IETF
rtgwg
rtgwg@jabber.ietf.org
Wednesday, November 12, 2014< ^ >
Room Configuration
Room Occupants

GMT+0
[18:58:30] dhruvdhody joins the room
[18:59:19] Eric Rosen joins the room
[19:00:17] Meetecho joins the room
[19:01:01] <dhruvdhody> Routing Area WG (rtgwg)
IETF 91 (Hawaii) Agenda
CHAIR(s):  Alvaro Retana <aretana@cisco.com>
Jeff Tantsura <jeff.tantsura@ericsson.com>
lets get started...
[19:01:29] <dhruvdhody> Administrivia...
[19:01:47] equinox joins the room
[19:03:03] <dhruvdhody> Chair’s Slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-7.pptx
[19:03:29] Michael Baer joins the room
[19:04:16] equinox is now known as David Lamparter
[19:05:27] Dhruv  Dhody joins the room
[19:09:31] sunseawq joins the room
[19:17:01] Sriganesh Kini joins the room
[19:18:59] Paolo Saviano joins the room
[19:19:39] A A joins the room
[19:20:11] Dhruv  Dhody leaves the room
[19:20:30] Dhruv Dhody joins the room
[19:20:58] Simon Romano joins the room
[19:21:17] Lorenzo Miniero joins the room
[19:22:53] Paolo Saviano leaves the room
[19:23:12] Lorenzo Miniero leaves the room
[19:23:34] A A leaves the room
[19:23:48] alex amirante joins the room
[19:24:02] adrianfarrel joins the room
[19:27:09] alex amirante leaves the room
[19:27:29] <dhruvdhody> John presenting: Use of BGP for routing in large-scale data centers    
https://tools.ietf.org/html/draft-ietf-rtgwg-bgp-routing-large-dc
[19:28:08] <dhruvdhody> Slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-2.pdf (incase you are not in meetecho)
[19:28:59] Alvaro Retana joins the room
[19:29:04] metricamerica joins the room
[19:29:22] metricamerica leaves the room
[19:32:45] Jeffrey Haas joins the room
[19:38:38] Christian  OFlaherty joins the room
[19:38:53] <adrianfarrel> If you are still on draft-ietf-rtgwg-bgp-routing-large-dc can someone ask at the mic how they think this work relates to the work done in BESS
[19:39:19] <adrianfarrel> (I am not on audio)
[19:39:28] <Dhruv Dhody> will do..
[19:40:16] <Dhruv Dhody> does that satisfy you :P
[19:42:52] <dhruvdhody> That was lou
[19:43:58] Martin Bjorklund joins the room
[19:45:50] <dhruvdhody> Sam: can u talk about multi-tenancy
[19:46:03] <dhruvdhody> John: using IP overlay of some sort
[19:46:06] dhruvdhody leaves the room
[19:46:22] dhruvdhody joins the room
[19:46:33] <dhruvdhody> acee:  what policy will you apply to outside of DC
[19:46:41] <dhruvdhody> John: no policy in our deployments
[19:47:54] <dhruvdhody> Alia: I talked to GROW/IDR; Sue had comments -07 please make sure you handle them
[19:48:19] <dhruvdhody> Jeff speaking..
[19:50:26] <dhruvdhody> Jeff: dis you consider sub 100 ms convergence?
[19:51:09] akatlas joins the room
[19:51:16] <dhruvdhody> Jeff: expand on failure analysis, guide operator who starting now
[19:51:45] <dhruvdhody> Alvero: aim to move from ad-sponser so that RTGWG help and provide inputs
[19:52:19] <dhruvdhody> Moving on......... Extended procedures and considerations for evaluating Loop-Free Alternates
http://tools.ietf.org/html/draft-chunduri-rtgwg-lfa-extended-procedures
By Uma Chunduri
[19:52:41] <dhruvdhody> Slides at http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-3.ppt
[19:52:51] <adrianfarrel> Hmmm, was there an answer to my question?
[19:53:38] Martin Bjorklund leaves the room
[19:54:02] Martin Bjorklund joins the room
[19:54:30] <Alvaro Retana> Adrian:  The answer to your question was "no relaitonship".  they are not using BGP to signal the virtualization in the network.
[19:55:06] <adrianfarrel> thanks
[20:00:22] <Jeffrey Haas> Has the rtg-cfg model been presented yet?
[20:00:28] <dhruvdhody> no
[20:00:36] <Jeffrey Haas> This is a *very* quiet jabber channel. :-)
[20:00:38] <Jeffrey Haas> Thanks.
[20:00:46] <dhruvdhody> another 15 mins
[20:01:42] <dhruvdhody> Acee on mic
[20:02:26] <dhruvdhody> Stewert on mic
[20:03:16] akatlas joins the room
[20:03:47] <dhruvdhody> Alvero: we need more ppl to read this draft
[20:03:51] akatlas leaves the room
[20:04:15] <dhruvdhody> Next....... LFA selection for Multi-Homed Prefixes
http://tools.ietf.org/html/draft-psarkar-rtgwg-multihomed-prefix-lfa
by Pushpasis Sarkar
[20:04:31] <dhruvdhody> Slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-5.pptx
[20:05:00] Martin Bjorklund leaves the room
[20:05:27] Martin Bjorklund joins the room
[20:06:08] Jeffrey Haas leaves the room
[20:08:42] A A joins the room
[20:09:07] Jeffrey Haas joins the room
[20:14:27] <dhruvdhody> That is Uma
[20:15:03] <dhruvdhody> Acee: since this is local repair, why do you care
[20:15:16] <dhruvdhody> for backward compatibility
[20:17:35] adrianfarrel leaves the room
[20:17:47] <dhruvdhody> Alia: it might be useful for this clarification as from implementation are coming, there is some improvement to the base RFC
[20:18:21] <dhruvdhody> Alia: (a) do a bis / (b) have informational draft with clarifications
[20:18:45] <dhruvdhody> Alvero: any thoughts?
[20:18:54] <dhruvdhody> discuss on the list
[20:19:17] <dhruvdhody> *************
[20:19:29] <dhruvdhody> A YANG Data Model for Routing Management
https://tools.ietf.org/html/draft-ietf-netmod-routing-cfg
By Lada
[20:19:44] <dhruvdhody> Slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-0.pdf
[20:20:03] Martin Bjorklund leaves the room
[20:20:06] FM EC joins the room
[20:20:26] Martin Bjorklund joins the room
[20:25:31] Simon Romano leaves the room
[20:25:45] Albert Tian joins the room
[20:25:57] Martin Bjorklund leaves the room
[20:26:04] <dhruvdhody> Rob on mic
[20:26:20] Martin Bjorklund joins the room
[20:26:25] <dhruvdhody> Rob: routing-instance is a VRF?
[20:27:53] FM EC leaves the room
[20:27:56] <dhruvdhody> lada: Not just VRF, that is why we have type, you can also use assign new rules per type
[20:33:42] Jeffrey Haas leaves the room
[20:35:21] <dhruvdhody> Rob: different vendor models is an issue for an operator, this would be a failure root.
[20:35:27] <dhruvdhody> Acee: that exist already
[20:35:47] <dhruvdhody> Start with standard, and each vendor modify the base
[20:36:27] <dhruvdhody> logical router is much more than routing, there is security, management etc ... this is just routing model not the whole system.
[20:36:57] <dhruvdhody> Jeff: lot of issues are taken care, instance issue is the biggest bottleneck
[20:37:21] <dhruvdhody> we need anchor to identify RIB based on context (VRF, process, chasis)
[20:40:44] <dhruvdhody> acee on mic
[20:41:16] <dhruvdhody> Acee: issue with route-filter, get the route-policy out of this ietf-routing
[20:42:40] zhutao cheng joins the room
[20:42:42] adrianfarrel joins the room
[20:43:40] zhutao cheng leaves the room
[20:44:04] <dhruvdhody> Acee: address and interface are disjoint in the model
[20:45:20] Clyde Wildes joins the room
[20:45:20] <dhruvdhody> alia on mic
[20:46:26] <dhruvdhody> where do we put route-filters if not here or acl.
[20:46:38] zhutao cheng joins the room
[20:46:50] <dhruvdhody> Acee: It should be in a different routing-policy model
[20:47:13] Stewart Bryant joins the room
[20:47:35] Evangelos Haleplidis joins the room
[20:47:50] <dhruvdhody> acee on mic
[20:47:53] zhutao cheng leaves the room
[20:48:16] <dhruvdhody> acee: putting stubs is useful; have multi-vendors team is important
[20:50:17] Stewart Bryant leaves the room
[20:50:30] zhutao cheng joins the room
[20:51:09] <dhruvdhody> Dan on mic
[20:51:56] Naiming Shen joins the room
[20:52:35] Naiming Shen leaves the room
[20:53:11] Naiming Shen joins the room
[20:54:56] <dhruvdhody> .............
[20:55:01] <dhruvdhody> Destination/Source Routing
http://tools.ietf.org/html/draft-lamparter-rtgwg-routing-extra-qualifiers
http://tools.ietf.org/html/draft-lamparter-rtgwg-dst-src-routing
David Lamparter
[20:55:31] <dhruvdhody> slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-4.pdf
[20:56:02] Naiming Shen leaves the room
[20:56:26] Martin Bjorklund leaves the room
[20:57:18] Naiming Shen joins the room
[20:59:12] Naiming Shen leaves the room
[20:59:14] Naiming Shen joins the room
[20:59:26] Naiming Shen leaves the room
[21:00:04] Naiming Shen joins the room
[21:01:04] <dhruvdhody> Fred: will multicast not be related to source-specefic multicast
[21:02:19] <dhruvdhody> Doug: Can we do strict RPF check at egress points?
[21:04:18] <dhruvdhody> Adrian: easier to facilitate IGP edges to allow some address beyond BCP 38
[21:04:40] <dhruvdhody> harder to deploy in legacy network
[21:06:09] <dhruvdhody> Fred/Alvero: This is not specefic to homenet, esp for campus network
[21:06:11] zhutao cheng leaves the room
[21:07:00] <dhruvdhody> Rob: scaling of edge box because of diff rules would be difficult
[21:07:39] Clyde Wildes leaves the room
[21:07:58] zhutao cheng joins the room
[21:08:20] <dhruvdhody> stewert: go for simplest solution, pay ISP to do this, think about operating a more complex routing protocol
[21:09:03] Alvaro Retana leaves the room
[21:09:11] Clyde Wildes joins the room
[21:09:24] zhutao cheng leaves the room
[21:09:26] <dhruvdhody> Rob: is there a problem? or is *this* is solution? there are multiple issues to be looked into
[21:09:57] Jeffrey Haas joins the room
[21:10:10] <dhruvdhody> Fred: what are the complexity involved here? operational issues, convergence, operations, inter-operability...
[21:10:54] <dhruvdhody> Its just extra TLV, not necessarily more complex, no change in computation
[21:11:10] zhutao cheng joins the room
[21:11:34] zhutao cheng leaves the room
[21:11:53] <dhruvdhody> humm. for use case
[21:12:00] <dhruvdhody> silence for not a problem
[21:12:31] <dhruvdhody> .....
[21:12:33] <dhruvdhody> A Framework for Secure Routing Protocols
https://tools.ietf.org/html/draft-atwood-rtgwg-secure-rtg
Bill Atwood
[21:12:39] zhutao cheng joins the room
[21:13:04] <dhruvdhody> slides: http://www.ietf.org/proceedings/91/slides/slides-91-rtgwg-8.pptx
[21:13:54] zhutao cheng leaves the room
[21:13:58] Jeffrey Haas leaves the room
[21:14:13] Alvaro Retana joins the room
[21:16:31] zhutao cheng joins the room
[21:17:49] zhutao cheng leaves the room
[21:23:19] <dhruvdhody> acee: yang model for keychain, is all we need and there is work going on there
[21:25:56] Naiming Shen leaves the room
[21:29:00] <dhruvdhody> Adrian: Draft name please, make sure slide name matches with data tracker
[21:30:11] <adrianfarrel> Answer was http://www.ietf.org/id/draft-atwood-rtgwg-secure-rtg-00.txt
[21:30:21] <dhruvdhody> Dutchman on mic :)
[21:32:05] adrianfarrel leaves the room
[21:32:29] <dhruvdhody> Uma on mic
[21:32:47] Albert Tian leaves the room
[21:33:25] <dhruvdhody> ---End of Meeting----
[21:33:28] Eric Rosen leaves the room
[21:33:31] <dhruvdhody> Till Dallas!
[21:33:35] dhruvdhody leaves the room
[21:33:38] Dhruv Dhody leaves the room
[21:34:17] Meetecho leaves the room
[21:34:19] Alvaro Retana leaves the room
[21:34:39] Christian  OFlaherty leaves the room
[21:34:55] Clyde Wildes leaves the room
[21:36:47] Evangelos Haleplidis leaves the room
[21:39:55] akatlas joins the room
[21:41:03] akatlas leaves the room
[21:41:08] adrianfarrel joins the room
[21:41:31] akatlas leaves the room
[21:42:57] Michael Baer leaves the room
[21:45:35] adrianfarrel leaves the room
[21:50:45] sunseawq leaves the room
[21:54:35] David Lamparter leaves the room
[22:20:37] dhruvdhody joins the room
[22:42:19] dhruvdhody leaves the room
[22:53:02] Stewart Bryant joins the room
[23:06:29] adrianfarrel joins the room
[23:15:03] Jeffrey Haas joins the room
[23:19:14] Jeffrey Haas leaves the room
[23:19:44] adrianfarrel leaves the room
[23:35:43] akatlas joins the room
[23:35:45] akatlas leaves the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!