Tuesday, 2015-03-17

*** david-lyle is now known as david-lyle_afk03:35
* ttx yawns09:00
asalkeldhi ttx09:00
ttxasalkeld: hi!09:00
ttx#startmeeting ptl_sync09:00
openstackMeeting started Tue Mar 17 09:00:28 2015 UTC and is due to finish in 60 minutes.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.09:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.09:00
openstackThe meeting name has been set to 'ptl_sync'09:00
ttx#topic Heat09:00
asalkeldttx sorry, i missed last weeks09:00
ttx#link https://launchpad.net/heat/+milestone/kilo-309:00
ttxasalkeld: not a problem, I skipped most of them anyway09:00
ttxSo... crunch time09:01
ttx6 BPs up09:01
asalkeldso most are very close to getting  merged09:01
ttxLet's go through them one by one09:01
asalkeldso i am not very stressed about them09:01
asalkeldok09:01
ttxhttps://blueprints.launchpad.net/heat/+spec/convergence-resource-table09:01
ttxOnly needing https://review.openstack.org/#/c/156693/ ?09:01
asalkeldthat's a one patch spec09:01
asalkeldyip09:02
asalkeldshould get in soon09:02
ttxhttps://blueprints.launchpad.net/heat/+spec/decouple-nested09:02
asalkeldthat's mine09:02
asalkeld*should* get it in time09:02
asalkeld3 patches up, all working well09:02
asalkeldsome disent amounst the minions;)09:02
ttxhttps://blueprints.launchpad.net/heat/+spec/convergence-lightweight-stack09:03
asalkeldthat might well miss09:03
asalkeldbut it't only one patch09:03
asalkeldso willing to wait until the death09:03
ttxI see two patches09:03
ttxIf that one misses, do you intend to request FFE for it ?09:04
asalkeldttx: it can just wait09:04
ttxIf not, maybe prioritize to Medium09:04
asalkeldok09:04
ttxI assume the first two would get exceptions09:04
ttxif they missed09:04
ttxhttps://blueprints.launchpad.net/heat/+spec/convergence-push-data09:04
asalkeldthat should be ok09:05
asalkeldagain one patch09:05
ttxOK, it's linked to the other09:05
ttxhttps://blueprints.launchpad.net/heat/+spec/mistral-resources-for-heat09:05
ttxLooks like 2 patches09:06
asalkeldso this is a bp for code that goes into contrib/09:06
asalkeldso IMO it can go in any time09:06
asalkeld(does not effect strings or deps)09:06
ttxyeah, at least could get an exception as "self-contained"09:06
asalkeldand gets installed seperately09:07
ttxhttps://blueprints.launchpad.net/heat/+spec/stack-breakpoint09:07
asalkeldthere are 2 or 3 guys trying hard to get that in09:07
asalkeldit should make it09:07
asalkeldneeded for tripelo09:07
ttxok we'll see how it goes09:07
asalkeldyeah, we can wait09:08
ttxOn the bugs side... anything k3-critical that we would not just defer to rc1 ?09:08
asalkeldthey are all on the edge of getting merged09:08
asalkeldi think just defer everything that's not merged09:08
ttx#info k3 bugs can be deferred to rc1 if incomplete09:09
ttx#info 6 BPs left, all still standing a reasonable chance of making it09:09
* asalkeld is hopeful09:09
ttxOK, that's all I guess. We'll make a new status point on Thursday (your evening, my morning)09:10
ttxhopefully for the final tag / early FFE decisions09:10
asalkeldok - same time?09:10
asalkeld(just want to make a reminder)09:10
ttxsounds good. Ping me if I don't talk09:10
asalkeldok, chat then09:10
ttxsee you!09:11
ttxzz_johnthetubagu: around?09:11
*** asalkeld has left #openstack-relmgr-office09:13
*** zz_johnthetubagu is now known as johnthetubaguy09:14
ttxjohnthetubaguy: o/09:17
johnthetubaguyttx: hi09:17
ttx#topic Nova09:17
ttx#link https://launchpad.net/nova/+milestone/kilo-309:17
ttxThat's still 12 BPs in progress09:17
ttxLet's see how far they are09:18
ttxhttps://blueprints.launchpad.net/nova/+spec/v2-on-v3-api09:18
ttxI see 4 patches on https://review.openstack.org/#/q/topic:bp/v2-on-v3-api,n,z09:18
ttxBut the last says it partially implements so it looks like there is more to it09:18
johnthetubaguyhmm, I thought they were done there now, pretty much, but I am rarely online the same time as those folks so its tricky relaying info09:19
ttxThose might be add-ons patches09:20
johnthetubaguyah, yes, I think a lot of this is now unit test clean up09:20
ttxok, if that's the case it would be good to stop mentioning the BP in the patches at some point and consider it "finished"09:20
johnthetubaguywell, general cleanups anyways09:20
ttxhttps://blueprints.launchpad.net/nova/+spec/api-microversions09:21
johnthetubaguyagreed, but folks do like the tracking09:21
ttxseems to be in the same situation09:21
johnthetubaguyyes09:21
ttx#info v2-on-v3-api and api-microversions are "mostly finished"09:21
ttxhttps://blueprints.launchpad.net/nova/+spec/cells-v2-mapping09:21
ttxThis one will be partial IIUC09:22
ttxin which case it is "mostly finished" too, I suspect09:22
johnthetubaguyyes, there are a few patches, I think they wanted, but as you say, its kinda partial anyways09:22
ttxhttps://blueprints.launchpad.net/nova/+spec/kilo-objects09:23
johnthetubaguythis one is basically tracking cleanups09:23
ttxFeels like another "what's in kilo is kilo"09:23
johnthetubaguyright09:23
johnthetubaguyits a tracking placeholder to help with reviews more than anything09:23
ttx#info cells-v2-mapping and kilo-objects are partial work anyway, so what's in kilo is kilo09:23
ttxhttps://blueprints.launchpad.net/nova/+spec/online-schema-changes09:24
johnthetubaguyone last patch there I think09:24
ttxpending on https://review.openstack.org/#/c/154521/09:24
ttx#info online-schema-changes pending on https://review.openstack.org/#/c/154521/ -- looking good09:24
ttxit says "partially implements" though09:24
ttxhttps://blueprints.launchpad.net/nova/+spec/make-resource-tracker-use-objects09:25
johnthetubaguyyeah, thats true, I will check, but I thought that was it09:26
ttxLots of patches up09:26
ttxThis one sounds more at risk09:26
ttxalthough it's marked partial too09:27
johnthetubaguyoh yes, I think you are right, that one was close to getting kicked out already09:27
ttx#info make-resource-tracker-use-objects might be so partial it could be considered deferred09:27
ttxhttps://blueprints.launchpad.net/nova/+spec/detach-service-from-computenode09:27
johnthetubaguyI will check out that one closer after the meeting09:28
ttxNot sure what's waiting on09:28
johnthetubaguyit might be done now, as much as it will be in kilo, let me check09:28
ttxyou can check after meeting, no worry09:29
johnthetubaguyyep is done, although its technically, a bit partial09:29
ttx#info detach-service-from-computenode might be considered finished09:29
ttxOK, medium stuff now09:29
ttxhttps://blueprints.launchpad.net/nova/+spec/keypair-x509-certificates09:29
ttxRemaining patches are novaclient and a bug fix09:30
johnthetubaguyyes, they need work really09:30
ttxso we could move them to RC1 if necessary09:30
ttxand consider this finished09:30
johnthetubaguyyes, I think thats probably easiest, I might open bugs for those patches09:31
ttx#info keypair-x509-certificates could just be considered finished and remaining bugs pushed to RC1 if need be09:31
ttxhttps://blueprints.launchpad.net/nova/+spec/v3-api-policy09:31
ttxThis one is marked partial, so probably another of those "what's in kilo is kilo"09:31
johnthetubaguypossibly, there are a few patches they really want in kilo, its on the etherpad I think09:32
ttxLoads of patches there09:32
johnthetubaguyyeah, its a streatch to get those all in09:32
ttx#info v3-api-policy needs work, will be partial09:32
ttxhttps://blueprints.launchpad.net/nova/+spec/ec2-api09:32
ttxIsn't that finished?09:33
johnthetubaguyI think it is09:33
ttx#info ec2-api is probably finished09:33
ttxhttps://blueprints.launchpad.net/nova/+spec/servicegroup-api-is-up-host-topic09:34
ttxA bit unclear what it's waiting on09:34
ttxalso marked partial09:34
johnthetubaguyI think its this one: https://review.openstack.org/#/c/149924/09:35
ttx#info servicegroup-api-is-up-host-topic pending on https://review.openstack.org/#/c/149924/09:36
johnthetubaguyits a clean up though, so I might just remove the blueprint tag, lets see09:36
ttxhttps://blueprints.launchpad.net/nova/+spec/hyper-v-test-refactoring09:36
ttxThat one looks forever partial09:36
ttxand candidate for deferral09:37
ttx#info hyper-v-test-refactoring may or may not make it09:37
johnthetubaguyyeah, I vote we defer the BP, I mean its unit tests, so they can merge later if needed I guess09:37
ttxyeah, I'm more afraid of the distration than of the impact09:37
ttxdistraction09:37
ttx#info 10 blueprints left, mostly partial & "what's in kilo will be kilo" work anyway09:38
ttxQuick look on the bugs side, see if any of those "critical" are actually k3-critical09:38
ttxlike regressions09:39
ttxor showstoppers09:39
ttxhttps://bugs.launchpad.net/nova/+bug/1383465 is unlikely to be solved in the next two days09:40
openstackLaunchpad bug 1383465 in OpenStack Compute (nova) "[pci-passthrough] nova-compute fails to start" [Critical,In progress] - Assigned to Yongli He (yongli-he)09:40
johnthetubaguynothing that stops kilo-3 from a quick look, right, thats the only big one09:40
johnthetubaguyit got discussed in the nova meeting again09:40
johnthetubaguytrying to move that forward before kilo ships09:40
johnthetubaguyit was targeted to kilo-1 I think09:40
ttxand passed forward09:41
ttxhttps://bugs.launchpad.net/nova/+bug/1296414 is another of those long-standing issues09:41
openstackLaunchpad bug 1296414 in OpenStack Compute (nova) "quotas not updated when periodic tasks or startup finish deletes" [Critical,In progress] - Assigned to Rajesh Tailor (rajesh-tailor)09:41
ttxok, so I'll push to RC1 anything not merged by tag time09:42
ttx#info Bugs to be deferred if not merged09:42
johnthetubaguysounds good09:42
johnthetubaguyI will ask around to see if there are any exceptions to that09:42
ttxjohnthetubaguy: how about we meet again on Thursday and formally close all those partial things, then see if anything in the pipe justifies waiting ?09:42
johnthetubaguythat sounds like a plan09:42
ttxsame time ?09:43
johnthetubaguyyes, I will pop that in my calendar now09:43
ttxsounds good, have a good day!09:43
johnthetubaguyyou too!09:43
dhellmannttx: ready when you are12:24
ttxdhellmann: you might be one hour early, but we can probably do it nevertheless12:25
dhellmannttx: oh, right, forgot you're not on dst yet12:25
ttxno pb, if you're available now, let's do this12:25
dhellmannsure thing12:26
ttx#topic Oslo12:26
ttxA quick look at milestone pages12:26
ttx#link https://launchpad.net/oslo/+milestone/kilo-312:26
ttxthere is a bug targeted: https://bugs.launchpad.net/cinder/+bug/141701812:26
openstackLaunchpad bug 1417018 in oslo.db "Cinder encounters dbapi error: NoSuchColumnError: "Could not locate column in row for column '%(140070887032400 anon)s.volumes_id'"" [Medium,In progress]12:26
ttxis that something we want to fix and sync in kilo ?12:27
dhellmannyeah, that one is actually done in oslo.db -- I'll be making a release later today12:27
dhellmannwe haven't settled on the best way to fix it in the incubator, but the oslo.db fix should remove the issue from cinder12:27
dhellmannI'll update the targets12:27
ttxok cool12:27
ttxhttps://launchpad.net/oslo/+milestone/next-kilo shows a bunch of PBR bugfixes12:28
ttxis there a PBR release planned ?12:28
ttxor would those actually appear in next-liberty ?12:28
dhellmannttx: not to include those items. I'll retarget them too12:28
ttxok12:28
ttxLooking at https://review.openstack.org/#/c/162656/12:29
dhellmannactually, I don't think whoever has been doing pbr releases has been using the script, so I'll verify that they're not already done12:29
ttxdo you think we should collect more +2s ? Or should I just W+1 it ?12:29
ttxnot sure why the subsequent reviewers didn't w+112:30
dhellmannyeah, sdague expressed some reservations and that may have made people hold off on the w+112:30
dhellmannI would like to get that in asap so we can also land the subsequent requirements updates to the projects12:30
dhellmannthe requirements freeze is this week, right?12:31
ttxyeah, timing is tight12:31
sdaguejogo is +1 on that now, I'll approve12:31
dhellmannsdague: thanks, I was going to check in with you on that in a bit anyway12:31
sdaguethat's going to trigger a ton of patch / test churn, btw12:31
ttxrequirements freeze is same as feature freeze yes12:31
dhellmannsdague: yeah, that's why I did it all in one patch :-)12:31
sdague:)12:31
ttxsdague: should we generally try do those one week earlier ?12:31
sdaguettx: probably12:31
sdagueI expect we'll consume all the test nodes on the proposal bot changes for that one12:32
ttxnote that those could technically land post-requirements freeze, since requirements freeze is mostly to guard against additions12:32
dhellmannI think we would have been ready a week earlier this cycle except for the deprecation warning issue12:32
dhellmannttx: ok, as long as it lands before the branches are made I think we're ok12:32
ttxso I'm fine with those landing post-FF pre-RC112:32
dhellmannI'm OK with that if we want to preserve test nodes12:33
ttxif we think approving this now is a bad idea12:33
dhellmannI can update the policy document to reflect that change, too12:33
ttx(requirements sync is part of the RC1 process)12:33
sdaguezuul still has capacity this morning12:33
sdagueso anyway, it's approved12:33
dhellmannok, thanks, sdague12:33
ttxdhellmann: I also wanted to pick your brains on which projects actually could be considered havign a final release per cycle (for the release:at-6mo-cycle-end tag application)12:34
ttxFirst, looking at Oslo libs12:34
ttxI suspect some of them don't really apply12:35
ttxfor example openstack-dev/pbr12:35
dhellmannI want them all to be considered as following that plan, although we had a couple of stragglers this time. pbr and taskflow, I think.12:35
dhellmannI need to get the pbr team inline with the rest of oslo12:35
ttxwhat about openstack-dev/cookiecutter ?12:35
ttx(and openstack-dev/oslo-cookiecutter)12:36
dhellmannoh, we don't release that, that's just a template for making new projects12:36
dhellmannyou use it straight from the git url12:36
ttxright, so those would not have release:* tags12:36
dhellmannright12:36
ttxso what should I do about taskflow/pbr ? Consider them in ?12:37
dhellmannyes, let's consider them in and I'll talk to the maintainers12:37
ttxSecond question is about python-*client, to piggyback on your discussion with David Lyle recently12:37
ttxI'm a bit confused if we should consider them in (as in having stable branches)12:38
ttxI think they might be release:independent, really12:38
dhellmannwell, we need to start doing that, I think. We're capping the versions we use in the stable branches, so if we have to fix something the only want to release a new version would be to have a stable branch of the client12:38
dhellmannI believe the d-g tests are already installing the services and tempest in separate site-packages (one is in a virtualenv, I'm not sure which)12:39
dhellmannthat lets us test  new clients from the outside of the cloud while using stable clients inside the cloud12:39
ttxhmm, ok. So they should be at-6mo-end12:39
ttxeven if technically they don't have stable branches until needed12:39
dhellmannyes, unless someone else comes up with an alternative to my stable-library spec12:40
ttxApplying tags results in uncovering interesting corner cases to say the least12:40
dhellmanntrue12:40
ttxat the very least helps clarify things, even for us12:41
ttxok, I think that's all I had in store for you12:41
dhellmannok, I had a couple for you12:41
ttxdo we have any lib left to release for kilo ? You mentioned pbr ?12:41
dhellmannthat policy on stable branches for libs should be prioritized: https://review.openstack.org/15507212:42
ttxyeah, I added it to todays meeting agenda12:42
dhellmannI have a backported fix for oslo.db that I will release today, but I don't anticipate any other releases12:42
dhellmannok, good12:42
ttx(cross-project meeting)12:42
dhellmannand do we know the dates for the fall summit yet?12:42
ttxin silence is consent mode12:42
ttxI think we do, let me check12:43
dhellmannI didn't see them on the event calendar on openstack.org, but I don't know how far out that actually goes12:43
ttxhttps://www.openstack.org/summit/tokyo-2015/12:43
ttxOctober 26 - 30, 201512:43
dhellmannah, ok thanks12:43
ttxThat may not be linked to yet, but the page is public12:43
ttxand the URL is not that hard to guess12:44
dhellmannyeah, I should have tried that :-)12:44
ttxactually it's linked12:44
ttxhttps://www.openstack.org/summit/12:44
ttxso consider it public :)12:44
dhellmannI was looking on https://www.openstack.org/community/events/12:45
dhellmannthe calendar includes the spring summit, so I thought if the dates were set for fall that would be there, too, but maybe it's too far out12:45
dhellmannanyway, thanks, I should have looked harder :-)12:45
dhellmannthat's all I had for this week12:45
ttxdhellmann: so in summary, next steps are merging requirements patch (but that's arguably non-oslo-specific12:46
ttxand get that policy approved12:46
ttxotherwise you're mostly done release-wise ?12:47
dhellmannyes, I believe that's it12:47
ttxdhellmann: we could push the kilo-3 tag on oslo-incubator now12:47
dhellmannwe'll obviously do patch releases if we need to, but the branches are created so we can do that12:47
ttxsince we did a 2015.1.0b212:47
dhellmannsure, we can push the tag. I think we said we would wait to do the stable branch on the incubator until the other projects had theirs, but the tag is easy12:48
dhellmanndo you want to do that, or shall I?12:48
ttxI'll do it now12:48
dhellmannok12:48
ttxdone12:49
dhellmanncool, thanks12:49
ttxok, thx and have a good day!12:50
dhellmannyou, too!12:50
SergeyLukjanovttx, /me ready when you are13:05
ttxSergeyLukjanov: hi!13:05
ttx#topic Sahara13:05
SergeyLukjanovttx, hello!13:05
SergeyLukjanov#link https://launchpad.net/sahara/+milestone/kilo-313:05
SergeyLukjanovso, we have mostly everything planned done13:06
ttx8 still up13:06
SergeyLukjanovexcept few things that are now candidates for FFE13:06
SergeyLukjanovstuff that is under review now is expected to be merged today/tomorrow13:06
ttxSergeyLukjanov: ok13:07
ttxWhat are the FFE candidates ? edp-job-types-endpoint ?13:07
SergeyLukjanovttx, yup, this one for sure, there is a core stuff merged already, but plugin-related things not yet merged now, so, we'll need some more time for it13:07
ttx#info edp-job-types-endpoint likely FFE13:08
SergeyLukjanovthe same for "Provide default templates for each plugin" https://blueprints.launchpad.net/sahara/+spec/default-templates13:08
ttx#info default-templates likely FFE13:08
SergeyLukjanovand probably for edp-datasource-placeholders as well13:08
ttx#info edp-datasource-placeholders likely FFE13:09
ttxthe others shall merge before the k3 date13:09
SergeyLukjanovyup13:09
ttxOK, any k3-blocking bug ? Or can I just defer them to rc1 isf they miss target ,13:09
ttx?13:09
SergeyLukjanovI expect sahara to be ready for tag cut on Thu EU morning13:09
SergeyLukjanovno k3-blocking bugs13:09
SergeyLukjanovI've cleaned up the bugs list, so, there is only few of them and I think they will  be merged till the k313:10
ttxOK. I had a question for you on dashboard, extra and image-elements13:10
SergeyLukjanovand anyway they aren't importnat13:10
SergeyLukjanovyeah13:10
ttxhow many of those are actually still in use ?13:10
SergeyLukjanovdashboard is only for stable/icehouse13:11
SergeyLukjanovstarting from Juno it's moved to horizon repo13:11
SergeyLukjanovextra and s-i-e are in use13:11
ttxwhat is the release model for extra and image-elements ?13:11
ttxonce at the end of the 6-month cycle ?13:11
SergeyLukjanovthe same as for sahara - tagging each 6 month ~ the same day with sahara13:11
ttxok, that works13:12
SergeyLukjanovit's needed for coupling images builded for sahara and tools used inside of them13:12
ttxI'll apply the corresponding tag as needed :)13:12
SergeyLukjanovokay, thx :)13:12
ttxyou dont use intermediary tags on those, right ?13:12
* SergeyLukjanov lurking for gov changes :)13:12
SergeyLukjanovttx, we're using the kilo-X tags as well13:12
ttxok13:13
ttxThat's all the questions I had, thx13:13
SergeyLukjanovttx, in fact all tags done for sahara are done for both extra and image-elements repos as well13:13
ttxack. have a good day!13:13
SergeyLukjanovoh, I have a thing to share13:13
ttxlistening13:13
SergeyLukjanovwe have an issue with src juno job for sahara client13:13
ttxmaybe #info it13:14
SergeyLukjanovwe're now investigating it and we'd like to make it non-voting13:14
SergeyLukjanovfor some time to avoid blocking client dev for kilo13:14
SergeyLukjanovheat-engine fails to load because saharaclient installed from source and heat checks versions13:14
SergeyLukjanovand that's extremely strange why only sahara job is failing13:15
SergeyLukjanovwe're not trying to understand it13:15
SergeyLukjanovhttps://review.openstack.org/#/c/165038/13:15
ttxok13:16
SergeyLukjanov#info gate-tempest-dsvm-neutron-src-python-saharaclient-juno is now failing due to the very strange issue - heat-engine fails to load because saharaclient installed from source and heat checks versions and finds incorrect version (not <=0.7.6 defined in requirements.txt)13:16
SergeyLukjanovnothing else from my side13:16
SergeyLukjanovthx13:16
ttxcheers13:16
SergeyLukjanovttx, have a good day!13:17
ttxdhellmann: https://review.openstack.org/#/c/162656/ failed gate on some horizon startup failure13:36
ttxInteresting: http://logstash.openstack.org/#eyJzZWFyY2giOiJcIm9wdC9zdGFjay9uZXcvZGV2c3RhY2svZXhlcmNpc2VzL2hvcml6b24uc2g6Mzk6ZGllXCIiLCJmaWVsZHMiOltdLCJvZmZzZXQiOjAsInRpbWVmcmFtZSI6IjE3MjgwMCIsImdyYXBobW9kZSI6ImNvdW50IiwidGltZSI6eyJ1c2VyX2ludGVydmFsIjowfSwic3RhbXAiOjE0MjY1OTk1MDQ0ODd913:38
*** johnthetubaguy is now known as zz_johnthetubagu13:43
*** zz_johnthetubagu is now known as johnthetubaguy13:51
*** thingee has joined #openstack-relmgr-office14:18
ttxthingee: if you want to go at your usual USA DST time, we can do it now14:25
ttxsince I'm around anyway14:26
thingeeha I was just thinking of that14:26
ttxyour call14:26
thingeesure sounds good to me14:27
ttx#topic Cinder14:27
ttx#link https://launchpad.net/cinder/+milestone/kilo-314:27
ttxSo you seem to be all-set featurewise14:27
ttxbut you might want to include a few more bugfixes before tagging µ?14:27
thingeeyea, we still got some we have to take of.14:28
thingeeand me as well ;)14:28
ttxAre any of those targeted bugs k3-critical ? Or can I just move them if they happen not to make it at tag time ?14:28
thingeeI just went through the ones targeted and think they have a fair shot. They're all in review.14:29
thingeeI removed the ones that have no reviews at this point.14:29
ttxOK, let's try to use the list as targets then14:29
ttxand tag when list is empty, and revise that on Thursday if necessary14:29
thingeesounds good14:30
ttxsee PM for a few security issues pings14:32
ttxThat's all I had14:32
ttxit's the calmest Cinder FF since a long time :)14:33
thingee:)14:33
thingeesee the ML. I'm not a popular person14:34
ttxI'll let you go back to debunking security issues and review bugfix patches.14:34
thingeethough I'm not sad to say no to patches coming in on March 10th for new features.14:34
ttxok, have a great day then14:35
thingeehave a great rest of your day!14:35
ttxnikhil_k: interested in speaking now ? I might have a call interrupting our usual time14:35
ttxmestery: same14:35
nikhil_kttx: sure, trying to do double meeting14:36
nikhil_kanother is audio14:36
ttxI'll go slowly14:36
ttx#topic Glance14:36
nikhil_kthanks14:36
ttx#link https://launchpad.net/glance/+milestone/kilo-314:36
ttx3 BPs left on the docket14:36
ttxyou sent an email on those, let me check14:37
ttxok, so those are all candidates for a FFE if they don't make it14:37
ttxHow far are they from completion ?14:38
ttxartifact-repository seems to be half-dozen patches away, but mostly proposed14:38
nikhil_kttx: yes, I'm tentative on all three at this point. Catalog Index Service is still pending some last minute discussion/blockers. Artifacts is stuck on lot of reviews and the policy strenghthening is waiting on one confirmation due to a possible conflict/overlap with another feature.14:38
ttxok, would be good to have at least one of those in by the deadline to limit the number of exceptions14:39
ttx#info last 3 BPs are likely to require exceptions14:39
nikhil_kttx: how long can we go with FFE?14:39
ttxnikhil_k: on the targeted bugs side, I assume nothing is k3-critical and everything can be deferred to rc1 if not fixed by Thursday14:40
nikhil_k1 week / 2 weeks?14:40
ttxnikhil_k: ideally max 10 days14:40
nikhil_kgotcha14:40
ttx(i.e. end of month)14:40
nikhil_ksure thing14:40
mesteryttx: here when ready14:40
nikhil_kttx: about the bugs14:40
ttx#info Bugs can be deferred if needed14:40
nikhil_kttx: this one https://review.openstack.org/#/c/119824/11 might be tricy https://launchpad.net/bugs/136543614:40
openstackLaunchpad bug 1365436 in Glance "Database schema differs from migrations" [High,In progress] - Assigned to Oleksii Chuprykov (ochuprykov)14:41
nikhil_kI will try to review soon14:41
nikhil_k(that's it from me)14:41
ttxnikhil_k: ok, keeping an eye on it14:41
ttx#info Bug 1365436 should be solved before k3 ideally14:42
openstackbug 1365436 in Glance "Database schema differs from migrations" [High,In progress] https://launchpad.net/bugs/1365436 - Assigned to Oleksii Chuprykov (ochuprykov)14:42
nikhil_kthanks!14:42
ttxnikhil_k: that is all I had. We'll revisit progress on Thursday14:42
ttxcheers14:43
nikhil_ksure14:43
ttxmestery: o/14:43
ttx#topic Neutron14:43
mesteryttx: o/14:43
ttx#link https://launchpad.net/neutron/+milestone/kilo-314:43
ttxhmm, lots of open stuff there14:43
mesteryYup, so many are very close14:44
mesteryBut the reality is I need to start moving some of them out14:44
mesteryWe had a gate issue which killed our Fri-Mon the past weekend which was awful :(14:44
ttxyou need to sort them so that those that are 99% there get priority reviews14:44
mesteryYes.14:44
ttxbecause otherwise they will all be at 98% on Thursday and just generate more pain14:45
ttx#info 34 open BPs14:45
mesteryIndeed, I'm on that post this meeting and will get it down to the actual 5-8 that will make it14:45
ttx#info need to sort between FFE candidates, nearly-there and wont-make-it14:45
ttxHow about we discuss again tomorrow and see how much you managed to clean up ?14:46
mesteryWorks for me.14:46
mesteryThe javelin/grenade issue just killed us, apologies for being behind a bit14:46
mesteryI spent yesterday sheparding that and trying to recheck some stuff. *sigh*14:47
ttxOn the bugs side, anything that should be k3-blocking ?14:47
* mestery finds the one14:47
ttxOr just move them all to RC1 as-needed ?14:47
mesteryWell, actually, no none are K3 blocking14:47
mesteryWe have a few that may be RC blockers, but not K314:47
ttxthey can be RC1-blocking alright14:47
ttxok, so I'll just defer the open ones when we tag14:47
mesteryThanks ttx!14:48
mestery++ to that14:48
ttxThat is all I had14:48
ttxTalk to you tomorrow then14:48
mesteryMe too, back to cleanup now14:48
mesteryyes sir14:48
mesterylater!14:48
ttxcheers14:48
*** david-lyle_afk is now known as david-lyle14:57
*** SlickNik has joined #openstack-relmgr-office15:08
ttxdavid-lyle: I'll be 5min late15:40
david-lylettx: no worries15:40
ttx#topic Horizon15:44
ttxdavid-lyle: hi!15:44
david-lylettx: hi15:44
ttx#link https://launchpad.net/horizon/+milestone/kilo-315:44
ttx22 still on the docket, time for some pre-FF cleanup15:45
david-lyleyes indeed15:45
*** thingee has quit IRC15:45
david-lylewill cull today, the process continues15:45
david-lylehad to resharpen cleaver15:45
ttxyeah, would be good to split between FFE candidates, nearly-there and wont-make-it15:46
ttxso that you can actually focus on the 2nd category rather than dilute review effort15:46
david-lyleI have some status updates to roll in too15:46
ttxHow about you process that and we discuss status of the remaining items tomorrow ?15:47
david-lyleI believe  the launch instance bp finished merging last night15:47
david-lylesure15:47
ttxOn the bugs side I can't find anything k3-critical15:47
ttxso i'll just defer whatever didn't make it at tag time15:47
david-lyleyeah, nothing critical in the current list15:48
david-lyleI'm sure something will spring up15:48
ttxI had a question about tuskar-ui release model, if any15:48
ttxsince it's a repo listed under horizon15:48
david-lyleIIRC I just tag it around release time15:49
ttxok, so one tag every 6 months around cycle end ?15:49
david-lylethat's been the cadence15:50
ttxok thx15:50
ttxdavid-lyle: and in that discussion here yesterday we said we'd probably need stable branches for django_openstack_ath, right15:51
david-lyleyes, trying to sort out what that means global-requirements wise15:51
ttxok15:51
ttxso I'll be talking to you again tomorrow, in the mean time please refine and cut the k3 list15:52
david-lyleI could create a stable branch for 1.1.7 and use that for icehouse and juno, but the keystoneclient dependency isn't exact with the global-requirements for those branches15:52
ttxI created l1 milestones if you want to defer stuff15:52
david-lyledoes that have to precisely match?15:53
david-lyleit requires keystoneclient slightly newer than the minimum15:53
ttxdavid-lyle: I think it does, but might want to cross-check with dhellmann15:53
david-lylethat's going to be a difficult shift then15:53
david-lylewill continue to look into it15:54
david-lylemight be nice to pin in requirements for current stable and cut a stable branch for kilo moving forward15:54
ttxack, off-sync discussion15:54
david-lylesure15:54
ttxsure, for kilo we can certainly converge15:54
ttxdavid-lyle: will talk to you more tomorrow15:55
david-lylettx: sounds good, thanks!15:55
ttxmorganfainberg: around,16:51
ttx?16:51
morganfainbergttx: o/16:51
ttx#topic Keystone16:51
ttx#link https://launchpad.net/keystone/+milestone/kilo-316:51
ttx3 Bps on the list16:52
ttxhttps://blueprints.launchpad.net/keystone/+spec/klw-tokens16:52
ttxThought that was mostly in16:52
morganfainbergKlw-tokens is almost done. We need a couple items reviewed should gate today.16:53
ttxok16:53
ttxhttps://blueprints.launchpad.net/keystone/+spec/mapping-enhancements16:53
ttxlooks done to me ?16:53
ttxor maybe pending https://review.openstack.org/#/c/163172/16:53
morganfainbergNot sure. Might be 1 patch16:53
morganfainbergIt's also very close.16:53
ttxhttps://blueprints.launchpad.net/keystone/+spec/model-timestamps16:54
ttxalso seemùs one patch away16:54
morganfainbergThe last one is a cleanup bp and should have no impact to api. It can move to post k316:54
morganfainbergand it will move to post k3 or l116:54
ttx#info 2 BPs very close should make it today/tomorrow, last one (model-timestamps) can be deferred16:54
morganfainbergYep.16:55
ttxOK, sounds pretty close16:55
ttxOn the bug side, I assume nothing is k3-critical and all non-merged can be deferred to rc116:55
morganfainbergThere will be 2 FFEs requested. One has all the code ready just needs review, can't land it all before freeze though. The second is landing a couple final patches for a feature.16:56
ttxI suggest we reconvene on Thursday to tag, or ping me if earlier16:56
morganfainbergBugs are fairly clear, nothing critical on my radar yet.16:56
ttxI had a few questions for you16:56
morganfainbergSure.16:56
ttxfirst would you be available at the cross-project meeting today to discuss progress (or absence thereof) on new client in openstack-sdk16:57
ttxwe discussed that back in December and promised a status update in Feb/Mar16:57
ttxcan be quick :)16:57
morganfainbergSure. And the answer is lack of progress afaik. I'll 2x check with jamielennox though.16:57
ttxsecond, I need to clarify a few release models for your secondary repositories16:58
ttxkeystonemiddleware, pycadf and python-keystoneclient-*16:58
ttxkeystonemiddleware seems released as-needed + one at the end of the cycle, is that a fair assumption ?16:58
morganfainbergPycadf is like Oslo packages16:59
ttxok so same as above16:59
morganfainbergMiddleware is like client and typically released as needed.16:59
morganfainbergAnd yes, one at end of cycle.16:59
ttxand same for the client-*16:59
morganfainbergYep.16:59
ttxOK makes all sense17:00
ttxThat is all I had17:00
ttxanythgin on your side ?17:00
morganfainbergNope.17:00
ttxalright then, talk to you Thursday if not earlier17:00
ttxnotmyname: around?17:00
notmynamegood morning17:00
ttx#topic Swift17:00
ttxnotmyname: hi!17:00
notmynamehello17:00
ttxSame question as for Morgan17:00
ttxwould you be available at the cross-project meeting today to discuss progress (or absence thereof) on new client in openstack-sdk17:01
ttxwe discussed that back in December and promised a status update in Feb/Mar17:01
ttxcan be quick17:01
notmynameif it's the first agenda item.17:01
ttxok, that's the plan17:01
notmynameI was double booked with something else for the bottom of the hour17:01
ttxnotmyname: second question would be on swift-bench -- I assume it's using the release:independent model, is that a good assumption ?17:02
notmynamesame as python-swiftclient17:02
ttxfwiw swift seems to be release:managed, release:at-end-of-6mo-cycle *and* release:independent17:02
ttxnotmyname: ok17:03
notmynamethe managed one is surprising. is that because of the end of cycle one?17:03
notmynameI'd expected end-of-6mo and independent17:03
ttxnotmyname: well, currently I still do the swift releases17:03
notmynameah, I see17:03
notmynamethat makes sense then17:04
ttxI agree that "managed" is tied to milestones, so I might not apply that17:04
ttxand do courtesy handling17:04
ttxand/or give you the tools and have you run with them17:04
ttxwe'll refine as we go17:04
notmynameok. that's something to consider (after kilo)17:04
ttxmy added value in the swift release process is pretty slim17:04
ttxif I can get my tools properly documented, that would be an option17:05
ttxanywat, post-kilo17:05
notmynamehmm... rotating that among swift-core would be interesting17:05
notmynameanyway, post-kilo17:05
ttxhow is development coming up ?17:05
ttxsince last week ?17:05
notmynamefor now, we're furiously working on EC stuff17:05
notmynametons to do17:06
notmynamebut here's the plan:17:06
notmynamewhen it's ready for master (end of next week (-ish), we'll freeze master and work on the merge to master17:06
notmynamelike with storage policies, we'll tag the current feature branch (for history) and build a new patchset for the merge. ie cleaned up work17:07
notmynameand, like with storage policies, I'd like to work with -infra to land it as one merge commit instead of a ff merge17:07
notmynameand that gets us to the april 10 (-ish) target for a RC17:07
ttxsounds good17:07
ttxwe'll keep track of that17:08
notmynamethere's a few other non-ec things I'm tracking17:08
notmynamebut the focus is ec17:08
ttxOK... anything else ?17:08
notmynameI proposed a bunch of tests to defcore17:08
notmynamethe in-tree functional tests17:09
notmynameto the objectstore capability17:09
notmynamein addition to the existing 15 tempest tests listed, I added references to 279 swift functional tests17:09
ttxheh, ok17:09
notmyname#link https://review.openstack.org/#/c/164865/17:10
ttxany other news ?17:10
notmynameso I think that's an important pattern for the TC->code->BoD interaction17:10
notmynameno, I think that's it. tons of work on EC. scary. committed to a beta. scope will be cut before the feature17:11
ttxalright, I'll let you run then17:11
ttxdevananda: you around ?17:11
ttxnotmyname: have a good week17:11
notmynamethanks. have a good day17:11
devanandattx: hi!17:11
ttxhere he is17:11
ttx#topic Ironic17:11
ttx#link https://launchpad.net/ironic/+milestone/kilo-317:11
ttxreading your email17:12
devanandaI can paste the cliff notes here for record keeping17:13
ttxsure17:13
devanandawe now have 7 BP open still17:13
devanandahttps://blueprints.launchpad.net/ironic/+spec/implement-cleaning-states hit an issue with devstack configuration that's blocking landing the code in ironic17:14
devanandaonce we can get that in (or work around it temporarily) that should be ready to merge17:14
ttxwe can easily grant FFE to cover for the landing issue if that needs a few days17:14
devananda3 more BPs are driver-specific and I'd really like to get them in, if possible17:15
devanandawe've already bumped several other driver-specific pieces to Liberty17:15
ttxdevananda: are they likely to hit the Thursday deadline ?17:15
ttxor will require FFE17:15
devanandaright now, I'm about 50/50 on whether all of this will land by thursday17:15
devanandabut - if it's not in by friday, I may just cut things that are driver-specific17:16
ttxthe driver-specific ones are ilo-cleaning-support ilo-properties-capabilities-discovery and... ?17:16
devanandauefi-secure-boot17:16
devanandaactually, sorry, uefi secure boot touches all the drivers17:17
ttxwhat about the last two (local-boot-support-with-partition-images & ipa-as-default-ramdisk) ? How close are they ?17:17
devanandait's the other one that might delay the release17:17
devanandaautomate-uefi-bios-iso-creation is the other driver-specific one, but it's also very small at this point and will probably land today17:17
devanandathose two, I think, are done, though they are tagged on a UEFI-related review, and I'm waiting for feedback from the developers as to why17:18
ttxok, let's discuss progress tomorrow (your morning) and see where we are at that point17:19
devanandasounds good17:19
ttxHad a question about ironic-pythonagent17:19
devanandasure17:19
ttxI suspect it's released as-needed17:19
ttxany reason we'd do a stable branch for it ?17:20
devanandaindeed17:20
devanandapossibly17:20
ttxi.e. have a release coinciding with the 6-month cycle ?17:20
devanandaone of the story arcs of this cycle has been adding micro version support17:20
devanandathere are two open reviews related to that on the server side, and a -lot- of work still in the client to fully utilize it17:20
devanandaas we're now versioning the server API, it might make sense to tag clients at the same time17:21
ttxhmm, ok17:21
devanandaOTOH, no future client release should ever be unable to talk to an old server17:21
devananda(within the supported window, which is TBD)17:21
devanandaso making a statement like "to use the juno server, you need th ejuno client" -- we expressly do not want to do that17:21
ttxtrying to see if release:at-6mo-cycle-end would apply to it17:22
ttxor if we'd need something finer for clients/libraries/agents17:22
devanandaI will no doubt tag a client somewhere around the time we tag kilo rc, and probably again around kilo-final, and after that i'll tag it again ...17:22
ttxright, sounds like "having a stable branch" should be separated from "releasing at the end of the cycle"17:23
devanandaI don't see a need to release the client and server in lock-step. and I don't want to, explicilty or implicitly, require folks to install an older client to talk to an older server17:23
devanandathe more i think through it, the less i think a stable branch of python-ironicclient makes sense17:24
ttxyeah, so including clients on the tag might communicate the wrong intent17:24
devanandaright17:24
ttxanyway, side discussion17:24
ttxthx for the input though17:24
devanandasure!17:24
ttxdevananda: questions on your side ?17:24
devanandattx: nope. I'm going to keep at the status page, trying to get everything green17:25
devanandatty tmw. same time?17:25
ttxalright, talk to you tomorrow17:25
ttxyes17:25
ttxSlickNik: around?17:25
*** johnthetubaguy is now known as zz_johnthetubagu17:26
SlickNikttx: o/17:27
ttx#topic Trove17:27
ttx#link https://launchpad.net/trove/+milestone/kilo-317:27
ttxLooks like 6 BPs still on the fence17:27
ttxhttps://blueprints.launchpad.net/trove/+spec/mgmt-show-deleted17:27
ttxThis one looks relatively close17:27
SlickNikYes, that should make it in by today, I'm guessing.17:28
SlickNikClose to the finish line.17:28
ttxhttps://blueprints.launchpad.net/trove/+spec/replication-v2 -- not very far either afaict17:28
ttxhttps://blueprints.launchpad.net/trove/+spec/vertica-db-support -- one patch pending too, but admittedly a large one ?17:29
ttxsame for https://blueprints.launchpad.net/trove/+spec/couchdb-plugin-trove17:29
ttxhttps://blueprints.launchpad.net/trove/+spec/implement-vertica-cluster -- depends on the other, so likely to be late17:30
ttxhttps://blueprints.launchpad.net/trove/+spec/guest-rpc-ping-pong -- close to merge17:30
SlickNikhttps://blueprints.launchpad.net/trove/+spec/implement-vertica-cluster I'm concerned about.17:30
ttxyes, it's also a feature needing testing, so deferring might be the rigth call17:31
ttxrather than pushing it late in17:31
ttxHow about we revisit tomorrow, once most of those will be in ?17:32
SlickNikttx: agreed. I'll talk with the author (sushil) to see what's his take on that.17:32
ttxWould be great to push most in in the mean time17:32
SlickNikttx: sounds good. I think at least 4-5 of those will have merged by then.17:32
ttxack17:32
ttxOn the bugs side, anything k3-critical ?17:32
SlickNikI'm working on getting folks to reviewthe patches by17:32
ttxOr can I just defer to RC1 anything that doesn't make it ?17:32
SlickNikttx: We've fixed all the ones that are critical already.17:33
SlickNikSo I'd defer anything in kilo-3 that doesn't make it (bugwise) to RC117:34
ttxalright then, I'll let you babysit patches and talk to you tomorrow17:34
SlickNiksounds good. will ping you if something comes up.17:34
ttxwfm17:35
ttx#endmeeting17:35
openstackMeeting ended Tue Mar 17 17:35:08 2015 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)17:35
openstackMinutes:        http://eavesdrop.openstack.org/meetings/ptl_sync/2015/ptl_sync.2015-03-17-09.00.html17:35
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/ptl_sync/2015/ptl_sync.2015-03-17-09.00.txt17:35
openstackLog:            http://eavesdrop.openstack.org/meetings/ptl_sync/2015/ptl_sync.2015-03-17-09.00.log.html17:35
SlickNikThanks ttx!17:35
notmynamettx: still around?19:17
ttxnotmyname: yes19:43
notmynamesee #openstack-dev19:43
*** asalkeld has joined #openstack-relmgr-office20:48
*** mestery has quit IRC20:55
*** mestery has joined #openstack-relmgr-office20:59
*** asalkeld has left #openstack-relmgr-office21:01
*** david-lyle is now known as david-lyle_afk23:50

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!