15:00:02 <ricolin> #startmeeting heat
15:00:03 <openstack> Meeting started Wed Jun 28 15:00:02 2017 UTC and is due to finish in 60 minutes.  The chair is ricolin. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:06 <openstack> The meeting name has been set to 'heat'
15:00:10 <ricolin> #topic roll call
15:01:05 <zaneb> hallo
15:01:09 <ramishra> Hey
15:01:11 <ricolin> O/
15:01:15 <kiennt> O/
15:01:29 <ricolin> HI kiennt
15:01:45 <kiennt> Hi, ricolin
15:02:04 <ricolin> #topic adding items to agenda
15:02:10 <ricolin> #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282017-06-28_1500_UTC.29
15:02:39 <ricolin> I don't have much to talk this week except for please do more review:)
15:03:46 <ricolin> #topic Weekly report
15:05:03 <ricolin> So we have face two gate issue this week, and now we already propose to turn off any nova quota cheeck for gate
15:05:13 <ricolin> and landed
15:05:38 <therve> They reverted the problematic change too, right?
15:05:46 <ramishra> therve: yeah
15:05:59 <ramishra> it broke many projects
15:06:04 <ricolin> yep
15:06:32 <ramishra> including python-novaclient jobs itself;)
15:06:43 <ricolin> but I think we still keep the quota out right?:)
15:07:04 <ramishra> yeah, it would not hurt
15:07:30 <ramishra> therve: It seems to break the amqp1 job badly, did you notice that?
15:07:59 <ramishra> I mean that devstack patch, not it's ok after the revert
15:08:05 <ramishra> s/not/now
15:08:22 <therve> ramishra, Yeah, on it
15:08:30 <therve> https://bugs.launchpad.net/oslo.messaging/+bug/1700835
15:08:32 <openstack> Launchpad bug 1700835 in oslo.messaging "[AMQP 1.0] vhost is ignored when using message routing" [Undecided,New] - Assigned to Ken Giusti (kgiusti)
15:08:45 <ramishra> therve: great:)
15:09:49 <ricolin> move on
15:10:03 <ricolin> #topic Testing rolling upgrade report (The guideline: https://review.openstack.org/#/c/475853/)
15:10:42 <kiennt> hi, i'm Kien from Fujitsu VietNam. This week, i setup the Kolla cluster and testing heat rolling upgrade in it
15:10:52 <kiennt> follow the patchset: https://review.openstack.org/#/c/475853/
15:11:55 <kiennt> During rolling upgrade, i send several async requests to heat-api (Create, List and Delete stack). Everything works fine, without downtime
15:12:33 <kiennt> You can info about test enviroment, scripts here: https://github.com/ntk148v/heat-rolling-upgrades
15:12:55 <kiennt> That's all about testing rolling upgrade.
15:13:52 <therve> kiennt, Do you think that can/should in CI somewhere?
15:13:52 <zaneb> kiennt: this is very cool work, thanks!
15:14:02 <therve> It sounds more useful than grenade
15:14:10 <ramishra> yeah, I think there is a governance requiment to test rolling upgrade in the gate
15:14:13 <ricolin> kiennt, so the test environment is ocata and newton?
15:14:37 <kiennt> ricolin: yep, newton to ocata.
15:15:27 <ricolin> kiennt, and like therve said do you think that you can help to put this in heat's gate so we can make sure about this all the time?:)
15:16:03 <kiennt> ricolin, therve: yes, i can, it's my main work now
15:16:13 <therve> Cool
15:16:16 <ricolin> kiennt, so cool!
15:16:51 <kiennt> i will diccuss with my PL about it in tomorrow.
15:17:11 <ramishra> newton to ocata? I thought we added number of db changes in ocata?
15:18:18 <ricolin> I think that test environment is newton->ocata because you only can get kolla to work with those version right?
15:18:36 <kiennt> ramishra: i backup the database schema before and after rolling upgrade, to compare: https://github.com/ntk148v/heat-rolling-upgrades/tree/master/db_backup_schema
15:19:01 <kiennt> ricolin: yes, i setup Kolla in these version.
15:19:40 <ramishra> so the old engines and new engines both work with new schema without any issues then??
15:19:43 <kiennt> May be the next days i will re-setup and test rolling upgrade from ocata -> pike
15:20:01 <kiennt> ramishra: yeah, everything seems fine
15:21:22 <ricolin> kiennt, That sounds perfect:)
15:21:52 <kiennt> ramishra: OSM using the simple strategy of 'never remove, never alter, only add'
15:22:02 <kiennt> so a new schema change is understandable by both old and new engines.
15:23:20 <ramishra> kiennt: yeah, that's the assumption, but we did some work around properties_data stuff in ocata I assume, would be pleasanty surprised if there are no issues:)
15:24:35 <ramishra> We would need some gate testing though as discussed earlier
15:25:11 <kiennt> ramishra: During my test, everything works fine. I will re-test it.
15:25:21 <ricolin> That will be awesome:)
15:25:45 <kiennt> About the gate testing, i will diccuss with my PL and give feedback asap :)
15:26:21 <ramishra> kiennt: I'm by no means saying it would not work:) Thanks for working on it.
15:26:55 <ricolin> kiennt, thanks I think we really need that if we can check on the rolling upgrade from each release to next
15:27:17 <kiennt> ramishra: Thank you.
15:27:24 <kiennt> ricolin: Sure, i will do it.
15:27:33 <ricolin> Okay, move on ?:)
15:27:57 <kiennt> yep, thanks
15:28:01 <ricolin> #topic Open discussion
15:29:23 <ricolin> we're approach to feature freeze, so would like cores to put more review on BPs if you can:)
15:29:50 <ramishra> so the neutron trunk plugin has landed
15:30:36 <therve> https://review.openstack.org/#/c/473700/ needs in
15:31:22 <ramishra> therve: probably the functional tests can land later.. take it's own time for review
15:31:38 <zaneb> I have a very long series that needs review, starting here: https://review.openstack.org/#/c/441475/
15:31:53 <therve> ramishra, Well, that's what going to happen anyway? :)
15:32:02 <therve> We need it to land in the release, though
15:32:23 <ramishra> yeah, I'm not that worried about the functional tests
15:32:28 <ramishra> therve: sure
15:32:34 <ricolin> ramishra, if you can give it a review and make sure about it, than I will give it a +1:)
15:34:29 <ramishra> I assume it doesn't test much other than creating/updating a trunk. It's not being used
15:35:51 <ramishra> zaneb: I somehow managed to get around the py3 issue for https://review.openstack.org/#/c/466802/
15:36:52 <zaneb> ramishra: oh cool, I will take another look
15:37:05 <ricolin> spoke aboutpy35
15:37:23 <ricolin> zaneb, please review https://review.openstack.org/#/c/425057/ :)
15:38:21 <zaneb> ricolin: ack
15:38:26 <ricolin> zaneb, thx:)
15:39:24 <ricolin> Also I would like to propose that we introduction some online video meeting tools like zoom?
15:40:10 <zaneb> I'm not familiar with zoom
15:40:18 <ricolin> so we can get use about setting up a global virtual meeting, and might be one of the format for PTG?
15:40:30 <ricolin> zaneb, any familiar tools?
15:40:43 <ricolin> zoom is kind of stable and easy to use
15:40:44 <zaneb> but there are occasions when being able to do e.g. a review walkthrough by video would be useful
15:41:07 <zaneb> as long as it is open to everyone in the community (i.e. not google hangouts)
15:41:29 <LanceHaig> +1 from me
15:41:42 <ricolin> LanceHaig :)
15:42:37 <ricolin> Maybe we can try to setup a zoom for next meeting and try to figure that works for all or not?
15:42:41 <therve> zaneb, I suspect you still need an account there, no?
15:42:49 <therve> Not sure what you mean by open
15:43:51 <LanceHaig> You will need a zoom account
15:43:56 <zaneb> therve: I mean I can't join google hangouts ;)
15:44:09 <ricolin> zaneb, why may I ask?
15:44:27 <zaneb> ricolin: apparently they don't believe I'm a real person
15:44:42 <zaneb> also, I don't want to :)
15:44:46 <ricolin> zaneb, but you're not!
15:44:50 <ricolin> zaneb, lol
15:44:54 <zaneb> :P
15:45:38 <ricolin> hangout is actually one of the top choice if zoom is not work for all:)
15:46:02 <LanceHaig> I would like to get some help as to why this review is failing pep8 please https://review.openstack.org/#/c/474953/
15:46:07 <ricolin> but anyway, I think we can just pick one and figure if that work for all
15:46:35 <LanceHaig> oops sorry
15:46:41 <LanceHaig> wrong channel
15:48:12 <zaneb> LanceHaig: http://logs.openstack.org/53/474953/1/check/gate-heat-pep8-ubuntu-xenial/f6f7f49/console.html.gz#_2017-06-16_12_08_39_999499 there is no .iteritems() in python3, so use .items() or six.iteritems()
15:49:52 <ricolin> LanceHaig, ^^^
15:50:00 <LanceHaig> thanks
15:50:03 <LanceHaig> sorry
15:50:31 <ricolin> LanceHaig, NP, that's actually something that you're welcome to put it in meeting:)
15:50:59 <ricolin> Anyway, anything that we should discussed about?:)
15:52:12 <ricolin> If no, I think we can call this meeting off:)
15:53:05 <ricolin> Thanks all for join
15:53:09 <ricolin> #endmeeting