15:00:47 <eglynn_> #startmeeting ceilometer
15:00:48 <openstack> Meeting started Thu Sep 25 15:00:47 2014 UTC and is due to finish in 60 minutes.  The chair is eglynn_. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:53 <openstack> The meeting name has been set to 'ceilometer'
15:01:17 <eglynn_> hey folks
15:01:25 <eglynn_> show of hands for the ceilo weekly?
15:01:25 <fabiog> o/
15:01:28 <llu-laptop> o/
15:01:29 <gordc> o/
15:01:42 <ildikov> _o_, more reading than writing (parallel meeting...)
15:02:22 <eglynn_> #topic Juno-RC1 status
15:02:25 <_nadya_> o/
15:02:35 <eglynn_> #link https://launchpad.net/ceilometer/+milestone/juno-rc1
15:02:44 <sileht> o/
15:02:50 <eglynn_> we've 5 bugs still outstanding
15:03:02 <eglynn_> I've bumped a few that didn't look like they'd make it
15:03:16 <DinaBelova> o/
15:03:30 <ityaptin> o/
15:03:40 <eglynn_> I'd like to delay RC1 until at least https://bugs.launchpad.net/bugs/1309555 and https://bugs.launchpad.net/bugs/1357869 are in
15:03:41 <uvirtbot> Launchpad bug 1309555 in ceilometer "Retry connection to mongodb on AutoReconnect exceptions" [Medium,In progress]
15:03:47 <gordc> i'm not sure we'll resolve this: https://bugs.launchpad.net/bugs/1348818
15:03:49 <uvirtbot> Launchpad bug 1348818 in keystone "Unittests do not succeed with random PYTHONHASHSEED value" [Medium,In progress]
15:04:09 <eglynn_> gordc: yeah, that one has been hanging around like a bad smell :)
15:04:24 <eglynn_> gordc: I'm not 100% sure it's still a kosher bug
15:04:37 <eglynn_> gordc: what's your feeling on it?
15:04:37 <DinaBelova> eglynn_, as for the mongo replica set retry - Igor has almost fixed it to your suggestion
15:04:47 <eglynn_> DinaBelova: nice! :)
15:04:49 <DinaBelova> he's testing it right now
15:04:51 <gordc> eglynn_: yeah... i'm sorta spinning wheels on that one so if anyone else has an idea, feel free to jump in
15:04:53 <DinaBelova> to be sure it works nice
15:05:21 <nealph_afk> o/
15:05:21 <eglynn_> OK, I'll try to take another look once I get the https://bugs.launchpad.net/bugs/1357869 squared away
15:05:23 <uvirtbot> Launchpad bug 1357869 in ceilometer "duplicate samples for network.services.lb.member" [Medium,In progress]
15:06:08 <eglynn_> (WIP patch is up for the sample dupes, needs a bit more polish and test coverage tho)
15:06:19 <eglynn_> what about https://bugs.launchpad.net/bugs/1372912 ?
15:06:20 <uvirtbot> Launchpad bug 1372912 in ceilometer "Unit tests at real HBase are too long" [Medium,In progress]
15:06:24 <nsaje-lunch> o/ sorry guys, was stuck in traffic
15:06:34 <_nadya_> https://review.openstack.org/#/c/115290/  looks good to me
15:06:51 <DinaBelova> eglynn_, for me as well
15:07:07 <eglynn_> _nadya_, DinaBelova: cool, use your super-powers! :)
15:07:10 <DinaBelova> just did not set +2 waiting gordc to take a look on :)
15:07:17 <_nadya_> gordc: what do you think?
15:07:18 <eglynn_> DinaBelova: a-ha, cool, got it
15:07:35 <DinaBelova> eglynn_, as gordc added several comments to it
15:07:40 <DinaBelova> with -1 :)
15:07:53 <eglynn_> DinaBelova: yep, I see that now
15:08:16 <gordc> _nadya_: i'm hesitant on that patch still to be honest... but i won't block since i don't have hbase set up
15:08:23 <eglynn_> OK, best to that discussion play out on gerrit I guess?
15:08:31 <DinaBelova> eglynn_, yeah
15:08:36 <_nadya_> I guess we have no chance to see HBase on devstack in Juno..
15:08:37 <gordc> if we can get that code tested in gate that's cool
15:08:54 <eglynn_> so the original plan was to cut RC1 by EoW
15:09:04 <DinaBelova> _nadya_, yeah - so that'll be just preparation to it
15:09:05 <eglynn_> but looks like we'll bleed into Monday
15:09:24 <gordc> _nadya_: i see... yeah than i'd prefer it not in.. (but again, i assume mirantis needs it in?)
15:10:15 <DinaBelova> gordc, actually this just gives us great opportunity to test all this structure together, not something special
15:10:42 <eglynn_> FYI Tuesday next week would be the drop-dead day for RC1, as all projects need to have a release candidate by end-of-month
15:10:59 <DinaBelova> eglynn_, a-ha, thanks!
15:11:11 <gordc> DinaBelova: gotcha... yeah, i'm cool with it if it actually gets tested... which i don't think it is right now.
15:11:18 <gordc> we can figure it out on gerrit though :)
15:11:24 <eglynn_> cool
15:11:30 <eglynn_> so let's gun for having everything we need landed on Monday if at all poss
15:12:04 <eglynn_> anything else on rc1?
15:12:28 <eglynn_> #topic TSDaaS/gnocchi status
15:12:47 <eglynn_> jd__: the floor is your's!
15:13:00 <jd__> yeah!
15:13:10 <jd__> so great progress on the archive policy front as it's ready to be merged
15:13:19 <jd__> the dispatcher saw great progress too thanks to sileht
15:13:33 <jd__> we have also support for some swift resources now
15:13:34 <eglynn_> cool, yeah nice work sileht!
15:13:43 <sileht> :)
15:13:48 <jd__> and well I'll still have things on the plate to do
15:13:53 <jd__> but everything's going great so far
15:14:31 <sileht> swift is better to generate a tone of sample :p
15:14:37 <sileht> (in devstack)
15:14:54 <eglynn_> so for now are we defering on the instance server_group attribute?
15:14:57 <eglynn_> (for alarming over heat autoscaling groups)
15:15:08 <jd__> eglynn_: we can add the attribute if you need it
15:15:14 <jd__> it's just one line of code :)
15:15:25 <eglynn_> yeah we can add that support in a patch subsequent to the dispatcher
15:15:29 <eglynn_> cool :)
15:16:06 <eglynn_> BTW amalagon is back and working on her custom aggregator patches again
15:16:31 <jd__> cool
15:16:35 <jd__> questions?
15:16:48 <eglynn_> she'll be using the new archive policy logic to introspect over the available granularities for the target entity
15:17:01 <eglynn_> ... first customer for that stuff :)
15:17:19 <jd__> nice
15:17:53 <eglynn_> sounds good, thanks jd__
15:17:59 <eglynn_> #topic Tempest status
15:18:14 <eglynn_> anything new in the world of Tempest?
15:18:22 <DinaBelova> eglynn_, really no progress from last week
15:18:29 <DinaBelova> at least what I know...
15:18:38 <DinaBelova> still the same kind of blocked thing
15:18:44 * eglynn_ needs to get up to speed on the latest re. the "libification" of Tempest
15:18:51 <DinaBelova> :D
15:19:03 <DinaBelova> yeah, that's true
15:19:11 <eglynn_> seeing as we'll likely be needing to use that in kilo
15:19:48 <DinaBelova> eglynn_, yeah, it looks like only one solution here...
15:20:09 <eglynn_> yep, the wind is very firmly blowing in that direction
15:20:25 <eglynn_> #topic Kilo summit planning
15:20:57 <eglynn_> so with the re-org on the summit format, the number of slot per session is way down on last time
15:21:29 <eglynn_> i.e. one day less of scheduled design session == 33% fewer slots to share out
15:21:37 <eglynn_> we had 10 slots in ATL
15:21:55 <eglynn_> looks like we'll only get 6 in Paris
15:22:03 <DinaBelova> wow
15:22:06 <nsaje> whew, better make them count then :)
15:22:30 <fabiog> eglynn_, these are 1hr slot?
15:22:39 <eglynn_> fabiog: 40mins
15:22:50 <DinaBelova> eglynn_, it looks like we'll have lots of joint sessions...
15:22:57 <eglynn_> yeap
15:23:05 * eglynn_ should fight to get back the missing 0.6666 slot ;)
15:23:30 <fabiog> eglynn_ I am wondering if it is better to have a "theme" per session or session and 1/2 so we can talk of related topics together
15:23:32 <DinaBelova> eglynn_, definitely you should :D
15:23:51 <eglynn_> fabiog: yeah, good point
15:24:02 <ildikov> eh, 6 does not sound too good, however I like that number in general :(
15:24:04 <DinaBelova> fabiog, yeah, after all the topics will be on place, it'll be possible
15:24:24 <eglynn_> so last we talked about using a google docs spreadsheet for proposals
15:24:36 <eglynn_> here's something simple to start ...
15:24:51 <eglynn_> #link https://docs.google.com/spreadsheets/d/1ISbSNe9kojkONFPIHH2oc4pkt3qunKzTPldfd97qJOg/edit?usp=sharing
15:25:06 <eglynn_> feel free to hack on the format
15:25:21 <eglynn_> (as well as adding actual proposals)
15:25:58 <eglynn_> the "Requires a full slot?" column is just a (y/n) to make it easier to find candidates for merging
15:26:31 <llu-laptop> eglynn_: what does theme mean?
15:27:09 <fabiog> llu-laptop, it could be a general topic like improving performance
15:27:10 <eglynn_> llu-laptop: Theme column similarly was meant to help identify similar topics to merge (e.g. gnocchi, SNMP, testing, performance, etc.)
15:27:27 <llu-laptop> ok
15:27:53 <eglynn_> so in general, I'd expect the proposer and the leader to be the same person
15:28:22 <eglynn_> and the session leader to be implicitly signing up to drive the implementation work
15:29:08 <eglynn_> but maybe we'll have some cases where an idea is thrown out on the table, to see if anyone will bite in terms of running with it
15:29:29 <eglynn_> anyhoo, happy hacking on that spreadsheet! :)
15:29:56 <ildikov> eglynn_: thanks for this docco!
15:30:18 <eglynn_> when should we discuss selecting proposals and matching slots?
15:30:28 <eglynn_> the weekly meeting after next maybe?
15:30:35 * jd__ wrote a first line to show example
15:30:42 <DinaBelova> jd__, hehe, "required a full slot? Variant: y/n" - "Maybe"
15:30:46 <DinaBelova> :)
15:30:50 <eglynn_> jd__: cool :)
15:31:07 <jd__> :-D
15:31:37 <eglynn_> so is two weeks long enough for folks to get their proposals in?
15:31:46 <DinaBelova> eglynn_, I suppose so
15:31:49 <jd__> should be
15:32:15 <jd__> I hope people that will propose stuff to discuss will actually work on them later :)
15:32:23 <ildikov> eglynn_: two weeks from today?
15:32:33 <eglynn_> ildikov: yep, that work?
15:33:03 <eglynn_> jd__: yeah, if you nominate yourself as leader for the topic, that would be the expectation all right
15:33:15 <ildikov> eglynn_: will try my best, but well more time is almost always better ;)
15:33:29 <eglynn_> #info aim to select design summit topics and match to sessions at Oct 9th weekly meeting
15:34:10 <eglynn_> note of course we also have the full day Friday for the "contributors meetup" (i.e. the pod++)
15:34:49 <eglynn_> ... despite Friday traditionally being the day everyone at summit is zombified ;)
15:35:24 <eglynn_> #topic Open discussion
15:36:14 <eglynn_> anyone familar with this "Win The Enterprise" management working group?
15:36:28 <eglynn_> seems to have come from a session in ATL
15:36:47 <eglynn_> #link https://wiki.openstack.org/wiki/Win_the_Enterprise
15:37:02 <gordc> eglynn_: never heard of it until you mentioned it this morning
15:37:15 <DinaBelova> eglynn_, as well for me
15:37:31 <eglynn_> I have a call with them later, for some reason they're interested in zabbix integration (as mentioned on the #os-ceilo channel earlier)
15:37:47 * gordc see intel email
15:38:02 <gordc> llu-laptop: you should know right? intel is a small company :)
15:38:30 <eglynn_> #link https://blueprints.launchpad.net/ceilometer/+spec/zabbix-agent-adoption
15:39:05 <llu-laptop> gordc: I've heard some gruops in intel is working on enterprise readiness, but don't know this :(
15:39:38 <ildikov> hmm, interesting...
15:40:08 <ildikov> eglynn_: was this topic discussed today on the channel?
15:40:12 <eglynn_> anyhoo, I'll try to smoke the motivation for using zabbix in this way (i.e. as an in-VM agent, as opposed to polling the zabbix API from the central agent)
15:40:39 <eglynn_> ildikov: yeah I mentioned it earlier, discussed briefly with gordc
15:41:03 <eglynn_> ildikov: note that it's a different approach to the one you -2'd on ceilo-specs
15:41:06 <gordc> ildikov: yeah. check logs for my name.
15:41:13 <gordc> ildikov: i called you out :)
15:41:31 <ildikov> eglynn_: sure, I quickly checked the BP you copied the link here of
15:41:37 <gordc> i blamed everything on you.
15:41:56 <ildikov> gordc: sorry, today I had a visit at the doctor... :(
15:42:11 <ildikov> gordc: so I'm already suffering for my sins no worries ;)
15:42:33 <gordc> ildikov: lol. it's not my fault.
15:42:45 <ildikov> eglynn_: gordc: anyway, thanks, I will check the BP and the logs also later today
15:43:17 <ildikov> gordc: really, I thought that it's some woodoo thing that's going on with me now :)
15:44:00 <gordc> ildikov: i cannot confirm or deny anything.
15:44:26 <ildikov> gordc: I thought so :)
15:44:30 <gordc> sidenote, is there any links to docs we added in juno?
15:45:07 <eglynn_> #link http://docs.openstack.org/admin-guide-cloud/content/ch_admin-openstack-telemetry.html
15:45:08 <ildikov> eglynn_: will your call be about this Zabbix integration stuff or it is just one topic from many?
15:45:25 <eglynn_> ildikov: just one topic from many on the agenda IIRC
15:45:28 <ildikov> eglynn_: thanks for the link :)
15:45:33 <gordc> eglynn_: cool cool. thanks
15:45:58 <ildikov> eglynn_: ok, cool, let me know if you herad there anything interesting please :)
15:46:12 <eglynn_> ildikov: absolutely!
15:46:59 <eglynn_> have we run out of stuff to chat about? :)
15:47:23 <eglynn_> cool enough, let's call it a wrap
15:47:26 <gordc> eglynn_: nothing from me.
15:47:33 <eglynn_> thanks folks for a productive meeting!
15:47:38 <gordc> laters
15:47:41 <eglynn_> #endmeeting ceilometer