20:00:13 <johnsom> #startmeeting Octavia
20:00:17 <openstack> Meeting started Wed Sep 28 20:00:13 2016 UTC and is due to finish in 60 minutes.  The chair is johnsom. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:00:20 <sbalukoff> Howdy folks!
20:00:21 <openstack> The meeting name has been set to 'octavia'
20:00:30 <johnsom> Hi there
20:01:01 <johnsom> #topic Announcements
20:01:13 <johnsom> Octavia 0.9.0 (newton) released!
20:01:20 <johnsom> #link http://lists.openstack.org/pipermail/openstack-announce/2016-September/001695.html
20:01:46 <sbalukoff> Woo-hoo!
20:01:47 <johnsom> As always I really appreciate the work everyone on the team has done this cycle.
20:02:08 <johnsom> It was a bit of a rough cycle for a lot of us, but we still got some good things done.
20:02:15 <johnsom> So, thank you!
20:02:17 <sbalukoff> Yep!
20:02:38 <johnsom> Next up: Ocata schedule
20:02:41 <sbalukoff> Also this means we're now open to merging stuff again.
20:02:49 <johnsom> #link https://releases.openstack.org/ocata/schedule.html
20:02:49 <sbalukoff> (Stuff that didn't make it into Newton)
20:03:18 <johnsom> Yes, that is the next topic here. Ocata.  We are free to start working on Ocata features.
20:03:34 <johnsom> The schedule is short due to the summit changes, etc.
20:03:41 <sbalukoff> The major initiative is the neutron-lbaas merge.
20:03:43 <johnsom> So, we need to be on our game for Ocata
20:03:57 <sbalukoff> As well as whatever is important for your organization. (eg. active-active)
20:04:33 <johnsom> Yeah, so what do you all think about trying to limit non-merge/bug fix related changes to landing in Ocata-1?
20:05:11 <sbalukoff> I'm reluctant to have a hard commitment for that, but I'd be all for that being a good guideline.
20:05:16 <johnsom> I am worried that as Ocata goes along, too many other reviews/feature changes will put the merge at risk
20:05:34 <sbalukoff> Ocata-1 is just 6 weeks out.
20:05:37 <johnsom> Yeah, I'm not putting it in stone, but floating the "idea"
20:06:47 <sbalukoff> So! If you have a major feature you want to get in (*cough* ACTIVE-ACTIVE *cough*), then NOW is the time to get those patches in good shape, and ask incessantly for reviews!
20:06:59 <johnsom> Yes
20:07:50 <johnsom> Ok, well, let's set that as a goal, new features landing in Ocata-1.
20:07:53 <perelman> hi
20:08:01 <sbalukoff> Speak of the devil!
20:08:03 <sbalukoff> So! If you have a major feature you want to get in (*cough* ACTIVE-ACTIVE *cough*), then NOW is the time to get those patches in good shape, and ask incessantly for reviews!
20:08:06 <johnsom> Hi there perelman
20:08:36 <perelman> i am trying to get them in a good shape
20:08:47 <sbalukoff> perelman: Since Ocata is a weird, short cycle, we're trying to get major features in by Ocata-1. (This is a "firm guideline" we're self-imposing.)
20:09:00 <sbalukoff> Good deal! Please poke people for specific reviews.
20:09:10 <perelman> there were some issues with some tests for my recent commits that it looks like problems of the testing environment
20:09:26 <sbalukoff> Because, given how many patches there are... well, it can help to know *what* we should be looking at first.
20:09:56 <johnsom> perelman If you need help figuring out test failures, etc.  Please don't hesitate to ask in the channel.
20:10:02 <sbalukoff> Yep.
20:10:04 <perelman> #link: https://review.openstack.org/#/c/313006/
20:10:13 <perelman> this one is the first patch
20:10:15 <johnsom> I for one am happy to try to help
20:10:20 <sbalukoff> I think we all would love it if we can get N+1 ACTIVE-ACTIVE in in this cycle.
20:10:50 <perelman> #link: https://review.openstack.org/#/c/317629/25
20:10:59 <johnsom> It needs to land, yes
20:11:09 <sbalukoff> perelman: Could you create an etherpad which lists the patches in order?
20:11:14 <perelman> #link: https://review.openstack.org/#/c/320422/19
20:11:34 <perelman> yes, sure. I will do it now
20:11:38 <sbalukoff> Thanks!
20:11:46 <johnsom> Yes, that would be very handy to have.  An etherpad with all of the patches, numerically ordered in the order you would like reviews.
20:12:00 <johnsom> Excellent idea.
20:12:15 <sbalukoff> Are we still on announcements?
20:12:26 <johnsom> Yes, I wasn't done talking about Ocata
20:12:32 <sbalukoff> Oh, sorry!
20:12:37 <johnsom> How do you all feel about using the launchpad milestones to track/project the merge activities?
20:12:48 <sbalukoff> We can give it a try, eh.
20:13:08 <johnsom> We got away from using launchpad for a while, but I'm a bit worried about getting all of the bits and pieces done for the merge.
20:13:10 <sbalukoff> Note that we're down a couple of major contributors (one of which is/was a core)
20:13:13 <johnsom> Some have dependencies, etc.
20:13:26 <johnsom> Yes, exactly.
20:14:02 <johnsom> I am expecting to increase my focus on upstream this cycle, so that will be good, but yeah, contributors we are a bit short on
20:14:02 <sbalukoff> That seems like a good way to communicate priorities to people looking to contribute.
20:14:25 <johnsom> Ok, if no one objects, I will take a first pass at tagging the RFEs with milestones.
20:14:40 <sbalukoff> As usual, I expect contributors to do essentially "what they want" or "what their employer tells them to work on", but it helps to know what are the priorities for the upstream project.
20:15:06 <johnsom> Yep.  It will help me gauge if we are going to make the merge happen during Ocata as well
20:15:17 <sbalukoff> Sounds good.
20:15:32 <johnsom> Ok
20:15:33 <johnsom> #topic Brief progress reports / bugs needing review
20:15:58 <johnsom> I have been focused on bug fixes, reviews, and a release...  grin
20:16:07 <sbalukoff> There are a few patches that have been in queue for a while with one +2 on them (usually mine) that were waiting for Newton to be tagged.
20:16:09 <perelman> #link: https://etherpad.openstack.org/p/Active-_Active_Topology_commits
20:16:12 <sbalukoff> These can now potentially be merged.
20:16:33 <perelman> i have added all the fresh commits for Active Active Topology there
20:16:34 <johnsom> Yep, I will be getting started on those today/tomorrow
20:16:42 <sbalukoff> I've also been concentrating on bug-fixes, documentation, and other stuff identified as a priority for Newton.
20:17:00 <sbalukoff> perelman: Nice!
20:17:37 <johnsom> Cool.  I might reorganize a bit.  Ok, someone is doing that
20:17:51 <johnsom> Nice, any other updates?
20:18:25 <johnsom> #topic Translation for Octavia
20:18:34 <johnsom> dougwig Are you here this week?
20:18:40 <dougwig> aye
20:18:48 <dougwig> got back yesterday.
20:19:36 <johnsom> Awesome.  So, I see neutron-lbaas is translated.  What do we need to do for Octavia?  I assume it will need translation as well for the merge.
20:20:04 <johnsom> #link https://github.com/openstack-infra/project-config/blob/master/gerrit/projects.yaml#L4812
20:20:48 <dougwig> does it have the translation hooks?  one sec.
20:21:35 <johnsom> The log messages are all tagged
20:21:36 <perelman> I saw some strange error in running stack script in some tests, but in others doing the same for the same commits...  0     0    0     0    0     0      0      0 --:--:--  0:04:37 --:--:--     0* After 21537ms connect time, move on! 2016-09-28 10:29:46.950 | * connect to 151.101.0.175 port 443 failed: Connection timed out 2016-09-28 10:29:46.950 | * Failed to connect to bootstrap.pypa.io port 443: Connection timed out
20:21:59 <perelman> http://logs.openstack.org/25/336925/14/check/gate-neutron-lbaasv2-dsvm-healthmonitor/0f6ff1c/logs/devstacklog.txt.gz
20:22:11 <johnsom> Yeah, that is an infra issue that is going on.  It is not your patch.
20:22:16 <sbalukoff> perelman: Er... let's do troubleshooting outside of this meeting.
20:22:23 <perelman> for the same patch other tests run it with connecting to the host without any issues
20:22:31 <perelman> any ideas, what could it be?
20:22:49 <dougwig> johnsom: the old way seems to be gone.  we'll need to dig into how to enable.
20:22:50 <perelman> ok, sorry...
20:22:54 <sbalukoff> No worries.
20:23:10 <perelman> should i open a bug for it or does the issue has open bug?
20:23:10 <johnsom> dougwig Ok.  Is that something you can help us with?
20:23:50 <sbalukoff> perelman: If it's an intermittent infra issue, just do a "recheck"
20:23:59 <dougwig> johnsom: ahh, it might be the hooks in setup.cfg.
20:24:03 <perelman> ok, thanks!
20:24:06 <dougwig> johnsom: yes, i'll find out
20:24:13 <johnsom> Excellent, thanks
20:24:28 <johnsom> #topic Debian package build job for Octavia
20:25:10 <johnsom> dougwig Another question I had for you.  What is this deb job for neutron-lbaas?
20:25:12 <johnsom> #link https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/projects.yaml#L1748
20:25:35 <johnsom> Is OpenStack packaging debs now?  Is this something we need to setup for Octavia?
20:25:36 <dougwig> i literally have no idea.  :)  let me ping someone.
20:25:46 <sbalukoff> Heh! Ok.
20:26:07 <sbalukoff> That'd be good to know... because if it is, I expect rpms will be coming right on the heels of that.
20:26:08 <johnsom> I feel so much better now that I wasn't the only one that didn't know....  grin
20:26:31 <dougwig> that was always magic to me.
20:26:51 <sbalukoff> If you've ever had to build RPMs in a CI, you know that it is mostly magic.
20:27:10 <sbalukoff> I don't envy those who have to maintain the infrastructure for that...
20:27:12 <johnsom> RPMs are more straight forward to me than the debs...
20:27:26 <johnsom> But, I started building rpms before I had to do debs, so...
20:28:07 <johnsom> Ok, we can follow up on that later dougwig.
20:28:15 <johnsom> #topic Open Discussion
20:28:29 <johnsom> Just trying to get things lined up for the merged project
20:29:08 <dougwig> cool.
20:29:25 <johnsom> perelman I'm pretty sure the "Failed to connect to bootstrap.pypa.io" issue is an infra issue.  A recheck has cleared it for me.  I'm not sure if they have a bug on it.
20:30:15 <johnsom> There have been a bunch of nasty DDoS attacks over the last two weeks that have been impacting some DCs, so it could just be fallout from that.
20:31:00 <johnsom> #link http://www.techradar.com/news/internet/here-s-how-security-cameras-drove-the-world-s-biggest-ddos-attack-ever-1329480
20:31:03 <dougwig> what is trying to connect to something other than the local pypi mirror?
20:31:13 <dougwig> is that dib?
20:31:28 <johnsom> OVH (which provides some hosts for infra) is called out in this article.  990Gbps, oye
20:31:40 <sbalukoff> Wow!
20:32:27 <dougwig> right, but if the job properly stays inside it's local bubble, it shouldn't matter.
20:33:10 <johnsom> #link http://logs.openstack.org/25/336925/14/check/gate-neutron-lbaasv2-dsvm-healthmonitor/0f6ff1c/logs/devstacklog.txt.gz#_2016-09-28_10_14_39_800
20:33:25 <johnsom> It looks like it is devstack, not DIB or our code
20:34:19 <sbalukoff> Has anyone heard from diltram lately?
20:34:23 <dougwig> "dib-run-parts' isn't dib ?
20:34:36 <sbalukoff> I'd really like to see his authentication patch merged sooner rather than later.
20:34:38 <johnsom> Yeah, I see that it's used down in DIB
20:34:46 <johnsom> sbalukoff He is travelling/vacation
20:34:56 <sbalukoff> Oh, ok.
20:35:03 <dougwig> namely "source-repository-pip-and-virtualenv"
20:36:54 <johnsom> dougwig Yep, I see it now.  Good catch.  Are you going to file a bug with them or want me to?
20:37:09 <dougwig> go for it
20:37:38 <johnsom> Hmm, actually it might just be something we need to override in our test setup.  hmmm, I will look into that.
20:38:07 <johnsom> We also need to make sure that that image isn't being built for the noop tests like the amp image is setup.
20:38:18 <johnsom> I see that gate was a noop test
20:39:14 <johnsom> Ok, I have notes on those two items.
20:39:30 <sbalukoff> Sweet.
20:39:32 <johnsom> Any other topics today?
20:39:40 <sbalukoff> Nothing from me.
20:40:14 <sbalukoff> Thanks folks!
20:40:18 <johnsom> dougwig there are two release related backports pending, can you kick those?
20:40:34 <dougwig> johnsom: yep, let me see if gertty is done updating yet
20:40:36 <johnsom> #link https://review.openstack.org/377995
20:40:45 <dougwig> johnsom: and it's not.  do you have links?
20:40:49 <johnsom> #link https://review.openstack.org/377994
20:41:18 <johnsom> Thanks.
20:41:32 <johnsom> Ok, I think that is it.  Thanks again for another good release.
20:41:50 <johnsom> #endmeeting