| jroll | devananda: that sounds correct, and we've decided that | 00:00 |
|---|---|---|
| *** doug-fish has left #openstack-infra | 00:00 | |
| JayF | devananda: that's how the jobs are running now | 00:00 |
| devananda | fantastic | 00:00 |
| jroll | devananda: but now we're deciding should devstack changes run ironic+agent | 00:00 |
| JayF | devananda: agent_ssh is tip-ironic + stable-IPA, agent_ssh-src is tip-IPA + master-ironic | 00:00 |
| devananda | oooh | 00:00 |
| *** dmsimard is now known as dmsimard_away | 00:00 | |
| JayF | s/stable-IPA/master-IPA/ | 00:00 |
| *** asettle has quit IRC | 00:00 | |
| JayF | because it uses the tarballs.o.o images | 00:00 |
| mtreinish | fungi: so did switching things over to prod fix it? | 00:00 |
| *** alex7376 is now known as asettle | 00:00 | |
| devananda | I would say, no | 00:00 |
| devananda | don't gate devstack on an IPA-specific job | 00:01 |
| jroll | devananda: devstack has IPA-specific code | 00:01 |
| fungi | mtreinish: about to find out. the 00:00 pulse only just ran, which is the first one since that change merged | 00:01 |
| devananda | jroll: so? | 00:01 |
| mtreinish | fungi: ah ok, I misread the timestamps... | 00:01 |
| jroll | devananda: so devstack changes can break that job | 00:01 |
| jroll | I'm not saying that there's a large chance, and I'm not sure if IPA job should vote on devstack | 00:02 |
| jroll | I think JayF is saying the same | 00:02 |
| devananda | jroll: sure. I mean, run the job, but it shouldn't be voting | 00:02 |
| JayF | I came here to ask exactly ^ that | 00:02 |
| jroll | this is why we're asking the channel :) | 00:02 |
| jroll | ok | 00:02 |
| jroll | yeah, ironic doesn't even vote, does it? | 00:02 |
| JayF | so I'm going to make my existing project-config PR do this: make IPA and Ironic voting (people can +1/-1 as to their opinion of if it's stale enough) | 00:03 |
| *** baoli has quit IRC | 00:03 | |
| JayF | and make a new project-config PR to add the agent_ssh-src job as nv to devstack | 00:03 |
| devananda | also, honestly, I'd rather it not run there at all. because that will cause IPA to enter the "serialized gate" | 00:03 |
| mtreinish | jroll, JayF: yeah I would suggest just use experimental, or have it nonvoting | 00:03 |
| jroll | yeah, either is fine with me | 00:03 |
| JayF | mtreinish: devananda: So I'll just ignore devstack today, and if Ironic gate gets broken by devstack I reserve the right to be slightly more annoyed than usual ;) | 00:03 |
| jroll | sure | 00:03 |
| jroll | devstack is easy to fix | 00:03 |
| devananda | s/taht will/I think that will, if I understand the way the gate serialization works, which actually I don't/ | 00:04 |
| fungi | devananda: though if ipa shares a job with ironic and ironic is in the serialized gate, then that's happening anyway | 00:04 |
| *** lttrl has joined #openstack-infra | 00:04 | |
| clarkb | fungi: where are we on locking bug tackers down and resyncing? is that a tomorrow morning thing? | 00:04 |
| fungi | clarkb: it's a tomorrow thing | 00:05 |
| devananda | fungi: bah humbug. can we unserialize all the things? | 00:05 |
| clarkb | devananda: this is why I get so confuzzled talking to you about big tent and gating :) | 00:05 |
| devananda | clarkb: also, I change my mind sometimes | 00:05 |
| mtreinish | fungi: ooh, can't find the yaml files: http://puppetboard.openstack.org/report/subunit-worker01.openstack.org/f67e176df6e8e1e02a9922445176b50b9e81fa62 | 00:06 |
| *** banix has quit IRC | 00:06 | |
| *** david-lyle is now known as david-lyle_afk | 00:06 | |
| fungi | mtreinish: yep! | 00:06 |
| *** rmcall has joined #openstack-infra | 00:06 | |
| fungi | mtreinish: that's from just now when i attempted to apply | 00:06 |
| devananda | clarkb: like, I totally want ironic and nova to co-gate -- just until their driver API is deemed as stable. and then I don't. | 00:06 |
| *** stevemar has joined #openstack-infra | 00:07 | |
| devananda | *wont | 00:07 |
| jroll | I wish all the APIs were stable | 00:07 |
| clarkb | right I get the intents. I just find a lot of people talking about this hand wave around what it actually means in the real gate | 00:07 |
| jroll | and (micro-)versioned | 00:07 |
| fungi | mtreinish: Could not evaluate: Could not retrieve information from environment production source(s) puppet:///modules/openstack_project/logstash/jenkins-subunit-worker.yaml | 00:07 |
| clarkb | what it actually means is no transitive integration tests | 00:07 |
| devananda | clarkb: yep! | 00:07 |
| clarkb | whcih is not what you want | 00:07 |
| devananda | wait. I think that is what I want | 00:07 |
| clarkb | because you have to hvae integration testing somewhere | 00:08 |
| mtreinish | fungi: did I screw up the path or filename? because I'm pretty sure it's in modules/openstack_project/files/logstash/ | 00:08 |
| clarkb | devananda: you never want to test that a cloud with ironic in it can boot nodes? | 00:08 |
| clarkb | you only want to test ironic in a bubble? | 00:08 |
| mtreinish | fungi: doh: s/jenkins-subunit-worker.yaml/jenkins-subunit.worker.yaml | 00:08 |
| devananda | example: if we integration test A+B and B+C, then on a change in C, I don't think we should runthe tests for A | 00:08 |
| clarkb | devananda: yes but thats not cogating... | 00:08 |
| clarkb | arg | 00:08 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Fix test to tolerate new GitPython version https://review.openstack.org/135434 | 00:08 |
| clarkb | you are still cogating in that case | 00:08 |
| fungi | mtreinish: ahh, yep. that took some squinting to notice | 00:08 |
| mtreinish | fungi: let me push a patch with the git mv | 00:08 |
| clarkb | you don't have to run the tests in that particular run against that particular change | 00:09 |
| clarkb | but you still line everything up in the same serialized but tested in parallel queue | 00:09 |
| devananda | ahhhh right | 00:09 |
| *** lttrl has quit IRC | 00:09 | |
| devananda | words. cogating || serialized | 00:09 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Add gerrit owner information to status.json https://review.openstack.org/135386 | 00:09 |
| *** lttrl has joined #openstack-infra | 00:09 | |
| * devananda has no answers | 00:10 | |
| fungi | only more questions? | 00:10 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Rename subunit worker yaml file https://review.openstack.org/135466 | 00:11 |
| mtreinish | fungi, clarkb: ^^^ | 00:11 |
| fungi | i have questions which sound like answers until you think about them and then realize that you're on an existential trip to questioning your questions | 00:11 |
| *** bswartz has joined #openstack-infra | 00:11 | |
| openstackgerrit | Jay Faulkner proposed openstack-infra/project-config: Make agent_ssh check job voting for Ironic & IPA https://review.openstack.org/134436 | 00:12 |
| JayF | devananda, others ^ | 00:12 |
| *** ZZelle has quit IRC | 00:12 | |
| *** iain-king has joined #openstack-infra | 00:14 | |
| *** gfidente has quit IRC | 00:14 | |
| *** camunoz has quit IRC | 00:15 | |
| *** r-daneel has quit IRC | 00:17 | |
| *** pc_m has quit IRC | 00:18 | |
| openstackgerrit | Devananda van der Veen proposed openstack-infra/project-config: Update Ironic jobs post-graduation https://review.openstack.org/126627 | 00:21 |
| devananda | also, ^ should finally enable voting by Ironic | 00:21 |
| devananda | mikal: your opinion on ^ is requested | 00:22 |
| clarkb | fungi: so do we have working subunit workers yet? | 00:23 |
| mtreinish | clarkb: one more patch: https://review.openstack.org/135466 | 00:23 |
| fungi | what mtreinish said | 00:23 |
| clarkb | approved | 00:23 |
| *** cikid has quit IRC | 00:23 | |
| *** liam_ has joined #openstack-infra | 00:24 | |
| adam_g | wtf | 00:25 |
| *** liam_ is now known as Guest50977 | 00:25 | |
| adam_g | flood has hit #openstack-ironic | 00:25 |
| devananda | erm | 00:25 |
| adam_g | clarkb, ^ | 00:25 |
| *** yamamoto has joined #openstack-infra | 00:25 | |
| JayF | We need some ops in #openstack-ironic | 00:25 |
| * JayF volunteers to actually get that access to kickban people | 00:25 | |
| JayF | both me and jroll know how to IRC admin, yo :) | 00:25 |
| jroll | lol | 00:26 |
| clarkb | hrm I am not in there /me dons cape | 00:26 |
| *** unicell has quit IRC | 00:26 | |
| *** Guest50977 has quit IRC | 00:26 | |
| openstackgerrit | Merged openstack-infra/system-config: Rename subunit worker yaml file https://review.openstack.org/135466 | 00:29 |
| openstackgerrit | Jay Faulkner proposed openstack-infra/project-config: Make agent_ssh check job voting for Ironic & IPA https://review.openstack.org/134436 | 00:30 |
| *** stevemar has quit IRC | 00:30 | |
| *** yamahata has joined #openstack-infra | 00:30 | |
| *** r1chardj0n3s has joined #openstack-infra | 00:31 | |
| r1chardj0n3s | hullo infra folk | 00:31 |
| *** banix has joined #openstack-infra | 00:31 | |
| *** MarkAtwood has quit IRC | 00:31 | |
| r1chardj0n3s | don't know whether you've been following the Horizon/angular mega-thread | 00:31 |
| clarkb | I read pieces of it | 00:31 |
| r1chardj0n3s | recently there was a proposal to stick with bower for handling 3rd party web packages | 00:31 |
| *** nelsnelson has quit IRC | 00:32 | |
| r1chardj0n3s | and introduce bower into global-requirements | 00:32 |
| r1chardj0n3s | would like to get your thoughts on that :) | 00:32 |
| *** camunoz has joined #openstack-infra | 00:32 | |
| clarkb | r1chardj0n3s: in atlanta I basically decided horizon was insane | 00:32 |
| clarkb | and let them xstatic | 00:32 |
| *** banix has quit IRC | 00:32 | |
| clarkb | I would be so happy to not do that anymore | 00:32 |
| r1chardj0n3s | xstatic is crazy | 00:32 |
| r1chardj0n3s | hence flaming-shame | 00:32 |
| r1chardj0n3s | :) | 00:32 |
| clarkb | and if bower is the way to not do that then +2 | 00:33 |
| clarkb | especially since we already know how to make that owrk | 00:33 |
| r1chardj0n3s | https://github.com/r1chardj0n3s/flaming-shame | 00:33 |
| r1chardj0n3s | cool. does storyboard have a global-requirements thing for bower? | 00:33 |
| clarkb | r1chardj0n3s: no because it doesn't participate in global reqs | 00:33 |
| clarkb | it isn't part of the the thing people ship and call openstack | 00:34 |
| clarkb | so it just did it | 00:34 |
| r1chardj0n3s | ah, ok | 00:34 |
| clarkb | but horizon is part of that thing so we will need to global reqs this stuff | 00:34 |
| clarkb | but I don't oppose it since it seems to be working well for storyboard | 00:34 |
| r1chardj0n3s | it is the sensible way of handling 3rd party web components | 00:34 |
| krotscheck | Ehn, the bower and npm repos occasionally go down. | 00:34 |
| krotscheck | Which requires node. | 00:34 |
| krotscheck | Which requires a compile on centos. | 00:35 |
| r1chardj0n3s | yes | 00:35 |
| krotscheck | Which makes RedHat unhappy | 00:35 |
| clarkb | does centos7 not do node properly? | 00:35 |
| krotscheck | Nope. | 00:35 |
| clarkb | centos6 definitely doesn't | 00:35 |
| clarkb | :/ | 00:35 |
| r1chardj0n3s | node is inevitable, storyboard has gone through that | 00:35 |
| r1chardj0n3s | well, to an extent it has, I realise it's "not openstack" so I should stop holding it up like that | 00:35 |
| krotscheck | r1chardj0n3s: I had this battle with the Redhat package guy in Atlanta. | 00:35 |
| r1chardj0n3s | ok, so the Linux folk just need to pull their heads in and accept that Node Is A Thing Now | 00:36 |
| krotscheck | r1chardj0n3s: His response was “OMG Node Sucks No We’re Never Including It" | 00:36 |
| r1chardj0n3s | sorry, but I can't put it more nicely | 00:36 |
| krotscheck | r1chardj0n3s: By all means. I have his email address. | 00:36 |
| krotscheck | If you can convince him, I will make you storyboard core on the spot. | 00:36 |
| *** yamahata has quit IRC | 00:36 | |
| r1chardj0n3s | redhat already packages *some* node-based tools | 00:36 |
| * krotscheck doesn’t actually have that power. | 00:36 | |
| *** banix has joined #openstack-infra | 00:37 | |
| r1chardj0n3s | (like, they have a *jasmine* package, but not a *karma* one) | 00:37 |
| *** yamahata has joined #openstack-infra | 00:37 | |
| clarkb | so another thing to keep in mind is that the project doesn't have to bend over for everything | 00:37 |
| clarkb | we try to be nice, but sometimes you have to draw a line in the sand | 00:37 |
| r1chardj0n3s | which project, Horizon, OpenStack or RedHat? | 00:37 |
| *** Sincler has quit IRC | 00:37 | |
| clarkb | OpenStack | 00:37 |
| r1chardj0n3s | yes | 00:37 |
| clarkb | might be worth asking TC to chime in? | 00:37 |
| r1chardj0n3s | agreed | 00:37 |
| *** todd_dsm has joined #openstack-infra | 00:37 | |
| clarkb | I feel like settling these larger ecosystem fights is one of the things we have them for | 00:37 |
| krotscheck | Is EPEL allowed? | 00:37 |
| *** yamamoto has quit IRC | 00:38 | |
| r1chardj0n3s | EPEL? | 00:38 |
| krotscheck | Because apparently there’s a node package for centos on that. | 00:38 |
| clarkb | krotscheck: yes I think so | 00:38 |
| krotscheck | Extra Packages For Enterprise Linux | 00:38 |
| r1chardj0n3s | oh, ok | 00:38 |
| krotscheck | So sudo yum install epel-release; sudo yum install nodejs | 00:38 |
| krotscheck | And suddenly that would do all the things. | 00:39 |
| clarkb | so I support our new bower overlords | 00:39 |
| clarkb | pretty sure mordred and jeblair do too | 00:39 |
| r1chardj0n3s | krotscheck: but that also implies that the node-based tools are packaged as well | 00:40 |
| krotscheck | r1chardj0n3s: yep | 00:40 |
| krotscheck | Spinning up a VM now to check. | 00:40 |
| r1chardj0n3s | well, at least karma, jasmine, protractor, selenium and phantomjs | 00:40 |
| *** Sukhdev has joined #openstack-infra | 00:40 | |
| r1chardj0n3s | those are the minimum requirements for horizon | 00:40 |
| clarkb | selenium is node now? | 00:40 |
| r1chardj0n3s | grunt we can do without | 00:40 |
| r1chardj0n3s | er | 00:40 |
| clarkb | we have used selenium for a long time iwthout worrying about node | 00:40 |
| r1chardj0n3s | there's a node-based selenium, yes? | 00:40 |
| clarkb | oh js bindings likely | 00:41 |
| r1chardj0n3s | ok, mebbe I take that back | 00:41 |
| r1chardj0n3s | yeah, bindings | 00:41 |
| r1chardj0n3s | which is typically "selenium for langauge x" | 00:41 |
| mtreinish | fungi, clarkb: ooh, it worked http://puppetboard.openstack.org/report/subunit-worker01.openstack.org/effcaf9d927240ac2af52ff70633b695598bbc42 | 00:41 |
| krotscheck | There’s an NPM package that includes selenium. | 00:41 |
| *** mmaglana has joined #openstack-infra | 00:41 | |
| r1chardj0n3s | yeah, but we can't NPM | 00:41 |
| krotscheck | That’s what storyboard uses. | 00:41 |
| krotscheck | r1chardj0n3s: Why not? | 00:41 |
| r1chardj0n3s | which is exciting given how incompatible NPM and deb/rpm are | 00:41 |
| r1chardj0n3s | because LINUX | 00:41 |
| r1chardj0n3s | ugh | 00:41 |
| clarkb | I think you can npm | 00:41 |
| *** banix has quit IRC | 00:41 | |
| clarkb | you npm in virtualenvs like storyboard | 00:41 |
| clarkb | (at least in unittesting land) | 00:42 |
| r1chardj0n3s | no, we have people who have drawn the line in the sand for OpenStack: if it's not in a distro-installable format, it's not allowed | 00:42 |
| r1chardj0n3s | so, no install from npm | 00:42 |
| clarkb | well sort of | 00:42 |
| r1chardj0n3s | (or bower, hence we have to repackage all that as well) | 00:42 |
| krotscheck | Urr… | 00:42 |
| clarkb | for our python deps we don't really say this has to already be in the distros | 00:42 |
| krotscheck | Ok, so if NPM is installable via distro | 00:42 |
| clarkb | its nice | 00:43 |
| clarkb | but sometimes isn't already the case | 00:43 |
| r1chardj0n3s | that's how we've gotten up to 93 messages in the angular/horizon discussion - it's been entirely because of that line in the sand | 00:43 |
| fungi | we can npm things to run test jobs. distros aren't going to ship anything which expects you to npm its dependencies | 00:43 |
| *** oomichi has joined #openstack-infra | 00:43 | |
| clarkb | right so distros would have to have js-foo packages for the things tests npm install | 00:43 |
| clarkb | which is no different than the python-foo packages | 00:43 |
| krotscheck | But they allow gem? Wat? | 00:43 |
| r1chardj0n3s | but apparently deployers wish to modify openstack and run tests and still not hit npm | 00:43 |
| r1chardj0n3s | hmm | 00:44 |
| clarkb | but I don't think we necessarily have to wait on the distros | 00:44 |
| r1chardj0n3s | I thought that openstack was sans gem/pypi once baked | 00:44 |
| fungi | exactly. things we need from npm now will get subsequently packaged by the distro packagers into their various preferred formats | 00:44 |
| clarkb | we certianly never have for the python things | 00:44 |
| clarkb | xstatic for example | 00:44 |
| clarkb | we threw a ton of new deps at distros | 00:44 |
| r1chardj0n3s | xstatic is repackaged by distros | 00:44 |
| clarkb | r1chardj0n3s: yes but not before we started using it | 00:44 |
| krotscheck | r1chardj0n3s: It is. And if you’re using npm and bower post bake, then you’r eusing those tools wrong. | 00:44 |
| r1chardj0n3s | yes | 00:44 |
| r1chardj0n3s | krotscheck: see my previous point: we bake, the deployer modifies and wants to *re-bake* without hitting npm. that's the story I've been told | 00:45 |
| jroll | that's absurd | 00:45 |
| krotscheck | r1chardj0n3s: Don’t they have to hit gem? | 00:45 |
| clarkb | or pypi | 00:45 |
| krotscheck | If they want to rebake? | 00:45 |
| fungi | in our ci toolchain we install all sorts of things from pypi and potentially other places. the distro packagers understand they'll need to repackage suitable versions of all that before release | 00:45 |
| jroll | why are you using distro packages if you're modifying the source | 00:45 |
| r1chardj0n3s | I cannot say | 00:45 |
| *** mmaglana has quit IRC | 00:45 | |
| r1chardj0n3s | sorry folks, but I'm simply too new to openstack to know the ins and outs of such deployer issues :( | 00:46 |
| r1chardj0n3s | I can't make the argument, I've only had it presented to me | 00:46 |
| clarkb | r1chardj0n3s: its ok I often say things that distro landers all look at me funny for | 00:46 |
| jroll | that's ok | 00:46 |
| clarkb | r1chardj0n3s: like apparnetly you can only build packages with other packaged things | 00:46 |
| jroll | I'm saying that's an invalid argument (I may be wrong) | 00:46 |
| clarkb | bootstrapping outside of package land is apparently the most evil thing ever | 00:46 |
| krotscheck | And let’s be honest here. By “Distro landers” we mean RedHat, right? | 00:46 |
| clarkb | krotscheck: well debian and canonical yell a fair bit too | 00:47 |
| *** banix has joined #openstack-infra | 00:47 | |
| mtreinish | fungi: does it look like things are running on subunit-worker01 | 00:47 |
| clarkb | they just seem to accept that the world moves forward a bit more than red hat | 00:47 |
| *** rmcall has quit IRC | 00:47 | |
| fungi | it's pretty standard that all distros want to be "self-hosting" (e.g. they can rebuild all their packages from source using their packages) | 00:47 |
| jroll | what I would do is to use npm and bower to build some artifact that ends up in the package, but idk what packagers would say about that | 00:47 |
| clarkb | fungi: ya I don't think its a bad thing | 00:47 |
| clarkb | I am just still baffled its the only way to do things | 00:47 |
| r1chardj0n3s | jroll: that would result in a ton of vendoring due to the way npm works | 00:47 |
| clarkb | fungi: and is one reason I think people love the image based deployment stuff so much | 00:48 |
| jroll | in web dev, people generally just build a tarball with dependencies and minified js/etc | 00:48 |
| r1chardj0n3s | npm is built over things having many, many very small dependencies | 00:48 |
| jroll | r1chardj0n3s: I understand, that's fine | 00:48 |
| clarkb | fungi: because then you can sort of avoid the package/distro problems and go straight to useable thing | 00:48 |
| fungi | they will extract everything they need which is hosted on the node, ruby, python et cetera package channels and repackage that source into something they can distribute. even things which are only needed to build other things, or to generate baked documentation, or to run tests | 00:48 |
| r1chardj0n3s | it'd all be so much simpler if we could just require that deployers modifying horizon have access to npm/bower | 00:49 |
| openstackgerrit | Merged openstack-infra/devstack-gate: Remove nosetests.xml reference https://review.openstack.org/131714 | 00:49 |
| *** rmcall has joined #openstack-infra | 00:49 | |
| clarkb | ianw: ^ | 00:49 |
| jroll | r1chardj0n3s: I think that's completely reasonable | 00:49 |
| fungi | so that you can rebuild any distro package, or test the source of any distro-provided source package, by only installing things provided from the tistro | 00:49 |
| *** rmcall has quit IRC | 00:49 | |
| r1chardj0n3s | but that would also imply that the bower packages are baked into a Horizon release package | 00:49 |
| r1chardj0n3s | which goes against the recent move to having them external through xstatic | 00:50 |
| krotscheck | I feel that ultimately the only way that’s going to work is if the horizon UI is extracted from openstack entirely and left as a stackforge thing. | 00:50 |
| r1chardj0n3s | and I'm pretty sure david-lyle_afk will have a conniption | 00:50 |
| fungi | mtreinish: yeah, no more puppet errors now | 00:50 |
| r1chardj0n3s | krotscheck: wow, that there's crazy talk :) | 00:50 |
| krotscheck | r1chardj0n3s: It solves the problem. | 00:50 |
| fungi | mtreinish: though i don't see the worker process running | 00:51 |
| r1chardj0n3s | horizon is so very unlike the rest of openstack, it makes a lot of sense | 00:51 |
| krotscheck | r1chardj0n3s: Remove the expectation that horizon-ui is part of openstack, and that you have to go build the UI yourself. | 00:51 |
| r1chardj0n3s | which other component of OpenStack is *designed* to be modified by deployers? | 00:51 |
| fungi | r1chardj0n3s: arguably, all of them? | 00:51 |
| adam_g | r1chardj0n3s, IIRC at one point horizon was bundling their own version of node-less | 00:51 |
| mtreinish | fungi: hmm, ok | 00:51 |
| r1chardj0n3s | really? not just configuration? | 00:51 |
| fungi | r1chardj0n3s: most of our largest deployers of openstack clouds are running forks | 00:52 |
| mtreinish | fungi: any ideas? | 00:52 |
| fungi | mtreinish: i'm checking logs now | 00:52 |
| clarkb | adam_g: it was the python compiler version thing | 00:52 |
| r1chardj0n3s | fungi: ah, ok | 00:52 |
| clarkb | but it didn't keep up with less proper | 00:52 |
| clarkb | and now they use pyscss or something | 00:52 |
| jroll | r1chardj0n3s: those deployers are also ok with using pypi, btw | 00:52 |
| jroll | and presumably npm | 00:52 |
| r1chardj0n3s | fungi: but I think that the linux-packaging argument isn't about the large deployers who run forks | 00:52 |
| adam_g | clarkb, i seem to remember it being a node-based css/js compiler | 00:52 |
| r1chardj0n3s | jroll: yeah | 00:52 |
| clarkb | adam_g: they wanted less css without the node so they went to the python compiler | 00:52 |
| clarkb | iirc | 00:53 |
| adam_g | clarkb, eventually, yeah | 00:53 |
| fungi | mtreinish: i manually started one via its initscript and its running now... it just didn't get started by puppet | 00:53 |
| clarkb | fungi: mtreinish ya I think thats normal for those services | 00:53 |
| fungi | r1chardj0n3s: agreed, the packaging argument is orthogonal to whether or not they're modifying the applications they're installing | 00:53 |
| clarkb | because we have a habit of not auto starting services | 00:53 |
| adam_g | clarkb, i just remember having to manually compile the js and css into the source package to avoid a dependency on node.js.. and getting yelled at by debian for it | 00:53 |
| fungi | mtreinish: i went ahead and started the other three now | 00:53 |
| clarkb | adam_g: oh prior to the python compiler? | 00:54 |
| clarkb | thats "awesome" | 00:54 |
| adam_g | clarkb, yeah, this is way back | 00:54 |
| mtreinish | fungi: ok awesome! | 00:54 |
| mtreinish | now I guess we'll see how badly I screwed up on subunit2sql | 00:54 |
| mtreinish | and the worker script | 00:54 |
| fungi | adam_g: well, they want to have separate packages for that stuff, built independently from source | 00:54 |
| r1chardj0n3s | hmm, thanks for the discussion folks. I suspect things will become significantly less clear if I propose the approach discussed here to the mailing list, but who knows... | 00:55 |
| clarkb | r1chardj0n3s: so the thing I would do is put out feeler questions to TC maybe? | 00:55 |
| r1chardj0n3s | how would I do that? | 00:56 |
| fungi | mtreinish: is there anything else i should be looking at now to make sure this is working? | 00:56 |
| clarkb | honestly I think this is really starting to get into TC territory. external entities are demanding we do one thing at the expense of dev cost | 00:56 |
| r1chardj0n3s | I guess mikal is a TC, so I could start there | 00:56 |
| clarkb | TC has ability to say "yes" or "no" to people | 00:56 |
| clarkb | (hopefully TC people don't tell me I am wrong about that) | 00:56 |
| clarkb | but I don't think we ever solve this if we just let distros fight horizon devs | 00:57 |
| mtreinish | fungi: I believe the worker log file should print out when it gets an event and when it's adding it to the db | 00:57 |
| mtreinish | if it does that more than once I guess it's working | 00:57 |
| mtreinish | (short of looking at the db) | 00:57 |
| *** dannywil_ has quit IRC | 00:57 | |
| *** wenlock has quit IRC | 00:58 | |
| r1chardj0n3s | mordred is the only TC who's chimed in on the discussion IIRC and his statement was (paraphrased) "let's be a little pragmatic about it" though I think he was being overly optimistic about the distros packaging the node stuff | 00:58 |
| fungi | mtreinish: firewalls? | 00:58 |
| fungi | mtreinish: Unable to connect to <gear.Connection 0x2af66d0 host: logstash.openstack.org port: 4730> | 00:58 |
| clarkb | that said it is also one project and the TC may try to avoid invading on not really corss project things | 00:58 |
| clarkb | fungi: mtreinish oh ya since gear didn't have ACLs in the way before time firewall is used | 00:59 |
| clarkb | let me show you the line to updat that | 00:59 |
| mtreinish | clarkb: oh ok | 01:00 |
| fungi | clarkb: if we open that, will the logstash indexer which is also running on here connect and start running wild, or is that trying to connect to something other than gearman? | 01:00 |
| clarkb | https://git.openstack.org/cgit/openstack-infra/system-config/tree/manifests/site.pp#n311 its that line | 01:00 |
| clarkb | fungi: ya it will I think | 01:00 |
| clarkb | so you may want to use a a different variable than elasticsearch_cleitns | 01:01 |
| fungi | so we need that other cleanup too | 01:01 |
| clarkb | or extend that array before passing to gearman_works | 01:01 |
| *** prometheanfire has joined #openstack-infra | 01:01 | |
| r1chardj0n3s | ok, back to the infra specifics: is there any point to having a global-requirements for bower dependencies given that horizon is the only project that's going to have them? | 01:01 |
| clarkb | in other news https://github.com/openstack/openstack/graphs/commit-activity welcome back from the summit everyone :) | 01:01 |
| clarkb | r1chardj0n3s: I think initially no | 01:02 |
| clarkb | global-requirements was created to solve the cross project dependency mess | 01:02 |
| *** banix has quit IRC | 01:02 | |
| fungi | okay, so tomorrow i need to refresh the git repo for puppet-httpd, buy a cert for groups.o.o and update bug imports for all the infra projects before locking them down | 01:02 |
| r1chardj0n3s | global-requirements exists for two reasons, IIUC: cross-project dependency mess *and* vetting of new versions, or am I making up the latter? | 01:02 |
| clarkb | and if only one project needs a certain type of requirements it probably isn't necessary to have that sort of machinery | 01:03 |
| clarkb | r1chardj0n3s: it does exist for the second reason too | 01:03 |
| mtreinish | fungi, clarkb: I can append that line into https://review.openstack.org/135463 | 01:03 |
| r1chardj0n3s | but if it's just one project then that project can vet version bumps | 01:03 |
| prometheanfire | if it helps, I'm not going to be packaging horizon because of all those extra deps | 01:03 |
| clarkb | r1chardj0n3s: but that can happen in the individual project as well (assuming people do that) | 01:03 |
| fungi | mtreinish: fine by me | 01:03 |
| r1chardj0n3s | prometheanfire: hi, what perspective are you coming from? | 01:03 |
| clarkb | prometheanfire: "those extra deps" being the xstatic deps or npm js deps or? | 01:03 |
| *** camunoz has quit IRC | 01:03 | |
| * krotscheck wonders if he should just implement a sane horizon client and then make all his millions relicensing it. | 01:04 | |
| fungi | prometheanfire: you package for gentoo, right? | 01:04 |
| mtreinish | clarkb: err, do I just need to add subunit-work01 to: https://git.openstack.org/cgit/openstack-infra/system-config/tree/manifests/site.pp#n16 | 01:04 |
| r1chardj0n3s | krotscheck: I was already well on the way to doing that but stopped because community ;) | 01:04 |
| prometheanfire | xstatic | 01:04 |
| prometheanfire | npn is annoying | 01:04 |
| prometheanfire | I package for gentoo | 01:04 |
| clarkb | mtreinish: you need to add it to the thing passed to gearman_workers on the line I linked. the naive way is to put it in the array on line 16 but that may allow the logstash worker on that node to connect and do terrile thngs | 01:05 |
| r1chardj0n3s | prometheanfire: ok, so there's a couple of possibilities here | 01:05 |
| clarkb | prometheanfire: ok so its an issue regardless of python produced deps or npm produced deps. that is good to know | 01:05 |
| krotscheck | So horizon’s not getting packaged anyway? | 01:05 |
| clarkb | mtreinish: let me write a patch | 01:05 |
| clarkb | krotscheck: on gentoo | 01:06 |
| krotscheck | Urm. Well then. Why do we care? | 01:06 |
| mtreinish | clarkb: I'm removing the logstash indexer class from the subunit worker at the same time | 01:06 |
| r1chardj0n3s | oh | 01:06 |
| clarkb | mtreinish: well I think there is a simple intermediate step | 01:06 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Remove unnecessary use of logstash::indexer from subunit_worker https://review.openstack.org/135463 | 01:06 |
| clarkb | give me a couple minutes | 01:06 |
| prometheanfire | krotscheck: I was packaging it for icehouse | 01:06 |
| mtreinish | clarkb: ^^^ | 01:06 |
| clarkb | oh you alread did it | 01:06 |
| clarkb | nevermind then :) | 01:06 |
| mtreinish | clarkb: ok sure | 01:06 |
| krotscheck | prometheanfire: Intersting. | 01:06 |
| prometheanfire | not very well, but it was there | 01:06 |
| prometheanfire | I THINK we have the npm thing somewhat fixed in gentoo | 01:06 |
| r1chardj0n3s | prometheanfire: ok, so how would you feel about packaging a horizon that has been built to include all the bower-based components, and has no npm dependencies (but also can no longer run its unit test suite) | 01:07 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Remove unnecessary use of logstash::indexer from subunit_worker https://review.openstack.org/135463 | 01:07 |
| *** Longgeek has joined #openstack-infra | 01:07 | |
| mtreinish | clarkb: let me know if that doesn't work | 01:07 |
| clarkb | fungi: you or someone will need to stop the running indexer and remove it with fire if that change merges | 01:07 |
| clarkb | but I think that change works | 01:07 |
| clarkb | mtreinish: you do have trailing whitespace though :P | 01:07 |
| prometheanfire | r1chardj0n3s: I'd rather have tests (I tend to add a test use flag so end users can run tests as well), but the first thing is getting it working | 01:07 |
| fungi | clarkb: i can stop the indexer asap | 01:07 |
| prometheanfire | that's most important | 01:07 |
| clarkb | nevermind ou fixed it | 01:07 |
| clarkb | fungi: careful puppet may restart it | 01:08 |
| r1chardj0n3s | prometheanfire: ok, but the tests require npm-based tools; that's unavoidable | 01:08 |
| prometheanfire | right | 01:08 |
| clarkb | mtreinish: lgtm | 01:08 |
| prometheanfire | we have RESTRICT=test for a reason | 01:08 |
| r1chardj0n3s | sorry, I don't know what that means | 01:08 |
| prometheanfire | gentooism | 01:08 |
| r1chardj0n3s | (oh, I got that ;) | 01:08 |
| prometheanfire | we can restrict testing if we feel like it on a per package basis | 01:09 |
| r1chardj0n3s | ok | 01:09 |
| krotscheck | Ok, I’m outta here. | 01:09 |
| krotscheck | Good luck :) | 01:09 |
| *** ayoung has quit IRC | 01:09 | |
| prometheanfire | otoh, I will likely be packaging more openstack stuffs (like heat) | 01:09 |
| *** yamamoto_ has joined #openstack-infra | 01:09 | |
| fungi | honestly, since the earliest days where debian struggled to package gnu imp, wordpress, mediawiki et cetera, i've felt like the web application scene is incongruent with distro packaging principles. square peg, round hole | 01:11 |
| *** Longgeek has quit IRC | 01:12 | |
| prometheanfire | we have a specific eclass that handles upgrades and the like, helps a ton | 01:12 |
| prometheanfire | webapp-config -U -h owncloud.foo.com owncloud 7.0.3 | 01:13 |
| clarkb | do debian python packages ship the sdist then shove files where they need to go? | 01:13 |
| * clarkb admits to be very naive when it comes to gentoo packaging | 01:13 | |
| prometheanfire | don't worry, it sucks everywhere | 01:14 |
| clarkb | er not debian | 01:14 |
| clarkb | gentoo | 01:14 |
| fungi | the openstack packager for debian doesn't use sdists at all. rearranges files into the package directly from git clones | 01:14 |
| *** banix has joined #openstack-infra | 01:14 | |
| clarkb | ya debian does it that other way | 01:14 |
| clarkb | which is grab git repo, mangle it because lol, spit out packag | 01:14 |
| r1chardj0n3s | thanks krotscheck | 01:15 |
| prometheanfire | crazyness, though we do the same with our git versions (stable branch git based) | 01:15 |
| fungi | i think the primary argument used is that not having an intermediate tarball-producing or -retrieving step makes it easier to consume and return upstream patches | 01:15 |
| clarkb | fungi: I guess that makes sense | 01:16 |
| mtreinish | hmm the arch approach seems like how I would do it: https://projects.archlinux.org/svntogit/community.git/tree/trunk/PKGBUILD?h=packages/python-oslo-config | 01:16 |
| mtreinish | although I find the prepare() phase hysterical... | 01:16 |
| fungi | so instead patches and the upstream source live together in parallel branches and then the packaging workflow bundles them up relative to one another to accommodate their upstream/packaging modifications split | 01:17 |
| clarkb | mtreinish: except it uses python setup.py install | 01:17 |
| mtreinish | clarkb: as opposed to pip install? | 01:17 |
| *** lttrl has quit IRC | 01:17 | |
| clarkb | mtreinish: or some external tool ensuring things go where the distro actually wants them | 01:17 |
| clarkb | setup.py install is arbitrary code and I could decide on next release to just completely hose arch | 01:18 |
| clarkb | as root :) | 01:18 |
| prometheanfire | here's how we do it for comparison | 01:18 |
| prometheanfire | http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/dev-python/oslo-config/oslo-config-1.4.0.ebuild?view=markup | 01:18 |
| * fungi hopes you would at worst dirty a chroot | 01:18 | |
| prometheanfire | heh, I know that arch guy, sit about 100 feet from him | 01:18 |
| clarkb | fungi: iirc arch packages literally just run that as root | 01:18 |
| mtreinish | clarkb: heh, well it does it all in a chroot so damage would be limited | 01:18 |
| clarkb | in he base install | 01:18 |
| mtreinish | fungi: yep | 01:18 |
| clarkb | oh it is chroot? | 01:18 |
| mtreinish | clarkb: the pkg building is done in a chroot by default | 01:19 |
| clarkb | mtreinish: is that "new" I remember the days of yaourt and thinking why am I doing this | 01:19 |
| fungi | not that it's even remotely hard to escape from a chroot if you're root in it | 01:19 |
| mtreinish | it warns you about doing root, because there are a bunch of sec implications running commands even in a chroot | 01:19 |
| mtreinish | yaourt still builds things in a chroot | 01:20 |
| prometheanfire | we don't allow network access or any access outside that chroot... | 01:20 |
| fungi | yeah, for example direct access to memory structures | 01:20 |
| mtreinish | I don'y recall a time when it wasn't that way, but I've only been running arch since like 2010, so maybe before that | 01:20 |
| *** manfred has joined #openstack-infra | 01:20 | |
| clarkb | mtreinish: I may have een mistaken too. But I ran arch way before 2010 | 01:20 |
| manfred | prometheanfire: yo | 01:21 |
| fungi | being able to write to arbitrary things in the chroot's /dev, /proc, /sys et cetera (if mounted) can pretty well get you direct root in the outer context | 01:21 |
| *** todd_dsm has quit IRC | 01:21 | |
| prometheanfire | manfred: yo | 01:21 |
| prometheanfire | manfred is who packaged that | 01:21 |
| clarkb | mtreinish: the license is wrong in the arch package | 01:21 |
| prometheanfire | manfred: does arch do this stuff in chroot or something? http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/dev-python/oslo-config/oslo-config-1.4.0.ebuild?view=markup | 01:21 |
| prometheanfire | manfred: how does it keep the system safe? | 01:22 |
| mtreinish | clarkb: heh, yeah the arch packages for openstack stuff are all terrible | 01:22 |
| mtreinish | and out of date too | 01:22 |
| manfred | for building packages? | 01:22 |
| prometheanfire | manfred: you have work to do :P | 01:22 |
| prometheanfire | manfred: ya, for build | 01:22 |
| clarkb | ya if I yaourt install oslo-config or whatever the current tool is today | 01:22 |
| clarkb | I really should go install arch again just because | 01:23 |
| clarkb | its been so long | 01:23 |
| *** henditan has joined #openstack-infra | 01:23 | |
| mtreinish | clarkb: yaourt is just the automated tool for things installing things in the aur (which are user contributed and have to be built) | 01:23 |
| manfred | you should just use extra-$arch-build, it generates a clean environment in /var/lib/archbuild/ and only installs the base system, base-devel and then whatever dependencies are in the PKGBUILD | 01:23 |
| mtreinish | most of the openstack packages are in there | 01:23 |
| manfred | it is actually a systemd-nspawn container that it runs the makepkg inside of | 01:24 |
| manfred | that is how we package them for the repositories at least, and at this point... that is how I build for myself as well from the aur | 01:24 |
| *** camunoz has joined #openstack-infra | 01:24 | |
| fungi | containerization alleviates a lot of those concerns, so at least it's not just a plan chroot | 01:25 |
| clarkb | ah yes I arched long before systemd | 01:25 |
| fungi | not that lxc isn't still riddled with context-escaping holes | 01:25 |
| openstackgerrit | Merged openstack-infra/system-config: Remove unnecessary use of logstash::indexer from subunit_worker https://review.openstack.org/135463 | 01:25 |
| clarkb | I arched when arch had an intaller | 01:25 |
| manfred | https://projects.archlinux.org/devtools.git/tree/arch-nspawn.in#n99 | 01:26 |
| mtreinish | manfred: oh wait you're the maintainer for that package I linked aren't you? :) | 01:26 |
| manfred | oslo-config? | 01:26 |
| prometheanfire | mtreinish: told you :P | 01:26 |
| mtreinish | yeah | 01:26 |
| manfred | yeah i think so | 01:26 |
| r1chardj0n3s | lol, every discussion I start these days ends up involving packaging minutae :) | 01:26 |
| r1chardj0n3s | we suck at software :) | 01:26 |
| fungi | r1chardj0n3s: our entire species, yes | 01:27 |
| clarkb | I am willing to bet other species capable of writing and executing software suck at it too | 01:27 |
| mtreinish | prometheanfire: sorry completely missed that :) | 01:27 |
| mtreinish | fungi: ok the change landed you have to kill the indexer again I think | 01:29 |
| mtreinish | it looks like puppet restarted it on the last run | 01:29 |
| *** ddieterly has joined #openstack-infra | 01:29 | |
| manfred | what about it? | 01:29 |
| fungi | mtreinish: yep, it did restart it, and i've stopped it again | 01:29 |
| mtreinish | fungi: ok cool | 01:29 |
| mtreinish | manfred: oh, nothing in particular. I was just slow and realized it after we were talking | 01:30 |
| clarkb | mtreinish: but yes pip install is much better | 01:30 |
| *** ddieterl_ has joined #openstack-infra | 01:30 | |
| manfred | ok :) | 01:30 |
| clarkb | but maybe I have just gotten so annoyed at easy_install that I do everything I can to avoid it | 01:30 |
| manfred | i was just told to come here... something about arch packaging <3 | 01:30 |
| *** unicell has joined #openstack-infra | 01:30 | |
| StevenK | Avoiding easy_install sounds like a brilliant plan | 01:31 |
| fungi | not_so_easy_install | 01:31 |
| fungi | manfred: yeah, we were mostly debating the security constraints around the arch package building, as it pertains to upstream code running on the package builder | 01:32 |
| clarkb | because I said don't run setup.py its arbitrary code | 01:33 |
| *** sarob has quit IRC | 01:33 | |
| prometheanfire | manfred: might be all that was | 01:33 |
| *** ddieterly has quit IRC | 01:33 | |
| clarkb | fungi: so is logstash subunit stuff doing things now? | 01:34 |
| adam_g | anyone free to tag this quick DSG patch? the multi node patch that merged today broke debug log archiving for ironic, which we're relying on to help fix a current gate bug. https://review.openstack.org/#/c/135418/ | 01:34 |
| fungi | clarkb: yeah, though i see exceptions getting thrown now | 01:35 |
| mtreinish | adam_g: +1 ffwiw | 01:35 |
| mtreinish | fungi: ooh, pastebin? | 01:35 |
| *** mwagner_lap has joined #openstack-infra | 01:36 | |
| clarkb | hrm did that break grenade too? | 01:36 |
| clarkb | adam_g: ^ | 01:36 |
| adam_g | mtreinish, thanks! | 01:36 |
| adam_g | clarkb, im not sure | 01:36 |
| adam_g | clarkb, kinda unclear where the the fallout from the grenade part of that would be showing | 01:36 |
| adam_g | it seems like it would have broken those runs entirely | 01:36 |
| clarkb | I have approved the change | 01:36 |
| clarkb | adam_g: ya but it didn't ... | 01:36 |
| clarkb | maybe greande doesn't rely on that info in localrc? | 01:37 |
| fungi | mtreinish: http://paste.openstack.org/show/134563/ | 01:37 |
| *** Sukhdev has quit IRC | 01:37 | |
| mtreinish | fungi: hmm, it's not where I expected it to fail | 01:38 |
| fungi | wow... first joey hess officially abdicates further debian community involvement, now russ allbery and colin watson are stepping down from the tech-ctte. shakeup | 01:39 |
| adam_g | clarkb, hmm *_PHASE may not be used at all, lookin now | 01:40 |
| *** gyee has quit IRC | 01:40 | |
| mtreinish | fungi: is that happening all the time? | 01:41 |
| fungi | mtreinish: looks like it's happening for every file | 01:41 |
| *** mmaglana has joined #openstack-infra | 01:42 | |
| mtreinish | I'm thinking it's related to how the script is being invoked in the daemon. Something with subprocess needs a cwd | 01:42 |
| mtreinish | because I didn't hit that with my local testing but I wasn't running the daemon | 01:42 |
| *** ChuckC has quit IRC | 01:43 | |
| mtreinish | clarkb: have you tried using subprocess inside of python-daemon before? | 01:43 |
| clarkb | mtreinish: the logstash stuff does it for crm114 | 01:44 |
| clarkb | jeblair: wrote that for the most part though | 01:44 |
| *** mmaglana has quit IRC | 01:46 | |
| *** shashankhegde has quit IRC | 01:47 | |
| adam_g | clarkb, see how this works out.. https://review.openstack.org/#/c/135479/ i believe the phase is determined by *_DEVSTACK_DIR/branch now | 01:47 |
| clarkb | cool | 01:48 |
| mtreinish | oh, I wonder if it's because I don't have the path for subunit-2to1 in the subprocess call | 01:48 |
| clarkb | mtreinish: that could be | 01:48 |
| clarkb | espeially if the init scripts set path to not include /usr/local/bin | 01:48 |
| fungi | mtreinish: likely it isn't in a path inherited in the subprocess environment | 01:48 |
| fungi | checking | 01:49 |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 01:49 | |
| fungi | according to proc/$PID/environ the daemon has PATH=/sbin:/usr/sbin:/bin:/usr/bin | 01:50 |
| fungi | so yes, that's your trouble | 01:50 |
| mtreinish | fungi: yep that would explain it | 01:50 |
| mtreinish | ok should I do the cheap fix and hard code the patch of the worker to /usr/local/bin | 01:50 |
| *** manfred has left #openstack-infra | 01:50 | |
| mtreinish | or make it not pip installed subunit on ubuntu specific? | 01:51 |
| fungi | you could instead extend the path in the initscript, right? | 01:51 |
| mtreinish | fungi: oh, yeah that should work | 01:51 |
| clarkb | I think we went that route for dib in nodepool | 01:52 |
| clarkb | let PATH do its thing | 01:52 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add /usr/local/bin to subunit-worker init PATH https://review.openstack.org/135480 | 01:52 |
| mtreinish | fungi, clarkb: ^^^ | 01:53 |
| *** banix has quit IRC | 01:53 | |
| mtreinish | I'm really digging the stat padding this week :) | 01:53 |
| fungi | mtreinish: infra is all about death by a thousand patches | 01:53 |
| fungi | (or testing your stuff in advance) | 01:54 |
| *** baoli has joined #openstack-infra | 01:54 | |
| mtreinish | fungi: I did test things :) but I guess just not enough | 01:54 |
| *** zz_dimtruck is now known as dimtruck | 01:55 | |
| fungi | you tested other things, which worked without any fuss | 01:55 |
| fungi | it's the things you didn't test which were broken ;) | 01:55 |
| fungi | i'm sure if you'd not tested those other things, they'd also have been broken | 01:56 |
| mtreinish | or my test was broken/incomplete (like here because I tested the worker script by just running it without init) | 01:56 |
| mtreinish | oh yeah I'm 100% on that | 01:56 |
| fungi | deploying and running software is complicated, no doubt about it. lots of surface area to collect bugs | 01:57 |
| mgagne | I see most projects are running run_tests.sh with set -e and some are collecting exit status with $?. How is that possible if set -e is used? I don't see them disabling it with set +e. | 02:00 |
| *** camunoz has quit IRC | 02:00 | |
| *** nosnos has joined #openstack-infra | 02:02 | |
| fungi | good question. most of us in here just run tox and see run_tests.sh as an anathema | 02:02 |
| fungi | we mostly don't use any project's run_tests.sh (with a couple exceptions) in upstream ci jobs, and so i assume they're in various degrees of broken | 02:03 |
| mgagne | fungi: ok so set -e was introduced, people added checks on $?, tests were passing all fine and they didn't see that failing command would break their logic on $? | 02:03 |
| fungi | mgagne: that would be my first guess, yeah | 02:04 |
| *** achanda_ has quit IRC | 02:04 | |
| mgagne | fungi: alright, I wasn't crazy then. thanks | 02:04 |
| fungi | probably the exit code handling logic predates someone adding set -e (or someone removing a set +e somewhere near the end) | 02:04 |
| *** ayoung has joined #openstack-infra | 02:05 | |
| openstackgerrit | Merged openstack-infra/system-config: Add /usr/local/bin to subunit-worker init PATH https://review.openstack.org/135480 | 02:07 |
| mtreinish | ok, here we go again... | 02:07 |
| *** bdpayne_ has quit IRC | 02:07 | |
| mtreinish | although I think that just missed the last puppet refresh | 02:08 |
| *** amcrn has quit IRC | 02:08 | |
| *** Ryan_Lane has quit IRC | 02:11 | |
| *** Ryan_Lane has joined #openstack-infra | 02:11 | |
| fungi | yeah, the puppetmaster will update its system-config clone at 02:15 and start the apply run to all the servers again then | 02:11 |
| * mtreinish gets another beer | 02:12 | |
| fungi | good idea | 02:12 |
| *** tsg has quit IRC | 02:13 | |
| mtreinish | my apt is super hot today, the heat went out while I was in Paris and air got in the pipes. The maint guy said I need to leave the heat on full blast for about 48hrs to flush everything | 02:13 |
| mtreinish | so lots of liquids... | 02:14 |
| fungi | ick | 02:14 |
| *** jerryz has joined #openstack-infra | 02:15 | |
| mtreinish | yeah it's no fun, it's weird though having the heat on full blast with the windows open | 02:16 |
| mtreinish | and it being -4c outside | 02:16 |
| clarkb | I came home to frozen former liquid water on the back porch | 02:16 |
| clarkb | in november | 02:16 |
| clarkb | that was weird. also my power went out | 02:16 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 02:18 | |
| mtreinish | clarkb: are you dumb like me and run your mail infrastructure from home? | 02:19 |
| mtreinish | I lost email my last day in paris because my router crapped out | 02:20 |
| clarkb | no | 02:20 |
| clarkb | didnt know until I got home | 02:20 |
| *** Ryan_Lane has quit IRC | 02:21 | |
| *** koolhead17 has joined #openstack-infra | 02:22 | |
| mtreinish | well that's good at least. I need to do something about my brain dead setup at some point | 02:22 |
| mtreinish | or at least setup a backup mx | 02:22 |
| mtreinish | ok cool, looks like the inits got updated | 02:23 |
| mtreinish | fungi: any new stack traces? | 02:25 |
| fungi | mtreinish: looks like the worker daemons didn't restart | 02:28 |
| fungi | not that anything about that patch should have caused them to i guess | 02:28 |
| fungi | mtreinish: looks like it may be working now? | 02:29 |
| fungi | i manually stopped and started all four workers | 02:29 |
| mtreinish | cool :) | 02:29 |
| fungi | no, wait, new exceptions | 02:30 |
| *** ryanpetrello has quit IRC | 02:32 | |
| fungi | mtreinish: http://paste.openstack.org/show/134575/ | 02:32 |
| *** Longgeek has joined #openstack-infra | 02:33 | |
| mtreinish | fungi: hmm not much I can do about that one :) there's no subunit file for that run | 02:33 |
| mtreinish | it looks like discovery failed | 02:34 |
| *** ryanpetrello has joined #openstack-infra | 02:34 | |
| mtreinish | although I'm not sure why the client added a gear job for it then | 02:34 |
| fungi | okay, so that's probably benign then | 02:35 |
| fungi | i do see it appearing to process other files at least | 02:35 |
| mtreinish | fungi: awesome | 02:35 |
| fungi | just the occasional 404 like that | 02:35 |
| fungi | presumably for jobs with no subunit files | 02:35 |
| mtreinish | fungi: can you run select * from runs; on the db? | 02:35 |
| mtreinish | fungi: yeah that's what I'd expect, but I'll double check the client to see why there are gear jobs being added for jobs that don't have subunit files | 02:36 |
| mtreinish | clarkb: ^^^ unless you know off the top of your head | 02:36 |
| mtreinish | oh, I guess it doesn't actually check does it | 02:37 |
| fungi | it'll take a few minutes. i need to find an instance with mysqlclient on it and access to the trove db and dig up the credentials | 02:37 |
| mtreinish | fungi: ok, no rush. I was just curious how many runs are being added (and that it's doing the right thing) | 02:38 |
| *** Longgeek has quit IRC | 02:38 | |
| fungi | 151 rows in set | 02:39 |
| *** ChuckC has joined #openstack-infra | 02:39 | |
| mtreinish | wow, that's more than I thought for like 10 min... | 02:39 |
| mtreinish | especially at 9:30p | 02:39 |
| fungi | 2783 rows in tests | 02:41 |
| mtreinish | fungi: I wonder if it's not filtering on just gate | 02:41 |
| mtreinish | fungi: yeah that makes sense, that's about the number of tempest tests that get run | 02:41 |
| mtreinish | test_runs should be ~= 2783 * 151 you don't need to check that | 02:42 |
| mtreinish | fungi: can you take a quick look at run_metadata and see if there is metadata for check jobs in there? | 02:42 |
| *** mmaglana has joined #openstack-infra | 02:42 | |
| *** slogan has quit IRC | 02:42 | |
| openstackgerrit | Merged openstack-infra/devstack-gate: Update usage of LOCALRC_OLDNEW https://review.openstack.org/135418 | 02:42 |
| fungi | | f877f571-34b3-411c-b360-982f3c58f028 | build_queue | check | 02:44 |
| fungi | | 52a7d80f-462b-46ac-b716-7994d15691f4 | 02:44 |
| fungi | | | 02:44 |
| fungi | grr stray newlines | 02:44 |
| fungi | but yes, check jobs too | 02:44 |
| mtreinish | ok that explains why so many | 02:44 |
| mtreinish | I guess my regex in the client config is wrong | 02:44 |
| clarkb | mtreinish remember things were queued for a while too | 02:45 |
| mtreinish | clarkb: oh yeah, I forgot that it's been queueing since the first patch landed... | 02:45 |
| mtreinish | hmm, http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/files/logstash/jenkins-log-client.yaml#n14 | 02:46 |
| mtreinish | I guess we could have tempest jobs in the check queue named gate* | 02:46 |
| *** iain-king has quit IRC | 02:46 | |
| *** mmaglana has quit IRC | 02:47 | |
| *** patrickeast has quit IRC | 02:47 | |
| *** camunoz has joined #openstack-infra | 02:49 | |
| *** chandankumar has joined #openstack-infra | 02:51 | |
| mtreinish | clarkb: ^^^ do you know of the top of your head? | 02:51 |
| * mtreinish has a feeling the little trove node will be too small pretty soon | 02:52 | |
| *** stevemar has joined #openstack-infra | 02:52 | |
| fungi | mtreinish: we definitely have some tempest jobs in check named "gate-.*" | 02:53 |
| fungi | mtreinish: you probably want to go by build_queue=="check" | 02:54 |
| mtreinish | fungi: I think I'll have to expand the client to filter based on the build_queue | 02:55 |
| *** bdpayne has joined #openstack-infra | 02:55 | |
| mtreinish | based on my quick glance at the code it only uses the job name right now | 02:55 |
| mtreinish | I'll do that tomorrow | 02:55 |
| fungi | mtreinish: likely, just pointing out that filtering on job name alone isn't going to get you gate--pipeline-only jobs | 02:56 |
| fungi | also, i'm heading to bed | 02:56 |
| fungi | night all | 02:56 |
| mtreinish | night | 02:56 |
| mtreinish | I'm calling it a night too | 02:56 |
| *** shashankhegde has joined #openstack-infra | 02:56 | |
| mtreinish | fungi, clarkb: thanks for all the help getting things running today | 02:56 |
| fungi | mtreinish: any time! | 02:57 |
| *** koolhead17 has quit IRC | 02:57 | |
| *** banix has joined #openstack-infra | 03:00 | |
| *** ddieterl_ has quit IRC | 03:01 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add support to log gearman client to filter on build-queue https://review.openstack.org/135493 | 03:03 |
| *** links has joined #openstack-infra | 03:04 | |
| *** mudassirlatif has quit IRC | 03:04 | |
| *** achanda has joined #openstack-infra | 03:04 | |
| *** bdpayne has quit IRC | 03:04 | |
| *** koolhead17 has joined #openstack-infra | 03:05 | |
| *** koolhead17 has joined #openstack-infra | 03:05 | |
| *** dims_ has quit IRC | 03:05 | |
| *** dims has joined #openstack-infra | 03:06 | |
| *** miqui has joined #openstack-infra | 03:07 | |
| *** achanda has quit IRC | 03:10 | |
| *** bradjones has quit IRC | 03:10 | |
| *** bradjones has joined #openstack-infra | 03:12 | |
| *** unicell has quit IRC | 03:13 | |
| *** harlowja is now known as harlowja_away | 03:16 | |
| *** chandankumar has quit IRC | 03:18 | |
| *** ryanpetrello has quit IRC | 03:20 | |
| *** nosnos has quit IRC | 03:20 | |
| *** tsg has joined #openstack-infra | 03:21 | |
| *** ryanpetrello has joined #openstack-infra | 03:22 | |
| *** tsg has quit IRC | 03:30 | |
| *** tsg has joined #openstack-infra | 03:31 | |
| *** ddieterly has joined #openstack-infra | 03:31 | |
| *** achanda has joined #openstack-infra | 03:32 | |
| *** changbl has joined #openstack-infra | 03:33 | |
| *** bhunter71 has quit IRC | 03:33 | |
| *** Longgeek has joined #openstack-infra | 03:34 | |
| *** ddieterly has quit IRC | 03:36 | |
| *** henditan has quit IRC | 03:36 | |
| *** MarkAtwood has joined #openstack-infra | 03:36 | |
| *** henditan has joined #openstack-infra | 03:37 | |
| *** Longgeek has quit IRC | 03:38 | |
| *** tsg_ has joined #openstack-infra | 03:40 | |
| *** tsg has quit IRC | 03:40 | |
| *** mmaglana has joined #openstack-infra | 03:43 | |
| *** bhuvan has quit IRC | 03:47 | |
| *** jamespage_ has joined #openstack-infra | 03:47 | |
| *** mmaglana has quit IRC | 03:48 | |
| *** funzo has quit IRC | 03:49 | |
| *** jamespage_ has quit IRC | 03:50 | |
| *** jamespage_ has joined #openstack-infra | 03:50 | |
| *** funzo has joined #openstack-infra | 03:51 | |
| *** ddieterly has joined #openstack-infra | 03:51 | |
| *** funzo has quit IRC | 03:55 | |
| *** ddieterly has quit IRC | 03:56 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/devstack-gate: Fix up sudo wildcard copy https://review.openstack.org/133914 | 04:02 |
| openstackgerrit | Ian Wienand proposed openstack-infra/devstack-gate: Copy all apache logs into output https://review.openstack.org/133881 | 04:03 |
| *** rmcall has joined #openstack-infra | 04:03 | |
| *** todd_dsm has joined #openstack-infra | 04:04 | |
| *** nosnos has joined #openstack-infra | 04:12 | |
| *** ayoung is now known as ayoung-ZZZzzzZzz | 04:13 | |
| *** koolhead17 has quit IRC | 04:14 | |
| *** koolhead17 has joined #openstack-infra | 04:15 | |
| *** SumitNaiksatam has quit IRC | 04:17 | |
| *** SumitNaiksatam has joined #openstack-infra | 04:17 | |
| *** rushiagr_away is now known as rushiagr | 04:18 | |
| *** koolhead17 has quit IRC | 04:19 | |
| *** todd_dsm has quit IRC | 04:20 | |
| *** banix has quit IRC | 04:21 | |
| *** sputnik13 has quit IRC | 04:23 | |
| *** otherwiseguy has joined #openstack-infra | 04:23 | |
| *** banix has joined #openstack-infra | 04:24 | |
| *** MarkAtwood has quit IRC | 04:26 | |
| *** mmaglana has joined #openstack-infra | 04:29 | |
| *** bswartz has quit IRC | 04:30 | |
| *** teran has quit IRC | 04:31 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 04:31 | |
| *** ddieterly has joined #openstack-infra | 04:31 | |
| *** bswartz has joined #openstack-infra | 04:32 | |
| *** MarkAtwood has joined #openstack-infra | 04:35 | |
| *** ddieterly has quit IRC | 04:36 | |
| *** SumitNaiksatam has quit IRC | 04:37 | |
| *** SumitNaiksatam has joined #openstack-infra | 04:39 | |
| *** bhuvan has joined #openstack-infra | 04:40 | |
| *** banix has quit IRC | 04:41 | |
| *** ivar-laz_ has joined #openstack-infra | 04:42 | |
| *** achanda has quit IRC | 04:42 | |
| *** ivar-lazzaro has quit IRC | 04:45 | |
| *** ivar-laz_ has quit IRC | 04:46 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for conf.sample oslo.db bug 1393559 https://review.openstack.org/135343 | 04:50 |
| uvirtbot | Launchpad bug 1393559 in heat "oslo.db release breaks sample config" [Undecided,Fix committed] https://launchpad.net/bugs/1393559 | 04:50 |
| *** chandankumar has joined #openstack-infra | 04:53 | |
| *** ryanpetrello has quit IRC | 04:54 | |
| *** tnurlygayanov has quit IRC | 04:55 | |
| *** fandi has quit IRC | 04:58 | |
| *** dimtruck is now known as zz_dimtruck | 05:01 | |
| *** marun has joined #openstack-infra | 05:03 | |
| *** MarkAtwood has quit IRC | 05:05 | |
| *** rmcall has quit IRC | 05:07 | |
| *** yfried has joined #openstack-infra | 05:10 | |
| *** shashankhegde has quit IRC | 05:15 | |
| *** yfried is now known as yfried|afk | 05:16 | |
| *** shashankhegde has joined #openstack-infra | 05:18 | |
| *** pcrews has quit IRC | 05:19 | |
| *** yfried|afk is now known as yfried | 05:19 | |
| *** achanda has joined #openstack-infra | 05:25 | |
| *** sputnik13 has joined #openstack-infra | 05:26 | |
| *** koolhead17 has joined #openstack-infra | 05:27 | |
| *** ddieterly has joined #openstack-infra | 05:31 | |
| *** teran has joined #openstack-infra | 05:32 | |
| *** marun has quit IRC | 05:32 | |
| *** talluri has joined #openstack-infra | 05:33 | |
| *** ddieterly has quit IRC | 05:36 | |
| *** teran has quit IRC | 05:36 | |
| *** tsg_ has quit IRC | 05:37 | |
| *** AlexF has joined #openstack-infra | 05:39 | |
| *** boris-42 has joined #openstack-infra | 05:39 | |
| *** yfried is now known as yfried|afk | 05:41 | |
| *** miqui has quit IRC | 05:44 | |
| *** yfried|afk has quit IRC | 05:45 | |
| *** koolhead17 has quit IRC | 05:48 | |
| *** dtalton has joined #openstack-infra | 05:50 | |
| *** k4n0 has joined #openstack-infra | 05:52 | |
| *** talluri has quit IRC | 05:53 | |
| *** talluri has joined #openstack-infra | 05:54 | |
| *** subscope has joined #openstack-infra | 05:54 | |
| *** vigneshvar has joined #openstack-infra | 05:54 | |
| *** AlexF has quit IRC | 05:55 | |
| *** subscope has quit IRC | 05:55 | |
| *** talluri has quit IRC | 05:55 | |
| *** rmcall has joined #openstack-infra | 05:55 | |
| *** subscope has joined #openstack-infra | 05:55 | |
| *** subscope has quit IRC | 05:56 | |
| *** amotoki_ has quit IRC | 05:57 | |
| *** AlexF has joined #openstack-infra | 05:58 | |
| *** sputnik13 has quit IRC | 06:00 | |
| *** links has quit IRC | 06:02 | |
| *** stevemar has quit IRC | 06:02 | |
| *** camunoz has quit IRC | 06:02 | |
| *** unicell has joined #openstack-infra | 06:03 | |
| *** AlexF has quit IRC | 06:04 | |
| *** melwitt has quit IRC | 06:06 | |
| *** sputnik13 has joined #openstack-infra | 06:06 | |
| *** koolhead17 has joined #openstack-infra | 06:07 | |
| *** shashankhegde has quit IRC | 06:08 | |
| *** jkt has quit IRC | 06:09 | |
| *** jkt has joined #openstack-infra | 06:10 | |
| *** BharatK has joined #openstack-infra | 06:10 | |
| *** david-ly_ has joined #openstack-infra | 06:11 | |
| *** david-lyle_afk has quit IRC | 06:12 | |
| *** asettle has quit IRC | 06:13 | |
| *** camunoz has joined #openstack-infra | 06:15 | |
| *** AlexF has joined #openstack-infra | 06:20 | |
| *** Longgeek has joined #openstack-infra | 06:23 | |
| *** camunoz has quit IRC | 06:24 | |
| *** AlexF has quit IRC | 06:25 | |
| *** koolhead17 has quit IRC | 06:25 | |
| *** koolhead17 has joined #openstack-infra | 06:26 | |
| *** otherwiseguy has quit IRC | 06:26 | |
| *** HeOS has quit IRC | 06:28 | |
| *** Longgeek has quit IRC | 06:28 | |
| *** koolhead17 has quit IRC | 06:30 | |
| *** ddieterly has joined #openstack-infra | 06:31 | |
| *** hdd has joined #openstack-infra | 06:31 | |
| *** amotoki has joined #openstack-infra | 06:33 | |
| *** teran has joined #openstack-infra | 06:33 | |
| *** chandankumar has quit IRC | 06:33 | |
| *** ddieterly has quit IRC | 06:35 | |
| *** fandi has joined #openstack-infra | 06:37 | |
| *** teran has quit IRC | 06:38 | |
| *** k4n0 has quit IRC | 06:39 | |
| *** AlexF has joined #openstack-infra | 06:40 | |
| *** camunoz has joined #openstack-infra | 06:40 | |
| *** AlexF has quit IRC | 06:42 | |
| *** dtalton has quit IRC | 06:42 | |
| *** rakhmerov_ has joined #openstack-infra | 06:43 | |
| *** rakhmerov_ is now known as rakhmerov__ | 06:43 | |
| *** esker has quit IRC | 06:44 | |
| *** rakhmerov has quit IRC | 06:44 | |
| *** koolhead17 has joined #openstack-infra | 06:44 | |
| *** koolhead17 has quit IRC | 06:46 | |
| *** rlandy has joined #openstack-infra | 06:47 | |
| *** achanda has quit IRC | 06:48 | |
| *** koolhead17 has joined #openstack-infra | 06:48 | |
| *** ildikov has quit IRC | 06:50 | |
| *** viglesias has quit IRC | 06:51 | |
| *** belmoreira has joined #openstack-infra | 06:51 | |
| *** AlexF has joined #openstack-infra | 06:52 | |
| *** _buzz has quit IRC | 06:52 | |
| *** Longgeek has joined #openstack-infra | 06:53 | |
| *** koolhead17 has quit IRC | 06:53 | |
| *** chandankumar has joined #openstack-infra | 06:53 | |
| *** _buzz has joined #openstack-infra | 06:54 | |
| *** koolhead17 has joined #openstack-infra | 06:54 | |
| *** k4n0 has joined #openstack-infra | 06:54 | |
| *** Longgeek has quit IRC | 06:55 | |
| *** Longgeek has joined #openstack-infra | 06:56 | |
| *** achanda has joined #openstack-infra | 06:57 | |
| *** rcarrillocruz has quit IRC | 06:57 | |
| *** rcarrillocruz has joined #openstack-infra | 06:58 | |
| *** ddieterly has joined #openstack-infra | 07:01 | |
| *** unicell has quit IRC | 07:01 | |
| *** fandi has quit IRC | 07:04 | |
| *** unicell has joined #openstack-infra | 07:05 | |
| *** ddieterly has quit IRC | 07:06 | |
| *** AlexF has quit IRC | 07:09 | |
| *** unicell has quit IRC | 07:09 | |
| *** rmcall has quit IRC | 07:13 | |
| *** armax has quit IRC | 07:16 | |
| *** david-ly_ has quit IRC | 07:18 | |
| *** ildikov has joined #openstack-infra | 07:19 | |
| *** yongli has quit IRC | 07:20 | |
| *** yongli has joined #openstack-infra | 07:21 | |
| *** fandi has joined #openstack-infra | 07:21 | |
| *** kevinbenton has quit IRC | 07:22 | |
| *** yfried|afk has joined #openstack-infra | 07:23 | |
| *** yfried|afk is now known as yfried | 07:23 | |
| *** achanda has quit IRC | 07:26 | |
| *** fabo has quit IRC | 07:26 | |
| *** fabo has joined #openstack-infra | 07:26 | |
| *** k4n0 has quit IRC | 07:31 | |
| *** amuller has joined #openstack-infra | 07:35 | |
| *** rakhmerov__ is now known as rakhmerov | 07:42 | |
| *** otter768 has quit IRC | 07:43 | |
| *** k4n0 has joined #openstack-infra | 07:44 | |
| *** boris-42 has quit IRC | 07:47 | |
| *** Ryan_Lane has joined #openstack-infra | 07:47 | |
| *** talluri has joined #openstack-infra | 07:48 | |
| *** HeOS has joined #openstack-infra | 07:49 | |
| *** Ryan_Lane has quit IRC | 07:52 | |
| yfried | clarkb: are we running icehouse on any of the project gates outside tempest, and are these gates using tempest/master? | 07:53 |
| *** rcarrillocruz has quit IRC | 07:54 | |
| *** rcarrillocruz has joined #openstack-infra | 07:54 | |
| *** talluri has quit IRC | 07:56 | |
| *** yamamoto_ has quit IRC | 08:02 | |
| *** teran has joined #openstack-infra | 08:04 | |
| *** teran has quit IRC | 08:08 | |
| *** talluri has joined #openstack-infra | 08:08 | |
| *** yolanda has quit IRC | 08:09 | |
| *** IvanBerezovskiy has joined #openstack-infra | 08:12 | |
| *** luqas has joined #openstack-infra | 08:13 | |
| *** talluri has quit IRC | 08:13 | |
| *** nadya has joined #openstack-infra | 08:14 | |
| *** nadya is now known as Guest31365 | 08:14 | |
| *** tnovacik has joined #openstack-infra | 08:18 | |
| *** links has joined #openstack-infra | 08:21 | |
| *** jgallard has joined #openstack-infra | 08:24 | |
| *** jawed has joined #openstack-infra | 08:31 | |
| *** MaxV has joined #openstack-infra | 08:37 | |
| *** arxcruz has joined #openstack-infra | 08:37 | |
| *** Guest31365 has quit IRC | 08:37 | |
| *** henditan has quit IRC | 08:41 | |
| *** nadya_ has joined #openstack-infra | 08:42 | |
| *** henditan has joined #openstack-infra | 08:42 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/devstack-gate: Add option for is_debian to support it https://review.openstack.org/130768 | 08:43 |
| *** koolhead17 has quit IRC | 08:45 | |
| *** koolhead17 has joined #openstack-infra | 08:46 | |
| *** koolhead17 has quit IRC | 08:46 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard: Add API call to return task statuses https://review.openstack.org/135221 | 08:46 |
| *** yolanda has joined #openstack-infra | 08:46 | |
| *** unicell has joined #openstack-infra | 08:47 | |
| *** jlibosva has joined #openstack-infra | 08:49 | |
| *** david-lyle_afk has joined #openstack-infra | 08:49 | |
| *** nadya_ has quit IRC | 08:49 | |
| *** jp_at_hp has joined #openstack-infra | 08:53 | |
| *** yamamoto_ has joined #openstack-infra | 08:53 | |
| *** ashaeron has joined #openstack-infra | 08:56 | |
| *** jcoufal has joined #openstack-infra | 08:58 | |
| *** jpich has joined #openstack-infra | 08:59 | |
| *** jawed has quit IRC | 08:59 | |
| *** jcoufal has quit IRC | 09:01 | |
| *** jcoufal has joined #openstack-infra | 09:01 | |
| *** jawed has joined #openstack-infra | 09:01 | |
| *** jcoufal has quit IRC | 09:01 | |
| *** jcoufal has joined #openstack-infra | 09:02 | |
| *** jedimike has joined #openstack-infra | 09:03 | |
| *** gilliard has joined #openstack-infra | 09:04 | |
| *** SumitNaiksatam has quit IRC | 09:06 | |
| *** SumitNaiksatam has joined #openstack-infra | 09:06 | |
| *** nosnos has quit IRC | 09:07 | |
| *** nosnos has joined #openstack-infra | 09:08 | |
| *** sergsh_away is now known as sergsh | 09:12 | |
| *** andreykurilin_ has joined #openstack-infra | 09:14 | |
| *** mfink has quit IRC | 09:16 | |
| *** oomichi has quit IRC | 09:16 | |
| *** dizquierdo has joined #openstack-infra | 09:17 | |
| *** yamamoto_ has quit IRC | 09:18 | |
| openstackgerrit | Steven Dake proposed openstack-infra/project-config: Initial magnum client import for stackforge https://review.openstack.org/135436 | 09:21 |
| *** nadya_ has joined #openstack-infra | 09:22 | |
| *** bo_sh has joined #openstack-infra | 09:24 | |
| *** bo_sh has left #openstack-infra | 09:24 | |
| *** Guest67947 has joined #openstack-infra | 09:25 | |
| *** Svedrin has quit IRC | 09:25 | |
| nibalizer | question: with a line like #3 in this https://review.openstack.org/#/c/124789/1/lib/puppet/provider/vs_port/ovs.rb | 09:27 |
| nibalizer | is that okay in openstack? | 09:27 |
| *** vipul has quit IRC | 09:27 | |
| nibalizer | in work contexts i've heard it both ways "stackoverflow is fine" and "we cant be sure of ownership and licensing on SO code so we cant touch it" | 09:27 |
| *** teran has joined #openstack-infra | 09:28 | |
| *** nadya_ has quit IRC | 09:28 | |
| *** henditan has quit IRC | 09:28 | |
| *** henditan has joined #openstack-infra | 09:29 | |
| *** dtantsur|afk is now known as dtantsur | 09:29 | |
| *** Svedrin has joined #openstack-infra | 09:33 | |
| *** henditan has quit IRC | 09:35 | |
| *** henditan has joined #openstack-infra | 09:36 | |
| *** Longgeek has quit IRC | 09:38 | |
| *** jamielennox is now known as jamielennox|away | 09:41 | |
| jedimike | nibalizer, I just saw your review of the dedupe module upstream, not sure how to split the module out... is there a document somewhere that shows how we want the modules organised? | 09:41 |
| nibalizer | so it could go in as a new project openstack-infra/puppet-dedup | 09:42 |
| *** luqas has quit IRC | 09:42 | |
| nibalizer | in which case you would do the procedure in http://ci.openstack.org/stackforge.html | 09:43 |
| nibalizer | except everywhere it says stackforge you do openstack-infra | 09:43 |
| *** henditan has quit IRC | 09:43 | |
| nibalizer | then you can add the dedup module to modules.env in system-config | 09:43 |
| nibalizer | and a role/profile class to the openstack module | 09:43 |
| nibalizer | does that make sense? | 09:44 |
| *** henditan has joined #openstack-infra | 09:44 | |
| nibalizer | now this is the first time i think we've added a new module since deciding to split out so idunno | 09:44 |
| nibalizer | others in infra might have different ideas | 09:44 |
| nibalizer | for instance i could see the argument to put the code and the implementation and the wiring all into one patch to system-config so we can evaluate if the dedup stuff is a good thing to add to the systems at all | 09:44 |
| *** otter768 has joined #openstack-infra | 09:44 | |
| jedimike | nibalizer, thanks :) i'll ask if it's a good thing to have later today when it's more active in here | 09:46 |
| nibalizer | okay | 09:46 |
| *** Longgeek has joined #openstack-infra | 09:49 | |
| *** otter768 has quit IRC | 09:49 | |
| *** pelix has joined #openstack-infra | 09:51 | |
| *** mpaolino has joined #openstack-infra | 09:53 | |
| *** skolekonov has joined #openstack-infra | 09:54 | |
| *** vrutkovs has joined #openstack-infra | 09:57 | |
| *** vipul has joined #openstack-infra | 09:59 | |
| *** IvanBerezovskiy has quit IRC | 10:00 | |
| *** dizquierdo has quit IRC | 10:02 | |
| *** BobBall_AWOL is now known as BobBall | 10:07 | |
| *** dizquierdo has joined #openstack-infra | 10:14 | |
| *** yfried is now known as yfried|afk | 10:16 | |
| *** nadya_ has joined #openstack-infra | 10:16 | |
| *** pblaho has joined #openstack-infra | 10:17 | |
| *** dmsimard_away is now known as dmsimard | 10:17 | |
| *** yamamoto has joined #openstack-infra | 10:18 | |
| *** jawed has quit IRC | 10:22 | |
| *** yamamoto has quit IRC | 10:23 | |
| *** reed has joined #openstack-infra | 10:27 | |
| *** talluri has joined #openstack-infra | 10:31 | |
| *** ldnunes has joined #openstack-infra | 10:32 | |
| *** talluri has quit IRC | 10:33 | |
| *** talluri has joined #openstack-infra | 10:33 | |
| jd__ | We have an issue on tooz, LP bugs are never updated, e.g. https://review.openstack.org/#/c/135077/ has a proper Closes-Bug but nothing happens on LP :( | 10:33 |
| *** talluri has quit IRC | 10:38 | |
| *** talluri has joined #openstack-infra | 10:39 | |
| *** boris-42 has joined #openstack-infra | 10:40 | |
| *** talluri has quit IRC | 10:44 | |
| *** jawed has joined #openstack-infra | 10:46 | |
| *** nfedotov has joined #openstack-infra | 10:47 | |
| *** talluri has joined #openstack-infra | 10:49 | |
| *** bhuvan has quit IRC | 10:50 | |
| *** luqas has joined #openstack-infra | 10:50 | |
| *** amotoki has quit IRC | 10:51 | |
| *** tnovacik has quit IRC | 10:54 | |
| *** adalbas has joined #openstack-infra | 10:54 | |
| *** fandi has quit IRC | 11:00 | |
| *** tnovacik has joined #openstack-infra | 11:03 | |
| *** andreykurilin__ has joined #openstack-infra | 11:03 | |
| *** nadya_ has quit IRC | 11:03 | |
| *** andreykurilin_ has quit IRC | 11:03 | |
| *** jgallard has quit IRC | 11:06 | |
| *** erlon has joined #openstack-infra | 11:07 | |
| *** nadya_ has joined #openstack-infra | 11:07 | |
| openstackgerrit | Merged openstack-infra/storyboard: Working Directory Management https://review.openstack.org/130649 | 11:08 |
| *** dims has quit IRC | 11:09 | |
| *** dims has joined #openstack-infra | 11:09 | |
| *** Guest67947 has quit IRC | 11:11 | |
| *** e0ne has joined #openstack-infra | 11:13 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 11:14 | |
| *** denis_makogon has joined #openstack-infra | 11:14 | |
| *** kmartin has quit IRC | 11:16 | |
| *** luqas has quit IRC | 11:16 | |
| *** teran has quit IRC | 11:18 | |
| *** marcusvrn has joined #openstack-infra | 11:18 | |
| *** talluri has quit IRC | 11:21 | |
| *** jawed has quit IRC | 11:23 | |
| *** unicell has quit IRC | 11:27 | |
| *** vrutkovs has quit IRC | 11:27 | |
| sileht | hi, infra cores can someone take a looks to: https://review.openstack.org/#/c/135266/ | 11:29 |
| *** aysyd has joined #openstack-infra | 11:30 | |
| *** yfried|afk is now known as yfried | 11:30 | |
| *** vrutkovs has joined #openstack-infra | 11:31 | |
| *** henditan has quit IRC | 11:33 | |
| *** teran has joined #openstack-infra | 11:36 | |
| *** BharatK has left #openstack-infra | 11:37 | |
| nibalizer | sileht: most infra cores are in the americas, so you'll have better luck asking in serveral hours | 11:37 |
| *** bhuvan has joined #openstack-infra | 11:37 | |
| *** teran has quit IRC | 11:40 | |
| *** bhuvan has quit IRC | 11:42 | |
| *** dmsimard is now known as dmsimard_away | 11:43 | |
| *** alexpilotti has joined #openstack-infra | 11:44 | |
| *** ekarlso- has quit IRC | 11:44 | |
| openstackgerrit | Merged openstack-infra/storyboard-webclient: Correct virtualenv to activate https://review.openstack.org/133167 | 11:45 |
| * mordred punches nibalizer | 11:45 | |
| *** otter768 has joined #openstack-infra | 11:45 | |
| *** marcusvrn has quit IRC | 11:47 | |
| mordred | sileht: +3 | 11:47 |
| *** ekarlso- has joined #openstack-infra | 11:49 | |
| nibalizer | mordred: :D | 11:50 |
| nibalizer | there is mordred randomness factor, it is a good thing | 11:50 |
| *** otter768 has quit IRC | 11:50 | |
| nibalizer | mordred: have you been to sleep yet? | 11:50 |
| mordred | nibalizer: yes! | 11:51 |
| mordred | nibalizer: but it was on an aeroplane | 11:51 |
| mordred | nibalizer: so, you know, milage may vary | 11:52 |
| openstackgerrit | Merged openstack-infra/project-config: oslo.messaging: typo fix https://review.openstack.org/135266 | 11:52 |
| jedimike | mordred, then if you're around, you might be able to tell me if this would be useful outside of gozer and how new modules would go in to system-config ;) https://review.openstack.org/#/c/135287/ | 11:52 |
| *** AJaeger has joined #openstack-infra | 11:53 | |
| jedimike | and this has 2 +2's, so an A would be appreciated :) https://review.openstack.org/#/c/114510/ | 11:54 |
| mordred | jedimike: so - I have sevearl thoughts, some of which might be in conflict with each other ... | 11:56 |
| mordred | jedimike: first, and most non-controversially, you need a license header in the python file | 11:56 |
| mordred | :) | 11:56 |
| jedimike | ah yes :) | 11:57 |
| * jedimike makes notes | 11:57 | |
| mordred | I think it should certainly be a standalone puppet module - whether it goes in stackforge or openstack-infra would be the next question | 11:57 |
| nibalizer | we could also throw it in github.com/puppet-community | 11:58 |
| mordred | and without an actual use of it in system-config, I think it would be hard to justify why it would be in openstack-infra | 11:58 |
| nibalizer | which is attempting to be the 'community maintained' group of puppet modules | 11:58 |
| mordred | nibalizer: you could, but then you'd have to use github | 11:58 |
| jedimike | do you not have apt or pypi snapshots? | 11:58 |
| nibalizer | mordred: indeed | 11:58 |
| mordred | jedimike: we do not, they do not serve a purpose for us | 11:59 |
| jedimike | ah ok, then it's not of too much use | 11:59 |
| mordred | NOW - and here's where I'm going to just ramble for a second, so take it with a grain of salt ... | 11:59 |
| nibalizer | jedimike: so you could post new patchsets to that review with examples of where it would be used in the system-config | 11:59 |
| nibalizer | which would help set the context | 11:59 |
| mordred | nibalizer: I mean, the code makes sense to me and I grok it - just not where infra would use it | 11:59 |
| mordred | reprepro already knows how to do efficient apt repo snapshotting, and I believe devpi also similarly may know how to do it for pypi repos | 12:00 |
| jedimike | yeah, it's very useful to gozer, saves us over 100gb disk space, but if we don't use snapshots here, then it's not really of use | 12:00 |
| mordred | SO - I just want to make sure I've asked the question "why aren't you just using reprepro to manage apt repos, and have you looked at devpi for product oriented pypi repos" | 12:01 |
| mordred | (recognizing that the code you have is 20 lines and solves the problem) | 12:01 |
| jedimike | I'm not sure I have all the info to answer those questions fully, but from previous experience with reprepro it seems overkill. We're mirroring the ubuntu and hlinux archives, and every now and then we take a snapshot to develop against a known state of the archive from a particular date. It's those snapshots that this dedupes | 12:04 |
| * nibalizer adds aptly to the list of cool tools that are untested but sound neat | 12:04 | |
| * jedimike notes that usually, more than half the things that jedimike explains are for the benefit of jedimike's understanding, not yours ;) | 12:06 | |
| mordred | :) | 12:06 |
| mordred | nibalizer: what, you mean: http://www.aptly.info/doc/aptly/snapshot/create/ ? | 12:07 |
| nibalizer | mordred: just look at the beautiful web design | 12:08 |
| nibalizer | that tool must work | 12:08 |
| mordred | hahahaahahah | 12:08 |
| *** mwagner_lap has quit IRC | 12:08 | |
| *** luqas has joined #openstack-infra | 12:09 | |
| nibalizer | this being an example of the other side of that principle: http://rsync.samba.org/ | 12:10 |
| jedimike | also I still have ptsd from the insane reprepro set up at The Other Place | 12:13 |
| *** teran has joined #openstack-infra | 12:17 | |
| mordred | fair | 12:18 |
| *** dmsimard_away is now known as dmsimard | 12:18 | |
| mordred | aptly and devpi both have mostly the UI one wants to manage such a thing | 12:18 |
| *** doug-fish has joined #openstack-infra | 12:18 | |
| *** Ashenhav has joined #openstack-infra | 12:19 | |
| mordred | but, if aptly is anything like devpi, it's not REALLY well suited for production as much as it's suited for dev workstation | 12:19 |
| jedimike | but also, https://review.openstack.org/#/c/114510/ has 2 +2's, so if anyone could +A, it would make me do the happy dance | 12:19 |
| *** hdd has quit IRC | 12:22 | |
| *** enikanorov has quit IRC | 12:22 | |
| *** cnesa has joined #openstack-infra | 12:22 | |
| *** pc_m has joined #openstack-infra | 12:23 | |
| *** enikanorov has joined #openstack-infra | 12:24 | |
| *** rediskin has quit IRC | 12:24 | |
| * AJaeger can't approve 114510 but likes to put https://review.openstack.org/128395 in the zuul review queue as well | 12:27 | |
| jlibosva | AJaeger: hi | 12:30 |
| AJaeger | hi jlibosva | 12:30 |
| jlibosva | AJaeger: is there an easy way to have job voting on master but not on stable? | 12:31 |
| jlibosva | AJaeger: I made a patch https://review.openstack.org/#/c/134532/ which accidentally got rid of greande-neutron job on stable juno | 12:31 |
| AJaeger | jlibosva: you need two jobs in that case | 12:32 |
| jlibosva | AJaeger: also two job-templates? | 12:32 |
| AJaeger | jlibosva: create a -nv job... | 12:32 |
| AJaeger | there are quite a few -nv jobs for dsvm jobs, have a look and copy. | 12:32 |
| *** Ashenhav has quit IRC | 12:33 | |
| AJaeger | jlibosva: sorry, can't be more verbose, I have to run for breakfast and then a full conference day... | 12:33 |
| jlibosva | AJaeger: ok, I'll have a look. Thanks for your help and enjoy conference :) | 12:33 |
| AJaeger | jlibosva: thanks. I'll try to find some time inbetween to review your patch quickly once it's up | 12:34 |
| jlibosva | AJaeger: thanks, I don't think it has a high priority as it's most likely gonna fail :) | 12:34 |
| *** Longgeek has quit IRC | 12:36 | |
| AJaeger | ;) | 12:36 |
| *** Longgeek has joined #openstack-infra | 12:36 | |
| *** AJaeger has quit IRC | 12:37 | |
| *** yfried is now known as yfried|afk | 12:37 | |
| *** ashenhav has joined #openstack-infra | 12:37 | |
| *** bswartz has quit IRC | 12:37 | |
| *** bswartz has joined #openstack-infra | 12:38 | |
| *** wuhg has joined #openstack-infra | 12:41 | |
| *** mase_x200 has joined #openstack-infra | 12:41 | |
| wuhg | Code Review - Error | 12:43 |
| wuhg | Wu Hong Guang <whg@hp.com> does not identify a registered user or group | 12:43 |
| wuhg | whg@hp.com can't be added as a reviewer , | 12:43 |
| wuhg | i don't have the problem before | 12:43 |
| wuhg | anyone can help me on this | 12:44 |
| wuhg | ? | 12:44 |
| *** AlexF has joined #openstack-infra | 12:49 | |
| *** ekarlso- has quit IRC | 12:50 | |
| *** ekarlso- has joined #openstack-infra | 12:50 | |
| *** yamamoto has joined #openstack-infra | 12:50 | |
| *** yfried|afk is now known as yfried | 12:51 | |
| *** tnurlygayanov has joined #openstack-infra | 12:52 | |
| *** mbacchi has joined #openstack-infra | 12:52 | |
| *** koolhead17 has joined #openstack-infra | 12:53 | |
| *** koolhead17 has joined #openstack-infra | 12:53 | |
| *** afazekas has joined #openstack-infra | 12:53 | |
| *** jgallard has joined #openstack-infra | 12:53 | |
| *** koolhead17 has quit IRC | 12:54 | |
| *** AlexF has quit IRC | 12:54 | |
| *** yamamoto has quit IRC | 12:55 | |
| *** dims has quit IRC | 12:58 | |
| *** dims has joined #openstack-infra | 12:58 | |
| *** kashyap has quit IRC | 13:00 | |
| *** eantyshev has joined #openstack-infra | 13:00 | |
| *** amotoki has joined #openstack-infra | 13:01 | |
| *** marcusvrn has joined #openstack-infra | 13:02 | |
| eantyshev | Hello, everyone! I'm looking at jenkins jobs names and guess what 'dsvm' stands for? | 13:02 |
| eantyshev | Is there any rule for job naming? | 13:02 |
| *** vigneshvar has quit IRC | 13:02 | |
| *** bswartz has quit IRC | 13:04 | |
| pc_m | eantyshev: DevStack VM? | 13:05 |
| *** ddieterly has joined #openstack-infra | 13:07 | |
| *** dims has quit IRC | 13:08 | |
| *** jamespage_ has quit IRC | 13:08 | |
| *** dims has joined #openstack-infra | 13:08 | |
| *** andreykurilin__ has quit IRC | 13:10 | |
| eantyshev | Thanks! But what does VM mean? | 13:10 |
| *** nadya_ has quit IRC | 13:10 | |
| pc_m | eantyshev: Virtual Machine | 13:11 |
| eantyshev | Oh, indeed! | 13:11 |
| pc_m | eantyshev: I don't know if there is a naming convention... hopefully an infra person can chime in. | 13:12 |
| wuhg | hi, everyone , my account have a problem with review.openstack.org | 13:12 |
| wuhg | is this the right room to ask this question? | 13:12 |
| *** ZZelle has joined #openstack-infra | 13:13 | |
| pc_m | salv-orlando: Do you know if naming convention for jenkins job names that eantyshev is asking above? | 13:13 |
| openstackgerrit | Spencer Krum proposed openstack-infra/infra-specs: Remove Modulefile from puppet-module spec https://review.openstack.org/134812 | 13:14 |
| *** rushiagr is now known as rushiagr_away | 13:14 | |
| *** Longgeek has quit IRC | 13:18 | |
| salv-orlando | pc | 13:20 |
| *** ddieterly has quit IRC | 13:20 | |
| salv-orlando | _pc_m, eantyshev: correct, dsvm == DevStack VM | 13:20 |
| dmsimard | Hi. Trying to figure out why some tests were removed from stackforge/puppet-ceph but I don't know where to look. We used to have unit tests but now it only tests for lint and syntax. Any hints ? | 13:20 |
| dmsimard | A general direction as to where to look would be enough for me to find what's wrong surely :) | 13:20 |
| *** rcarrillocruz has quit IRC | 13:21 | |
| *** ddieterly has joined #openstack-infra | 13:21 | |
| *** rcarrillocruz has joined #openstack-infra | 13:22 | |
| *** bswartz has joined #openstack-infra | 13:26 | |
| *** markmcclain has joined #openstack-infra | 13:27 | |
| *** markmcclain has quit IRC | 13:27 | |
| *** markmcclain has joined #openstack-infra | 13:28 | |
| anteaya | dmsimard: tests for all projects are in the project-config repo | 13:28 |
| nibalizer | anteaya: oh you are online i have questions | 13:28 |
| anteaya | dmsimard: I'd suggest running git blame on zuul/layout.yaml | 13:29 |
| anteaya | nibalizer: I appear to be online, I can listen to your questions | 13:29 |
| nibalizer | anteaya: is this okay? https://review.openstack.org/#/c/124789/1/lib/puppet/provider/vs_port/ovs.rb | 13:29 |
| anteaya | dmsimard: that should serve as a starting point | 13:29 |
| nibalizer | the stackoverflow bits? | 13:29 |
| nibalizer | are we allowed to consume from SO like that? | 13:29 |
| dmsimard | anteaya: Thanks. I'll check. I was already in project-config but was looking elsewhere :) | 13:29 |
| *** kgiusti has joined #openstack-infra | 13:30 | |
| anteaya | dmsimard: glad you were in the right repo | 13:30 |
| jpich | wuhg: It should the correct place, though your problem isn't very clear. Is the issue when you're uploading a new patch from the CLI, or when someone tries to add you as a reviewer on the web interface? | 13:31 |
| anteaya | nibalizer: I don't know, it appears that the relevant information is the regex for uuid: [0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12} | 13:31 |
| *** bhunter71 has joined #openstack-infra | 13:31 | |
| anteaya | nibalizer: if yes, I would suggest copying the regex in the comment and using the url as the source | 13:32 |
| anteaya | I would have no argument with using stackoverflow as a reference | 13:32 |
| anteaya | as for licencing, I have never been able to hold my own in the licensing conversations | 13:32 |
| anteaya | so I would suggest talking to fungi or jeblair since they have more knowledge in licensing that I do | 13:33 |
| *** ayoung-ZZZzzzZzz is now known as ayoung | 13:33 | |
| anteaya | eantyshev: there are some formats for jenkins jobs, looking in project-config/jenkins/jobs should give you examples | 13:34 |
| anteaya | I don't know as we have written out the format anywhere as a reference | 13:35 |
| nibalizer | anteaya: okay thanks | 13:35 |
| *** weshay has joined #openstack-infra | 13:39 | |
| anteaya | that's the best I have | 13:39 |
| dmsimard | anteaya: Found the culprit.. this commit: https://github.com/openstack-infra/project-config/commit/64bac91927a6df10cfc40fff317bf19e345b6e9b | 13:41 |
| dmsimard | I'll check how to fix it | 13:41 |
| dmsimard | When the person changed out the jobs for the templated versions, he left out puppet-module-unit-jobs from puppet-ceph | 13:42 |
| anteaya | dmsimard: ah | 13:42 |
| anteaya | should be as easy as adding this line: - name: puppet-module-unit-jobs to puppet-ceph | 13:43 |
| dmsimard | I don't know if straight-up adding puppet-module-unit-jobs to the list of jobs will fix it, trying to crawl up and look where that's referenced | 13:43 |
| anteaya | crawl away | 13:43 |
| anteaya | you will learn something | 13:43 |
| nibalizer | dmsimard: huh | 13:43 |
| nibalizer | what happened? | 13:44 |
| dmsimard | nibalizer: Jenkins only runs puppet-lint and syntax check against stackforge/puppet-ceph, it doesn't run spec tests anymore | 13:44 |
| anteaya | dmsimard: so in that patch puppet-ceph still has puppet-unit-2.7 and 3.1 on it | 13:44 |
| nibalizer | dmsimard: im seeing that hrm let me look deeper | 13:45 |
| nibalizer | i think i was the one who turned off 2.7 and 3.1 | 13:45 |
| *** dprince has joined #openstack-infra | 13:45 | |
| *** otter768 has joined #openstack-infra | 13:46 | |
| nibalizer | ya so it looks like the problem is line 5026 should have puppet-module-unit-jobs | 13:47 |
| dmsimard | nibalizer: Yup. That simple, right ? | 13:47 |
| nibalizer | not sure why/how that got skipped | 13:47 |
| dmsimard | nibalizer: I'll submit a fix.. | 13:47 |
| nibalizer | sweet do it | 13:47 |
| nibalizer | also note that we stopped testing 2.7 and early 3.x versions | 13:47 |
| *** hdd has joined #openstack-infra | 13:48 | |
| dmsimard | nibalizer: Yeah, that's fine, I think 2.7 was non-voting already for us, we wanted to drop support for it. | 13:48 |
| *** ryanpetrello has joined #openstack-infra | 13:49 | |
| nibalizer | okay cool | 13:49 |
| *** Hal has joined #openstack-infra | 13:50 | |
| *** Longgeek has joined #openstack-infra | 13:50 | |
| openstackgerrit | David Moreau Simard proposed openstack-infra/project-config: Re-enable spec/unit tests for puppet-ceph https://review.openstack.org/135602 | 13:50 |
| *** Hal is now known as Guest53254 | 13:50 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/requirements: Add glance_store, kite, python-kiteclient to projects.txt https://review.openstack.org/135603 | 13:50 |
| dmsimard | Hrm, let me fix that commit message - it's not referencing the commt sha1 | 13:51 |
| dmsimard | commit* | 13:51 |
| *** otter768 has quit IRC | 13:51 | |
| wuhg | jpich:the problem happends when i tries to add as a reviewer on the web interface | 13:51 |
| *** miqui has joined #openstack-infra | 13:51 | |
| wuhg | jpich:i can do this before , | 13:51 |
| openstackgerrit | David Moreau Simard proposed openstack-infra/project-config: Re-enable spec/unit tests for puppet-ceph https://review.openstack.org/135602 | 13:52 |
| *** ildikov has quit IRC | 13:52 | |
| *** xchu has joined #openstack-infra | 13:52 | |
| *** AlexF has joined #openstack-infra | 13:53 | |
| *** jistr has joined #openstack-infra | 13:53 | |
| anteaya | wuhg: what web interface are you using? are you following this guide? http://docs.openstack.org/infra/manual/developers.html#getting-started | 13:53 |
| *** zz_dimtruck is now known as dimtruck | 13:53 | |
| *** jistr is now known as jistr|mtg | 13:53 | |
| *** xchu has quit IRC | 13:54 | |
| wuhg | anteaya:yes,i follow this guide | 13:54 |
| openstackgerrit | Merged openstack-infra/project-config: Split out github as it's own module https://review.openstack.org/126448 | 13:54 |
| wuhg | i can add me as a reviewer a few days ago | 13:54 |
| *** dalgaaf__ has joined #openstack-infra | 13:55 | |
| anteaya | I don't understand what you mean by add you as a reviewer | 13:56 |
| anteaya | do you mean to subscribe to a patch? | 13:56 |
| *** BharatK has joined #openstack-infra | 13:56 | |
| anteaya | as in you fill in your name and you get email updates on the patch? | 13:56 |
| *** eharney has joined #openstack-infra | 13:56 | |
| anteaya | nibalizer: this commit message makes much more sense now: https://review.openstack.org/#/c/134834/4 thank you | 13:56 |
| *** yamahata has quit IRC | 13:57 | |
| openstackgerrit | Merged openstack-infra/project-config: Publish python-barbicanclient docs https://review.openstack.org/135449 | 13:57 |
| wuhg | anteaya:when i submit a patch , i usally add a reviewer on from web interface | 13:57 |
| *** yamahata has joined #openstack-infra | 13:57 | |
| anteaya | nibalizer: do you need to register a namespace on puppet forge prior to publishing? if so, can you demonstrate you have registered the name by linking a url in a comment on the patch | 13:57 |
| *** dtantsur is now known as dtantsur|lunch | 13:58 | |
| anteaya | nibalizer: we like to see this before using the publish-to-pypi job to demonstrate the name is registered with pypi, it avoids name conflicts | 13:58 |
| *** ddieterly has quit IRC | 13:58 | |
| anteaya | wuhg: can you share the url for the patch you are trying to add a reviewer to? | 13:58 |
| *** vrutkovs|mtg has joined #openstack-infra | 13:59 | |
| *** krtaylor has quit IRC | 13:59 | |
| mmedvede | The https://review.openstack.org/126448 got merged, is somebody shepharding it? nibalizer, your upstream module is not up to date on that one\ | 13:59 |
| wuhg | https://review.openstack.org/#/c/135273/ | 13:59 |
| anteaya | nibalizer: would make sense we should have this for puppet forge too | 13:59 |
| *** ihrachyshka has joined #openstack-infra | 13:59 | |
| anteaya | sdague: are you shepherding the jenkins module split? | 14:00 |
| wuhg | anteaya:there is a add reviewer button (the second table) | 14:00 |
| *** vrutkovs has quit IRC | 14:00 | |
| anteaya | yes I just successfully added myself as a reviewer | 14:00 |
| sdague | anteaya: I was reviewing changes that seemed to make sense | 14:00 |
| anteaya | sdague: well mmedvede has concerns | 14:01 |
| anteaya | and modules splits need shepherds | 14:01 |
| sdague | on which patches? | 14:01 |
| anteaya | to ensure upstream is current | 14:01 |
| anteaya | puppet module splits | 14:01 |
| openstackgerrit | Merged openstack-infra/project-config: Enable Ironic periodic-stable jobs https://review.openstack.org/135133 | 14:01 |
| wuhg | anteaya:but when i add whg@hp.com ,it report error | 14:01 |
| *** nadya_ has joined #openstack-infra | 14:01 | |
| sdague | ok, can we WIP them then? | 14:01 |
| nibalizer | anteaya: we've got https://forge.puppetlabs.com/openstackci already | 14:02 |
| anteaya | fungi seemed to think comments were enough | 14:02 |
| sdague | or have some other indication in the commit message that their is a synchronization point ? | 14:02 |
| anteaya | sdague: can you take it up with fungi? | 14:02 |
| sdague | ok, sure | 14:02 |
| *** esker has joined #openstack-infra | 14:02 | |
| anteaya | thanks | 14:02 |
| sdague | I'm happy to not review any of the puppet parts | 14:02 |
| anteaya | I'm all for what works and apparently were aren't there yet | 14:02 |
| anteaya | sdague: I don't want you to feel blocked on your reviews | 14:02 |
| sdague | ok, so if something isn't working, that really does need to be WIP | 14:03 |
| *** dimtruck is now known as zz_dimtruck | 14:03 | |
| anteaya | I am in agreement | 14:03 |
| sdague | at least in my opinion | 14:03 |
| anteaya | I share your opinion | 14:03 |
| anteaya | it was discussed at the last infra meeting though, for reference | 14:03 |
| mmedvede | sdague: we are still getting a better workflow figured out. The one that got merged was not updated with latest recommendations (it was one of the first splits) | 14:03 |
| sdague | mmedvede: ok, is there a revert? | 14:04 |
| *** mpaolino has quit IRC | 14:04 | |
| anteaya | wuhg: why do you want to add yourself as a reviewer to your own patch? | 14:04 |
| anteaya | wuhg: you already are subscribed | 14:04 |
| anteaya | wuhg: can you subscribe to a patch that is not owned by you? | 14:04 |
| *** ihrachyshka has quit IRC | 14:04 | |
| wuhg | anteaya:i can't be added to any patch | 14:05 |
| anteaya | can you try adding Wu Hong Guang | 14:05 |
| wuhg | anteaya:it report the same error | 14:05 |
| anteaya | I'll try to subscribe you to some of my patches | 14:05 |
| mmedvede | sdague: no revert is needed, I believe. But someone now needs to make sure the counterpart patch gets taken care of. And the upstream module source would get updated | 14:06 |
| wuhg | anteaya:yes ,i tryied Wu Hong Guang ,but no lucy | 14:06 |
| nibalizer | yea... | 14:06 |
| nibalizer | mmedvede: so weird thing happened | 14:07 |
| mmedvede | nibalizer: would you be able to update your github upstream? | 14:07 |
| nibalizer | i used git subtree split exactly as we generated the procedure | 14:07 |
| nibalizer | and it didn't pull in the changes from 2014 | 14:07 |
| *** ashenhav has quit IRC | 14:07 | |
| nibalizer | i just reproduced it and yea no changes form 2014 | 14:07 |
| nibalizer | mmedvede: you want me to try to push changes in really fast before the sync happens? | 14:08 |
| anteaya | wuhg: I see the same | 14:08 |
| nibalizer | i dont have a thing to push, otherwise i would try | 14:08 |
| nibalizer | mmedvede: can you do the subtree split on your system and see if you get the same results? | 14:08 |
| wuhg | anteaya: i tried to subscribe to other's patch too, but got same error | 14:08 |
| anteaya | wuhg: go to this page when you are signed into review.openstack.org: https://review.openstack.org/#/settings/agreements | 14:08 |
| anteaya | wuhg: do you see a line that says: VerifiedICLAOpenStack Individual Contributor License Agreement | 14:09 |
| mmedvede | nibalizer: I am trying subtree now. I do not think we can get it in time though. I remember last time this happened we needed to force update the project? | 14:09 |
| wuhg | VerifiedICLAOpenStack Individual Contributor License Agreemen | 14:09 |
| wuhg | anteay:sure | 14:09 |
| anteaya | wuhg: then I don't know what is going on with your account, somehow the database doesn't recognize you as a registered user | 14:10 |
| wuhg | i have been bale to subscrube any patch | 14:10 |
| *** ihrachyshka has joined #openstack-infra | 14:10 | |
| anteaya | wuhg: go to this page: https://review.openstack.org/#/settings/ | 14:10 |
| anteaya | what is your Account ID? | 14:10 |
| ekarlso- | https://review.openstack.org/#/c/134711/ < anyone got a clue on why that hasn't merged yet even though i has a +2 / +w ? | 14:10 |
| nibalizer | mmedvede: ya two options, one is to have fungi force push, the other is to do submit those patches as reviews | 14:10 |
| wuhg | Username | 14:10 |
| wuhg | whg | 14:10 |
| nibalizer | im much more concerned that the subtree split workflow we have may not be correct | 14:10 |
| *** mwagner_lap has joined #openstack-infra | 14:11 | |
| wuhg | email:whg@hp.com | 14:11 |
| anteaya | wuhg: Account ID, it will be a number | 14:11 |
| anteaya | wuhg: just the Account ID please | 14:11 |
| *** krtaylor has joined #openstack-infra | 14:11 | |
| wuhg | 581 | 14:11 |
| mmedvede | nibalizer: yes, subtree split is not robust. It does not track renames either, as I talked before | 14:11 |
| *** pblaho has quit IRC | 14:11 | |
| anteaya | wuhg: wow, how long have you been signed up with gerrit? | 14:12 |
| wuhg | anteaya:581 | 14:12 |
| mmedvede | nibalizer: but it is faster, and we can check if it does proper job | 14:12 |
| *** dkranz has joined #openstack-infra | 14:13 | |
| *** mestery has quit IRC | 14:14 | |
| anteaya | wuhg: what is your entry for Registered? That is the line above Account ID? | 14:15 |
| *** ociuhandu has joined #openstack-infra | 14:15 | |
| wuhg | RegisteredJul 23, 2011 3:13 AM | 14:16 |
| *** marcusvrn has quit IRC | 14:16 | |
| anteaya | okay the number might be right then | 14:17 |
| openstackgerrit | Sean Dague proposed openstack-infra/project-config: add support for running devstack unit tests https://review.openstack.org/135075 | 14:17 |
| *** emagana has joined #openstack-infra | 14:17 | |
| *** AJaeger has joined #openstack-infra | 14:17 | |
| anteaya | wuhg: I'm guessing something funny happened with your gerrit database entry but we won't know until someone with database access takes a look | 14:17 |
| anteaya | wuhg: can you stay online a few hours so that we can ask questions if we need to? | 14:18 |
| wuhg | anteaya:yes ,i can be online for about another 2 hours | 14:19 |
| anteaya | wuhg: thanks, hopefully fungi will take a look once he is online | 14:20 |
| wuhg | anteaya:many thanks for your help | 14:20 |
| mmedvede | nibalizer: tried the subtree split for github, did not catch 2014 changes either | 14:20 |
| jeblair | ttx: it's raining in berkeley! :) | 14:21 |
| anteaya | fungi: wuhg is apparently not part of a registered group according to gerrit, account id 581, as evidenced by trying to subscribe Wu Hong Guang to a patch | 14:21 |
| anteaya | and (s)he was able to successfully subscribe several days ago, so this is a recent change | 14:22 |
| ttx | jeblair: sunny here | 14:22 |
| jeblair | ha! | 14:22 |
| ttx | pure blue sky, a bit chilly though (12°C) | 14:22 |
| *** rushiagr_away is now known as rushiagr | 14:22 | |
| fungi | anteaya: yeah, i think i answered a question about that account on ask.o.o a few days ago but saw no follow-up. i'll look in a sec | 14:23 |
| *** mpaolino has joined #openstack-infra | 14:23 | |
| anteaya | okay thanks | 14:23 |
| anteaya | cross referencing with ask.o.o never crossed my mind | 14:23 |
| fungi | once i skim through this mountain of scrollback | 14:25 |
| anteaya | yep | 14:25 |
| pc_m | anteaya: FYI: there are two entries for wuhg, one under HP email and one under Gmail email - same name. | 14:26 |
| *** bcourt has joined #openstack-infra | 14:26 | |
| *** mestery has joined #openstack-infra | 14:26 | |
| * pc_m saw the response and poked around but didn't know what was wrong | 14:26 | |
| wuhg | fungi:can you take a look at my acoount whg@hp.com in 2 hours? | 14:27 |
| *** AJaeger has quit IRC | 14:27 | |
| *** hdd has quit IRC | 14:28 | |
| *** Kiall has joined #openstack-infra | 14:28 | |
| fungi | jd__: you need to convert the tooz bug supervisor to a group and add the OpenStack Infra (hudson-openstack) account as a member. also http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/projects.yaml#n1718 needs to be updated to add "groups: - python-tooz" since the project name on lp is different from git | 14:29 |
| *** funzo has joined #openstack-infra | 14:29 | |
| mmedvede | nibalizer: the two commits that subtree split missed are noop (one reverts the other). So we are good | 14:29 |
| *** Sincler has joined #openstack-infra | 14:30 | |
| *** ddieterly has joined #openstack-infra | 14:30 | |
| mmedvede | nibalizer: as long as somebody watches it before the https://review.openstack.org/#/c/126450 is taken care of | 14:31 |
| anteaya | pc_m: nice spelunking, that is probably the issue | 14:31 |
| wuhg | pc_m:yes, a few days ago , a added gmail | 14:31 |
| wuhg | pc_m:yes, a few days ago , I added gmail | 14:31 |
| pc_m | wuhg: worked before then? | 14:32 |
| wuhg | yes , it worked before i add the gmail | 14:32 |
| anteaya | wuhg: how did you add the gmail? | 14:33 |
| *** openstackgerrit has quit IRC | 14:33 | |
| fungi | sdague: item #6 at http://specs.openstack.org/openstack-infra/infra-specs/specs/puppet-modules.html#work-items describes what we want in the commit messages | 14:33 |
| *** chandankumar has quit IRC | 14:33 | |
| jeblair | nibalizer, anteaya, fungi, sdague: good morning (i'm jetlagged) | 14:33 |
| *** openstackgerrit has joined #openstack-infra | 14:34 | |
| * pc_m curious as I was thinking of using gmail account | 14:34 | |
| anteaya | jeblair: welcome back | 14:34 |
| *** che-arne has joined #openstack-infra | 14:34 | |
| wuhg | anteaya: under Contact Information ->register a new mail | 14:34 |
| *** doude has quit IRC | 14:34 | |
| *** derekh has joined #openstack-infra | 14:34 | |
| anteaya | pc_m: many folks have multiple emails, including gmail | 14:34 |
| anteaya | pc_m: for most users adding a new email is not an issue, this is the first time I have seen this | 14:34 |
| pc_m | anteaya: Just under the same account (IOW, can one have multiple emails associated w/one account)? | 14:34 |
| *** zul has joined #openstack-infra | 14:35 | |
| wuhg | yes , on launchpad ,we can have more than one email | 14:35 |
| anteaya | wuhg: which is your prefered email? | 14:35 |
| anteaya | pc_m: yes, multiple emails with the same account are fine | 14:35 |
| openstackgerrit | Merged openstack-infra/project-config: Puppet Jenkins uses storyboard https://review.openstack.org/135420 | 14:35 |
| openstackgerrit | Merged openstack-infra/project-config: tooz: gate against py{27,34}-mysql https://review.openstack.org/135206 | 14:35 |
| openstackgerrit | Merged openstack-infra/project-config: Enable dib service for heat-dsvm-functional https://review.openstack.org/134422 | 14:35 |
| pc_m | anteaya: So wuhg shows as two entries, are these two accounts (with the same name)? | 14:35 |
| *** doude has joined #openstack-infra | 14:35 | |
| wuhg | i preferer email is whg@hp.com, gmail is the backup | 14:35 |
| *** Ark has joined #openstack-infra | 14:36 | |
| anteaya | pc_m: where are you seeing two entries? | 14:36 |
| anteaya | pc_m: can you link me? | 14:36 |
| nibalizer | mmedvede: oh so thats why that fel out? | 14:36 |
| nibalizer | interesting | 14:36 |
| *** Ark is now known as Guest72001 | 14:36 | |
| nibalizer | jeblair: ohai! | 14:36 |
| sdague | fungi: ok, so can we WIP things that aren't in that state? I think I accidentally approved a thing because it said ready for review, and seemed fine | 14:36 |
| pc_m | anteaya: Hang on. It was a few hours ago that I poked around. | 14:36 |
| fungi | sdague: yes, wip would be a good idea for those | 14:37 |
| openstackgerrit | Merged openstack-infra/project-config: Create new stackforge project for striker https://review.openstack.org/132966 | 14:37 |
| openstackgerrit | Merged openstack-infra/project-config: Ensure sideways ironic grenade job is voting https://review.openstack.org/134618 | 14:37 |
| *** Guest72001 has quit IRC | 14:37 | |
| fungi | wuhg: account 581 has your hp.com e-mail address listed (i don't see a gmail account listed for it) | 14:38 |
| *** markmcclain has quit IRC | 14:38 | |
| wuhg | pc_m: i deleted gmail half an hour ago | 14:38 |
| asselin | fungi, sdague nibalizer I proposed this change to help with premature merges. But I can update to WIP. https://review.openstack.org/#/c/135452/ | 14:39 |
| wuhg | fungi: i deleted gmail half an hour ago | 14:39 |
| pc_m | anteaya: I looked up "People" and then those in the CLA accepted group. Wu shows up 2x there: https://review.openstack.org/#/admin/groups/324,members | 14:39 |
| wuhg | fungi: my account sill is not registerd even i i deleted gmail half an hour ago | 14:39 |
| pc_m | anteaya: So wondering if there are two separate accounts. | 14:40 |
| fungi | wuhg: i see the problem. you have two accounts. 13981 is also associated with the same e-mail address | 14:40 |
| wuhg | fungi:i just add a email acount to account id 581 | 14:40 |
| mmedvede | sdague, fungi, nibalizer, asselin: I am looking through history to find reasoning why it was decided not to use WIP | 14:41 |
| anteaya | pc_m: good eye | 14:41 |
| pc_m | anteaya: thanks | 14:41 |
| fungi | wuhg: account 13981 was created by Wu Hong Guang on 2014-11-14 using a primary e-mail of flankw@gmail.com and secondary e-mail matching your hp.com address | 14:41 |
| *** eantyshev has quit IRC | 14:41 | |
| fungi | wuhg: are you using that second account at all? the problem is that it's in conflict with your earlier account | 14:42 |
| wuhg | fungi:please delete the gmail | 14:42 |
| *** r-daneel has joined #openstack-infra | 14:42 | |
| fungi | it seems to have filled in contact info and signed a cla | 14:42 |
| *** esker has quit IRC | 14:42 | |
| wuhg | no , i didn't signed cla for gmail | 14:43 |
| anteaya | pc_m: so add as many additional emails to your gerrit account as you wish, just don't create another account, the creation of a second account causes problems | 14:43 |
| fungi | wuhg: well, i'm trying to figure out how it got created, because gerrit auto-creates accounts based on openid authentication events, so if i don't take the correct action there's a chance it will be recreated | 14:43 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add support to log gearman client to filter on build-queue https://review.openstack.org/135493 | 14:43 |
| *** ecosta has joined #openstack-infra | 14:43 | |
| *** esker has joined #openstack-infra | 14:43 | |
| *** ecosta has quit IRC | 14:44 | |
| fungi | wuhg: and i'm trying to permanently fix it so that these accounts don't wind up back in conflict again | 14:44 |
| *** ecosta has joined #openstack-infra | 14:44 | |
| wuhg | fungi:ok ,could you let the whg@hp.com work asap? | 14:44 |
| anteaya | wuhg: how many launchpad accounts do you have? | 14:44 |
| fungi | wuhg: i would like to, can you please describe what you did back on november 14 which led to creation of the second gerrit account? | 14:44 |
| *** esker has quit IRC | 14:44 | |
| wuhg | i have only 1 account launchpad | 14:45 |
| *** zz_gondoi is now known as gondoi | 14:45 | |
| fungi | wuhg: also, one of those accounts is using an ssh username of whg and the other has an ssh username of wuhg | 14:45 |
| fungi | wuhg: i want to make sure that if i disable the newer one it doesn't stop you from being able to upload new changes | 14:46 |
| pc_m | anteaya: good to know. Thanks. | 14:46 |
| *** cpowell has joined #openstack-infra | 14:46 | |
| pc_m | anteaya: so is it the 'preferred' address as the one that everyone sees? | 14:46 |
| wuhg | fungi:what i did is : add a gmail to whg@hp.com account ,then set gmail as prefered email | 14:47 |
| anteaya | pc_m: that is my understanding, it also is the email that summit passes and election ballots get mailed to | 14:47 |
| wuhg | fungi:nothing else | 14:47 |
| fungi | wuhg: did you maybe change your address or username in launchpad? | 14:47 |
| wuhg | gungi:yes i did ,but not succeed | 14:48 |
| pc_m | anteaya: fungi: so the preferred email is the one used in reviews as well? | 14:48 |
| fungi | wuhg: it looks like i have two different launchpad openid urls for these accounts, one with an lp username of whg and the other with an lp username of wuhg | 14:48 |
| fungi | pc_m: yes | 14:49 |
| *** otherwiseguy has joined #openstack-infra | 14:49 | |
| fungi | wuhg: are you going to continue trying to use the wuhg launchpad account? | 14:49 |
| *** openstackgerrit has quit IRC | 14:49 | |
| pc_m | fungi: so if you have code out for revuew, I take it that changing the preferred email is not a good idea? | 14:49 |
| *** krtaylor has quit IRC | 14:49 | |
| *** openstackgerrit has joined #openstack-infra | 14:49 | |
| wuhg | gungi:i only use whg | 14:49 |
| wuhg | on launchpad | 14:49 |
| *** ecosta has quit IRC | 14:50 | |
| fungi | wuhg: okay, thanks. that's what i needed to know. i'll disable the other conflicting account then | 14:50 |
| *** pradk has joined #openstack-infra | 14:50 | |
| *** amotoki has quit IRC | 14:50 | |
| wuhg | gungi:ok | 14:51 |
| fungi | pc_m: it's complicated... the list of e-mail addresses associated with your gerrit account are used to determine who you can submit changes as, but the one set as preferred is the one where you'll receive e-mail notifications | 14:51 |
| *** armax has joined #openstack-infra | 14:51 | |
| fungi | wuhg: it should be working normally now | 14:51 |
| *** jaypipes has joined #openstack-infra | 14:51 | |
| *** nosnos has quit IRC | 14:52 | |
| wuhg | yes ,it works again | 14:52 |
| wuhg | good work gungi: | 14:53 |
| fungi | wuhg: you're welcome, and sorry for the delay i was just making sure we took the correct action based on your particular situation | 14:54 |
| *** e0ne has quit IRC | 14:54 | |
| wuhg | fungi:one more question:can i link a gmail to 581? | 14:54 |
| *** stevemar has joined #openstack-infra | 14:54 | |
| wuhg | fungi:in case i don't own whg@hp.com | 14:55 |
| fungi | wuhg: yes, in https://review.openstack.org/#/settings/contact click the "register new email" button and add the new address you want associated with the account | 14:55 |
| *** zul has quit IRC | 14:55 | |
| *** pawel-palucki has joined #openstack-infra | 14:55 | |
| *** amitgandhinz has joined #openstack-infra | 14:55 | |
| wuhg | fungi:ok , i'll do it again | 14:56 |
| fungi | wuhg: then, once you complete the registration process based on the confirmation e-mail you receive, you should be able to switch it to primary and then, if you want, delete the hp.com address after that | 14:56 |
| *** MaxV has quit IRC | 14:57 | |
| *** tonytan4ever has joined #openstack-infra | 14:57 | |
| wuhg | fungi:my job with hp.com will not be lost after doing that? | 14:57 |
| *** gordc has joined #openstack-infra | 14:57 | |
| *** doude has quit IRC | 14:57 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add subunit-worker01 to cacti https://review.openstack.org/135615 | 14:57 |
| fungi | wuhg: no, your changes and comments are associated to your gerrit account id number | 14:57 |
| *** MaxV has joined #openstack-infra | 14:58 | |
| gordc | anyone aware of issues with ipaddr? it appears to breaking ceilometer py33 gate. | 14:58 |
| fungi | wuhg: however, deleting an e-mail address from gerrit which is still associated with your launchpad account will likely not work, since the next time you authenticate via lp openid it will re-add that e-mail address | 14:58 |
| fungi | gordc: got a link to a log? | 14:58 |
| gordc | fungi: http://logs.openstack.org/40/134940/4/check/gate-ceilometer-python33/27162e0/console.html | 14:58 |
| *** mase_x200 has quit IRC | 14:58 | |
| wuhg | fungi:many thanks ,great help | 14:58 |
| *** doude has joined #openstack-infra | 14:59 | |
| gordc | fungi: not sure what is bringing in ipaddr... seems to conincide with wsme release... but ipaddr appears to be on same release for a while. | 15:00 |
| fungi | gordc: did wsme just start depending on it? | 15:00 |
| fungi | gordc: also, does wsme get tested against py33? | 15:00 |
| gordc | fungi: not that i can tell... i search wsme for ipaddr reference and it's been there for over a year. | 15:01 |
| fungi | gordc: and wsme does get tested against python 3.3 it seems | 15:01 |
| dhellmann | fungi, gordc : is something installing from a universal wheel? | 15:01 |
| fungi | dhellmann: tarball (ipaddr-2.1.11.tar.gz) | 15:02 |
| dhellmann | fungi: yeah, we have wsme set to build universal wheels, but we shouldn't do that because different versions depend on different packages | 15:02 |
| mmedvede | sdague, fungi, asselin, nibalizer: I found the discussion. We did not use WIP for module splits because it would have made trolling for reviews necessary. We agreed to make commit messages be the gatekeeper from unwanted merges (we are not there yet, apparently). http://eavesdrop.openstack.org/meetings/infra/2014/infra.2014-10-28-19.01.log.html | 15:02 |
| * gordc is not sure. | 15:02 | |
| *** krtaylor has joined #openstack-infra | 15:02 | |
| dhellmann | fungi: and it looks like we uploaded the universal wheel to pypi | 15:02 |
| fungi | mmedvede: entry #6 at http://specs.openstack.org/openstack-infra/infra-specs/specs/puppet-modules.html#work-items | 15:03 |
| *** MaxV has quit IRC | 15:03 | |
| *** thedodd has joined #openstack-infra | 15:03 | |
| fungi | dhellmann: is wsme set for universal in its setup.cfg then? | 15:03 |
| dhellmann | fungi: I have to move physically, but I'll be back online in a few minutes. We should try removing the universal wheel from pypi and then I'll change that flag in setup.cfg (yes) | 15:03 |
| dhellmann | fungi: can you remove the wheel? I can't seem to login to pypi from this laptop for some reason | 15:04 |
| dhellmann | fungi: bbiab | 15:04 |
| gordc | dhellmann: thanks for looking into it (i've no idea how packaging works) | 15:04 |
| gordc | fungi: ^ same thanks. | 15:04 |
| fungi | dhellmann: and yeah, it did install WSME-0.6.2-py2.py3-none-any.whl | 15:04 |
| *** mriedem has joined #openstack-infra | 15:04 | |
| fungi | dhellmann: deleting now | 15:05 |
| *** denis_makogon has quit IRC | 15:05 | |
| *** dustins has joined #openstack-infra | 15:06 | |
| *** e0ne has joined #openstack-infra | 15:06 | |
| *** esker has joined #openstack-infra | 15:06 | |
| *** yfried is now known as yfried|brb | 15:07 | |
| mmedvede | fungi: this helps, but it does not prevent a reviewer from merging the commit with the Needed-By section. The addition asselin put to the spec might, though (https://review.openstack.org/#/c/135452/) | 15:07 |
| *** esker has quit IRC | 15:07 | |
| *** nelsnelson has joined #openstack-infra | 15:07 | |
| fungi | mmedvede: yeah, it was waiting on an additional reviewer to +2 | 15:08 |
| *** otherwiseguy has quit IRC | 15:08 | |
| fungi | dhellmann: gordc: i've deleted WSME-0.6.2-py2.py3-none-any.whl from pypi and our bandersnatch mirrors should follow suit in a matter of minutes | 15:08 |
| *** doude has quit IRC | 15:08 | |
| *** jamespage_ has joined #openstack-infra | 15:10 | |
| *** jamespage__ has joined #openstack-infra | 15:12 | |
| clarkb | we should disable that job on wsme | 15:13 |
| fungi | clarkb: dhellmann is going to modify wsme to stop claiming it has a universal wheel | 15:14 |
| *** marcusvrn has joined #openstack-infra | 15:14 | |
| fungi | clarkb: then the py2 wheel will be the only one uploaded | 15:14 |
| *** jamespage_ has quit IRC | 15:14 | |
| clarkb | gotcha | 15:14 |
| *** zz_jgrimm is now known as jgrimm | 15:14 | |
| *** belmoreira has quit IRC | 15:14 | |
| *** xyang0 has joined #openstack-infra | 15:15 | |
| *** ihrachyshka has quit IRC | 15:16 | |
| *** ddieterly has quit IRC | 15:16 | |
| *** zul has joined #openstack-infra | 15:16 | |
| *** unicell has joined #openstack-infra | 15:16 | |
| *** denis_makogon has joined #openstack-infra | 15:16 | |
| *** ddieterly has joined #openstack-infra | 15:17 | |
| gordc | fungi: huge thanks | 15:17 |
| clarkb | fungi mtreinish logstash queue fell to zero :) | 15:20 |
| *** yfried|brb is now known as yfried|afk | 15:20 | |
| mtreinish | clarkb: cool :) | 15:20 |
| dhellmann | fungi: I'm back online. So if I remove the wheel = 1 in setup.cfg, what sort of wheels will we get? Just py2, or separate py2 and py3? | 15:20 |
| mtreinish | clarkb: do you think that means 1 subunit worker is enough for now? | 15:21 |
| clarkb | mtreinish ya we should watch it but it caught up on a couple weeks of work so its probably ok | 15:21 |
| clarkb | er 1.5ish weeks | 15:21 |
| fungi | dhellmann: just py2 | 15:21 |
| *** jamespage__ has quit IRC | 15:21 | |
| openstackgerrit | Merged openstack-infra/release-tools: Allow spec2bp.py to work for Oslo projects https://review.openstack.org/134274 | 15:22 |
| clarkb | I thought there was a separate jjb job for that? | 15:22 |
| fungi | dhellmann: we'd need to consider adding py3-specific wheel jobs to run under (preferably now) python 3.4 | 15:22 |
| *** enikanorov has quit IRC | 15:22 | |
| fungi | clarkb: we have a separate py3wheel job of some description? | 15:22 |
| clarkb | I think it was uplaod universal and upload py2 jobs | 15:23 |
| *** enikanorov has joined #openstack-infra | 15:23 | |
| mtreinish | clarkb: heh, well to be fair it was a light 1.5 weeks. Everyone was on vacation :) | 15:23 |
| mtreinish | clarkb: I've got 2 patches up on it right now: https://review.openstack.org/135493 and https://review.openstack.org/135615 they might help a bit | 15:23 |
| mtreinish | I'm also working on the mysql proxy module so we can have read-only access to the db externally | 15:24 |
| fungi | clarkb: oh, looking | 15:24 |
| *** yfried|afk is now known as yfried|brb | 15:25 | |
| *** yfried|brb is now known as yfried | 15:25 | |
| dhellmann | fungi: ok, I'm not sure how big of a priority it is since most of the openstack packages work with universal wheels | 15:26 |
| fungi | clarkb: it looks like we just have a single macro which does the wheel building from a slave script which is name-agnostic | 15:27 |
| dhellmann | fungi, gordc : https://review.openstack.org/135626 | 15:27 |
| clarkb | yfried I dont understand your question from 8 hours ago | 15:27 |
| fungi | clarkb: so i believe we'll get an interpreter-specific wheel name | 15:27 |
| *** amotoki has joined #openstack-infra | 15:28 | |
| clarkb | for building but upload to pypi is separate | 15:28 |
| clarkb | does it try all the options? | 15:28 |
| fungi | dhellmann: that ought to work | 15:28 |
| dhellmann | fungi: after the tests are done I'll slam it through and do another release | 15:28 |
| *** MaxV has joined #openstack-infra | 15:29 | |
| *** markmcclain has joined #openstack-infra | 15:29 | |
| fungi | clarkb: i'm not sure what you mean. the upload script just takes a filename | 15:29 |
| fungi | clarkb: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/scripts/pypi-wheel-upload.sh#n28 is the closest we get to caring whether it's universal in our toolchain, from what i can see | 15:30 |
| clarkb | ok that must have changed. original jobs were wheel type specific | 15:31 |
| *** andreykurilin_ has joined #openstack-infra | 15:31 | |
| fungi | ahh, yeah, we've had a bunch of changes around wheel-building early on since we took a while to get it working righth | 15:31 |
| clarkb | https://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/pypi-jobs.yaml#n1 is the job and it doesnt care | 15:31 |
| fungi | right | 15:32 |
| *** Ark has joined #openstack-infra | 15:33 | |
| *** Ark is now known as Guest45888 | 15:33 | |
| *** rediskin has joined #openstack-infra | 15:36 | |
| *** pcrews has joined #openstack-infra | 15:36 | |
| *** vrutkovs has joined #openstack-infra | 15:37 | |
| *** amuller has quit IRC | 15:38 | |
| *** vrutkovs|mtg has quit IRC | 15:39 | |
| *** wenlock has joined #openstack-infra | 15:40 | |
| *** tnovacik has quit IRC | 15:40 | |
| sdague | mmedvede: so the problem is when sub groups agree to review processes that are outside the norm, there is a communication lag to get everyone in sync. The WIP bit is universal. | 15:40 |
| sdague | anyway, if stuff isn't ready for landing, please WIP it. I review on a ton of projects and having different projects have different ways to designate ready to land is just error prone. | 15:41 |
| *** dtantsur|lunch is now known as dtantsur | 15:42 | |
| fungi | sdague: the trick is that "ready to land" may be a very narrow window dependent on external factors, and changed indefinitely marked wip generally won't get any reviews | 15:43 |
| fungi | s/changed/changes/ | 15:43 |
| johnthetubaguy | fungi: clarkb: hey, I noticed some of the infra builds are failing in the rackspace cloud, seems like the image: 6643b3ab-aace-49f8-97be-25dd10d9c8e0 in DFW has gone bad, its the devstack-precise-rax-dfw-* I think, let me know if I can get you folks more info | 15:43 |
| mmedvede | sdague: you think if we explicitly would have assigned core specified in the commit message, WIP would be desirable as well? | 15:43 |
| fungi | the trick is to get a sufficient number of core reviews on the change without letting it get approved prematurely | 15:44 |
| *** jerryz has quit IRC | 15:44 | |
| fungi | and in a timely enough fashion that it neither starts to merge conflict with the branch nor the external dependencies slip out of freshness | 15:45 |
| sdague | fungi: right, but you do see the problem with building different review process for different parts of openstack, right? | 15:45 |
| *** mrmartin has joined #openstack-infra | 15:46 | |
| *** prometheanfire has left #openstack-infra | 15:46 | |
| fungi | sdague: agreed it's a challenge. the needs for this differ from code which gets released | 15:46 |
| johnthetubaguy | fungi: clarkb: seems like the builds are failing because Xen can't load the PV kernel, it could be missing some metadata to boot HVM, or it might be that xen can't read the grub.conf for some reason | 15:46 |
| mmedvede | sdague: I see your point and agree, I also consider fungi's reasoning | 15:46 |
| jeblair | sdague, fungi: this is part of the reason for wanting a formalized replacement for "procedural -2", which i have on our todo list from the summit | 15:47 |
| jeblair | at least, it's a related issue | 15:47 |
| sdague | jeblair: right, that makes sense | 15:47 |
| *** otter768 has joined #openstack-infra | 15:47 | |
| *** krtaylor has quit IRC | 15:47 | |
| fungi | jeblair: yep, i agree that would help tremendously for things like this | 15:47 |
| yfried | clarkb: I'm wondering if there's any gate (except for the tempest gate) that is running tests from tempest/master against an icehouse neutron cloud | 15:47 |
| *** otherwiseguy has joined #openstack-infra | 15:48 | |
| *** achanda has joined #openstack-infra | 15:48 | |
| sdague | but today, we've got WIP. Which the only reason for it not being used is because people are ignoring it. But it gets ignored because it's not landable. So now what I'm hearing is we need to implement our own virtual wip which is don't +A these changes based on some other non gerrit reason. | 15:50 |
| fungi | sdague: we already frequently list cross-project dependencies in commit messages to that end | 15:51 |
| sdague | sure | 15:52 |
| sdague | I guess I missed that in the one that I let through (or it wasn't there) | 15:52 |
| *** otter768 has quit IRC | 15:52 | |
| fungi | sdague: it probably wasn't there | 15:52 |
| sdague | so if the commit messages are sufficiently clear, I'm good | 15:52 |
| *** yamamoto has joined #openstack-infra | 15:52 | |
| fungi | sdague: i never meant to suggest you did anything wrong there, just trying to determine the best way to clean up | 15:53 |
| sdague | ok | 15:53 |
| sdague | I guess where the conversation started this morning was a little more on the "why did you do that?" | 15:53 |
| fungi | i likely missed the start of that conversation | 15:53 |
| anteaya | well it was more, now that you did, are you going to see the whole process through | 15:54 |
| anteaya | since the two patches go together | 15:54 |
| mmedvede | sdague: I think no one meant that. When I noticed it merged, I just wanted to bring attention to it of interested parties | 15:54 |
| fungi | yfried: change 135624,1 is currently in the check pipeline at http://status.openstack.org/zuul/ and is a stable/icehouse neutron change running check-tempest-dsvm-neutron-heat-slow, check-tempest-dsvm-neutron, check-tempest-dsvm-neutron-pg, gate-tempest-dsvm-neutron-large-ops (plus a couple duplicates of those and a devstack-based functional test) all of which use tempest's master branch and are | 15:54 |
| fungi | voting | 15:54 |
| *** thedodd has quit IRC | 15:55 | |
| fungi | well, the functional test may not use bits of tempest, but the rest of those do | 15:55 |
| fungi | yfried: does that answer your question? | 15:55 |
| *** pcrews has quit IRC | 15:55 | |
| *** achanda has quit IRC | 15:56 | |
| *** marcusvrn has quit IRC | 15:56 | |
| jedimike | fungi, would you be ok with +A'ing https://review.openstack.org/#/c/114510/ ? It has 2 +2's now | 15:56 |
| *** yamamoto has quit IRC | 15:57 | |
| *** achanda has joined #openstack-infra | 15:57 | |
| nibalizer | sdague: also a) nothing bad happened and b) literally nothing bad could have happend with that approval | 15:58 |
| nibalizer | it is impossible for that change to have broken anything in infra | 15:58 |
| fungi | jedimike: it looks like you addressed jeblair's comments and clarkb +2'd the previous change. jhesketh doesn't say why he didn't approve it but seems fine to me | 15:58 |
| yfried | fungi: not completely. I don't know if this is running against tempest/master of some old tempest | 15:58 |
| nibalizer | it could have made more work for people but, it didn't even do that | 15:58 |
| sdague | nibalizer: yeh, ok. Just trying to make sure. So that I don't accidentally break things in the future. | 15:59 |
| fungi | yfried: there is no "some old tempest" | 15:59 |
| fungi | yfried: there is only master | 15:59 |
| *** yamamoto has joined #openstack-infra | 15:59 | |
| yfried | fungi: that's what I'm afraid of | 15:59 |
| sdague | fungi: there is only zuul | 15:59 |
| sdague | :) | 15:59 |
| dhellmann | fungi: wsme 0.6.3 is tagged and only the py2 wheel was built, as expected | 15:59 |
| fungi | yfried: the last stable branch of tempest was havana, since deleted | 15:59 |
| fungi | dhellmann: perfect--thanks for figuring that out so quickly! | 16:00 |
| *** krtaylor has joined #openstack-infra | 16:00 | |
| *** patrickeast has joined #openstack-infra | 16:00 | |
| *** pcrews has joined #openstack-infra | 16:00 | |
| *** hashar has joined #openstack-infra | 16:00 | |
| yfried | fungi: ok | 16:00 |
| yfried | fungi: tnx | 16:00 |
| *** chandankumar has joined #openstack-infra | 16:01 | |
| *** Sukhdev has joined #openstack-infra | 16:02 | |
| *** patrickeast has quit IRC | 16:02 | |
| dhellmann | fungi: thanks for the assist! | 16:02 |
| *** Longgeek has quit IRC | 16:03 | |
| *** patrickeast has joined #openstack-infra | 16:03 | |
| *** liam_ has joined #openstack-infra | 16:04 | |
| *** liam_ is now known as Guest9691 | 16:04 | |
| jedimike | fungi, thanks :) | 16:04 |
| *** Guest9691 has quit IRC | 16:05 | |
| nibalizer | sdague: its hard, you either cowboy too much and break, or you overburden things with procedure and oversight | 16:09 |
| nibalizer | the middle is elusive | 16:09 |
| *** achanda has quit IRC | 16:10 | |
| *** funzo has quit IRC | 16:10 | |
| *** banix has joined #openstack-infra | 16:12 | |
| *** ashaeron has quit IRC | 16:13 | |
| *** jamespage_ has joined #openstack-infra | 16:14 | |
| *** rlandy has quit IRC | 16:14 | |
| *** rmcall has joined #openstack-infra | 16:15 | |
| *** atiwari has joined #openstack-infra | 16:15 | |
| *** kashyap has joined #openstack-infra | 16:17 | |
| sdague | nibalizer: or you put in checks to -1 you when your cowboying is dangerous | 16:19 |
| *** hashar has quit IRC | 16:19 | |
| *** VijayTripathi has joined #openstack-infra | 16:19 | |
| *** vigneshvar has joined #openstack-infra | 16:20 | |
| *** dangers_away is now known as dangers | 16:23 | |
| openstackgerrit | David Moreau Simard proposed openstack-infra/project-config: Re-enable spec/unit tests for puppet-ceph https://review.openstack.org/135602 | 16:24 |
| *** marun has joined #openstack-infra | 16:25 | |
| *** BobBall is now known as BobBall_Away | 16:28 | |
| *** marun_ has joined #openstack-infra | 16:28 | |
| *** afazekas has quit IRC | 16:29 | |
| *** afazekas has joined #openstack-infra | 16:30 | |
| *** marun has quit IRC | 16:30 | |
| krotscheck | r1chardj0n3s, clarkb: FWIW, nodejs, npm, and bower all install brilliantly using epel-release on centos7 | 16:30 |
| clarkb | nice | 16:33 |
| *** yfried has quit IRC | 16:33 | |
| *** david-lyle_afk is now known as david-lyle | 16:34 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add mysql-proxy to enable read-only access to a db https://review.openstack.org/122311 | 16:34 |
| *** marun_ has quit IRC | 16:34 | |
| dtantsur | hi! any hints, how to create a stable branch for my stackforge project? | 16:34 |
| mtreinish | clarkb, fungi: ^^^ I still need to test that, I'm having a hard time getting mysql-proxy to even work manually with read-only | 16:34 |
| *** marun_ has joined #openstack-infra | 16:35 | |
| sdague | fungi / anteaya / clarkb - turning on devstack unit tests would be appreciated - https://review.openstack.org/#/c/135075/ - then will work on some for all the libs from pypi work | 16:37 |
| *** arxcruz has quit IRC | 16:38 | |
| anteaya | sdague: you want to make a new job voting right away? | 16:38 |
| sdague | anteaya: yes | 16:38 |
| sdague | my confidence level is high | 16:39 |
| *** vrutkovs has left #openstack-infra | 16:39 | |
| sdague | unless there is a reason you believe it is incorrect | 16:39 |
| anteaya | I just like to see the new job workflow of either experiemental or non-voting | 16:39 |
| anteaya | so I asked | 16:40 |
| anteaya | if this is what you want, so be it | 16:40 |
| *** sputnik13 has quit IRC | 16:41 | |
| anteaya | sdague: fungi has a comment, I was about to approve with a comment unless you want to respond to fungi first | 16:42 |
| fungi | anteaya: my comment was for the benefit of subsequent reviewers, summarizing earlier questions which had already been answered | 16:43 |
| anteaya | okay | 16:43 |
| *** gyee has joined #openstack-infra | 16:45 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: Fix specs repo location calculation https://review.openstack.org/135652 | 16:46 |
| openstackgerrit | Julien Danjou proposed openstack-infra/project-config: Move tooz from Stackforge to OpenStack https://review.openstack.org/135215 | 16:46 |
| anteaya | so nibalizer what you are saying is that all modules uploaded to puppet forge will land in this namespace? https://forge.puppetlabs.com/openstackci | 16:46 |
| *** doude has joined #openstack-infra | 16:48 | |
| nibalizer | anteaya: yes | 16:48 |
| *** dmellado has quit IRC | 16:48 | |
| anteaya | grand thanks | 16:49 |
| *** jcoufal has quit IRC | 16:49 | |
| krotscheck | fungi: Are we ready to send out the StoryBoard email? | 16:50 |
| *** BharatK has left #openstack-infra | 16:52 | |
| *** doug-fish has quit IRC | 16:53 | |
| openstackgerrit | Merged openstack-infra/project-config: add support for running devstack unit tests https://review.openstack.org/135075 | 16:53 |
| *** marun__ has joined #openstack-infra | 16:53 | |
| *** doug-fish has joined #openstack-infra | 16:54 | |
| fungi | krotscheck: no, i still need to update the imports and lock down lp after the meeting i'm in wraps up | 16:54 |
| krotscheck | fungi: KK | 16:54 |
| *** krtaylor has quit IRC | 16:54 | |
| *** jamespage_ has quit IRC | 16:54 | |
| *** marun_ has quit IRC | 16:55 | |
| *** belmoreira has joined #openstack-infra | 16:55 | |
| fungi | krotscheck: but it's next on my hit list | 16:55 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for infra bug 1365046 https://review.openstack.org/135654 | 16:56 |
| uvirtbot | Launchpad bug 1365046 in openstack-ci "Job failed due to no devstack directory" [Undecided,Confirmed] https://launchpad.net/bugs/1365046 | 16:56 |
| *** k4n0 has quit IRC | 16:56 | |
| anteaya | mtreinish and jeblair this adds voting ironic jobs to tempest, devstack, devstack-gate would like to hear from you on it: https://review.openstack.org/#/c/126627/ (also nova, but mikal is probably asleep) | 16:57 |
| *** sarob has joined #openstack-infra | 16:58 | |
| jeblair | anteaya: ack, thx | 16:59 |
| *** AlexF has quit IRC | 17:00 | |
| *** AlexF_ has joined #openstack-infra | 17:00 | |
| anteaya | np | 17:00 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/devstack-gate: Give a more specific error message when giving up in git_remote_update https://review.openstack.org/135656 | 17:01 |
| mriedem | could use some eyes on that so we can get a better e-r query on that bug ^ | 17:01 |
| *** SumitNaiksatam has quit IRC | 17:01 | |
| *** skolekonov has quit IRC | 17:01 | |
| mtreinish | anteaya: sure I'm fine with that | 17:01 |
| anteaya | mtreinish: thanks for the review | 17:01 |
| *** esker has joined #openstack-infra | 17:02 | |
| *** sputnik13 has joined #openstack-infra | 17:02 | |
| *** marun__ is now known as marun | 17:02 | |
| *** sputnik13 has quit IRC | 17:03 | |
| *** Longgeek has joined #openstack-infra | 17:03 | |
| *** Guest53254 has quit IRC | 17:04 | |
| *** kmartin has joined #openstack-infra | 17:04 | |
| *** dustins has quit IRC | 17:04 | |
| *** garyh has joined #openstack-infra | 17:05 | |
| *** patrickeast has quit IRC | 17:05 | |
| *** groknix has quit IRC | 17:05 | |
| *** chandankumar has quit IRC | 17:06 | |
| *** krtaylor has joined #openstack-infra | 17:07 | |
| notmyname | how do I find the associated patch from jenkins output (eg someone gave me a link to a jenkins error, but I can't find the patch) | 17:07 |
| *** gondoi is now known as zz_gondoi | 17:07 | |
| jeblair | notmyname: it should be in a link near the top of the jenkins console log | 17:07 |
| *** mudassirlatif has joined #openstack-infra | 17:07 | |
| jeblair | 2014-11-19 17:07:07.817 | Triggered by: https://review.openstack.org/135356 | 17:08 |
| jeblair | notmyname: eg ^ | 17:08 |
| notmyname | jeblair: https://jenkins.openstack.org/job/swift-propose-translation-update/63/consoleFull | 17:08 |
| *** Guest45888 has quit IRC | 17:08 | |
| *** Longgeek has quit IRC | 17:08 | |
| *** mmaglana has joined #openstack-infra | 17:09 | |
| *** nfedotov has quit IRC | 17:09 | |
| jeblair | notmyname: ah, that happens after a change merges and runs with master | 17:09 |
| *** dkranz has quit IRC | 17:09 | |
| jeblair | (ie, is a post-merge job) | 17:09 |
| *** Sukhdev has quit IRC | 17:09 | |
| notmyname | jeblair: ok. so is there a patch? or not? (I clearly don't know what's happening) | 17:09 |
| fungi | HEAD is now at dba3fc3 Merge "Prefer X-Backend-Timestamp for X-Newest" | 17:10 |
| jeblair | notmyname: nope, no patch. basically each time a change merges, that job runs and checks out tip-of-master and runs with that | 17:10 |
| notmyname | ok | 17:10 |
| jeblair | notmyname: so yeah, it checked out what fungi pasted | 17:10 |
| notmyname | right. which is the current HEAD | 17:10 |
| notmyname | the beginning of this story is that seagate proposed 100% translation of swift to chinese. and this aparantly is the result of the cron to merge the translation | 17:11 |
| fungi | notmyname: corresponding change parent of that merge is https://review.openstack.org/133869 | 17:11 |
| notmyname | but it failed (there's another patch in zuul now for that...) | 17:11 |
| notmyname | fungi: ya, that patch doesn't seem to have any problems | 17:11 |
| *** amitgandhinz has quit IRC | 17:12 | |
| notmyname | https://review.openstack.org/#/c/135606/ should fix the translation part. I think | 17:12 |
| notmyname | I'm not clear on the translation workflow, so that's where some of my confusion is coming from | 17:12 |
| anteaya | sdake: can you help me out here: https://review.openstack.org/#/c/135436/ I read this patch as "we don't really need this code, it doesnt' do anything, but everyone else has this so we are following the crowd" | 17:12 |
| jeblair | notmyname: i think https://review.openstack.org/#/c/135606/ needs to merge then, after it does so, the next translation proposal job should succeed | 17:12 |
| anteaya | sdake: I'm not for creating repos for that reason | 17:13 |
| fungi | notmyname: also this recently added documentation explains how the translations update/import mechanisms currently work if you're interested https://wiki.openstack.org/wiki/Translations/Infrastructure | 17:13 |
| jeblair | notmyname: i mis-spoke earlier, that translation proposal job is not post-merge, but rather runs nightly, so you should see the result next "day" after that fix merges | 17:14 |
| notmyname | fungi: thanks | 17:14 |
| jeblair | for various values of "day" | 17:14 |
| notmyname | jeblair: ya, I'm told 0600UTC | 17:14 |
| *** dizquierdo has quit IRC | 17:14 | |
| jeblair | (the similarly named but completely different upstream-translation-update runs post-merge :) | 17:14 |
| mtreinish | fungi: is there an existing server it makes sense to run the read-only mysql proxy on? | 17:15 |
| fungi | mtreinish: the words "public database proxy" give me security hives. do you have some details on how it would work? | 17:17 |
| fungi | i'd need to know a little about the design before i could speculate on somewhere "safe" to implement such a thing | 17:18 |
| fungi | currently closest service i think we have is the elasticsearch api proxy on logstash.o.o | 17:18 |
| *** mmaglana has quit IRC | 17:18 | |
| fungi | so that _might_ be a good home, depending on the details | 17:18 |
| jeblair | not a terrible place to put it, i think. similar security profile for the data it's dealing with. | 17:19 |
| *** jpich has quit IRC | 17:19 | |
| clarkb | ++ | 17:19 |
| mtreinish | fungi: yeah so there is a lua script (which is packaged with mysql_proxy) to split the reads and writes which mysql proxy uses to filter the queries | 17:19 |
| mtreinish | http://dev.mysql.com/doc/mysql-proxy/en/mysql-proxy-scripting.html | 17:20 |
| jeblair | clarkb: good morning; are you back on Breakfast Burrito Standard Time? | 17:20 |
| mtreinish | if you only configure mysql-proxy to have read-only backends when a update or insert query comes in, it'll just drop the connection because there isn't a backend avail | 17:20 |
| mtreinish | fungi, jeblair, clarkb: ok I'll put it on there | 17:21 |
| *** todd_dsm has joined #openstack-infra | 17:21 | |
| *** amitgandhinz has joined #openstack-infra | 17:21 | |
| fungi | mtreinish: yep, seems reasonable, and i agree logstash.o.o has a similar security profile | 17:21 |
| fungi | (and is somewhat related to the target uses as well_ | 17:22 |
| *** packet has joined #openstack-infra | 17:22 | |
| mtreinish | the only thing I'm not sure how to handle is in my local testing I needed to have an admin user and pass set to get mysql-proxy to start | 17:24 |
| mtreinish | is that another value to put in heira? | 17:24 |
| *** JayJ has joined #openstack-infra | 17:25 | |
| fungi | mtreinish: the password at least. you may just put the account name in the openstack_project class manifest in question and pass it through to your proxy module | 17:25 |
| *** mmaglana has joined #openstack-infra | 17:25 | |
| fungi | but the password does need to come from hiera | 17:25 |
| mtreinish | fungi: ok | 17:26 |
| fungi | mtreinish: also possible that the existing db account and password for subunit2sql would work there? | 17:26 |
| fungi | mtreinish: if we need a separate account (for functional or security reasons) then i'll need to add it to the trove instance manually | 17:27 |
| mtreinish | fungi: so sure I think we could reuse them for the admin creds, although I don't think SergeyLukjanov put them in hiera as separates, just the combined uri | 17:29 |
| mtreinish | I'm sure there is a sec implication of that though :) | 17:29 |
| *** pradk has quit IRC | 17:30 | |
| *** sergsh is now known as sergsh_away | 17:30 | |
| fungi | mtreinish: oh, we might want to split that uri out in that case (separate username, password and hostname hiera entries, and move the format string for the url into the manifest) | 17:30 |
| *** mpaolino has quit IRC | 17:31 | |
| *** HeOS has quit IRC | 17:31 | |
| mtreinish | fungi: yeah, we'll need to do that anyway because mysql_proxy needs host:port | 17:32 |
| fungi | mtreinish: the risk there depends on whether the new account would need more/less privs than the existing account. if they both need access to all the same things then at best the only security benefit of separating them is for auditing purposes | 17:32 |
| fungi | mtreinish: however if the proxy can, for example, get by with a read-only database account then there may be benefit to separating them | 17:33 |
| mtreinish | fungi: well they're different privs, it's only to access the mysql_proxy admin interface (which is a fake db) | 17:33 |
| openstackgerrit | Stefano Maffulli proposed openstack-infra/system-config: Adding more details about the server and link to wiki page https://review.openstack.org/135671 | 17:33 |
| mtreinish | fungi: we'll probably need separate "public" account which has read privs on the subunit2sql db so people can actually use the endpoint | 17:33 |
| *** AlexF_ has quit IRC | 17:34 | |
| *** luqas has quit IRC | 17:34 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add mysql-proxy to enable read-only access to a db https://review.openstack.org/122311 | 17:35 |
| fungi | mtreinish: okay, so i'm conflating two different accounts here... we need a read-only database account in the trove instance, and we also need a proxy admin credential (which is not something that exists in the trove database at all) | 17:35 |
| *** atan8 has joined #openstack-infra | 17:35 | |
| mtreinish | fungi: correct | 17:35 |
| *** emagana has quit IRC | 17:35 | |
| *** emagana has joined #openstack-infra | 17:36 | |
| *** thedodd has joined #openstack-infra | 17:36 | |
| fungi | mtreinish: yep, so go ahead and submit a patch passing in the appropriate new credential variables from hiera (whatever it makes sense to call them) and i'll generate and add them to our hiera file and set up the additional trove account we need | 17:36 |
| mtreinish | fungi: ^^^ that patch should show what's needed on the mysql proxy side | 17:36 |
| *** links has quit IRC | 17:37 | |
| fungi | cool. it's on my list to try to get to after storyboard migration next steps, puppet-httpd repo refresh and groups.o.o https cert ordering | 17:37 |
| *** rushiagr is now known as rushiagr_away | 17:37 | |
| openstackgerrit | Merged openstack-infra/devstack-gate: Turn off nova_allow_move_to_same_host if in aiopcpu https://review.openstack.org/135044 | 17:37 |
| mtreinish | fungi: sure, no rush, it's a low prio. I'm going to go and find some food at some point, just ping me whenever | 17:38 |
| *** e0ne has quit IRC | 17:38 | |
| *** mmaglana has quit IRC | 17:38 | |
| *** andreykurilin_ has quit IRC | 17:39 | |
| *** amotoki has quit IRC | 17:39 | |
| *** emagana has quit IRC | 17:40 | |
| *** vigneshvar has quit IRC | 17:40 | |
| *** dtalton has joined #openstack-infra | 17:41 | |
| *** ildikov has joined #openstack-infra | 17:41 | |
| *** derekh has quit IRC | 17:41 | |
| dtantsur | hi folks! any hints on who and what creates stable branches for stackforge projects? and how? | 17:42 |
| fungi | dtantsur: it depends on that project's gerrit acl | 17:42 |
| anteaya | dtantsur: that depends on the permission on the project | 17:42 |
| sc68cal | ttx: I'm looking to post a review for openstack-infra/irc-meetings but it's missing a .gitreview file in the repo | 17:42 |
| anteaya | dtantsur: which project are we discussing? | 17:42 |
| fungi | dtantsur: a group needs to be delegated create permission on refs/heads/* and depending on the branch name may also need exclusive group permission overrides on refs/heads/proposed/* and refs/heads/stable/* | 17:43 |
| dtantsur | anteaya, fungi, ironic-discoverd | 17:43 |
| *** emagana has joined #openstack-infra | 17:44 | |
| dtantsur | it has very basic acl's copy-pasted from some docs | 17:44 |
| fungi | sc68cal: step 1. create a .gitreview file and propose that as the first change to that project | 17:44 |
| sc68cal | fungi: yep, on it :) | 17:44 |
| *** armax has quit IRC | 17:44 | |
| anteaya | dtantsur: here is the acl: http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/stackforge/ironic-discoverd.config | 17:45 |
| anteaya | dtantsur: no ability to create branches | 17:45 |
| dtantsur | aha | 17:45 |
| anteaya | dtantsur: so the first step is edit the acl for the project | 17:45 |
| sdake | anteaya we absolutely need the repo | 17:46 |
| sdake | the code upstream is incomplete | 17:46 |
| sdake | but we plan to work on it | 17:46 |
| *** doug-fish1 has joined #openstack-infra | 17:46 | |
| sdake | anteaya sorry if that wasn't clear I can change it if you like | 17:46 |
| dtantsur | anteaya, what should I put there (looks like a can do it via Gerrit UI, right?) | 17:46 |
| *** Ark has joined #openstack-infra | 17:46 | |
| fungi | dtantsur: you can't do it via the gerrit ui. we manage the gerrit acls through a configuration management toolchain | 17:46 |
| *** Ark is now known as Guest82162 | 17:47 | |
| fungi | dtantsur: you need to propose a change to openstack-infra/project-config editing that file anteaya linked | 17:47 |
| dtantsur | fungi, ok, so the last question is what exactly I should put to this file | 17:47 |
| *** tnovacik has joined #openstack-infra | 17:47 | |
| *** hdd has joined #openstack-infra | 17:47 | |
| fungi | dtantsur: a group needs to be delegated create permission on refs/heads/* and depending on the branch name may also need exclusive group permission overrides on refs/heads/proposed/* and refs/heads/stable/* | 17:47 |
| jkt | jeblair: I might be missing something about the "reviewed" logic within Gertty. It appears that if I'm just a reviewer on some change, I sent my -1, when the original change owner sends an update, gertty still shows the change as "reviewed" | 17:48 |
| anteaya | sdake: okay great, can the commit message reflect the fact that the repo is necessary to your goals and not just because everyone else is doing it? | 17:48 |
| *** otter768 has joined #openstack-infra | 17:48 | |
| *** dkranz has joined #openstack-infra | 17:48 | |
| jkt | jeblair: what I'd like to ask is whether this is by design (the comment in gertty/sync.py suggest so), because then I don't get the point of the "reviewed" flag at all | 17:48 |
| *** VijayTripathi has quit IRC | 17:48 | |
| dtantsur | fungi, ok lemme try to figure out... | 17:48 |
| *** doug-fish has quit IRC | 17:48 | |
| fungi | dtantsur: see https://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/stackforge/fuel.config as a reasonable example | 17:49 |
| *** bhuvan has joined #openstack-infra | 17:49 | |
| dtantsur | fungi, oh wonderful, thank you! | 17:49 |
| fungi | dtantsur: also be warned, there's no clean way to delegate branch _deletion_ so don't create branches unless you really want them, as a gerrit admin has to delete branches you created by mistake | 17:50 |
| jeblair | jkt: since gerrit comments aren't actually threaded (currently), there's no way to know whether any subsequent activity was related to your comment or not. so gertty assumes that once you have left a non-zero vote, the change has been reviewed. if you leave a comment with no vote, the review flag will be cleared on subsequent updates. | 17:50 |
| dtantsur | fungi, I'll be careful | 17:50 |
| *** harlowja_away is now known as harlowja | 17:51 | |
| jeblair | jkt: basically leaving a non-zero vote is a signal to gertty that you have made a decision about the change and don't want to see it again; you would need out-of-band communication to know to revisit that | 17:51 |
| *** sputnik13 has joined #openstack-infra | 17:51 | |
| jeblair | jkt: that's at least the workflow i'm happy with, but i'm totally open to configurable options to change that if you need something else | 17:51 |
| sc68cal | ttx: https://review.openstack.org/135675 | 17:51 |
| openstackgerrit | Dmitry Tantsur proposed openstack-infra/project-config: Allow creating stable/XXX branches for ironic-discoverd https://review.openstack.org/135679 | 17:52 |
| dtantsur | fungi, ^^^ looking good? | 17:52 |
| *** mmaglana has joined #openstack-infra | 17:52 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/storyboard: Add API call to return task statuses https://review.openstack.org/135221 | 17:52 |
| *** tonytan4ever has quit IRC | 17:52 | |
| *** nadya_ has quit IRC | 17:52 | |
| fungi | dtantsur: that should work | 17:53 |
| *** otter768 has quit IRC | 17:53 | |
| dtantsur | thanks again! | 17:54 |
| vponomaryov | Hello everyone. Have strange problem with not found logs: http://logs.openstack.org/97/135597/4/check/check-manilaclient-dsvm-neutron-functional/0b58724/console.html#_2014-11-19_15_59_36_159 | 17:54 |
| sc68cal | fungi: ah, didn't know you were core for that repo, I only saw ttx in the list of cores | 17:54 |
| sc68cal | fungi: thanks :) | 17:54 |
| sc68cal | ahhh, it's got infra-core in the access. cool. | 17:55 |
| fungi | sc68cal: the infra-core is also included indirectly | 17:55 |
| fungi | sc68cal: yep | 17:55 |
| *** MaxV has quit IRC | 17:55 | |
| *** markmcclain has quit IRC | 17:55 | |
| fungi | sc68cal: i expect ttx to grow that core team to whoever is willing to help review meeting changes over time | 17:55 |
| * sc68cal now feels stupid for drawing bandwith from people who have more important things to do | 17:55 | |
| anteaya | sc68cal: never feel stupid about asking a question | 17:56 |
| *** jlibosva has quit IRC | 17:56 | |
| *** jlibosva has joined #openstack-infra | 17:57 | |
| jkt | jeblair: right, that makes sense. | 17:57 |
| jkt | jeblair: I guess I see what went wrong this time -- my gerrit is configured to "carry on" the labels upon trivial rebase and also commit message change | 17:58 |
| *** reed has quit IRC | 17:58 | |
| *** dtantsur is now known as dtantsur|afk | 17:58 | |
| openstackgerrit | Steven Dake proposed openstack-infra/project-config: Initial magnum client import for stackforge https://review.openstack.org/135436 | 17:58 |
| sdake | anteaya ok all set I hope :) | 17:59 |
| *** thedodd has quit IRC | 17:59 | |
| *** erlon has quit IRC | 17:59 | |
| *** vigneshvar has joined #openstack-infra | 18:00 | |
| fungi | jkt: aha... was it a commit message change you wanted to see pop back up for review? we specifically don't carry review votes across for commit message changes since we do have people reviewing commit messages | 18:00 |
| fungi | jkt: so you might consider adjusting your gerrit similarly if your community does the same | 18:00 |
| *** rmcall has quit IRC | 18:01 | |
| *** melwitt has joined #openstack-infra | 18:01 | |
| fungi | jkt: we do still carry over votes on trivial rebases (which is marginally dangerous since gerrit identifies these by comparing git patch-id, which ignores whitespace) | 18:01 |
| *** esker has quit IRC | 18:01 | |
| anteaya | sdake: thanks, that remedies my cargo cult concerns | 18:01 |
| *** rmcall has joined #openstack-infra | 18:01 | |
| jkt | fungi: it's always about balancing different sorts of evil, yeah | 18:01 |
| *** jlibosva has quit IRC | 18:02 | |
| jkt | fungi: I asked here because I found out that this was just a commit-message-change | 18:03 |
| *** armax has joined #openstack-infra | 18:03 | |
| anteaya | sdake: now address andreas's comments? | 18:03 |
| jkt | fungi: so it was mainly just some confusion on my side | 18:03 |
| fungi | jkt: yep, we ran with votes carried over on commit message changes for a while, but ultimately our reviewers want to be able to -1 for bad commit messages and have those votes reset if a new patchset potentially only corrects issues with the commit message | 18:04 |
| *** AlexF has joined #openstack-infra | 18:05 | |
| vponomaryov | Who knows, what does mean part "**/*" in code line here - https://github.com/openstack-infra/project-config/blob/3a1b91d6/jenkins/jobs/macros.yaml#L493 ? | 18:05 |
| jkt | do you guys have some cure for that gerrit's misfeature of always including just "new version of a changeset got added" in e-mails? | 18:05 |
| jkt | I see that both the web interface and `git push`' output contain info about what got changed, but not the outgoing e-mail | 18:06 |
| *** patrickeast has joined #openstack-infra | 18:06 | |
| *** ivar-lazzaro has joined #openstack-infra | 18:08 | |
| sdake | oh there were comments | 18:09 |
| sdake | sorry you hit me just as I woke up | 18:09 |
| sdake | and my brain isn't operational yet :) | 18:09 |
| anteaya | vponomaryov: in a terminal cd down in to a dir and run `ls **/*` | 18:09 |
| anteaya | that is the best way I can explain it | 18:09 |
| *** pelix has quit IRC | 18:09 | |
| anteaya | sdake: ah the need for coffee | 18:09 |
| *** links has joined #openstack-infra | 18:10 | |
| *** dtalton has quit IRC | 18:10 | |
| sdake | in my case pepsi but same idea :) | 18:10 |
| *** yamamoto has quit IRC | 18:11 | |
| anteaya | I can't do caffinee at all myself | 18:11 |
| anteaya | but I have learned how important it is to some | 18:11 |
| fungi | vponomaryov: recurses the subdirectory tree | 18:11 |
| openstackgerrit | Steven Dake proposed openstack-infra/project-config: Initial magnum client import for stackforge https://review.openstack.org/135436 | 18:11 |
| anteaya | those are the words | 18:11 |
| fungi | i made them up | 18:12 |
| *** reed has joined #openstack-infra | 18:12 | |
| *** tnovacik has quit IRC | 18:13 | |
| fungi | krotscheck: i'm starting on refreshing the lp->sb imports now. i'm going to play with https://launchpad.net/gear bug settings first since it's fairly low-volume. once we're happy with the result i'll repeat the exercise on the rest | 18:13 |
| anteaya | they are effective, I could find any at all | 18:14 |
| jeblair | fungi, krotscheck: where's the draft email? | 18:15 |
| vponomaryov | anteaya, fungi: thanks | 18:15 |
| fungi | jeblair: https://etherpad.openstack.org/p/storyboard-migration-email | 18:15 |
| krotscheck | jeblair: etherpad.openstack.org/p/storyboard-migration-email | 18:15 |
| jeblair | thx | 18:15 |
| anteaya | sdake: +2, thanks | 18:16 |
| *** dustins has joined #openstack-infra | 18:17 | |
| *** armax has quit IRC | 18:18 | |
| *** jedimike has quit IRC | 18:18 | |
| *** sputnik13 has quit IRC | 18:18 | |
| *** andreykurilin_ has joined #openstack-infra | 18:18 | |
| *** armax has joined #openstack-infra | 18:19 | |
| *** __TheDodd__ has joined #openstack-infra | 18:20 | |
| fungi | jeblair: krotscheck: i think https://launchpad.net/gear is set up as best we're going to manage for this... unless we can get lp to add storyboard.o.o as a registered bug tracker | 18:20 |
| *** VijayTripathi has joined #openstack-infra | 18:20 | |
| *** rfolco has joined #openstack-infra | 18:20 | |
| fungi | jeblair: krotscheck: i configured bugs for gear to "handled somewhere else" and then added a line in the project description with the url to the tracker | 18:20 |
| jeblair | fungi: that seems about right to me | 18:21 |
| jeblair | we should probably consider friendly project urls a requirement before we move anything more than infra :) | 18:22 |
| fungi | yep | 18:22 |
| fungi | that would be helpful | 18:22 |
| jeblair | (if we implement that, we'll have to change a bunch of lp pages, i wouldn't wish that on other projects) | 18:22 |
| *** kgiusti has quit IRC | 18:22 | |
| *** kgiusti has joined #openstack-infra | 18:22 | |
| krotscheck | What would that take? | 18:22 |
| krotscheck | Registered launchpad. | 18:23 |
| krotscheck | Sorry | 18:23 |
| krotscheck | Registered bugtracker | 18:23 |
| *** yamahata has quit IRC | 18:23 | |
| *** sputnik13 has joined #openstack-infra | 18:23 | |
| jeblair | krotscheck: hacking on lp | 18:24 |
| krotscheck | jeblair: oh god I just broke out in hives. | 18:24 |
| fungi | krotscheck: i have no idea, but StevenK, lifeless or Daviey might have details on that | 18:24 |
| *** andreykurilin_ has quit IRC | 18:24 | |
| fungi | they probably have a table of bug tracker url patterns or something | 18:24 |
| *** otter768 has joined #openstack-infra | 18:25 | |
| fungi | likely something canonical is going to want sooner rather than later anyway, since they like to link package bugs against our upstream bugs | 18:25 |
| *** teran has quit IRC | 18:26 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:26 | |
| *** jp_at_hp has quit IRC | 18:28 | |
| jkt | krotscheck: right now, storyboard greets me with a dashboard saying "Michael Krotscheck created this story." in the "Recent Events" -- I don't think it is particularly user-friendly | 18:29 |
| *** AlexF has quit IRC | 18:29 | |
| jkt | krotscheck: also, can I somehow close a bugreport that I reported? | 18:29 |
| krotscheck | jkt: Yep, yolanda is working on that. | 18:29 |
| krotscheck | jkt: Change the task status to invalid or merged. | 18:29 |
| anteaya | hey who has reviewed infra manual patches today? https://review.openstack.org/#/q/status:open+project:openstack-infra/infra-manual,n,z | 18:30 |
| jkt | krotscheck: I don't see a button for that in https://storyboard.openstack.org/#!/story/2000007 | 18:30 |
| krotscheck | jkt: We’ve been focusing on getting backend features landed, and haven’t had a chance to focus on UX until now. | 18:30 |
| *** achanda has joined #openstack-infra | 18:30 | |
| krotscheck | jkt: That story has no tasks. | 18:30 |
| jkt | krotscheck: "This story has no tasks." -- I guess I deleted them a couple of days ago in an attempt to, umm, "close this one" | 18:30 |
| krotscheck | jkt: Well, that effectively closes it. | 18:31 |
| *** zul has quit IRC | 18:31 | |
| *** SumitNaiksatam has joined #openstack-infra | 18:31 | |
| jkt | krotscheck: kind of -- it isn't visible in any tab within https://storyboard.openstack.org/#!/project/698 | 18:32 |
| *** otherwiseguy has quit IRC | 18:33 | |
| *** achanda has quit IRC | 18:33 | |
| *** achanda has joined #openstack-infra | 18:33 | |
| *** sarob has quit IRC | 18:33 | |
| jkt | krotscheck: the UI also appears to give no indication that there were some tasks within this story, and that they got deleted | 18:33 |
| *** dangers is now known as dangers_away | 18:33 | |
| *** e0ne has joined #openstack-infra | 18:34 | |
| krotscheck | jkt: I see your activity in the events timeline. | 18:34 |
| jkt | krotscheck: but anyway, if it's more or less known that the usability is not ideal, I probably shouldn't be reporting it here as you likely know all that already | 18:34 |
| krotscheck | So the data’s there. | 18:34 |
| fungi | krotscheck: possibly a stupid idea, but would it make sense to refuse to delete the only task on a story, so as to avoid taskless stories? | 18:34 |
| *** __TheDodd__ has quit IRC | 18:34 | |
| jkt | krotscheck: eh right, "deleted" is right there, indeed | 18:34 |
| *** achanda has quit IRC | 18:34 | |
| krotscheck | fungi: Not stupid at all. Something we’ve considered. | 18:34 |
| krotscheck | fungi: Honestly I haven’t really sat down and done a priority map of actions that our different users might want to take on that screen. | 18:35 |
| fungi | could help reconcile some lack of intuitiveness there | 18:35 |
| jkt | but I cannot "access" the deleted task, right? | 18:35 |
| *** links has quit IRC | 18:35 | |
| krotscheck | jkt: No, tasks get hard-deleted. | 18:36 |
| fungi | that was an explicit design decision iirc, whether to implement "soft" deletion of tasks | 18:36 |
| fungi | and it was deemed at the time to be not worth the additional complication i think? | 18:37 |
| *** arosen has left #openstack-infra | 18:38 | |
| *** arosen has joined #openstack-infra | 18:38 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add mysql-proxy to enable read-only access to a db https://review.openstack.org/122311 | 18:39 |
| *** jgallard has quit IRC | 18:39 | |
| *** achanda has joined #openstack-infra | 18:39 | |
| *** banix has quit IRC | 18:40 | |
| jkt | fungi: interesting, https://review.openstack.org/#/c/103667/2/specs/storyboard_permissions.rst has Thierry's comment on this, that this is a feature which should be restricted to admins for hard corner cases such as spam | 18:41 |
| *** emagana has quit IRC | 18:41 | |
| *** emagana has joined #openstack-infra | 18:42 | |
| jkt | unless there's an important difference between a story and a task in this context | 18:42 |
| fungi | jkt: there is. a story has tasks | 18:42 |
| *** tonytan4ever has joined #openstack-infra | 18:42 | |
| krotscheck | jkt: We had a longer discussion during a meeting where we went through all the different resource types and decided wehther we’d allow hard/soft deletes or not. | 18:42 |
| krotscheck | jkt: It ended up being either “Hard deletes” or “No delete at all" | 18:43 |
| *** pawel-palucki has quit IRC | 18:43 | |
| *** marun has quit IRC | 18:43 | |
| fungi | zaro: it looks like we're going to need to adjust the delegation on https://launchpad.net/gearman-plugin since right now you're the only one who can configure that project in lp | 18:43 |
| *** wuhg has quit IRC | 18:43 | |
| *** radez_g0n3 is now known as radez | 18:44 | |
| rcarrillocruz | hey folks, may i please get another core review on https://review.openstack.org/#/c/132194/ and https://review.openstack.org/#/c/132283/ ? it's just pending another +2... | 18:44 |
| *** arosen has quit IRC | 18:45 | |
| jkt | fungi: I got that, but what I'm saying is that maybe the option of only allowing hard deletes in cases of obvious spam (and hence restricted to admin) would make a lot of sense | 18:46 |
| *** emagana has quit IRC | 18:46 | |
| fungi | jkt: maybe, though the reasons given during the meeting where that was discussed were sufficiently convincing in favor of the current behavior. i just don't recall what those were now | 18:47 |
| *** marun has joined #openstack-infra | 18:49 | |
| *** packet has quit IRC | 18:49 | |
| openstackgerrit | Adam Gandelman proposed openstack-infra/system-config: Add iptables rule to log dropped packets https://review.openstack.org/135694 | 18:50 |
| *** bradjones has quit IRC | 18:52 | |
| *** emagana has joined #openstack-infra | 18:52 | |
| openstackgerrit | Stefano Maffulli proposed openstack-infra/system-config: Adding more details about the server and link to wiki page https://review.openstack.org/135671 | 18:53 |
| *** armax has quit IRC | 18:54 | |
| *** armax has joined #openstack-infra | 18:55 | |
| *** pcrews has quit IRC | 18:57 | |
| *** Ryan_Lane has joined #openstack-infra | 18:57 | |
| *** pcrews has joined #openstack-infra | 18:57 | |
| *** andreykurilin_ has joined #openstack-infra | 18:58 | |
| *** dannywilson has joined #openstack-infra | 18:58 | |
| *** sarob has joined #openstack-infra | 18:59 | |
| *** otherwiseguy has joined #openstack-infra | 19:00 | |
| *** mfink has joined #openstack-infra | 19:00 | |
| *** dims has quit IRC | 19:01 | |
| *** dannywilson has quit IRC | 19:02 | |
| *** marun has quit IRC | 19:03 | |
| *** marun_ has joined #openstack-infra | 19:03 | |
| *** packet has joined #openstack-infra | 19:03 | |
| *** yfried has joined #openstack-infra | 19:04 | |
| *** hdd has quit IRC | 19:04 | |
| openstackgerrit | Merged openstack-infra/project-config: Allow creating stable/XXX branches for ironic-discoverd https://review.openstack.org/135679 | 19:05 |
| *** cpowell_ has joined #openstack-infra | 19:05 | |
| *** thedodd has joined #openstack-infra | 19:05 | |
| *** MaxV has joined #openstack-infra | 19:06 | |
| *** cpowell has quit IRC | 19:07 | |
| *** dannywilson has joined #openstack-infra | 19:07 | |
| *** zz_dimtruck is now known as dimtruck | 19:07 | |
| *** emagana has quit IRC | 19:10 | |
| *** MaxV has quit IRC | 19:10 | |
| *** emagana has joined #openstack-infra | 19:11 | |
| *** marun_ has quit IRC | 19:11 | |
| anteaya | this patch to remove the jenkins EnvInject plugin from production has 5 +1's, 3 +2's and a happy current jenkins run: https://review.openstack.org/#/c/101101/ | 19:11 |
| *** yamamoto has joined #openstack-infra | 19:11 | |
| anteaya | some system-config core feel like giving it a push? | 19:12 |
| *** marun_ has joined #openstack-infra | 19:12 | |
| openstackgerrit | Merged openstack-infra/project-config: Use ip addresses for localhost in beaker tests https://review.openstack.org/135442 | 19:13 |
| *** emagana_ has joined #openstack-infra | 19:13 | |
| *** mjturek has quit IRC | 19:13 | |
| *** emagana has quit IRC | 19:13 | |
| *** VijayTripathi1 has joined #openstack-infra | 19:14 | |
| *** VijayTripathi has quit IRC | 19:14 | |
| *** ryanpetrello_ has joined #openstack-infra | 19:15 | |
| *** EntropyWorks has joined #openstack-infra | 19:16 | |
| *** yamamoto has quit IRC | 19:16 | |
| *** marun has joined #openstack-infra | 19:17 | |
| *** dimtruck is now known as zz_dimtruck | 19:17 | |
| *** andreykurilin_ has quit IRC | 19:17 | |
| *** ryanpetrello has quit IRC | 19:17 | |
| *** ryanpetrello_ is now known as ryanpetrello | 19:17 | |
| *** marun_ has quit IRC | 19:19 | |
| *** marun has quit IRC | 19:22 | |
| sdague | anteaya: thanks for reviewing the devstack uni tests job, looks like it's operating well from here | 19:25 |
| boris-42 | anteaya: hi | 19:25 |
| *** HeOS has joined #openstack-infra | 19:25 | |
| *** armax has quit IRC | 19:26 | |
| *** belmoreira has quit IRC | 19:26 | |
| *** zul has joined #openstack-infra | 19:26 | |
| *** armax has joined #openstack-infra | 19:29 | |
| *** aysyd has quit IRC | 19:29 | |
| *** armax has quit IRC | 19:30 | |
| *** markmcclain has joined #openstack-infra | 19:30 | |
| *** markmcclain has quit IRC | 19:31 | |
| *** julim has quit IRC | 19:33 | |
| *** ivar-laz_ has joined #openstack-infra | 19:34 | |
| *** ivar-laz_ has quit IRC | 19:34 | |
| *** baoli has quit IRC | 19:34 | |
| *** ivar-laz_ has joined #openstack-infra | 19:35 | |
| *** Sukhdev has joined #openstack-infra | 19:35 | |
| fungi | krotscheck: we have a minor oopsie here... i didn't check your import list closely enough obviously. https://launchpad.net/pypi-mirror is not our openstack-infra/pypi-mirror project | 19:35 |
| *** armax has joined #openstack-infra | 19:35 | |
| krotscheck | fungi: Hrm. | 19:36 |
| krotscheck | fungi: That might require some SQL magic. | 19:36 |
| fungi | krotscheck: i assume we'll want to delete some rows somewhere | 19:36 |
| krotscheck | fungi: Especially since it means that we imported unrelated users. | 19:36 |
| krotscheck | fungi: And unrelated tags | 19:36 |
| fungi | krotscheck: not too many hopefully... that project had only about a dozen bugs | 19:37 |
| *** marun has joined #openstack-infra | 19:37 | |
| *** ivar-lazzaro has quit IRC | 19:37 | |
| krotscheck | fungi: Alright, I’m closing something out, I’ll put it on my list for this afternoon. | 19:37 |
| *** rlandy has joined #openstack-infra | 19:37 | |
| *** zz_dimtruck is now known as dimtruck | 19:38 | |
| krotscheck | fungi: I’ll give you list of SQL queries to run that _should_ work. | 19:38 |
| fungi | krotscheck: appreciated. i'll move forward with the remaining refreshes and lp adjustments | 19:38 |
| fungi | nearly done with the ones i can do | 19:38 |
| *** ivar-laz_ has quit IRC | 19:38 | |
| krotscheck | fungi: Neat | 19:39 |
| *** ivar-lazzaro has joined #openstack-infra | 19:39 | |
| fungi | krotscheck: i'm tracking my progress at the bottom of https://etherpad.openstack.org/p/infra-storyboard-migration | 19:39 |
| *** marun_ has joined #openstack-infra | 19:42 | |
| *** homeless has joined #openstack-infra | 19:44 | |
| *** marun has quit IRC | 19:44 | |
| *** sarob has quit IRC | 19:45 | |
| *** dimtruck is now known as zz_dimtruck | 19:47 | |
| *** otherwiseguy has quit IRC | 19:50 | |
| fungi | mtreinish: would you mind terribly making OpenStack Continuous Integration Administrators (openstack-ci-admins) the maintainer team for https://launchpad.net/subunit2sql ? | 19:50 |
| *** ivar-laz_ has joined #openstack-infra | 19:51 | |
| mtreinish | fungi: no I don't think that should be a problem | 19:51 |
| mtreinish | I barely use the lp anyway, I think it has 1 bug | 19:51 |
| *** marun_ has quit IRC | 19:51 | |
| *** otherwiseguy has joined #openstack-infra | 19:52 | |
| fungi | mtreinish: yep, so does storyboard now ;) | 19:52 |
| mtreinish | fungi: I guess the better question is what do I have to click to change that? | 19:52 |
| mtreinish | nm, I found it | 19:52 |
| fungi | mtreinish: yeah, change details i think | 19:52 |
| *** marun_ has joined #openstack-infra | 19:53 | |
| *** VijayTripathi1 has quit IRC | 19:53 | |
| mtreinish | fungi: ok should be maintained by openstack-ci-admins now | 19:53 |
| fungi | mtreinish: yep, perfect--thanks! | 19:53 |
| fungi | zaro: if you're around, i need you to do the same for https://launchpad.net/gearman-plugin | 19:54 |
| *** ivar-lazzaro has quit IRC | 19:54 | |
| *** MaxV has joined #openstack-infra | 19:55 | |
| *** thedodd has quit IRC | 19:58 | |
| *** matrohon has joined #openstack-infra | 19:58 | |
| *** zul has quit IRC | 19:59 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add mysql-proxy to enable read-only access to a db https://review.openstack.org/122311 | 20:01 |
| *** ivar-laz_ has quit IRC | 20:01 | |
| *** emagana_ has quit IRC | 20:01 | |
| *** jamespage_ has joined #openstack-infra | 20:02 | |
| *** emagana has joined #openstack-infra | 20:02 | |
| *** vigneshvar has quit IRC | 20:03 | |
| *** liam__ has joined #openstack-infra | 20:03 | |
| *** liam__ has quit IRC | 20:03 | |
| *** tonytan4ever has quit IRC | 20:05 | |
| *** emagana_ has joined #openstack-infra | 20:05 | |
| *** otherwiseguy has quit IRC | 20:05 | |
| *** emagana has quit IRC | 20:06 | |
| *** sabeen1 has quit IRC | 20:07 | |
| *** jamespage_ has quit IRC | 20:07 | |
| *** sabeen2 has joined #openstack-infra | 20:07 | |
| *** matrohon is now known as matrohon_ | 20:09 | |
| *** ivar-lazzaro has joined #openstack-infra | 20:09 | |
| *** wenlock has quit IRC | 20:09 | |
| *** marun_ has quit IRC | 20:09 | |
| *** matrohon_ is now known as matrohon | 20:09 | |
| *** wenlock has joined #openstack-infra | 20:09 | |
| *** banix has joined #openstack-infra | 20:09 | |
| *** thedodd has joined #openstack-infra | 20:09 | |
| *** david-lyle is now known as david-lyle_afk | 20:10 | |
| *** zul has joined #openstack-infra | 20:10 | |
| *** markmcclain has joined #openstack-infra | 20:11 | |
| *** ldnunes has quit IRC | 20:11 | |
| openstackgerrit | Marton Kiss proposed openstack-infra/system-config: Enable SSL in groups-dev.openstack.org https://review.openstack.org/135708 | 20:12 |
| tristanC | Folks, for your information stable/icehouse neutron tests are blocked until this https://review.openstack.org/#/c/135692/ (global-requirement update) get merged... | 20:13 |
| *** jamespage_ has joined #openstack-infra | 20:14 | |
| *** jamespage_ has quit IRC | 20:14 | |
| *** MaxV has quit IRC | 20:16 | |
| *** dmsimard is now known as dmsimard_away | 20:16 | |
| mtreinish | tristanC: +2 | 20:16 |
| *** MaxV has joined #openstack-infra | 20:16 | |
| dstufft | Do y'all still support pip < 1.4? | 20:16 |
| mtreinish | thanks for respinning it | 20:16 |
| dstufft | (Asking because of https://github.com/openstack/requirements/blob/master/global-requirements.txt#L135-L140) | 20:16 |
| *** dims has joined #openstack-infra | 20:16 | |
| *** sputnik13 has quit IRC | 20:17 | |
| fungi | tristanC: though you need a second stable core to approve it. normal reqs reviewers don't have access | 20:17 |
| *** otherwiseguy has joined #openstack-infra | 20:17 | |
| dstufft | mordred: clarkb fungi lifeless ^ (fwiw if PEP 440 lands for pip 6.0 that is gonna break and I think it should, that spec doesn't make sense) | 20:19 |
| fungi | dstufft: is the multiple ranges a problem? | 20:19 |
| fungi | jogo: doesn't hacking use that too ^ ? | 20:19 |
| dstufft | fungi: well the overlap doesn't make sense, if you treat the , as an AND then no versions fit that overlap, if you treat it as an OR than everything version matches that overlap | 20:20 |
| *** marun has joined #openstack-infra | 20:20 | |
| *** ociuhandu has quit IRC | 20:21 | |
| fungi | dstufft: the way i read that, it's attempting (poorly) to exclude sqlalchemy 0.9 versions prior to 0.9.7 | 20:21 |
| krotscheck | It’s that last 5% of usability that is always the hardest :/ | 20:21 |
| fungi | dstufft: but yeah, it's a bunch of anded inequalities which end up as the universal set | 20:22 |
| fungi | dstufft: so i agree basically unparseable | 20:22 |
| sdague | dstufft: whether or not it makes sense, this is the way pip works today | 20:22 |
| sdague | so it's a defacto interface, changing that probably requires some warning | 20:23 |
| fungi | dstufft: anyway, we do still expect pip to work on ubuntu 12.04 (which carries pip 1.0 i believe) | 20:23 |
| *** unicell has quit IRC | 20:24 | |
| *** wenlock has quit IRC | 20:25 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 20:25 | |
| fungi | hrm, my recollection about hacking (ab)using that to support discrete ranges of versions is outdated. now i remember that's why we special-case skip hacking in requirements update proposal patches | 20:25 |
| *** ivar-lazzaro has quit IRC | 20:26 | |
| *** wenlock has joined #openstack-infra | 20:26 | |
| *** rfolco has quit IRC | 20:27 | |
| dstufft | sdague: well it "works" primarily as an accident and it's documented that using conflicting ranges is "meaningless" and may produce "bizare results" | 20:27 |
| sdague | dstufft: ok, well is there a new spec to support exclusion / inclusion ranges? | 20:28 |
| sdague | because that's actually important to us, having to != 30 versions is kind of not fun | 20:28 |
| dstufft | PEP 440 defines the specifier syntax | 20:29 |
| *** sarob has joined #openstack-infra | 20:31 | |
| fungi | dstufft: are there good examples of how the inclusive/exclusive ordered comparisons are intended to work? and is there any way to make that coexist with something which will produce a sane fallback in older (1.0) pip versions? | 20:31 |
| jogo | fungi: I was using multiple ranges to bump the reqs for hacking | 20:32 |
| jogo | because we pin flake8 and pep8 etc | 20:32 |
| fungi | jogo: oh, right, so it was an intermittent thing | 20:32 |
| jogo | fungi: yeah, and I am sure there are other ways to work around that issue | 20:33 |
| fungi | jogo: anyway, sounds like we need to avoid that because new pip will be teh unhappiez | 20:33 |
| dstufft | specifically it's not multiple ranges, but conflicting ranges | 20:33 |
| dstufft | it supports multiple ranges just fine | 20:33 |
| jogo | dstufft:we were using things like: ==1.1,==1,2 | 20:33 |
| *** talluri has joined #openstack-infra | 20:34 | |
| fungi | dstufft: i'm not sure what you mean by conflicting ranges. how can you have multiple ranges which don't conflict, or rather what is the quality which would make two ranges conflict? | 20:35 |
| dstufft | fungi: the inclusive operations work pretty much exactly how they do today, greater than/less than equal to the given version inclusive, the exclusive ones are a little different, something like foo < 1.0 won't match 1.0a1 anymore because it's defined as "less than 1.0 and not 1.*" | 20:35 |
| dstufft | fungi: well taking the SQLALchemy spec you guys have, SQLAlchemy>=0.8.4,<=0.8.99,>=0.9.7,<=0.9.99, it's impossible to have a single version that satisified both <=0.8.99 AND >=0.9.7 | 20:36 |
| *** AJaeger has joined #openstack-infra | 20:36 | |
| fungi | dstufft: in the sqlalchemy example we are trying to accomplish "supports (0.8.4 through 0.8.99) AND (0.9.7 through 0.9.99)" | 20:37 |
| dstufft | the comments indicate that the reason for doing that horrible hack is because SQLAlchemy released a pre-release to PypI and you didn't want to match that | 20:37 |
| *** zz_gondoi is now known as gondoi | 20:38 | |
| *** julim has joined #openstack-infra | 20:38 | |
| fungi | dstufft: maybe, though i took that as simply an example of potential versions it was trying to avoid | 20:39 |
| *** julim has quit IRC | 20:39 | |
| fungi | dstufft: looking on pypi now, the only version i see which we might have tried to avoid was SQLAlchemy-0.8.0b2.tar.gz | 20:40 |
| *** talluri has quit IRC | 20:40 | |
| *** amcrn has joined #openstack-infra | 20:40 | |
| fungi | except that's already less than the stated minimum there | 20:40 |
| dstufft | fungi: So in terms of a constraint what you're really looking for is (>0.8.4 AND <0.8.99) OR (>0.9.7 AND < 0.9.99) | 20:41 |
| *** julim has joined #openstack-infra | 20:41 | |
| *** dustins has quit IRC | 20:42 | |
| *** wenlock has quit IRC | 20:42 | |
| fungi | dstufft: yep, though in this case we could probably make it SQLAlchemy>=0.8.4,!=0.9.0,!=0.9.1,!=0.9.2,!=0.9.3,!=0.9.4,!=0.9.5,!=0.9.6,<0.10 | 20:42 |
| *** wenlock has joined #openstack-infra | 20:43 | |
| fungi | dstufft: and assuming their devs don't release any new versions earlier than 0.9.7 we should be fine | 20:43 |
| dstufft | There isn't a way to do that in PEP 440 without a chain of != like you just specified, the downside of the chain you just posted is that <0.10 will match 0.10 pre-releases in pip < 1.4 if they are uploaded | 20:43 |
| fungi | dstufft: ahh, so <=0.9.99 still there i guess | 20:44 |
| fungi | or something hackish like that | 20:44 |
| dstufft | yea | 20:44 |
| *** dustins has joined #openstack-infra | 20:45 | |
| *** AlexF has joined #openstack-infra | 20:45 | |
| dstufft | if you want proper support for OR that'd probably require a new PEP and discussion on distutils-sig at this point | 20:46 |
| *** achanda has quit IRC | 20:46 | |
| *** AlexF has quit IRC | 20:46 | |
| dstufft | (for the record, the reason I noticed this was because I was reading through global requirements to see if anything but SQLAlchemy would break here, I don't see anything else other than that) | 20:47 |
| fungi | dstufft: thanks for checking! we should probably update that before we start using the brave new pep-440 world pip | 20:48 |
| dstufft | (also for the record, the number of digits in a specifier is siginfcant in PEP 440, >7 is different than >7.0, >7 implies a !=7.* rule, while >7.0 implies a !=7.0.* rule, 7.1 will get excluded for a !=7.* rule, but not a !=7.0.* rule) | 20:49 |
| fungi | dstufft: i expect in about a year we'll be able to consider seriously dropping old pip support since the oldest distros we'll be actively supporting for server releases will be ubuntu 14.04 and centos/rhel 7 | 20:50 |
| fungi | at the moment we're still testing stable branches on ubuntu 12.04 and centos 6 though | 20:51 |
| dstufft | yea | 20:51 |
| dstufft | that'll be good, pip 1.4+ makes people uploading pre-releases to PyPI way saner | 20:52 |
| *** jamespage_ has joined #openstack-infra | 20:52 | |
| dstufft | though it won't be until the PEP 440 stuff lands that the --pre flag isn't broken if people upload any bdist_dumbs :/ | 20:52 |
| fungi | though i think we've got more and more people seeing the light with semver-like versioning and getting away from prerelease version numbers | 20:53 |
| fungi | which seriously uncomplicates version number comparisons | 20:53 |
| dstufft | (untangling this mess sucks, because you more or less have an under documented, if it's oducmented at all, ball of twine where things are monkeypatching other things and it's not entirely clear what you're allowed to rely on and what you're not, so we break things more now than some folks would like, and i feel bad about that, but hopefully this will end up with something saner at the end) | 20:54 |
| dstufft | fungi: y'all will like ~= once you can depend on whatever version of pip gets PEP 440 (shooting for 6.0) | 20:54 |
| *** AJaeger has quit IRC | 20:54 | |
| dstufft | ~= is basically the semver operator | 20:54 |
| *** achanda has joined #openstack-infra | 20:55 | |
| fungi | dstufft: right, it's the overheating spacebar | 20:55 |
| *** emagana_ has quit IRC | 20:55 | |
| dstufft | ~=2.0 is >=2.0,<3 | 20:55 |
| openstackgerrit | Aaron Rosen proposed openstack-infra/project-config: Create congressmiddleware repo https://review.openstack.org/133567 | 20:56 |
| fungi | yeah, i meant the not entirely clear what you're allowed to rely on is the overheating spacebar problem | 20:56 |
| fungi | a la http://xkcd.com/1172/ | 20:56 |
| dstufft | fungi: yea | 20:56 |
| openstackgerrit | Cedric Brandily proposed openstack-infra/git-review: Support authentication in run_http_exc https://review.openstack.org/124410 | 20:56 |
| openstackgerrit | Cedric Brandily proposed openstack-infra/git-review: Handle correctly http error raise in run_http_exc https://review.openstack.org/124396 | 20:56 |
| dstufft | and things were broke for a long time, so people hacked around things to make them work, and we're now breaking a lot of those hacks :/ | 20:56 |
| fungi | dstufft: yep, i do not envy | 20:57 |
| dstufft | the automatic wheel cache is gonna break things I think | 20:57 |
| dstufft | once we get that in | 20:57 |
| dstufft | (future plan is to instead of using setuptools to install stuff, we always use setuptools to build a Wheel unless one is available from PyPI, and then install from that Wheel, and cache that wheel locally so you only build once) | 20:57 |
| *** ivar-lazzaro has joined #openstack-infra | 20:57 | |
| fungi | dstufft: sneaky | 20:58 |
| fungi | that sounds efficient | 20:58 |
| dstufft | it'll speed things up A LOT for people, and it reduces the complexity of pip because we reduce the ways to install things | 20:58 |
| dstufft | (right now setup.py install, setup.py develop, Wheel) | 20:59 |
| dstufft | Not sure how to handle setup.py develop (aka -e) yet | 20:59 |
| dstufft | it might just have to stay special | 20:59 |
| *** emagana_ has joined #openstack-infra | 20:59 | |
| *** e0ne has quit IRC | 21:00 | |
| *** talluri has joined #openstack-infra | 21:01 | |
| *** emagana_ has quit IRC | 21:01 | |
| *** emagana has joined #openstack-infra | 21:02 | |
| openstackgerrit | Merged openstack/requirements: Add pykmip https://review.openstack.org/114037 | 21:03 |
| *** julim has quit IRC | 21:04 | |
| *** julim has joined #openstack-infra | 21:07 | |
| *** achanda_ has joined #openstack-infra | 21:09 | |
| *** andreykurilin_ has joined #openstack-infra | 21:09 | |
| *** otherwiseguy has quit IRC | 21:09 | |
| *** jungleboyj has joined #openstack-infra | 21:10 | |
| *** ivar-lazzaro has quit IRC | 21:10 | |
| jungleboyj | Has anyone noticed that gate-grenade-dsvm seems to be failing for stable/juno right now? | 21:10 |
| *** ivar-lazzaro has joined #openstack-infra | 21:10 | |
| *** otherwiseguy has joined #openstack-infra | 21:11 | |
| *** todd_dsm has quit IRC | 21:11 | |
| mtreinish | jungleboyj: it's probably caused by: https://review.openstack.org/135692 | 21:11 |
| mtreinish | once that lands try it again | 21:11 |
| *** shakamunyi_ has joined #openstack-infra | 21:12 | |
| *** talluri has quit IRC | 21:12 | |
| *** achanda has quit IRC | 21:12 | |
| *** AlexF has joined #openstack-infra | 21:12 | |
| jungleboyj | mtreinish: Thank you. That looks like the problem I am seeing. I think we need to get that into stable/juno as well. Should I propose that? | 21:13 |
| fungi | jungleboyj: or ask dims if he's around | 21:13 |
| *** yamamoto has joined #openstack-infra | 21:14 | |
| jungleboyj | dims: ^^ | 21:14 |
| dims | jungleboyj: url to a failure? | 21:14 |
| jungleboyj | dims: http://logs.openstack.org/13/130213/1/gate/gate-grenade-dsvm/aa56da8/logs/grenade.sh.txt.gz | 21:14 |
| dims | looking | 21:14 |
| fungi | dims: asking for a stable/juno backport of the suds-jurko reqs change | 21:14 |
| dims | yep | 21:15 |
| *** ivar-lazzaro has quit IRC | 21:15 | |
| jungleboyj | Have a couple Cinder changes that have hit that today. | 21:15 |
| fungi | dims: from https://review.openstack.org/#/q/If91a87ede96a82a2a0d181ee0d4fdc102dc0299d,n,z it looks like you only backported from master to stable/icehouse | 21:15 |
| fungi | so just wondering if there was a reason not to do it on juno | 21:15 |
| dims | fungi: yep, wanted to be sure we saw failures before filing one | 21:15 |
| fungi | ahh, perfect | 21:15 |
| *** JayJ has quit IRC | 21:15 | |
| jungleboyj | :-) /me has been useful today | 21:16 |
| *** JayJ has joined #openstack-infra | 21:16 | |
| *** cnesa2 has joined #openstack-infra | 21:16 | |
| *** cnesa has quit IRC | 21:16 | |
| *** achanda_ has quit IRC | 21:17 | |
| dims | jungleboyj: yay! | 21:17 |
| *** achanda has joined #openstack-infra | 21:17 | |
| jungleboyj | dims: You going to do the cherry-pick then? | 21:17 |
| dims | jungleboyj: on my way, yes | 21:17 |
| jungleboyj | Cool. Thanks! | 21:17 |
| jungleboyj | dims: No bug to reverify againstt? | 21:18 |
| *** talluri has joined #openstack-infra | 21:18 | |
| *** yamamoto has quit IRC | 21:18 | |
| *** talluri has quit IRC | 21:20 | |
| dims | jungleboyj: tristan beat me to it - https://review.openstack.org/#/c/135721/ | 21:21 |
| jkt | playing around with zuul & gerrit -- it seems that it's perfectly OK to refer to a non-existing job from a project's configuration | 21:21 |
| dims | jungleboyj: no | 21:21 |
| jkt | if I understand this correctly, it would be rather hard to check this up-front, but anyway, I'm a bit surprised that I don't see any trace of the launched jobs in zuul-server's log, not even in the -d mode | 21:22 |
| *** AlexF has quit IRC | 21:22 | |
| *** fandi has joined #openstack-infra | 21:22 | |
| jkt | also, shouldn't zuul somehow notice that no gearman worker (I have no workers right now) "responded" to the job in question, and raise some fuss about that in the logs? | 21:22 |
| fungi | jkt: we check it up-front with a job called gate-project-config-layout | 21:23 |
| jkt | ah, well, I mean in my private setup | 21:23 |
| fungi | jkt: right, you could check it too with a similar (or the same) script ;) | 21:23 |
| fungi | jkt: as for there being no gearman worker available for a job, that's normal for us under load. the job stays in a pending state until an available worker takes the work request for it | 21:24 |
| jkt | fungi: do you happen to have a link for an actual definition of the gate-project-config-layout? | 21:25 |
| *** armax has quit IRC | 21:25 | |
| fungi | jkt: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/infra.yaml#n30 | 21:25 |
| *** mfink_ has quit IRC | 21:25 | |
| *** gyee has quit IRC | 21:26 | |
| fungi | jkt: basically check out the openstack-infra/project-config repo and then run 'tox -e zuul' in it | 21:26 |
| jkt | fungi: and https://git.openstack.org/cgit/openstack-infra/project-config/tree/tox.ini#n52 , right? | 21:26 |
| * jkt needs some time to wrap his head around these levels of indirections | 21:26 | |
| fungi | jkt: yep | 21:26 |
| *** mfink_ has joined #openstack-infra | 21:27 | |
| fungi | jkt: which in turn is running tools/run-layout.sh followed by tools/layout-checks.py | 21:27 |
| *** achanda has quit IRC | 21:27 | |
| *** Ryan_Lane has quit IRC | 21:27 | |
| *** emagana has quit IRC | 21:27 | |
| jeblair | jkt: if zuul is triggered and wants to run a job for which no worker is registered, it will make a fuss and return "NOT_REGISTERED" as the build result for that job | 21:28 |
| *** emagana has joined #openstack-infra | 21:28 | |
| *** gyee has joined #openstack-infra | 21:28 | |
| jkt | so as hashar(?) told me during the summit, the jjb is also used for processing these templates, not just for preparing jobs within Jenkins | 21:28 |
| *** gyee has quit IRC | 21:28 | |
| *** mfink_ has quit IRC | 21:28 | |
| *** kgiusti has left #openstack-infra | 21:29 | |
| *** gyee has joined #openstack-infra | 21:29 | |
| jeblair | jkt: it doesn't make a fuss about it until it's an issue though | 21:29 |
| *** mfink_ has joined #openstack-infra | 21:29 | |
| *** lttrl has joined #openstack-infra | 21:29 | |
| jkt | jeblair: how come http://fpaste.org/152285/14164325/ happens, then? | 21:29 |
| *** gyee has quit IRC | 21:29 | |
| fungi | jkt: well, it does that in service of preparing the jobs for jenkins, but there has been recent libification added to jjb allowing us to also use it to do other things with the templated job data (like translate them into some format used by a different kind of job system) | 21:30 |
| *** gyee has joined #openstack-infra | 21:30 | |
| jkt | using 2.0.0-285-g8c81dc3, with just zuul-server running | 21:30 |
| jkt | maybe I would hit that as soon as I start the zuul-merger? | 21:30 |
| *** achanda has joined #openstack-infra | 21:30 | |
| *** achanda has quit IRC | 21:30 | |
| *** armax has joined #openstack-infra | 21:31 | |
| jeblair | jkt: i've never used "zuul show running-jobs". however, i feel that is not the output i would expect from that command in that circumstance. :( | 21:31 |
| *** achanda has joined #openstack-infra | 21:31 | |
| *** gyee has quit IRC | 21:31 | |
| fungi | jkt: yeah, the job will be in a pending state if it hasn't been serviced by a merger yet | 21:31 |
| *** mfink_ has quit IRC | 21:31 | |
| *** mfink_ has joined #openstack-infra | 21:32 | |
| jkt | ok, let's start a merger and see if it breaks | 21:32 |
| fungi | er, if the changeish hasn't been serviced by a merger yet that is | 21:32 |
| jeblair | jkt: in particular, i do not think that is a list of "running jobs". it looks like it is a list of "known or suspected job names" | 21:32 |
| *** emagana has quit IRC | 21:33 | |
| jeblair | jkt: the merger will throw a fuss about "non/existing/project", but will clone "project-foo" and otherwise continue to operate correctly | 21:33 |
| * clarkb is almost back | 21:33 | |
| jeblair | (assuming project-foo exists) | 21:33 |
| *** baoli has joined #openstack-infra | 21:33 | |
| *** Ryan_Lane has joined #openstack-infra | 21:34 | |
| fungi | krotscheck: so i've locked down bugs for all the projects in the etherpad with the exception of gearman-plugin (need zaro's help to delegate control) and openstack-ci (because elastic-recheck still uses it, but i did refresh its import again just now) | 21:34 |
| jkt | yes, project-foo is an actual project within gerrit, the project which triggered these jobs | 21:34 |
| *** sarob has quit IRC | 21:34 | |
| jeblair | jkt: so if you porpose a change to project-foo, you should get a report back with "testing-job: SUCCESS" and "non-existing-job: NOT_REGISTERED". | 21:34 |
| fungi | krotscheck: we can probably proceed with the announcement once everyone's okay with the wording | 21:34 |
| *** Sukhdev has quit IRC | 21:34 | |
| clarkb | fungi woot | 21:35 |
| jeblair | jkt: (after starting the merger) | 21:35 |
| krotscheck | fungi: Cool. | 21:35 |
| jeblair | oh right, i had some wording thoughts... 1 sec | 21:35 |
| fungi | i haven't really reviewed the wording yet because i've been too lazy^H^H^H^Hsidetracked by other stuff. looking now | 21:36 |
| clarkb | fungi zaro is afk until after turkey day | 21:37 |
| clarkb | do we have any other options there? | 21:37 |
| fungi | clarkb: we can ask a lp admin for control of the project i guess | 21:37 |
| fungi | clarkb: or we can just live with it as-is for now and refresh its import when he gets back | 21:38 |
| clarkb | fungi krotscheck where is announcement thibg? | 21:38 |
| jeblair | https://etherpad.openstack.org/p/storyboard-migration-email | 21:38 |
| *** dizquierdo has joined #openstack-infra | 21:39 | |
| *** AJaeger has joined #openstack-infra | 21:40 | |
| *** AlexF has joined #openstack-infra | 21:40 | |
| *** Sincler has quit IRC | 21:40 | |
| *** jamespage_ has quit IRC | 21:41 | |
| *** Guest82162 has quit IRC | 21:42 | |
| fungi | krotscheck: does the call for storyboard bug reports maybe merit a link to where to report them? (seems like it should be common sense, but...) | 21:42 |
| *** dustins has quit IRC | 21:43 | |
| krotscheck | fungi: Never underestimate the power of TL/DR | 21:43 |
| krotscheck | i.e. yes. | 21:43 |
| *** salv-orlando_ has joined #openstack-infra | 21:44 | |
| *** AlexF has quit IRC | 21:44 | |
| fungi | krotscheck: a very geeky part of me is mildly disappointed that storyboard isn't self-hosted as https://storyboard.openstack.org/#!/project/0 | 21:44 |
| *** salv-orlando has quit IRC | 21:44 | |
| *** salv-orlando_ is now known as salv-orlando | 21:44 | |
| jkt | jeblair: (back to gertty and the "reviewed" flag) -- I don't actually have that apply-votes-on-commitmsg-updated, just on the trivial rebase | 21:45 |
| *** Sincler has joined #openstack-infra | 21:45 | |
| jeblair | jkt: so your vote was cleared in gerrit but the review flag was not cleared in gertty? | 21:45 |
| jkt | it seems so, yeah | 21:46 |
| *** emagana has joined #openstack-infra | 21:46 | |
| jkt | jeblair: I'm running with your patch for selective refreshes | 21:46 |
| jeblair | jkt: sounds vaguely bug-like. if that continues to happen or is otherwise reproducible, it may be worth running with '-d' and examining the log around when the sync was updated. | 21:47 |
| jkt | I have copyMinScore and copyAllScoresOnTrivialRebase, which confused me a bit | 21:47 |
| jkt | all right, I can do that -- do you want me to run this without your patch, or with it? | 21:47 |
| fungi | jkt: that matches our setup then | 21:48 |
| krotscheck | fungi: That’s easily changed :) | 21:48 |
| clarkb | jeblair I would appreciate your thoughts on https://review.openstack.org/#/c/135070/5 dont need full review just a hell no or ok :) | 21:49 |
| jeblair | jkt: the selective refresh patch should, at worst, only cause the ui not to update (so you can test that by hitting esc and then returning to the same screen) | 21:49 |
| jeblair | jkt: so i think as long as that's not the case, should be fine to run with the patch. | 21:50 |
| fungi | jkt: just to confirm, your All-Projects project.config doesn't use "copyAllScoresIfNoCodeChange = true"? | 21:50 |
| jungleboyj | dims: Approved. | 21:51 |
| jeblair | krotscheck: (et al) i have inserted a paragraph that i'm not entirely sure i'm happy with, but i thought that something mentioning the general story-arc of what we're trying to do with storyboard would be a good thing | 21:51 |
| dims | jungleboyj: thanks | 21:51 |
| jungleboyj | dims: Welcome. Thanks for getting a fix out there. | 21:51 |
| *** xyang0 has quit IRC | 21:51 | |
| fungi | jeblair: seems fine to me | 21:52 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config: Import puppet-openstack-cloud to Stackforge https://review.openstack.org/135729 | 21:53 |
| AJaeger | 21:54 | |
| *** jamielennox|away is now known as jamielennox | 21:54 | |
| jkt | fungi: yes | 21:54 |
| jeblair | clarkb: +1 | 21:54 |
| fungi | jkt: cool, just making sure since it's a really, really confusingly-named config option | 21:55 |
| openstackgerrit | Matthew Treinish proposed stackforge/gertty: Fix vote order in review dialog box https://review.openstack.org/135732 | 21:55 |
| clarkb | jeblair danke | 21:55 |
| mtreinish | jeblair: ^^^ finally got around to fixing my vote ordering complaint on gertty | 21:56 |
| mtreinish | sorry it took so long | 21:56 |
| *** emagana has quit IRC | 21:56 | |
| *** mrmartin has quit IRC | 21:56 | |
| *** emagana has joined #openstack-infra | 21:56 | |
| jeblair | fungi: disable-dont-fail-to-break: true? | 21:56 |
| jeblair | mtreinish: woot, thanks! | 21:57 |
| fungi | jeblair: that's the one which turns on non-working | 21:57 |
| fungi | or is it that it allows you to turn off the preventative measures against working? i can't remember now | 21:58 |
| *** shakamunyi_ has quit IRC | 21:59 | |
| *** esker has joined #openstack-infra | 21:59 | |
| *** Ryan_Lane has quit IRC | 22:01 | |
| *** cpowell_ has quit IRC | 22:01 | |
| *** AJaeger has quit IRC | 22:02 | |
| *** Ryan_Lane has joined #openstack-infra | 22:02 | |
| *** emagana has quit IRC | 22:02 | |
| *** emagana has joined #openstack-infra | 22:03 | |
| jkt | jeblair: so this http://jkt.flaska.net/tmp/log-new-change-created-ctrl-r-but-not-shown-in-the-change-list-view is what I have in my log when I push a completely new change (initial revision) and press ctrl-r in the view of "Unreviewed changes in project-foo" | 22:04 |
| lifeless | fungi: don't we upgrade to latest pip as the first thing we do ? | 22:04 |
| jkt | jeblair: that screen does not get updated here | 22:04 |
| *** rmcall has quit IRC | 22:05 | |
| jkt | jeblair: I suppose that it will after I Esc and go forward again, though | 22:05 |
| jkt | jeblair: right, the new change is visible after that step | 22:05 |
| lifeless | dstufft: what we need is a transition plan where we aren't broken at any stage along it | 22:05 |
| lifeless | dstufft: e.g. pip supports saner language, migrate to that, pip drops old language | 22:06 |
| *** AlexF has joined #openstack-infra | 22:06 | |
| jkt | jeblair: at that point in time, the log looks like this: http://jkt.flaska.net/tmp/log-2-after-esc-and-forward | 22:06 |
| *** rmcall has joined #openstack-infra | 22:07 | |
| *** zul has quit IRC | 22:07 | |
| *** emagana has quit IRC | 22:07 | |
| anteaya | sdague: thanks for the update | 22:08 |
| *** andreykurilin_ has quit IRC | 22:08 | |
| *** ivar-lazzaro has joined #openstack-infra | 22:08 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard-webclient: Streamlined task edit form. https://review.openstack.org/135739 | 22:08 |
| *** andreykurilin_ has joined #openstack-infra | 22:08 | |
| *** Sukhdev has joined #openstack-infra | 22:08 | |
| openstackgerrit | Merged openstack-infra/project-config: Initial magnum client import for stackforge https://review.openstack.org/135436 | 22:09 |
| dstufft | lifeless: well the problem is that there isn't a good way to do a phase in, we don't have the information avialable to us to let people opt in. I mean the person invoking pip could opt in or out, but that's the wrong person to do that because it should be the person who wrote the specifiers | 22:09 |
| *** miqui has quit IRC | 22:09 | |
| *** banix has quit IRC | 22:09 | |
| *** openstackgerrit has quit IRC | 22:10 | |
| *** openstackgerrit has joined #openstack-infra | 22:10 | |
| ianw | clarkb: https://review.openstack.org/133914 & https://review.openstack.org/133881 <- follow-on to the nosetest.xml copy removal if you get a chance | 22:10 |
| *** mwagner_lap has quit IRC | 22:10 | |
| *** zul has joined #openstack-infra | 22:11 | |
| openstackgerrit | Merged openstack-infra/project-config: Re-enable spec/unit tests for puppet-ceph https://review.openstack.org/135602 | 22:11 |
| krotscheck | jeblair: LGTM | 22:12 |
| anteaya | boris-42: how can I help you? | 22:12 |
| fungi | lifeless: the case for pip 1.0 is i think not that we intend to use it in our test systems, just that we expect, e.g., someone on a current supported ubuntu precise machine to be able to pip install python-openstackclient or whatever without first needing to find a new version of pip | 22:12 |
| boris-42 | anteaya: hi there | 22:12 |
| boris-42 | anteaya: so about rally-fakevirt stuff | 22:12 |
| boris-42 | anteaya: it will require creating new job template with new name, and removing old* | 22:13 |
| boris-42 | anteaya: does that work for you? | 22:13 |
| fungi | krotscheck: jeblair: the announcement looks good to me... i'm heading out to eat now but didn't want my absence to hold up sending it | 22:14 |
| anteaya | boris-42: if that enables you to fix the name that that is what it takes | 22:15 |
| boris-42 | anteaya: okay I'll do | 22:15 |
| anteaya | I won't know exactly if that works for me or not until I see the code | 22:15 |
| boris-42 | anteaya: it will be something like copy-paste lol=) | 22:15 |
| anteaya | but based on what you have said so far, I do believe it is a good direction to start | 22:15 |
| *** AlexF has quit IRC | 22:15 | |
| anteaya | or how about use a template that already exists? | 22:15 |
| boris-42 | anteaya: hm I need to change the name of template* | 22:16 |
| anteaya | that is rather the point of templates | 22:16 |
| boris-42 | anteaya: that is what you are asking* | 22:16 |
| *** emagana has joined #openstack-infra | 22:16 | |
| boris-42 | anteaya: I am not sure that I am able to that in templates, like reusing templates | 22:16 |
| anteaya | what I am saying is that rally generally needs to be more aware of how it names its jobs | 22:16 |
| anteaya | and the two examples I have so far is the libvert one and the name-name one | 22:16 |
| anteaya | I havne't reviewed a lot of rally jobs | 22:17 |
| anteaya | but so far my impression is that the naming is sloppy and poorly thought out | 22:17 |
| anteaya | I'm hoping you can improve that going forward | 22:17 |
| boris-42 | anteaya: did you see my reply? | 22:17 |
| boris-42 | anteaya: about name-name | 22:17 |
| boris-42 | anteaya: it's not name-name* | 22:17 |
| anteaya | no, I got back from my walk and saw your ping | 22:17 |
| openstackgerrit | Jaroslav Henner proposed openstack-infra/jenkins-job-builder: Allow multiple comment-added events in gerrit trig. https://review.openstack.org/124568 | 22:17 |
| anteaya | so I replied to your ping | 22:17 |
| lifeless | fungi: I don't think thats a particularly needed thing though | 22:18 |
| boris-42 | anteaya: ah okay | 22:18 |
| boris-42 | anteaya: one sec | 22:18 |
| lifeless | fungi: *lots* of stuff utterly fails to install with system default pip these days | 22:18 |
| boris-42 | anteaya: https://review.openstack.org/#/c/135132/ | 22:18 |
| lifeless | fungi: wheels will dtrt half the time, for instance. | 22:18 |
| *** matrohon has quit IRC | 22:18 | |
| boris-42 | anteaya: so to be short: it's not name-name it's <template-name>-<part_of_file_name_in_project> | 22:18 |
| anteaya | lemme is not an english word, the words you are looking for are let me | 22:19 |
| *** dizquierdo has quit IRC | 22:19 | |
| boris-42 | anteaya: okay.. | 22:19 |
| anteaya | if you are trying to convince me your names are properly thought out using slang is a poor way to change my mind | 22:19 |
| *** dkranz has quit IRC | 22:20 | |
| anteaya | whatever the reason, using projectname-projectname appears sloppy to my mind | 22:20 |
| *** jamespage_ has joined #openstack-infra | 22:20 | |
| anteaya | you will note I did +2 that particular patch | 22:20 |
| anteaya | so I don't feel the need to invest a lot of time in this particular issue | 22:21 |
| boris-42 | anteaya: but probably zaqar team in future will need N jobs of rally | 22:21 |
| boris-42 | anteaya: like zaqar-base, zaqar-some_other, zaqar-zaqar | 22:21 |
| anteaya | but my general impression is tht you aren't paying attention to job naming format and the expectations of the consumers of the jobs | 22:21 |
| boris-42 | anteaya: if you would like I can change.. to zaqar-any-other | 22:21 |
| anteaya | see my above comments | 22:21 |
| *** jamespage_ has quit IRC | 22:21 | |
| anteaya | focus for a minute on the part that said +2 | 22:21 |
| boris-42 | anteaya: sure* I am just trying to make people happy | 22:22 |
| boris-42 | anteaya: if you think that it can be better propose your ideas | 22:22 |
| anteaya | paying attention to what I am trying to tell you rather than fighting over every detail would make me happy | 22:22 |
| anteaya | noting that I did just get back from my walk I am going to take a few minutes now and shed a layer of outer clothing, my socks are too hot | 22:23 |
| *** gordc has quit IRC | 22:24 | |
| *** radez is now known as radez_g0n3 | 22:27 | |
| *** kevinbenton has joined #openstack-infra | 22:27 | |
| clarkb | fungi: do you have those publishing site review numbers handy? | 22:27 |
| * clarkb is finally sitting down in a place where work can happen | 22:27 | |
| *** dprince has quit IRC | 22:28 | |
| *** baoli_ has joined #openstack-infra | 22:28 | |
| *** baoli has quit IRC | 22:29 | |
| *** marun has quit IRC | 22:29 | |
| *** packet has quit IRC | 22:30 | |
| clarkb | ianw: I am reviewing those log changes now | 22:31 |
| *** stevemar has quit IRC | 22:32 | |
| *** AlexF has joined #openstack-infra | 22:33 | |
| *** emagana has quit IRC | 22:34 | |
| *** emagana has joined #openstack-infra | 22:35 | |
| *** AlexF has quit IRC | 22:35 | |
| *** esker has quit IRC | 22:38 | |
| *** weshay has quit IRC | 22:38 | |
| *** dizquierdo has joined #openstack-infra | 22:38 | |
| *** ryanpetrello has quit IRC | 22:39 | |
| *** Sincler has quit IRC | 22:39 | |
| *** baoli_ has quit IRC | 22:39 | |
| *** baoli has joined #openstack-infra | 22:39 | |
| *** emagana has quit IRC | 22:39 | |
| ianw | clarkb: thanks | 22:41 |
| *** jistr|mtg has quit IRC | 22:42 | |
| ianw | clarkb: what is aiopcpu short for? asynchronous-I/O physical-cpu? | 22:42 |
| *** marun has joined #openstack-infra | 22:42 | |
| clarkb | ianw: all in one plus cpu | 22:42 |
| clarkb | ianw: basically its an all in one devstack install with a second nova compute node | 22:43 |
| clarkb | so we get a 2 node cloud :) with fake l2 networking via gre | 22:43 |
| anteaya | is that our mulitnode setup? | 22:43 |
| clarkb | ya | 22:43 |
| anteaya | cool | 22:43 |
| *** zul has quit IRC | 22:43 | |
| ianw | clarkb: cool, i got the what from the reviews, but the joining of aio + pcpu was weirding me out :) | 22:45 |
| *** ruhe_ has joined #openstack-infra | 22:45 | |
| clarkb | ya we sort of argued about it initially then decided not to bike shed :) | 22:45 |
| clarkb | its the thing afazekas came up with and we went with it | 22:45 |
| *** ruhe_ has left #openstack-infra | 22:46 | |
| *** rmcall has quit IRC | 22:47 | |
| *** rmcall has joined #openstack-infra | 22:49 | |
| *** teran has joined #openstack-infra | 22:49 | |
| *** marun_ has joined #openstack-infra | 22:49 | |
| *** david-lyle_afk is now known as david-lyle | 22:50 | |
| *** marun has quit IRC | 22:50 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config: Import puppet-openstack-cloud to Stackforge https://review.openstack.org/135729 | 22:51 |
| *** thedodd has quit IRC | 22:51 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: Pull in tempest patch to enable live migration https://review.openstack.org/135768 | 22:53 |
| *** rushiagr_away has quit IRC | 22:54 | |
| *** gyee has joined #openstack-infra | 22:55 | |
| *** emagana has joined #openstack-infra | 22:55 | |
| *** gyee has quit IRC | 22:55 | |
| *** MaxV has quit IRC | 22:56 | |
| *** mriedem has quit IRC | 22:57 | |
| *** shashankhegde has joined #openstack-infra | 22:57 | |
| *** rushiagr_away has joined #openstack-infra | 22:57 | |
| openstackgerrit | Adam Gandelman proposed openstack-infra/elastic-recheck: Omit ironic agent_ssh jobs from 139099 https://review.openstack.org/135770 | 22:59 |
| shashankhegde | Hi | 23:00 |
| shashankhegde | I was wondering if this is the right place for devstack questions | 23:00 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for infra bug 1365046 https://review.openstack.org/135654 | 23:00 |
| uvirtbot | Launchpad bug 1365046 in openstack-ci "Job failed due to no devstack directory" [Undecided,Confirmed] https://launchpad.net/bugs/1365046 | 23:00 |
| clarkb | shashankhegde: typically we point at the #openstack-qa channel | 23:00 |
| clarkb | but asking here is fine too | 23:00 |
| clarkb | we might just ask you to move after if we have no idea | 23:00 |
| shashankhegde | clarkb: Oh.. ok | 23:01 |
| *** jamespage_ has joined #openstack-infra | 23:01 | |
| openstackgerrit | Merged openstack-infra/system-config: Add subunit-worker01 to cacti https://review.openstack.org/135615 | 23:01 |
| shashankhegde | I was running into issues installing oslo.vmware through devstack when I tried to install stable/icehouse | 23:01 |
| clarkb | ah ye sI think that is a known issue | 23:02 |
| clarkb | mtreinish: ^ you had the link to the fix? | 23:02 |
| shashankhegde | Oh | 23:02 |
| *** mwagner_lap has joined #openstack-infra | 23:02 | |
| dtroyer | shashankhegde, clarkb: https://review.openstack.org/#/c/135692/ | 23:02 |
| shashankhegde | I guess the issue is that there is no stable/icehouse branch for oslo.vmware | 23:03 |
| bhunter71 | a question; when gerrit replicates openstack repos to github.com, is it using the git/ssh protocol? If so, which keypair is it using for this? | 23:03 |
| *** r-daneel has quit IRC | 23:03 | |
| *** jamespage_ has quit IRC | 23:03 | |
| clarkb | bhunter71: it is and it uses a keypair for the user gerrit runs as | 23:03 |
| *** otherwiseguy has quit IRC | 23:04 | |
| bhunter71 | so, yeah, I ssh-keygen'd into ~gerrit2/.ssh and registered that with github. sounds like i should be good. | 23:05 |
| shashankhegde | dtroyer: Thanks for the info! | 23:05 |
| clarkb | ya that should do it | 23:05 |
| *** ivar-laz_ has joined #openstack-infra | 23:05 | |
| jeblair | krotscheck: cool; msg lgtm now (btw, i do think we need infra in the subject (as it is currently written) -- otherwise it looks a bit like openstack has moved to storyboard which just might freak some people out) | 23:06 |
| bhunter71 | clarkb, thanks much! | 23:06 |
| *** ivar-laz_ has quit IRC | 23:06 | |
| *** armax has quit IRC | 23:06 | |
| *** ivar-laz_ has joined #openstack-infra | 23:07 | |
| jungleboyj | dims: gate still failed with your change in place: http://logs.openstack.org/21/135721/1/check/check-grenade-dsvm/c72c84c/logs/grenade.sh.txt.gz | 23:07 |
| *** ddieterly has quit IRC | 23:07 | |
| *** ivar-lazzaro has quit IRC | 23:08 | |
| *** mbacchi has quit IRC | 23:09 | |
| *** gondoi is now known as zz_gondoi | 23:10 | |
| *** rlandy has quit IRC | 23:11 | |
| *** markmcclain has quit IRC | 23:11 | |
| *** dkliban is now known as dkliban_afk | 23:12 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config: Import puppet-openstack-cloud to Stackforge https://review.openstack.org/135729 | 23:14 |
| *** amitgandhinz has quit IRC | 23:14 | |
| *** dalgaaf__ has quit IRC | 23:17 | |
| dims | jungleboyj: probably needs the icehouse one to merge first | 23:17 |
| dims | looking | 23:17 |
| *** gyee has joined #openstack-infra | 23:18 | |
| *** dizquierdo has quit IRC | 23:18 | |
| dims | jungleboyj: 135692 is still in the process of merging | 23:19 |
| fungi | clarkb: https://review.openstack.org/132466 and https://review.openstack.org/133302 are what you were requesting | 23:19 |
| jungleboyj | dims: Ok, so that one has to merge first. | 23:20 |
| jeblair | jkt: i can't reproduce that. if i push up a new change for review while on a change list for a project and wait 1 minute, it shows up in the refresh. also if i push a new change and hit ctrl-r it shows up immediately | 23:21 |
| *** pc_m has quit IRC | 23:22 | |
| fungi | ianw: working on seeding puppet-httpd now... just a sec and i should have it in place | 23:22 |
| nibalizer | fungi: youre my hero | 23:22 |
| *** ddieterly has joined #openstack-infra | 23:23 | |
| jeblair | jkt: what project list were you on at the time? | 23:24 |
| *** wenlock has quit IRC | 23:24 | |
| jeblair | jkt: from the logs, it sort of looks like you were looking at the list of changes for project key #1, while the new change was for project key #2 | 23:24 |
| clarkb | fungi: first one lgtm any reason to not approve it? | 23:24 |
| fungi | clarkb: approve at will | 23:25 |
| clarkb | done | 23:25 |
| *** wenlock has joined #openstack-infra | 23:25 | |
| openstackgerrit | Ken Dreyer proposed openstack-infra/jenkins-job-builder: rm argparse from requirements.txt https://review.openstack.org/135775 | 23:26 |
| openstackgerrit | James E. Blair proposed stackforge/gertty: Improve debug logging of sync events https://review.openstack.org/135776 | 23:26 |
| jeblair | jkt: ^ added that for debugging this situation | 23:26 |
| clarkb | jeblair: did you see zuul got borked by new gitpython? | 23:27 |
| jeblair | clarkb: no, i find that delightful though! | 23:27 |
| clarkb | dims pushed up a pin for gitpython probably worth considering that or investing some time to make zuul happy with new gitpython | 23:27 |
| clarkb | it looks like we get exceptions in some cases where a false value was returned previously | 23:27 |
| clarkb | :/ | 23:27 |
| jkt | jeblair: I was definitely on the listing of the same project (the other one has no changes) | 23:27 |
| fungi | nibalizer: ianw: puppet-httpd should be open for business now | 23:30 |
| *** ivar-laz_ has quit IRC | 23:30 | |
| dims | clarkb: all i know was how to make the tests happy, not sure what else will break | 23:30 |
| fungi | nibalizer: ianw: i did not import any extraneous branches or tags, merely set master to the same as the upstream 0.0.4 commit and then force pushed that in | 23:30 |
| jkt | jeblair: ...but of course *now* my new change indeed does appear as soon as I ctrl-r | 23:31 |
| jkt | jeblair: could it be that it was the first time I used gertty when I hit these problems? | 23:31 |
| dims | clarkb: jeblair: i pushed that review so i could get the tests happy for the thing i wanted which was gerrit owner information in status.json (https://review.openstack.org/#/c/135386/) | 23:32 |
| clarkb | dims: ya Ithink it may be something we should just do unitl we can be a bit more sure we do the update right | 23:32 |
| clarkb | but I am also a lazy hacker | 23:32 |
| fungi | clarkb: laziness is one of the three virtues after all | 23:33 |
| dims | clarkb: want me to push a review for pinning the version? :) | 23:33 |
| fungi | (impatience and hubris being the other two) | 23:33 |
| dims | hehe | 23:33 |
| clarkb | dims: ya why don't we start with that | 23:33 |
| dims | ack | 23:34 |
| mtreinish | fungi: but isn't sloth one of the deadly sins :) | 23:34 |
| fungi | mtreinish: not in my religion | 23:34 |
| clarkb | :) | 23:34 |
| jkt | jeblair: hmm, it behaves correctly even after rmeoving the sqlite DB and restarting | 23:34 |
| openstackgerrit | Merged openstack-infra/system-config: Add a governance.openstack.org site for policies https://review.openstack.org/132466 | 23:34 |
| *** marun_ has quit IRC | 23:34 | |
| * clarkb awaits krotscheck's announcement about storyboard | 23:34 | |
| * krotscheck is in expense report hell righ tnow. | 23:34 | |
| ianw | fungi: awesome! welcome to 2012 :) | 23:35 |
| fungi | ianw: 2012 called, they want their apache module back | 23:35 |
| jeblair | dims, clarkb: ++pin until real software development happens :) | 23:35 |
| dims | jeblair: yes sir! | 23:36 |
| jeblair | jkt: ok, i think i'm going to merge those patches now then, and let's just keep an eye out to see if it happens again | 23:36 |
| clarkb | krotscheck: I am so sorry | 23:36 |
| openstackgerrit | Joe Gordon proposed openstack-infra/devstack-gate: DO NOT MERGE: Pull in tempest patch to enable live migration https://review.openstack.org/135768 | 23:37 |
| openstackgerrit | Merged stackforge/gertty: Add ability to sort change list https://review.openstack.org/122383 | 23:37 |
| openstackgerrit | Merged stackforge/gertty: Selectively refresh screen https://review.openstack.org/132324 | 23:37 |
| krotscheck | clarkb: It would have been easier if french people took amex. | 23:37 |
| openstackgerrit | Merged stackforge/gertty: Make owner name in change screen a search https://review.openstack.org/132374 | 23:37 |
| openstackgerrit | Merged stackforge/gertty: Improve debug logging of sync events https://review.openstack.org/135776 | 23:37 |
| fungi | krotscheck: or if your employer provided you with a company credit card whose name didn't start with "american" and then expect you to use it in other countries | 23:37 |
| krotscheck | fungi: Why do you hate ‘Murica! | 23:38 |
| * fungi is not at all surprised most french establishments were disinterested in "american" express | 23:38 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/puppet-httpd: Add .gitreview https://review.openstack.org/135781 | 23:39 |
| *** julim has quit IRC | 23:39 | |
| openstackgerrit | Mikhail S Medvedev proposed openstack-infra/puppet-kibana: Add standard puppet module files and .gitreview https://review.openstack.org/131771 | 23:40 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Pin GitPython version to prevent test failure https://review.openstack.org/135782 | 23:40 |
| jkt | jeblair: could you please take a look at the gertty patches I send during the last few weeks? no hurries, definitely, but I would also hate if they got lost | 23:40 |
| jeblair | jkt: no prob, i'm looking at the gertty review queue now :) | 23:40 |
| jeblair | (i'm just doing the easy ones first) | 23:41 |
| clarkb | fungi: hey they helped us overthrow the oppressive taxation without representation regime | 23:41 |
| *** jungleboyj has quit IRC | 23:42 | |
| clarkb | fungi: in fact there is a large mural of george washington in versaille | 23:42 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Add gerrit owner information to status.json https://review.openstack.org/135386 | 23:42 |
| clarkb | oh wait they didn't like that group of people. nevermind | 23:42 |
| jkt | that's very cool, thanks | 23:42 |
| *** Adri2000 has quit IRC | 23:42 | |
| fungi | clarkb: yeah, they were also none too keen on the english armada | 23:42 |
| fungi | minor squabbles over control of the atlantic | 23:43 |
| fungi | and territories | 23:43 |
| fungi | the usual | 23:43 |
| ianw | fungi: hmm, possible i got something wrong with the setup ... doesn't seem to be running any ci jobs | 23:43 |
| fungi | ianw: looking now | 23:44 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Pin GitPython version to prevent test failure https://review.openstack.org/135782 | 23:44 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Add gerrit owner information to status.json https://review.openstack.org/135386 | 23:44 |
| openstackgerrit | Merged stackforge/gertty: Nicer exit on CTRL-c https://review.openstack.org/133002 | 23:45 |
| fungi | ianw: looks fine to me. i see four pending jobs for 135781,1 in the zuul status | 23:45 |
| clarkb | dims: I think we want to pin to the RC | 23:45 |
| openstackgerrit | Merged stackforge/gertty: type fix in help message https://review.openstack.org/133275 | 23:45 |
| clarkb | the non RC is the new thing iirc | 23:45 |
| dims | 0.3.2.1 is the new thing | 23:45 |
| fungi | ianw: oh, though your first change is actually going to need to get it passing the lint job | 23:45 |
| ianw | fungi: oh, ok, just took a little while to get there | 23:45 |
| clarkb | oh hrm let me check their git log real quick | 23:45 |
| clarkb | because I am paranoid like that | 23:45 |
| ianw | fungi: yep, will examine the logs | 23:45 |
| dims | clarkb: good call | 23:46 |
| clarkb | ya 0.3.2 is from the 14th | 23:46 |
| clarkb | so thats super new too | 23:46 |
| dims | ok switching to RC1 | 23:46 |
| *** sarob has joined #openstack-infra | 23:46 | |
| fungi | clarkb: i was good with that change because it finally got zuul no longer listing a release candidate in its requirements, but maybe it's a question of priorities | 23:46 |
| clarkb | fungi ya but it won't work | 23:47 |
| *** zul has joined #openstack-infra | 23:47 | |
| fungi | oh | 23:47 |
| clarkb | unless is the not .1 fine with zuul? | 23:47 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Pin GitPython version to prevent test failure https://review.openstack.org/135782 | 23:47 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Add gerrit owner information to status.json https://review.openstack.org/135386 | 23:47 |
| dims | done | 23:48 |
| mtreinish | clarkb, fungi: we might need to use bigger nodes for the subunit2sql workers: http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=2387&rra_id=all | 23:48 |
| clarkb | that swap usage | 23:49 |
| *** Adri2000 has joined #openstack-infra | 23:49 | |
| *** shashankhegde has quit IRC | 23:49 | |
| *** Adri2000 has quit IRC | 23:49 | |
| *** Adri2000 has joined #openstack-infra | 23:49 | |
| mtreinish | clarkb: oops, here's mem: http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=2385&rra_id=all | 23:49 |
| mtreinish | it's going through both | 23:49 |
| clarkb | ya | 23:49 |
| clarkb | I am guessing that inflating the subunit stuff is heavy on memory use | 23:50 |
| jeblair | yowza | 23:50 |
| *** mfink has quit IRC | 23:50 | |
| mtreinish | clarkb: I guess, but I don't remember it being too intense when I did local testing though | 23:50 |
| jeblair | there are 4 running, is that intended? | 23:51 |
| jeblair | 3296 logstash 20 0 4567m 2.0g 1520 S 0 26.5 131:48.94 subunit-gearman | 23:51 |
| jeblair | 3332 logstash 20 0 4557m 1.9g 1484 S 1 25.1 128:18.63 subunit-gearman | 23:51 |
| jeblair | 3263 logstash 20 0 4037m 1.7g 1380 S 1 23.1 130:48.40 subunit-gearman | 23:51 |
| jeblair | 3370 logstash 20 0 3929m 1.4g 1328 S 0 18.9 120:36.84 subunit-gearman | 23:51 |
| mtreinish | jeblair: yeah the manifest starts up 4 daemons per worker | 23:51 |
| mtreinish | I guess we could drop that number down too | 23:52 |
| fungi | heh. i can rebuild a bigger one if needed | 23:52 |
| openstackgerrit | Jan Kundrát proposed stackforge/gertty: Add mouse wheel scrolling https://review.openstack.org/133550 | 23:52 |
| jeblair | mtreinish: how about we restart them all so we have more graph data over time? | 23:52 |
| fungi | that sounds like a good first task | 23:52 |
| mtreinish | jeblair: sure sounds good | 23:52 |
| fungi | though they've only been running for about a day right? | 23:53 |
| clarkb | fungi: just under ya | 23:53 |
| mtreinish | fungi: yeah we got them going late yesterday | 23:53 |
| fungi | we could stop all but one of the workers too and profile just a single one i guess | 23:53 |
| jeblair | fungi: sounds good. i'll stop b,c,d and restart a | 23:54 |
| *** esker has joined #openstack-infra | 23:54 | |
| jeblair | done | 23:54 |
| *** andreykurilin_ has quit IRC | 23:55 | |
| *** atiwari has quit IRC | 23:55 | |
| jeblair | jkt, fungi: can you see my comment in 124239 and let me know what you think? | 23:56 |
| *** shashankhegde has joined #openstack-infra | 23:56 | |
| mtreinish | jeblair: will puppet restart the other 3 on the next refresh? | 23:56 |
| clarkb | mtreinish: no | 23:56 |
| mtreinish | clarkb: ok | 23:57 |
| openstackgerrit | Ian Wienand proposed openstack-infra/puppet-httpd: Add rake "lint" target, fix errors https://review.openstack.org/135788 | 23:58 |
| clarkb | mtreinish: we sort of made that intentionally that way | 23:58 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Pin GitPython version to prevent test failure https://review.openstack.org/135782 | 23:59 |
| *** rediskin has left #openstack-infra | 23:59 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/zuul: Add gerrit owner information to status.json https://review.openstack.org/135386 | 23:59 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/system-config: Add mysql-proxy to enable read-only access to a db https://review.openstack.org/122311 | 23:59 |
| fungi | jeblair: responded. i agree for a second reason | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!