16:00:01 <smcginnis> #startmeeting Cinder
16:00:01 <openstack> Meeting started Wed Nov 30 16:00:01 2016 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:05 <openstack> The meeting name has been set to 'cinder'
16:00:19 <smcginnis> ping dulek duncant eharney geguileo winston-d e0ne jungleboyj jgriffith thingee smcginnis hemna xyang1 tbarron scottda erlon rhedlind jbernard _alastor_ bluex patrickeast dongwenjuan JaniceLee cFouts Thelo vivekd adrianofr mtanino yuriy_n17 karlamrhein diablo_rojo jay.xu jgregor baumann rajinir wilson-l reduxio wanghao thrawn01 chris_morrell stevemar watanabe.isao,tommylike.hu mdovgal ildikov
16:00:23 <e0ne> hi
16:00:26 <jungleboyj> o/
16:00:27 <winston-1> o/
16:00:28 <geguileo> hi! o/
16:00:29 <smcginnis> Hey everyone
16:00:31 <xyang1> hi
16:00:32 <eharney> hi
16:00:33 <viks_> hi
16:00:41 <hemna> yough
16:00:42 <patrickeast> Hi
16:00:43 <erlon-airlong> hey
16:01:04 <smcginnis> #topic Announcements
16:01:13 <smcginnis> The usual:
16:01:13 <mdovgal> hi
16:01:14 <smcginnis> #link https://etherpad.openstack.org/p/cinder-spec-review-tracking Review focus
16:01:19 <Cibo> hello
16:01:32 <smcginnis> Please pay attention to those driver patches.
16:01:38 <guyr-infinidat> Hi
16:01:38 <diablo_rojo> Hello :)
16:01:48 <scottda> hi
16:01:58 <tbarron> hi
16:02:14 <erlon-airlong> smcginnis: just an explanation about the Hitachi FC CI
16:02:14 <smcginnis> Other than that, I just wanted to mention the Bugsmash event is going on right now.
16:02:35 <hemna> bugsmash?
16:02:43 <erlon-airlong> smcginnis: we have 2 patches for the iSCSI and FC driver and the CI reports for both drivers
16:02:44 <smcginnis> So if you have extra cycles to review any of those bug fix patches while the event is taking place, we've got some dedicated attention right now.
16:02:47 <hemna> everyone walks around squishing bugs outside?
16:02:51 <hemna> oh the horror
16:02:53 <smcginnis> hemna: Hah
16:03:06 <e0ne> hemna: :)
16:03:08 <smcginnis> hemna: Huawei and Intel's event they sponsor each cycle. ;)
16:03:15 <hemna> :)
16:03:16 <erlon-airlong> smcginnis: so, in the first patch of the series, theres no iSCSI driver and the CI will fail
16:03:29 <scottda> Are you at the bugsmash smcginnis ? In China?
16:03:31 <smcginnis> erlon-airlong: But the FC one passes?
16:03:34 <hemna> erlon-airlong, urls ?
16:03:34 <smcginnis> scottda: Yep
16:03:41 <erlon-airlong> smcginnis: yes
16:03:46 <diablo_rojo> scottda, Of course he is
16:03:53 <smcginnis> erlon-airlong: OK, that makes snese then.
16:03:54 <smcginnis> *sense
16:03:56 <erlon-airlong> hemna: the first driver in the etherepad
16:04:15 <smcginnis> I thought at first it meant none of it would pass. THat's fine then.
16:04:30 <hemna> erlon-airlong, thanks
16:04:31 <winston-d> smcginnis: oh, so we are in same timezone now. :)
16:04:49 <erlon-airlong> smcginnis: oh no, the FC passes in the first patch, both passes in the second
16:04:51 <smcginnis> winston-d: I was going to say - I'm impressed you actually make it to these meeetings. ;)
16:05:02 <smcginnis> erlon-airlong: +1
16:05:12 <smcginnis> #topic Using py34 vs py35
16:05:31 <smcginnis> Movinng along, we have environments for both py34 and py35 in places.
16:05:42 <smcginnis> I believe we are ultimately trying to get to py35.
16:05:54 <smcginnis> And that is what is included by defaultnow in xenial.
16:05:58 <hemna> fwiw, I had to manually build py34 for ubuntu 16.04, from source
16:05:59 <eharney> is this about gate jobs?    tox -e py35 works now
16:06:22 <smcginnis> I waned to bring up the idea if we want to get rid of py34 for default local tox jobs and just do 35.
16:06:24 <Swanson> hi
16:06:27 <hemna> py34 isn't available from canonical for 16.04 afaik
16:06:33 <smcginnis> I would expect them to catch the same issues.
16:06:44 <hemna> smcginnis, do we still support 14.04 ?
16:06:44 <eharney> 3.4 is already gone from Fedora too
16:06:47 <smcginnis> hemna: Yep, and e=hence the point of my frustration. ;)
16:07:07 <hemna> FYI, this works http://devmartin.com/blog/2016/04/creating-a-virtual-environment-with-python3.4-on-ubuntu-16.04-xenial-xerus/
16:07:11 <smcginnis> hemna: Yes, but easier to add 35 to 14 than 34 to 16
16:07:24 <hemna> but IMHO, we shouldn't have to do that unless we are still trying to support 14.04
16:07:25 <smcginnis> hemna: Oh, nice reference!
16:07:45 <e0ne> hemna: +1
16:08:01 <hemna> smcginnis, probably should ask this question in the cross project and see what other projects are doing w/ py34
16:08:03 <smcginnis> I would think the 35 job would catch anything that would break in 34 in 99% of cases.
16:08:23 <smcginnis> hemna: Well, I wouldn't remove the gate job, just the locally run stuff.
16:08:30 <tbarron> fedora25 has py34 and py35 now getting rid of py34 makes sense I think
16:08:44 <smcginnis> But in general that may be a good cross project discussion too.
16:09:16 <smcginnis> OK, I may put up a patch to remove locally run 34 stuff then since it seems most (so far) are agreeable.
16:09:20 <tbarron> was going to propose just 27 and 35 in manila once I realized xenial comes with 35
16:09:22 <mdovgal> i have some problems with starting tox/py35 on my local vm(
16:09:27 <smcginnis> tbarron: +1
16:09:32 <hemna> tbarron, +1
16:09:32 <jungleboyj> smcginnis: I don't have any concerns.
16:09:36 <smcginnis> mdovgal: Have you installed 35?
16:09:39 <mdovgal> p
16:09:47 <jgriffith_away> death to 27
16:09:52 <smcginnis> jgriffith_away: Hah!
16:09:56 <winston-d> wait, we'd drop the support for 14.04?
16:09:58 <hemna> :)
16:10:02 <smcginnis> jgriffith: Soon if things go according to plan.
16:10:10 <diablo_rojo> jgriffith lol
16:10:14 <jgriffith> winston-d: no, they were asking that as a question
16:10:15 <mdovgal> i have 3.5.2 version with 16.04 ubuntu
16:10:18 <smcginnis> winston-d: No, just py34 for non-gate tests.
16:10:37 <hemna> I wasn't suggesting it, just asking
16:10:48 <winston-d> Oh, man. We've just finished upgrading from 12.04 to 14.04.
16:10:55 <jgriffith> winston-d: :)
16:10:55 <hemna> lolz
16:10:59 <smcginnis> OK, well I just wanted to bring it up for now. If/when I put up a patch, feel free to express concerns on that.
16:11:02 <smcginnis> winston-d: :)
16:11:15 <smcginnis> #topic NEC Third party CI
16:11:28 <smcginnis> stakeda: You're up. :)
16:11:32 <stakeda> hi
16:11:51 <smcginnis> stakeda: Did you have a question about CI?
16:12:18 <stakeda> Yes, we are testing with sandbox and it is almost done
16:12:35 <stakeda> is there any requirements to get approval?
16:13:06 <smcginnis> stakeda: Nope, just once you think it is stable, have it start testing against the cinder patches.
16:13:22 <smcginnis> stakeda: Disable reporting until you know things are working right.
16:13:53 <smcginnis> stakeda: No other approval needed to have it test and comment on cinder patches.
16:14:09 <stakeda> okey, i will start reporting when i think our ci is stable
16:14:31 <smcginnis> Oopos
16:14:48 <smcginnis> Sorry, terminal issues on my end.
16:15:01 <smcginnis> stakeda: Great, feel free to ping us in channel if you have questions.
16:15:13 <stakeda> thanks
16:15:29 <jgriffith> smcginnis: You're not dieing are you?
16:15:42 <smcginnis> geguileo: Sorry, going to give diablo_rojo a quick minute for the ptg.
16:15:42 <jgriffith> "terminal issues" doesn't sound good at all
16:15:56 <geguileo> smcginnis: np
16:15:56 <smcginnis> jgriffith: Hah!
16:16:01 <diablo_rojo> smcginnis, please dont die
16:16:08 <smcginnis> #topic PTG announcement
16:16:08 * jungleboyj would be sad
16:16:18 <smcginnis> diablo_rojo: OK, have at it.
16:16:32 <diablo_rojo> So, as of now we only have like a few CInder people registered
16:16:52 <erlon-airlong> diablo_rojo: do you have numbers?
16:16:53 <diablo_rojo> Prices are gonna go up sometime in January just a heads up, but what I'm wondering is what is holding people back?
16:16:57 * smcginnis imagines four of us sitting in a ballroom playing cards
16:17:04 <diablo_rojo> 6 cinder people have registered
16:17:09 <scottda> I don't have approval from my company yet
16:17:16 <diablo_rojo> ameade, smcginnis patrickeast eharney dulek  and jungleboyj
16:17:29 <xyang> diablo_rojo: I registered too
16:17:30 <diablo_rojo> scottda, I figured that was the case for you and hemna
16:17:35 <diablo_rojo> Oh yes
16:17:41 <erlon-airlong> diablo_rojo: scottda: same
16:17:47 <diablo_rojo> xyang, you were under manila and CInder
16:17:54 <diablo_rojo> sorry :)
16:17:55 * jungleboyj didn't have approval.  :-)
16:18:04 <xyang> diablo_rojo: no problem:)
16:18:06 <diablo_rojo> jungleboyj is a rebel
16:18:14 <diablo_rojo> So, jgriffith ?
16:18:16 <jgriffith> diablo_rojo: I wasn't able to get the registration to work when I tried it... haven't tried again lately
16:18:25 <diablo_rojo> It should be good now
16:18:28 <erlon-airlong> diablo_rojo: how many people used to attend the midcicle?
16:18:35 <jgriffith> diablo_rojo: I'll give it a go after the meeting
16:18:39 <jgriffith> diablo_rojo: thanks!
16:18:41 <scottda> erlon-airlong: 20-25
16:18:48 <diablo_rojo> jgriffith, sweet, I'll bring the skateboard
16:18:51 <jungleboyj> What scottda said.  :-)
16:19:00 <smcginnis> Please harass your management to get approval. For it to be successful we need folks there.
16:19:08 <diablo_rojo> smcginnis, +1
16:19:15 <erlon-airlong> scottda: hmm, Id say we can spect the same number in the PTG?
16:19:15 <jungleboyj> Or be a rebel
16:19:19 <diablo_rojo> THere is also a travel support program you can take advantage of
16:19:27 <diablo_rojo> erlon-airlong, yeah thats the hope
16:19:28 <e0ne> I didn't have approval yet too:(
16:19:32 <smcginnis> (or we can all meetup up in Ft Collins on our own)
16:19:33 <Cibo> is it about some sort of an event?
16:19:36 * smcginnis i kidding
16:19:46 * smcginnis is mostly kidding
16:19:51 <diablo_rojo> e0ne, you can apply for the TSP and if you dont get funding you can have that as a backup
16:19:56 <scottda> Ha. Ft Collins is Cold in Jan
16:20:04 <diablo_rojo> e0ne, and if you do get funding you can just let them give it to someone else
16:20:12 <erlon-airlong> Cibo: yes, it a design meeting
16:20:16 <diablo_rojo> scottda, cold? Not like MN
16:20:19 <e0ne> diablo_rojo: thanks for the hint
16:20:20 <mtanino> Cibo: https://www.openstack.org/ptg/#tab_faq
16:20:25 <smcginnis> #link http://www.openstack.org/ptg PTG Info
16:20:27 <scottda> Cold, but Sunny
16:20:33 <diablo_rojo> e0ne, no problem :)
16:20:47 <diablo_rojo> scottda, and hemna go harass managers :)
16:20:52 <xyang> diablo_rojo: travel support seems to be difficult to get?  hemna said he didn't get approved last time
16:21:08 <jgriffith> diablo_rojo: I'll bring mine :)  https://www.muirskate.com/photos/products/2352/product_Comet_TMKF__set_.png
16:21:09 <diablo_rojo> So far there have only been 27 requests for the PTG
16:21:30 <diablo_rojo> jgriffith, Oh please do :)
16:21:37 <Swanson> Presumably mid cycle attendance is a leading indicator of corporate support.
16:21:46 <hemna> oh yah ptg...
16:21:46 <diablo_rojo> xyang, they have more spots for this than they did for the summit
16:22:06 <xyang> diablo_rojo: ok, thanks
16:22:13 <diablo_rojo> xyang, no problem :)
16:22:30 <smcginnis> diablo_rojo: OK, let's move on. Thanks!
16:22:38 <diablo_rojo> so, moral of the story, GO REGISTER FOR THE PTG
16:22:42 <diablo_rojo> All done :)
16:22:45 <diablo_rojo> Thanks smcginnis :)
16:22:46 <smcginnis> #link Replication patches
16:22:49 <smcginnis> diablo_rojo: Thanks.
16:22:50 <Cibo> 1 sec
16:22:55 <smcginnis> geguileo: OK, all yours.
16:23:00 <Cibo> small talk aboun JovianDSS
16:23:02 <geguileo> thanks
16:23:14 <geguileo> I wanted to report the progress on the replication stuff we discussed in Barcelona.
16:23:22 <geguileo> There are now a good number of patches up for review (bug fixes, specs, devref update) with only the Secondary Promotion left to do.
16:23:24 <smcginnis> Cibo: We can talk about it at the end of the agenda.
16:23:30 <Cibo> ok
16:23:36 <hemna> registering costs money
16:23:40 <hemna> so, it's not as simple as that
16:23:40 <geguileo> They are all in order in the etherpad: https://etherpad.openstack.org/p/cinder-spec-review-tracking
16:23:56 <smcginnis> geguileo: Great, that really helps.
16:23:57 * hemna frowns on making devs pay to attend a meeting about an open source project
16:23:59 <patrickeast> geguileo: awesome!
16:24:27 <geguileo> The secondary promotion will get a little delayed because I have to get back to the A/A patches
16:24:47 <geguileo> That was mostly it, let people know they are there, including the freeze mechanism fix
16:25:04 <geguileo> And I would also like to beg for reviews on the RBD replication patch, that got FFE approved and still has no core reviews: https://review.openstack.org/333565
16:25:14 <geguileo> pleaaaaaaaaaaaaaaaaaase
16:25:19 <smcginnis> geguileo: Is that on the etherpad?
16:25:20 <smcginnis> :)
16:25:24 <geguileo> smcginnis: yes
16:25:29 <smcginnis> geguileo: OK, great.
16:25:51 <geguileo> It would be "fun" if I had to request another FFE in this cycle for the same patch
16:25:52 <geguileo> lol
16:25:55 <e0ne> geguileo: I'm going to test it again this week
16:26:01 <geguileo> e0ne: Thanks!!
16:26:03 <patrickeast> geguileo: I'd you want I can maybe help with the promotion stuff, I've got that on my backlog to improve too
16:26:15 <geguileo> patrickeast: That would be great!
16:26:25 <smcginnis> geguileo: Want to move into the next topic?
16:26:28 <geguileo> patrickeast: We can discuss it in the cinder channel
16:26:36 <geguileo> smcginnis: Sure
16:26:37 <patrickeast> geguileo: sounds good
16:26:41 <smcginnis> #topic Failure codes
16:27:06 <smcginnis> #link https://github.com/openstack/api-wg/blob/master/guidelines/http.rst#failure-code-clarifications Failure guidelines
16:27:13 <geguileo> So we are not following the HTTP guidelines
16:27:20 <geguileo> In terms of failure codes
16:27:33 <geguileo> And in this review https://review.openstack.org/401941
16:27:50 <smcginnis> Tricky thing now is we need to microversion it to change it. :/
16:27:50 <geguileo> It was asked, do we really want to make all the changes required to adhere to it?
16:27:52 <erlon-airlong> geguileo: what is the backward  compatibility issues we can have if we change?
16:28:13 <smcginnis> Too bad we didn't identify this before creating v3.
16:28:15 <geguileo> erlon-airlong: It's mosly returning 400 instead of 404
16:28:16 <erlon-airlong> geguileo: can we return both? :)
16:28:21 <scottda> geguileo: I think we should move in the direction of adhering to the guidelines
16:28:23 <geguileo> And returning 403 instead of 413
16:28:24 <eharney> we should just do what the api-wg documents as the right thing
16:28:25 <geguileo> I think
16:28:26 <e0ne> erlon-airlong: no
16:28:42 <geguileo> erlon-airlong: XD XD XD
16:29:02 <geguileo> erlon-airlong: That would be great, but like e0ne said, it's not possible
16:29:04 <e0ne> eharney: I agree with you. we should follow api-wg guidelines
16:29:04 <erlon-airlong> e0ne: geguileo: my concern is people that have scripts based on that
16:29:16 <eharney> erlon-airlong: that's solved by microversions
16:29:20 <geguileo> erlon-airlong: I know, that's always the main problem  :-(
16:29:27 <scottda> erlon-airlong: We will microversion the changes. Old scripts will still work.
16:29:34 <geguileo> eharney: Yup
16:29:44 <scottda> don't ask for new microversion, you get the old return codes.
16:29:50 <erlon-airlong> eharney: hmmm, so, that solve the problem
16:29:54 <geguileo> We can try to gather all related changes into the same microversion to limit the crazyness of it all
16:29:56 <scottda> Do ask for new microversion, you get New return codes
16:30:17 <scottda> geguileo: About how many changes are there?
16:30:20 <smcginnis> geguileo: +1
16:30:25 <geguileo> scottda: Right now only 1
16:30:30 <erlon-airlong> geguileo: yeah one patch for all change
16:30:39 <geguileo> scottda: But we can add all other changes to that one
16:30:44 <erlon-airlong> geguileo: is there any changes in the CLI?
16:30:57 <geguileo> And for the Quota stuff I think it will be mostly just changing the conde in the exception
16:31:10 <geguileo> (although I haven't checked it)
16:31:19 <geguileo> erlon-airlong: I don't think it should
16:31:31 <geguileo> erlon-airlong: Because the CLI usually just returns whatever error it gets
16:31:49 <geguileo> I don't think we are doing fancy stuff with status codes 404 or 413
16:32:08 <geguileo> But it would also be solved with microversions if a change is required
16:32:27 <erlon-airlong> geguileo: yeap
16:32:44 <geguileo> So we all agree we should try to adhere to the guidelines
16:32:51 <scottda> +1
16:32:53 <smcginnis> geguileo: Sounds good to me.
16:32:55 <winston-d> +1
16:32:55 <geguileo> And group patches to limit the number of microversions?
16:33:00 <smcginnis> +1
16:33:02 <e0ne> +1
16:33:06 <jungleboyj> +1
16:33:07 <patrickeast> +1
16:33:15 <erlon-airlong> geguileo: +1
16:33:36 <geguileo> OK, I'll add it as a comment to the patch
16:33:45 <smcginnis> geguileo: Thanks!
16:33:46 <erlon-airlong> geguileo: I vote for 1 patch that does all changes and mv bump
16:33:47 <geguileo> And ask DuncanT to remove his -2
16:34:24 <jgriffith> meh; I'm kinda with DuncanT WRT not seeing much value in changing but oh well
16:34:45 <geguileo> jgriffith: Yeah, I don't have a strong opinion on this one
16:35:01 <geguileo> jgriffith: That's why I wanted to gather everyones opinion
16:35:16 <mdovgal> erlon-airlong, +1
16:35:23 <jgriffith> geguileo: yeah, good call!
16:35:54 <geguileo> smcginnis: next topic?
16:35:57 <geguileo> ;-)
16:36:06 <smcginnis> #topic IBMStorage Driver Update
16:36:14 <jungleboyj> Hey!
16:36:15 <smcginnis> jungleboyj: ringleader?
16:36:23 <jungleboyj> smcginnis: Yes, that is me.
16:36:43 <jungleboyj> So, we now have the XCLI library for talking to XIV out in pypi .
16:36:54 <smcginnis> Nice!
16:37:11 <erlon-airlong> jungleboyj: links?
16:37:20 <jungleboyj> We are working very hard to get the opensourced code for the driver pushed out later today or for sure this week so we can start reviews.
16:37:42 <jungleboyj> hemna: Do you the link to pypi handy?
16:37:44 <scottda> #link https://pypi.python.org/pypi/pyxcli/1.1.0
16:37:52 <jungleboyj> scottda: Thanks!
16:38:13 <jungleboyj> Can we get the driver added to the review focus list one we have it pushed up?
16:38:30 <smcginnis> jungleboyj: Sure, that would be good.
16:38:43 <jungleboyj> smcginnis: Cool.
16:39:01 <smcginnis> jungleboyj: Thanks to IBM for changing that.
16:39:14 <jungleboyj> So, otherwise, just a heads up that we have gotten through the legal issues and are working to get the code there as fast as possible given the limited runway.
16:39:41 <jungleboyj> Obviously CI has already been there and shouldn't be a blocker.
16:40:07 <jungleboyj> Any questions?
16:40:12 * jgriffith cough cough cough
16:40:16 <smcginnis> I'm good.
16:40:21 <jungleboyj> jgriffith: Yes sir?
16:40:44 <jgriffith> oh nuthin
16:40:55 * jungleboyj laughs.  Ok.
16:41:22 <smcginnis> jungleboyj: OK, thanks.
16:41:28 <jungleboyj> smcginnis: Thank you!
16:41:31 <smcginnis> #topic Open discussion
16:41:42 <erlon-airlong> Cibo: ?
16:41:45 <Cibo> #topic JovianDSS Driver Update
16:41:48 <smcginnis> Cibo: You wanted to bring up omething?
16:41:53 <smcginnis> *something
16:42:03 <Cibo> so I have driver code almost done
16:42:35 <Cibo> and CI that is posting to http://openstack-ci-logs-open-e.s3-website-eu-west-1.amazonaws.com/logs
16:43:20 <Cibo> as I see I need to fix what is left and start to leave comments on gerrit
16:43:57 <Cibo> any notice, advises...
16:44:13 <Cibo> code review https://review.openstack.org/#/c/399967/
16:45:10 <e0ne> Cibo: you patch still doesn't pass pep8 checks :(
16:46:04 <xyang> Cibo: please add unit tests too
16:46:20 <smcginnis> Cibo: Also needs a release note.
16:46:34 <Cibo> e0ne: yep, I have changed options name and now as I see I have to update by hands options file
16:47:21 <Cibo> cinder/opts.py if I am write
16:47:26 <e0ne> Cibo: please, take a look on http://docs.openstack.org/developer/hacking/. you've got some imports that don't follow guidelines
16:48:05 <smcginnis> Cibo: The opts.py file gets updated by running tox -e genopts.
16:48:35 <smcginnis> Cibo: To be honest, this is looking more likely for Pike. But if you can address all issues, there's still some time.
16:49:23 <Cibo> smcginnis: I am really tending to do it all in time
16:49:53 <smcginnis> Cibo: Good luck.
16:49:56 <smcginnis> Any other topics?
16:50:36 <smcginnis> Going once..
16:51:00 <smcginnis> Going twice..
16:51:10 <smcginnis> OK, thank you everyone.
16:51:32 <e0ne> see you next week
16:51:36 <Swanson> toodles
16:51:40 <smcginnis> #endmeeting