From mailnull@www1.ietf.org Wed Nov 20 14:25:53 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28218 for ; Wed, 20 Nov 2002 14:25:53 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gAKJS6b07014 for trigtran-archive@odin.ietf.org; Wed, 20 Nov 2002 14:28:06 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJS6v07011 for ; Wed, 20 Nov 2002 14:28:06 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28210 for ; Wed, 20 Nov 2002 14:25:22 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJS3v07003; Wed, 20 Nov 2002 14:28:03 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJRXv06947 for ; Wed, 20 Nov 2002 14:27:33 -0500 Received: from gandalf.icr.a-star.edu.sg (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28203 for ; Wed, 20 Nov 2002 14:24:48 -0500 (EST) Received: from gwserver.icr.a-star.edu.sg (gwserver.icr.a-star.edu.sg [137.132.31.253]) by gandalf.icr.a-star.edu.sg (8.12.2+Sun/8.12.2) with ESMTP id gAKJSTXw000497 for ; Thu, 21 Nov 2002 03:28:29 +0800 (SGT) Received: from yoyo ([204.42.66.233]) by gwserver.icr.a-star.edu.sg; Thu, 21 Nov 2002 03:21:06 +0800 Message-ID: <002101c290ca$bed21df0$e9422acc@yoyo> From: "Paul" To: Date: Wed, 20 Nov 2002 14:26:32 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_001E_01C290A0.D27F9960" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , This is a multi-part message in MIME format. ------=_NextPart_000_001E_01C290A0.D27F9960 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable subscribe tanpaul@cwc.nus.edu.sg ------=_NextPart_000_001E_01C290A0.D27F9960 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

subscribe tanpaul@cwc.nus.edu.sg

------=_NextPart_000_001E_01C290A0.D27F9960-- _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 20 14:37:51 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28567 for ; Wed, 20 Nov 2002 14:37:51 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gAKJe4C08281 for trigtran-archive@odin.ietf.org; Wed, 20 Nov 2002 14:40:04 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJe3v08278 for ; Wed, 20 Nov 2002 14:40:03 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28556 for ; Wed, 20 Nov 2002 14:37:20 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJe1v08261; Wed, 20 Nov 2002 14:40:01 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKJdSv08208 for ; Wed, 20 Nov 2002 14:39:28 -0500 Received: from nit.isi.edu (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28546 for ; Wed, 20 Nov 2002 14:36:44 -0500 (EST) Received: (from falk@localhost) by nit.isi.edu (8.11.6/8.11.6) id gAKJdAa29968; Wed, 20 Nov 2002 11:39:10 -0800 Date: Wed, 20 Nov 2002 11:39:10 -0800 From: Aaron Falk To: Paul Cc: trigtran@ietf.org Subject: Re: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg Message-ID: <20021120193910.GD29764@isi.edu> References: <002101c290ca$bed21df0$e9422acc@yoyo> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <002101c290ca$bed21df0$e9422acc@yoyo> User-Agent: Mutt/1.4i Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Finally, some traffic on this list. I thought I wasn't subscribed. :) BTW, Paul, you need to send that message to trigtran-request@ietf.org. --aaron Paul wrote: > subscribe tanpaul@cwc.nus.edu.sg _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 20 15:32:51 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00119 for ; Wed, 20 Nov 2002 15:32:50 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gAKKZ4O11615 for trigtran-archive@odin.ietf.org; Wed, 20 Nov 2002 15:35:04 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKKZ4v11612 for ; Wed, 20 Nov 2002 15:35:04 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00108 for ; Wed, 20 Nov 2002 15:32:20 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKKZ2v11592; Wed, 20 Nov 2002 15:35:02 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKKYMv11533 for ; Wed, 20 Nov 2002 15:34:22 -0500 Received: from gandalf.icr.a-star.edu.sg (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00094 for ; Wed, 20 Nov 2002 15:31:34 -0500 (EST) Received: from gwserver.icr.a-star.edu.sg (gwserver.icr.a-star.edu.sg [137.132.31.253]) by gandalf.icr.a-star.edu.sg (8.12.2+Sun/8.12.2) with ESMTP id gAKKZEXw000917 for ; Thu, 21 Nov 2002 04:35:14 +0800 (SGT) Received: from yoyo ([204.42.66.233]) by gwserver.icr.a-star.edu.sg; Thu, 21 Nov 2002 04:27:56 +0800 Message-ID: <001001c290d4$104a5a90$e9422acc@yoyo> From: "Paul" To: "Aaron Falk" Cc: References: <002101c290ca$bed21df0$e9422acc@yoyo> <20021120193910.GD29764@isi.edu> Subject: Re: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg Date: Wed, 20 Nov 2002 15:33:10 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Content-Transfer-Encoding: 7bit Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Hi Aaron, thanks. so this is the second message on the list after the 55th IETF TrigTrans BOF meeting. Cheers, Paul ----- Original Message ----- From: "Aaron Falk" To: "Paul" Cc: Sent: Wednesday, November 20, 2002 2:39 PM Subject: Re: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg > Finally, some traffic on this list. I thought I wasn't subscribed. :) > > BTW, Paul, you need to send that message to trigtran-request@ietf.org. > > --aaron > > Paul wrote: > > subscribe tanpaul@cwc.nus.edu.sg > _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 20 16:17:59 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01312 for ; Wed, 20 Nov 2002 16:17:59 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gAKLKCC15074 for trigtran-archive@odin.ietf.org; Wed, 20 Nov 2002 16:20:12 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLKCv15067 for ; Wed, 20 Nov 2002 16:20:12 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01212 for ; Wed, 20 Nov 2002 16:17:25 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLK8v14896; Wed, 20 Nov 2002 16:20:08 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLJbv14772 for ; Wed, 20 Nov 2002 16:19:37 -0500 Received: from MAIL.cynetanetworks.com (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01125 for ; Wed, 20 Nov 2002 16:16:51 -0500 (EST) X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Date: Wed, 20 Nov 2002 15:19:28 -0600 Message-ID: <9255B6CD76A88943A3062F7D4E6432F5103090@mail.cynetanetworks.com> Thread-Topic: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg Thread-Index: AcKQ1GZvarMzGM9dQMCaCuJE6j1cKwABTGJA From: "Spencer Dawkins" To: Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gAKLJbv14773 Subject: [Trigtran] Trigtran Traffic and Low-Hanging Fruit Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Additional messages is a Step In The Right Direction. If we were turning up the volume AND adjusting the transmitter, my suggestion is to focus on what we identified as "low-hanging fruit" an hour ago. While we're gathering minutes, and figuring out what ELSE went on in the BOF, could we gather our thoughts on link/up? o What scenarios would benefit from a "far-end TRIGTRAN indication of LINK-UP"? o What are reasonable responses to this indication? o What are the pitfalls to avoid when making these responses? Thanks, Spencer, for Spencer and Carl > -----Original Message----- > From: Paul [mailto:tanpaul@cwc.nus.edu.sg] > Sent: Wednesday, November 20, 2002 3:33 PM > To: Aaron Falk > Cc: trigtran@ietf.org > Subject: Re: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg > > > Hi Aaron, > > thanks. > > so this is the second message on the list after the 55th > IETF TrigTrans BOF > meeting. > > Cheers, > Paul _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 20 16:42:04 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01948 for ; Wed, 20 Nov 2002 16:42:04 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gAKLiJ117078 for trigtran-archive@odin.ietf.org; Wed, 20 Nov 2002 16:44:19 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLiJv17075 for ; Wed, 20 Nov 2002 16:44:19 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01922 for ; Wed, 20 Nov 2002 16:41:33 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLiAv17034; Wed, 20 Nov 2002 16:44:10 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gAKLh2v16929 for ; Wed, 20 Nov 2002 16:43:02 -0500 Received: from emailO.iomega.com (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA01853 for ; Wed, 20 Nov 2002 16:40:15 -0500 (EST) Received: from royntex01.iomegacorp.com by emailO.iomega.com via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 20 Nov 2002 21:42:55 UT Received: from sdontfps01.iomegacorp.com ([147.178.169.11]) by royntex01.iomegacorp.com with Microsoft SMTPSVC(5.0.2195.4453); Wed, 20 Nov 2002 14:42:53 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: [Trigtran] Trigtran Traffic and Low-Hanging Fruit Date: Wed, 20 Nov 2002 13:42:52 -0800 Message-ID: Thread-Topic: [Trigtran] Trigtran Traffic and Low-Hanging Fruit Thread-Index: AcKQ1GZvarMzGM9dQMCaCuJE6j1cKwABTGJAAADcJ5A= From: "Sukanta Ganguly" To: X-OriginalArrivalTime: 20 Nov 2002 21:42:53.0354 (UTC) FILETIME=[C7C4F0A0:01C290DD] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gAKLh2v16930 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit For the far-end TRIGTRAN indication of link up some sort of state notion needs to be maintained. One of the important reasons to have a link up indication is that the previous communication channel, which was put on hold due to link down, can be continued. It does have a value of having a pause, continue model within the transport. Also, I do believe that this is a transport layer play and the unfortunate part of it is that the transport can get heavier as a consequence. If what I have said above makes sense to others then we got to have some notion of what constitutes the state of the channel? How can we keep it light but still retain all the required information to have the pause and continue working in place. Just some thoughts that came up in my mind. SG -----Original Message----- From: Spencer Dawkins [mailto:sdawkins@cynetanetworks.com] Sent: Wednesday, November 20, 2002 1:19 PM To: trigtran@ietf.org Subject: [Trigtran] Trigtran Traffic and Low-Hanging Fruit Additional messages is a Step In The Right Direction. If we were turning up the volume AND adjusting the transmitter, my suggestion is to focus on what we identified as "low-hanging fruit" an hour ago. While we're gathering minutes, and figuring out what ELSE went on in the BOF, could we gather our thoughts on link/up? o What scenarios would benefit from a "far-end TRIGTRAN indication of LINK-UP"? o What are reasonable responses to this indication? o What are the pitfalls to avoid when making these responses? Thanks, Spencer, for Spencer and Carl > -----Original Message----- > From: Paul [mailto:tanpaul@cwc.nus.edu.sg] > Sent: Wednesday, November 20, 2002 3:33 PM > To: Aaron Falk > Cc: trigtran@ietf.org > Subject: Re: [Trigtran] subscribe tanpaul@cwc.nus.edu.sg > > > Hi Aaron, > > thanks. > > so this is the second message on the list after the 55th > IETF TrigTrans BOF > meeting. > > Cheers, > Paul _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 07:22:53 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07140 for ; Wed, 27 Nov 2002 07:22:53 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARCP8921494 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 07:25:08 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARCP8v21491 for ; Wed, 27 Nov 2002 07:25:08 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA07125 for ; Wed, 27 Nov 2002 07:22:22 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARCP5v21483; Wed, 27 Nov 2002 07:25:05 -0500 Received: from albatross.wise.edt.ericsson.se (albatross-ext.wise.edt.ericsson.se [193.180.251.49]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARCOSv21452 for ; Wed, 27 Nov 2002 07:24:28 -0500 Received: from eed.ericsson.se (mailhost.eed.ericsson.se [164.48.133.33]) by albatross.wise.edt.ericsson.se (8.12.1/8.12.1/WIREfire-1.4) with ESMTP id gARCORKV012191 for ; Wed, 27 Nov 2002 13:24:27 +0100 (MET) Received: from res0010384da36.eed.ericsson.se (dhcp5-197 [164.48.135.197]) by eed.ericsson.se (8.8.8+Sun/1.1.mit) with ESMTP id NAA25666 for ; Wed, 27 Nov 2002 13:24:26 +0100 (MET) Message-Id: <5.1.0.14.0.20021127131254.028629b8@mailhost.eed.ericsson.se> X-Sender: eedrel@mailhost.eed.ericsson.se X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Wed, 27 Nov 2002 13:26:57 +0100 To: trigtran@www1.ietf.org From: Reiner Ludwig Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: [Trigtran] TCP & Link Outages Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , at the BOF in Atlanta we were talking about the link layer advice from draft-ietf-pilc-link-design-12.txt in case of link outages. This is the relevant text from section 8.2: It is therefore highly desirable that a subnetwork subject to outages not silently discard packets during an outage. Ideally, it should define an interface to the next higher layer (i.e., IP) that allows it to refuse packets during an outage, and to automatically ask IP for new packets when it is again able to deliver them. If it cannot do this, then the subnetwork should hold onto at least some of the packets it accepts during an outage and attempt to deliver them when the subnetwork comes back up. When packets are discarded, IP should be notified so that the appropriate ICMP messages can be sent. Note that it is *not* necessary to completely avoid dropping packets during an outage. The purpose of holding onto a packet during an outage, either in the subnetwork or at the IP layer, is so that its eventual delivery will implicitly notify TCP that the subnetwork is again operational. This is to enhance performance, not to ensure reliability -- a task, that as discussed earlier, can only be done properly on an end-to-end basis. If you would like to see real world measurements that show how nicely this works for TCP, you can check slide no. 12 from a talk I gave at the PILC WG meeting in San Diego. It's available here http://pilc.grc.nasa.gov/ and the link is titled "Flow Adaptive ARQ Proposal". Slide no. 13 shows how poorely TCP performs if this advice is not followed. Having said this, I'm not convinced that we need a "Link UP" trigger, nor a "Link DOWN trigger. ///Reiner _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 11:25:48 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17245 for ; Wed, 27 Nov 2002 11:25:48 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARGS5605182 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 11:28:05 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARGS5v05175 for ; Wed, 27 Nov 2002 11:28:05 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17238 for ; Wed, 27 Nov 2002 11:25:17 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARGS3v05165; Wed, 27 Nov 2002 11:28:03 -0500 Received: from MAIL.cynetanetworks.com (64-3-223-229.dia.xo.com [64.3.223.229] (may be forged)) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARGLEv04921 for ; Wed, 27 Nov 2002 11:21:14 -0500 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 10:21:05 -0600 Message-ID: <9255B6CD76A88943A3062F7D4E6432F510309B@mail.cynetanetworks.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWGLpvW3/E4V59Svy4QGo+nshyBwAFrtbw From: "Spencer Dawkins" To: Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARGLEv04922 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Reiner is raising a topic Sally mentioned in her talk at TRIGTRAN last week - do we need explicit triggers, or would implicit triggers be sufficient? Retransmitting a packet when an interface becomes active looks like an implicit trigger to me. I think the key question here is whether we think there are additional interesting triggers beyond link up/link down. If (no), the implicit trigger mechanism described in the PILC LINK document would be sufficient, if (yes), we're probably looking at an explicit trigger. Sally also raised a minor question - is the retransmitted packet useful, or just the fact that it was sent? I assume that the answer is "yes for TCP, no for most other packets", but am still thinking about this, and welcome other views. Spencer > -----Original Message----- > From: Reiner Ludwig [mailto:Reiner.Ludwig@eed.ericsson.se] > Sent: Wednesday, November 27, 2002 6:27 AM > To: trigtran@www1.ietf.org > Subject: [Trigtran] TCP & Link Outages > [ deleted down to ] > If you would like to see real world measurements that show > how nicely this > works for TCP, you can check slide no. 12 from a talk I gave > at the PILC WG > meeting in San Diego. It's available here > http://pilc.grc.nasa.gov/ and the > link is titled "Flow Adaptive ARQ Proposal". > > Slide no. 13 shows how poorely TCP performs if this advice is > not followed. > > Having said this, I'm not convinced that we need a "Link UP" > trigger, nor a > "Link DOWN trigger. _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 11:46:46 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18395 for ; Wed, 27 Nov 2002 11:46:46 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARGn3D07432 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 11:49:03 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARGn3v07429 for ; Wed, 27 Nov 2002 11:49:03 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18374 for ; Wed, 27 Nov 2002 11:46:15 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARGn1v07400; Wed, 27 Nov 2002 11:49:01 -0500 Received: from mail.eecis.udel.edu (louie.udel.edu [128.4.40.12]) by www1.ietf.org (8.11.6/8.11.6) with SMTP id gARGmHv07354 for ; Wed, 27 Nov 2002 11:48:17 -0500 Received: from ren.eecis.udel.edu by mail.eecis.udel.edu id aa18487; 27 Nov 2002 11:47 EST Date: Wed, 27 Nov 2002 11:47:33 -0500 (EST) From: "Armando L. Caro Jr." Reply-To: "Armando L. Caro Jr." To: Spencer Dawkins cc: trigtran@www1.ietf.org MMDF-Warning: Parse error in original version of preceding line at mail.eecis.udel.edu Subject: RE: [Trigtran] TCP & Link Outages In-Reply-To: <9255B6CD76A88943A3062F7D4E6432F510309B@mail.cynetanetworks.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , On Wed, 27 Nov 2002, Spencer Dawkins wrote: > Reiner is raising a topic Sally mentioned in her talk at TRIGTRAN last > week - do we need explicit triggers, or would implicit triggers be > sufficient? Retransmitting a packet when an interface becomes active > looks like an implicit trigger to me. > > I think the key question here is whether we think there are additional > interesting triggers beyond link up/link down. If (no), the implicit > trigger mechanism described in the PILC LINK document would be > sufficient, if (yes), we're probably looking at an explicit trigger. Since I haven't read all the documents and haven't followed all the work very closely, please feel free correct/stop me... My interest in triggers stems from the issue of doing AIMD congestion control on connections which involve at least one mobile host. I believe that when a host moves to a new domain, its TCP connections should slow start again. I am not sure yet how this behavior would fit into the planned framework, but I believe explicit triggers would be needed. ~armando 0-- --0 | Armando L. Caro Jr. | acaro@cis.udel.edu | | University of Delaware | www.cis.udel.edu/~acaro | 0-- --0 _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 12:12:24 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19681 for ; Wed, 27 Nov 2002 12:12:24 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARHEgQ09956 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 12:14:42 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHEfv09953 for ; Wed, 27 Nov 2002 12:14:41 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19605 for ; Wed, 27 Nov 2002 12:11:53 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHEcv09895; Wed, 27 Nov 2002 12:14:38 -0500 Received: from mgw-dax2.ext.nokia.com (mgw-dax2.ext.nokia.com [63.78.179.217]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHAAv09394 for ; Wed, 27 Nov 2002 12:10:10 -0500 Received: from davir04nok.americas.nokia.com (davir04nok.americas.nokia.com [172.18.242.87]) by mgw-dax2.ext.nokia.com (Switch-2.2.1/Switch-2.2.0) with ESMTP id gARHA6U26516 for ; Wed, 27 Nov 2002 11:10:06 -0600 (CST) Received: from daebh002.NOE.Nokia.com (unverified) by davir04nok.americas.nokia.com (Content Technologies SMTPRS 4.2.5) with ESMTP id ; Wed, 27 Nov 2002 11:09:57 -0600 Received: from daebe004.NOE.Nokia.com ([172.18.242.201]) by daebh002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.5329); Wed, 27 Nov 2002 11:09:53 -0600 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 11:09:52 -0600 Message-ID: <025E7DD4182874489CC2F61EE0FA19CEA87F@daebe004.americas.nokia.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWGLpvW3/E4V59Svy4QGo+nshyBwAFrtbwAAEje0A= From: To: , X-OriginalArrivalTime: 27 Nov 2002 17:09:53.0166 (UTC) FILETIME=[CD4EDAE0:01C29637] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARHAAv09396 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Hi, I guess there is a practical aspect of explicit triggers that needs mentioning. In many cases--depending upon where the TCP sender is located--explicit triggers (like ICMP or any other signaling protocol for that matter) will be blocked by firewalls. So even if we do have explicit triggers, there will always be a need to have implicit mechanisms built into the transport layer. But on the other hand, implicit triggers tend to be too broad and this certainly hurts the performance. So whether we should have an explicit trigger or an implicit triggers is not necessarily an either-or issue. --Yogesh > -----Original Message----- > From: ext Spencer Dawkins [mailto:sdawkins@cynetanetworks.com] > Sent: Wednesday, November 27, 2002 10:21 AM > To: trigtran@www1.ietf.org > Subject: RE: [Trigtran] TCP & Link Outages > > > Reiner is raising a topic Sally mentioned in her talk at > TRIGTRAN last week - do we need explicit triggers, or would > implicit triggers be sufficient? Retransmitting a packet when > an interface becomes active looks like an implicit trigger to me. > > I think the key question here is whether we think there are > additional interesting triggers beyond link up/link down. If > (no), the implicit trigger mechanism described in the PILC > LINK document would be sufficient, if (yes), we're probably > looking at an explicit trigger. > > Sally also raised a minor question - is the retransmitted > packet useful, or just the fact that it was sent? I assume > that the answer is "yes for TCP, no for most other packets", > but am still thinking about this, and welcome other views. > > Spencer > > > -----Original Message----- > > From: Reiner Ludwig [mailto:Reiner.Ludwig@eed.ericsson.se] > > Sent: Wednesday, November 27, 2002 6:27 AM > > To: trigtran@www1.ietf.org > > Subject: [Trigtran] TCP & Link Outages > > > > [ deleted down to ] > > > If you would like to see real world measurements that show > > how nicely this > > works for TCP, you can check slide no. 12 from a talk I gave > > at the PILC WG > > meeting in San Diego. It's available here > > http://pilc.grc.nasa.gov/ and the > > link is titled "Flow Adaptive ARQ Proposal". > > > > Slide no. 13 shows how poorely TCP performs if this advice is > > not followed. > > > > Having said this, I'm not convinced that we need a "Link UP" > > trigger, nor a > > "Link DOWN trigger. > _______________________________________________ > Trigtran mailing list > Trigtran@www1.ietf.org > https://www1.ietf.org/mailman/listinfo/trigtran > _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 12:42:47 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21227 for ; Wed, 27 Nov 2002 12:42:47 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARHj5H12423 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 12:45:05 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHj4v12420 for ; Wed, 27 Nov 2002 12:45:04 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21207 for ; Wed, 27 Nov 2002 12:42:16 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHj2v12409; Wed, 27 Nov 2002 12:45:02 -0500 Received: from courier.cs.helsinki.fi (courier.cs.helsinki.fi [128.214.9.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARHiYv12367 for ; Wed, 27 Nov 2002 12:44:34 -0500 Received: from melkki.cs.Helsinki.FI (melkki.cs.helsinki.fi [::ffff:128.214.48.122]) (IDENT: root, TLS: TLSv1/SSLv3,168bits,DES-CBC3-SHA) by courier.cs.helsinki.fi with esmtp; Wed, 27 Nov 2002 19:44:26 +0200 Received: from melkki.cs.Helsinki.FI (localhost [127.0.0.1]) by melkki.cs.Helsinki.FI (8.12.5/8.12.5) with ESMTP id gARHiOiZ024251; Wed, 27 Nov 2002 19:44:24 +0200 Received: from localhost (sarolaht@localhost) by melkki.cs.Helsinki.FI (8.12.5/8.12.5/Submit) with ESMTP id gARHiNM6024247; Wed, 27 Nov 2002 19:44:23 +0200 X-Authentication-Warning: melkki.cs.Helsinki.FI: sarolaht owned process doing -bs Date: Wed, 27 Nov 2002 19:44:23 +0200 (EET) From: Pasi Sarolahti To: "Armando L. Caro Jr." cc: Spencer Dawkins , trigtran@www1.ietf.org Subject: RE: [Trigtran] TCP & Link Outages In-Reply-To: Message-ID: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit On Wed, 27 Nov 2002, Armando L. Caro Jr. wrote: > My interest in triggers stems from the issue of doing AIMD congestion > control on connections which involve at least one mobile host. I believe > that when a host moves to a new domain, its TCP connections should slow > start again. I am not sure yet how this behavior would fit into the > planned framework, but I believe explicit triggers would be needed. Exactly. It could be as simple as a new ICMP "reinitialize congestion control" message type. If a firewall blocks it, that wouldn't be too fatal since the standard TCP/SCTP/DCCP congestion control rules would still be applied. - Pasi _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 14:25:49 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25169 for ; Wed, 27 Nov 2002 14:25:49 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARJS6318815 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 14:28:06 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJS6v18812 for ; Wed, 27 Nov 2002 14:28:06 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25159 for ; Wed, 27 Nov 2002 14:25:18 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJS3v18804; Wed, 27 Nov 2002 14:28:03 -0500 Received: from MAIL.cynetanetworks.com (64-3-223-229.dia.xo.com [64.3.223.229] (may be forged)) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJR5v18778 for ; Wed, 27 Nov 2002 14:27:06 -0500 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 13:26:56 -0600 Message-ID: <9255B6CD76A88943A3062F7D4E6432F543B949@mail.cynetanetworks.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWGLpvW3/E4V59Svy4QGo+nshyBwAFrtbwAAEje0AABaN8QA== From: "Spencer Dawkins" To: Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARJR6v18779 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Hi, Yogesh, Yes, exactly. We're not turning off any of the implicit mechanisms that exist today, the question is whether explicit triggers would be significantly useful, in addition. Definitely not either/or. Spencer > -----Original Message----- > From: Yogesh.Swami@nokia.com [mailto:Yogesh.Swami@nokia.com] > Sent: Wednesday, November 27, 2002 11:10 AM > To: Spencer Dawkins; trigtran@www1.ietf.org > Subject: RE: [Trigtran] TCP & Link Outages > > > Hi, > > I guess there is a practical aspect of explicit triggers that > needs mentioning. In many cases--depending upon where the TCP > sender is located--explicit triggers (like ICMP or any other > signaling protocol for that matter) will be blocked by firewalls. > So even if we do have explicit triggers, there will always be a > need to have implicit mechanisms built into the transport layer. > But on the other hand, implicit triggers tend to be too broad and > this certainly hurts the performance. > > So whether we should have an explicit trigger or an implicit > triggers is not necessarily an either-or issue. > > --Yogesh _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 14:50:48 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25858 for ; Wed, 27 Nov 2002 14:50:48 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARJr5o20403 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 14:53:05 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJr5v20396 for ; Wed, 27 Nov 2002 14:53:05 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25839 for ; Wed, 27 Nov 2002 14:50:17 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJr3v20383; Wed, 27 Nov 2002 14:53:03 -0500 Received: from mgw-dax2.ext.nokia.com (mgw-dax2.ext.nokia.com [63.78.179.217]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARJqTv20359 for ; Wed, 27 Nov 2002 14:52:29 -0500 Received: from davir02nok.americas.nokia.com (davir02nok.americas.nokia.com [172.18.242.85]) by mgw-dax2.ext.nokia.com (Switch-2.2.1/Switch-2.2.0) with ESMTP id gARJqPU22146 for ; Wed, 27 Nov 2002 13:52:25 -0600 (CST) Received: from daebh001.NOE.Nokia.com (unverified) by davir02nok.americas.nokia.com (Content Technologies SMTPRS 4.2.5) with ESMTP id ; Wed, 27 Nov 2002 13:52:24 -0600 Received: from daebe004.NOE.Nokia.com ([172.18.242.201]) by daebh001.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.5329); Wed, 27 Nov 2002 11:52:24 -0800 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 13:52:23 -0600 Message-ID: <025E7DD4182874489CC2F61EE0FA19CEA882@daebe004.americas.nokia.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWPdY6um/CUq+TS+ekxBJl84RczQADTXUQ From: To: , Cc: , X-OriginalArrivalTime: 27 Nov 2002 19:52:24.0214 (UTC) FILETIME=[8162B360:01C2964E] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARJqUv20360 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit > Exactly. It could be as simple as a new ICMP "reinitialize congestion > control" message type. If a firewall blocks it, that wouldn't > be too fatal > since the standard TCP/SCTP/DCCP congestion control rules > would still be > applied. > > - Pasi It will be interesting to see how this will work, given all the security issues involved. It's naive to think that a TCP sender can have a security association with a random router in the network ==> we cannot have any public/private key schemes. So the bulls other eye for security will be to use cookies. Again, its not possible to have a cryptographically safe cookie exchange without any security association between the "smart router" and the TCP sender. But even if there did exist such a mechanism, all these cookies will have to carried as either IP options (not my favorite) or as TCP options which is already too crowded. In addition all these cookies will have to be carried in each and every packet. The third choice is to use TCP sequence numbers as a loose form of cookie exchange. But anyone who can eavesdrop on any link between the TCP sender and TCP receiver can get all the needed information. So before we say we can solve congestion related problems using explicit signaling we should first have a complete solution for securing such signals. This discussion is meaningless, otherwise. --Yogesh _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 15:23:48 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA26782 for ; Wed, 27 Nov 2002 15:23:48 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARKQ6p22367 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 15:26:06 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARKQ6v22364 for ; Wed, 27 Nov 2002 15:26:06 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA26771 for ; Wed, 27 Nov 2002 15:23:17 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARKQ2v22355; Wed, 27 Nov 2002 15:26:02 -0500 Received: from MAIL.cynetanetworks.com (64-3-223-229.dia.xo.com [64.3.223.229] (may be forged)) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARKPov22329 for ; Wed, 27 Nov 2002 15:25:51 -0500 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 14:25:40 -0600 Message-ID: <9255B6CD76A88943A3062F7D4E6432F510309C@mail.cynetanetworks.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWPdY6um/CUq+TS+ekxBJl84RczQADTXUQAAEXM6A= From: "Spencer Dawkins" To: Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARKPpv22330 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Hi, Yogesh, Keeping in mind that the TRIGTRAN problem statement draft had more questions than answers about security ... It's probably worth keeping in mind that anyone who can dork with these explicit messages can dork with at least some of today's implicit triggers - if you're on the path to a receiver, just drop some packets, for instance, or forge ACKs guessing sequence numbers and overrun the bandwidth-delay product when the sender responds to them. But we do need to look at who trusts who, and why, in each scenario we look at. Spencer > -----Original Message----- > From: Yogesh.Swami@nokia.com [mailto:Yogesh.Swami@nokia.com] > Sent: Wednesday, November 27, 2002 1:52 PM > To: sarolaht@cs.Helsinki.FI; acaro@acm.org > Cc: Spencer Dawkins; trigtran@www1.ietf.org > Subject: RE: [Trigtran] TCP & Link Outages > > > > > Exactly. It could be as simple as a new ICMP "reinitialize > congestion > > control" message type. If a firewall blocks it, that wouldn't > > be too fatal > > since the standard TCP/SCTP/DCCP congestion control rules > > would still be > > applied. > > > > - Pasi > > It will be interesting to see how this will work, given all > the security > issues involved. [ deleted down to ] > > So before we say we can solve congestion related problems > using explicit > signaling we should first have a complete solution for securing such > signals. This discussion is meaningless, otherwise. > > --Yogesh > _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran From mailnull@www1.ietf.org Wed Nov 27 16:35:56 2002 Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA28661 for ; Wed, 27 Nov 2002 16:35:56 -0500 (EST) Received: (from mailnull@localhost) by www1.ietf.org (8.11.6/8.11.6) id gARLcFo26920 for trigtran-archive@odin.ietf.org; Wed, 27 Nov 2002 16:38:15 -0500 Received: from ietf.org (odin.ietf.org [132.151.1.176]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARLcEv26917 for ; Wed, 27 Nov 2002 16:38:14 -0500 Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA28575 for ; Wed, 27 Nov 2002 16:35:25 -0500 (EST) Received: from www1.ietf.org (localhost.localdomain [127.0.0.1]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARLcBv26904; Wed, 27 Nov 2002 16:38:11 -0500 Received: from mgw-dax1.ext.nokia.com (mgw-dax1.ext.nokia.com [63.78.179.216]) by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gARLbPv26784 for ; Wed, 27 Nov 2002 16:37:25 -0500 Received: from davir01nok.americas.nokia.com (davir01nok.americas.nokia.com [172.18.242.84]) by mgw-dax1.ext.nokia.com (Switch-2.2.1/Switch-2.2.0) with ESMTP id gARLbLE22545 for ; Wed, 27 Nov 2002 15:37:21 -0600 (CST) Received: from daebh002.NOE.Nokia.com (unverified) by davir01nok.americas.nokia.com (Content Technologies SMTPRS 4.2.5) with ESMTP id ; Wed, 27 Nov 2002 15:37:20 -0600 Received: from daebe004.NOE.Nokia.com ([172.18.242.201]) by daebh002.NOE.Nokia.com with Microsoft SMTPSVC(5.0.2195.5329); Wed, 27 Nov 2002 15:36:54 -0600 X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: [Trigtran] TCP & Link Outages Date: Wed, 27 Nov 2002 15:36:53 -0600 Message-ID: <025E7DD4182874489CC2F61EE0FA19CEA884@daebe004.americas.nokia.com> Thread-Topic: [Trigtran] TCP & Link Outages Thread-Index: AcKWPdY6um/CUq+TS+ekxBJl84RczQADTXUQAAEXM6AAAQhX4A== From: To: , X-OriginalArrivalTime: 27 Nov 2002 21:36:54.0435 (UTC) FILETIME=[1ABA7F30:01C2965D] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gARLbPv26785 Sender: trigtran-admin@www1.ietf.org Errors-To: trigtran-admin@www1.ietf.org X-BeenThere: trigtran@www1.ietf.org X-Mailman-Version: 2.0.12 Precedence: bulk List-Unsubscribe: , List-Id: Triggers for Transport List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 8bit Content-Transfer-Encoding: 8bit Hi, please see my comments below: > Hi, Yogesh, > > Keeping in mind that the TRIGTRAN problem statement draft had > more questions than answers about security ... > > It's probably worth keeping in mind that anyone who can dork > with these explicit messages can dork with at least some of > today's implicit triggers - if you're on the path to a > receiver, just drop some packets, for instance, or forge ACKs I was not saying that TCP as it is does not have any security holes, but with explicit triggers we will have even more holes than what we already have. So I doubt, for example, if any web sever manager will be willing to use a high performance TCP server with twice as many security holes than a low performance TCP server with fewer security holes. (BTW, as a side note, dropping a packet is far more difficult than forging an ACK etc...) I think TRIGTRANS will be useful if we can have better performance without having to make other compromises. But I don't know how much better the performance could be. For example (which was used by others), if there is a bit error loss every 1000 packets (with an average cwnd of 10), its probably okay to have the congestion window halved even though the loss was not due to congestion. Clearly there is a need to weigh these options and I personally don't know the answer. --yogesh _______________________________________________ Trigtran mailing list Trigtran@www1.ietf.org https://www1.ietf.org/mailman/listinfo/trigtran