07:00:28 <prometheanfire> #startmeeting requirements
07:00:29 <openstack> Meeting started Wed Dec 13 07:00:28 2017 UTC and is due to finish in 60 minutes.  The chair is prometheanfire. Information about MeetBot at http://wiki.debian.org/MeetBot.
07:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
07:00:32 <openstack> The meeting name has been set to 'requirements'
07:00:36 <prometheanfire> tonyb, prometheanfire, number80, dirk, coolsvap, toabctl, smcginnis
07:00:39 <dirk> o/
07:00:39 <prometheanfire> #topic rollcall
07:00:40 <prometheanfire> o/
07:01:41 <toabctl> hi
07:01:41 <tonyb> \o
07:02:03 * tonyb is cooking dinenr so I might be a little laggy
07:02:07 <prometheanfire> tonyb: welcome back :P
07:02:17 <tonyb> \o/
07:03:20 <prometheanfire> #topic Any controversies in the Queue?
07:03:57 <dirk> the kubernetes thing?
07:04:14 <dirk> https://review.openstack.org/#/c/526925/
07:04:18 <dirk> #link https://review.openstack.org/#/c/526925/
07:04:26 <prometheanfire> I think a decision needs to be made about newton, it's dead and not gating (and not, on purpose, a part of the new zuul), I think we should close the open reviews for newton
07:04:45 <dirk> wfm
07:04:47 <prometheanfire> dirk: ya, I was hoping to get a better response, but upstream seems like they want to uncap, so I'll vote
07:04:59 <prometheanfire> dirk: the newton thing worksforyou?
07:05:08 <dirk> prometheanfire: sorry, yes, newton closing +1
07:05:24 <dirk> prometheanfire: I see there are new updates in the kubernetes changeset that I haven't consumed yet
07:05:29 * dirk preferred sleeping instead
07:05:58 <coolsvap> o/
07:06:05 <tonyb> prometheanfire: Yeah anything on newton shoudl just be abandoned as a matter of course
07:06:15 <tonyb> we can't kill the branch yet but newton is EOL
07:06:27 <prometheanfire> k, I'll do that, ya, I know we can't kill the old branches
07:06:44 <tonyb> that kubernetes this is potentially disruptive so we shoudl get buy-in from affects projects
07:07:15 <tonyb> also it'd be nice if we could do something better than just cp the websocket-client
07:07:21 <dirk> #agreed closing open reviews about stable/newton branch as the rest of newton is EOL and gating is no longer working
07:08:15 <prometheanfire> tonyb: think we should push upstream for a realease?  I think we should maybe a 4.0.1 or something
07:08:48 <tonyb> prometheanfire: We shoudl request it and/or research as to why it's there
07:08:58 <tonyb> they may have a compelling reason
07:09:26 <prometheanfire> no, they are in progress on uncapping it
07:09:46 <prometheanfire> https://github.com/kubernetes-incubator/client-python/issues/413
07:09:51 <prometheanfire> #link https://github.com/kubernetes-incubator/client-python/issues/413
07:10:04 <dirk> prometheanfire: can you add that link to the git commit message?
07:11:02 <dirk> https://github.com/kubernetes-incubator/client-python/commit/c4aac96342a1c3444b3eedf0a9da63353e25cf3d
07:11:12 <dirk> the commit message on why the cap there isn't very telling. but it says it has issues
07:11:42 <tonyb> prometheanfire: okay well then I guess we'ev done what we can.  Having the cap is gross but as long as we're confident we can get it removed before we cut queens
07:12:18 <prometheanfire> tonyb: I think so
07:12:51 <tonyb> cool
07:12:59 <tonyb> the last comment is pretty positive
07:13:18 <leyal> dirk , according to dims that was the issue for the cap  - https://github.com/kubernetes-incubator/client-python/issues/262 , i tried to reproduce it in all websocket-client without success ..
07:14:07 <prometheanfire> ok, I'll submit a PR refrencing it then
07:14:41 <dirk> leyal: great pointer!
07:14:48 <dirk> prometheanfire: looks like we want to include this link: https://github.com/kubernetes-incubator/client-python/pull/299
07:15:27 <prometheanfire> sure
07:15:41 <dirk> tonyb: I wonder -- is any project affected by the webclient cap that is not *also* using kubernetes?
07:16:04 <tonyb> dirk: give me 5 .....
07:16:07 * dirk can spent 15 minutes on scripting it to find the answer himself or just ask tony who probably has the magic script ready ;-)
07:17:18 <tonyb> dirk: I have the tool, but the data is out of date so it'll take a little while
07:17:29 <tonyb> or I could just be hacky ....
07:18:10 <prometheanfire> http://codesearch.openstack.org/?q=websocket-client&i=nope&files=.*requirements.*&repos=
07:18:13 <prometheanfire> that'd work
07:18:31 <prometheanfire> and it's not in setup.*
07:18:31 <openstackgerrit> Eyal Leshem proposed openstack/requirements master: Use kubernetes client  4.0.0  https://review.openstack.org/526925
07:18:40 <tonyb> [tony@thor openstack]$ grep websocket `grep -l kubernetes */*/*requirements.txt`
07:18:43 <tonyb> openstack/requirements/global-requirements.txt:websocket-client>=0.33.0  # LGPLv2+
07:18:46 <tonyb> openstack/rpm-packaging/global-requirements.txt:websocket-client>=0.33.0  # LGPLv2+
07:18:47 <prometheanfire> leyal: ping
07:18:49 <tonyb> openstack/rpm-packaging/requirements.txt:websocket-client>=0.33.0  # LGPLv2+
07:18:52 <tonyb> so the answer is no
07:19:07 <leyal> prometheanfire , pong
07:19:40 <tonyb> dirk: unless openstack/rpm-packaging/requirements.txt is a legit hit I thought it might be a false positive
07:20:48 <prometheanfire> leyal: we are talking about your patch :D
07:22:07 <leyal> prometheanfire , yep i am following , thanks :) , i wait for response in  https://github.com/kubernetes-incubator/client-python/issues/413 , but i didn't mange to reproduce the issue in all version that avialble in pypi..
07:22:31 <prometheanfire> leyal: ya, I'm making a PR to remove the cap now
07:23:26 <dirk> tonyb: thats a false positive
07:23:34 <dirk> tonyb: rpm-packaging has a full copy of requirements/*txt files
07:23:55 <tonyb> dirk: okay, that was more or less what I thought
07:23:56 <leyal> prometheanfire  , thanks - so the plan is to insert it now with the cap , and remove it when the PR will be accepted ?
07:24:02 <dirk> so agreement to move forward with the cap?
07:24:10 <prometheanfire> leyal: we'd also need an upstream release
07:24:11 <dirk> as it doesn't seem to affect anyone :)
07:24:19 <prometheanfire> dirk: yep, agreed
07:24:47 <dirk> #agreed insert a cap on webclient-socket until kubernetes 4.0.1 removes the cap
07:24:52 <dirk> tonyb: can you revisit your -1 ?
07:25:10 <tonyb> dirk: I only just added it.
07:25:50 <tonyb> dirk: I still think bumping 3 major releases in a single bound without reaching out to the 3 consumers would be a bad idea
07:26:51 <prometheanfire> leyal: do you mind sending a email to the mailing list asking for feedback from the other consumers of kubernets-client?
07:27:21 <dirk> tonyb: ah, yes, I agree
07:27:38 <leyal> prometheanfire, will do - need to ask about the cap or about the update for 4.0.0 ?
07:27:43 <dirk> to be fair the bump to 4.0 is something we're undoing in the uc update as of eternity already
07:28:00 <prometheanfire> no need to ask about the cap, more about the big jump in kube-client version
07:28:24 <prometheanfire> https://github.com/kubernetes-incubator/client-python/pull/416 btw
07:28:35 <tonyb> dirk: Yeah, and gettign to a modern library would be good.
07:29:03 <prometheanfire> ok, moving on then
07:29:13 <prometheanfire> any other controversies in the queue?
07:29:43 <dirk> not from me
07:29:54 <prometheanfire> ok, moving on
07:29:56 <prometheanfire> #topic PTG
07:30:25 <prometheanfire> nothing here, sent in the request for time and space, but no progress past that
07:31:13 <dirk> space an time is an illusion anyway ;-)
07:31:16 <dirk> and
07:31:39 <prometheanfire> something like that
07:32:00 <prometheanfire> and?
07:32:03 <prometheanfire> oh, sed
07:32:10 <prometheanfire> #topic open discussion
07:33:44 <prometheanfire> ok, closing in a couple of min
07:33:52 <dirk> One question: is anyone looking at why propose constraints update  is no longer working??
07:34:22 <dirk> It seems to be missing for this week
07:34:24 * tonyb didn't know it was broken ... clearly I've been distracted
07:34:26 <prometheanfire> bot not running?
07:34:31 <prometheanfire> ya, me too
07:34:39 <dirk> And i cannot find logfiles
07:34:53 <dirk> Well, zuul periodic run I think
07:35:38 <dirk> It used to be somewhere on logs.o.o but I can't see it there anymore
07:35:46 <tonyb> gimme 5
07:35:56 <prometheanfire> k, guess that's something to look into :|
07:36:14 <tonyb> http://logs.openstack.org/e3/e31a2077720bb7b95a16fb18d13011eeb592c126/post/
07:36:26 <tonyb> is the post pipeline for the current master
07:36:34 <tonyb> propose updates is there
07:36:39 <dirk> Sorry
07:36:51 <tonyb> oh propose *constraints* update
07:36:52 <prometheanfire> zuul has been having problems, maybe that messed with things the last couple of days
07:36:53 <dirk> I meant periodic generate constraints
07:36:59 <tonyb> that wont be in that list ....
07:37:28 <dirk> So tonyb takes up the action item?
07:37:49 <prometheanfire> welcome back, here's more work :P
07:37:57 <dirk> :-)
07:38:06 <tonyb> last periodic pipeline ... http://logs.openstack.org/periodic/git.openstack.org/openstack/requirements/master/
07:38:28 <tonyb> it looks like some how it's now omitted from the list of jobs
07:38:41 <prometheanfire> :|
07:38:51 <dirk> Ah, so that's the new location
07:39:05 <tonyb> dirk: Sure I'll look at it tomororow but zuulv3 is still outside my wheelhouse
07:39:38 <dirk> I can do some grepping during my meetings
07:39:59 <dirk> I will ping you if I find it before your tomorrow
07:40:11 <tonyb> dirk: cool
07:40:17 <prometheanfire> k
07:40:24 <prometheanfire> and I'll go to sleep :P
07:40:32 <prometheanfire> anything else?
07:40:47 <dirk> Nope
07:40:57 <prometheanfire> #endmeeting