#opendaylight-nic: nic_weekly
Meeting started by tbachman at 16:02:11 UTC
(full logs).
Meeting summary
- agenda (tbachman, 16:02:24)
- https://meetings.opendaylight.org/opendaylight-nic/2015/nic_weekly/opendaylight-nic-nic_weekly.2015-02-06-16.09.html
Minutes from last week’s meeting (tbachman,
16:02:43)
- Agenda: PTL Election; Clarify SFC prototype in
release plan; thread discussions (tbachman,
16:03:29)
- agenda item: how to keep pressing key
topics (tbachman,
16:04:34)
- PTL Election (tbachman, 16:04:41)
- ShaunWackerly1 (ShaunWackerly1,
16:05:34)
- dbainbri (dbainbri,
16:05:37)
- devond (devond,
16:05:42)
- dmentze (dmentze,
16:05:45)
- tbachman (tbachman,
16:05:47)
- alagalah (alagalah,
16:05:53)
- gzhao (gzhao,
16:05:56)
- devond says last call for PTL election
(ShaunWackerly1,
16:06:09)
- devond has pulled his election nomination for
PTL; dmentze has added his (tbachman,
16:06:24)
- current nominations are dbainbri and
dmentze (tbachman,
16:06:31)
- ShaunWackerly1 asks how long the election
should take (tbachman,
16:06:50)
- election period will last for 1 week
(ShaunWackerly1,
16:06:51)
- https://wiki.opendaylight.org/view/CondorcetElection4PTLs
Wiki page showing how to setup PTL elections (tbachman,
16:07:51)
- devond asks what the criteria is to add
committers (tbachman,
16:08:11)
- phrobb says that it requires contributions to
the project, and project votes on whether to add the committer, then
brings this to the TSC (tbachman,
16:08:32)
- Clarify SFC Prototype in release plan (tbachman, 16:08:47)
- devond asks if anyone aside from dlenrow can
speak to this (tbachman,
16:09:21)
- dmentze says we’ll probably need dlenrow for
this (tbachman,
16:09:32)
- ACTION: devond says
he’ll contact dlenrow to get him at the next meeting to
discuss (tbachman,
16:09:52)
- How to keep pressing forward on key topics (ShaunWackerly1, 16:10:31)
- devond suggests options like f2f meetings or
more frequent webex (ShaunWackerly1,
16:11:11)
- dbainbri says f2f would be ideal, but non-local
people would need a way to be involved (ShaunWackerly1,
16:11:38)
- mlemay asks if anyting from ONF meeting will be
included in our NIC project (ShaunWackerly1,
16:12:08)
- dbainbri says ONF has lots of interest in
intent, but some things need clarified in terms of how ONF will work
on intent (ShaunWackerly1,
16:12:35)
- dbainbri says subgroups are looking at models,
should be looked at and pulled in (ShaunWackerly1,
16:13:40)
- devond asks if we can identify some key topics
to focus on (ShaunWackerly1,
16:14:06)
- dbainbri suggests "endpoint" definition,
SFC (ShaunWackerly1,
16:14:18)
- dmentze suggests multi-writer and multi-api,
and the difference between them (ShaunWackerly1,
16:14:42)
- dmentze says he has prototype code to share
with multi-writer (ShaunWackerly1,
16:15:20)
- mlemay says he also has code to share, would
like to discuss offline details (ShaunWackerly1,
16:15:40)
- dbainbri asks mlemay if his company will be
funded by ONF (ShaunWackerly1,
16:16:10)
- mlemay says that it is being decided
(ShaunWackerly1,
16:16:25)
- dbainbri notes that dlenrow (in the ONF
meeting) mentioned that ONF would make code available in git repo,
and that code would be utilized by ONOS and ODL (ShaunWackerly1,
16:17:04)
- mlemay says that politics get involved
(ShaunWackerly1,
16:17:33)
- dbainbri asks if everyone from NIC needs to
work on ONF (ShaunWackerly1,
16:18:00)
- mlemay says that NIC and ONF would need a
similar set of primitives need to be exposed (ShaunWackerly1,
16:18:19)
- ShaunWackerly1 asks how we can guarantee we've
got the same primitives (ShaunWackerly1,
16:19:20)
- mlemay says that's what cross-project meetings
are for (ShaunWackerly1,
16:19:31)
- dmentze asks how this will factor into the
policy engine that NIC is going to develop, how will it generate
native code (ShaunWackerly1,
16:20:34)
- mlemay says that the ONF stuff will sit on top,
above policy level (ShaunWackerly1,
16:20:58)
- dbainbri says he was under the impression that
ONF would develop the policy engine underneath (ShaunWackerly1,
16:21:24)
- dmentze says that if ONF wants to participate
in ODL, they are welcome to (ShaunWackerly1,
16:21:49)
- mlemay says that ONF wanted NIC to be the
reference implementation of what ONF wanted to do (ShaunWackerly1,
16:22:06)
- mlemay says ONF would bring standards, ODL
would bring implementation (ShaunWackerly1,
16:22:28)
- ShaunWackerly1 asks if agreement was made
between ONF and NIC (ShaunWackerly1,
16:24:00)
- mlemay says that ONF and dlenrow came up with a
compromise (not on behalf of NIC), it is more of an
experiment (ShaunWackerly1,
16:24:32)
- dbainbri says that NIC still retains the right
to not use ONF code (ShaunWackerly1,
16:24:53)
- mlemay says that we will remain flexible, that
another higher level runtime project may be what interfaces with
ONF (ShaunWackerly1,
16:25:36)
- mlemay mentions that the project could be
GBP (ShaunWackerly1,
16:25:50)
- mlemay says it is more politics than
technology (ShaunWackerly1,
16:26:23)
- mlemay says that the goal was an experiment to
try and come up with a way for ONF to work with ODL (ShaunWackerly1,
16:27:44)
- dbainbri adds "and ONOS" (ShaunWackerly1,
16:27:51)
- mlemay says will try to send more info about
ONF meetings, perhaps NIC can influence ONF's git repo (ShaunWackerly1,
16:28:38)
- dmentze notes that ONF is interested in
northbound API, asks if ONF has reviewed NIC discussions on NB
API (ShaunWackerly1,
16:29:01)
- mlemay says discussions in ONF are over whether
they will even have an API, or will it just be
primitives/models (ShaunWackerly1,
16:29:29)
- dbainbri notes that individuals in ONF have
pinged him, regarding discussions (ShaunWackerly1,
16:30:03)
- dmentze says that if ONF wants to participate,
they will need to speak up in NIC (ShaunWackerly1,
16:30:19)
- dbainbri says that ONF is restructuring to
focus on intent (ShaunWackerly1,
16:30:33)
- dbainbri says they might not have voiced
opinions because they don't know what they are doing yet
(ShaunWackerly1,
16:31:20)
- ("they" being ONF) (ShaunWackerly1,
16:31:27)
- mlemay says he and dbainbri may be able to help
bridge the gap (ShaunWackerly1,
16:32:10)
- dbainbri says he's not too involved with ONF,
but dlenrow is the chair of the ONF NBI group and dlenrow is also
involved in NIC (ShaunWackerly1,
16:32:37)
- phrobb says that as we figure out which
software ONF is building, we need to keep licenses in view. ONF and
ONOS are under apache, ODL is under EPL. (ShaunWackerly1,
16:33:27)
- phrobb says licenses would prevent moving code
from ONF to ODL easily (ShaunWackerly1,
16:34:34)
- dbainbri says that since it's "open source",
NIC folks could go over to ONF and develop code, swap between
projects (ShaunWackerly1,
16:35:21)
- dbainbri is scared by forking and having our
own copy of code (ShaunWackerly1,
16:35:34)
- phrobb says ONF is meant for upstreaming
w/apache license (ShaunWackerly1,
16:36:01)
- mlemay says he will try to keep informed of the
ONF work (ShaunWackerly1,
16:37:00)
- devond asks for more topics to discuss in the
future (ShaunWackerly1,
16:38:08)
- ShaunWackerly1 says we should decide on
classifier/filter (or not) (ShaunWackerly1,
16:38:32)
- dmentze says we need use cases defined before
API discussions (ShaunWackerly1,
16:38:47)
- ShaunWackerly1 notes that API is frozen on
3/19, need to freeze use cases prior (ShaunWackerly1,
16:39:34)
- dbainbri says use cases can be developed in
tandem to the API discussions (ShaunWackerly1,
16:40:15)
- dmentze and dbainbri agree that we should pick
a use case deadline (ShaunWackerly1,
16:40:37)
- mlemay suggests a topic, that we discuss
primitives (ShaunWackerly1,
16:41:03)
- ACTION: mlemay will
send out some primitives from the HP intent summit, and notes from
ONF discussions (ShaunWackerly1,
16:41:55)
- devond asks if we need a special use case
meeting (ShaunWackerly1,
16:42:53)
- ShaunWackerly1 says the only reason we'd need
that is in the presence of disagreement over whether a use case is
valid (ShaunWackerly1,
16:43:11)
- dmentze says he'd rather present advanced
topics (like multi-writer and SFC) in a f2f meeting, suggests we
could share the floor with ONF to present and listen (ShaunWackerly1,
16:44:06)
- devond says use cases will be discussed over
email, asks if we meet f2f next week (ShaunWackerly1,
16:44:37)
- dmentze says that f2f meeting should be the
following week (2/23-27?) (ShaunWackerly1,
16:45:02)
- dbainbri asks if multi-writer is predicated on
use cases (ShaunWackerly1,
16:45:17)
- dmentze says he can meet sooner than 2 weeks
from now, if desired (ShaunWackerly1,
16:45:41)
- mlemay and dbainbri say that f2f scheduling may
be problematic (ShaunWackerly1,
16:46:18)
- dmentze asks for poll regarding f2f meeting
regarding multi-writer and SFC (ShaunWackerly1,
16:46:35)
- dmentze asks who would be interested in f2f
meeting (ShaunWackerly1,
16:48:04)
- F2F Devond Roseville (devond,
16:48:25)
- dmentze bay area (dmentze,
16:48:31)
- DavidB - bay area (dbainbri,
16:48:38)
- F2F ShaunWackerly1 Roseville (ShaunWackerly1,
16:48:39)
- uchau Roseville (uchau,
16:48:48)
- keith bay area (dbainbri,
16:48:51)
- mlemay is based on ottowa, and can travel to
bay area (ShaunWackerly1,
16:49:47)
- dmentze (tbachman,
16:50:10)
- dmentze suggests bay area would give best
chance for participants (ShaunWackerly1,
16:50:17)
- dmentze roseville - but wiling to travel to bay
areal (tbachman,
16:50:20)
- dbainbri offers to donate ciena offices
(ShaunWackerly1,
16:50:30)
- dbainbri asks for a date (ShaunWackerly1,
16:51:27)
- dmentze proposes 2/18 or 2/19 (ShaunWackerly1,
16:51:37)
- ACTION: devond will
email mailing list to ask for interest and solidify the date
(ShaunWackerly1,
16:52:42)
- devond asks for pre-work that's necessary
before the meeting (ShaunWackerly1,
16:54:12)
- dmentze asks what is the best way to offer code
for review prior to meeting (ShaunWackerly1,
16:54:54)
- dbainbri suggests git (along with
tbachman) (ShaunWackerly1,
16:55:06)
- dbainbri asks for explanation of reasoning,
along with code (ShaunWackerly1,
16:55:29)
- ACTION: dbainbri will
secure meeting location for f2f (ShaunWackerly1,
16:56:15)
Meeting ended at 16:56:53 UTC
(full logs).
Action items
- devond says he’ll contact dlenrow to get him at the next meeting to discuss
- mlemay will send out some primitives from the HP intent summit, and notes from ONF discussions
- devond will email mailing list to ask for interest and solidify the date
- dbainbri will secure meeting location for f2f
Action items, by person
- dbainbri
- dbainbri will secure meeting location for f2f
- devond
- devond says he’ll contact dlenrow to get him at the next meeting to discuss
- devond will email mailing list to ask for interest and solidify the date
People present (lines said)
- ShaunWackerly1 (96)
- tbachman (48)
- odl_meetbot (6)
- alagalah (4)
- dbainbri (4)
- dmentze (3)
- uchau (3)
- devond (2)
- gzhao (2)
Generated by MeetBot 0.1.4.