IETF
tcpm@jabber.ietf.org
Wednesday, March 29, 2017< ^ >
jishac has set the subject to: TCP Maintenance and Minor Extensions WG
Room Configuration
Room Occupants

GMT+0
[12:24:51] ekaterina joins the room
[12:49:10] Samuel Jero joins the room
[13:01:49] Samuel Jero leaves the room
[13:02:07] sjero joins the room
[13:02:43] sjero leaves the room
[13:02:44] Samuel Jero joins the room
[13:04:52] ekaterina leaves the room
[13:26:13] ekaterina joins the room
[13:42:31] ekaterina leaves the room
[13:47:46] Meetecho joins the room
[13:47:54] Joseph joins the room
[13:53:27] craigt joins the room
[13:53:43] craigt yawns
[13:54:42] ekaterina joins the room
[13:54:52] Simon Romano joins the room
[13:55:13] Simon Pietro Romano joins the room
[13:55:13] Mark Allman joins the room
[13:55:33] Lorenzo Miniero joins the room
[13:55:36] ekaterina leaves the room
[13:55:44] ekaterina joins the room
[13:55:52] Simon Pietro Romano leaves the room
[13:56:27] Michael Tüxen joins the room
[13:57:07] tuexen joins the room
[13:58:26] tuexen leaves the room
[13:58:30] Joe Touch joins the room
[13:58:43] Yoshi Nishida joins the room
[13:59:28] <Michael Tüxen> Yes
[13:59:33] Samuel Jero_ joins the room
[13:59:35] <Mark Allman> yes
[14:00:10] yuki goto joins the room
[14:00:51] Bob Briscoe joins the room
[14:01:20] <craigt> https://datatracker.ietf.org/meeting/98/agenda/tcpm/
[14:01:48] mallman@xmpp.rg.net joins the room
[14:02:20] ekaterina leaves the room
[14:02:52] John Leslie joins the room
[14:05:12] ekaterina joins the room
[14:08:11] Okke Timm joins the room
[14:09:55] <Joseph> Wes briefing EDO - Requesting Implementations
[14:12:03] <Joe Touch> That is incorrect. Please read the draft cited. It SHOULD work with GRO - what we saw was a bug in GRO.
[14:18:58] Fernando Gont joins the room
[14:19:15] Iheanyi Obasi joins the room
[14:19:50] <Mark Allman> have a cup of coffee, joe :-)
[14:21:26] Padma Bhooma joins the room
[14:21:31] Roland Bless joins the room
[14:21:33] Fernando Gont leaves the room
[14:21:35] <Joe Touch> Huawei- please contact me. I'd be glad to hear more about the use case you are considering.
[14:21:37] Fernando Gont joins the room
[14:22:12] Fernando Gont leaves the room
[14:22:15] Fernando Gont joins the room
[14:22:27] frodek joins the room
[14:24:31] Fernando Gont leaves the room
[14:25:15] gryning joins the room
[14:27:35] <Joseph> Wes now speaking to 798bis
[14:29:02] <gryning> 6161 is elliptic curve
[14:29:11] gryning is now known as craigt
[14:29:59] <craigt> it's 6191
[14:32:41] Alan Ford joins the room
[14:34:29] Michael Tüxen leaves the room
[14:34:31] Michael Tüxen joins the room
[14:35:09] Michael Tüxen leaves the room
[14:35:11] Michael Tüxen joins the room
[14:38:11] craigt leaves the room: Disconnected: timeout during writing
[14:40:58] <craigt> "implementers may wants to consider... ...."
[14:41:42] Aaron Falk joins the room
[14:44:58] <Joseph> Yuchung Cheng now presenting RACK
[14:48:48] <Mark Allman> (i dont track this draft) but RACK is a 20 year old idea from paxson.  i think he said you just need to wait 10 or 20msec.  not something based on the RTT.  i am not sure what the intuition behind using an RTT-based time.  
[14:53:22] Joe Touch leaves the room
[14:53:37] Michael Tüxen leaves the room
[14:53:45] Michael Tüxen joins the room
[14:53:53] <Joseph> Last in a flight? Then why not just use existing mechanics like SACK and the RTO
[14:54:02] <Joseph> Basing this timer on RTT seems like you just have another RTO, the RTO is already adjusting based on RTT.
[14:55:10] <Mark Allman> because the rto is long
[14:57:16] Safiqul Islam joins the room
[15:07:06] Joe Touch joins the room
[15:13:16] <Meetecho> Bob Briscoe: we couldn't hear you in Meetecho either
[15:13:22] <Meetecho> have you checked your mic?
[15:14:22] Bob Briscoe leaves the room
[15:14:24] Bob Briscoe joins the room
[15:14:55] <Meetecho> Bob Briscoe: just FYI, when you're granted the floor, check if the volume levels below your name "dance"
[15:15:19] <Meetecho> if they don't, it means we couldn't get your mic for some reason, or it's muted in the mixer (OS setting perhaps)
[15:15:27] Joe Touch leaves the room
[15:16:01] Joe Touch joins the room
[15:16:06] <Bob Briscoe> It timed out just as I stuck my hand up (been regularly timing out the whole time). I have the session back now, so I'll interrupt after the next slide.
[15:16:20] <Meetecho> (y)
[15:25:24] Mirja Kühlewind joins the room
[15:27:45] ekaterina leaves the room
[15:29:58] <Joseph> Mark Allman presenting RTO Consider
[15:30:20] <Joseph> Mark you have a lot of distortion
[15:30:25] <Joseph> to much gain maybe
[15:30:40] Simon Pietro Romano joins the room
[15:30:52] <Samuel Jero_> Better
[15:31:01] <Joseph> Great
[15:31:29] <Meetecho> he's a bit loud in Meetecho as well, so it might be his mic: should be understandable though
[15:34:27] <Joseph> History part 2
[15:34:31] <Joseph> UDP
[15:36:54] ekaterina joins the room
[15:38:27] <Joseph> Take more than one is where you left off
[15:46:28] Okke Timm leaves the room
[15:46:46] ekaterina leaves the room
[15:47:07] ekaterina joins the room
[15:54:19] Padma Bhooma leaves the room
[15:54:22] <Joseph> There is already wiggle room all over this doc with all the SHOULDS and MAYS
[15:54:40] Padma Bhooma joins the room
[15:56:22] <Joseph> That sounds orthogonal to this document, I think Tim needs to make a doc
[15:56:26] Bob Briscoe leaves the room
[15:56:28] Bob Briscoe joins the room
[15:56:28] <Joseph> I agree
[16:02:26] <Joseph> Updating the UDP rfc because of the exponential backoff?
[16:02:35] <Mark Allman> that is my read of the room, too
[16:02:45] <Mark Allman> i dont think michael's read is right
[16:03:03] Aaron Falk leaves the room
[16:03:03] Alan Ford leaves the room
[16:03:08] Padma Bhooma leaves the room
[16:03:08] Joe Touch leaves the room
[16:03:08] John Leslie leaves the room
[16:03:29] Lorenzo Miniero leaves the room
[16:05:54] Padma Bhooma joins the room
[16:06:11] Lorenzo Miniero joins the room
[16:06:28] Alan Ford joins the room
[16:06:38] <Joseph> Regarding wiggle room for the exponential backoff:  The draft says "... such that the next firing requires a longer interval."  Which is wiggle room, it does not specifically say that it needs to be double the previous value.  It seems to imply something that is exponentially increasing only which can be any number of different approaches
[16:07:49] Alan Ford leaves the room
[16:07:52] Aaron Falk joins the room
[16:08:34] <mallman@xmpp.rg.net> i dont think your read is correct.  the draft says "the value of the RTO MUST be exponentially backed
        off"
[16:09:43] <mallman@xmpp.rg.net> that is, it does not say "double it".  but, it says something more strongly than "imply[ing] something that is exponentially increasing"
[16:10:07] Lorenzo Miniero leaves the room
[16:10:08] Joe Touch joins the room
[16:10:59] John Leslie joins the room
[16:11:35] Safiqul Islam leaves the room
[16:13:21] <Joseph> I guess I don't understand the difference in our statements, you're implying that "exponentially backed off" is stronger than "exponentially increasing".  So I'm happy to discuss that with you offline
[16:14:10] argc argv joins the room
[16:14:26] <Mark Allman> i am pushing back against this notion that the document IMPLIES exponential backoff.  it does not.  it says it.
[16:14:58] <Joseph> Ah OK
[16:14:58] <Joseph> Yes I agree
[16:15:08] <Joseph> It is a MUST and explicit
[16:28:51] Iheanyi Obasi leaves the room
[16:30:08] Simon Pietro Romano leaves the room
[16:31:44] Joe Touch leaves the room
[16:33:49] Mark Allman leaves the room
[16:34:44] ekaterina leaves the room
[16:35:05] Samuel Jero leaves the room
[16:35:22] Aaron Falk leaves the room
[16:35:23] Roland Bless leaves the room
[16:35:24] Michael Tüxen leaves the room
[16:35:24] argc argv leaves the room
[16:35:24] Mirja Kühlewind leaves the room
[16:35:24] Padma Bhooma leaves the room
[16:35:24] Samuel Jero_ leaves the room
[16:35:24] Yoshi Nishida leaves the room
[16:35:24] yuki goto leaves the room
[16:35:24] John Leslie leaves the room
[16:35:24] Bob Briscoe leaves the room
[16:36:31] Meetecho leaves the room
[16:36:58] Joseph leaves the room
[16:41:16] ekaterina joins the room
[16:47:20] ekaterina leaves the room
[16:47:24] ekaterina joins the room
[16:47:38] frodek leaves the room
[16:51:25] craigt leaves the room: Disconnected: No route to host
[16:56:28] ekaterina leaves the room
[17:34:03] ekaterina joins the room
[17:37:19] ekaterina leaves the room
[17:38:33] ekaterina joins the room
[17:42:17] ekaterina leaves the room
[17:44:34] gryning joins the room
[18:00:47] mallman@xmpp.rg.net leaves the room
[18:18:04] mallman@xmpp.rg.net joins the room
[18:20:38] ekaterina joins the room
[18:21:52] ekaterina leaves the room
[18:34:20] gryning leaves the room: Disconnected: No route to host
[18:40:38] mallman@xmpp.rg.net leaves the room
[18:41:47] mallman@xmpp.rg.net joins the room
[18:54:59] mallman@xmpp.rg.net leaves the room
[19:38:18] gryning joins the room
[19:51:16] ekaterina joins the room
[19:56:21] ekaterina leaves the room
[20:04:29] ekaterina joins the room
[20:10:57] ekaterina leaves the room
[20:21:07] gryning leaves the room: Disconnected: closed
[20:22:05] ekaterina joins the room
[20:53:15] frodek joins the room
[20:53:39] frodek leaves the room
[21:15:13] gryning joins the room
[21:38:00] gryning leaves the room: Disconnected: No route to host
[22:28:38] Simon Romano leaves the room
[23:12:58] ekaterina leaves the room
[23:26:59] ekaterina joins the room
Powered by ejabberd - robust, scalable and extensible XMPP server Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!