14:59:16 <frankbrockners> #startmeeting OPNFV BGS/Genesis weekly meeting
14:59:16 <collabot> Meeting started Mon Aug 17 14:59:16 2015 UTC.  The chair is frankbrockners. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:59:16 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:59:16 <collabot> The meeting name has been set to 'opnfv_bgs_genesis_weekly_meeting'
14:59:32 <frankbrockners> #chair pbandzi
14:59:32 <collabot> Current chairs: frankbrockners pbandzi
14:59:36 <szilard_> #info Szilard Cserey
14:59:37 <pbandzi> #info Peter Bandzi
14:59:46 <fdegir> #info Fatih Degirmenci
14:59:51 <amaged_> Hi frank, peter
14:59:55 <frankbrockners> Hi folks, I won't be able to attend the first 30min...
15:00:05 <frankbrockners> pbandzi will drive the meeting
15:00:14 <trozet> #info Tim Rozet
15:00:33 <frankbrockners> #info agenda for today: https://wiki.opnfv.org/meetings/bgs#aug172015
15:01:03 <frankbrockners> pbandzi... please take over.. talk to all of you in a bit...
15:01:07 <pbandzi> sure
15:01:29 <pbandzi> #topic LF lab reconfig status
15:01:44 <dduffey> #info David Duffey
15:02:07 <pbandzi> #info trzoet mentioned in octopus meeting:  Fuel fixes were put in.  A common clean that should work between installers was added as well.  We tested Foreman to make sure it was working, but for some reason it's nodes cannot ping on the public interface
15:02:16 <pbandzi> trozet
15:02:39 <chenshuai> #info chenshuai
15:02:52 <ChrisPriceAB> #info chris price
15:03:08 <trozet> #info pbandzi figured out the issue was puppet-opendaylight has been upgraded to Lithium, and the wrong feature set is being booted (for lithium) so public interface has no flows and cannot push traffic
15:03:24 <[1]JonasB> #info Jonas Bjurel
15:03:28 <pbandzi> thanks I was trying to write it
15:03:42 <trozet> #info trozet temporarily pinned puppet-opendaylight to Arno version.  Will contact dfarrell07_con to get an SR1 version of puppet-odl
15:04:09 <trozet> #info new redeploy is running with change: https://build.opnfv.org/ci/job/genesis-foreman-deploy-master/49/console
15:04:26 <trozet> #info If Foreman passes, Fuel is next up to verify on Pod2
15:04:55 <pbandzi> #info now we are reconfiguring POD1 : 4 virtual deployments will be running there (2 centos, 2 ubuntu jumpohosts)
15:05:27 <trozet> pbandzi: I thought konstantin said we had to wait till after this week?  Are we able to do it sooner?
15:05:49 <pbandzi> #info currently waitng for more public neworks -> will lbe provided by LF next week (this week they are traveling... -> most of them is not in the office)
15:06:54 <pbandzi> trozet: we need to wait for them.. I am instaling servers in the meantime.. and when subnets will be available we just change IPs and can proceed
15:07:07 <trozet> pbandzi: ok cool thanks
15:07:30 <pbandzi> #topic Arno SR1 - brief status
15:08:03 <pbandzi> anyone can info news for this topic?
15:08:24 <ChrisPriceAB> Guys, on Arno SR1 we need to address the TSC on a discussion around the proposed dates and content.
15:08:51 <pbandzi> #info <ChrisPriceAB> Guys, on Arno SR1 we need to address the TSC on a discussion around the proposed dates and content.
15:09:12 <trozet> ChrisPriceAB: when will that discussion happen?
15:09:15 <ChrisPriceAB> #info The TSC needs to be ready to vote for the release and should previously have set the release date.  :)
15:09:28 <ChrisPriceAB> I would hope we can do that tomorrow, it's kind of urgent.
15:09:39 <[1]JonasB> #info For fuel, all component work/coding for SR1 is done. Starting to rebase the build system today. Deploy mods are ready for SR1
15:09:44 <ChrisPriceAB> But needs input from this team to drive the discussion
15:10:46 <trozet> ChrisPriceAB: the date we had discussed in BGS was end of sept.
15:12:08 <ChrisPriceAB> Yes Frank mentioned that to me today, our SR1 date has some impact on the B-Release planning.  It's important we understand the dependencies.  Is the proposed date clear and do we have all dependencies covered from an SR1 perspective?
15:12:41 <trozet> ChrisPriceAB:  I think Fuel and Foreman have said they are comfortable with that date
15:12:52 <trozet> ChrisPriceAB: with code freeze mid sept.
15:13:21 <trozet> ChrisPriceAB: release includes fixes and minor changes to the installers with keeping major versions the same (Helium+Juno)
15:13:33 <[1]JonasB> I am confident :-)
15:13:54 <ChrisPriceAB> Ok, that's fine.  Do we see any impacts on the main branch (activities we need to postpone), or is this relatively well seperated from impacting brahmaputra activity?
15:14:08 <pbandzi> #info agreed release date: end of Sep, code freeze: mid Sep,
15:14:39 <trozet> ChrisPriceAB: For Foreman this is separate from B release activities there is no impact for us
15:15:12 <[1]JonasB> No issues for Fuel either
15:15:30 <ChrisPriceAB> Ok, then we'll take it to the TSC tomorrow and rubber stamp.  :)
15:15:34 <trozet> ChrisPriceAB: the one dependency that I am concerned about is getting virtual deploy verification
15:15:46 <trozet> ChrisPriceAB: so that I can verify fixes for SR1 when I actually submit a patch
15:16:05 <trozet> ChrisPriceAB: but we are all working towards that so
15:16:29 <pbandzi> #agreed release date: end of Sep, code freeze: mid Sep, (no issues both teams), TSC to confirm tomorrow
15:16:33 <fdegir> trozet: I think it should be nice to have
15:16:51 <fdegir> for SR1
15:16:52 <ChrisPriceAB> trozet: Ok, that's related to the lab re-config and any needed patches.  Can release without, prefer not to?
15:16:57 <fdegir> musth have Brahmaputra
15:17:23 <trozet> ChrisPriceAB: we can release without it its just takes more effort
15:17:48 <ChrisPriceAB> Ok, thanks.
15:19:12 <pbandzi> trozet: one installer with centos7 is already availble on pod1
15:19:22 <trozet> pbandzi: oh?
15:19:31 <trozet> pbandzi: with public network connectivity?
15:19:33 <pbandzi> if it is so crucial you can use that also for virtual deployment
15:19:36 <pbandzi> yes
15:19:39 <trozet> great!
15:19:50 <pbandzi> the one which was used by fuel previously
15:20:09 <trozet> pbandzi: I'll update the jjb to use it for virtual deployment
15:20:17 <pbandzi> trozet: ok
15:20:36 <pbandzi> next topic
15:20:44 <ChrisPriceAB> Can we propose fixed dates?  (CF - Sept 8, Rel - 22) or (CF Sept 15, Rel Sept 29)
15:21:00 <trozet> 2nd one please :)
15:22:00 <ChrisPriceAB> ok
15:22:22 <pbandzi> #agreed fixed dates for R1:  (CF Sept 15, Rel Sept 29)
15:22:31 <trozet> [1]JonasB are you ok with that?
15:24:00 <[1]JonasB> Yes
15:24:04 <fdegir> pbandzi: what is that "one installer with centos7 is already availble on pod1"
15:24:37 <trozet> fdegir: I assume its just the POD1 jumphost still intact?
15:24:39 <pbandzi> fdegir, at least I hope so.. Is Szilard using it?
15:24:53 <fdegir> pbandzi: it needs to be connected to jenkins as slave
15:25:04 <fdegir> in order for trozet to use it for virt deploy from jenkins
15:26:05 <fdegir> trozet: jumphost looks unchanged
15:26:11 <pbandzi> fdegir: yes I mean unused that no one else is using it now and then trozet may use it for virtual deployment
15:26:36 <trozet> fdegir: is it ok if i update jjb to use pod1 for virtual deploy and then add it as part of jenkins-verify?
15:26:44 <pbandzi> I think that in this case that host will not be restarted and will stay untuched from jenkins point of view
15:27:02 <fdegir> trozet: I think it should be ok
15:27:08 <fdegir> szilard_ is working on pod2
15:27:17 <trozet> fdegir: ok will make sure it works first before I add it as part of jenkins verify
15:27:18 <fdegir> and the rest of the machines from pod1 waiting for IPs
15:27:32 <fdegir> so pod1 jumphost is unused
15:28:12 <pbandzi> #info POD1 jumhost will be used for foreman virtual deployment CI
15:28:45 <pbandzi> #topic Genesis - next steps
15:28:59 <chenshuai> Hi guys, has genesis proposal been approved by TSC?
15:29:05 <frankbrockners> hey guys - I'm back
15:29:12 <pbandzi> by this time I hope frankbrockners is here
15:29:15 <pbandzi> ok
15:29:17 <frankbrockners> #info Genesis has been approved by TSC
15:29:22 <chenshuai> great
15:29:32 <chenshuai> some installers issues I think should be discussed and managed by genesis.
15:29:45 <frankbrockners> #info project infra  (genesisreq repo and Jira have been setup
15:30:09 <frankbrockners> #info we can start to move requirements from the sandbox to Jira
15:30:49 <frankbrockners> #info For Jira, it would make sense to create some groupings - similar to the headlines in the sandbox (we can use Epics for that purpose)
15:31:22 <frankbrockners> #info Frank will propose some inital structure - which we can then jointly evolve
15:32:18 <frankbrockners> #info Am also fiddling around with gerrit rules.pl to make sure that we have everyone voting on patches before they go into the repo. Not done yet (anyone out there who knows Prolog and wants to help?)
15:33:17 <frankbrockners> #info We'll also need to (eventually) update the genesis work procedures (i.e. how to work with Genesis): https://wiki.opnfv.org/genesis/genesis_work_procedures - please review and ammend/update
15:33:35 <frankbrockners> #info Once complete we can let all other teams know
15:34:06 <frankbrockners> this would probably also help with chenshuai's question/note earlier
15:34:48 <frankbrockners> For now, this is it on Genesis update - i.e. will put some structure in place in Jira and then we can start to raise tickets.
15:35:14 <frankbrockners> Additional thoughts on what else we need to get done as next steps for Genesis?
15:35:15 <trozet> frankbrockners: is there some way we could have a wiki pull from JIRA to create the approved requirements?
15:35:31 <trozet> create them in  a webpage
15:35:39 <trozet> rather than manually editing the wiki to update what is in JIRA?
15:36:17 <fdegir> trozet: https://www.dokuwiki.org/plugin:jiralinks
15:36:19 <frankbrockners> trozet: Not sure... Let's move away from the googledoc and put everything into Jira. Detailed requirements should become patches that go into the repo anyway
15:36:21 <fdegir> haven't read it fully
15:36:43 <trozet> thanks fdegir
15:36:54 <fdegir> given that the tool (wikis, confluence) discussion going on
15:37:03 <fdegir> I think it would be good to raise this question
15:37:22 <frankbrockners> yup - confluence nicely integrates with jira
15:37:26 <fdegir> +1
15:37:32 <frankbrockners> (not a surprise - same company)
15:38:14 <chenshuai> Jira is greate, no access to googledoc in china
15:39:38 <frankbrockners> what I thought as process for Genesis is: high-level ticket in Jira -> create detailed patch (could be code, could be doc/rst) -> all committers vote -> if we have >3, then we merge, otherwise we abandon
15:39:47 <frankbrockners> ok for everyone?
15:40:38 <narindergupta> +1
15:40:58 <trozet> +1
15:41:15 <[1]JonasB> +1
15:41:20 <lmcdasm2> i am assuming the answer to the JIRA ticket would be an indication of whether to proceed or not - so that we dont have people contributing things and then it get voted down
15:41:21 <lmcdasm2> +1
15:41:26 <lmcdasm2> (otherwise :) )
15:42:06 <frankbrockners> lmcdasm2 - yup - we should update the jira ticket according to the vote
15:43:16 <frankbrockners> we could also do without jira tickets... but they help keep some structure and jira could nicely replace the current sandbox to collect ideas which might become common requirements
15:43:31 <frankbrockners> let's #agree the above
15:44:00 <frankbrockners> #agree high level process for Genesis: create high-level ticket in Jira -> create detailed patch (could be code, could be doc/rst) -> all committers vote -> if we have >3, then we merge, otherwise we abandon
15:44:23 <frankbrockners> any other thoughts on immediate next steps?
15:44:42 <frankbrockners> if not, we can finish 15min early ;-)
15:44:46 <pbandzi> #agreed high level process for Genesis: create high-level ticket in Jira -> create detailed patch (could be code, could be doc/rst) -> all committers vote -> if we have >3, then we merge, otherwise we abandon
15:44:55 <pbandzi> just 'd' at the end
15:45:14 <frankbrockners> thanks pbandzi
15:45:36 <frankbrockners> any other topic that needs discussion?
15:46:05 <frankbrockners> ... if not - thanks pbandzi for hosting the session and everyone else for participating.
15:46:09 * ChrisPriceAB ponders an alternative to cookies...
15:46:38 <frankbrockners> #endmeeting