#opendaylight-meeting: tsc
Meeting started by colindixon at 17:03:22 UTC
(full logs).
Meeting summary
- agenda bashing and roll call (colindixon, 17:03:27)
- colindixon (colindixon,
17:03:29)
- skitt (skitt,
17:03:32)
- rovarga (on voice) (colindixon,
17:03:34)
- abhijitkumbhare (abhijitkumbhare,
17:03:36)
- https://wiki.opendaylight.org/index.php?title=TSC:Main&oldid=53466#Agenda
(colindixon,
17:03:38)
- https://meetings.opendaylight.org/opendaylight-meeting/2017/tsc/opendaylight-meeting-tsc.2017-03-24-03.30.html
last week's meeting minutes (colindixon,
17:03:45)
- Hideyuki (hideyuki,
17:03:55)
- jamoluhrsen (jamoluhrsen,
17:04:03)
- katie (katiez,
17:04:08)
- Thanh Ha (zxiiro,
17:04:22)
- ACTION: colindixon to
e-mail about an informal ODL gathering at ONS (abeit a bit
late) (colindixon,
17:04:24)
- AGREED: there will be
no TSC call next week (decided a few weeks ago) (colindixon,
17:04:40)
- next DDF planning thread https://lists.opendaylight.org/pipermail/tsc/2017-March/006851.html
(colindixon,
17:05:29)
- LuisGomez (LuisGomez,
17:05:33)
- https://wiki.opendaylight.org/view/Events:Nitrogen_Dev_Forum
Submit topics to the Nitrogen DDF here. (CaseyODL,
17:06:07)
- ACTION: colindixon,
zxiiro and phrobb to come up with a proposal for tracking project
activity in a positive way (colindixon,
17:06:14)
- ACTION: phrobb and
tykeal to look into an ODL infra micro-datacenter in a box to make
things work better at tutorials (colindixon,
17:06:15)
- ACTION: rovarga,
phrobb, CaseyODL, and colindixon to work with marketing about maybe
commenting Boron-SR3's RFC-compatible RESTCONF (colindixon,
17:06:16)
- https://wiki.opendaylight.org/view/Events:Nitrogen_Dev_Forum
Submit topics to the Nitrogen DDF here. (CaseyODL,
17:06:51)
- ACTION: colindixon or
katie to send out nastygrams about being removed from the
release (colindixon,
17:06:55)
- ACTION: colindixon to
try to either find people to document how to be compatible with an
OpenDaylight release with participating in the OpenDaylight
simultaneous release (colindixon,
17:07:07)
- ACTION: colindixon to
create postmortem topics for Carbon and add the clustering breakage
to it (colindixon,
17:07:11)
- TSC mailing list (colindixon, 17:07:58)
- see agenda, mostly it's future DDF planning
(above) and nitrogen planning (below) (colindixon,
17:08:15)
- Events (colindixon, 17:08:20)
- https://www.opendaylight.org/global-events
(colindixon,
17:08:26)
- https://wiki.opendaylight.org/view/Events:Main
(colindixon,
17:08:31)
- ONS is next week Sunday to Thursday
(colindixon,
17:08:41)
- the DDF will have a wiki page up soon to start
populating topics for the one end of May and start of June
(colindixon,
17:09:47)
- shague (shague,
17:09:59)
- CaseyODL thinks that the booth demos are closed
for the ONS booth (colindixon,
17:10:17)
- boron (colindixon, 17:10:23)
- still churning on Boron-SR3 (colindixon,
17:10:39)
- https://docs.google.com/spreadsheets/d/1zImtd764e-hOgJAxoJKl85fxHCPu2agLfqsBtf13zQY/edit#gid=1776065453
spreadsheet (colindixon,
17:10:54)
- zxiiro says we're waiting on 14 projects to OK
it for now, colindixon asks if we know what fraction of them are the
exact same failures that were OKed in the previous spin (colindixon,
17:11:30)
- jamoluhrsen and zxiiro say they'll look at it
after this call (colindixon,
17:11:39)
- ACTION: colindixon to
work with jamoluhrsen and zxiiro on automating "real vs. fake"
failures for future releasing (colindixon,
17:12:50)
- ONS has the following Demo times available:
Tues, 3 and 4PM. Wed: 2, 3 and 4PM. Thrus: 11AM, 12 and 1:30PM. If
you want to showcase a demo in the main LF booth at ONS, please
contact me. (CaseyODL,
17:14:41)
- jamoluhrsen, edwarnicke, zxiiro and colindixon
say that the right solution to this is really to try very hard to
get projects to fix their tests, most of them aren't very hard to
fix (colindixon,
17:15:45)
- carbon (colindixon, 17:17:21)
- katiez reminds all projects to please respond
about karaf 4 (don't ignore it) (colindixon,
17:18:34)
- the following projects are missing M4: atrium
capwap cardinal centinel integration/distribution next of-config
ttp yang-push (colindixon,
17:19:06)
- the following (offset 0/1) projecs are missing
M5: netconf, infrautils and genius (colindixon,
17:19:36)
- ACTION: anipbu to
work with katiez to validate the M4 status reports (use colindixon
if need be) (colindixon,
17:20:21)
- anipbu (anipbu,
17:20:49)
- https://bugs.opendaylight.org/show_bug.cgi?id=7976
(katiez,
17:21:23)
- https://bugs.opendaylight.org/show_bug.cgi?id=8021
(katiez,
17:21:38)
- two API change waivers: https://bugs.opendaylight.org/show_bug.cgi?id=7976
(colindixon,
17:21:53)
- two API change waivers: https://bugs.opendaylight.org/show_bug.cgi?id=8021
(colindixon,
17:21:58)
- rovarga says that they should affect nobody in
OpenDaylight, but might affect downstream, one is a typo fix another
is a race condition fix which requires a small API change
(colindixon,
17:22:28)
- rovarga asks if we're still planning to cut the
carbon branch today or hold off (colindixon,
17:23:23)
- jamoluhrsen notes that adding a new branch will
increase CSIT from boron + carbon running daily to boron + carbon +
nitrogen running daily? (colindixon,
17:24:33)
- colindixon asks how we're doing with Karaf 4?
jamoluhrsen says that during the integration call the problem is
that RESTCONF doesn't work in Karaf 4 and everything is blocked on
that. (colindixon,
17:27:23)
- vrpolak notes that many jobs to a RESTCONF
check to get a 200 OK, and they never get it (colindixon,
17:28:12)
- skitt says that RESTCONF never comes up in the
traces that vrpolak sees (colindixon,
17:28:22)
- ACTION: rovarga will
take a look at the isuse with RESTCONF as soon as vrpolak files a
bug for it (colindixon,
17:29:39)
- AGREED: we will cut
stable/carbon branches on Monday, April 10th at 10pm UTC unless
otherwise decided (colindixon,
17:32:37)
- ACTION: colindixon to
statt a thread as to when branch makes sense for
stable/carbon (colindixon,
17:32:51)
- https://bugs.opendaylight.org/show_bug.cgi?id=8117
<-- the karaf4 restconf bug (vrpolak,
17:34:03)
- skitt says that the bug tracker has 24 issues
open, but that may be more like 20 (colindixon,
17:34:10)
- skitt says that skitt has had lots of requests
for help with Karaf 4 migrations recently, he thinks that many of
the 20 projects are just starting or have yet to start (colindixon,
17:35:26)
- https://bugs.opendaylight.org/showdependencytree.cgi?id=4219&hide_resolved=1
(skitt,
17:35:47)
- rovarga says that the issue in GBP is actually
caludio helping and discovering projects which don't follow best
practices upstream of GBP (colindixon,
17:36:28)
- rovarga says that claudio says that things get
worse as they move to offset 2 (colindixon,
17:36:43)
- skitt notes that some offset 2 projects don't
have the SingleFeatureTest enabled, which means it's usually
broken (colindixon,
17:36:58)
- ACTION: colindixon to
work with katiez to let projects that don't have karaf 4 done that
they will be dropped from the release without it (colindixon,
17:37:40)
- skitt notes that almost everyone has moved to
features-parent, but need to do pacakging jar to get the
SingleFeatureTest to run (colindixon,
17:38:13)
- skitt says it would be really good if people
could test their karaf4 code now (colindixon,
17:39:48)
- to test with Karaf 4, switch from you're local
karaf disgtributions use of karaf-parent to karaf4-parent and use
the features4 repository instead of features (colindixon,
17:40:59)
- colindixon asks about Karaf 5, skitt says Karaf
4.1 is easy, Karaf 4.2 or 5.0 will use Java 9 which is a
wildcard (colindixon,
17:41:31)
- skitt says that the next fun upgrade after
Karaf 4 will be guava (colindixon,
17:41:41)
- skitt qualifies "easy" as "not too hard"
(skitt,
17:41:49)
- system integration and test (colindixon, 17:41:58)
- we have reduced the numbers of times we need to
run our CSIT jobs, yay! (colindixon,
17:42:29)
- infrastructure (colindixon, 17:42:49)
- none this week (colindixon,
17:42:52)
- gaps in release management (colindixon, 17:43:02)
- colindixon notes that you may have been hearing
more from him, jamoluhrsen and zxiiro during the last few weeks with
Boron-SR3 and carbon as we transition from different people in
release management roles (colindixon,
17:44:17)
- colindixon notes that we would love help
(colindixon,
17:44:52)
- CaseyODL notes that one major thing people are
commenting about is the test failures being reported for builds and
then easier to update things frequently (colindixon,
17:46:13)
- CaseyODL also notes that keeping blocking bugs
up-to-date is key (colindixon,
17:47:00)
- https://bugs.opendaylight.org/buglist.cgi?bug_severity=blocker&bug_status=UNCONFIRMED&bug_status=CONFIRMED&bug_status=IN_PROGRESS&bug_status=WAITING_FOR_REVIEW&f1=cf_target_milestone&f2=cf_target_milestone&j_top=OR&list_id=75953&o1=substring&o2=substring&query_format=advanced&resolution=---&v1=Boron&v2=---
blocking bugs list (zxiiro,
17:47:51)
- https://wiki.opendaylight.org/view/Weather#Blocker_Bugs
(colindixon,
17:48:02)
- katiez asks where blocking bugs come from,
colindixon says the authoritative place where people track bugs is
in the tracking spreadsheet here:
https://docs.google.com/spreadsheets/d/1MufwajG-YwGU7dth97x07VVRznVx4UgPHkfRFk6-bQg/edit#gid=1293046707
(for carbon) (colindixon,
17:50:00)
- periodically, people queriy the buzilla for
blocking bugs to find them and move them to the speadsheet
(colindixon,
17:50:27)
- ACTION: zxiiro and
katiez to work together to describe how to to generate the build
failure reports from distribution jobs (colindixon,
17:51:07)
- https://docs.google.com/spreadsheets/d/1zImtd764e-hOgJAxoJKl85fxHCPu2agLfqsBtf13zQY/edit#gid=1703426731
(colindixon,
17:51:25)
- https://docs.google.com/spreadsheets/d/1zImtd764e-hOgJAxoJKl85fxHCPu2agLfqsBtf13zQY/edit#gid=1703426731
columns D and E are semi-automatically generated on distribution
jobs (colindixon,
17:51:51)
- conference call platform platforms (colindixon, 17:53:23)
- CaseyODL and phrobb say that the LF uses Zoom
effectively for cross-OS conference calling with large numbers of
participants (colindixon,
17:53:49)
- https://bugs.opendaylight.org/show_bug.cgi?id=8091
related to release engineering, this but is asking for notifications
on merge failiures for active releases, which should signify a major
problem (colindixon,
17:55:09)
- CaseyODL notes that Zoom has nearly universal
support for OSes (Linux, Mac, Windows) (colindixon,
17:55:46)
- it allows for local and cloud recording to mp4
files as well as anyone to record (or not) which means you don't
need to have the host code to record and easliy get recordings
out (colindixon,
17:56:15)
- license only comes with 1 GB of cloud storage,
but can be upraded to $10/GB/mo, but local recording allows for
that (colindixon,
17:56:44)
- each license $14.99/mo for 50 people, can go up
to 100 people for $40/mo, can allow some licenses to be community
administered to let phrobb and CaseyODL get out of the way from
people scheduling meetings (colindixon,
17:57:46)
- can get suage reports, if people want to use
them (colindixon,
17:57:57)
- starting at $100/mo, we can get international
dial-ins, credit for dial ins to numbers (colindixon,
17:59:28)
- edwarnicke asks if does who's talking
notifcation + can mute people for them + recording indication
(colindixon,
18:00:24)
- colindixon says that he has motivated humans at
the Kernel projects + TWS when we have one (colindixon,
18:01:02)
- edwarnicke asks if multilple concurrent
meetings to go on at the same time, need a separate license for each
meeting, we could have multiple licenses (colindixon,
18:01:39)
- CaseyODL said that having ~6 licenses which
would more than take care of our needs a lower cost than what we do
now (colindixon,
18:03:06)
- abhijitkumbhare says given that we usually ask
people before recording, how will that work with local recording,
CaseyODL say that by default only admins can record, can be done
per-call to allow only certain people to record if we want to make
sure to notify people (colindixon,
18:04:02)
- ACTION: CaseyODL to
send mail about Zoom to relevant lists (colindixon,
18:04:54)
- security mailng list (colindixon, 18:05:00)
- the mailing list is defunct, rovarga says that
there are only 2 people reading it and neither seem to have time
respond (colindixon,
18:05:27)
- cookies (colindixon, 18:05:42)
- ACTION: colindixon to
add security community population to the TSC next time (colindixon,
18:06:15)
- cookies (colindixon, 18:06:16)
Meeting ended at 18:06:17 UTC
(full logs).
Action items
- colindixon to e-mail about an informal ODL gathering at ONS (abeit a bit late)
- colindixon, zxiiro and phrobb to come up with a proposal for tracking project activity in a positive way
- phrobb and tykeal to look into an ODL infra micro-datacenter in a box to make things work better at tutorials
- rovarga, phrobb, CaseyODL, and colindixon to work with marketing about maybe commenting Boron-SR3's RFC-compatible RESTCONF
- colindixon or katie to send out nastygrams about being removed from the release
- colindixon to try to either find people to document how to be compatible with an OpenDaylight release with participating in the OpenDaylight simultaneous release
- colindixon to create postmortem topics for Carbon and add the clustering breakage to it
- colindixon to work with jamoluhrsen and zxiiro on automating "real vs. fake" failures for future releasing
- anipbu to work with katiez to validate the M4 status reports (use colindixon if need be)
- rovarga will take a look at the isuse with RESTCONF as soon as vrpolak files a bug for it
- colindixon to statt a thread as to when branch makes sense for stable/carbon
- colindixon to work with katiez to let projects that don't have karaf 4 done that they will be dropped from the release without it
- zxiiro and katiez to work together to describe how to to generate the build failure reports from distribution jobs
- CaseyODL to send mail about Zoom to relevant lists
- colindixon to add security community population to the TSC next time
Action items, by person
- anipbu
- anipbu to work with katiez to validate the M4 status reports (use colindixon if need be)
- CaseyODL
- rovarga, phrobb, CaseyODL, and colindixon to work with marketing about maybe commenting Boron-SR3's RFC-compatible RESTCONF
- CaseyODL to send mail about Zoom to relevant lists
- colindixon
- colindixon to e-mail about an informal ODL gathering at ONS (abeit a bit late)
- colindixon, zxiiro and phrobb to come up with a proposal for tracking project activity in a positive way
- rovarga, phrobb, CaseyODL, and colindixon to work with marketing about maybe commenting Boron-SR3's RFC-compatible RESTCONF
- colindixon or katie to send out nastygrams about being removed from the release
- colindixon to try to either find people to document how to be compatible with an OpenDaylight release with participating in the OpenDaylight simultaneous release
- colindixon to create postmortem topics for Carbon and add the clustering breakage to it
- colindixon to work with jamoluhrsen and zxiiro on automating "real vs. fake" failures for future releasing
- anipbu to work with katiez to validate the M4 status reports (use colindixon if need be)
- colindixon to statt a thread as to when branch makes sense for stable/carbon
- colindixon to work with katiez to let projects that don't have karaf 4 done that they will be dropped from the release without it
- colindixon to add security community population to the TSC next time
- jamoluhrsen
- colindixon to work with jamoluhrsen and zxiiro on automating "real vs. fake" failures for future releasing
- katiez
- anipbu to work with katiez to validate the M4 status reports (use colindixon if need be)
- colindixon to work with katiez to let projects that don't have karaf 4 done that they will be dropped from the release without it
- zxiiro and katiez to work together to describe how to to generate the build failure reports from distribution jobs
- vrpolak
- rovarga will take a look at the isuse with RESTCONF as soon as vrpolak files a bug for it
- zxiiro
- colindixon, zxiiro and phrobb to come up with a proposal for tracking project activity in a positive way
- colindixon to work with jamoluhrsen and zxiiro on automating "real vs. fake" failures for future releasing
- zxiiro and katiez to work together to describe how to to generate the build failure reports from distribution jobs
People present (lines said)
- colindixon (102)
- katiez (8)
- jamoluhrsen (7)
- skitt (6)
- odl_meetbot (6)
- CaseyODL (5)
- shague (4)
- zxiiro (4)
- anipbu (3)
- vrpolak (2)
- LuisGomez (1)
- hideyuki (1)
- abhijitkumbhare (1)
Generated by MeetBot 0.1.4.