mail_from (ndj1@mailcity.com) Received: from mexpru.mcsa.net.mx (mexpru.mcsa.net.mx [200.33.47.221]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id EAA01877 for ; Sun, 20 Dec 1998 04:53:38 -0800 (PST) mail_from (ndj1@mailcity.com) From: ndj1@mailcity.com Received: from jamie (167.aebc.com [204.244.170.167]) by mexpru.mcsa.net.mx (8.8.5/SCO5) with SMTP id DAA22375; Sun, 20 Dec 1998 03:57:54 -0600 (CST) Date: Sun, 20 Dec 1998 03:57:54 -0600 (CST) Message-Id: <199812200957.DAA22375@mexpru.mcsa.net.mx> To: connie@excite.com Subject: re: Your Account info. Sender: owner-tcp-impl@cthulhu.engr.sgi.com Precedence: bulk From owner-tcp-impl@lerc.nasa.gov Wed Dec 23 16:13:18 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id QAA10151 for ; Wed, 23 Dec 1998 16:13:17 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id OAA07927; Wed, 23 Dec 1998 14:24:40 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from ietf.org (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id OAA07920; Wed, 23 Dec 1998 14:24:37 -0500 (EST) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.8.5/8.8.7a) with ESMTP id OAA04770; Wed, 23 Dec 1998 14:24:34 -0500 (EST) Message-Id: <199812231924.OAA04770@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: IETF-Announce:;;@ns.cnri.reston.va.us Cc: tcp-impl@lerc.nasa.gov From: Internet-Drafts@ietf.org Reply-to: Internet-Drafts@ietf.org Subject: I-D ACTION:draft-ietf-tcpimpl-cong-control-02.txt Date: Wed, 23 Dec 1998 14:24:33 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the TCP Implementation Working Group of the IETF. Title : TCP Congestion Control Author(s) : M. Allman, V. Paxson, W. Stevens Filename : draft-ietf-tcpimpl-cong-control-02.txt Pages : 11 Date : 22-Dec-98 This document defines TCP's four intertwined congestion control algorithms: slow start, congestion avoidance, fast retransmit, and fast recovery. In addition, the document specifies how TCP should begin transmission after a relatively long idle period, as well as discussing various acknowledgment generation methods. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-tcpimpl-cong-control-02.txt". Internet-Drafts directories are located at: Africa: ftp.is.co.za Europe: ftp.nordu.net ftp.nic.it Pacific Rim: munnari.oz.au US East Coast: ftp.ietf.org US West Coast: ftp.isi.edu Internet-Drafts are also available by mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <19981222172428.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-tcpimpl-cong-control-02.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <19981222172428.I-D@ietf.org> --OtherAccess-- --NextPart-- From owner-tcp-impl@lerc.nasa.gov Wed Dec 23 20:06:35 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id UAA14843 for ; Wed, 23 Dec 1998 20:06:35 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id SAA26657; Wed, 23 Dec 1998 18:40:11 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from saba.cs.washington.edu (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id SAA26492; Wed, 23 Dec 1998 18:40:07 -0500 (EST) Received: from localhost (cardwell@localhost) by saba.cs.washington.edu (8.8.8+CS/7.2ws+) with SMTP id PAA31903; Wed, 23 Dec 1998 15:40:05 -0800 Date: Wed, 23 Dec 1998 15:40:05 -0800 (PST) From: Neal Cardwell To: Mark Allman cc: tcp-impl@lerc.nasa.gov, Vern Paxson Subject: Re: new version of TCP Congestion Control In-Reply-To: <199812181751.MAA07588@guns.lerc.nasa.gov> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk There seems to be a small ambiguity in section 3.1, "Slow Start and Congestion Avoidance": The slow start algorithm is used when cwnd < ssthresh, while the congestion avoidance algorithm is used when cwnd > ssthresh. When cwnd and ssthresh are equal the sender may use either slow start or congestion avoidance. During slow start, a TCP increments cwnd by at most MSS bytes for each ACK received that acknowledges new data. Slow start ends when cwnd exceeds ssthresh (or, optionally, when it reaches it, as noted above); or when cwnd reaches rwnd; or when congestion is observed. This last sentence says that slow start ends when cwnd reaches rwnd. This seems to confuse things a bit. I suspect this sentence is alluding to the fact that the exponential growth in sending rate that we usually associate with slow start is going to stop when cwnd hits the rwnd ceiling. But it's my understanding from reading the popular TCP implementations that the more fundamental aspect of slow start -- the increase of cwnd by SMSS bytes for each ACK received -- continues even when cwnd is above rwnd, as long as cwnd is still below (or at) ssthresh. So either a) TCP is supposed to exit slow start (stop the exponential growth of cwnd) when cwnd reaches rwnd. This makes a lot of sense, but i don't think this is what most TCP implementations do, and i don't think this is what the draft is proposing. If it is proposing this, then i think the draft should be more explicit about what TCP is supposed to do when cwnd hits rwnd. or b) TCP is not supposed to exit slow start (stop the exponential growth of cwnd) when cwnd reaches rwnd. In this case, i guess the "or when cwnd reaches rwnd; " phrase above should be omitted from the draft. Happy Holidays! neal ps: btw, i guess the "MSS" in the second paragraph above should be "SMSS". From owner-tcp-impl@lerc.nasa.gov Thu Dec 24 12:11:10 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id MAA05337 for ; Thu, 24 Dec 1998 12:11:09 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id KAA24016; Thu, 24 Dec 1998 10:36:55 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from ietf.org (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id KAA24009; Thu, 24 Dec 1998 10:36:52 -0500 (EST) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.8.5/8.8.7a) with ESMTP id KAA00538; Thu, 24 Dec 1998 10:36:49 -0500 (EST) Message-Id: <199812241536.KAA00538@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: IETF-Announce:;;@ns.cnri.reston.va.us Cc: tcp-impl@lerc.nasa.gov From: Internet-Drafts@ietf.org Reply-to: Internet-Drafts@ietf.org Subject: I-D ACTION:draft-ietf-tcpimpl-cong-control-02.txt Date: Thu, 24 Dec 1998 10:36:48 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the TCP Implementation Working Group of the IETF. Title : TCP Congestion Control Author(s) : M. Allman, V. Paxson, W. Stevens Filename : draft-ietf-tcpimpl-cong-control-02.txt Pages : 11 Date : 22-Dec-98 This document defines TCP's four intertwined congestion control algorithms: slow start, congestion avoidance, fast retransmit, and fast recovery. In addition, the document specifies how TCP should begin transmission after a relatively long idle period, as well as discussing various acknowledgment generation methods. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-tcpimpl-cong-control-02.txt". Internet-Drafts directories are located at: Africa: ftp.is.co.za Europe: ftp.nordu.net ftp.nic.it Pacific Rim: munnari.oz.au US East Coast: ftp.ietf.org US West Coast: ftp.isi.edu Internet-Drafts are also available by mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <19981222172428.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-tcpimpl-cong-control-02.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-tcpimpl-cong-control-02.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <19981222172428.I-D@ietf.org> --OtherAccess-- --NextPart-- From owner-tcp-impl@lerc.nasa.gov Thu Dec 24 12:58:03 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id MAA08538 for ; Thu, 24 Dec 1998 12:58:02 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id LAA28905; Thu, 24 Dec 1998 11:47:28 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from ietf.org (fw01.lerc.nasa.gov [139.88.145.14]) by assateague-fi.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id LAA28900; Thu, 24 Dec 1998 11:47:26 -0500 (EST) Received: from CNRI.Reston.VA.US (localhost [127.0.0.1]) by ietf.org (8.8.5/8.8.7a) with ESMTP id LAA03859; Thu, 24 Dec 1998 11:47:22 -0500 (EST) Message-Id: <199812241647.LAA03859@ietf.org> Mime-Version: 1.0 Content-Type: Multipart/Mixed; Boundary="NextPart" To: IETF-Announce:;;@ns.cnri.reston.va.us Cc: tcp-impl@lerc.nasa.gov From: Internet-Drafts@ietf.org Reply-to: Internet-Drafts@ietf.org Subject: I-D ACTION:draft-ietf-tcpimpl-newreno-01.txt Date: Thu, 24 Dec 1998 11:47:22 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk --NextPart A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the TCP Implementation Working Group of the IETF. Title : The NewReno Modification to TCP's Fast Recovery Algorithm Author(s) : S. Floyd, T. Henderson Filename : draft-ietf-tcpimpl-newreno-01.txt Pages : 11 Date : 23-Dec-98 RFC 2001 [RFC2001] documents the following four intertwined TCP congestion control algorithms: Slow Start, Congestion Avoidance, Fast Retransmit, and Fast Recovery. RFC 2001-bis [RFC2001-bis] explicitly allows certain modifications of these algorithms, including modifications that use the TCP Selective Acknowledgement (SACK) option [MMFR96], and modifications that respond to ``partial acknowledgments'' (ACKs which cover new data, but not all the data outstanding when loss was detected) in the absence of SACK. This document describes a specific algorithm for responding to partial acknowledgments, referred to as NewReno. This response to partial acknowledgments was first proposed by Janey Hoe in [Hoe95]. A URL for this Internet-Draft is: http://www.ietf.org/internet-drafts/draft-ietf-tcpimpl-newreno-01.txt Internet-Drafts are also available by anonymous FTP. Login with the username "anonymous" and a password of your e-mail address. After logging in, type "cd internet-drafts" and then "get draft-ietf-tcpimpl-newreno-01.txt". Internet-Drafts directories are located at: Africa: ftp.is.co.za Europe: ftp.nordu.net ftp.nic.it Pacific Rim: munnari.oz.au US East Coast: ftp.ietf.org US West Coast: ftp.isi.edu Internet-Drafts are also available by mail. Send a message to: mailserv@ietf.org. In the body type: "FILE /internet-drafts/draft-ietf-tcpimpl-newreno-01.txt". NOTE: The mail server at ietf.org can return the document in MIME-encoded form by using the "mpack" utility. To use this feature, insert the command "ENCODING mime" before the "FILE" command. To decode the response(s), you will need "munpack" or a MIME-compliant mail reader. Different MIME-compliant mail readers exhibit different behavior, especially when dealing with "multipart" MIME messages (i.e. documents which have been split up into multiple messages), so check your local documentation on how to manipulate these messages. Below is the data which will enable a MIME compliant mail reader implementation to automatically retrieve the ASCII version of the Internet-Draft. --NextPart Content-Type: Multipart/Alternative; Boundary="OtherAccess" --OtherAccess Content-Type: Message/External-body; access-type="mail-server"; server="mailserv@ietf.org" Content-Type: text/plain Content-ID: <19981223132700.I-D@ietf.org> ENCODING mime FILE /internet-drafts/draft-ietf-tcpimpl-newreno-01.txt --OtherAccess Content-Type: Message/External-body; name="draft-ietf-tcpimpl-newreno-01.txt"; site="ftp.ietf.org"; access-type="anon-ftp"; directory="internet-drafts" Content-Type: text/plain Content-ID: <19981223132700.I-D@ietf.org> --OtherAccess-- --NextPart-- From owner-tcp-impl@cthulhu.engr.sgi.com Mon Dec 28 04:27:29 1998 Received: from sgi.sgi.com (SGI.COM [192.48.153.1]) by ietf.org (8.8.5/8.8.7a) with ESMTP id EAA17749 for ; Mon, 28 Dec 1998 04:27:28 -0500 (EST) Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id BAA00940; Mon, 28 Dec 1998 01:22:20 -0800 (PST) mail_from (owner-tcp-impl@cthulhu.engr.sgi.com) Received: (from majordomo-owner@localhost) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id BAA89401 for tcp-impl-list; Mon, 28 Dec 1998 01:13:29 -0800 (PST) mail_from (owner-tcp-impl@relay.engr.sgi.com) Received: from sgi.sgi.com (sgi.engr.sgi.com [192.26.80.37]) by cthulhu.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id BAA74853 for ; Mon, 28 Dec 1998 01:13:28 -0800 (PST) mail_from (tempting.tear-outs@wolf.net) Received: from mail.wolf.net (wolf1.wolf.net [209.49.165.6]) by sgi.sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via SMTP id BAA05773 for ; Mon, 28 Dec 1998 01:13:27 -0800 (PST) mail_from (tempting.tear-outs@wolf.net) Received: from [207.240.233.19] [207.240.233.19] by mail.wolf.net with ESMTP (SMTPD32-4.07) id AE99DBF0228; Mon, 28 Dec 1998 03:17:29 EST5EDT Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 28 Dec 1998 00:28:06 -0400 To: Tempting Tear-Outs From: Tempting Tear-Outs Subject: 055 ===>> FREE 1 yr USA Magazine Sub sent worldwide-200+ Choices! Up to $50.00 value! Sender: owner-tcp-impl@cthulhu.engr.sgi.com Precedence: bulk To be removed from our mailing list, send a blank email message, with the subject of "remove" to: tempting.tear-outs@wolf.net -- FOR MORE INFO: please "cut out" the below form on the "cut" lines shown, and fax it, for the fastest reply to: 1-718-227-9125 (this is a fax # in the USA) or send via smail (first class mail or airmail) to: Tempting Tear-Outs Att. Free-catalogue-by-email Dept 3835 Richmond Ave. Suite #200 Staten Island NY 10312-3828 USA SORRY, BUT.... our software is not set up to accept the below form via return email; WE CAN ONLY acknowledge forms sent in via fax or smail. --> IMPORTANT complete directions, to ensure that you get a reply, and more info follow, below the reply form and the catalogue options. *------------cut here/begin-------------------------------------------* Name (First Middle Last): Internet email address: Smail home address: City-State-Zip: Country: Work Tel. #: Work Fax #: Home Tel. #: Home Fax #: Cellular (Mobile) Tel. #: Beeper (Pager) Tel. #: How did you hear about us (name of person/company who referred you or the area of the internet that you saw us mentioned in): Referred by: Tempting Tear-Outs 0122798-l-aaa Name of USA mags you currently get on the newsstand or in the store: Name of USA mags you currently get on a subscription basis, through the mail: Name of USA mags you would like price quotes on when we call you: Catalogue version desired (list number of choice below): *------------cut here/end--------------------------------------------* CATALOGUE VERSION CHOICES: 1. This version can be read by everyone, no matter what type of computer you use, or what type of software you use. It is a simple format, with just our entire catalogue pasted into the body of a single email message, 316K in size. If you use pine or elm on a unix system or an advanced software version such as Eudora Pro 3.0 or later, you will most likely receive it as a single email message. However, if your software limits incoming email messages to a certain size, say 32K or so, then your software will split it into multiple email message parts. Whether you receive it as a single email message or multiple part email messages, you can easily paste it into one whole text document with your word processor, in about 10 minutes or so. 2. For more advanced computer users: attached plain ascii text file ~316K - you must know how to download an attached text file and then be able to locate it on your hard drive or system home directory; it can then be opened with any pc or mac word processing software. If in doubt, don't ask for this version. This isn't for internet *newbies.* Better to order option 1 and spend a few minutes pasting them into one whole text document with your word processor, than to waste hours trying to figure how to deal with this option. This version is great for doing keyword searches and jumping around within the catalogue with your word processing software, if your normal email reading software doesn't allow this. VERY IMPORTANT DIRECTIONS TO ENSURE THAT YOU GET A REPLY: 1. you must call from an "unblocked number," ie. one that is not blocked from caller id. We are very sorry for this requirement, but our fax software requires this before it allows an incoming fax call to connect. If you have a blocked number, you must first unblock it. In most cases this means dialing *82 from a touch-tone phone (or 1182 from a rotary phone) before you dial 1-718-227-9125. NOTE: If you are not sure if your number is blocked, just try dialing our fax # normally. If you don't get a recording telling you your number is blocked, your number has been transmitted and you may press the start button on your fax when you hear the fax tone from our fax. 2. no reply forms can be accepted by email....only via fax or smail. 3. your form must be typewritten or printed out on your computer printer before you fax it; sorry, but *no* handwritten forms will be acknowledged. If you can't find someone with a typewriter or a computer printer, we apologize for not being able to reply to you. 4. faxes with cover pages will be rejected. You must send *only* the reply form. 5. forms not *completely* filled in will not be acknowledged. 6. you will receive a reply within 1 business day directly from the company making the offer via email. Therefore you must have an email address. If you read this message, then you must have an email address, or access to one, at least. :-) 7. your fax must not exceed 1 page in length. Faxes of 2 or more pages will be sensed, then auto-terminated and deleted. Your fax goes directly onto our 5.0 gigabyte hard drive and we must limit all incoming faxes to 1 page. 8. all faxes must begin with: *------------cut here/begin-------------------------------------------* and must end with: *------------cut here/end--------------------------------------------* 9. Any fax not conforming to this format will be sensed by our software, then auto-terminated and deleted from the hard drive, before any human ever gets to see it. 10. The type on your fax must be dark and legible. If in doubt, please print it out darker before faxing it in. If we can't read it, we can't reply to you or send you our FREE catalogue. :-( 11. If this all seems too complicated for faxing, just do it the old fashioned way via smail!!! WHO WE ARE: Tempting Tear-Outs is an advertising company that brings potential new customers to the companies they advertise for. MORE ABOUT THE COMPANY MAKING THE FREE OFFER: The company making the offer is a magazine subscription agency based in the USA. They have over 1,100 popular USA titles available to be shipped to *any* country, including of course, to anywhere in the USA! They offer a FREE 1 yr. subscription to your choice of over 200 of the titles in their catalogue to any new customer using them for the first time. The dollar value of the freebies, based on the subscription prices directly from the publishers, ranges from $6.97 all the way up to $50.00! For new customers in the USA, there is no charge for FPH (foreign postage & handling), so the freebie is 100% free! For new customers living overseas, the only charge on the freebie would be for the FPH (foreign postage & handling). Their president has been in the magazine subscription business since 1973 and they are very customer-service oriented. They will even help you with address changes on your magazines, even if you move from one country to another country. They have thousands of happy customers in over 59 countries. Their price guarantee is very simple: they guarantee that their subscription prices are the lowest available and they will BEAT any legitimate, verifiable offer before you pay them or match it afterwards, by refunding you the difference in price PLUS the cost of the postage stamp you would use sending in the special offer to them, even 6 months after you pay them, as long as it was current at the time of your offer. Does that sound fair? Wouldn't it be great if everything you bought came with that price guarantee? Sometimes they are less than half of the next best deal out there, sometimes just a little cheaper, but always you get the lowest rates without having to shop around. With 1,100+ titles on their list, they would like to think that they have also the best selection around! Within the USA, for their USA customers, they are cheaper than all their competitors and even the publishers themselves. This is their price guarantee. The 1 yr. freebie that you get with your first order is completely free! Overseas, (even after you factor in the cost of the FPH (foreign postage & handling) and the conversion from USA Dollars to your currency), on the average, they are generally around one-fourth to one-half of what the newsstands overseas charge locally for USA magazines. On some titles they are as little as one-tenth of what the newsstands charge. They are also the cheapest subscription source for delivery overseas, including directly from the publishers themselves! Some publishers don't even offer subscriptions overseas.........but overseas subscriptions are this company's specialty! They feel that magazines should not be a luxury overseas. In the USA, people buy magazines and then toss them after reading them for just a few minutes or hours. They are so cheap in the USA! Well, this company would like to make it the same way for their overseas customers. They are also cheaper than all their competitors in the USA and overseas, including the publishers themselves! It is also *highly unlikely* you will find any of their USA competitors calling you overseas, in order to offer that personal touch, just to sell you a couple of magazines! But that is what this company specializes in and loves doing! Around one-half their business comes from overseas, so they are very patient with new customers who only speak limited English as a 2nd language. Subscription prices quoted for overseas consist of the subscription price, plus the FPH. You add the two together and that is your total cost. The exception is the 1 yr. freebie you get with your first order. On that title, you pay *only* the FPH for the 1 yr. term. Their prices are so cheap because when you deal with them, you cut-out all the middlemen. HERE IS HOW YOU CAN GET MORE INFO AND GET STARTED WITH THEM: Simply fax or smail back to us the reply form listed at the top of this message. We will then forward your form on to the subscription agency. They will then email their "big and juicy" catalogue to you, in whichever of the two formats you chose. The catalogue is FREE and makes for hours of fascinating reading, on its own. It includes the complete list of freebies, a complete list of all the titles they sell, as well as detailed descriptions on most of the titles, along with lists of titles by category of interest and their terms of sale. They will then give you a friendly, no-pressure, no obligation, 5-minute call to go over how they work and to answer any questions that you might have, as well as give you up-to-the minute price quotes on any titles you might be considering. They will call you in whatever country you live in, taking the time difference into account. As they like to emphasize the personal touch they give to each new customer, all first-time orders can only be done via phone, so they can answer all your questions completely and personally. Once you have placed your first order via phone, you will be able to place future orders and make inquiries on your account, get price quotes, etc., all via email, if that is most convenient for you. Within the USA, they accept payment via check over the phone, Mastercard, Visa, American Express, Diner's Club and Carte Blanche. Overseas, they accept Mastercard, Visa, American Express, Diner's Club and Carte Blanche, even if your credit card is a local one in local currency (that most merchants in the USA would not normally be willing to accept). That's our introduction of our client that we represent. We hope that we have piqued your interest and that you will take the next step to get their free catalogue! Thank you for your time and interest. -- Tempting Tear-Outs. For more info on advertising rates, please write us on your company letterhead, w/business card, via smail to: Tempting Tear-Outs, 3835 Richmond Ave. Suite #200, Staten Island NY 10312-3828, USA. From owner-tcp-impl@lerc.nasa.gov Mon Dec 28 19:46:27 1998 Received: from assateague.lerc.nasa.gov (assateague-fi.lerc.nasa.gov [139.88.112.23]) by ietf.org (8.8.5/8.8.7a) with ESMTP id TAA00800 for ; Mon, 28 Dec 1998 19:46:26 -0500 (EST) Received: (listserv@localhost) by assateague.lerc.nasa.gov (NASA LeRC 8.7.4.1/2.01-main) id SAA20554; Mon, 28 Dec 1998 18:06:38 -0500 (EST) X-Authentication-Warning: assateague-fi.lerc.nasa.gov: listserv set sender to owner-tcp-impl@lerc.nasa.gov using -f Received: from tuvok.lerc.nasa.gov (ppp21.lerc.nasa.gov [139.88.123.31]) by assateague.lerc.nasa.gov with ESMTP (NASA LeRC 8.7.4.1/2.01-main) id SAA20550; Mon, 28 Dec 1998 18:06:34 -0500 (EST) Received: from tuvok.lerc.nasa.gov (localhost [127.0.0.1]) by tuvok.lerc.nasa.gov (8.8.5/8.8.5) with ESMTP id SAA00790; Mon, 28 Dec 1998 18:07:41 -0500 (EST) Message-Id: <199812282307.SAA00790@tuvok.lerc.nasa.gov> To: Neal Cardwell From: Mark Allman Reply-To: mallman@lerc.nasa.gov cc: tcp-impl@lerc.nasa.gov, Vern Paxson Subject: Re: new version of TCP Congestion Control Organization: Late Night Hackers, NASA LeRC, Cleveland, Ohio Song-of-the-Day: Heavy Fuel Date: Mon, 28 Dec 1998 18:07:40 -0500 Sender: owner-tcp-impl@lerc.nasa.gov Precedence: bulk > This last sentence says that slow start ends when cwnd reaches > rwnd. Neal- Thanks for pointing this out. Depending on the implementation, cwnd may grow well past rwnd. Although, I think it would also be fine for an implementation to cap cwnd growth at rwnd, as well. In any case, I think it is an implementation issue. So, we are going to to remove the language that indicates that slow start ends when rwnd is reached and send out a new version of the document soon (and, submit the document to the IESG). Oh, and thanks for catching the remaining "MSS" reference. I made another pass and hopefully have caught all of them now. Thanks, allman