IETF
DNSSD
dnssd@jabber.ietf.org
Monday, April 4, 2016< ^ >
timc has set the subject to: DNSSD WG at IETF95
Room Configuration
Room Occupants

GMT+0
[00:39:36] timc leaves the room
[00:41:56] timc joins the room
[01:01:36] timc leaves the room
[01:45:40] timc joins the room
[02:27:36] timc leaves the room
[03:40:42] timc joins the room
[03:53:06] timc leaves the room
[03:55:42] timc joins the room
[04:08:08] timc leaves the room
[04:31:48] timc joins the room
[04:44:06] timc leaves the room
[05:26:06] timc joins the room
[05:38:06] timc leaves the room
[06:13:20] timc joins the room
[06:25:38] timc leaves the room
[07:03:28] timc joins the room
[07:16:38] timc leaves the room
[07:52:54] timc joins the room
[08:05:07] timc leaves the room
[08:13:12] timc joins the room
[08:25:38] timc leaves the room
[08:44:04] timc joins the room
[09:20:39] timc leaves the room
[09:21:53] timc joins the room
[09:34:07] timc leaves the room
[09:34:50] timc joins the room
[10:35:07] timc leaves the room
[10:47:50] timc joins the room
[11:09:37] timc leaves the room
[11:09:37] timc joins the room
[11:17:07] timc leaves the room
[12:14:07] timc joins the room
[12:16:07] timc leaves the room
[12:16:10] timc joins the room
[12:17:07] timc leaves the room
[12:23:23] timc joins the room
[12:30:37] timc leaves the room
[12:32:56] timc joins the room
[12:46:37] timc leaves the room
[12:57:14] timc joins the room
[13:03:39] timc leaves the room
[13:59:45] timc joins the room
[13:59:54] timc leaves the room
[18:40:04] Meetecho joins the room
[18:45:31] each joins the room
[18:45:44] Timothy Morizot joins the room
[18:47:04] jacques latour joins the room
[18:47:12] Wentao Shang joins the room
[18:47:13] Jorge Cano joins the room
[18:47:16] Fearghas McKay joins the room
[18:48:09] Hosnieh Rafiee joins the room
[18:49:19] Michael Breuer joins the room
[18:49:28] Joseph Yee joins the room
[18:49:30] <Hosnieh Rafiee> :)
[18:50:52] Sara Dickinson joins the room
[18:51:24] doug o joins the room
[18:52:04] Kerry Lynn joins the room
[18:52:17] <Hosnieh Rafiee> it is too loud...
[18:52:17] Suzanne (scribe) joins the room
[18:52:40] <Suzanne (scribe)> Meeting is starting
[18:52:48] <Suzanne (scribe)> Note well the note well
[18:53:07] KY Hong joins the room
[18:53:13] Dave Thaler joins the room
[18:54:29] Andrew McConachie joins the room
[18:54:40] Olafur joins the room
[18:54:50] <Suzanne (scribe)> document statuses (statusi?)
[18:55:16] <Suzanne (scribe)> p. 6 of chairs slides (URL in a sec)
[18:55:57] Andrew Sullivan joins the room
[18:56:15] Timothy Morizot leaves the room
[18:56:22] mellon joins the room
[18:56:35] <mellon> I wasn't able to get to etherpad.
[18:56:35] Fearghas McKay leaves the room
[18:56:37] <Suzanne (scribe)> https://www.ietf.org/proceedings/95/slides/slides-95-dnssd-3.pdf
[18:57:07] <Suzanne (scribe)> Staurt Cheshire presenting summary of comments on hybrid proxy draft
[18:57:07] <mellon> Here is a google doc in which I am taking notes:
https://docs.google.com/document/d/1EvenkFSIQ7pOvlyTHiE2Y9GMfRqplM4ebRT9Sofv4rQ/edit?usp=sharing
[18:58:26] Pierre Peloso joins the room
[18:58:53] Shinta Sato joins the room
[18:58:56] <Suzanne (scribe)> (for remote: sorry if there seems to be a lot of background noise in the audio; there is a lot of sound bleeding between rooms in this venue)
[18:59:08] <Suzanne (scribe)> slide 4
[18:59:15] <Suzanne (scribe)> slide 5
[18:59:35] DanYork joins the room
[19:00:32] <Suzanne (scribe)> slide 6
[19:01:07] Fearghas McKay joins the room
[19:01:46] john dickinson joins the room
[19:03:01] <Suzanne (scribe)> slide 7
[19:04:03] <Suzanne (scribe)> for remote: if you have a question or comment, please put the string MIC in at the beginning and I'll go to the mic for you.
[19:05:16] <Suzanne (scribe)> Ted Lemon: wasn't suggesting clients should assume TTL=0; if you're using DNS not mDNS, you can send 0 because you don't care as much about repeating query— chattiness less of an issue than in some environments
[19:05:32] <Suzanne (scribe)> (did I get that Ted?)
[19:05:44] liman joins the room
[19:06:45] Sara Dickinson leaves the room
[19:06:48] <Suzanne (scribe)> Ted Lemon: on how to resolve .local names outside of the local link: dnssd is looking at DNS scope for service discovery beyond a single link
[19:08:04] <Suzanne (scribe)> slide 8
[19:09:00] Wentao Shang leaves the room
[19:11:53] <Suzanne (scribe)> Ted Lemon at the mic
[19:12:01] Sara Dickinson joins the room
[19:13:01] <Suzanne (scribe)> Ted: prior conversation conclusion: if you have a service publishing in DNS using DNS updates, "Link Merging" may not be too hard. Could also just "notice DNS announcements" and monitor for merges needed.
[19:13:41] <Suzanne (scribe)> Dave Thaler at the mic: may also be that zones you're using to proxy shouldn't be the same one you're updating
[19:14:26] <Suzanne (scribe)> Tim: thinks the draft is adequate on this point
[19:16:46] <Suzanne (scribe)> Ted Lemon: beig realistic about reacting to the devices we actually have, not the ones we wish we had
[19:16:50] Fearghas McKay leaves the room
[19:17:10] john dickinson leaves the room
[19:17:30] rdroms joins the room
[19:18:43] <Suzanne (scribe)> Dave Thaler: you can have the same first label in different zones; question is about terminating label; .local as opposed to .building1
[19:19:10] <jacques latour> is it me or the screen is blinking? morse code?
[19:19:28] <Kerry Lynn> Probably time to take this debate offline...
[19:19:34] <Suzanne (scribe)> Ted Lemon: no value at all to this; having multiple zones where the same name might appear is a disaster waiting to happen
[19:19:44] Jorge Cano leaves the room
[19:20:01] Joseph Yee leaves the room
[19:20:41] <Suzanne (scribe)> Dave Thaler: auth server for local zone, updates should go to that server. Pas de deux Ted, Dave, and a new doc that needs to be written.
[19:20:58] <Suzanne (scribe)> slide 9
[19:21:53] Shinta Sato leaves the room
[19:22:35] Shinta Sato joins the room
[19:22:47] <Suzanne (scribe)> slide 10
[19:22:51] Andrew McConachie leaves the room
[19:23:35] Lorenzo Miniero joins the room
[19:24:12] <Suzanne (scribe)> Dave Thaler at the mic: prnciple of least surprise suggests don't just drop those punycode queries
[19:24:32] <Andrew Sullivan> FWIW, I am deeply uneasy about that idea
[19:24:51] <Suzanne (scribe)> End of the slide deck about draft-ietf-dnssd-hybrid
[19:25:19] <Andrew Sullivan> This very issue was part of what the label interop thing was supposed to try to make better, and having intermediate systems doing IDNA mapping sounds to me like a way to create problems
[19:25:34] <Dave Thaler> @Andrew: consider it as if it's stored in the punycode form in the zone, but an alternative update protocol is used that registers it passing UTF-8 in the update and it gets convierted before being stored.   The result would be the same.
[19:25:35] <Suzanne (scribe)> Ralph Droms at the mic, no hat: impact of link-scoped addresses to be discussed?
[19:25:43] <Andrew Sullivan> I won't say, "Bad idea," yet, because I haven't thought it through completely
[19:25:46] <Dave Thaler> you may be equally uneasy about that too though :)
[19:25:51] <Andrew Sullivan> but I think it has ugly potential
[19:26:09] <Andrew Sullivan> Ugh.  Yes, I might be
[19:26:58] <Dave Thaler> the alternative is that every client has to know for each zone (which may not be in its own network) whether to use UTF-8 or Punycode, and that I don't think is practical.
[19:28:28] <Suzanne (scribe)> Ted Lemon at the mic: on ULAs and homenet, doesn't want ULAs for all the hosts— you want the hybrid proxy on the homenet to answer only ULAs
[19:28:41] <Andrew Sullivan> No, I think what we concluded with label-interop was that you query both, and maybe the client has a theory about when labels are likely to be A-labels and when they're likely to be UTF-8
[19:28:54] <Dave Thaler> ah, true
[19:29:18] <Andrew Sullivan> Put the smarts at the edge, I think, not in the intermediate networks
[19:29:52] <Suzanne (scribe)> Tim, Stuart, and Ted are finding more documents they need to write— this one for the homenet configuration/administration of the proxy
[19:30:01] <Dave Thaler> ok yes I agree that approach would be consistent with your doc, and I think you're up next so feel free to say so in your talk :)
[19:30:39] <Suzanne (scribe)> Tim: next step: definitely needs a -04; Ralph and Tim will discuss with Terry whether to re-do the WGLC or move it ahead
[19:31:19] <Suzanne (scribe)> Tom Pusateri presenting
[19:31:20] <Suzanne (scribe)> https://www.ietf.org/proceedings/95/slides/slides-95-dnssd-5.pdf
[19:31:25] <Suzanne (scribe)> slide 3
[19:32:16] Timothy Morizot joins the room
[19:32:44] Alexander Ilin joins the room
[19:34:44] <Suzanne (scribe)> slide 6
[19:35:03] <Suzanne (scribe)> Ted Lemon at the mic: best to avoid special cases where possible
[19:35:14] <Suzanne (scribe)> slide 7
[19:36:57] <Suzanne (scribe)> slide 8
[19:37:20] Alexander Ilin leaves the room
[19:37:34] <Suzanne (scribe)> STuart at the mic: can we have an opinion on these options for keepalive
[19:37:43] Alexander Ilin joins the room
[19:37:57] <Suzanne (scribe)> Ted Lemon at the mic: TCP keepalives not that viable; empty DNS message may be simplest/cheapest way todo it
[19:38:22] <Suzanne (scribe)> Bernie Volz: maybe consider something bidirectional
[19:40:04] <Suzanne (scribe)> Andrew Sullivan at the mic: not sure we want to reinvent the Keepalive spec; this really needs discussion on the list, could be some sharp edges
[19:40:48] <Suzanne (scribe)> Stuart at the mic: what exactly is in the DNSOP doc? the current one doesn't discuss keepalive
[19:41:22] <Suzanne (scribe)> Andrew at the mic: trying to avoid state in the server; negotiate connection and warn when it's about to be torn down instead
[19:42:08] <Suzanne (scribe)> Andrew: really nervous about the empty query idea, would want very much to have all the relevant standards docs available for the analysis
[19:42:26] <Suzanne (scribe)> SHane Kerr at the mic: what's the motivation for insisting on keeping the connection open?
[19:42:46] <Suzanne (scribe)> Stuart: entire reason for this mechanism is to avoid polling; duplicate multicast behavior beyond the local link
[19:43:26] Sara Dickinson leaves the room
[19:43:59] <Suzanne (scribe)> Shane: but is keepalive just a cheper way to poll?
[19:44:13] <Suzanne (scribe)> Dave Thaler at the mic: no, it's not bound to polling intervals
[19:44:43] <Suzanne (scribe)> (Shane. Stuart, and Dave at the mic: rock stars!)
[19:45:36] <Suzanne (scribe)> Dave Thaler: has an opinion on TCP vs. DNS: strongly prefers DNS layer, doesn't care on details except don't invent a new message
[19:46:57] <Suzanne (scribe)> Wes Hardaker at the mic: normal queries will work just as well as keepalives
[19:46:57] Olafur leaves the room
[19:47:23] <Andrew Sullivan> The thing I'm worried about is that a message with QR=0 and QRCOUNT=0 is likely to look like something bogus and to be filtered in some places.
[19:48:02] Olafur joins the room
[19:48:37] Sara Dickinson joins the room
[19:48:42] <Andrew Sullivan> Ted's point about telling the client that it has to support it is well-advised
[19:49:28] <Suzanne (scribe)> slide 9
[19:50:08] Carl Mehner joins the room
[19:50:21] <Andrew Sullivan> I believe the client is already supposed to close the connection conventionally.  I didn't understand (in the draft) why this is a problem, but I admit I didn't read carefully enough
[19:50:37] <Suzanne (scribe)> Bernie Volz at the mic: consider how this problem is solved in web space, but sending lots of RSTs is often to be avoided
[19:50:55] <Suzanne (scribe)> slide 10
[19:52:21] Sara Dickinson leaves the room
[19:52:24] <Suzanne (scribe)> End of the deck, now on next steps
[19:52:48] Sara Dickinson joins the room
[19:53:21] <Suzanne (scribe)> Ralph & Tim will send summary to the list
[19:53:50] <Suzanne (scribe)> Andrew Sullivan to the front, no slides
[19:54:25] <Suzanne (scribe)> on https://tools.ietf.org/html/draft-ietf-dnssd-mdns-dns-interop-02
[19:55:05] <Suzanne (scribe)> status of review, etc….haven't heard much
[19:55:11] <Suzanne (scribe)> Ted Lemon to the mic: draft looks good
[19:55:59] <Suzanne (scribe)> Dave Thaler to the mic: refer to jabber room discussion supporting the draft re: punycode/UTF-8
[19:56:10] <Suzanne (scribe)> Dave Thaler: needs some clarification of the problem
[19:58:14] <Suzanne (scribe)> next steps: minor rev needed; any other comments welcome
[19:58:41] <Suzanne (scribe)> Christian on https://tools.ietf.org/html/draft-huitema-dnssd-privacy-00
[19:58:50] KY Hong leaves the room
[19:59:12] <Suzanne (scribe)> https://www.ietf.org/proceedings/95/slides/slides-95-dnssd-0.pdf
[20:01:07] <Suzanne (scribe)> slide 3
[20:06:29] <Suzanne (scribe)> slide 6
[20:07:40] <Suzanne (scribe)> Tim: it's good to see this work as further followup to perpass; any comments?
[20:08:36] <Suzanne (scribe)> Ted Lemon at the mic: do I even want to publish my laptop's name o na strange network? is this just so someone with a laptop on a public net who wants to keep privacy but also be ableto use the printer in the back of the coffeeshop
[20:09:08] <doug o> Special Use domains can  also provide a level of privacy.
[20:09:46] <Suzanne (scribe)> Stuart Cheshire at the mic: this is useful important work
[20:09:48] Shinta Sato leaves the room
[20:10:06] <Suzanne (scribe)> room for guidance that spme mobile devices don't need to advertise services at all
[20:10:14] Shinta Sato joins the room
[20:10:56] Sara Dickinson leaves the room
[20:10:58] doug o leaves the room
[20:11:08] Sara Dickinson joins the room
[20:11:27] doug o joins the room
[20:11:27] <Suzanne (scribe)> Tim: this isn't exactly on charter, but it does seem like good work to do; will speak with Terry
[20:11:40] <Suzanne (scribe)> Dan York at the mic: what thought has been given to keydist?
[20:13:35] <Suzanne (scribe)> Tom Pusateri at the mic: joining a wireless network; choosing to join is a trust decision
[20:14:37] liman leaves the room
[20:15:12] <Suzanne (scribe)> wrapup!
[20:15:12] Olafur leaves the room
[20:15:21] <Suzanne (scribe)> hybrid proxy: Stuart will do 04
[20:15:37] <Suzanne (scribe)> push: needs some work before WGLC
[20:15:55] <Suzanne (scribe)> label-interop: Andrew and Dave will produce a new rev, probably minor changes
[20:16:10] <Suzanne (scribe)> prviacy work: please discuss on the list; this seems like good work to carry forward
[20:16:54] <Suzanne (scribe)> will be talking about how to take the group forward after hybrid-proxy and push documents; there will be discussion with the AD & the group about other documents that might come up regarding how to use those buildingblcks
[20:16:57] Sara Dickinson leaves the room
[20:18:03] <Suzanne (scribe)> need to continue on discussion of DNS vs TCP keepalive
[20:18:08] <Suzanne (scribe)> need to decide what to do with threats doc
[20:18:16] DanYork leaves the room
[20:18:22] jacques latour leaves the room
[20:18:38] Suzanne (scribe) leaves the room
[20:18:59] Andrew Sullivan leaves the room
[20:19:08] rdroms leaves the room
[20:19:18] mellon leaves the room
[20:19:32] Lorenzo Miniero leaves the room
[20:19:33] Alexander Ilin leaves the room
[20:19:33] Pierre Peloso leaves the room
[20:19:33] Timothy Morizot leaves the room
[20:19:33] Michael Breuer leaves the room
[20:19:34] Carl Mehner leaves the room
[20:19:34] Kerry Lynn leaves the room
[20:19:48] each leaves the room
[20:20:12] Hosnieh Rafiee leaves the room
[20:20:37] doug o leaves the room
[20:24:06] Andrew Sullivan joins the room
[20:29:53] Dave Thaler joins the room
[20:30:08] Dave Thaler leaves the room
[20:33:04] Meetecho leaves the room
[20:36:54] Olafur joins the room
[20:37:05] Olafur leaves the room
[20:38:31] Suzanne (scribe) joins the room
[20:39:25] Andrew Sullivan leaves the room
[20:40:48] Suzanne (scribe) leaves the room
[20:44:32] DanYork joins the room
[20:44:38] DanYork leaves the room
[20:45:04] Andrew Sullivan joins the room
[20:45:14] Andrew Sullivan leaves the room
[20:45:45] rdroms joins the room
[21:17:08] each joins the room
[21:17:13] each leaves the room
[21:28:08] rdroms leaves the room
[21:35:24] rdroms joins the room
[21:52:38] rdroms leaves the room
[21:52:46] rdroms joins the room
[22:01:38] rdroms leaves the room
[22:01:41] rdroms joins the room
[22:03:08] rdroms leaves the room
[22:08:33] rdroms joins the room
[22:13:08] rdroms leaves the room
[22:13:14] rdroms joins the room
[22:55:38] rdroms leaves the room
[22:57:26] Shinta Sato leaves the room
[23:00:08] Dave Thaler leaves the room
Powered by ejabberd - robust, scalable and extensible XMPP server Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!