| *** ybathia has quit IRC | 00:00 | |
| *** ybathia has joined #openstack-infra | 00:00 | |
| *** ybathia has quit IRC | 00:00 | |
| timrc | Anyway, I think jeblair had a good suggestion to use an orderFile, so that's what I'll do | 00:01 |
|---|---|---|
| *** bswartz has joined #openstack-infra | 00:02 | |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 00:02 |
| jeblair | clarkb, mordred: ^ passes all local tests | 00:03 |
| clarkb | jeblair: woo | 00:03 |
| *** Daisy_ has quit IRC | 00:04 | |
| jeblair | clarkb: note in particular the diff in 19..20 -- your suggestion about setting the value to 'fake_v6' uncovered a bug in the fakes (though the program logic is okay) | 00:04 |
| clarkb | jeblair: neat | 00:05 |
| *** sergmelikyan has joined #openstack-infra | 00:05 | |
| jeblair | anyway, gotta run now; hopefully that's close, if not final. | 00:05 |
| openstackgerrit | Dolph Mathews proposed openstack/requirements: Add pep257 as a requirement https://review.openstack.org/229685 | 00:06 |
| *** ashleighfarnham has quit IRC | 00:06 | |
| *** signed8bit has quit IRC | 00:08 | |
| *** sergmelikyan has quit IRC | 00:10 | |
| openstackgerrit | Dolph Mathews proposed openstack-dev/hacking: Add support for PEP257 https://review.openstack.org/229686 | 00:10 |
| *** odgrim has joined #openstack-infra | 00:11 | |
| *** geoffarn_ has quit IRC | 00:11 | |
| *** geoffarnold has joined #openstack-infra | 00:11 | |
| *** Sukhdev has quit IRC | 00:13 | |
| *** Sukhdev has joined #openstack-infra | 00:13 | |
| *** abitha has quit IRC | 00:15 | |
| *** adrian_otto has joined #openstack-infra | 00:16 | |
| *** amotoki has joined #openstack-infra | 00:17 | |
| *** asettle has joined #openstack-infra | 00:18 | |
| waynr | clarkb: i might not have been clear enough...what i am trying to do is depend on my local experimental copy of JJB as a package, so I need to build a package that has a version I can point to in requirements.txt with an appropriate entry in dependency_links in setup.py as described here: http://peak.telecommunity.com/DevCenter/setuptools#dependencies-that-aren-t-in-pypi | 00:18 |
| *** mriedem1 has joined #openstack-infra | 00:18 | |
| *** odgrim has quit IRC | 00:19 | |
| clarkb | waynr: just change the git path | 00:19 |
| waynr | up until now I have just been using JJB as a command line too installed using "pip install ." from a submodule in the repo that contains my job configs | 00:19 |
| clarkb | waynr: you can pip intsall a fs path too | 00:19 |
| clarkb | waynr: for example pip install /foo/bar/jenkins-job-builder | 00:19 |
| waynr | so the command you suggested will upload a usable package to github for me? | 00:20 |
| adrian_otto | I have a question. We are adding a functional test for magnum to test the new mesos bay type: https://review.openstack.org/225212 which raised the question if we could build and host an image that would accelerate the gate test by reducing the steps from here: | 00:20 |
| adrian_otto | http://git.openstack.org/cgit/openstack/magnum/tree/magnum/templates/heat-mesos/README.md | 00:20 |
| adrian_otto | into a simple download into glance | 00:20 |
| clarkb | waynr: if you need it to not be local you can python setup.py sdist then copy the resulting tar.gz somewhere then install that instead | 00:20 |
| adrian_otto | ideas? | 00:20 |
| waynr | clarkb: I have a new python tool I am working on that is not JJB, i will be running "pip install ." in the top directory of that tool, that command should be able to pull my experimental copy of jjb as a setuptools package | 00:20 |
| clarkb | adrian_otto: projects like ironic python agent and trove both build images that they save on tarballs.openstack.org then the jobs use them | 00:21 |
| waynr | okay, I think sdist is the command I was looking for...the problem with that so far is that the resulting version is not what I was expecting | 00:21 |
| *** pvaneck has quit IRC | 00:21 | |
| clarkb | waynr: so in your requirements.txt point at wherever that JJB version is | 00:21 |
| clarkb | you have many options git/sdist/fs | 00:21 |
| *** mriedem has quit IRC | 00:21 | |
| *** amotoki has quit IRC | 00:22 | |
| smcginnis | anteaya: ping | 00:22 |
| waynr | ie, git describe in the jenkins-job-builder repo is 1.2.0-154-GITSHA | 00:22 |
| adrian_otto | clarkb: sorry you lost me. How do I get an image file onto tarballs.openstack.org? | 00:22 |
| waynr | but the version of the package created with sdist is 0.0.1.dev1314 | 00:22 |
| clarkb | adrian_otto: you would setup a post job in zuul+jenkins to run the image build and copy it to tarballs.openstack.org. Look at ironic-python-agent for examples | 00:23 |
| adrian_otto | thanks! | 00:23 |
| clarkb | adrian_otto: so that whenever your image build details change you build a new image | 00:23 |
| *** sdake has quit IRC | 00:23 | |
| adrian_otto | ok, thanks clarkb | 00:23 |
| openstackgerrit | Dolph Mathews proposed openstack/requirements: Add pep257 as a requirement https://review.openstack.org/229685 | 00:23 |
| waynr | i guess i could just hard-pin to that version anyway and indicate the zip file listed in dependency_links provides that specific version | 00:23 |
| waynr | i have to pin to a specific version either way | 00:24 |
| *** adrian_otto has left #openstack-infra | 00:24 | |
| anteaya | smcginnis: hello there | 00:24 |
| anteaya | smcginnis: how can I help? | 00:24 |
| clarkb | waynr: as an alternative manage the install of jjb independently so that it doens't have to be so hard coded in whatever thing depends on it | 00:24 |
| clarkb | thats less portable but can be handy for dev | 00:24 |
| smcginnis | anteaya: Hey! Sorry to disrupt your evening. Newbie PTL question. | 00:24 |
| smcginnis | anteaya: We have a cinder third party CI account that I think can be reenabled. | 00:25 |
| smcginnis | anteaya: What is the correct procedure for this? | 00:25 |
| waynr | yeah it's handy for dev but so far I am the only person in my org doing dev work and i'd rather be depending on our experimental JJB as a setuptools package than a submodule moving forward | 00:25 |
| anteaya | smcginnis: no worries, my evening is filling out expense reports, please interrupt | 00:25 |
| waynr | dev work on JJB that is | 00:25 |
| smcginnis | anteaya: ;) | 00:25 |
| anteaya | smcginnis: and thank you for asking newbie ptl questions :) | 00:25 |
| anteaya | smcginnis: tell me the name | 00:25 |
| smcginnis | anteaya: Don't worry, I have plenty of them. | 00:25 |
| anteaya | the channel logs will serve as archive | 00:26 |
| smcginnis | It's for the Microsoft iSCSI CI. I don't have the actual account name. | 00:26 |
| smcginnis | anteaya: I can get that | 00:26 |
| anteaya | okey dokey | 00:26 |
| smcginnis | anteaya: But thought I would check first if there is anything else needed. | 00:26 |
| anteaya | smcginnis: good question and no | 00:26 |
| *** tsekiyama has quit IRC | 00:26 | |
| anteaya | you as ptl asking in the channel (directing at me is fine but not required is all that is needed) | 00:27 |
| *** larainema is now known as larainema_holida | 00:27 | |
| anteaya | mailing list works too in a pinch (reply to the disable post on third party announce, please) | 00:27 |
| *** tiswanso has joined #openstack-infra | 00:27 | |
| smcginnis | anteaya: I figured someone else could answer, but I know you always know what's going on. ;) | 00:27 |
| *** enikanorov has joined #openstack-infra | 00:27 | |
| *** Daisy has joined #openstack-infra | 00:27 | |
| anteaya | but irc tends to be the fastest | 00:27 |
| anteaya | I give that impression | 00:27 |
| anteaya | thanks for continuing to buy into it :) | 00:27 |
| smcginnis | heh | 00:28 |
| *** enikanorov_ has quit IRC | 00:28 | |
| *** mtanino has quit IRC | 00:29 | |
| anteaya | if someone with root access can reenable gerrit id 13394 smcginnis and I would be grateful: http://paste.openstack.org/show/474979/ | 00:29 |
| anteaya | nibalizer: ^^ one for you if you want to learn how | 00:29 |
| *** mudassirlatif has joined #openstack-infra | 00:29 | |
| openstackgerrit | Merged openstack/diskimage-builder: Use --nodeps when installing fedora-release https://review.openstack.org/227123 | 00:30 |
| anteaya | clarkb: you still about? | 00:30 |
| nibalizer | anteaya: oh sure | 00:30 |
| nibalizer | do we have docs? | 00:30 |
| anteaya | nibalizer: ah yay | 00:30 |
| anteaya | nibalizer: don't think so | 00:30 |
| anteaya | nibalizer: in the gerrit db you find them by gerrit id | 00:31 |
| anteaya | http://paste.openstack.org/show/474979/ | 00:31 |
| clarkb | anteaya: ya | 00:31 |
| clarkb | anteaya: I am caught up on sb let me do that | 00:31 |
| anteaya | ah clarkb can you walk nibalizer through his first re-enable? | 00:31 |
| clarkb | oh I can do that instead | 00:31 |
| anteaya | thanks | 00:31 |
| nibalizer | yes please | 00:31 |
| anteaya | better for nibalizer | 00:31 |
| smcginnis | anteaya, nibalizer, clarkb: Thanks for your help! | 00:31 |
| clarkb | nibalizer: the docs are gerrit's docs | 00:32 |
| nibalizer | haha | 00:32 |
| *** rlandy has quit IRC | 00:32 | |
| nibalizer | in what way is the account disabled? | 00:32 |
| anteaya | smcginnis: glad you have one to be re-enabled, yay | 00:32 |
| *** geoffarnold has quit IRC | 00:32 | |
| smcginnis | anteaya: Yes, very much goodness IMO. :) | 00:32 |
| clarkb | nibalizer: https://review.openstack.org/Documentation/index.html | 00:32 |
| *** Daisy has quit IRC | 00:32 | |
| anteaya | smcginnis: a good start | 00:32 |
| *** geoffarnold has joined #openstack-infra | 00:32 | |
| smcginnis | anteaya: Hopefully! | 00:33 |
| clarkb | nibalizer: from there you get https://review.openstack.org/Documentation/cmd-index.html and from there https://review.openstack.org/Documentation/cmd-set-account.html | 00:33 |
| anteaya | smcginnis: keep up the newbie ptl questions | 00:33 |
| clarkb | nibalizer: set-account lets you change certain aspects of an account without doing it in the db directly which is unfun | 00:33 |
| smcginnis | anteaya: Oh, don't worry. I'll have plenty. ;) | 00:33 |
| anteaya | smcginnis: well some days will be better than others to be sure, celebrate the small victories when you have them | 00:33 |
| anteaya | smcginnis: great | 00:33 |
| clarkb | nibalizer: so what you need to do is ssh -p 29418 nibz@review.openstack.org gerrit set-account --active 13394 | 00:33 |
| nibalizer | ok | 00:34 |
| clarkb | nibalizer: that says flip the active bit to active in gerrit for account id 13394 | 00:34 |
| clarkb | nibalizer: the inverse is --inactive $accountid | 00:34 |
| anteaya | clarkb: ah cool, don't have to scour the db | 00:34 |
| clarkb | nibalizer: which you would do if the ptl came and said the account was spamming or whatever | 00:34 |
| anteaya | nibalizer: then post to the third-party-announce mailing list so we tell them the are deactivated and why and how to be re-enabled | 00:35 |
| smcginnis | clarkb, nibalizer: Hopefully I don't need to ask for that. | 00:35 |
| smcginnis | much | 00:35 |
| anteaya | or tell pleia2 or myself to do that | 00:35 |
| anteaya | smcginnis: much | 00:35 |
| smcginnis | :) | 00:35 |
| anteaya | smcginnis: it has been a pretty good release | 00:35 |
| anteaya | not much business | 00:35 |
| nibalizer | okay i've activated it | 00:37 |
| nibalizer | cool | 00:37 |
| nibalizer | thanks clarkb and anteaya | 00:37 |
| smcginnis | Thanks everyone. | 00:37 |
| anteaya | nibalizer: thank you for doing it | 00:38 |
| anteaya | found the original disable post: http://lists.openstack.org/pipermail/third-party-announce/2015-July/000251.html | 00:39 |
| anteaya | will reply | 00:39 |
| nibalizer | thanks anteaya | 00:39 |
| anteaya | nibalizer: might be an idea to subscribe to third-party-announce if you haven't already | 00:39 |
| anteaya | yolanda: ^^ too | 00:39 |
| nibalizer | sure | 00:39 |
| *** Rockyg has quit IRC | 00:41 | |
| *** ddieterly has quit IRC | 00:44 | |
| *** ddieterl_ has joined #openstack-infra | 00:44 | |
| *** tjones has quit IRC | 00:44 | |
| *** tiswanso has quit IRC | 00:46 | |
| *** rellerreller has quit IRC | 00:51 | |
| *** mriedem has joined #openstack-infra | 00:51 | |
| *** geoffarnold has quit IRC | 00:53 | |
| *** mriedem1 has quit IRC | 00:53 | |
| *** geoffarnold has joined #openstack-infra | 00:54 | |
| *** shashank_hegde has quit IRC | 00:56 | |
| *** Sukhdev has quit IRC | 00:57 | |
| openstackgerrit | Matthew Treinish proposed openstack/os-testr: Force utf8 encoding on subunit attachments output https://review.openstack.org/229527 | 00:58 |
| *** tiswanso has joined #openstack-infra | 01:00 | |
| *** bharathm has quit IRC | 01:01 | |
| *** julim has quit IRC | 01:01 | |
| *** tiswanso_ has joined #openstack-infra | 01:02 | |
| *** bpokorny has quit IRC | 01:03 | |
| *** Rockyg has joined #openstack-infra | 01:03 | |
| *** enikanorov_ has joined #openstack-infra | 01:04 | |
| *** tiswanso has quit IRC | 01:05 | |
| *** lawrancejing has joined #openstack-infra | 01:06 | |
| *** enikanorov has quit IRC | 01:06 | |
| *** waynr has quit IRC | 01:08 | |
| *** waynr has joined #openstack-infra | 01:08 | |
| *** Swami has quit IRC | 01:10 | |
| *** amotoki has joined #openstack-infra | 01:11 | |
| *** alexsyip has quit IRC | 01:14 | |
| timrc | clarkb, Headed to Seattle next week. Any BBQ recommendations ;-)? | 01:14 |
| *** geoffarnold has quit IRC | 01:15 | |
| *** geoffarnold has joined #openstack-infra | 01:15 | |
| *** amotoki has quit IRC | 01:15 | |
| *** alexsyip has joined #openstack-infra | 01:16 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for juno ebtables kernel fail https://review.openstack.org/229696 | 01:16 |
| *** alexsyip has quit IRC | 01:17 | |
| fungi | timrc: if you like fish bbq, sure ;) | 01:18 |
| *** julim has joined #openstack-infra | 01:20 | |
| timrc | fungi, Hah :) I was sort of trolling there. | 01:20 |
| ruagair | BBQ salmon | 01:20 |
| * ruagair notes it's awfully close to lunch... | 01:21 | |
| *** doug-fish has joined #openstack-infra | 01:21 | |
| timrc | I mean there is smoked salmon... *shudder* | 01:22 |
| fungi | tasty stuff that | 01:23 |
| fungi | all manner of fish can be smoked to great effect | 01:23 |
| anteaya | clarkb: is it still brisket season? | 01:23 |
| * anteaya wonders how long brisket season lasts | 01:23 | |
| jroll | anteaya: every season is brisket season, so long as you're ok with being outside all day in said season | 01:24 |
| jroll | I've done a brisket in the snow :P | 01:24 |
| *** otter768 has joined #openstack-infra | 01:24 | |
| anteaya | jroll: good to know | 01:24 |
| anteaya | jroll: where is home for you? | 01:24 |
| * anteaya wonders who to send to jroll for brisket | 01:25 | |
| jroll | anteaya: then, michigan. currently california | 01:25 |
| anteaya | ah | 01:25 |
| anteaya | more snow in michigan definitely | 01:25 |
| anteaya | where in michigan? | 01:25 |
| anteaya | <-- ontario | 01:25 |
| jroll | anteaya: and given real estate pricing in california, I don't have anywhere for a smoker. wait to send people til when I move back :) | 01:25 |
| jroll | aha | 01:25 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for juno ebtables kernel fail https://review.openstack.org/229696 | 01:25 |
| *** mwagner_ has joined #openstack-infra | 01:25 | |
| anteaya | jroll: are you planning on returning to the land of snow and smoke? | 01:25 |
| jroll | grew up in grand rapids, college in flint, then metro detroit for a year or two | 01:25 |
| anteaya | nice | 01:26 |
| jroll | yes, next year hopefully :) | 01:26 |
| timrc | fungi, It's weird, salmon is one of the only fish I don't care for. I can eat its relative the rainbow trout, but it has to be cooked just right. | 01:26 |
| anteaya | wonderful | 01:26 |
| *** yamamoto has joined #openstack-infra | 01:26 | |
| anteaya | jroll: you can attend pycon.ca next year then | 01:26 |
| anteaya | I hope to have more details after I attend, to give folks a bit more of a heads up | 01:26 |
| jroll | anteaya: yes! | 01:26 |
| anteaya | timrc: interesting | 01:26 |
| timrc | Well actually probably 99.9% of the fish in the ocean I probably not find to be palatable but of the common species sold in stores... :) | 01:26 |
| jroll | <3 pycon, it's been a bit | 01:27 |
| anteaya | my father raised rainbow trout for about 15 years | 01:27 |
| anteaya | jroll: nice, yes I think pycon.ca gave themselves a pass when pyconus was in montreal | 01:27 |
| timrc | anteaya, There are days I want to quit this industry and go do something like that :) | 01:27 |
| jroll | right | 01:27 |
| timrc | I have a few farms for sale bookmarked :) | 01:28 |
| *** bharathm has joined #openstack-infra | 01:28 | |
| anteaya | timrc: well get yourself some oxygen tanks to pump oxygen directly into the water | 01:28 |
| anteaya | we didn't have that | 01:28 |
| anteaya | so the summer of 88 | 01:29 |
| anteaya | humid and hot took all the o2 out of the air and water | 01:29 |
| anteaya | we had pumps to put water into the air and capture o2 when it fell back in | 01:29 |
| anteaya | but you needed o2 in the air for it to work | 01:29 |
| *** armax has quit IRC | 01:29 | |
| anteaya | lost everything including the breeders | 01:29 |
| anteaya | it was a sad time | 01:30 |
| anteaya | so yeah, farms are fun | 01:30 |
| anteaya | but you need the weather to co-operatre | 01:30 |
| anteaya | I've always wanted an orchard | 01:30 |
| *** yamamoto has quit IRC | 01:32 | |
| *** geoffarnold is now known as geoffarnoldX | 01:33 | |
| *** geoffarnold has joined #openstack-infra | 01:36 | |
| *** odgrim has joined #openstack-infra | 01:38 | |
| *** kushal has quit IRC | 01:38 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/project-config: Add a job for testing amqp/proton driver https://review.openstack.org/229701 | 01:39 |
| timrc | anteaya, Yeah I guess the moral of this story is that ever job has its hazards :) | 01:39 |
| *** Rockyg has quit IRC | 01:39 | |
| *** mudassirlatif has quit IRC | 01:40 | |
| anteaya | well we lost a lot of money | 01:40 |
| *** geoffarnoldX has quit IRC | 01:41 | |
| anteaya | and had to figure out how to get rid of a lot of dead fish in a hurry | 01:41 |
| *** mudassirlatif has joined #openstack-infra | 01:41 | |
| *** mudassirlatif has quit IRC | 01:41 | |
| anteaya | but we didn't get hurt | 01:41 |
| anteaya | and he turned back to building houses and digging in the dirt | 01:41 |
| anteaya | which was always his strength anyway | 01:41 |
| anteaya | so I think it is more like try for your dream | 01:42 |
| anteaya | and have the ability to change if that doesn't work | 01:43 |
| *** unicell has quit IRC | 01:43 | |
| timrc | anteaya, I wouldn't even know where to begin with farming at all. I'm an avid fisher person though so I'm at least intrigued. | 01:43 |
| anteaya | is it really a fish farm you want? | 01:43 |
| timrc | Can I have the fish farm for revenue and the private trout pond for angling enjoyment? | 01:44 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/requirements: Bump minimum os-testr version to 0.4.1 https://review.openstack.org/229129 | 01:44 |
| timrc | lol so off topic here, sorry :P | 01:44 |
| *** yamamoto has joined #openstack-infra | 01:44 | |
| anteaya | timrc: okay if you want fish get property where you have the source of the stream on your property | 01:44 |
| anteaya | my father didn't, folks upstream let the beavers damn it up and that slowed the water flow, warming the water | 01:45 |
| anteaya | and warm water doesn't carry as much o2 in it | 01:45 |
| anteaya | so that is the land consideration | 01:45 |
| timrc | So a spring fed stream would be best? | 01:45 |
| anteaya | yes | 01:45 |
| anteaya | and find out from folks around for 50+ years if the spring has ever stopped running | 01:46 |
| *** sridhar_ram has quit IRC | 01:46 | |
| anteaya | like if the water table has ever gotten so low it stopped flowing | 01:46 |
| *** angdraug has quit IRC | 01:46 | |
| anteaya | my dad owned his own heavy equipment and moved the land around to suit him so his ponds were were he wanted | 01:47 |
| anteaya | small fish up top, large pond in the middle for fishing out of, large breeders below that and then a settling pond so the feces would settle out before it went down river | 01:47 |
| anteaya | and a hatchery for the eggs and fry | 01:48 |
| anteaya | picking over the eggs is awesome | 01:48 |
| anteaya | like meditation | 01:48 |
| anteaya | I'll tell you more in tokyo if you want | 01:48 |
| anteaya | can't describe it on irc | 01:48 |
| *** woodster_ has quit IRC | 01:49 | |
| anteaya | and o2 tanks for the breeders | 01:49 |
| anteaya | and be able to change the water flow so if the other fish die the breeders have fresh water | 01:49 |
| anteaya | if you can save the breeders you can recover in a year or two from losing everything else | 01:50 |
| *** yamamoto has quit IRC | 01:53 | |
| *** odgrim has quit IRC | 01:53 | |
| *** julim has quit IRC | 01:58 | |
| openstackgerrit | David Pursehouse proposed openstack-infra/lodgeit: Remove unused pagination lib https://review.openstack.org/229702 | 01:59 |
| *** ybathia has joined #openstack-infra | 01:59 | |
| openstackgerrit | John L. Villalovos proposed openstack-infra/infra-manual: Use 'https://' instead of 'git://' https://review.openstack.org/229703 | 01:59 |
| * ruagair ran dairy goats and alpacas in Tasmania for about 8 years. | 02:00 | |
| timrc | anteaya, I'm definitely interesting in knowing more :) Sounds like a perfect conversation for Tokyo. | 02:00 |
| timrc | er interested* | 02:00 |
| timrc | ruagair, Neat! | 02:00 |
| ruagair | Had a small business making sopa form goat's milk and selling alpaca fleece locally. | 02:01 |
| ruagair | wow, typing not a thing today. | 02:02 |
| *** doug-fish has quit IRC | 02:05 | |
| *** edmondsw has quit IRC | 02:05 | |
| *** amotoki has joined #openstack-infra | 02:05 | |
| *** rfolco has quit IRC | 02:06 | |
| anteaya | timrc: sounds good | 02:07 |
| *** tjones has joined #openstack-infra | 02:09 | |
| *** jerrygb has joined #openstack-infra | 02:09 | |
| *** amotoki has quit IRC | 02:09 | |
| *** tjones has quit IRC | 02:10 | |
| *** thorst has joined #openstack-infra | 02:13 | |
| *** niedbalski has quit IRC | 02:14 | |
| *** markvoelker has quit IRC | 02:15 | |
| *** shashank_hegde has joined #openstack-infra | 02:15 | |
| clarkb | timrc there is a hole in the wall downtown that is decent | 02:16 |
| clarkb | I forget the name let me look it up | 02:16 |
| *** thorst has quit IRC | 02:16 | |
| clarkb | lol its called hole in the wall | 02:17 |
| clarkb | also zaro has said pecos is good but I never want | 02:17 |
| *** lawrancejing has quit IRC | 02:20 | |
| openstackgerrit | David Pursehouse proposed openstack-infra/lodgeit: Correct command to start server in README https://review.openstack.org/229707 | 02:24 |
| *** yamahata has quit IRC | 02:26 | |
| anteaya | clarkb: ha ha ha | 02:26 |
| *** amotoki has joined #openstack-infra | 02:28 | |
| *** dims__ has quit IRC | 02:29 | |
| *** weshay_xchat has quit IRC | 02:29 | |
| *** dims_ has joined #openstack-infra | 02:29 | |
| *** odgrim has joined #openstack-infra | 02:34 | |
| openstackgerrit | Merged openstack-infra/lodgeit: lodgeit.py: Set default URL to paste.openstack.org https://review.openstack.org/216077 | 02:34 |
| openstackgerrit | Merged openstack-infra/lodgeit: Offer download of lodgeit.py with link to plain file on cgit https://review.openstack.org/216078 | 02:34 |
| *** odgrim has quit IRC | 02:34 | |
| *** asettle has quit IRC | 02:35 | |
| *** ybathia has quit IRC | 02:36 | |
| timrc | clarkb, Nice :) I can remember a place called "hole in the wall" | 02:37 |
| *** rguillebert has quit IRC | 02:38 | |
| *** yamamoto has joined #openstack-infra | 02:39 | |
| fungi | docaedo: in hiera now as apps_ssl_cert_file_contents, apps_ssl_chain_file_contents and apps_ssl_key_file_contents | 02:39 |
| *** tjones has joined #openstack-infra | 02:46 | |
| *** amotoki has quit IRC | 02:49 | |
| *** mriedem has quit IRC | 02:51 | |
| *** ddieterl_ has quit IRC | 02:53 | |
| *** dra_ is now known as denisra | 02:54 | |
| *** enikanorov has joined #openstack-infra | 02:57 | |
| *** enikanorov_ has quit IRC | 03:00 | |
| *** kengo_sakai has joined #openstack-infra | 03:01 | |
| *** geoffarnold has quit IRC | 03:01 | |
| *** geoffarnold has joined #openstack-infra | 03:02 | |
| *** amotoki has joined #openstack-infra | 03:02 | |
| *** amotoki_ has joined #openstack-infra | 03:04 | |
| *** dahoo has quit IRC | 03:05 | |
| *** amotoki has quit IRC | 03:08 | |
| *** amotoki_ has quit IRC | 03:10 | |
| *** otter768 has quit IRC | 03:11 | |
| *** otter768 has joined #openstack-infra | 03:12 | |
| *** ociuhandu has joined #openstack-infra | 03:12 | |
| *** dims__ has joined #openstack-infra | 03:15 | |
| *** amotoki has joined #openstack-infra | 03:15 | |
| *** markvoelker has joined #openstack-infra | 03:15 | |
| *** dims___ has joined #openstack-infra | 03:16 | |
| *** tjones has quit IRC | 03:18 | |
| *** dims____ has joined #openstack-infra | 03:18 | |
| *** dims_ has quit IRC | 03:18 | |
| *** lawrancejing has joined #openstack-infra | 03:18 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Turn image logging level up to DEBUG https://review.openstack.org/229718 | 03:20 |
| *** markvoelker has quit IRC | 03:20 | |
| *** dims__ has quit IRC | 03:20 | |
| *** BharatK has joined #openstack-infra | 03:21 | |
| *** dims___ has quit IRC | 03:21 | |
| *** otter768 has quit IRC | 03:21 | |
| *** puranamr has joined #openstack-infra | 03:23 | |
| *** tiswanso_ has quit IRC | 03:24 | |
| *** r-daneel has quit IRC | 03:26 | |
| *** flepied has joined #openstack-infra | 03:31 | |
| *** kengo_sa_ has joined #openstack-infra | 03:32 | |
| *** enikanorov_ has joined #openstack-infra | 03:32 | |
| *** ashleighfarnham has joined #openstack-infra | 03:33 | |
| *** spzala has quit IRC | 03:34 | |
| *** enikanorov has quit IRC | 03:34 | |
| *** kengo_sakai has quit IRC | 03:34 | |
| *** puranamr has quit IRC | 03:36 | |
| *** armax has joined #openstack-infra | 03:36 | |
| *** ayoung-meeting has quit IRC | 03:36 | |
| *** dims____ has quit IRC | 03:37 | |
| *** amotoki has quit IRC | 03:40 | |
| *** salv-orl_ has joined #openstack-infra | 03:43 | |
| *** geoffarn_ has joined #openstack-infra | 03:44 | |
| *** geoffarnold has quit IRC | 03:45 | |
| *** salv-orlando has quit IRC | 03:46 | |
| *** amotoki has joined #openstack-infra | 03:52 | |
| *** links has joined #openstack-infra | 03:55 | |
| *** puranamr has joined #openstack-infra | 03:56 | |
| *** jerrygb has quit IRC | 03:57 | |
| *** jerrygb has joined #openstack-infra | 03:58 | |
| *** boris-42 has quit IRC | 03:59 | |
| *** sdake has joined #openstack-infra | 04:00 | |
| *** geoffarn_ has quit IRC | 04:05 | |
| *** geoffarnold has joined #openstack-infra | 04:06 | |
| *** puranamr has quit IRC | 04:06 | |
| *** harlowja_at_home has joined #openstack-infra | 04:10 | |
| *** lawrancejing has quit IRC | 04:10 | |
| *** markvoelker has joined #openstack-infra | 04:16 | |
| *** hdd has joined #openstack-infra | 04:17 | |
| *** abitha has joined #openstack-infra | 04:19 | |
| *** markvoelker has quit IRC | 04:20 | |
| *** sabeen1 has joined #openstack-infra | 04:21 | |
| *** maishsk has quit IRC | 04:21 | |
| *** Sukhdev has joined #openstack-infra | 04:21 | |
| *** sdake_ has joined #openstack-infra | 04:24 | |
| *** ddieterly has joined #openstack-infra | 04:24 | |
| *** aukhan has joined #openstack-infra | 04:25 | |
| *** geoffarnold has quit IRC | 04:26 | |
| *** geoffarnold has joined #openstack-infra | 04:27 | |
| *** sdake has quit IRC | 04:28 | |
| *** maishsk has joined #openstack-infra | 04:28 | |
| *** ddieterly has quit IRC | 04:29 | |
| *** maishsk has quit IRC | 04:30 | |
| *** flepied has quit IRC | 04:30 | |
| *** stevemar has joined #openstack-infra | 04:30 | |
| *** ashleigh_ has joined #openstack-infra | 04:32 | |
| *** ashleighfarnham has quit IRC | 04:32 | |
| *** StevenK has quit IRC | 04:35 | |
| *** StevenK has joined #openstack-infra | 04:36 | |
| *** dahoo has joined #openstack-infra | 04:38 | |
| *** gildub has quit IRC | 04:41 | |
| *** flepied has joined #openstack-infra | 04:42 | |
| *** unicell has joined #openstack-infra | 04:46 | |
| *** BharatK has quit IRC | 04:48 | |
| *** geoffarnold has quit IRC | 04:48 | |
| *** signed8bit has joined #openstack-infra | 04:48 | |
| *** geoffarnold has joined #openstack-infra | 04:48 | |
| *** dahoo has quit IRC | 04:49 | |
| *** asettle has joined #openstack-infra | 04:50 | |
| *** asettle has quit IRC | 04:50 | |
| *** KanagarajM has joined #openstack-infra | 04:52 | |
| *** Sree has joined #openstack-infra | 04:52 | |
| *** signed8b_ has joined #openstack-infra | 04:54 | |
| *** asettle has joined #openstack-infra | 04:55 | |
| *** gildub has joined #openstack-infra | 04:57 | |
| *** signed8bit has quit IRC | 04:57 | |
| zaro | timrc, clarkb : IMO pecos is much better than hole in the wall | 05:00 |
| *** sabeen1 has quit IRC | 05:00 | |
| timrc | zaro, Good to know.. | 05:02 |
| *** Sree has quit IRC | 05:03 | |
| *** KanagarajM has quit IRC | 05:03 | |
| *** Sree has joined #openstack-infra | 05:03 | |
| timrc | jeblair, clarkb, fungi, jhesketh or anyone else know why the y axis is fractional? http://graphite.openstack.org/render/?width=1228&height=654&_salt=1443675711.364&target=stats.gerrit.event.ref-updated | 05:03 |
| *** bharathm has quit IRC | 05:04 | |
| *** bharathm has joined #openstack-infra | 05:05 | |
| timrc | I'd expect 'statsd.incr('gerrit.event.%s' % event.type)' to increment by 1, for example. | 05:07 |
| *** Sree has quit IRC | 05:07 | |
| *** geoffarnold has quit IRC | 05:09 | |
| *** geoffarnold has joined #openstack-infra | 05:09 | |
| *** KanagarajM has joined #openstack-infra | 05:12 | |
| *** pcrews has quit IRC | 05:16 | |
| *** pcrews has joined #openstack-infra | 05:16 | |
| *** harlowja_at_home has quit IRC | 05:19 | |
| *** BharatK has joined #openstack-infra | 05:19 | |
| *** sdake_ is now known as sdake | 05:21 | |
| *** otter768 has joined #openstack-infra | 05:22 | |
| *** signed8b_ has quit IRC | 05:25 | |
| *** signed8bit has joined #openstack-infra | 05:25 | |
| *** ddieterly has joined #openstack-infra | 05:25 | |
| *** otter768 has quit IRC | 05:28 | |
| *** flepied has quit IRC | 05:28 | |
| *** hdd has quit IRC | 05:29 | |
| *** ddieterly has quit IRC | 05:30 | |
| *** geoffarnold has quit IRC | 05:30 | |
| *** geoffarnold has joined #openstack-infra | 05:31 | |
| *** geoffarnold is now known as geoffarnoldX | 05:33 | |
| *** geoffarnoldX is now known as geoffarnold | 05:33 | |
| *** geoffarnold is now known as geoffarnoldX | 05:34 | |
| *** kushal has joined #openstack-infra | 05:34 | |
| *** oomichi has quit IRC | 05:35 | |
| *** nadya has joined #openstack-infra | 05:36 | |
| *** KanagarajM has quit IRC | 05:38 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 05:41 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Skip caching for dnf (Fedora 22) builds https://review.openstack.org/229728 | 05:43 |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Add diskimage-builder lint job https://review.openstack.org/229729 | 05:43 |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Update dib elements to standard headers https://review.openstack.org/229730 | 05:43 |
| ianw | ^ clarkb / fungi : f22 jobs are just looping around in failure ATM, see above | 05:44 |
| ianw | sorry i must have missed this -- i need to move my test env somewhere that cloning every openstack repo & downloading all the disk-images is practcial | 05:45 |
| *** flepied has joined #openstack-infra | 05:49 | |
| *** geoffarnoldX has quit IRC | 05:52 | |
| *** geoffarnold has joined #openstack-infra | 05:52 | |
| *** gildub has quit IRC | 05:52 | |
| *** gildub has joined #openstack-infra | 05:53 | |
| *** kiran-r has joined #openstack-infra | 05:54 | |
| *** Sukhdev has quit IRC | 05:56 | |
| *** Sukhdev has joined #openstack-infra | 05:56 | |
| *** nadya has quit IRC | 05:57 | |
| ruagair | ouch ianw :-/ | 05:58 |
| *** oomichi has joined #openstack-infra | 05:59 | |
| *** Sukhdev has quit IRC | 05:59 | |
| *** oomichi has quit IRC | 06:03 | |
| *** odgrim has joined #openstack-infra | 06:04 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack-infra/project-config: Normalize projects.yaml https://review.openstack.org/229735 | 06:05 |
| *** k4n0 has joined #openstack-infra | 06:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements: Updated from generate-constraints https://review.openstack.org/226816 | 06:08 |
| ruagair | Does any know if Storyboard has gone through any schema changes since April? | 06:08 |
| ruagair | (the database schema) | 06:08 |
| ruagair | In particular dropping the "username" column from the storyboard.users table? | 06:09 |
| *** flepied has quit IRC | 06:10 | |
| *** geoffarnold has quit IRC | 06:13 | |
| *** geoffarnold has joined #openstack-infra | 06:13 | |
| *** AJaeger has joined #openstack-infra | 06:14 | |
| *** odgrim has quit IRC | 06:15 | |
| *** sergmelikyan has joined #openstack-infra | 06:17 | |
| *** markvoelker has joined #openstack-infra | 06:17 | |
| *** scheuran has joined #openstack-infra | 06:18 | |
| *** AJaeger has quit IRC | 06:21 | |
| *** markvoelker has quit IRC | 06:21 | |
| *** AJaeger has joined #openstack-infra | 06:22 | |
| *** shardy has joined #openstack-infra | 06:25 | |
| openstackgerrit | Merged openstack-infra/project-config: Add non-voting job to test Cinder with Block Device Driver backend https://review.openstack.org/228857 | 06:25 |
| *** tobiash has quit IRC | 06:26 | |
| *** ddieterly has joined #openstack-infra | 06:26 | |
| *** ddieterly has quit IRC | 06:30 | |
| *** tobiash has joined #openstack-infra | 06:33 | |
| *** abitha has quit IRC | 06:35 | |
| *** ParsectiX has joined #openstack-infra | 06:36 | |
| openstackgerrit | azher ullah khan proposed openstack-infra/system-config: Setting ansible playbooks path variable https://review.openstack.org/229744 | 06:37 |
| *** armax has quit IRC | 06:39 | |
| *** kiran-r has quit IRC | 06:39 | |
| *** matbu_ has joined #openstack-infra | 06:40 | |
| *** nadya has joined #openstack-infra | 06:40 | |
| *** shashank_hegde has quit IRC | 06:42 | |
| *** sergmelikyan has quit IRC | 06:42 | |
| *** matbu has quit IRC | 06:42 | |
| *** asmigala has joined #openstack-infra | 06:46 | |
| *** peristeri has joined #openstack-infra | 06:47 | |
| *** ihrachys has joined #openstack-infra | 06:48 | |
| *** ihrachys has quit IRC | 06:49 | |
| openstackgerrit | Merged openstack-infra/project-config: Normalize projects.yaml https://review.openstack.org/229735 | 06:50 |
| *** ajmiller_ has joined #openstack-infra | 06:51 | |
| *** rvasilets has joined #openstack-infra | 06:53 | |
| *** ajmiller has quit IRC | 06:53 | |
| *** vponomaryov has joined #openstack-infra | 06:54 | |
| *** aarefiev has joined #openstack-infra | 06:54 | |
| *** HeOS has quit IRC | 06:54 | |
| *** olaph has quit IRC | 06:55 | |
| *** svasheka has joined #openstack-infra | 06:55 | |
| *** geoffarnold has quit IRC | 06:55 | |
| *** geoffarnold has joined #openstack-infra | 06:55 | |
| *** andreykurilin has joined #openstack-infra | 06:57 | |
| openstackgerrit | Merged openstack-infra/project-config: Add job to test Add Cinder API with Apache https://review.openstack.org/208498 | 06:58 |
| *** ParsectiX has quit IRC | 06:58 | |
| *** devvesa has joined #openstack-infra | 06:58 | |
| *** claudiub has joined #openstack-infra | 06:59 | |
| *** mwagone_ has joined #openstack-infra | 07:00 | |
| *** otter768 has joined #openstack-infra | 07:00 | |
| *** mwagone has quit IRC | 07:03 | |
| *** kiran-r has joined #openstack-infra | 07:04 | |
| *** ParsectiX has joined #openstack-infra | 07:04 | |
| *** kiran-r has quit IRC | 07:05 | |
| *** kiranr has joined #openstack-infra | 07:05 | |
| *** kiranr is now known as kiran-r | 07:06 | |
| *** boris-42 has joined #openstack-infra | 07:07 | |
| *** wolsen has quit IRC | 07:08 | |
| *** sergmelikyan has joined #openstack-infra | 07:09 | |
| *** rossella_s has joined #openstack-infra | 07:11 | |
| *** arxcruz has joined #openstack-infra | 07:12 | |
| *** geoffarnold has quit IRC | 07:16 | |
| *** ihrachys has joined #openstack-infra | 07:16 | |
| *** geoffarnold has joined #openstack-infra | 07:17 | |
| *** bharathm has quit IRC | 07:17 | |
| openstackgerrit | Merged openstack-infra/project-config: Create puppet-barbican repository https://review.openstack.org/229643 | 07:17 |
| *** ihrachys has quit IRC | 07:17 | |
| openstackgerrit | Merged openstack-infra/project-config: Add puppet-openstack_spec_helper project https://review.openstack.org/229126 | 07:18 |
| openstackgerrit | Merged openstack-infra/project-config: Added new Openstack repository for PureStorage Cinder Fuel plugin https://review.openstack.org/226662 | 07:18 |
| *** aeng has quit IRC | 07:22 | |
| *** notnownikki has joined #openstack-infra | 07:22 | |
| *** matrohon has joined #openstack-infra | 07:24 | |
| openstackgerrit | Merged openstack-infra/system-config: Fix project URL for Nodepool https://review.openstack.org/229671 | 07:25 |
| *** ifarkas has joined #openstack-infra | 07:25 | |
| *** camunoz has quit IRC | 07:26 | |
| *** ddieterly has joined #openstack-infra | 07:27 | |
| openstackgerrit | Merged openstack-infra/system-config: Third party CI ops should subscribe to -infra list https://review.openstack.org/223233 | 07:27 |
| frickler | yolanda: how about upgrading your +1 on https://review.openstack.org/190526 to a +A now that you have received the power? ;) | 07:29 |
| yolanda | frickler, sounds like a good idea :) | 07:29 |
| *** asettle has quit IRC | 07:31 | |
| *** ddieterly has quit IRC | 07:31 | |
| *** claudiub has quit IRC | 07:32 | |
| peristeri | anteaya, Hi, I'm interested in helping with Gerrit. Is there any way I could help? (I used to develop in Java) | 07:35 |
| frickler | yolanda: thx, I was hoping for a Workflow+1 though, since the review has had a number of +2s for some time now. ;;) | 07:36 |
| yolanda | frickler, for nodepool i prefer to do it when some other cores are present | 07:37 |
| yolanda | i wanted to land some other change as well, so i will look in a few hours | 07:37 |
| frickler | peristeri: best talk to zaro directly, though I gather he is on US pacific time, so better retry a bit later | 07:37 |
| frickler | yolanda: o.k., understood, thx anyway | 07:38 |
| yolanda | anytime | 07:38 |
| peristeri | frickler, cool, I'll send zaro an email. | 07:39 |
| *** jtomasek has joined #openstack-infra | 07:39 | |
| *** dingyichen has quit IRC | 07:40 | |
| openstackgerrit | Jesse Pretorius proposed openstack-infra/project-config: Add pep8 and bashate non-voting checks to openstack-ansible https://review.openstack.org/228406 | 07:40 |
| *** jlanoux has joined #openstack-infra | 07:42 | |
| *** markus_z has joined #openstack-infra | 07:43 | |
| *** dingyichen has joined #openstack-infra | 07:43 | |
| *** dchen has joined #openstack-infra | 07:43 | |
| *** sergmelikyan has quit IRC | 07:44 | |
| nibalizer | yolanda: o/ | 07:46 |
| nibalizer | yolanda: did you test your ssh access? | 07:46 |
| nibalizer | do you need reviews on anything? | 07:46 |
| *** openstackgerrit has quit IRC | 07:46 | |
| *** zeih has quit IRC | 07:46 | |
| *** zeih has joined #openstack-infra | 07:46 | |
| *** openstackgerrit has joined #openstack-infra | 07:46 | |
| *** e0ne has joined #openstack-infra | 07:47 | |
| openstackgerrit | Merged openstack-infra/puppet-logstash: Create a class to install and run curator https://review.openstack.org/222741 | 07:47 |
| yolanda | hi nibalizer, yes, i have access to all instances | 07:47 |
| yolanda | i tested a few | 07:47 |
| yolanda | nibalizer, if you can do some reviews for puppet modules it will help. Specially this one https://review.openstack.org/#/c/200289/ is blocked | 07:48 |
| *** e0ne has quit IRC | 07:49 | |
| *** mgrosz has joined #openstack-infra | 07:50 | |
| *** Ala has joined #openstack-infra | 07:51 | |
| nibalizer | ok i will do that | 07:51 |
| mgrosz | Hello, I need help...: I have this first push/review to Vitrage project https://review.openstack.org/#/c/229748/. The files that are added are based on the new Project introduction here http://docs.openstack.org/infra/manual/creators.html, using the new project template described here "Choosing the Right cookiecutter Template". | 07:53 |
| *** sergmelikyan has joined #openstack-infra | 07:53 | |
| *** HeOS has joined #openstack-infra | 07:53 | |
| mgrosz | Now I pushed the changes, but it fails in Jenkins..... What am I missing? | 07:53 |
| *** e0ne has joined #openstack-infra | 07:54 | |
| *** e0ne has quit IRC | 07:56 | |
| AJaeger | mgrosz: you need to adjust your change so that it passes the tests. Have a look at the log files of the failures | 07:56 |
| AJaeger | Or run locally to reproduce | 07:56 |
| *** rlandy has joined #openstack-infra | 07:56 | |
| *** thingee has joined #openstack-infra | 07:57 | |
| *** sergmelikyan has quit IRC | 07:58 | |
| openstackgerrit | Merged openstack-infra/puppet-openstackci: Make jenkins password a required parameter https://review.openstack.org/200289 | 07:59 |
| *** fhubik has joined #openstack-infra | 08:00 | |
| *** olaph has joined #openstack-infra | 08:06 | |
| *** Sree has joined #openstack-infra | 08:08 | |
| *** jordanP has joined #openstack-infra | 08:08 | |
| *** derekh has joined #openstack-infra | 08:08 | |
| *** sergmelikyan has joined #openstack-infra | 08:10 | |
| *** fhubik is now known as fhubik_brb | 08:10 | |
| *** Sree has quit IRC | 08:11 | |
| *** Sree has joined #openstack-infra | 08:12 | |
| *** Sree has quit IRC | 08:16 | |
| *** dizquierdo has joined #openstack-infra | 08:16 | |
| *** alexpilotti has joined #openstack-infra | 08:17 | |
| *** dguitarbite has quit IRC | 08:18 | |
| *** bharathm has joined #openstack-infra | 08:18 | |
| *** markvoelker has joined #openstack-infra | 08:18 | |
| *** jistr has joined #openstack-infra | 08:18 | |
| openstackgerrit | Augustina Ragwitz proposed openstack-infra/project-config: (WIP) split out into testable units https://review.openstack.org/229772 | 08:19 |
| *** sflanigan has quit IRC | 08:19 | |
| *** ihrachys has joined #openstack-infra | 08:20 | |
| openstackgerrit | Jesse Pretorius proposed openstack-infra/project-config: Add pep8 and bashate non-voting checks to openstack-ansible https://review.openstack.org/228406 | 08:20 |
| *** markvoelker has quit IRC | 08:22 | |
| *** bharathm has quit IRC | 08:23 | |
| *** mgrosz has quit IRC | 08:24 | |
| *** mgrosz has joined #openstack-infra | 08:24 | |
| *** jlanoux has quit IRC | 08:24 | |
| *** e0ne has joined #openstack-infra | 08:26 | |
| *** ddieterly has joined #openstack-infra | 08:28 | |
| *** ihrachys has quit IRC | 08:30 | |
| *** thingee has quit IRC | 08:31 | |
| *** dguitarbite has joined #openstack-infra | 08:32 | |
| *** ddieterly has quit IRC | 08:32 | |
| *** claudiub has joined #openstack-infra | 08:34 | |
| *** e0ne has quit IRC | 08:36 | |
| *** fhubik_brb is now known as fhubik | 08:36 | |
| *** e0ne has joined #openstack-infra | 08:38 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use tripleo.sh to build packages https://review.openstack.org/229425 | 08:38 |
| *** dchen has quit IRC | 08:40 | |
| *** geoffarnold has quit IRC | 08:40 | |
| *** geoffarnold has joined #openstack-infra | 08:41 | |
| *** mgrosz has quit IRC | 08:43 | |
| *** rguillebert has joined #openstack-infra | 08:44 | |
| *** sergmelikyan has quit IRC | 08:45 | |
| *** salv-orlando has joined #openstack-infra | 08:46 | |
| *** salv-or__ has joined #openstack-infra | 08:46 | |
| *** mgrosz has joined #openstack-infra | 08:47 | |
| *** salv-orl_ has quit IRC | 08:49 | |
| *** oomichi has joined #openstack-infra | 08:50 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack-infra/project-config: Add devstack-plugin-bdd to PROJECTS for tempest-dsvm-full-bdd job https://review.openstack.org/229775 | 08:50 |
| *** salv-orlando has quit IRC | 08:50 | |
| *** BharatK has quit IRC | 08:51 | |
| *** jlanoux has joined #openstack-infra | 08:51 | |
| *** dingyichen has quit IRC | 08:51 | |
| *** mgrosz has quit IRC | 08:53 | |
| *** jlanoux has quit IRC | 08:58 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use tripleo.sh to build packages https://review.openstack.org/229425 | 08:58 |
| *** ihrachys has joined #openstack-infra | 08:59 | |
| *** jlanoux has joined #openstack-infra | 08:59 | |
| ianw | AJaeger: not sure what you mean with https://review.openstack.org/#/c/229729/1/tox.ini ? | 09:01 |
| *** geoffarnold has quit IRC | 09:02 | |
| AJaeger | ianw: let me double check... | 09:02 |
| *** sergmelikyan has joined #openstack-infra | 09:02 | |
| *** geoffarnold has joined #openstack-infra | 09:02 | |
| AJaeger | ianw: Johannes confused me I fear, checking a bit more... | 09:03 |
| AJaeger | ianw: .tox/dib/bin/dib-lint uses flake8 but you're not installing flake8 in the tox environment | 09:04 |
| *** BharatK has joined #openstack-infra | 09:04 | |
| ianw | AJaeger: hmm, yeah ok, maybe it only gets run in the context of tox within dib | 09:04 |
| AJaeger | ianw: I have flake8 installed systemwide, so don't get a failure. | 09:07 |
| ianw | AJaeger: yeah, me too :) | 09:08 |
| AJaeger | ianw: See Johannes comment where it failed | 09:08 |
| AJaeger | or temporarily move flake8 | 09:08 |
| AJaeger | so, my suggestion was to add "flake8" to the new dib tox environment | 09:09 |
| AJaeger | or add "-r test-requirements.txt" | 09:09 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add flake8 to requirements https://review.openstack.org/229780 | 09:10 |
| ianw | AJaeger: ^ so that's the real solution, i'll propose a work-around since that will take some weeks to get merged & into a release | 09:10 |
| *** deepakcs has joined #openstack-infra | 09:11 | |
| AJaeger | ianw: ok, thanks | 09:11 |
| AJaeger | if I should change my -1 back to a +2, leave a comment so that others understand it. | 09:12 |
| *** ParsectiX has quit IRC | 09:12 | |
| *** shardy_ has joined #openstack-infra | 09:13 | |
| *** alexpilotti has quit IRC | 09:13 | |
| *** sambetts has joined #openstack-infra | 09:14 | |
| *** shardy has quit IRC | 09:14 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Update dib elements to standard headers https://review.openstack.org/229730 | 09:18 |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Add diskimage-builder lint job https://review.openstack.org/229729 | 09:18 |
| *** shardy_ has quit IRC | 09:18 | |
| ianw | AJaeger: ^ thanks! good catch there | 09:18 |
| *** shardy has joined #openstack-infra | 09:19 | |
| *** sergmelikyan has quit IRC | 09:21 | |
| AJaeger | ianw: LGTM. Did you see my comment for doing this lint only for nodepool changes? | 09:23 |
| *** geoffarn_ has joined #openstack-infra | 09:24 | |
| *** signed8bit_ZZZzz has quit IRC | 09:24 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Test a update in trunk repository version https://review.openstack.org/229789 | 09:24 |
| *** ParsectiX has joined #openstack-infra | 09:26 | |
| *** salv-orlando has joined #openstack-infra | 09:26 | |
| *** sergmelikyan has joined #openstack-infra | 09:27 | |
| *** geoffarnold has quit IRC | 09:28 | |
| *** ddieterly has joined #openstack-infra | 09:29 | |
| *** salv-or__ has quit IRC | 09:30 | |
| *** kiran-r has quit IRC | 09:30 | |
| *** kiran-r has joined #openstack-infra | 09:32 | |
| *** marzif has joined #openstack-infra | 09:32 | |
| *** ddieterly has quit IRC | 09:33 | |
| *** jtomasek has quit IRC | 09:35 | |
| openstackgerrit | Mateusz Matuszkowiak proposed openstack-infra/project-config: Added new Openstack repository for Fuel Mistral plugin integration https://review.openstack.org/225596 | 09:36 |
| *** sdake_ has joined #openstack-infra | 09:38 | |
| *** Sree has joined #openstack-infra | 09:38 | |
| *** pelix has joined #openstack-infra | 09:41 | |
| *** sdake has quit IRC | 09:42 | |
| *** salv-orl_ has joined #openstack-infra | 09:42 | |
| *** Sree has quit IRC | 09:44 | |
| *** salv-orlando has quit IRC | 09:45 | |
| openstackgerrit | Ivan Kolodyazhny proposed openstack-infra/project-config: Fix cinder-dsvm-apache job https://review.openstack.org/229800 | 09:48 |
| *** boris-42 has quit IRC | 09:49 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/111011 | 09:51 |
| *** jtomasek has joined #openstack-infra | 09:51 | |
| openstackgerrit | Mateusz Matuszkowiak proposed openstack-infra/project-config: Added new Openstack repository for ScaleIO Cinder Fuel plugin https://review.openstack.org/229801 | 09:51 |
| *** e0ne has quit IRC | 09:53 | |
| *** mgrosz has joined #openstack-infra | 09:53 | |
| *** mgrosz has quit IRC | 09:54 | |
| *** dims has joined #openstack-infra | 09:55 | |
| *** Kiall has quit IRC | 09:55 | |
| *** Kiall has joined #openstack-infra | 09:56 | |
| *** e0ne has joined #openstack-infra | 09:56 | |
| *** jtomasek has quit IRC | 09:57 | |
| *** alexpilotti has joined #openstack-infra | 09:58 | |
| *** dizquierdo has quit IRC | 10:00 | |
| e0ne | AJaeger: thank you for the quick feedback and reviews! I appreciate your help | 10:01 |
| *** dizquierdo has joined #openstack-infra | 10:01 | |
| *** mgrosz has joined #openstack-infra | 10:02 | |
| *** akscram has quit IRC | 10:03 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/project-config: Add a job for testing amqp/proton driver https://review.openstack.org/229701 | 10:05 |
| *** bharathm has joined #openstack-infra | 10:07 | |
| *** mgrosz has quit IRC | 10:07 | |
| *** stevemar has quit IRC | 10:08 | |
| *** stevemar has joined #openstack-infra | 10:09 | |
| *** bharathm has quit IRC | 10:11 | |
| *** amotoki_ has joined #openstack-infra | 10:11 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/puppet-gerrit: Allow optional configuration of jeepyb https://review.openstack.org/229809 | 10:11 |
| *** ZZelle has quit IRC | 10:12 | |
| *** ZZelle has joined #openstack-infra | 10:13 | |
| *** stevemar has quit IRC | 10:13 | |
| *** markvoelker has joined #openstack-infra | 10:19 | |
| *** amotoki_ has quit IRC | 10:19 | |
| *** ajmiller__ has joined #openstack-infra | 10:20 | |
| *** kiran-r has quit IRC | 10:21 | |
| *** fhubik is now known as fhubik_brb | 10:22 | |
| *** hashar has joined #openstack-infra | 10:23 | |
| *** markvoelker has quit IRC | 10:23 | |
| *** ajmiller_ has quit IRC | 10:25 | |
| *** ajmiller has joined #openstack-infra | 10:26 | |
| *** Hal has quit IRC | 10:26 | |
| *** mwagner_ has quit IRC | 10:26 | |
| *** geoffarnold has joined #openstack-infra | 10:26 | |
| *** geoffarn_ has quit IRC | 10:26 | |
| *** fhubik_brb is now known as fhubik | 10:27 | |
| *** ajmiller__ has quit IRC | 10:28 | |
| *** ddieterly has joined #openstack-infra | 10:29 | |
| *** ddieterly has quit IRC | 10:34 | |
| *** alexpilotti has quit IRC | 10:35 | |
| lennyb | Hi , what is a new home of subunit2html.py ? | 10:35 |
| *** alexpilotti has joined #openstack-infra | 10:36 | |
| *** asselin_ has joined #openstack-infra | 10:37 | |
| *** ajmiller has quit IRC | 10:38 | |
| *** asselin__ has quit IRC | 10:40 | |
| *** BharatK has quit IRC | 10:42 | |
| *** ajmiller has joined #openstack-infra | 10:43 | |
| *** amotoki_ has joined #openstack-infra | 10:46 | |
| *** nikil89_ has joined #openstack-infra | 10:46 | |
| *** geoffarnold has quit IRC | 10:47 | |
| *** geoffarnold has joined #openstack-infra | 10:47 | |
| *** pc_m has joined #openstack-infra | 10:50 | |
| *** abregman has joined #openstack-infra | 10:54 | |
| *** BharatK has joined #openstack-infra | 10:55 | |
| *** abregman is now known as abregman|pto | 10:56 | |
| AJaeger | lennyb: repo openstack/os-testr, file os_testr/subunit2html.py | 10:57 |
| *** trown|outttypeww is now known as trown | 11:00 | |
| *** marzif has quit IRC | 11:04 | |
| openstackgerrit | Andreas Jaeger proposed openstack/requirements: Rename stackforge/octavia -> openstack/octavia https://review.openstack.org/229822 | 11:04 |
| *** marzif has joined #openstack-infra | 11:05 | |
| *** aukhan has quit IRC | 11:07 | |
| *** bharathm has joined #openstack-infra | 11:08 | |
| *** geoffarnold has quit IRC | 11:08 | |
| *** geoffarnold has joined #openstack-infra | 11:09 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Test a update in trunk repository version https://review.openstack.org/229789 | 11:12 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Remove sudo when calling tripleo.sh https://review.openstack.org/229823 | 11:12 |
| *** bharathm has quit IRC | 11:13 | |
| *** marzif has quit IRC | 11:14 | |
| *** marzif has joined #openstack-infra | 11:15 | |
| *** marzif has quit IRC | 11:16 | |
| *** marzif has joined #openstack-infra | 11:16 | |
| *** arxcruz has quit IRC | 11:18 | |
| *** Daisy has joined #openstack-infra | 11:19 | |
| *** eantyshev has joined #openstack-infra | 11:19 | |
| *** ildikov has joined #openstack-infra | 11:25 | |
| *** marzif has quit IRC | 11:26 | |
| *** kgiusti has quit IRC | 11:26 | |
| *** doug-fish has joined #openstack-infra | 11:26 | |
| *** marzif has joined #openstack-infra | 11:26 | |
| *** geoffarnold has quit IRC | 11:30 | |
| *** geoffarnold has joined #openstack-infra | 11:30 | |
| *** ddieterly has joined #openstack-infra | 11:30 | |
| *** kengo_sa_ has quit IRC | 11:30 | |
| *** baoli has joined #openstack-infra | 11:31 | |
| *** markvoelker has joined #openstack-infra | 11:34 | |
| *** ddieterly has quit IRC | 11:34 | |
| lennyb | AJaeger: thanks | 11:35 |
| *** arxcruz has joined #openstack-infra | 11:35 | |
| *** marzif has quit IRC | 11:36 | |
| *** marzif has joined #openstack-infra | 11:37 | |
| *** ldnunes has joined #openstack-infra | 11:37 | |
| *** markvoelker has quit IRC | 11:39 | |
| *** AJaeger has quit IRC | 11:45 | |
| *** AJaeger has joined #openstack-infra | 11:46 | |
| *** yamamoto has quit IRC | 11:46 | |
| *** AJaeger has quit IRC | 11:46 | |
| *** marzif has quit IRC | 11:46 | |
| *** dprince has joined #openstack-infra | 11:47 | |
| *** dtantsur|afk is now known as dtantsur | 11:47 | |
| *** Daisy has quit IRC | 11:47 | |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/project-config: rpmlint checks for rpm-packages repo https://review.openstack.org/229840 | 11:47 |
| *** Daisy has joined #openstack-infra | 11:48 | |
| *** AJaeger has joined #openstack-infra | 11:48 | |
| *** marzif has joined #openstack-infra | 11:49 | |
| *** geoffarnold has quit IRC | 11:51 | |
| *** geoffarnold has joined #openstack-infra | 11:51 | |
| *** jtomasek has joined #openstack-infra | 11:52 | |
| *** baoli has quit IRC | 11:52 | |
| *** amotoki has quit IRC | 11:52 | |
| *** amotoki_ is now known as amotoki | 11:52 | |
| *** Daisy has quit IRC | 11:52 | |
| *** cdent has joined #openstack-infra | 11:54 | |
| *** flepied has joined #openstack-infra | 11:59 | |
| *** akscram has joined #openstack-infra | 12:01 | |
| *** markvoelker has joined #openstack-infra | 12:01 | |
| *** gordc has joined #openstack-infra | 12:01 | |
| *** fhubik is now known as fhubik_brb | 12:03 | |
| *** nikil89_ is now known as naggappan | 12:03 | |
| *** aarefiev has quit IRC | 12:04 | |
| *** marcusvrn_ has joined #openstack-infra | 12:06 | |
| *** marzif has quit IRC | 12:06 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/111011 | 12:07 |
| *** marzif has joined #openstack-infra | 12:07 | |
| *** dizquierdo has quit IRC | 12:09 | |
| *** kiran-r has joined #openstack-infra | 12:09 | |
| *** denisra has quit IRC | 12:10 | |
| *** pblaho has joined #openstack-infra | 12:11 | |
| *** moravec has quit IRC | 12:11 | |
| *** jtomasek has quit IRC | 12:11 | |
| *** aarefiev has joined #openstack-infra | 12:12 | |
| *** geoffarnold has quit IRC | 12:12 | |
| *** geoffarnold has joined #openstack-infra | 12:13 | |
| *** Hal has joined #openstack-infra | 12:14 | |
| *** k4n0 has quit IRC | 12:14 | |
| *** marzif has quit IRC | 12:17 | |
| *** jerrygb has quit IRC | 12:17 | |
| *** marzif has joined #openstack-infra | 12:17 | |
| *** jtomasek has joined #openstack-infra | 12:18 | |
| *** kiran-r has quit IRC | 12:21 | |
| *** regXboi has joined #openstack-infra | 12:21 | |
| *** k4n0 has joined #openstack-infra | 12:22 | |
| *** k4n0 has quit IRC | 12:22 | |
| *** mwagner_ has joined #openstack-infra | 12:23 | |
| *** ociuhandu has quit IRC | 12:24 | |
| *** weshay_xchat has joined #openstack-infra | 12:25 | |
| *** deepakcs has quit IRC | 12:25 | |
| *** annegentle has joined #openstack-infra | 12:25 | |
| *** dzamboni has joined #openstack-infra | 12:27 | |
| *** weshay_xchat is now known as weshay | 12:27 | |
| *** pradk has joined #openstack-infra | 12:28 | |
| *** baoli has joined #openstack-infra | 12:28 | |
| *** glauco_ has joined #openstack-infra | 12:31 | |
| *** ddieterly has joined #openstack-infra | 12:31 | |
| *** spzala has joined #openstack-infra | 12:31 | |
| *** Daisy has joined #openstack-infra | 12:31 | |
| *** geoffarnold has quit IRC | 12:34 | |
| *** amuller has joined #openstack-infra | 12:34 | |
| *** geoffarn_ has joined #openstack-infra | 12:34 | |
| *** ddieterly has quit IRC | 12:35 | |
| *** BharatK has quit IRC | 12:35 | |
| *** tlian has quit IRC | 12:36 | |
| *** yamamoto has joined #openstack-infra | 12:38 | |
| *** yamamoto has quit IRC | 12:38 | |
| *** yamamoto has joined #openstack-infra | 12:38 | |
| *** stevemar has joined #openstack-infra | 12:41 | |
| *** salv-orl_ has quit IRC | 12:42 | |
| *** edmondsw has joined #openstack-infra | 12:42 | |
| openstackgerrit | Merged openstack-infra/project-config: Skip caching for dnf (Fedora 22) builds https://review.openstack.org/229728 | 12:42 |
| *** stevemar has quit IRC | 12:46 | |
| *** glauco__ has joined #openstack-infra | 12:46 | |
| *** glauco__ has quit IRC | 12:46 | |
| *** naggappan is now known as nikil89_ | 12:47 | |
| *** stevemar has joined #openstack-infra | 12:47 | |
| *** dkehn has quit IRC | 12:47 | |
| *** nikil89_ is now known as naggappan | 12:47 | |
| *** tlian has joined #openstack-infra | 12:47 | |
| *** puranamr has joined #openstack-infra | 12:48 | |
| *** dkehn has joined #openstack-infra | 12:48 | |
| *** BharatK has joined #openstack-infra | 12:49 | |
| *** glauco_ has quit IRC | 12:49 | |
| *** fhubik_brb is now known as fhubik | 12:49 | |
| *** pfallenop has quit IRC | 12:49 | |
| *** pfallenop has joined #openstack-infra | 12:50 | |
| *** glauco_ has joined #openstack-infra | 12:50 | |
| openstackgerrit | Merged openstack-infra/project-config: Add diskimage-builder lint job https://review.openstack.org/229729 | 12:51 |
| *** glauco_ has quit IRC | 12:51 | |
| *** stevemar has quit IRC | 12:52 | |
| *** julim has joined #openstack-infra | 12:52 | |
| *** puranamr has quit IRC | 12:52 | |
| openstackgerrit | Merged openstack-infra/puppet-gerrit: Setup preconditions for acceptance tests. https://review.openstack.org/219400 | 12:53 |
| *** pfallenop has quit IRC | 12:53 | |
| *** pfallenop has joined #openstack-infra | 12:54 | |
| *** geoffarn_ has quit IRC | 12:55 | |
| *** glauco_ has joined #openstack-infra | 12:55 | |
| *** glauco_ has quit IRC | 12:55 | |
| *** geoffarnold has joined #openstack-infra | 12:56 | |
| *** bharathm has joined #openstack-infra | 12:57 | |
| *** glauco has joined #openstack-infra | 12:57 | |
| *** glauco has quit IRC | 12:57 | |
| *** jaypipes has joined #openstack-infra | 12:58 | |
| *** thorst has joined #openstack-infra | 12:58 | |
| *** thorst has quit IRC | 12:59 | |
| *** glauco has joined #openstack-infra | 12:59 | |
| *** glauco has quit IRC | 12:59 | |
| *** bharathm has quit IRC | 13:01 | |
| *** glauco has joined #openstack-infra | 13:01 | |
| *** glauco has quit IRC | 13:02 | |
| *** glauco has joined #openstack-infra | 13:02 | |
| *** nadya has quit IRC | 13:03 | |
| *** glauco has quit IRC | 13:05 | |
| *** glauco_ has joined #openstack-infra | 13:07 | |
| *** moravec has joined #openstack-infra | 13:10 | |
| *** ddieterly has joined #openstack-infra | 13:11 | |
| *** glauco_ has quit IRC | 13:12 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/229868 | 13:14 |
| *** ddieterly has quit IRC | 13:16 | |
| *** rfolco has joined #openstack-infra | 13:16 | |
| *** geoffarnold has quit IRC | 13:16 | |
| *** geoffarn_ has joined #openstack-infra | 13:17 | |
| *** Zhongjun has joined #openstack-infra | 13:18 | |
| *** marzif has quit IRC | 13:20 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 13:20 |
| *** marzif has joined #openstack-infra | 13:20 | |
| *** glauco has joined #openstack-infra | 13:21 | |
| *** mriedem has joined #openstack-infra | 13:23 | |
| yolanda | jeblair, fungi, when you come online, i'd like to land the secure.conf change https://review.openstack.org/#/c/189762/ | 13:25 |
| *** marzif has quit IRC | 13:25 | |
| *** sabeen1 has joined #openstack-infra | 13:25 | |
| yolanda | but it means a restart of nodepool, are we ok to do it? or how do we land these kind of changes? | 13:25 |
| *** ayoung has joined #openstack-infra | 13:25 | |
| *** signed8bit has joined #openstack-infra | 13:26 | |
| *** denisra has joined #openstack-infra | 13:27 | |
| *** bltavares has joined #openstack-infra | 13:27 | |
| *** dramalho has joined #openstack-infra | 13:28 | |
| *** tiswanso has joined #openstack-infra | 13:29 | |
| fungi | yolanda: well, it simply doesn't take effect until nodepool is restarted, right? | 13:31 |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/project-config: rpmlint checks for rpm-packages repo https://review.openstack.org/229840 | 13:31 |
| *** alivigni has joined #openstack-infra | 13:31 | |
| fungi | looks like our last nodepool restart was 6 days ago (september 25) | 13:32 |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/project-config: rpmlint checks for rpm-packages repo https://review.openstack.org/229840 | 13:32 |
| yolanda | fungi yes, but if the code goes there, and there is no restart, and then you need to restart it...and this change fails, is like having a retarded bomb there | 13:32 |
| yolanda | i normally like to land changes and test at the same moment | 13:32 |
| fungi | there have been no new changes landed to the nodepool repo in that time, so if we merge that one we can be relatively assured that restarting it will be testing only that change | 13:33 |
| yolanda | how do you proceed with that? | 13:33 |
| *** geoffarn_ is now known as geoffarnold | 13:33 | |
| yolanda | fungi, do you have some scheduled restarts for nodepool? | 13:33 |
| fungi | generally once the change is showing up on nodepool.openstack.org in /opt/nodepool's git head, then stop and start the nodepool init script (i think graceful restarts may still not be working quite right?) | 13:33 |
| yolanda | fungi, but a nodepool restart causes all node in "building" state to be lost. In downstream it normally is a problem, and we try to do it at times with no load | 13:34 |
| yolanda | how do you handle it? | 13:34 |
| fungi | looks like we're not really behind on demand at the moment, so this is probably a safe time to do it | 13:34 |
| *** trown is now known as trown|brb | 13:35 | |
| yolanda | fungi, so am i ok to do it? i also normally take a backup of the old nodepool code, to safely replace it if something goes wrong | 13:35 |
| fungi | that is to say, there is no appreciable zuul job queue backlog and we're not riding our providers' quotas | 13:35 |
| *** denisra has quit IRC | 13:35 | |
| fungi | yolanda: the old nodepool code is HEAD^1 in this case, so you can simply switch back by disabling puppet agent and doing a little git manipulation and another pip install from it | 13:36 |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/project-config: rpmlint checks for rpm-packages repo https://review.openstack.org/229840 | 13:36 |
| *** kgiusti has joined #openstack-infra | 13:36 | |
| *** Hal has quit IRC | 13:36 | |
| *** bcourt has joined #openstack-infra | 13:37 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for novaclient legacy bdm bug 1501435 https://review.openstack.org/229882 | 13:37 |
| openstack | bug 1501435 in python-novaclient "osc 1.7 no longer can boot a server from volume" [Critical,In progress] https://launchpad.net/bugs/1501435 - Assigned to John Garbutt (johngarbutt) | 13:37 |
| *** ajmiller has quit IRC | 13:38 | |
| *** dkranz has joined #openstack-infra | 13:38 | |
| *** ajmiller has joined #openstack-infra | 13:38 | |
| *** geoffarnold has quit IRC | 13:39 | |
| yolanda | ok, going to approve it, and will baby-sit | 13:39 |
| fungi | thanks! | 13:39 |
| *** fhubik is now known as fhubik_brb | 13:39 | |
| *** denisra has joined #openstack-infra | 13:40 | |
| *** xyang1 has joined #openstack-infra | 13:41 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for cinder v2 heat bug 1500780 https://review.openstack.org/229093 | 13:42 |
| openstack | bug 1500780 in tempest "orchestration.stacks.test_volumes test is failing" [High,Fix released] https://launchpad.net/bugs/1500780 - Assigned to Jordan Pittier (jordan-pittier) | 13:42 |
| *** Sree has joined #openstack-infra | 13:42 | |
| *** fhubik_brb is now known as fhubik | 13:42 | |
| openstackgerrit | Michal Vanco proposed openstack-infra/jenkins-job-builder: Support test trend graph in Build Flow Test Aggregator https://review.openstack.org/223942 | 13:43 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for gnocchi functional setup fail https://review.openstack.org/229889 | 13:43 |
| *** ddieterly has joined #openstack-infra | 13:44 | |
| *** abregman|pto is now known as abregman | 13:45 | |
| *** denisra has quit IRC | 13:46 | |
| *** ddieterly has quit IRC | 13:47 | |
| *** ddieterly has joined #openstack-infra | 13:47 | |
| *** trown|brb is now known as trown | 13:47 | |
| *** woodster_ has joined #openstack-infra | 13:48 | |
| *** ddieterly has quit IRC | 13:49 | |
| *** mrmartin has joined #openstack-infra | 13:49 | |
| *** denisra has joined #openstack-infra | 13:49 | |
| *** dkehn has quit IRC | 13:51 | |
| *** nelsnelson has quit IRC | 13:52 | |
| openstackgerrit | Merged openstack-infra/grafyaml: Add test coverate for grafana.py https://review.openstack.org/226986 | 13:52 |
| *** armax has joined #openstack-infra | 13:52 | |
| *** armax has quit IRC | 13:52 | |
| yolanda | fungi, so secure.conf failed , i need a rebase | 13:53 |
| yolanda | so it will need reviews again | 13:53 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/229868 | 13:53 |
| *** mattt has left #openstack-infra | 13:54 | |
| fungi | i'm happy to give teh rebase a quick look, but generally we're not so rigorous with getting lots of reviews on a rebase of something which already got approved | 13:54 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/229868 | 13:54 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/nodepool: Update nodepool to use secure.conf file https://review.openstack.org/189762 | 13:54 |
| *** stevemar has joined #openstack-infra | 13:54 | |
| fungi | in the past i've often just reapproved something after rebasing it, depending on the complexity of the patch and the merge conflict | 13:55 |
| yolanda | so asselin fixed, it's a problem on some new tests added | 13:55 |
| *** armax has joined #openstack-infra | 13:55 | |
| fungi | cool. looking now | 13:55 |
| asselin_ | yolanda, actually I just rebased it via gerrit | 13:55 |
| *** ddieterly has joined #openstack-infra | 13:56 | |
| asselin_ | yolanda, fungi I will fail again..sorry...just waking up | 13:56 |
| yolanda | asselin_, it should fail.. if there are new tests, the secure.conf parameter needs to be added | 13:56 |
| fungi | yeah, it doesn't so much need a rebase as some modifications i guess | 13:57 |
| *** links has quit IRC | 13:57 | |
| clarkb | why doesnt the test base class handle that? | 13:57 |
| dansmith | paste.openstack is always extremely slow for me.. I assume I'm not the only one, right? | 13:58 |
| *** dkehn has joined #openstack-infra | 13:58 | |
| *** ParsectiX has quit IRC | 13:58 | |
| fungi | dansmith: it gets blasted by crawlers reindexing it over and over from links in our html-based irc channel logs i think | 13:58 |
| fungi | looking to see if that robots.txt addition made it in yet | 13:58 |
| dansmith | fungi: ah | 13:59 |
| *** stevemar has quit IRC | 13:59 | |
| Clint | fungi: it's definitely not serving up robots.txt | 13:59 |
| dansmith | fungi: it's slow enough that I just don't use it, but people still send me stuff with it and it's frustrating :/ | 13:59 |
| yolanda | so clarkb, i added this new parameter to don't be so disruptive on the nodepool tests, as they were managing the configfile setup. But i think you are right, as secure file will not change | 13:59 |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Add validate command https://review.openstack.org/226554 | 13:59 |
| yolanda | is not the same case as the nodepool yaml file, where you test different cases | 13:59 |
| yolanda | asselin_, what do you think about adding the self.setup_secure() ad useNodepool level? | 14:00 |
| clarkb | pstch argb should just set it | 14:00 |
| clarkb | no reason to be explicit I dont think | 14:00 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for novaclient legacy bdm bug 1501435 https://review.openstack.org/229882 | 14:01 |
| openstack | bug 1501435 in python-openstackclient "osc 1.7 no longer can boot a server from volume" [Critical,New] https://launchpad.net/bugs/1501435 | 14:01 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for gnocchi functional setup fail https://review.openstack.org/229889 | 14:01 |
| clarkb | same for useNodepool | 14:01 |
| asselin_ | yolanda, in this change or a separate change? | 14:02 |
| yolanda | in this change | 14:02 |
| yolanda | refactor that calls to secure.conf on each test, that is causing us pain | 14:02 |
| fungi | yeah, looks like paste.o.o is beset by more bots posting search engine keyword bait | 14:02 |
| fungi | i'll try to squash them with more temporary iptables rules | 14:02 |
| asselin_ | yolanda, ok will look into that | 14:03 |
| *** Ala has quit IRC | 14:03 | |
| yolanda | thx asselin | 14:03 |
| *** eharney has joined #openstack-infra | 14:03 | |
| *** bcourt has quit IRC | 14:04 | |
| asselin_ | yolanda, I'm going to have breakfast and wakeup first | 14:04 |
| yolanda | asselin_ sure :) | 14:04 |
| fungi | also "yahoo! slurp[tm]" is indexing paste.o.o now | 14:04 |
| pabelanger | morning | 14:05 |
| *** rossella_s has quit IRC | 14:06 | |
| *** rossella_s has joined #openstack-infra | 14:06 | |
| clarkb | yolanda asselin you also need to remove dburi from setup_config() | 14:06 |
| clarkb | images dir as well apparebtly? | 14:07 |
| clarkb | why is images dir in secure.conf | 14:07 |
| *** dims has quit IRC | 14:10 | |
| *** david-lyle has quit IRC | 14:11 | |
| *** stevemar has joined #openstack-infra | 14:11 | |
| *** AJaeger has quit IRC | 14:12 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:12 | |
| fungi | dansmith: see if it's responding any faster for you now. i firewalled off a couple of offenders | 14:12 |
| openstackgerrit | Armando Migliaccio proposed openstack-infra/project-config: Skip lbaasv2-dsvm-minimal on doc changes https://review.openstack.org/229680 | 14:12 |
| dansmith | fungi: definitely | 14:13 |
| fungi | yeah, i think it's the combination of search engines and the spambots taking advantage of search engine indexing causing the slowdowns | 14:13 |
| *** david-lyle has joined #openstack-infra | 14:13 | |
| fungi | so in theory the robots.txt addition should stop the indexing, and after a while the spambots will hopefully realize it's no longer an attractive target again | 14:14 |
| Clint | that's a reasonable theory | 14:14 |
| *** signed8bit is now known as signed8bit_ZZZzz | 14:14 | |
| *** flip214 has quit IRC | 14:15 | |
| yolanda | clarkb so dburi can be removed yes, as it's legacy. About images path, is there on previous patches, not related to this secure.conf patchset, i cannot tell | 14:15 |
| dansmith | fungi: yeah, I'm sure it's hard to hide from the ones that don't follow the rules anyway | 14:15 |
| dansmith | fungi: seems like there should be some rate-limiting built in to throttle offenders | 14:15 |
| clarkb | yolanda its in setuo config where it belongs but its also in the secure.conf where it doesnt belong | 14:16 |
| clarkb | *in setup secure.conf | 14:16 |
| fungi | dansmith: unfortunately our fork of lodgeit was (at last check) the only one with anything close to actively maintained | 14:16 |
| dansmith | heh | 14:16 |
| fungi | so if it's going to get new features, it's probably up to us to write them | 14:17 |
| *** signed8bit_ZZZzz is now known as signed8bit | 14:17 | |
| *** dizquierdo has joined #openstack-infra | 14:17 | |
| *** jsavak has joined #openstack-infra | 14:17 | |
| yolanda | ok so it needs to be removed you are right, seems that this test part needs a bit of work | 14:17 |
| mrmartin | morning | 14:17 |
| *** gildub has quit IRC | 14:18 | |
| fungi | Clint: the reason for my theory is that i just blacklisted one ip address posting things like http://paste.openstack.org/show/475064/ and one search engine indexing the server | 14:18 |
| fungi | (and this is far from the first time seeing it) | 14:18 |
| Clint | multilingual | 14:19 |
| *** Hal has joined #openstack-infra | 14:20 | |
| *** bcourt has joined #openstack-infra | 14:21 | |
| eantyshev | Hello! This zuul change https://review.openstack.org/207094 has received the first +2, and I welcome more reviewers | 14:21 |
| *** jsavak has quit IRC | 14:21 | |
| *** eantyshev has left #openstack-infra | 14:22 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/elastic-recheck: Filter the uncategorized page on only tempest and grenade https://review.openstack.org/229909 | 14:22 |
| *** jsavak has joined #openstack-infra | 14:22 | |
| *** nelsnelson has joined #openstack-infra | 14:23 | |
| *** fhubik is now known as fhubik_brb | 14:23 | |
| fungi | looks like greghaynes's robots.txt support for puppet-lodgeit is https://review.openstack.org/228610 | 14:24 |
| *** fhubik_brb is now known as fhubik | 14:24 | |
| *** fhubik has quit IRC | 14:24 | |
| fungi | er, actually that's in lodgeit not puppet-lodgeit | 14:25 |
| yolanda | ah, i was wondering why i didn't see it :) | 14:25 |
| *** arxcruz has quit IRC | 14:25 | |
| clarkb | is there a reason apache should host that? | 14:25 |
| openstackgerrit | Merged openstack-infra/grafyaml: Add docs for grafana-dashboards command https://review.openstack.org/228601 | 14:25 |
| clarkb | maybe greghaynes will know when he wakes | 14:26 |
| Clint | fungi: isn't that gonna render as text/html? | 14:26 |
| fungi | seems like we could pretty easily just alias it in the vhost config, yeah | 14:26 |
| fungi | Clint: quite possibly? | 14:26 |
| * Clint asks in a better place. | 14:26 | |
| *** mtanino has joined #openstack-infra | 14:27 | |
| clarkb | fungi yes same way etherpad does it | 14:27 |
| fungi | render_to_response() does look like it might have that effect, yes | 14:27 |
| fungi | clarkb: greghaynes: i agree the apache solution is simpler and we basically already have it implemented elsewhere | 14:28 |
| *** zz_dimtruck is now known as dimtruck | 14:29 | |
| *** vmb_out is now known as vmbrasseur | 14:29 | |
| *** jsavak has quit IRC | 14:31 | |
| *** Ala has joined #openstack-infra | 14:31 | |
| anteaya | peristeri: hello and welcome | 14:32 |
| anteaya | peristeri: yes you can help, thank you for offering | 14:32 |
| peristeri | anteaya, Great, I sent an email to Remo to get an update of what is the current status of the bugs. | 14:33 |
| anteaya | remo? | 14:33 |
| anteaya | do you mean zaro? | 14:33 |
| anteaya | peristeri: here is the most recent status: https://etherpad.openstack.org/p/I0eV4IgkZS | 14:34 |
| peristeri | anteaya, yes zaro. | 14:35 |
| *** sridhar_ram has joined #openstack-infra | 14:36 | |
| anteaya | wonderful | 14:36 |
| *** bltavares has quit IRC | 14:36 | |
| *** nelsnelson has quit IRC | 14:36 | |
| anteaya | he'll be online in a few hours | 14:36 |
| anteaya | peristeri: in the mean time he and I put together the above etherpad | 14:36 |
| *** nelsnelson has joined #openstack-infra | 14:36 | |
| anteaya | with links to the issues and offered fixes that have stalled | 14:36 |
| peristeri | anteaya, Great, I will look into it. | 14:37 |
| anteaya | peristeri: let me know if you need more explaination | 14:37 |
| anteaya | peristeri: thank you! | 14:37 |
| peristeri | anteaya, sure. | 14:37 |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Reorder tox.ini envlist https://review.openstack.org/224865 | 14:39 |
| *** jsavak has joined #openstack-infra | 14:39 | |
| *** nelsnelson has quit IRC | 14:40 | |
| pabelanger | stepping away for 45mins, running some errands with family | 14:40 |
| mordred | Shrews, clarkb: https://review.openstack.org/#/q/status:open+project:openstack-infra/shade+branch:master+topic:private-test,n,z | 14:41 |
| mordred | should be ready for review | 14:41 |
| mordred | and I think we should land all three before landing the nodepool patch | 14:41 |
| *** stevemar has quit IRC | 14:41 | |
| *** stevemar has joined #openstack-infra | 14:42 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 14:42 |
| mordred | (just depends-on the nodepool on those patches) | 14:42 |
| pabelanger | yolanda: timrc: https://review.openstack.org/#/c/226554/ adds the validate command to grafyaml | 14:43 |
| anteaya | pabelanger: enjoy errands | 14:43 |
| pabelanger | anteaya: doctors. only so much fun | 14:44 |
| anteaya | pabelanger: ah I hope everyone is in good health and they say, go home | 14:44 |
| anteaya | :) | 14:44 |
| *** Daisy has quit IRC | 14:44 | |
| *** Daisy has joined #openstack-infra | 14:45 | |
| *** bltavares has joined #openstack-infra | 14:45 | |
| *** peristeri has quit IRC | 14:45 | |
| *** bharathm has joined #openstack-infra | 14:45 | |
| *** stevemar has quit IRC | 14:46 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/elastic-recheck: Filter the uncategorized page on only tempest and grenade https://review.openstack.org/229909 | 14:46 |
| *** stevemar has joined #openstack-infra | 14:46 | |
| *** tjones has joined #openstack-infra | 14:47 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Tell os-client-config that we prefer IPv6 https://review.openstack.org/229923 | 14:48 |
| mrmartin | fungi: I've fixed the puppet-community's git resource module, so the refresh on ref commits works well there. I'm just waiting an approval on the pull request: https://github.com/puppet-community/puppet-git_resource/pulls I like to add it to modules.env, and I prepared a changeset required for askbot's infra. | 14:48 |
| *** rmoats has joined #openstack-infra | 14:49 | |
| *** rmoats has quit IRC | 14:49 | |
| *** stevemar_ has joined #openstack-infra | 14:49 | |
| *** jtomasek has quit IRC | 14:50 | |
| *** bharathm has quit IRC | 14:50 | |
| *** stevemar has quit IRC | 14:51 | |
| *** Sree has quit IRC | 14:53 | |
| timrc | pabelanger, I'll take a look in a few. I have sick kids with me today. | 14:53 |
| *** Sree has joined #openstack-infra | 14:53 | |
| fungi | mrmartin: sounds great! | 14:54 |
| *** stevemar has joined #openstack-infra | 14:55 | |
| fungi | nibalizer: crinkle: ^ (not sure if you're able to help get that pr reviewed) | 14:55 |
| *** tonytan4ever has joined #openstack-infra | 14:56 | |
| *** geoffarnold has joined #openstack-infra | 14:56 | |
| anteaya | timrc: :( may the hot soup and repetitive video be yours | 14:57 |
| *** Sree has quit IRC | 14:57 | |
| *** stevemar_ has quit IRC | 14:58 | |
| *** tsekiyama has joined #openstack-infra | 14:58 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 14:58 | |
| *** jlanoux has quit IRC | 14:58 | |
| *** jaypipes has quit IRC | 14:59 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 15:00 | |
| openstackgerrit | Chris Dent proposed openstack-infra/project-config: Enable ceilometer devstack plugin for tempest jobs https://review.openstack.org/229940 | 15:00 |
| *** hashar has quit IRC | 15:00 | |
| *** Swami has joined #openstack-infra | 15:01 | |
| *** jsavak has quit IRC | 15:02 | |
| *** bltavares has quit IRC | 15:02 | |
| *** bltavare_ has joined #openstack-infra | 15:02 | |
| EmilienM | pabelanger: I think you can safely abandon https://review.openstack.org/220527 | 15:02 |
| *** bcourt has quit IRC | 15:04 | |
| *** jsavak has joined #openstack-infra | 15:05 | |
| *** wolsen has joined #openstack-infra | 15:05 | |
| *** sdake has joined #openstack-infra | 15:06 | |
| *** bpokorny has joined #openstack-infra | 15:08 | |
| openstackgerrit | Chris Dent proposed openstack-infra/project-config: Enable ceilometer devstack plugin for tempest jobs https://review.openstack.org/229940 | 15:08 |
| openstackgerrit | Dmitry Teselkin proposed openstack-infra/project-config: [WIP] rpmlint checks for rpm-packages repo https://review.openstack.org/229840 | 15:09 |
| *** jsavak has quit IRC | 15:09 | |
| *** nadya has joined #openstack-infra | 15:10 | |
| openstackgerrit | Doug Hellmann proposed openstack-dev/pbr: Restore warnerrors behavior https://review.openstack.org/229951 | 15:10 |
| *** sdake_ has quit IRC | 15:10 | |
| *** jsavak has joined #openstack-infra | 15:12 | |
| openstackgerrit | Matt Riedemann proposed openstack/requirements: Raise upper-constraints on novaclient to 2.30.1 https://review.openstack.org/229954 | 15:13 |
| *** mrmartin has quit IRC | 15:13 | |
| mriedem | dhellmann: sdague: ttx: ^ | 15:13 |
| sdague | mriedem: +2 | 15:13 |
| mriedem | wrong branch | 15:13 |
| sdague | gdi :) | 15:14 |
| *** tonytan4ever has quit IRC | 15:14 | |
| mriedem | now it's up | 15:15 |
| *** abregman is now known as abregman|afk | 15:15 | |
| mriedem | https://review.openstack.org/#/c/229956/ | 15:15 |
| anteaya | I guess our bot doesn't catch stable branches on requirements | 15:16 |
| anteaya | should it> | 15:16 |
| anteaya | > | 15:16 |
| anteaya | ? | 15:16 |
| wznoinsk | fungi: re the git mirroring software, I'm interested in what you use to keep local git copies of OS git repos - I'm thinking of having local git mirror with OS code here so asking people that already may have an experience | 15:16 |
| openstackgerrit | Merged openstack-infra/project-config: Add release jobs to the octavia project https://review.openstack.org/229518 | 15:16 |
| *** jsavak has quit IRC | 15:16 | |
| wznoinsk | fungi: GROK's area I'd say | 15:17 |
| fungi | wznoinsk: i use a script like this in a nightly cron job http://paste.openstack.org/show/475085/ | 15:17 |
| anteaya | mestery: can you rebase 229564 when you get a chance? | 15:17 |
| mestery | anteaya: On it | 15:18 |
| fungi | wznoinsk: i also have a separate script which blows away all non-remote-tracking branches and resets all the tracking branches to the origin state, which i run manually on occasion as cleanup | 15:18 |
| anteaya | mestery: thank you | 15:18 |
| *** jsavak has joined #openstack-infra | 15:18 | |
| *** jerrygb has joined #openstack-infra | 15:19 | |
| *** jerrygb has quit IRC | 15:19 | |
| openstackgerrit | Merged openstack-infra/project-config: Add devstack-plugin-bdd to PROJECTS for tempest-dsvm-full-bdd job https://review.openstack.org/229775 | 15:19 |
| *** jerrygb has joined #openstack-infra | 15:19 | |
| *** abregman|afk has quit IRC | 15:20 | |
| openstackgerrit | Kyle Mestery proposed openstack-infra/project-config: Add pypi publishing jobs for neutron sub-projects https://review.openstack.org/229564 | 15:21 |
| mestery | anteaya: ^^^^^ | 15:21 |
| anteaya | thank you | 15:21 |
| *** nadya has quit IRC | 15:21 | |
| *** bcourt has joined #openstack-infra | 15:21 | |
| *** shardy_ has joined #openstack-infra | 15:22 | |
| *** nadya has joined #openstack-infra | 15:22 | |
| *** jsavak has quit IRC | 15:23 | |
| *** jeadre has quit IRC | 15:24 | |
| *** shardy has quit IRC | 15:24 | |
| anteaya | mestery: you are waiting on the pypi creation for two projects, if I might suggest, remove those two from your current patch and add them as a child patch | 15:24 |
| *** jsavak has joined #openstack-infra | 15:24 | |
| anteaya | mestery: that way those who have complied can get your attention and the stragglers can wait | 15:25 |
| *** jeadre has joined #openstack-infra | 15:25 | |
| mestery | anteaya: That's not a bad idea to be honest, I'll take a look at that. | 15:25 |
| *** ajmiller_ has joined #openstack-infra | 15:25 | |
| *** samuelBartel has joined #openstack-infra | 15:26 | |
| anteaya | mestery: it keeps you moving, rather than waiting for the one who maybe asleep at the switch | 15:26 |
| Shrews | mordred: by any chance did you test 229140 against our ansible modules? | 15:26 |
| *** tonytan4ever has joined #openstack-infra | 15:27 | |
| *** scheuran has quit IRC | 15:27 | |
| *** ajmiller has quit IRC | 15:27 | |
| *** shardy_ has quit IRC | 15:27 | |
| TheJulia | win 19 | 15:27 |
| TheJulia | doh | 15:28 |
| *** ajmiller has joined #openstack-infra | 15:28 | |
| *** shardy has joined #openstack-infra | 15:28 | |
| *** jsavak has quit IRC | 15:29 | |
| mriedem | sdague: is an upper-constraints change on master needed immediately or can we just wait for the bot? | 15:30 |
| mriedem | for novaclient 2.31 | 15:30 |
| *** jsavak has joined #openstack-infra | 15:30 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 15:30 | |
| *** jaypipes has joined #openstack-infra | 15:30 | |
| *** dims has joined #openstack-infra | 15:30 | |
| sdague | we should push it | 15:30 |
| mriedem | ok | 15:30 |
| sdague | because I think there is going to be a timing issue otherwise | 15:30 |
| mriedem | push it real good? | 15:30 |
| sdague | :P | 15:30 |
| * fungi cringes at terrible salt-n-pepa reference | 15:31 | |
| *** bltavare_ has quit IRC | 15:32 | |
| *** ajmiller_ has quit IRC | 15:32 | |
| openstackgerrit | Matt Riedemann proposed openstack/requirements: Raise upper-constraints on python-novaclient to 2.31.0 https://review.openstack.org/229962 | 15:32 |
| mriedem | not terrible | 15:32 |
| mriedem | classic | 15:32 |
| anteaya | terrible | 15:33 |
| * mriedem is still waiting for his investment to turnaround on this style http://thehowardtheatre.com/assets/Salt-N-Pepa-4_652x367.jpg | 15:33 | |
| anteaya | ha ha ha | 15:33 |
| anteaya | oh can I see you in a picture with that | 15:33 |
| anteaya | showing your mn tatoo? | 15:33 |
| mordred | Shrews: nope - next on the list - I did test it by running shade-inventory and looking at the output | 15:34 |
| mordred | Shrews: oh - however, I just added rax to the mix and found a bug | 15:34 |
| *** jsavak has quit IRC | 15:34 | |
| mordred | Shrews: I'm starting to think that we should build a set of fakes for each of the clouds we know of that is different | 15:34 |
| *** markvan__ is now known as markvan | 15:34 | |
| *** dramalho has quit IRC | 15:34 | |
| *** jsavak has joined #openstack-infra | 15:34 | |
| mordred | "this is what rackspace returns" - "this is what hp returns" etc | 15:34 |
| Shrews | mordred: hrm, don't know if i like that. then we have to periodically check to make sure the truth is still the trush | 15:35 |
| Shrews | truth | 15:35 |
| mordred | yah | 15:35 |
| *** jlanoux has joined #openstack-infra | 15:35 | |
| *** sridhar_ram has quit IRC | 15:35 | |
| mordred | well - what I mean is - we have set of different things that we know of, and shade is supposed to handle all of them | 15:35 |
| mordred | if a cloud changes data format to a new form we've never seen, nothing we can do about that - and if it stops being weird, that's fine, shade is still supposed to handle the form of data they _were_ giving | 15:36 |
| mordred | OR | 15:36 |
| mordred | we could just keep doing what we're doing no | 15:36 |
| mordred | now | 15:36 |
| mordred | I don't feel terribly strongly about it | 15:36 |
| Shrews | ok, yeah, i can see that as being useful. a defined set of use cases | 15:36 |
| Shrews | not tied to specific vendors | 15:37 |
| *** zeih has quit IRC | 15:37 | |
| Shrews | so much to do | 15:37 |
| *** angdraug has joined #openstack-infra | 15:38 | |
| *** persia has quit IRC | 15:38 | |
| openstackgerrit | Zara proposed openstack-infra/storyboard-webclient: Subscriptions https://review.openstack.org/223562 | 15:38 |
| wznoinsk | fungi: so you recreate git repos everynight if I get it correctly, do you know how long it takes? (I'm assuming you're on some 100MB+ connection to the review.openstack.org) | 15:39 |
| *** signed8bit_ZZZzz is now known as signed8bit | 15:39 | |
| *** niedbalski has joined #openstack-infra | 15:41 | |
| *** Daisy has quit IRC | 15:42 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/release-tools: Fix ImportError: No module named simplejson https://review.openstack.org/229969 | 15:42 |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate: Use overlay always to make SERVICE_HOST setup easy https://review.openstack.org/229970 | 15:42 |
| *** Daisy has joined #openstack-infra | 15:42 | |
| fungi | wznoinsk: i don't recreate them every night. that script only clones if the repository doesn't already exist locally, otherwise it just does a remote update | 15:43 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Provide short-circuit for finding server networks https://review.openstack.org/229140 | 15:44 |
| mordred | Shrews: ^^ that fixes it for rackspace now | 15:44 |
| fungi | wznoinsk: the clone operation follows a conditional test for whether the local destination exists | 15:44 |
| mordred | Shrews: I have verified that shade-inventory runs a reasonable amount of API calls and that the data returned into the inventory is correct | 15:45 |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: update grenade branch logic for mitaka https://review.openstack.org/229363 | 15:45 |
| Shrews | mordred: awesome | 15:46 |
| *** Daisy has quit IRC | 15:47 | |
| sdague | fungi: so just maybe, we have enough patches in flight we can change branches today | 15:47 |
| *** tonytan4ever has quit IRC | 15:47 | |
| Shrews | mordred: the root of that change set needs requirements.txt updated | 15:48 |
| fungi | sdague: should i keep closer tabs on #openstack-relmgr-office today? any help you need in speeding things up? | 15:48 |
| sdague | no, now that the client release is out there, I think I can stack up tests with depends-on | 15:49 |
| sdague | we'll see if that all works | 15:49 |
| fungi | or i guess #openstack-qa for branching devstack/grenade | 15:49 |
| fungi | since those aren't under release management | 15:49 |
| sdague | I'll poke you if I need anything from you | 15:49 |
| fungi | sdague: please do. happy to help however i can | 15:49 |
| mordred | Shrews: ++ | 15:49 |
| sdague | I just wanted to give you the heads up | 15:49 |
| fungi | appreciated! | 15:49 |
| *** Daisy has joined #openstack-infra | 15:50 | |
| *** ociuhandu has joined #openstack-infra | 15:52 | |
| *** e0ne has quit IRC | 15:53 | |
| *** sridhar_ram has joined #openstack-infra | 15:55 | |
| openstackgerrit | Chris Dent proposed openstack-infra/project-config: Enable ceilometer devstack plugin for tempest jobs https://review.openstack.org/229940 | 15:57 |
| *** Swami has quit IRC | 15:57 | |
| *** hdd has joined #openstack-infra | 15:58 | |
| pleia2 | good morning | 16:00 |
| *** sergmelikyan has quit IRC | 16:00 | |
| *** sergmelikyan has joined #openstack-infra | 16:01 | |
| *** moravec has quit IRC | 16:03 | |
| openstackgerrit | Tom Fifield proposed openstack-infra/groups-static-pages: Only need one URL for Official User Group https://review.openstack.org/229989 | 16:03 |
| anteaya | morning pleia2 | 16:04 |
| *** moravec has joined #openstack-infra | 16:05 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 16:05 | |
| *** abregman|afk has joined #openstack-infra | 16:07 | |
| *** sergmelikyan has quit IRC | 16:07 | |
| *** thingee has joined #openstack-infra | 16:08 | |
| *** samuelBartel has quit IRC | 16:08 | |
| openstackgerrit | Jim Rollenhagen proposed openstack-infra/project-config: Allow ironic-python-agent to have stable/managed releases https://review.openstack.org/229993 | 16:09 |
| *** ildikov has quit IRC | 16:09 | |
| *** unicell has quit IRC | 16:09 | |
| *** Swanson has joined #openstack-infra | 16:09 | |
| *** markus_z has quit IRC | 16:09 | |
| *** tonytan4ever has joined #openstack-infra | 16:10 | |
| *** thiagop has joined #openstack-infra | 16:10 | |
| *** mudassirlatif has joined #openstack-infra | 16:10 | |
| *** Daisy has quit IRC | 16:11 | |
| *** Daisy has joined #openstack-infra | 16:12 | |
| anteaya | jroll: yeah looking at that patch | 16:12 |
| *** jlanoux has quit IRC | 16:12 | |
| anteaya | jroll: um, I read the commit message about the motivation, but has someone tried to use permissions and been blocked? | 16:12 |
| jroll | anteaya: thanks! not sure about the tags thing | 16:13 |
| *** jlanoux has joined #openstack-infra | 16:13 | |
| jroll | anteaya: oh, let me provide context | 16:13 |
| anteaya | jroll: I don't recall seeing a [access "refs/heads/stable/liberty"] stanza before | 16:13 |
| *** r-daneel has joined #openstack-infra | 16:13 | |
| jroll | anteaya: https://review.openstack.org/#/c/229983/ | 16:13 |
| anteaya | perhaps it is required and I just don't know | 16:13 |
| *** dramalho has joined #openstack-infra | 16:13 | |
| jroll | and yeah, I hadn't seen that either, but it's in ironic.config so I'm rolling with it | 16:13 |
| anteaya | jroll: fair enough as a starting point | 16:14 |
| *** signed8bit_ZZZzz is now known as signed8bit | 16:14 | |
| jroll | anteaya: I assume it's related to the changes in release management | 16:14 |
| anteaya | I am pushing back just on the fact I haven't seen it before, continuing to dig | 16:14 |
| anteaya | which I haven't been able to keep up with, so that's on me | 16:14 |
| *** trown is now known as trown|lunch | 16:15 | |
| *** odgrim has joined #openstack-infra | 16:15 | |
| jroll | anteaya: right, I also pointed release management folks at it, so hopefully they can point it out | 16:15 |
| *** Daisy has quit IRC | 16:16 | |
| anteaya | jroll: well look at that, nova has the same thing: http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/openstack/nova.config | 16:17 |
| *** salv-orlando has joined #openstack-infra | 16:17 | |
| jroll | \o/ | 16:18 |
| *** jaypipes has quit IRC | 16:18 | |
| *** shashank_hegde has joined #openstack-infra | 16:20 | |
| *** salv-orlando has quit IRC | 16:20 | |
| *** eye_ has joined #openstack-infra | 16:21 | |
| *** mriedem is now known as mriedem_lunch | 16:22 | |
| *** jerrygb has left #openstack-infra | 16:22 | |
| *** Daisy has joined #openstack-infra | 16:23 | |
| anteaya | well you are following what others are doing | 16:23 |
| anteaya | the confusion is mine | 16:23 |
| *** thingee has quit IRC | 16:23 | |
| *** bltavares has joined #openstack-infra | 16:23 | |
| anteaya | as I thought I had remembered from last week fungi telling me that this wasn't necessary | 16:24 |
| anteaya | but I possibly am miss remembering | 16:24 |
| anteaya | as there seems to be quite the crowd doing this | 16:24 |
| openstackgerrit | Merged openstack/os-testr: Force utf8 encoding on subunit attachments output https://review.openstack.org/229527 | 16:25 |
| *** stevemar_ has joined #openstack-infra | 16:25 | |
| *** matrohon has quit IRC | 16:26 | |
| *** ajmiller has quit IRC | 16:26 | |
| *** shayneburgess has quit IRC | 16:26 | |
| *** nadya has quit IRC | 16:26 | |
| *** vipul has quit IRC | 16:27 | |
| *** yamahata has joined #openstack-infra | 16:27 | |
| fungi | it also may just be le cult de cargo | 16:27 |
| fungi | looking now | 16:27 |
| *** stevemar has quit IRC | 16:27 | |
| *** ajmiller has joined #openstack-infra | 16:27 | |
| dramalho | clarkb: Could can look? https://review.openstack.org/#/c/222694/6 | 16:28 |
| *** vipul has joined #openstack-infra | 16:28 | |
| *** shayneburgess has joined #openstack-infra | 16:28 | |
| *** jistr has quit IRC | 16:29 | |
| *** hdd has quit IRC | 16:29 | |
| fungi | anteaya: oh, refs/heads/stable/liberty is a temporary block granting release management and project milestone teams exclusive control, to prevent stable-maint teams taking over the stable/liberty branches prior to the release | 16:29 |
| *** dtantsur is now known as dtantsur|afk | 16:29 | |
| anteaya | fungi: ah my fault, thank you | 16:29 |
| *** hdd has joined #openstack-infra | 16:30 | |
| fungi | it gets added at rc1 and removed on or shortly after release day | 16:30 |
| anteaya | perhaps it has happened in releases past and i have been out to lunch? | 16:30 |
| *** mriedem_lunch is now known as mriedem | 16:30 | |
| fungi | we did it first for kilo | 16:30 |
| fungi | so it's a new-ish process | 16:31 |
| *** bdemers has quit IRC | 16:31 | |
| anteaya | ah | 16:31 |
| *** ameade has quit IRC | 16:31 | |
| anteaya | I was out to lunch on kilo then | 16:31 |
| anteaya | thank you | 16:31 |
| *** abregman|afk has quit IRC | 16:31 | |
| anteaya | jroll: welcome to the cargo cult, nice copy/pasting | 16:31 |
| fungi | in previous releases we instead used a temporary "proposed" branch during rc | 16:31 |
| dims | haha ++ | 16:31 |
| anteaya | jroll: but in this case you are correct | 16:31 |
| *** bdemers has joined #openstack-infra | 16:31 | |
| jroll | anteaya: I'm great at cargo culting | 16:31 |
| jroll | fungi: thanks for the explanation :) | 16:32 |
| anteaya | jroll: just stay in step, noone will no you are making it up | 16:32 |
| fungi | which while not needing temporary acls, had problems of its own | 16:32 |
| anteaya | know | 16:32 |
| anteaya | fungi: I remember proposed | 16:32 |
| *** Sukhdev has joined #openstack-infra | 16:32 | |
| *** ameade has joined #openstack-infra | 16:33 | |
| *** ashleigh_ has quit IRC | 16:33 | |
| *** sergmelikyan has joined #openstack-infra | 16:34 | |
| *** bharathm has joined #openstack-infra | 16:34 | |
| *** jlanoux has quit IRC | 16:34 | |
| *** ddieterly has quit IRC | 16:35 | |
| anteaya | ttx: nice juggling | 16:35 |
| *** devvesa has quit IRC | 16:35 | |
| *** kashyap has joined #openstack-infra | 16:36 | |
| kashyap | Any rough idea how many jobs the CI infra runs each day? | 16:37 |
| sigmavirus24 | Does the release infrastructure sign packages and upload the signatures with twine? (Or can you point me at the repo that has that automation?) | 16:37 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Pin delorean to a pre mock version https://review.openstack.org/230011 | 16:37 |
| anteaya | ttx: I'm liking some of the condensed names | 16:37 |
| *** ZZelle_ has joined #openstack-infra | 16:37 | |
| anteaya | ttx: wondering if we can work an apostrophe in perhaps in future | 16:37 |
| anteaya | Puppet'O | 16:37 |
| anteaya | O'Ansible | 16:38 |
| *** bharathm has quit IRC | 16:38 | |
| anteaya | kashyap: you can start with the graphs at the bottom of http://status.openstack.org/zuul/ | 16:38 |
| anteaya | kashyap: zuul jobs launched per hour might give you a rough estimate | 16:39 |
| *** salv-orlando has joined #openstack-infra | 16:39 | |
| *** salv-orlando has quit IRC | 16:39 | |
| kashyap | anteaya: Ah, right - thanks for the pointer | 16:40 |
| anteaya | kashyap: welcome hope it fits your needs | 16:40 |
| zaro | morning | 16:40 |
| anteaya | morning zaro | 16:40 |
| *** salv-orlando has joined #openstack-infra | 16:40 | |
| fungi | sigmavirus24: not yet http://specs.openstack.org/openstack-infra/infra-specs/specs/artifact-signing.html | 16:40 |
| kashyap | anteaya: Was making some presentation & wanted to include a point about the kind of bugs the CI infra allows to find | 16:40 |
| anteaya | zaro: if you look in scrollback we have another respondant to the call for gerrit help | 16:40 |
| kashyap | (If you're curious why I asked.) | 16:41 |
| *** sergmelikyan has quit IRC | 16:41 | |
| anteaya | kashyap: ah wonderful, I hope the presentation goes well | 16:41 |
| kashyap | Thx! | 16:41 |
| anteaya | kashyap: I'm assuming you wanted to know, but thanks for sharing | 16:41 |
| anteaya | kashyap: we have a repo with presentation if you want to take a look at some of them | 16:41 |
| sigmavirus24 | fungi: okay. Was going to ask y'all to test https://github.com/pypa/twine/pull/138 if you had a chance since it was broken between 1.5.0 and 1.6.x and I just learned about it to fix it | 16:41 |
| kashyap | anteaya: Got a pointer to it? | 16:41 |
| anteaya | kashyap: yes: http://git.openstack.org/cgit/openstack-infra/publications/ | 16:42 |
| sigmavirus24 | (If anyone else signs packages and uploads the packages+signatures with twine and could test ^ though, I'd love it) | 16:42 |
| *** Swami has joined #openstack-infra | 16:42 | |
| fungi | kashyap: those graphs are rendered real-time from data in graphite.openstack.org so feel free to poke around its webui or just tweak parameters on one of those image urls from the zuul status page to get something suitable for your use (the parameter names are pretty self-explanatory) | 16:42 |
| zaro | anteaya: thanks. looks like he/she dropped off. | 16:42 |
| anteaya | kashyap: you can view those presentations here: http://docs.openstack.org/infra/publications/ | 16:42 |
| fungi | sigmavirus24: yeah, i saw you mention the bug earlier (can't recall if it was in a bug report or thread on distutils-sig ml) | 16:43 |
| anteaya | zaro: yes, hopefully (s)he will return, sounded like it | 16:43 |
| *** amit213 has joined #openstack-infra | 16:43 | |
| sigmavirus24 | fungi: yeah we added a dependency for large file support and the person who did it, did it a little carelessly. There was a different bug around discovering the files appropriately too that was fixed in 1.6.2. Also you may be in #pypa-dev where we kind of discussed this a little | 16:44 |
| *** moravec has quit IRC | 16:44 | |
| kashyap | fungi: Sweet. Thanks for the info. I'm making in LaTex (my drawing in TikZ module skills are not up2date), for now only ASCII diagrams and numbers :-) | 16:45 |
| *** tonytan4ever has quit IRC | 16:45 | |
| *** sergmelikyan has joined #openstack-infra | 16:46 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/devstack-gate: Use overlay always to make SERVICE_HOST setup easy https://review.openstack.org/229970 | 16:46 |
| fungi | kashyap: oh, you can change the render engine on the graph url to something like "raw" and get a json blob of values too | 16:46 |
| fungi | ahh, it's &rawData=true | 16:46 |
| fungi | kashyap: like http://graphite.openstack.org/render/?target=summarize(stats_counts.zuul.pipeline.check.all_jobs,%226months%22)&from=-6months&rawData=true | 16:47 |
| kashyap | fungi: Neat, /me clicks | 16:47 |
| * anteaya imposes self-restraint to not troll the #openstack channel | 16:47 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Trigger image uploads using gearman https://review.openstack.org/197722 | 16:47 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Trigger image builds using gearman https://review.openstack.org/195175 | 16:47 |
| sigmavirus24 | anteaya: uhoh, what's going on? | 16:47 |
| anteaya | just read the last statement | 16:48 |
| anteaya | I can't stop laughing | 16:48 |
| * sigmavirus24 doesn't idle in there for obvious reasons | 16:48 | |
| fungi | http://eavesdrop.openstack.org/irclogs/%23openstack/latest.log.html | 16:48 |
| kashyap | fungi: That's useful detail, much appreciated | 16:48 |
| *** nadya has joined #openstack-infra | 16:48 | |
| fungi | sigmavirus24: for when you don't want to idle | 16:48 |
| *** ajmiller has quit IRC | 16:48 | |
| anteaya | sigmavirus24: ttp://eavesdrop.openstack.org/irclogs/%23openstack/%23openstack.2015-10-01.log.html#t2015-10-01T16:15:07 | 16:49 |
| wznoinsk | fungi: thanks for that script, can I reuse it? | 16:49 |
| sigmavirus24 | thanks fungi and anteaya | 16:49 |
| anteaya | sigmavirus24: welcome | 16:49 |
| *** signed8bit is now known as signed8bit_ZZZzz | 16:49 | |
| *** ajmiller has joined #openstack-infra | 16:49 | |
| anteaya | yeah not idling in all the channels is a good strategy | 16:49 |
| fungi | wznoinsk: feel free. it's short enough i would just consider it public domain/no copyright | 16:49 |
| sigmavirus24 | anteaya: now I get it :D | 16:49 |
| anteaya | sigmavirus24: ha ha ha | 16:49 |
| wznoinsk | fungi: thanks, saved me some time to script-out the process myself | 16:49 |
| *** nadya has quit IRC | 16:50 | |
| fungi | yw | 16:50 |
| sigmavirus24 | anteaya: someone should put the festivus poll in #openstack so we can have an airing of greivances | 16:50 |
| sigmavirus24 | I'll join for the Glance part of it =P | 16:50 |
| anteaya | sigmavirus24: ah naw | 16:50 |
| fungi | i suspect we'll _all_ join for the glance part of it | 16:50 |
| anteaya | oh festivus | 16:50 |
| *** nadya has joined #openstack-infra | 16:50 | |
| anteaya | who was telling me about that | 16:50 |
| anteaya | travisT | 16:50 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Add dib deletion test https://review.openstack.org/215275 | 16:50 |
| sigmavirus24 | I only mean that because I can help fix some of that if my employer ever gives me time to work on OS again | 16:51 |
| *** bharathm has joined #openstack-infra | 16:51 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Trigger image delete using gearman https://review.openstack.org/198114 | 16:51 |
| sigmavirus24 | anteaya: yeah, Searchlight *loves* us ;P | 16:51 |
| * sigmavirus24 says as a Searchlight core | 16:51 | |
| anteaya | sigmavirus24: ah yes, let's get you back working on openstack things | 16:51 |
| anteaya | sigmavirus24: hmmmm, are you goign to summit? | 16:51 |
| sigmavirus24 | sadly not | 16:51 |
| sigmavirus24 | family obligations and such | 16:51 |
| *** signed8bit_ZZZzz is now known as signed8bit | 16:52 | |
| anteaya | ah well glad you have folks who love you | 16:52 |
| anteaya | that lasts longer than tokyo | 16:52 |
| sigmavirus24 | love might be a bit strong of a word choice =P | 16:52 |
| anteaya | well a trip anyway | 16:52 |
| openstackgerrit | Danilo Ramalho proposed openstack-infra/puppet-gerrit: Move test assertions closer to apply. https://review.openstack.org/227571 | 16:52 |
| anteaya | yes sorry, <appropriate descrition of choice> | 16:52 |
| anteaya | better? | 16:52 |
| *** erlon has joined #openstack-infra | 16:53 | |
| sigmavirus24 | lol thanks :P | 16:53 |
| pabelanger | EmilienM: Yup, updated reviews you commented on | 16:53 |
| anteaya | welcome | 16:53 |
| *** glauco has quit IRC | 16:53 | |
| *** ddieterly has joined #openstack-infra | 16:53 | |
| EmilienM | pabelanger: thank you :) | 16:54 |
| *** moravec has joined #openstack-infra | 16:54 | |
| openstackgerrit | Merged openstack-infra/grafyaml: Reorder tox.ini envlist https://review.openstack.org/224865 | 16:54 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/requirements: Update python-novaclient for new release 2.31.0 https://review.openstack.org/230015 | 16:54 |
| *** e0ne has joined #openstack-infra | 16:54 | |
| *** dramalho_ has joined #openstack-infra | 16:55 | |
| *** e0ne has quit IRC | 16:55 | |
| *** jsavak has quit IRC | 16:55 | |
| *** bltavare_ has joined #openstack-infra | 16:55 | |
| *** glauco_ has joined #openstack-infra | 16:55 | |
| *** markvoelker has quit IRC | 16:55 | |
| *** dramalho_ has quit IRC | 16:55 | |
| *** glauco_ has quit IRC | 16:55 | |
| *** jsavak has joined #openstack-infra | 16:55 | |
| *** bltavare_ has quit IRC | 16:55 | |
| *** markvoelker has joined #openstack-infra | 16:56 | |
| *** trown|lunch is now known as trown | 16:56 | |
| *** unicell has joined #openstack-infra | 16:56 | |
| *** dramalho_ has joined #openstack-infra | 16:56 | |
| *** dramalho_ has quit IRC | 16:56 | |
| *** dramalho has quit IRC | 16:56 | |
| *** geoffarnold has quit IRC | 16:56 | |
| *** bltavare_ has joined #openstack-infra | 16:57 | |
| *** persia has joined #openstack-infra | 16:57 | |
| *** bltavare_ has quit IRC | 16:57 | |
| *** glauco has joined #openstack-infra | 16:57 | |
| *** glauco has quit IRC | 16:57 | |
| *** dramalho has joined #openstack-infra | 16:57 | |
| *** bltavare_ has joined #openstack-infra | 16:57 | |
| *** bltavare_ has quit IRC | 16:57 | |
| *** dramalho_ has joined #openstack-infra | 16:57 | |
| *** bltavare_ has joined #openstack-infra | 16:58 | |
| *** bltavares has quit IRC | 16:58 | |
| *** glauco has joined #openstack-infra | 16:58 | |
| *** dramalho has quit IRC | 16:58 | |
| *** dramalho_ has quit IRC | 16:58 | |
| *** bltavare_ has quit IRC | 16:58 | |
| *** notnownikki has quit IRC | 16:59 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Added recent version of npm to release slave https://review.openstack.org/229381 | 16:59 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Remove distro-package npm from release slave. https://review.openstack.org/229380 | 16:59 |
| *** bltavares has joined #openstack-infra | 16:59 | |
| *** puranamr has joined #openstack-infra | 16:59 | |
| *** dramalho has joined #openstack-infra | 16:59 | |
| *** bltavares has quit IRC | 16:59 | |
| *** dramalho has quit IRC | 16:59 | |
| *** baoli has quit IRC | 17:00 | |
| *** nadya has quit IRC | 17:00 | |
| fungi | krotscheck: interesting... so npm can gracefully uninstall itself? | 17:01 |
| *** dramalho has joined #openstack-infra | 17:01 | |
| *** puranamr has quit IRC | 17:01 | |
| *** glauco__ has joined #openstack-infra | 17:01 | |
| *** derekh has quit IRC | 17:01 | |
| *** dramalho has quit IRC | 17:01 | |
| *** nadya has joined #openstack-infra | 17:01 | |
| *** glauco__ has quit IRC | 17:01 | |
| *** baoli has joined #openstack-infra | 17:01 | |
| krotscheck | fungi: THe more recent versions, yes. | 17:01 |
| *** bltavares has joined #openstack-infra | 17:01 | |
| *** bltavares has quit IRC | 17:01 | |
| *** dramalho has joined #openstack-infra | 17:02 | |
| *** sdake_ has joined #openstack-infra | 17:02 | |
| *** dramalho has quit IRC | 17:02 | |
| krotscheck | fungi: That particular patch still left some leftover bits in the /man[1-7]/ folders, which seem to have been leftover debian cruft. | 17:02 |
| *** glauco has quit IRC | 17:02 | |
| *** bltavares has joined #openstack-infra | 17:02 | |
| krotscheck | So, amusingly enough, npm is better at uninstalling itself than debian packaging is. | 17:02 |
| *** dramalho has joined #openstack-infra | 17:02 | |
| *** dramalho has quit IRC | 17:02 | |
| sambetts | Hey infra, can i get some eyes on this https://review.openstack.org/#/c/229060/ when someones got a spare moment | 17:02 |
| *** mriedem has quit IRC | 17:02 | |
| *** bltavares has quit IRC | 17:02 | |
| *** dims_ has joined #openstack-infra | 17:03 | |
| *** maishsk has joined #openstack-infra | 17:03 | |
| *** glauco_ has joined #openstack-infra | 17:03 | |
| *** glauco_ has quit IRC | 17:03 | |
| krotscheck | fungi: Yes it was a rebase. habit | 17:03 |
| *** sdake_ has quit IRC | 17:04 | |
| fungi | krotscheck: the nodejs base class installs an npm package i take it? | 17:04 |
| *** dramalho has joined #openstack-infra | 17:04 | |
| *** bltavares has joined #openstack-infra | 17:04 | |
| *** dims has quit IRC | 17:04 | |
| *** bltavares has quit IRC | 17:04 | |
| *** dramalho has quit IRC | 17:04 | |
| krotscheck | fungi: It does only when you're using the nodesource repos, which is automatic for that module unless explicitly disabled. | 17:04 |
| *** glauco__ has joined #openstack-infra | 17:04 | |
| krotscheck | fungi: It's... not optimal, but it works. | 17:05 |
| *** glauco__ has quit IRC | 17:05 | |
| *** sabeen1 has quit IRC | 17:05 | |
| *** maishsk has quit IRC | 17:05 | |
| *** sdake has quit IRC | 17:05 | |
| *** Piet has quit IRC | 17:06 | |
| *** sabeen has joined #openstack-infra | 17:06 | |
| *** sabeen has quit IRC | 17:06 | |
| *** bltavares has joined #openstack-infra | 17:06 | |
| *** ildikov has joined #openstack-infra | 17:06 | |
| *** bltavares has quit IRC | 17:06 | |
| fungi | krotscheck: why not keep the (temporarily unused) npmrc template? removing it and adding it back again creates a bit of a git blame roadblock | 17:06 |
| *** HeOS has quit IRC | 17:07 | |
| krotscheck | krotscheck: Well, the git blame thing will point to me anyway. | 17:07 |
| krotscheck | fungi: ^^ | 17:07 |
| fungi | hah | 17:07 |
| krotscheck | fungi: And, well, I was going for clean slate. | 17:07 |
| *** mriedem has joined #openstack-infra | 17:07 | |
| *** dramalho_ has joined #openstack-infra | 17:07 | |
| *** dramalho_ has quit IRC | 17:07 | |
| *** bltavares has joined #openstack-infra | 17:07 | |
| krotscheck | It was easier for me to test that in my puppet sandbox too. | 17:07 |
| *** bltavares has quit IRC | 17:07 | |
| fungi | but yeah, you could just leave it behind in the first patch then git mv it in the second | 17:07 |
| *** asmigala has quit IRC | 17:08 | |
| fungi | the contents seem unchanged | 17:08 |
| *** sabeen1 has joined #openstack-infra | 17:08 | |
| fungi | and it's just temporary cruft in the repo that way | 17:08 |
| *** bltavares has joined #openstack-infra | 17:09 | |
| *** dramalho has joined #openstack-infra | 17:09 | |
| *** bltavares has quit IRC | 17:09 | |
| *** dramalho has quit IRC | 17:09 | |
| krotscheck | fungi: Want me to send up a new patchset? | 17:09 |
| *** dramalho has joined #openstack-infra | 17:09 | |
| krotscheck | fungi: The content should be different, if it's not then I did something wrong..... | 17:10 |
| * krotscheck goes to check | 17:10 | |
| *** dramalho_ has joined #openstack-infra | 17:10 | |
| fungi | ahh, you're adding two lines | 17:10 |
| *** bltavares has joined #openstack-infra | 17:10 | |
| krotscheck | fungi: Right- the difference is that it now (again) contains tag-version-prefix and sign-git-tag | 17:10 |
| *** dramalho_ has quit IRC | 17:10 | |
| *** tsekiyama has quit IRC | 17:11 | |
| krotscheck | Which I removed from the first iteration of this patch way back when and moved into the lgobal config. | 17:11 |
| *** bltavares has quit IRC | 17:11 | |
| *** Daisy_ has joined #openstack-infra | 17:11 | |
| fungi | so, yeah, another reason to keep that file instead of removing and readding. easier to see what's changed between them | 17:11 |
| krotscheck | kk, will come up with a new one. | 17:11 |
| krotscheck | (Doctor's appointment for the munchkin, afk for a bit) | 17:11 |
| fungi | thanks. otherwise i'm cool with this and will +2 once you do | 17:11 |
| anteaya | krotscheck: hope all is well | 17:11 |
| *** Ala has quit IRC | 17:12 | |
| *** bltavares has joined #openstack-infra | 17:12 | |
| *** dramalho_ has joined #openstack-infra | 17:12 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 17:13 |
| *** bltavares has quit IRC | 17:13 | |
| *** dramalho_ has quit IRC | 17:13 | |
| *** austin81 has joined #openstack-infra | 17:13 | |
| greghaynes | clarkb: ooo ^ | 17:13 |
| greghaynes | clarkb: what motivated that? | 17:13 |
| clarkb | greghaynes: I want better testing of node life | 17:14 |
| *** dramalho has quit IRC | 17:14 | |
| clarkb | greghaynes: since thats core function of nodepool and realized that zmq events were completely untested | 17:14 |
| *** Daisy has quit IRC | 17:14 | |
| greghaynes | node life sounds like an openstack life insurance plan | 17:14 |
| clarkb | next stpe which may actually be required here is better testing of the jenkins stuff | 17:14 |
| fungi | node life mutual | 17:14 |
| greghaynes | nice | 17:14 |
| *** dramalho has joined #openstack-infra | 17:14 | |
| *** dramalho has quit IRC | 17:14 | |
| *** glauco_ has joined #openstack-infra | 17:14 | |
| *** glauco_ has quit IRC | 17:15 | |
| clarkb | I also think my yield in those tests is insufficient and prone to failure | 17:15 |
| *** bltavares has joined #openstack-infra | 17:15 | |
| clarkb | but push early and often | 17:15 |
| anteaya | ah our summit friday is with ironic in the morning then we change rooms and are with qa and relmgt in the afternoon | 17:15 |
| *** bltavares has quit IRC | 17:15 | |
| *** dramalho has joined #openstack-infra | 17:15 | |
| fungi | anteaya: yep. ironic only has a half-day scheduled | 17:15 |
| *** dramalho has quit IRC | 17:15 | |
| fungi | and we wanted some overlap with them because infra cloud bare metal goodness | 17:15 |
| anteaya | yup, makes sense | 17:16 |
| *** glauco has joined #openstack-infra | 17:16 | |
| *** glauco has quit IRC | 17:16 | |
| fungi | and as always there's plenty of back-scratching to go around between infra/qa/relmgmt to last a half day | 17:16 |
| *** glauco_ has joined #openstack-infra | 17:16 | |
| *** glauco_ has quit IRC | 17:16 | |
| *** dramalho has joined #openstack-infra | 17:16 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 17:17 |
| *** dramalho_ has joined #openstack-infra | 17:17 | |
| *** dramalho_ has quit IRC | 17:17 | |
| *** dramalho has quit IRC | 17:17 | |
| *** Daisy_ has quit IRC | 17:17 | |
| *** glauco has joined #openstack-infra | 17:17 | |
| *** dramalho has joined #openstack-infra | 17:18 | |
| *** stevemar_ has quit IRC | 17:18 | |
| *** ihrachys has quit IRC | 17:19 | |
| *** ajmiller has quit IRC | 17:21 | |
| *** ajmiller has joined #openstack-infra | 17:23 | |
| anteaya | fungi: great have added those details to the etherpad | 17:23 |
| anteaya | fungi: you might need to drum up some design summit ideas business | 17:24 |
| *** jian5397 has joined #openstack-infra | 17:24 | |
| *** jsavak has quit IRC | 17:24 | |
| *** jian5397 is now known as michaelxin | 17:25 | |
| openstackgerrit | Clint Adams proposed openstack-infra/puppet-httpd: Allow specification of locations to not be proxied https://review.openstack.org/230029 | 17:25 |
| openstackgerrit | James E. Blair proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/230030 | 17:25 |
| *** jsavak has joined #openstack-infra | 17:25 | |
| anteaya | Clint: have you an example of someone wanting to exclude urls from being proxied? | 17:26 |
| *** abitha has joined #openstack-infra | 17:26 | |
| Clint | anteaya: robots.txt on paste.o.o, perhaps | 17:27 |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Add the ability to parse directories https://review.openstack.org/227438 | 17:27 |
| anteaya | Clint: I wondered if that was it, thank you | 17:27 |
| * Clint nods. | 17:27 | |
| *** geoffarnold has joined #openstack-infra | 17:29 | |
| *** mrmartin has joined #openstack-infra | 17:30 | |
| openstackgerrit | Glauco Oliveira proposed openstack-infra/puppet-openstack-health: Add provisioning scripts https://review.openstack.org/227849 | 17:30 |
| openstackgerrit | James E. Blair proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/230030 | 17:31 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 17:31 |
| openstackgerrit | James E. Blair proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/230030 | 17:33 |
| *** ashleighfarnham has joined #openstack-infra | 17:33 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Added recent version of npm to release slave https://review.openstack.org/229381 | 17:33 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Remove distro-package npm from release slave. https://review.openstack.org/229380 | 17:33 |
| *** ybathia has joined #openstack-infra | 17:34 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/project-config: [WIP] Add grafana dashboard for status.o.o/zuul https://review.openstack.org/224834 | 17:34 |
| *** odgrim has quit IRC | 17:34 | |
| bswartz | what kind of power is required to modify the topic in a gerrit change, if it's incorrect? | 17:34 |
| Clint | same access as to make a new patchset | 17:35 |
| bswartz | but anyone can make a new patchset.... that doesn't change the topic in gerrit | 17:35 |
| *** amuller is now known as amuller_afk | 17:35 | |
| greghaynes | crinkle: ohai | 17:36 |
| fungi | bswartz: Clint: according to our global config, all registered users can edit any change's topic directly through the webui https://review.openstack.org/gitweb?p=All-Projects.git;a=blob_plain;f=project.config | 17:36 |
| anteaya | if you want to change it in the gui, I believe it is patch owner and project core reviewers | 17:36 |
| crinkle | greghaynes: hello | 17:36 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Provide short-circuit for finding server networks https://review.openstack.org/229140 | 17:36 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Provide shortcut around has_service https://review.openstack.org/230032 | 17:36 |
| anteaya | fungi: can they? | 17:36 |
| *** signed8bit is now known as signed8bit_ZZZzz | 17:36 | |
| greghaynes | crinkle: so the plan for infra-cloud (this should be documented) was to use glean + config-drive to configure networking | 17:36 |
| crinkle | greghaynes: okay i am missing those bits | 17:36 |
| zaro | pelix: had a question for you in https://review.openstack.org/#/c/226107/ | 17:36 |
| bswartz | fungi, Clint: it seems that only works if the topic is blank -- once it's set, there is no edit button | 17:36 |
| fungi | anteaya: we made that update a month or two ago (check git history on our gerrit config documentation in system-config for the approximate date) | 17:36 |
| *** ociuhandu has quit IRC | 17:36 | |
| anteaya | bswartz: you can also download the patch with git review -d and then upload it with git review -t <topic> | 17:36 |
| clarkb | jeblair: I am a little confused around why accessIPv4's value changes in the ipv6 change | 17:36 |
| crinkle | greghaynes: which makes it hard to get new nodes up | 17:36 |
| Clint | bswartz: you can do it through git push or git review | 17:36 |
| greghaynes | crinkle: dhcp is there for bifrost (so we can pxe boot the hosts) | 17:36 |
| jeblair | clarkb, fungi: related to our discussion yesterday, can you take a look at https://review.openstack.org/230030 ? mordred and i have been riffing on how to return the "best" ip address for ansible, (potentially nodepool, but probably not), etc, in the most ipv6 friendly way. | 17:36 |
| anteaya | fungi: thanks will do | 17:37 |
| jeblair | clarkb: oh yay you already are :) | 17:37 |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Add the ability to parse directories https://review.openstack.org/227438 | 17:37 |
| greghaynes | crinkle: When you say missing... | 17:37 |
| clarkb | jeblair: :) | 17:37 |
| greghaynes | crinkle: There are some missing parts. This is the issue we had with vlans and that - bifrost didnt support generating config drives that specified vlan info | 17:37 |
| fungi | bswartz: have an example change where you don't see an edit button for the topic when you're logged in? | 17:37 |
| *** ashleighfarnham has quit IRC | 17:37 | |
| greghaynes | crinkle: https://review.openstack.org/#/c/216431/ is for that | 17:37 |
| fungi | i thought it was supposed to be allowing it for everyone now | 17:37 |
| fungi | jeblair: thanks, reviewing | 17:38 |
| jeblair | clarkb: because there used to be a "prefer ipv6" variable, but we're inverting that to "force ipv4", so the default behavior in that method changes. | 17:38 |
| *** glauco has quit IRC | 17:38 | |
| jeblair | clarkb: my understand is that none of this is actually in use yet, so that's okay. | 17:38 |
| *** dramalho_ has joined #openstack-infra | 17:38 | |
| anteaya | fungi: thank you http://git.openstack.org/cgit/openstack-infra/system-config/commit/doc/source/gerrit.rst?id=afcfe2b904a3b57190b7a8c050aa34071b85ada8 | 17:38 |
| *** glauco__ has joined #openstack-infra | 17:38 | |
| *** glauco__ has quit IRC | 17:39 | |
| *** dramalho_ has quit IRC | 17:39 | |
| jeblair | clarkb: so i attempted to ensure both paths are still tested. but i guess the spirit of the change is that you should get ipv6 by default, so since that test looked like the most "defaulty" test, i changed its expected results. | 17:39 |
| clarkb | jeblair: mordred I think you can just getaddrinfo for checking if ipv6 is available | 17:39 |
| *** Sukhdev has quit IRC | 17:39 | |
| anteaya | Clint: we tend to favour folks use git review, some people have gotten themselves into issues not using it | 17:39 |
| fungi | anteaya: we try to use that documentation as a proxy for change control over the global config, since it's not directly under change review itself | 17:39 |
| clarkb | jeblair: mordred the INET6 data you get back has scope info and if global I think you ar egood | 17:39 |
| anteaya | fungi: ah okay thank you | 17:39 |
| clarkb | that way you don't need to rely on another dep or check with routing tables | 17:39 |
| crinkle | greghaynes: so that patch would help set the vlan, what about setting the private ip on eth2? | 17:39 |
| anteaya | fungi: yes you change it via the gui | 17:39 |
| Clint | anteaya: i've been getting myself into issues using it too | 17:40 |
| anteaya | Clint: git push? | 17:40 |
| greghaynes | crinkle: the glean + configdrive networking does work though - I have deployed images on the 10.x net (untagged) using bifrost there... | 17:40 |
| *** dramalho_ has joined #openstack-infra | 17:40 | |
| *** glauco__ has joined #openstack-infra | 17:40 | |
| *** dramalho_ has quit IRC | 17:40 | |
| mordred | clarkb: getaddrinfo on what? | 17:40 |
| greghaynes | crinkle: yea, we should do both, bifrist 'just works' for setting the untagged networking | 17:40 |
| greghaynes | crinkle: it doesnt really have a notion of two networks though, so that is an issue | 17:40 |
| bswartz | fungi: https://review.openstack.org/#/c/228159/ | 17:40 |
| *** jsavak has quit IRC | 17:40 | |
| greghaynes | crinkle: I think we can get by with just the public networking on deployed nodes at first, but IMBW | 17:41 |
| bswartz | fungi: you'll not that the topic has an incorrect bug number | 17:41 |
| bswartz | note | 17:41 |
| *** glauco has joined #openstack-infra | 17:41 | |
| *** glauco has quit IRC | 17:41 | |
| *** dramalho_ has joined #openstack-infra | 17:41 | |
| anteaya | bswartz: that patch has merged | 17:41 |
| fungi | bswartz: oh! well you can't edit the topic on closed changes, only open ones | 17:41 |
| mordred | clarkb: also, the above patch takes about 0.00006 seconds to run and just looks at the state of the interfaces, so should be pretty minimally invasive | 17:41 |
| Clint | anteaya: no, git review | 17:41 |
| SpamapS | Hey so if I started dumping _a lot_ more numbers into graphite.. would that stress it out? I'm thinking about some of the counters from dstat in devstack going in. | 17:41 |
| *** signed8bit_ZZZzz is now known as signed8bit | 17:41 | |
| anteaya | Clint: oh, what problems are you having with git review? | 17:41 |
| bswartz | anteaya: thanks! | 17:41 |
| *** dramalho_ has quit IRC | 17:41 | |
| fungi | bswartz: if the change is merged or abandoned you can't change a number of things, including the topic | 17:41 |
| *** kushal has quit IRC | 17:41 | |
| greghaynes | SpamapS: Do you mean a lot of values for certain stats or a lot more stats | 17:42 |
| crinkle | greghaynes: that's not what bifrost is currently doing | 17:42 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 17:42 |
| mordred | SpamapS: I would argue that if it stresses it out, it would be worthwhile to investigate scaling it out | 17:42 |
| clarkb | mordred: getaddrinfo on None gives you yourself right? | 17:42 |
| *** glauco___ has joined #openstack-infra | 17:42 | |
| SpamapS | mordred: Yes indeed. | 17:42 |
| jeblair | mordred: well, i think it uses the linux netlink api to get the routing table | 17:42 |
| *** glauco___ has quit IRC | 17:42 | |
| SpamapS | greghaynes: both. :) | 17:42 |
| mordred | SpamapS: however, I do not believe our build slaves have network access to our graphite | 17:42 |
| greghaynes | SpamapS: Basically graphite sucks at scaling out lots of different stats but does well at lots of input for each stat | 17:42 |
| *** dramalho has quit IRC | 17:42 | |
| *** glauco has joined #openstack-infra | 17:42 | |
| *** dramalho has joined #openstack-infra | 17:42 | |
| Clint | anteaya: it does things i don't expect when multiple semi-related patches are involved | 17:42 |
| *** dramalho has quit IRC | 17:42 | |
| anteaya | Clint: using git review is our supported tool for interacting with gerrit so I would like to help you smooth out the issues you are having with it | 17:42 |
| SpamapS | basically, there are probably 20 counters that will prove useful in chasing performance in OpenStack, and I'd like to at the very least throw the end-counter into graphite from every run of at least one devstack job. | 17:43 |
| Clint | there's also it changing topics when i forget to use -t, but that's just pebkac | 17:43 |
| clarkb | hrm I misread the docs None is useless /me reads more | 17:43 |
| anteaya | Clint: okay can you describe or link to an example? | 17:43 |
| openstackgerrit | Merged openstack-infra/puppet-graphite: Add acceptance tests for puppet-graphite. https://review.openstack.org/224252 | 17:43 |
| mordred | clarkb: that did not work | 17:43 |
| greghaynes | SpamapS: 20 seems like a small number to me | 17:43 |
| pelix | zaro: the intent is just to only show those log messages that were generated during the failing test alongside the test failure | 17:43 |
| *** dramalho has joined #openstack-infra | 17:43 | |
| *** dramalho has quit IRC | 17:43 | |
| SpamapS | greghaynes: Yeah I agree, but it might take doing 100 before we find the 20 that we need. | 17:43 |
| *** glauco has quit IRC | 17:43 | |
| *** glauco has joined #openstack-infra | 17:44 | |
| clarkb | mordred: ya only one or the other can be None | 17:44 |
| *** glauco has quit IRC | 17:44 | |
| greghaynes | SpamapS: yea, Id hope we can handle that... People get into issues when they dynamically generate stats... but I'd hope we can manage that many | 17:44 |
| crinkle | greghaynes: there's lots of local changes in /opt/stack/bifrost but none of them are related to 216431, you think applying that patch will make it "just work"? | 17:44 |
| clarkb | mordred: but not both args | 17:44 |
| *** glauco__ has quit IRC | 17:44 | |
| jeblair | greghaynes, SpamapS: yeah, we have 10,000s of keys, 100 more won't hurt | 17:45 |
| *** dramalho has joined #openstack-infra | 17:45 | |
| *** dramalho has quit IRC | 17:45 | |
| *** jsavak has joined #openstack-infra | 17:45 | |
| zaro | pelix: i though i compared it to how it works on master and didn't see much of a difference. am i seeing it wrong? | 17:45 |
| Clint | anteaya: mmm.. something like patch C depends on patch A, patch C depends on patch B, patch A and B are independent of each other, when doing git-review on patch C, it will rebase and push patch B on top of patch A or the other way around | 17:45 |
| jeblair | greghaynes, SpamapS: but yeah, each key becomes a large file on disk in perpetuity, so 100 keys per devstack run is awkward; but 100 keys where each devstack run contributes to the same 100 keys is not. | 17:45 |
| *** dramalho has joined #openstack-infra | 17:46 | |
| mordred | jeblair: how do we open network access to graphite to our devstack hosts? | 17:46 |
| greghaynes | crinkle: not likely. Those local changes were me hacking up the vlan and such patches IIRC (theres really not a great way to dev stuff like that). Might be a good idea to wipe out those changes and start applying whats up there to see what works | 17:46 |
| anteaya | Clint: have you a link to an example that you describe? | 17:46 |
| clarkb | mordred: socket.getaddrinfo(socket.gethostname(), None, socket.AF_INET6) | 17:46 |
| greghaynes | crinkle: That is what I was doing with making the new deploy host, basically | 17:46 |
| fungi | Clint: that sounds odd. git commits can't have multiple parents unless they're merge commits | 17:46 |
| clarkb | mordred: but that relies on /etc/hosts being setup properly | 17:46 |
| clarkb | rather than routes | 17:46 |
| *** dramalho_ has joined #openstack-infra | 17:46 | |
| fungi | Clint: multiple discrete parents i mean | 17:46 |
| jeblair | mordred: i don't know :) that's still a problem (i was focusing on one at a time) | 17:47 |
| crinkle | greghaynes: okay | 17:47 |
| clarkb | which is probably less reliable | 17:47 |
| *** dramalho has quit IRC | 17:47 | |
| crinkle | greghaynes: what about glean? are there patches in flight that i need from that? | 17:47 |
| SpamapS | jeblair: Right, it would be the latter. | 17:47 |
| clarkb | wait why would we want devstack hosts to talk to graphite? | 17:47 |
| *** dramalho_ has quit IRC | 17:47 | |
| pelix | zaro: checkout http://logs.openstack.org/42/223942/3/check/gate-jenkins-job-builder-python27/3d1e064/console.html compared to http://logs.openstack.org/07/226107/3/check/gate-jenkins-job-builder-python27/c4bcd99/console.html | 17:47 |
| *** glauco__ has joined #openstack-infra | 17:47 | |
| greghaynes | crinkle: https://review.openstack.org/#/c/209784/ will be needed | 17:47 |
| jeblair | clarkb: SpamapS is exploring the idea of recording devstack metrics in graphite. so at the end of each run, report 20-100 counters for that run. | 17:47 |
| *** nadya has quit IRC | 17:47 | |
| clarkb | I think any stats/logs/etc that you want from a job to go into $service goes through a system like subunit2sql and elasticserach | 17:47 |
| anteaya | Clint: I'm trying to picture a situation where one patch depends on two other patches but the two parent patches have no relationship | 17:48 |
| greghaynes | crinkle: also, both probably need work | 17:48 |
| openstackgerrit | Merged openstack-infra/project-config: Allow ironic-python-agent to have stable/managed releases https://review.openstack.org/229993 | 17:48 |
| clarkb | jeblair: right but you don't want them to talk directly, you use the existing pull method | 17:48 |
| pelix | first one has the log output generated from all the tests appearing in the console. Means that when a test fails, any useful log output will be interspersed with logs from working tests. | 17:48 |
| fungi | Clint: anyway, git-review should dtrt with a linear series of dependent commits | 17:48 |
| *** dizquierdo has quit IRC | 17:48 | |
| *** glauco__ has quit IRC | 17:48 | |
| jeblair | clarkb: yeah, i'm not arguing that. :) | 17:48 |
| *** geoffarnold has quit IRC | 17:48 | |
| SpamapS | mtreinish: hey, say that bit about gearman/zmq over here, and tag clarkb ;) | 17:48 |
| crinkle | greghaynes: fun | 17:48 |
| *** dramalho_ has joined #openstack-infra | 17:48 | |
| *** dramalho_ has quit IRC | 17:48 | |
| mordred | clarkb: we had not gotten to talking about the how of that part yet :) | 17:48 |
| Clint | anteaya: https://review.openstack.org/#/c/228656/ , i had to use Depends-On instead | 17:48 |
| *** glauco has joined #openstack-infra | 17:48 | |
| *** glauco has quit IRC | 17:48 | |
| Clint | fungi: the problem was that they weren't linear | 17:48 |
| clarkb | mordred: the more I htink about it the more I think the existing code is probably mots correct | 17:49 |
| jeblair | clarkb: but you have an idea! :) | 17:49 |
| mtreinish | SpamapS: clarkb is omnipresent (or just on qa too) he already saw it :) | 17:49 |
| *** dramalho_ has joined #openstack-infra | 17:49 | |
| SpamapS | clarkb: oh is there already something that could read a line out of the logs and turn that into statsd input? | 17:49 |
| *** glauco has joined #openstack-infra | 17:49 | |
| fungi | Clint: that sounds like a git limitation, not a git-review limitation. how did you represent the nonlinear dependency in git? | 17:49 |
| jeblair | clarkb: thinking maybe something like the elasticsearch/subunit2sql parser that post-processes jobs and inserts data into graphite? | 17:50 |
| *** glauco has quit IRC | 17:50 | |
| Clint | fungi: as a linear series of patches | 17:50 |
| mtreinish | SpamapS: look at: http://git.openstack.org/cgit/openstack-infra/puppet-log_processor/tree/files/log-gearman-client.py | 17:50 |
| jeblair | clarkb: maybe a 3rd thing, or maybe add it to one of the existing ones? | 17:50 |
| clarkb | SpamapS: there is already a thing that watches for job complete events and submits gearman jobs based on that | 17:50 |
| mtreinish | SpamapS: and an example worker: http://git.openstack.org/cgit/openstack-infra/puppet-subunit2sql/tree/files/subunit-gearman-worker.py | 17:50 |
| *** dramalh__ has joined #openstack-infra | 17:50 | |
| clarkb | jeblair: ya something like that | 17:50 |
| pelix | zaro: you might notice that some deprecation warnings from libraries JJB uses were hidden in the log output previously as well | 17:50 |
| fungi | Clint: so if the series of patches was linear, git-review should have dealt with it fine | 17:50 |
| pelix | see http://logs.openstack.org/07/226107/3/check/gate-jenkins-job-builder-python27/c4bcd99/console.html#_2015-09-30_15_32_35_839 | 17:50 |
| clarkb | as an alternative I think you can have logstash do graphite output too | 17:50 |
| *** jsavak has quit IRC | 17:51 | |
| Clint | fungi: it did so by resubmitting the first two in a dependency chain | 17:51 |
| *** dramalh__ has quit IRC | 17:51 | |
| clarkb | so you would add a second output based on things that it piped through but that ties us more tightly to logstash which is :/ | 17:51 |
| *** glauco has joined #openstack-infra | 17:51 | |
| *** glauco has quit IRC | 17:51 | |
| clarkb | however if you do it that way then other deployments of openstack can use it easily | 17:51 |
| *** dramalh__ has joined #openstack-infra | 17:51 | |
| *** dramalh__ has quit IRC | 17:51 | |
| *** jsavak has joined #openstack-infra | 17:51 | |
| *** pelix has quit IRC | 17:51 | |
| fungi | Clint: i'm losing chronology on this i guess. i thought you said you had already linearized them | 17:51 |
| *** dramalh__ has joined #openstack-infra | 17:52 | |
| *** jordanP has quit IRC | 17:52 | |
| SpamapS | So if there's already infra in place for this, and we're just adding counters.. do I need an infra spec? or just submit a change which adds the counters to statsd's configuration? | 17:52 |
| *** dramalh__ has quit IRC | 17:52 | |
| SpamapS | I'm working on a QA spec to add more counters and this output. | 17:52 |
| fungi | Clint: were you trying to establish dependence between changes in different local branches? (if so, that's a merge) | 17:52 |
| *** dramalho has joined #openstack-infra | 17:52 | |
| clarkb | SpamapS: I think you just add a change to emit the counters, the system is pretty well established I don't think it needs a spec | 17:52 |
| SpamapS | clarkb: ^5 thanks | 17:52 |
| greghaynes | Not sure if it matters, but the stats will be recorded as happening when they are emitted if youre going through statsd I think | 17:53 |
| jeblair | SpamapS: basically the new infra component is something that processes gearman jobs from the log-processor, reads the counters from devstack logs, and emits to statsd. | 17:53 |
| *** glauco_ has joined #openstack-infra | 17:53 | |
| *** glauco_ has quit IRC | 17:53 | |
| *** dramalho_ has quit IRC | 17:53 | |
| jeblair | greghaynes, SpamapS: good point, maybe skip statsd and emit directly to graphite? | 17:53 |
| *** dprince has quit IRC | 17:54 | |
| jeblair | greghaynes, SpamapS: though honestly, not sure it matters that much? | 17:54 |
| greghaynes | yea, if youre going for long term trends then it doesnt matter too much | 17:54 |
| Clint | fungi: only so that the child patch would work. the other two could have gone in independently. | 17:54 |
| jeblair | greghaynes, SpamapS: i mean, we're basically trying to trend over time, right? whether $time is now or 10 minutes ago, probably not a big deal? | 17:54 |
| *** glauco_ has joined #openstack-infra | 17:55 | |
| *** glauco_ has quit IRC | 17:55 | |
| *** tonytan4ever has joined #openstack-infra | 17:56 | |
| *** dramalho has quit IRC | 17:56 | |
| *** glauco has joined #openstack-infra | 17:56 | |
| *** glauco has quit IRC | 17:56 | |
| fungi | Clint: ahh, yes so that's still a git limitation and not a git-review one | 17:56 |
| *** dramalho has joined #openstack-infra | 17:56 | |
| *** dramalho has quit IRC | 17:56 | |
| fungi | Clint: or rather, our insistence on not reviewing gratuitous merge commits | 17:57 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 17:57 |
| *** glauco_ has joined #openstack-infra | 17:57 | |
| *** dramalho has joined #openstack-infra | 17:57 | |
| *** glauco_ has quit IRC | 17:57 | |
| *** dramalho has quit IRC | 17:57 | |
| *** Zhongjun has quit IRC | 17:57 | |
| fungi | Clint: regardless, git-review shouldn't do anything out of the ordinary in that case. either you create a linear patch series or you don't | 17:57 |
| *** hdd has quit IRC | 17:57 | |
| Clint | fungi: i maintain that it everything would have worked out fine if it hadn't resubmitted the earlier patches | 17:58 |
| Clint | s/it // | 17:58 |
| *** dramalho has joined #openstack-infra | 17:58 | |
| *** dramalho has quit IRC | 17:58 | |
| fungi | Clint: that sounds like a rebase cleanliness issue. when rebasing only alter the commits which need altering (either to change their content or because a parent changed) | 17:58 |
| *** glauco has joined #openstack-infra | 17:58 | |
| *** ociuhandu has joined #openstack-infra | 17:58 | |
| *** glauco has quit IRC | 17:58 | |
| *** SumitNaiksatam has joined #openstack-infra | 17:59 | |
| fungi | git-review will mention pushing all of them again, but will only actually upload the ones whose commit ids changed | 17:59 |
| Clint | fungi: the commit ids changed because it rebased one patch onto the other | 17:59 |
| fungi | yep, that changes the commit id | 17:59 |
| fungi | the parent commit id is part of the data which goes into calculating the git commit id | 17:59 |
| Clint | yup | 18:00 |
| fungi | i'm still not understanding what's wrong | 18:00 |
| *** ajmiller has quit IRC | 18:00 | |
| fungi | if the parent changes you need to push that as a new patchset, because new commit id | 18:00 |
| Clint | the parent didn't need to change | 18:00 |
| *** ajmiller has joined #openstack-infra | 18:01 | |
| fungi | if the parent didn't change then it wouldn't have been uploaded | 18:01 |
| Clint | the final patch could have been rebased onto master in whatever order the other patches landed in master | 18:01 |
| openstackgerrit | Merged openstack-infra/project-config: Add new project pymod2pkg to Packaging-RPM team https://review.openstack.org/227891 | 18:02 |
| Clint | the ordering was forced by a reverse dependency unnecessarily | 18:02 |
| fungi | oh, so ultimately the issue here is that we disallow pushing merge commits | 18:03 |
| *** doug-fish is now known as duct-tape-wannab | 18:03 | |
| fungi | because there's no actual way to represent n-way nonlinear dependencies in git without merge commits | 18:03 |
| *** duct-tape-wannab is now known as ducttape-wannabe | 18:03 | |
| fungi | however, not using git-review doesn't get around that limitation | 18:04 |
| Clint | fungi: i gotta hop on a plane, we can argue about this later or possibly never | 18:04 |
| fungi | heh, okay | 18:04 |
| *** mrmartin has quit IRC | 18:05 | |
| *** Piet has joined #openstack-infra | 18:05 | |
| *** rossella_s has quit IRC | 18:06 | |
| openstackgerrit | Merged openstack-infra/shade: Dont throw exception on missing service https://review.openstack.org/229664 | 18:06 |
| fungi | i still suspect there's some fundamentally incorrect assumption there about what git-review is or isn't doing. what you're describing isn't going to be able to get uploaded to gerrit regardless of whether you use git-review | 18:06 |
| *** rossella_s has joined #openstack-infra | 18:06 | |
| *** ivar-lazzaro has joined #openstack-infra | 18:07 | |
| *** shardy is now known as shardy_afk | 18:08 | |
| *** nelsnelson has joined #openstack-infra | 18:08 | |
| openstackgerrit | Merged openstack-infra/project-config: Enable ceilometer devstack plugin for tempest jobs https://review.openstack.org/229940 | 18:08 |
| *** BharatK has quit IRC | 18:09 | |
| *** ivar-lazzaro has quit IRC | 18:09 | |
| jeblair | fungi, yolanda: what's the status of nodepool re secure.conf / restart, etc? | 18:09 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 18:09 |
| clarkb | jeblair: the patch merge conflicted because it requires every test be changed | 18:09 |
| *** dprince has joined #openstack-infra | 18:09 | |
| jeblair | neato | 18:10 |
| clarkb | jeblair: I have suggested that instead of chasing rebases the tests be fixed to hide that in the base class | 18:10 |
| clarkb | its all boiler plate anyways that doesn't change between tests | 18:10 |
| jeblair | clarkb: ++ | 18:10 |
| *** tsekiyama has joined #openstack-infra | 18:10 | |
| *** ivar-lazzaro has joined #openstack-infra | 18:10 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Provide shortcut around has_service https://review.openstack.org/230032 | 18:10 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Fix mis-named has_service entry https://review.openstack.org/230060 | 18:10 |
| *** amotoki has quit IRC | 18:10 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Refactor grafana.py test coverage https://review.openstack.org/230061 | 18:11 |
| openstackgerrit | Merged openstack-infra/puppet-cgit: Fix git-daemon refresh logic https://review.openstack.org/228038 | 18:11 |
| *** Sukhdev has joined #openstack-infra | 18:12 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/grafyaml: Clean up tox.ini https://review.openstack.org/224866 | 18:12 |
| *** asselin__ has joined #openstack-infra | 18:13 | |
| *** kushal has joined #openstack-infra | 18:13 | |
| jeblair | mordred: the nodepool shade change now depends on a shade change that also relates to an occ change and later potential shade improvement. do we just need to land the one shade change (229140), or do we need to worry about any others? | 18:13 |
| *** odgrim has joined #openstack-infra | 18:14 | |
| mordred | jeblair: I think we need to land several changes - let me pull together a list for you | 18:14 |
| jeblair | mordred: ok. my goal is to restart nodepool with the shade change today. as you put together that list, do you think that's achievable? :) | 18:15 |
| mordred | jeblair: I do - we need to get occ landed/cut | 18:16 |
| jeblair | https://etherpad.openstack.org/p/nodepool-shade | 18:16 |
| *** asselin_ has quit IRC | 18:17 | |
| *** moravec has quit IRC | 18:18 | |
| *** pvaneck has joined #openstack-infra | 18:18 | |
| clarkb | so apparently pep8 doens't warn you about undefined names | 18:18 |
| clarkb | or flake8 rather | 18:18 |
| clarkb | (I have been testing pep8 locally but not unittests bceause don't have a mysql and keep getting bitten on that. This'll learn me) | 18:19 |
| jeblair | clarkb: pep8 is disabled on nodepool because of the 'select' line in tox.ini. | 18:19 |
| clarkb | jeblair: its completely disabled? | 18:19 |
| * clarkb looks | 18:20 | |
| jeblair | clarkb: the only thing we are testing with pep8 is H231 | 18:20 |
| clarkb | that would explain it | 18:20 |
| jeblair | i's been like that for a year, and there are like 4 violations or something :) | 18:20 |
| jeblair | still, i would like the flake8 output and support fixing it; just hasn't bubbled up. | 18:21 |
| clarkb | kk | 18:21 |
| *** HeOS has joined #openstack-infra | 18:21 | |
| *** nelsnelson has quit IRC | 18:21 | |
| jeblair | mordred: do we need 230030? | 18:22 |
| clarkb | jeblair: I don't think its required for ovh | 18:22 |
| *** moravec has joined #openstack-infra | 18:22 | |
| openstackgerrit | YAMAMOTO Takashi proposed openstack-infra/project-config: tap-as-a-service: Add python3 job https://review.openstack.org/230063 | 18:22 |
| mordred | jeblair: no, we do not | 18:23 |
| mordred | jeblair: it's a nice-to-have | 18:23 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config: puppet: split integration job in 2 https://review.openstack.org/230064 | 18:23 |
| jeblair | mordred: ya, maybe we can get it in the same shade release? if not, maybe not hold up the process on it? | 18:24 |
| mordred | yah. that was what I was thinking | 18:24 |
| openstackgerrit | Christian Berendt proposed openstack/os-testr: Fix syntax of the README file https://review.openstack.org/230065 | 18:25 |
| *** nelsnelson has joined #openstack-infra | 18:25 | |
| *** jsavak has quit IRC | 18:26 | |
| docaedo | fungi: regarding the cert for app catalog, you wrote "in hiera now as apps_ssl_cert_file_contents, apps_ssl_chain_file_contents and apps_ssl_key_file_contents" | 18:26 |
| *** jsavak has joined #openstack-infra | 18:26 | |
| docaedo | fungi: but I'm not sure how I'm supposed to reference that, is there a site you can tell me to look at that would be a good example? | 18:26 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 18:26 |
| *** maishsk has joined #openstack-infra | 18:26 | |
| mordred | Shrews: if you get a sec, could you help us out with reviews on the stuff in here: https://etherpad.openstack.org/p/nodepool-shade ? | 18:26 |
| docaedo | (i.e. one of the other sites in infra repo) | 18:27 |
| fungi | docaedo: those are the hiera keys you would reference in openstack-infra/system-config:manifests/site.pp | 18:27 |
| fungi | docaedo: within the node definition for the apps.openstack.org server | 18:27 |
| *** marzif has joined #openstack-infra | 18:27 | |
| Shrews | mordred: i've been trying to figure out which ones are urgent. looks like all of them | 18:27 |
| fungi | basically pass those as class variables to what's being instantiated there so that they can be consumed by the puppet module doing the apps-site setup | 18:27 |
| mordred | Shrews: ALL THINGS ARE URGENT AND HIGHEST PRIORITY PLZ FIX NOW KTHX | 18:28 |
| *** tsekiyama has quit IRC | 18:28 | |
| clarkb | BLOCKER! | 18:28 |
| Shrews | mordred: oh, time for a coffee break then. be back next week | 18:28 |
| docaedo | fungi: I think I follow | 18:28 |
| fungi | docaedo: you should see numerous examples in that file | 18:29 |
| *** sridhar_ram has quit IRC | 18:30 | |
| docaedo | fungi: thanks yeah looking now and seeing things that make some little bit of sense | 18:30 |
| openstackgerrit | Dolph Mathews proposed openstack/requirements: Add freezegun to global requirements https://review.openstack.org/228603 | 18:31 |
| *** openstackgerrit has quit IRC | 18:31 | |
| *** openstackgerrit has joined #openstack-infra | 18:31 | |
| *** angdraug has quit IRC | 18:32 | |
| SpamapS | jeblair: correct, the time that the count happens is not as important as the general series. | 18:32 |
| SpamapS | In fact, I'm wondering if we only want to emit this particular stat _on success_ | 18:32 |
| SpamapS | set of stats rather | 18:32 |
| SpamapS | as in, this is where the latest patch set to land put us. | 18:32 |
| jeblair | SpamapS: probably so | 18:33 |
| SpamapS | yes actually, it can only be on success | 18:33 |
| *** tjones has quit IRC | 18:33 | |
| SpamapS | otherwise there'll be tons of gate churn garbage | 18:33 |
| SpamapS | the idea is to get a series of data with trends, and then be able to go ahead and notice relatively stable counters, and put hard upper and lower bounds on them. | 18:34 |
| clarkb | you will also have to account for $cloud | 18:34 |
| jeblair | SpamapS: makes sense | 18:34 |
| clarkb | and $driverofvariousthings | 18:34 |
| jeblair | clarkb: hopefully by account for $cloud, we can just set appropriate bounds | 18:34 |
| SpamapS | clarkb: thats an interesting aspect of these, and one I have been thinking about. | 18:34 |
| jeblair | clarkb: i do not think we should have "expect this metric to take this long on cloud X" | 18:35 |
| SpamapS | I think for some counters, cloud should not matter. | 18:35 |
| SpamapS | like, the number of InnoDB rows read, should be stable. | 18:35 |
| clarkb | jeblair: except from data we know that it matters | 18:35 |
| *** jsavak has quit IRC | 18:35 | |
| clarkb | jeblair: theneutron full job averages ~50 minutes on one cloud nad ~72 on another according to sdagues numbers yesterday | 18:35 |
| *** yamamoto has quit IRC | 18:35 | |
| *** yamamoto has joined #openstack-infra | 18:36 | |
| greghaynes | Maybe break them up into "Cloud A, Cloud B" rather than cloud name | 18:36 |
| *** jsavak has joined #openstack-infra | 18:36 | |
| SpamapS | also as much as possible the point of this exercise is to identify performance cost metrics that do _not_ depend on the _speed_ of the underlying infrastructure. | 18:36 |
| clarkb | I don't think you want your counters to be collapsing that as such a large separation will make the data useless | 18:36 |
| clarkb | or at least less useful in aggregate because one cloud can drown out the other | 18:36 |
| SpamapS | greghaynes: yeah there may be counters that are entirely stable based on whatever is under the hood, and those would probably good to leverage. | 18:36 |
| jeblair | clarkb: i think SpamapS's approach is a good one. i'm heavily resistent to encoding per-cloud performance data though. | 18:37 |
| greghaynes | SpamapS: Yea, I think thats what clarkb is saying | 18:37 |
| SpamapS | First pass, I'd discard any that were unstable across clouds. | 18:37 |
| greghaynes | but yes, politics | 18:37 |
| clarkb | sure I don't want to make it a contest | 18:37 |
| SpamapS | discard from the upper/lower bounds I should say | 18:37 |
| jeblair | oh i'm not concerned about politics... | 18:37 |
| clarkb | but if you are measuring time to boot VM that will have drastically different numbers | 18:37 |
| jeblair | i just want there to be no bar to adding new clouds. | 18:37 |
| clarkb | and you won't really see variation in one or the other because the large set from the other cloud will drown it out | 18:37 |
| jeblair | clarkb: do we know why it varies so far in the neutron tests (but presumably not in others)? | 18:38 |
| SpamapS | time based metrics won't work for this approach | 18:38 |
| SpamapS | because ideally, one can iterate on driving these counters down on their laptop. | 18:38 |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-diskimage_builder: Fix apt issue. https://review.openstack.org/220505 | 18:38 |
| clarkb | jeblair: the current theory is that neutrons ipv6 tests which boot 2 nodes each are more cpu intensive (via qemu) and some clouds do poorly at that | 18:38 |
| jeblair | SpamapS: right 'counters' vs 'timers' in statsd/graphite speak | 18:38 |
| SpamapS | so if you have a patch set whose purpose is to reduce database rows read by making better use of indexes, for instance.. | 18:38 |
| *** nadya has joined #openstack-infra | 18:39 | |
| SpamapS | that should measure mostly the same on all clouds and your laptop | 18:39 |
| greghaynes | Seems like a neat idea | 18:39 |
| Shrews | mordred: jeblair: https://review.openstack.org/229941 is a pretty major API breaking change | 18:39 |
| clarkb | oh if you are just looking at that I am not sure why you need the gate data? its should be stable right? | 18:39 |
| clarkb | eg anyone can generate that number and its fixed | 18:39 |
| SpamapS | the lower level stuff will likely have too much collateral damage from performance hiccups to be useful, but I'm hopeful maybe basic I/O numbers will be somewhat stable too, like blocks read/written. | 18:39 |
| *** tonytan4ever has quit IRC | 18:40 | |
| SpamapS | clarkb: I need the gate to prove the stability of the number. | 18:40 |
| fungi | yeah, we've seen wide performance swings within a single provider just depending on shared load distribution/spikes across tenants | 18:41 |
| clarkb | SpamapS: thats one way to do it yes | 18:41 |
| *** sergmelikyan has quit IRC | 18:41 | |
| SpamapS | clarkb: we could also ask everybody to run devstack with this turned on and confirm.. I just feel like the robots do that better than the humans. | 18:41 |
| clarkb | but it seems like the most difficult to deal with way | 18:41 |
| clarkb | SpamapS: no no using robots is fine | 18:41 |
| fungi | so counting operations seems reasonable. timing operations much less so | 18:41 |
| mtreinish | SpamapS: you should take a look at some of the graphs from subunit2sql tings are super variable | 18:41 |
| jeblair | Shrews: my understanding from mordred was that nothing used that yet, but maybe he can expand. | 18:42 |
| SpamapS | Maybe there are other robots that are simpler to setup? | 18:42 |
| clarkb | but if you have to filter for success vs failure and potentially clouds and specific pipelines you are just spending effort | 18:42 |
| clarkb | instead use the periodic jobs or similar | 18:42 |
| SpamapS | mtreinish: can you give me a starting point for that. I'm embarrassed to say I've never looked. | 18:42 |
| mtreinish | SpamapS: yes, I can. One sec | 18:42 |
| Shrews | mordred: jeblair: uh, nevermind... i didn't look far enough ahead | 18:42 |
| clarkb | SpamapS: yes we run tests once a day for all the things they are a much better source of this type of data | 18:42 |
| mtreinish | SpamapS: http://blog.kortar.org/?p=212 | 18:42 |
| *** dzamboni has quit IRC | 18:42 | |
| mtreinish | I wrote a blog post on how to get started :) | 18:43 |
| mtreinish | clarkb: is it all the things, the jobs in periodic is a weird set of things | 18:44 |
| *** yamamoto has quit IRC | 18:44 | |
| greghaynes | SpamapS: I have an argument for recording timing data id like to make when it wouldnt be derailing technical discussion | 18:44 |
| clarkb | mtreinish: it should be all the things + super set of things that want to be periodic only | 18:44 |
| *** amuller_afk is now known as amuller | 18:45 | |
| *** tsekiyama has joined #openstack-infra | 18:45 | |
| *** marzif has quit IRC | 18:45 | |
| mtreinish | ah, ok. where do the results get put pushed for the other stuff. I've only ever seen the things which are explicitly in the periodic pipeline | 18:45 |
| * mtreinish probably should get back to the dashboard to solve that problem | 18:45 | |
| clarkb | mtreinish: its the same thing | 18:46 |
| SpamapS | clarkb: ok, so the daily periodics would work, but would take a lot longer to produce enough data. | 18:46 |
| *** yamamoto has joined #openstack-infra | 18:46 | |
| clarkb | SpamapS: yes but the data isn't known to be compromised | 18:46 |
| *** tjones has joined #openstack-infra | 18:46 | |
| SpamapS | clarkb: is there something significantly simpler about using that data that makes it worth waiting 30 or so days to get trends? | 18:46 |
| clarkb | SpamapS: yes its not compromised | 18:46 |
| greghaynes | You can also make periodic jobs that run more often | 18:47 |
| SpamapS | Is it on physical dedicated hardware, or like, just that it's one cloud? | 18:47 |
| clarkb | SpamapS: you know for all of the jobs that run there that all of the code tested passed all of its tests at once | 18:47 |
| clarkb | SpamapS: no no just compared to the gate. A test may pass in the gate but be completely broken and some other tests for the change would fail detecting that | 18:47 |
| clarkb | SpamapS: but you would load up that data anyways even though its known bad | 18:47 |
| openstackgerrit | Michael Johnson proposed openstack/requirements: Exclude coverage 4.0 https://review.openstack.org/228581 | 18:48 |
| SpamapS | clarkb: good poitn | 18:48 |
| SpamapS | point | 18:48 |
| *** julim has quit IRC | 18:48 | |
| SpamapS | unless we can somehow delay recording until after commit? | 18:48 |
| SpamapS | after merge I guess I should say | 18:48 |
| SpamapS | but that sounds like "moar work" | 18:48 |
| clarkb | ya you could set it up as a set of zuul post jobs that proxy the work into graphite or something | 18:49 |
| openstackgerrit | James E. Blair proposed openstack-infra/shade: Return IPv6 address for interface_ip on request https://review.openstack.org/230030 | 18:49 |
| clarkb | there are other ways of tackling that | 18:49 |
| SpamapS | And to greghaynes's point, we can just make a job that runs every 4 hours for a week and get a month's worth of data, then back it off. | 18:49 |
| SpamapS | I mean | 18:49 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/shade: Stop checking for E129 https://review.openstack.org/230074 | 18:50 |
| fungi | jeblair: your 230030 seems to have angered the worshippers of E129. we may want to evict them via tox.ini? ^ | 18:50 |
| *** tonytan4ever has joined #openstack-infra | 18:50 | |
| SpamapS | I could also just go backwards | 18:50 |
| jeblair | fungi: i have banished them, thanks! | 18:50 |
| fungi | oh! excellent | 18:50 |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-diskimage_builder: Fix apt issue. https://review.openstack.org/220505 | 18:50 |
| SpamapS | set up a test that spins up a node and runs the tests with HEAD^0, then HEAD^1 .. | 18:50 |
| mtreinish | SpamapS: that was always on the table to make periodic stuff happen more frequently | 18:51 |
| SpamapS | though I can't go backwards easily with pip | 18:51 |
| SpamapS | and also HEAD^0 of what.. ;) | 18:51 |
| mtreinish | the pre-req for that was getting the results dashboard up | 18:51 |
| SpamapS | more like I could just have git checkout based on time | 18:51 |
| SpamapS | Actually | 18:52 |
| greghaynes | SpamapS: dependencies make that even more wierd | 18:52 |
| SpamapS | I'm being silly | 18:52 |
| *** sergmelikyan has joined #openstack-infra | 18:52 | |
| SpamapS | can just run with HEAD _now_ | 18:52 |
| SpamapS | 100 times | 18:52 |
| *** sergmelikyan has quit IRC | 18:52 | |
| jeblair | clarkb, SpamapS: there are lots of interesting suggestions here... but is there anything wrong in _practice_ with just using gate+SUCCESS ? | 18:52 |
| SpamapS | and get _actual_ stability numbers | 18:52 |
| jeblair | we use that all the time for similar things... | 18:52 |
| jeblair | (e-r graphs, for instance) | 18:53 |
| SpamapS | jeblair: clarkb brought up a good point which is that if we wanted to just re-use the current mechanism we'd be graphing the number for gate fails too. | 18:53 |
| *** AJaeger has joined #openstack-infra | 18:53 | |
| jeblair | yes, it might record a value for a change that didn't merge, but in aggregate, it's pretty right) | 18:53 |
| *** ashleighfarnham has joined #openstack-infra | 18:53 | |
| SpamapS | the specific job may be successful | 18:53 |
| mtreinish | jeblair: I agree, that should be fine | 18:53 |
| *** ZZelle_ has quit IRC | 18:53 | |
| *** angdraug has joined #openstack-infra | 18:53 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 18:53 | |
| docaedo | fungi: should I update the puppet-apps_site commit to include "depends on" for the system-config commit I'm about to make? Not sure how I should reference between the two | 18:53 |
| jeblair | SpamapS: jobs that filed to merge for some reason after having been approved and after having passed all of the gate tests previously | 18:53 |
| SpamapS | right... | 18:54 |
| jeblair | SpamapS: in practice, i don't think it's "bad" data for the purpose under discussion | 18:54 |
| *** glauco_ has joined #openstack-infra | 18:54 | |
| anteaya | docaedo: do you need one patch to make tests pass on the other? | 18:54 |
| fungi | docaedo: other way around. the system-config change to pass those variables needs to depends-on the puppet-apps_site change which consumes them so that testing is validated | 18:54 |
| SpamapS | There's also the psychological benefit of having it this way | 18:54 |
| mtreinish | jeblair, SpamapS: I treat the subunit2sql data that way | 18:54 |
| SpamapS | which is that you can point to a graph and say "This is where the trend upward started" | 18:54 |
| jeblair | SpamapS: if you were to try to use it to track down the exact time something changed, it would be insufficient, but it's already insufficient for that and not designed for that :) | 18:54 |
| jeblair | SpamapS: yeah, you will be able to say "this is _about_ where the trend upward started" :) | 18:55 |
| docaedo | fungi: thanks | 18:55 |
| mtreinish | jeblair: which is normally good enough :) | 18:55 |
| SpamapS | the one time it's ok to wave your hands... at a general area of a graph | 18:55 |
| jeblair | SpamapS: with gate data, that's probably a 1 hour window. expanding that to a 4h window with periodic jobs doesn't improve that :) | 18:55 |
| *** nadya has quit IRC | 18:56 | |
| openstackgerrit | Christopher Aedo proposed openstack-infra/system-config: Add SSL support to App Catlog site https://review.openstack.org/230078 | 18:56 |
| jeblair | SpamapS: post jobs would improve it, for sure (except they are lower priority and often run much later than gate jobs, so you'd have to correlate a couple of sources of data) | 18:56 |
| greghaynes | also fails should be noise-like so it will filter out fine | 18:56 |
| *** glauco__ has joined #openstack-infra | 18:56 | |
| greghaynes | I would be very surprised if fails always were slower or faster | 18:57 |
| clarkb | jeblair: thats a good pont about post | 18:57 |
| fungi | greghaynes: that assumes a high-enough signal-to-noise ratio for you to be able to find the signal | 18:57 |
| jeblair | SpamapS: anyway, point being, it's fine if we want to do something else for some reason, but i would not eliminate gate based just on the false-success problem; it's an existing pattern that works for similar situations | 18:57 |
| greghaynes | fungi: just need a low pass filter ;) | 18:57 |
| *** notnownikki has joined #openstack-infra | 18:57 | |
| greghaynes | :p | 18:57 |
| SpamapS | jeblair: no, I think I got side-tracked by the word "compromised" .. ;) | 18:57 |
| clarkb | jeblair: well everywhere else we do it we want to look at fail data | 18:58 |
| fungi | greghaynes: we get plenty of failure modes where performance is degraded and the job runs longer, even possibly reaching the configured timeout | 18:58 |
| SpamapS | but... it's an acceptable compromise. | 18:58 |
| clarkb | jeblair: this is the first time someone has said we only want successful data that I know of | 18:58 |
| jeblair | SpamapS: you need to leave _something_ for other researchers to pick apart in your paper, yeah? | 18:58 |
| SpamapS | we could still do fail data | 18:58 |
| SpamapS | It might come in handy. | 18:58 |
| *** glauco_ has quit IRC | 18:58 | |
| greghaynes | fungi: yea, so it might skew up. Still it shouldnt have a trend I hope | 18:58 |
| mtreinish | greghaynes: yeah, it's very hard to find a trend like that in the data | 18:58 |
| SpamapS | the graph isn't harder to make if we say success only, right? | 18:58 |
| mtreinish | but it doesn't mean that there isn't one, like fungi said it's really noisy | 18:58 |
| *** signed8bit_ZZZzz is now known as signed8bit | 18:59 | |
| SpamapS | jeblair: Yeah the game is no fun if there aren't holes. ;) | 18:59 |
| jeblair | clarkb: i think mtreinish pointed out subunit2sql is fairly similar here? | 18:59 |
| clarkb | jeblair: I think subunit2sql looks at fails | 18:59 |
| mtreinish | clarkb: it looks at everything | 18:59 |
| * SpamapS is reading mtreinish's blog post now | 18:59 | |
| clarkb | right so thats different | 18:59 |
| mtreinish | clarkb: but for performance graphing I filter on success | 18:59 |
| mtreinish | for failure rate stuff obviously not | 18:59 |
| docaedo | fungi: I think 230078 is right, just let me know, gracias | 19:00 |
| fungi | greghaynes: in fact, quick fails are mostly the domain of setup-related problems. if setup is successful, most jobs still continue to run all their tests | 19:00 |
| *** moravec has quit IRC | 19:00 | |
| fungi | even if most of them are failing | 19:00 |
| clarkb | hrm I may have taken the wrong approach with the event handler tests /me tries again | 19:00 |
| *** jsavak has quit IRC | 19:01 | |
| *** jsavak has joined #openstack-infra | 19:01 | |
| fungi | greghaynes: and for many (if not most) tests, the failure is not detected until some assert at the end, so runtime is probably similar in many cases | 19:01 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/release-tools: do not update constraints list on each new release https://review.openstack.org/230080 | 19:01 |
| greghaynes | fungi: Yea. I think it would only be an issue if failures had a trend of their own... but we kinda just need to play with the data I think | 19:02 |
| jeblair | in summary, gate gets you the most amount of data, closest to HEAD in real-time, with the caveat that sometimes the data reported will be based on a change that didn't actually land. post gets you a guarantee that the change is in head, but sometimes it will fail, so you will miss some changes. periodic is what it is. :) | 19:02 |
| fungi | we have proof that periodic jobs can fail too ;_ | 19:02 |
| jeblair | in fact, some of them never succeed! ;) | 19:02 |
| clarkb | fungi: yes but that would could arguably be considered valid data | 19:02 |
| anteaya | ha ha ha | 19:02 |
| fungi | disappointingly many in fact | 19:02 |
| fungi | clarkb: true enough | 19:03 |
| SpamapS | It seems to me that having the counters for failures won't compromise the goal of identifying and gating on stable counters for success, but it might add useful data to correlate and debug problems. | 19:03 |
| *** hashar has joined #openstack-infra | 19:03 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add new graph type to show run_time variance by metadata https://review.openstack.org/210569 | 19:04 |
| *** gordc has quit IRC | 19:04 | |
| fungi | if you keep separate buckets for fail data and pass data that might make it easier to determine whether the data from failures is comparable/useful | 19:04 |
| *** tjones has quit IRC | 19:04 | |
| *** stevemar has joined #openstack-infra | 19:04 | |
| fungi | rather than comingling and assuming the failures don't compromise your data integrity | 19:05 |
| jeblair | SpamapS: keep in mind we're talking 2 kinds of failures here: failure of the job in question (easy to detect), and failure of the change to merge because of something external to the job in question (hard to detect) | 19:05 |
| jeblair | SpamapS: and yeah, the marginal cost to bucket "failure or success of the job in question" is low and easy to do. i'd recommend that. | 19:05 |
| jeblair | SpamapS: though it's quite possible that we never get any good data from failures. :) | 19:06 |
| *** ducttape-wannabe is now known as doug-fish | 19:06 | |
| jeblair | filtering for "success of the job in question" only works too | 19:06 |
| greghaynes | SpamapS: ok, my piece on timing data - what youre trying to do by measuring row accesses/ reads/etc is build up a model of perfomance using some other measurements. That lets you drive down on the model, which is nice, but you also need to be able to answer the question of whether your model reflects reality and how well. You dont need precise data for that, just an even more general trend you can | 19:07 |
| greghaynes | try and correlate, but you definitely want to try and answer that question too. | 19:07 |
| greghaynes | wow I got a two parter | 19:07 |
| jeblair | clarkb, fungi: would you mind reviewing https://review.openstack.org/226751 ? | 19:07 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 19:08 |
| jeblair | oh | 19:08 |
| jeblair | morded uploaded a new patchset there and didn't say why :/ | 19:08 |
| greghaynes | SpamapS: so tl;dr I agree that driving down on timing is a bad idea because of variance in the data, but it is a requriement for ensuring that your model is at all sane | 19:08 |
| mordred | jeblair: wait - what did I do? | 19:08 |
| *** jsavak has quit IRC | 19:08 | |
| clarkb | also not sure what recheck crd is supposed to do? | 19:08 |
| jeblair | mordred: you updated 226751 after it passed tests and i don't know why | 19:08 |
| clarkb | the unittests consume shade from releases | 19:09 |
| *** jsavak has joined #openstack-infra | 19:09 | |
| clarkb | oh that expains it | 19:09 |
| jeblair | clarkb: it's supposed to run the tests again | 19:09 |
| clarkb | jeblair: ya I see that the unitests were passing and only integration didnt | 19:09 |
| clarkb | but latest fails all three | 19:09 |
| *** ociuhandu has quit IRC | 19:09 | |
| jeblair | clarkb: well, latest failed because of a merge failure so we have no results | 19:09 |
| jeblair | clarkb: thus the recheck | 19:09 |
| mordred | jeblair: the dvsm error it had was a real error I was concerned about - so I added more debugging lines to try to figure out why we were seeing it | 19:10 |
| mordred | jeblair: specifically, this: http://logs.openstack.org/51/226751/20/check/gate-dsvm-nodepool/e77cb88/logs/screen-nodepool.txt.gz#_2015-10-01_00_59_14_404 | 19:11 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/release-tools: Rework translation-cleanup https://review.openstack.org/227454 | 19:11 |
| mordred | jeblair: touches the area we're poking at and would be bad if it didn't work inproduction | 19:11 |
| *** ZZelle_ has joined #openstack-infra | 19:11 | |
| mordred | dhellmann: around? could you give us a hand with https://review.openstack.org/230085 ? | 19:12 |
| jeblair | mordred, clarkb: do we run shade in zuul-cloner with nodepool? | 19:12 |
| jeblair | or, well, do we run shade from zuul? | 19:12 |
| jeblair | i guess that's how we should say that? :) | 19:13 |
| mordred | jeblair: I'm not sure - looking | 19:13 |
| jeblair | so, trying again: mordred, clarkb: do we run shade from zuul in the nodepool dsvm job? | 19:13 |
| SpamapS | greghaynes: Yes I agree on your thoughts that time based things are necessary, especially when we are working on things like response time which will depend heavily on the parallelism and concurrency of the system. | 19:14 |
| mordred | jeblair: it does not appear so | 19:14 |
| *** glauco_ has joined #openstack-infra | 19:14 | |
| mordred | jeblair: export PROJECTS="openstack-infra/nodepool $PROJECTS" | 19:14 |
| *** glauco_ has quit IRC | 19:14 | |
| clarkb | jeblair: no | 19:14 |
| jeblair | mordred: i agree | 19:14 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/project-config: puppet: split integration job in 2 https://review.openstack.org/230064 | 19:14 |
| SpamapS | greghaynes: this approach doesn't do anything to stand in the way of that, it's just us focusing on things that we can prevent from backsliding easily. | 19:14 |
| SpamapS | greghaynes: we can get trends on time based, but I'm pretty sure we're never going to be able to gate on it. | 19:14 |
| jeblair | mordred: okay, so your change to 226751 won't do anything anyway -- are we basically back at square 1 with not knowing what caused the error in ps20? | 19:15 |
| *** hdd has joined #openstack-infra | 19:15 | |
| mtreinish | SpamapS: yeah it's not really feesible to gate on anything based on trend date because of the inherent noise | 19:15 |
| greghaynes | SpamapS: Yes, my thinking though is you dont ever want to gate on it but you want it to be there so you can go back and evaluate whether what the stats youre driving down on are actually useful | 19:15 |
| jeblair | mtreinish: i think we can gate on trend data, just probably not _timing_ trend data | 19:16 |
| jeblair | mtreinish: SpamapS is proposing some non-noisy metrics that could, possibly, one day, be gateable :) | 19:16 |
| *** glauco__ has quit IRC | 19:16 | |
| SpamapS | greghaynes: indeed, correlating the two will be quite useful in running down the causes. | 19:16 |
| *** glauco_ has joined #openstack-infra | 19:16 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 19:17 |
| jeblair | mordred: do you have any more information about 226751? | 19:17 |
| SpamapS | "We reduced SQL rows read by 10% and that resulted in LESS concurrency in the system because we exposed a locking race" <-- something I've had happen to me before. | 19:17 |
| *** tsekiyama has quit IRC | 19:18 | |
| *** david-lyle has quit IRC | 19:18 | |
| greghaynes | Yea, exactly. Whenever youre driving something with a model you always want to also store the actual data youre modeling just so you remain based in reality | 19:18 |
| odyssey4me | any chance for us getting one more set of eyes on this? https://review.openstack.org/228406 | 19:19 |
| SpamapS | Or another fun one.. we switched from a single node InnoDB cluster for session storage to a 4 node MySQL cluster, which reduced overall response time, but added 10x network traffic to the system because it was MySQL 4.1 and NDB crashed every 5s, resulting in worse system performance in general because of saturated trunks between the redundant switches hosting the NDB nodes. | 19:19 |
| jeblair | mordred: er, okay. i'm going to go get lunch. i'll try to look into 226751 if you haven't by the time i get back. | 19:19 |
| SpamapS | So yeah, correlating the reality with the counters is critical. | 19:20 |
| openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci: Fix pep8 errors in tripleo-jobs.py https://review.openstack.org/230091 | 19:20 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add new graph type to show run_time variance by metadata https://review.openstack.org/210569 | 19:20 |
| SpamapS | greghaynes: it's just that.. the variances in our reality are staggering. ;) | 19:21 |
| *** kushal has quit IRC | 19:21 | |
| * SpamapS wanders off to an indian buffet | 19:21 | |
| greghaynes | oh man, indian sounds good | 19:21 |
| greghaynes | -> lunch too | 19:21 |
| *** moravec has joined #openstack-infra | 19:21 | |
| mordred | jeblair: yah. I'm digging in right now | 19:22 |
| docaedo | fungi: gate check failed with "Error: Parameter path failed on File[undef]: File paths must be fully qualified, not 'undef' at /etc/puppet/modules/apps_site/manifests/init.pp:86" (http://logs.openstack.org/78/230078/1/check/gate-infra-puppet-apply-bare-trusty/696e62b/console.html) | 19:22 |
| mordred | jeblair: will hopefully have info for you when you get back | 19:22 |
| docaedo | fungi: but I thought hiera would have filled that stuff in so it wouldn't be undefined? | 19:22 |
| *** notnownikki has quit IRC | 19:22 | |
| jeblair | mordred: is the auto-ip stuff enabled? | 19:22 |
| mordred | jeblair: yah. should be | 19:23 |
| jeblair | mordred: k. lunch. good luck. | 19:23 |
| *** matbu_ has quit IRC | 19:23 | |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-diskimage_builder: Fix apt issue. https://review.openstack.org/220505 | 19:23 |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-diskimage_builder: Add acceptance tests for puppet-diskimage_builder. https://review.openstack.org/220262 | 19:23 |
| fungi | docaedo: Error: Parameter path failed on File[undef]: File paths must be fully qualified, not 'undef' at /etc/puppet/modules/apps_site/manifests/init.pp:86 | 19:24 |
| tristanC | jd__: Should we use "jd__" or just "jd" as your irc name in TC candidate list ? | 19:24 |
| fungi | docaedo: oh, you saw it too | 19:24 |
| *** matbu has joined #openstack-infra | 19:25 | |
| docaedo | fungi: right, but I'm confused I guess because setting those values to undef was the review advice I got | 19:25 |
| crinkle | erm i don't think a file path of '' would be any better | 19:26 |
| fungi | docaedo: it looks like maybe the contents value is being filled causing that conditional block to evaluate? | 19:26 |
| docaedo | crinkle: no way, this is totally your fault ;) | 19:26 |
| *** david-lyle has joined #openstack-infra | 19:27 | |
| docaedo | fungi: I have to confess I have not peeled back enough infra layers to understand what's breaking. I mainly just copied other things (like from gerrit host) that appeared to work since they were merged. | 19:28 |
| *** matrohon has joined #openstack-infra | 19:29 | |
| fungi | docaedo: in this case we're passing in "XXX" for $ssl_chain_file_contents so it wants to create that file, but the module says the default $ssl_chain_file is undef | 19:29 |
| fungi | docaedo: i expect you need to set filenames/paths you want to use for those three in 230078 | 19:30 |
| tristanC | jd__: I put jd__ as it's your actual ircname here and your openstack profile says so too. Please let us know before tomorrow if you'd prefer just "jd" | 19:30 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 19:31 |
| tristanC | small details matters :) | 19:32 |
| *** yamamoto has quit IRC | 19:32 | |
| crinkle | docaedo: you can see gerrit sets ssl_chain_file so it doesn't have this problem http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/review.pp#n54 | 19:33 |
| *** shardy_afk is now known as shardy | 19:33 | |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: DNM: test oslo.service backports https://review.openstack.org/230101 | 19:33 |
| docaedo | fungi: thanks, I'll make some adjustments | 19:33 |
| *** puranamr has joined #openstack-infra | 19:33 | |
| fungi | docaedo: perhaps a cleaner solution is to amend the apps_site module to pick different default filenames if the contents are passed in for them | 19:33 |
| docaedo | crinkle: thank you for that pointer | 19:33 |
| anteaya | tristanC: thank you for your attention to detail! | 19:33 |
| *** jsavak has quit IRC | 19:33 | |
| anteaya | tristanC: you are so good at that | 19:33 |
| *** julim has joined #openstack-infra | 19:33 | |
| *** jsavak has joined #openstack-infra | 19:34 | |
| *** sergmelikyan has joined #openstack-infra | 19:34 | |
| dhellmann | mordred: looking at that release now | 19:34 |
| fungi | docaedo: yeah, i think i see the logic error in the apps_site module there actually | 19:34 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Provide short-circuit for finding server networks https://review.openstack.org/229140 | 19:35 |
| mordred | dhellmann: thanks! | 19:35 |
| dhellmann | mordred: is "Added SWITCHengines vendor file" a feature? | 19:35 |
| dhellmann | mordred: "Add support for Catalyst as vendor" almost certainly is, no? | 19:35 |
| tristanC | anteaya: you're welcome :-) | 19:35 |
| *** hdd has quit IRC | 19:35 | |
| crinkle | fungi: i think all the sites have this logic error | 19:36 |
| mordred | dhellmann: they're both exactly the same level of feature - I'm not sure if I think of those as features or not | 19:36 |
| crinkle | fungi: http://git.openstack.org/cgit/openstack-infra/puppet-askbot/tree/manifests/site/ssl.pp for example | 19:36 |
| anteaya | tristanC: :) | 19:36 |
| mordred | dhellmann: I could probably make a salient argument in both directions ... how are you defining it? | 19:36 |
| dhellmann | mordred: ok. normally if it smells like a feature we want a min version bump. | 19:36 |
| mordred | dhellmann: ok. I can respin with a 1.8.0 number | 19:36 |
| dhellmann | mordred: sounds good, thanks | 19:36 |
| mordred | dhellmann: done | 19:37 |
| mordred | dhellmann: code review ftw | 19:37 |
| *** glauco___ has joined #openstack-infra | 19:37 | |
| *** moravec has quit IRC | 19:37 | |
| fungi | crinkle: the askbot module does, yes, i saw earlier that it's being worked around in the global site manifest | 19:37 |
| dhellmann | mordred: ++ | 19:39 |
| *** glauco_ has quit IRC | 19:40 | |
| fungi | docaedo: crinkle: probably we should set all three of those file paths undef for the default, then if content is also undef use the snakeoil paths and don't use a chain cert file, but if content is provided then create the files at a sane base location like /etc/ssl/*/$fqdn.??? or at the specified path | 19:40 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 19:40 |
| fungi | that makes it possible (but not necessary) to override the cert/key paths on the filesystem when having puppet create them for you | 19:41 |
| *** moravec has joined #openstack-infra | 19:41 | |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-openstackci: Pin fog-google gem to known working version. https://review.openstack.org/230104 | 19:42 |
| openstackgerrit | Merged openstack-infra/project-config: Add pep8 and bashate non-voting checks to openstack-ansible https://review.openstack.org/228406 | 19:42 |
| *** bltavares has joined #openstack-infra | 19:42 | |
| mordred | fungi: all of your words sound like words | 19:42 |
| docaedo | fungi: that makes sense to me, but I'll need to beg for help from someone like crinkle who really gets this :) | 19:42 |
| bltavares | Hi there folks. I noticed that a bunch of gate jobs (gate-openstackci-beaker-trusty-dsvm) on every review was failing because of a bad transitive dependency. | 19:43 |
| bltavares | I pinned the version on the project that was causing the failures here https://review.openstack.org/#/c/230104/ | 19:44 |
| bltavares | I hope that is a good way to fix that for other modules (: | 19:44 |
| mordred | bltavares: I was wondering why we cared about fog-google :) | 19:45 |
| *** puranamr has quit IRC | 19:45 | |
| crinkle | docaedo: fungi i think it would be sufficient for now to set ssl_chain_file in site.pp and fix the logic across the modules later | 19:45 |
| bltavares | mordred: transitive dependency. rspec-beaker -> fog -> fog-google >= 0.0.2 | 19:45 |
| bltavares | or something like that hehe | 19:46 |
| crinkle | bltavares: they are fixing it in beaker and will hopefully release soon https://github.com/puppetlabs/beaker/commit/4f94d048a6fb048fa9323772237af511c425061c | 19:46 |
| EmilienM | crinkle: can you look https://review.openstack.org/#/c/230064/ when you got time? | 19:46 |
| *** michaelxin has quit IRC | 19:46 | |
| *** sambetts has quit IRC | 19:47 | |
| crinkle | EmilienM: i'm about to run off to an appointment | 19:47 |
| bltavares | crinkle: oh, cool. I was following the discussion on fog-google but not on beaker. | 19:47 |
| EmilienM | crinkle: ok np | 19:47 |
| EmilienM | bltavares: puppetlabs is about to release beaker today | 19:47 |
| EmilienM | (with the fix) | 19:47 |
| bltavares | I will abandon the change adding that information. Thanks folks | 19:47 |
| mtreinish | anteaya: you asked for a graph of the run variance, I did one without a date filter: http://i.imgur.com/g3OrUnI.jpg | 19:48 |
| mtreinish | I really need to figure out how to tell matplotlib to space the labels a bit better | 19:48 |
| *** sambetts has joined #openstack-infra | 19:48 | |
| mordred | mtreinish: hahahahahahaha | 19:48 |
| austin81 | mtreinish: oh dear god my eyes | 19:48 |
| mordred | those labels are amazing | 19:48 |
| mordred | mtreinish: I truly believe you could print and frame that and sell it as art | 19:48 |
| austin81 | mordred: ++ | 19:48 |
| mtreinish | haha | 19:49 |
| anteaya | mtreinish: ha ha ha | 19:49 |
| anteaya | charlie and the chocolate factory | 19:50 |
| *** Goneri has joined #openstack-infra | 19:50 | |
| anteaya | embiggen it, even better | 19:50 |
| *** claudiub has quit IRC | 19:50 | |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: DNM: test oslo.service backports https://review.openstack.org/230101 | 19:50 |
| anteaya | mtreinish: thank you | 19:50 |
| docaedo | crinkle or fungi - if I add $ssl_chain_file = '/etc/ssl/certs/intermediate.pem' in site.pp would that fix it? | 19:51 |
| mtreinish | anteaya: I'll read the matplot docs a bit more and see how to force to space the ticks a bit more (or make the label font smaller) | 19:52 |
| mtreinish | so you can actually see the job names | 19:52 |
| anteaya | oh I can see job names | 19:52 |
| anteaya | all of them | 19:52 |
| anteaya | in a lovely pile | 19:53 |
| anteaya | new meaning to word cloud | 19:53 |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: DNM: test devstack work around https://review.openstack.org/230108 | 19:54 |
| *** yolanda has quit IRC | 19:54 | |
| *** ociuhandu has joined #openstack-infra | 19:54 | |
| *** sergmelikyan has quit IRC | 19:55 | |
| sdague | jeblair: just another thanks for Depends-On | 19:55 |
| *** ifarkas has quit IRC | 19:55 | |
| *** gordc has joined #openstack-infra | 19:55 | |
| *** yolanda has joined #openstack-infra | 19:55 | |
| sdague | 230101,2 and 230108,1 show some fun testing of possible fixes | 19:55 |
| dims_ | sdague: jeblair: Hearty +1 | 19:56 |
| *** jsavak has quit IRC | 19:57 | |
| *** moravec has quit IRC | 20:00 | |
| *** denisra has quit IRC | 20:00 | |
| *** moravec has joined #openstack-infra | 20:01 | |
| *** tjones has joined #openstack-infra | 20:01 | |
| *** AJaeger has quit IRC | 20:02 | |
| greghaynes | mtreinish: hah, neat graph | 20:02 |
| *** nelsnelson has quit IRC | 20:02 | |
| *** jsavak has joined #openstack-infra | 20:03 | |
| *** signed8bit has quit IRC | 20:03 | |
| *** ociuhandu has quit IRC | 20:04 | |
| *** aeng has joined #openstack-infra | 20:04 | |
| *** ociuhandu has joined #openstack-infra | 20:06 | |
| *** jsavak has quit IRC | 20:07 | |
| *** odgrim has quit IRC | 20:08 | |
| *** jsavak has joined #openstack-infra | 20:08 | |
| jd__ | tristanC: jd__ is fine | 20:09 |
| *** jtomasek has joined #openstack-infra | 20:10 | |
| mtreinish | greghaynes, mordred, anteaya, austin81: here is is with somewhat legible labels: http://i.imgur.com/jEHfteW.jpg | 20:12 |
| *** tiswanso has quit IRC | 20:12 | |
| mtreinish | I'm gonna look at swapping the x and y so you don't have to crink your neck to read it | 20:12 |
| greghaynes | looking at it - seems like the inner quartile isnt *that* bad | 20:12 |
| greghaynes | for most of the tests | 20:13 |
| mtreinish | greghaynes: heh, yeah there are a few jobs which are noticeably worse then others | 20:14 |
| greghaynes | mtreinish: it also seems like the lower bound on a lot of them lines up too well - are those maybe failures right after boot? | 20:16 |
| *** hdd has joined #openstack-infra | 20:16 | |
| jeblair | sdague: \o/ thanks! | 20:17 |
| mtreinish | greghaynes: the query here is only runs without any failures | 20:17 |
| mtreinish | greghaynes: oh, this is for the run as a whole, not individual tests | 20:18 |
| mtreinish | it's grouped by build_name | 20:18 |
| *** tsekiyama has joined #openstack-infra | 20:18 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 20:19 |
| jeblair | mordred: why are you switching to interface_ip in 226751? | 20:19 |
| mtreinish | I could do it for variance of an individual tests too, although there would be a lot more boxes with whiskers. We have about 1000-1600 tests in a full run | 20:19 |
| jeblair | mordred: afaik, that's not expected to work until the next shade release with our changes; meanwhile, the code we have should work with the current shade release | 20:19 |
| *** flepied has quit IRC | 20:20 | |
| clarkb | greghaynes: its more likely that the min and max values are related to cloud things | 20:21 |
| clarkb | greghaynes: and the non outlier results are more representative | 20:22 |
| greghaynes | "cloud things" | 20:22 |
| asselin | I'm trying to run nodepool unit tests but I get mysql connection refused errors. Are you supposed to have mysql installed? http://paste.openstack.org/show/475119/ | 20:22 |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Use shade for all OpenStack interactions (jeblair's version) https://review.openstack.org/230122 | 20:22 |
| mordred | jeblair: interface_ip should work in both versions - it's been in the interface for a while | 20:22 |
| mordred | jeblair: I mean, it won't fix anything and we can revert that - but it should work | 20:22 |
| * asselin reads the docs | 20:22 | |
| clarkb | greghaynes: ya network slowness, cpu time, etc | 20:23 |
| clarkb | noisy neighbors | 20:23 |
| anteaya | mtreinish: neutron-dvr, neutron-full and full2 are out on there own | 20:23 |
| jeblair | mordred: ok. it feels like a tangent to me atm -- i mean, public_v4 _should_ work. | 20:23 |
| anteaya | their | 20:23 |
| jeblair | mordred: anyway, i forked it ^ :) | 20:23 |
| mordred | jeblair: kk | 20:23 |
| *** tsekiyama has quit IRC | 20:23 | |
| anteaya | mtreinish: let's try to smooth our the whiskers | 20:24 |
| mtreinish | anteaya: they're seperate job definitions | 20:24 |
| anteaya | mtreinish: yes and they have many blue crosses in the 7000 to 10000 range | 20:25 |
| *** odgrim has joined #openstack-infra | 20:26 | |
| mtreinish | oh, those are outliers | 20:26 |
| anteaya | yes, I think I was trying to say those jobs have some outliers | 20:27 |
| anteaya | I think that was my point | 20:27 |
| mtreinish | anteaya: I was just confused by your wording. | 20:28 |
| *** geoffarnold has joined #openstack-infra | 20:29 | |
| anteaya | asselin: I concur something running in the unit tests is looking for a MySQL to connect to | 20:29 |
| anteaya | mtreinish: ah happens often | 20:29 |
| anteaya | mtreinish: sorry about that | 20:29 |
| asselin | anteaya, I got it. It was all in the README | 20:30 |
| anteaya | I was trying to say these 3 jobs appear to be not like the others | 20:30 |
| anteaya | asselin: ah helpful README's | 20:30 |
| *** geoffarn_ has joined #openstack-infra | 20:30 | |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Use shade for all OpenStack interactions (jeblair's version) https://review.openstack.org/230122 | 20:31 |
| jeblair | mordred: ^ check out diff PS1..PS2 there | 20:31 |
| fungi | docaedo: that alone won't "fix" it since you'll start overwriting the snakeoil cert/key on the system. you'd at least need to pass in sane paths for all three files | 20:31 |
| anteaya | jeblair: I'm curious, I see you are the patch owner and have given yourself co-authored by in the commit message, why? | 20:32 |
| jeblair | anteaya: because i am not the patch _author_ | 20:32 |
| anteaya | ah thank you | 20:32 |
| fungi | docaedo: my recommendation in 226984 would work with the current site.pp and would also avoid overwriting the snakeoil files if someone did pass content in for them without specifying a path | 20:32 |
| *** geoffarnold has quit IRC | 20:33 | |
| docaedo | fungi: ok your suggestion seems sane, I just don't have the puppet skillz at hand to make that happen - will give me something new to learn unless someone else is bored and wants to educate me :) | 20:35 |
| *** cdent has quit IRC | 20:35 | |
| *** odgrim1 has joined #openstack-infra | 20:35 | |
| *** odgrim has quit IRC | 20:35 | |
| fungi | docaedo: i might have time to take a stab at it later, but am mostly occupied moving everything up off my ground floor for at least a few more hours | 20:36 |
| openstackgerrit | Merged openstack-infra/release-tools: Sync with global requirements https://review.openstack.org/227086 | 20:36 |
| *** amuller is now known as amuller_afk | 20:36 | |
| docaedo | fungi: no problem, prepping for potential disaster is a little more important than helping me make SSL cert work for app catalog | 20:37 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 20:37 |
| clarkb | this test is so close | 20:37 |
| ianw | http://nodepool.openstack.org/image.log <- we have f22 images built! thanks fungi for adding that workaround. pabelanger has tested latest versions of dnf and it's going to make the whole caching story on par with yum, we think. | 20:37 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Destroy previous test env VM's earlier https://review.openstack.org/230129 | 20:37 |
| *** hdd has quit IRC | 20:37 | |
| *** abregman has joined #openstack-infra | 20:37 | |
| docaedo | fungi: maybe crinkle will take a shot at it later (or maybe even krinkle will come back and help!) | 20:38 |
| *** sergmelikyan has joined #openstack-infra | 20:39 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add new graph type to show run_time variance by metadata https://review.openstack.org/210569 | 20:40 |
| glauco___ | I noticed that recently we added puppet-nodejs as part of system-config/modules.env | 20:40 |
| glauco___ | but there is a know bug on the version we are using | 20:40 |
| *** glauco___ is now known as glauco | 20:40 | |
| nibalizer | we can likely just bump | 20:40 |
| nibalizer | link to the bug? | 20:41 |
| glauco | nibalizer: This is the bug https://github.com/willdurand/puppet-nodejs/issues/50 | 20:42 |
| mordred | jeblair: HEADDESK | 20:42 |
| glauco | nibalizer: It was fixed on version 1.8.0 | 20:42 |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources: Summit Application API https://review.openstack.org/221964 | 20:42 |
| mordred | jeblair: oh. thank you as well - that reminds me of the one additional patch I wanted to write | 20:42 |
| glauco | nibalizer: https://github.com/willdurand/puppet-nodejs/blob/v1.8.0/lib/puppet/provider/package/npm.rb | 20:42 |
| *** yamamoto has joined #openstack-infra | 20:42 | |
| jeblair | mordred: NO MORE PATCHES! NO MORE PATCHES! :) | 20:42 |
| mordred | jeblair: which is to make create_server/get_server/list_servers _ALWAYS_ call get_openstack_vars on the server so that consumers do not have to deal with this | 20:43 |
| mordred | jeblair: but we can do that as a follow on | 20:43 |
| glauco | nibalizer: I am actually using on module that needs node-gyp and it fails :/ | 20:43 |
| jeblair | mordred: can i push 230122 over 226751? | 20:43 |
| mordred | jeblair: yes | 20:43 |
| nibalizer | glauco: we use this one https://github.com/puppet-community/puppet-nodejs | 20:43 |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 20:44 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 20:45 |
| glauco | nibalizer: my bad! | 20:45 |
| glauco | That seem to be already fixed in there | 20:45 |
| *** markvoelker has quit IRC | 20:46 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-infra/devstack-gate: DNM: test oslo.service backports https://review.openstack.org/230101 | 20:46 |
| *** markvoelker has joined #openstack-infra | 20:46 | |
| *** derekh has joined #openstack-infra | 20:46 | |
| *** hdd has joined #openstack-infra | 20:46 | |
| *** tsekiyama has joined #openstack-infra | 20:46 | |
| *** markvoelker has quit IRC | 20:47 | |
| *** hashar has quit IRC | 20:47 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Update query for ebtables kernel nova bug 1501558 https://review.openstack.org/230136 | 20:47 |
| openstack | bug 1501558 in OpenStack Compute (nova) "nova-net: libvirtError: Error while building firewall: Some rules could not be created for interface: Unable to update the kernel" [High,Confirmed] https://launchpad.net/bugs/1501558 | 20:47 |
| *** markvoelker has joined #openstack-infra | 20:47 | |
| *** dkranz has quit IRC | 20:47 | |
| *** yamamoto has quit IRC | 20:47 | |
| *** bltavares has quit IRC | 20:52 | |
| *** bltavares has joined #openstack-infra | 20:52 | |
| *** eharney has quit IRC | 20:53 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Pin delorean to a pre mock version https://review.openstack.org/230011 | 20:54 |
| *** ekarlso has quit IRC | 20:55 | |
| *** sridhar_ram has joined #openstack-infra | 20:55 | |
| *** geoffarn_ has quit IRC | 20:55 | |
| *** geoffarnold has joined #openstack-infra | 20:56 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 20:56 |
| clarkb | the threading around this is all kinds of fun | 20:56 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Update query for ebtables kernel nova bug 1501558 https://review.openstack.org/230136 | 20:57 |
| openstack | bug 1501558 in OpenStack Compute (nova) "nova-net: libvirtError: Error while building firewall: Some rules could not be created for interface: Unable to update the kernel" [High,Confirmed] https://launchpad.net/bugs/1501558 | 20:57 |
| clarkb | if this last patch doesn't work I give up for now. | 20:58 |
| clarkb | mordred: jeblair so we need 226751 and it should be good now? | 20:58 |
| *** ekarlso has joined #openstack-infra | 20:58 | |
| *** matrohon has quit IRC | 20:58 | |
| *** tsekiyama has quit IRC | 20:58 | |
| greghaynes | clarkb: thats the spirit! | 20:59 |
| mordred | clarkb: yes | 20:59 |
| *** FallenPegasus has joined #openstack-infra | 20:59 | |
| *** gildub has joined #openstack-infra | 21:00 | |
| *** derekh has quit IRC | 21:00 | |
| jeblair | mordred, clarkb: the dsvm tests for it under the other change id are close to finishing: https://jenkins05.openstack.org/job/gate-dsvm-nodepool/20/console | 21:00 |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Check pyflakes rules https://review.openstack.org/230141 | 21:01 |
| clarkb | jeblair: ^ and that will enable just pyflakes rules to catch var and import problems | 21:01 |
| *** jsavak has quit IRC | 21:02 | |
| clarkb | greghaynes: on the other hand if it does work I think I can remove some of my monkey patching to reduce async | 21:02 |
| *** asselin_ has joined #openstack-infra | 21:02 | |
| *** jsavak has joined #openstack-infra | 21:03 | |
| *** mudassirlatif_ has joined #openstack-infra | 21:04 | |
| *** jsavak has quit IRC | 21:04 | |
| *** alivigni has quit IRC | 21:04 | |
| *** asselin__ has quit IRC | 21:06 | |
| *** mudassirlatif has quit IRC | 21:07 | |
| *** mudassirlatif_ is now known as mudassirlatif | 21:07 | |
| *** asselin_ has quit IRC | 21:07 | |
| *** signed8bit has joined #openstack-infra | 21:07 | |
| *** signed8b_ has joined #openstack-infra | 21:09 | |
| *** jpeeler has quit IRC | 21:12 | |
| *** moravec has quit IRC | 21:13 | |
| *** signed8b_ is now known as signed8bit_ZZZzz | 21:13 | |
| *** signed8bit has quit IRC | 21:13 | |
| *** tsekiyama has joined #openstack-infra | 21:14 | |
| *** moravec has joined #openstack-infra | 21:14 | |
| *** sergmelikyan has quit IRC | 21:15 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/nodepool: Test ZMQ event handler https://review.openstack.org/230023 | 21:16 |
| jeblair | clarkb, mordred: hrm, that dsvm test still doesn't look great | 21:16 |
| jeblair | it hasn't finished yet, but i think it should have. | 21:17 |
| *** geoffarnold has quit IRC | 21:17 | |
| *** geoffarnold has joined #openstack-infra | 21:17 | |
| *** signed8bit_ZZZzz has quit IRC | 21:18 | |
| *** jsavak has joined #openstack-infra | 21:19 | |
| *** denisra has joined #openstack-infra | 21:19 | |
| *** Goneri has quit IRC | 21:19 | |
| openstackgerrit | Augustina Ragwitz proposed openstack-infra/project-config: (WIP) split out into testable units https://review.openstack.org/229772 | 21:20 |
| EmilienM | bltavares: https://github.com/puppetlabs/beaker/releases/tag/2.25.0 | 21:22 |
| bltavares | EmilienM: nice! thanks for letting me know. I will trigger the tests again (: | 21:23 |
| clarkb | jeblair: jenkins says 8 more minutes? | 21:23 |
| jeblair | clarkb: i logged into the host and see the same error in the log file | 21:23 |
| jeblair | clarkb: i'll let it finish and then debug on-host | 21:23 |
| clarkb | ok | 21:23 |
| mordred | jeblair: the same "can't find IP" error? | 21:24 |
| jeblair | mordred: yup | 21:24 |
| mordred | jeblair: that error is very confusing to me | 21:24 |
| *** FallenPegasus has quit IRC | 21:24 | |
| EmilienM | bltavares: cool :) | 21:25 |
| *** shardy has quit IRC | 21:26 | |
| *** moravec has quit IRC | 21:26 | |
| *** jsavak has quit IRC | 21:27 | |
| *** jsavak has joined #openstack-infra | 21:28 | |
| jeblair | mordred: while we're waiting -- why do we have get_openstack_vars at all? | 21:28 |
| EmilienM | fungi: if it's fine for infra folks, we can have 2 more jobs https://review.openstack.org/#/c/230064/ - if it's too much, we can drop upgrade jobs | 21:28 |
| jeblair | mordred: why not have whatever "server" object shade returns implement all of that internally? | 21:28 |
| mordred | jeblair: yes. | 21:28 |
| mordred | jeblair: that was the next patch I was going to write | 21:29 |
| mordred | jeblair: the other patches were to make it be able to be inexpensive enough that that would not become double-plus bad | 21:29 |
| mordred | (the short-circuit network call patch) | 21:29 |
| jeblair | mordred: ok | 21:29 |
| *** trown is now known as trown|outttypeww | 21:30 | |
| jeblair | mordred, clarkb: http://logs.openstack.org/22/230122/2/check/gate-dsvm-nodepool/c1a3157//logs/screen-nodepool.txt.gz#_2015-10-01_21_27_05_545 | 21:30 |
| *** signed8bit has joined #openstack-infra | 21:30 | |
| jeblair | digging into that now | 21:30 |
| mordred | jeblair: so - in one of my patches I tried to add debug logging statements to all the exceptions related to can't find IP | 21:31 |
| mordred | jeblair: (we log the server struct in that case in one of the cases, but not the other two) | 21:31 |
| jeblair | clarkb: this devstack plugin thing is pretty cool :) | 21:32 |
| anteaya | I'm seeing a failure on not being able to build a wheel for Pillow | 21:33 |
| clarkb | jeblair: ya it can be pretty handy | 21:34 |
| anteaya | http://logs.openstack.org/80/229680/2/check/gate-project-config-layout/66e431a/console.html#_2015-10-01_21_13_56_705 | 21:34 |
| *** e0ne has joined #openstack-infra | 21:34 | |
| clarkb | jeblair: mordred also fwiw that job is setup to use neutron ovs with floating ips | 21:34 |
| *** FallenPegasus has joined #openstack-infra | 21:34 | |
| clarkb | jeblair: mordred so it should resemble hpcloud | 21:34 |
| *** rfolco has quit IRC | 21:35 | |
| *** annegentle has quit IRC | 21:35 | |
| anteaya | 11 hits in the last 30 minutes: message:"Failed building wheel for Pillow" | 21:37 |
| anteaya | something that we expect to provide pillow isn't doing so | 21:37 |
| clarkb | anteaya: looking now | 21:38 |
| *** geoffarnold has quit IRC | 21:38 | |
| *** boris-42 has joined #openstack-infra | 21:38 | |
| anteaya | clarkb: thank you | 21:38 |
| *** geoffarnold has joined #openstack-infra | 21:38 | |
| clarkb | looks like its a dep of sphinx and requires libjpeg? | 21:38 |
| anteaya | yeah it is image based | 21:38 |
| jeblair | mordred: oh... my "fix" didn't change anything -- i think you can recursively call get_hostvars_from_server on its own output, because in all cases, input and output is a Bunch() | 21:39 |
| anteaya | I can support your initial theory | 21:39 |
| jeblair | mordred: so that was an optimization, not a fix, as it turns out | 21:39 |
| *** regXboi has quit IRC | 21:39 | |
| jeblair | mordred: i pasted a repr() of what we get in the etherpad https://etherpad.openstack.org/p/nodepool-shade | 21:39 |
| jeblair | mordred: public_v4='' | 21:40 |
| jeblair | mordred: (interestingly, public_v6 is set | 21:40 |
| mordred | there is no public address | 21:41 |
| mordred | addresses={u'private': [{u'OS-EXT-IPS-MAC:mac_addr': u'fa:16:3e:36:99:a4', u'version': 4, u'addr': u'10.1.0.114', u'OS-EXT-IPS:type': u'fixed'}, {u'OS-EXT-IPS-MAC:mac_addr': u'fa:16:3e:36:99:a4', u'version': 6, u'addr': u'fd89:b548:10f9:0:f816:3eff:fe36:99a4', u'OS-EXT-IPS:type': u'fixed'}]} | 21:41 |
| *** tjones has quit IRC | 21:41 | |
| mordred | so it is quite correct in saying that it cannot find one | 21:41 |
| mordred | now - why was there not one | 21:42 |
| jeblair | mordred: i'm guessing because we don't actually have a public ip allocation to give devstack? | 21:42 |
| mordred | well - no, there should be floating ips in devstack | 21:42 |
| jeblair | mordred: would they be non-rfc1918 though? | 21:42 |
| mordred | this is why "public" can't just be determined by looking at rfc1918 ranges | 21:42 |
| mordred | yes. we don't use that to determine their public-ness | 21:42 |
| jeblair | mordred: oh, i could have sworn i saw a function in shade to do that | 21:43 |
| *** salv-orl_ has joined #openstack-infra | 21:43 | |
| mordred | that's a very late fall back if nothing else works attempt | 21:43 |
| mordred | but the main thing we look for is addresses on a neutron network marked "router:external" - and then for floating ips | 21:43 |
| clarkb | the floating ips are in 172.24.5.0/24 | 21:43 |
| *** mriedem has quit IRC | 21:43 | |
| *** david-lyle has quit IRC | 21:43 | |
| *** rlandy has quit IRC | 21:43 | |
| mordred | doesn't matter | 21:43 |
| jeblair | mordred: 23.253.110.184 is the devstack host i have held | 21:43 |
| mordred | this server does not hav ea floating ip | 21:43 |
| jeblair | mordred: if you want to poke at it | 21:44 |
| clarkb | ah that would do it | 21:44 |
| mordred | the IP detection is not failing | 21:44 |
| mordred | something failed in IP creation | 21:44 |
| jeblair | mordred: want me to run it with shade debug? | 21:44 |
| mordred | sigh | 21:44 |
| mordred | nope | 21:44 |
| mordred | I see it | 21:44 |
| *** weshay has quit IRC | 21:44 | |
| mordred | it's silly. you're going to like it | 21:44 |
| jeblair | i love those | 21:45 |
| mordred | jeblair: I mean- I DO think we should be running these tests with shade debug turned on | 21:45 |
| mordred | return self._client.get_openstack_vars(self._client.create_server( | 21:45 |
| mordred | wait=False, auto_ip=True, **create_args)) | 21:45 |
| mordred | anybody else see the bug? | 21:45 |
| clarkb | mordred: wait=False? | 21:46 |
| clarkb | I dunno | 21:46 |
| mordred | yup. auto_ip can't do anyting with wait=False | 21:46 |
| *** salv-orlando has quit IRC | 21:46 | |
| mordred | because there is nothing to attach the ip _to_ | 21:46 |
| *** gordc has quit IRC | 21:46 | |
| *** ayoung has quit IRC | 21:46 | |
| *** kgiusti has quit IRC | 21:47 | |
| mordred | jeblair: I apologize, my brain should have caught this before - I thnk we will need one more patch, because I do not think we want wait=True in nodepool | 21:47 |
| jeblair | mordred: if we set wait to false, other threads will still multiplex, right? | 21:47 |
| *** ldnunes has quit IRC | 21:47 | |
| jeblair | er, wait to true, rather | 21:48 |
| jeblair | we could even just drop nodepool's waitforserver... | 21:48 |
| openstackgerrit | Bruno Tavares proposed openstack-infra/puppet-diskimage_builder: Add acceptance tests for puppet-diskimage_builder. https://review.openstack.org/220262 | 21:48 |
| *** annegentle has joined #openstack-infra | 21:49 | |
| *** austin81 has left #openstack-infra | 21:50 | |
| Shrews | mordred: oh, that's a fun one | 21:50 |
| *** tjones has joined #openstack-infra | 21:51 | |
| *** shashank_hegde has quit IRC | 21:51 | |
| *** johnsom has joined #openstack-infra | 21:52 | |
| *** spzala has quit IRC | 21:54 | |
| *** spzala has joined #openstack-infra | 21:55 | |
| *** annegentle has quit IRC | 21:56 | |
| *** david-lyle has joined #openstack-infra | 21:56 | |
| *** pc_m has quit IRC | 21:57 | |
| *** nelsnelson has joined #openstack-infra | 21:57 | |
| *** doug-fish has quit IRC | 21:57 | |
| *** nelsnelson has quit IRC | 21:58 | |
| openstackgerrit | Anita Kuno proposed openstack/requirements: Pin Pillow as the latest release is not building https://review.openstack.org/230167 | 21:58 |
| *** doug-fish has joined #openstack-infra | 21:58 | |
| *** nelsnelson has joined #openstack-infra | 21:58 | |
| anteaya | clarkb: thank you | 21:58 |
| clarkb | anteaya: can you include a link to https://github.com/python-pillow/Pillow/commit/29601ca119adf3c02967ae9a8aea28d034bafe29 breaking builds because libjpeg-dev is required and we don't have it installed on the test nodes in the commit message | 22:00 |
| anteaya | I can and will | 22:00 |
| clarkb | great I can approve once thats done | 22:00 |
| anteaya | thanks | 22:00 |
| clarkb | then we can work on rebuilding images with libjpeg installed | 22:01 |
| clarkb | I can work on a patch to make that happen now | 22:01 |
| ruagair | o/ | 22:01 |
| pleia2 | ergh, releases breaking things | 22:01 |
| bltavares | EmilienM: just to let you know, the gates are green again \o | 22:02 |
| *** amuller_afk is now known as amuller | 22:02 | |
| mordred | Shrews: logic mismatch between how we cache and process in shade and how nodepool needs to - I've got an easy fix coming and a more complex one to talk about once I've dumped this thought into code | 22:02 |
| EmilienM | bltavares: \o/ | 22:02 |
| openstackgerrit | Anita Kuno proposed openstack/requirements: Pin Pillow as the latest release is not building https://review.openstack.org/230167 | 22:03 |
| anteaya | clarkb: let me know if further edits are required | 22:03 |
| clarkb | hrm I am thinking about this more and devstack runs wont be affected due to constraints but unittsts/docs etc will be | 22:04 |
| *** sambetts has quit IRC | 22:04 | |
| clarkb | and it will affect basically everything because sphinx | 22:04 |
| clarkb | I think | 22:04 |
| *** e0ne has quit IRC | 22:04 | |
| openstackgerrit | Glauco Oliveira proposed openstack-infra/puppet-openstack-health: Add provisioning scripts https://review.openstack.org/227849 | 22:04 |
| clarkb | it may just be quicker to rebuild images. Can work both at the same time | 22:04 |
| clarkb | anteaya: +2 | 22:05 |
| * anteaya is fine with what clarkb suggests | 22:05 | |
| anteaya | thank you | 22:05 |
| anteaya | pleia2: feel like reviewing a requirements patch? | 22:05 |
| anteaya | 230167 | 22:06 |
| *** rossella_s has quit IRC | 22:06 | |
| *** jsavak has quit IRC | 22:06 | |
| *** rossella_s has joined #openstack-infra | 22:06 | |
| pleia2 | anteaya: yep, looks good | 22:06 |
| anteaya | pleia2: thank you | 22:06 |
| *** tonytan4ever has quit IRC | 22:06 | |
| *** sambetts has joined #openstack-infra | 22:06 | |
| *** bltavares has quit IRC | 22:07 | |
| *** bltavares has joined #openstack-infra | 22:07 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 22:07 |
| *** bltavares has quit IRC | 22:07 | |
| *** bltavares has joined #openstack-infra | 22:09 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config: Add Pillow C deps to thick slaves https://review.openstack.org/230175 | 22:13 |
| openstackgerrit | Glauco Oliveira proposed openstack-infra/puppet-openstack-health: Add provisioning scripts https://review.openstack.org/227849 | 22:13 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Plumb wait and timout down to add_auto_ip https://review.openstack.org/230178 | 22:14 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 22:14 |
| mordred | jeblair: ^^ kk. those two should get us | 22:14 |
| clarkb | mordred: why not two calls? boot node | 22:15 |
| clarkb | mordred: then attach fip | 22:15 |
| mordred | clarkb: that is what we're doing | 22:15 |
| jeblair | mordred: ack | 22:15 |
| clarkb | then you can keep your existing method as a rwapper around that | 22:15 |
| mordred | clarkb: nodepool needs to do its optimization of doing the list once ever X seconds and having the thread calls check the list | 22:16 |
| clarkb | mordred: I don't see where in the latest change its floating ip | 22:16 |
| *** geoffarn_ has joined #openstack-infra | 22:16 | |
| *** glauco has quit IRC | 22:16 | |
| mordred | clarkb: add_ips_to_server(server) | 22:16 |
| *** nijaba has quit IRC | 22:16 | |
| mordred | gah. typo | 22:16 |
| mordred | jeblair: one sec | 22:16 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 22:16 |
| mordred | jeblair: _that_ | 22:16 |
| *** ociuhandu has quit IRC | 22:17 | |
| jeblair | mordred: i'm not sure the commit message makes sense in https://review.openstack.org/230178 | 22:17 |
| clarkb | mordred: maybe that shouldn't go in wait for server? | 22:17 |
| jeblair | mordred: maybe you meant to name the other method? | 22:17 |
| mordred | jeblair: yup. one sec | 22:18 |
| clarkb | and I think it should probably be add_floating_ip_to_server if thats what it is doing | 22:18 |
| clarkb | as I think you can add arbitrary IPs otherwise | 22:18 |
| mordred | clarkb: no - because it's not necessarily adding a floating ip | 22:18 |
| mordred | it's adding a floating ip if that's what it ndeeds to do | 22:19 |
| mordred | or it's doing nothing | 22:19 |
| mordred | or, if you pass in a specified ip, it'll add that | 22:19 |
| clarkb | mordred: I feel like we are encoding way too much business logic into shade there | 22:19 |
| *** nelsnelson has quit IRC | 22:19 | |
| mordred | clarkb: that is shade's sole purpose in life | 22:19 |
| clarkb | shade should make getting a floating IP easy if you need that | 22:19 |
| mordred | no | 22:19 |
| mordred | nonono | 22:19 |
| mordred | you should not hav eto know about floating ips vs. not | 22:19 |
| zaro | dhellmann: ping | 22:20 |
| mordred | because openstack is broken | 22:20 |
| clarkb | mordred: ok | 22:20 |
| mordred | and the entire idea that different clouds get you a public ip differently is stupid | 22:20 |
| reed | Shrews, looks like something broke on shade get_image between version 0.11 and 0.14 | 22:20 |
| *** geoffarnold has quit IRC | 22:20 | |
| *** Goneri has joined #openstack-infra | 22:20 | |
| mordred | clarkb: it should probably be caled "ensure_public_ip" | 22:20 |
| mordred | clarkb: so taht it's clearer | 22:20 |
| clarkb | mordred: that works | 22:20 |
| clarkb | mordred: and I think I would bubble the call up a layer above wait for server | 22:21 |
| mordred | clarkb: I'll add an enure_public_ip method as an alias as a follow on | 22:21 |
| mordred | clarkb: not sure what you mean? | 22:21 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Plumb wait and timout down to add_auto_ip https://review.openstack.org/230178 | 22:21 |
| clarkb | mordred: you have overloaded wait for server to mean wait for server to boot and wait for ip to attach | 22:21 |
| *** nelsnelson has joined #openstack-infra | 22:21 | |
| clarkb | mordred: when it should be wait for server then wait for ip as distinct steps | 22:21 |
| Shrews | reed: can you be more specific? | 22:22 |
| mordred | clarkb: can we take a step back - because I feel like we're talking about two different things | 22:22 |
| clarkb | reed: don't rule out glanceclient either | 22:22 |
| clarkb | mordred: line 283 https://review.openstack.org/#/c/226751/27/nodepool/provider_manager.py | 22:22 |
| *** FallenPegasus has quit IRC | 22:22 | |
| jeblair | clarkb: ideally i would like it to be one -- but we can't use that in nodepool because shade's caching is insufficient. | 22:22 |
| reed | clarkb, that's correct, it could be glanceclient's fault too | 22:22 |
| jeblair | clarkb: i would like shade to give me in one call, a server with a public ip | 22:22 |
| *** mwagner_ has quit IRC | 22:23 | |
| reed | Shrews, when I try to get image list from DreamCompute I get: OpenStackCloudException: Error fetching image list: 404 Not Found: The resource could not be found. (HTTP 404) (Inner Exception: ) | 22:23 |
| jeblair | clarkb: but since shade doesn't cache server lists the way nodepool does (yet), we still need 2 steps in nodepool | 22:23 |
| clarkb | jeblair: the problem is that openstack doesn't make it possible to express that if you want async calls | 22:23 |
| mordred | "wait for server without a public ip" is never a thing we want - but we have the shade calls broken into two steps so that nodepool can do advanced caching | 22:23 |
| Shrews | reed: ok, yeah, we fixed that | 22:23 |
| *** achanda has joined #openstack-infra | 22:23 | |
| Shrews | reed: https://github.com/openstack-infra/shade/commit/bccce1857b7c57e1f320ef7bd7299b76b18926dd | 22:23 |
| jeblair | clarkb: yep, which is why we should work around that in shade until openstack comes down from whatever drugs its on | 22:23 |
| clarkb | jeblair: but we aren't working around it in shade is what I am saying | 22:24 |
| clarkb | jeblair: because we can't | 22:24 |
| clarkb | it wouldbe cool if we could | 22:24 |
| reed | Shrews, ah! thanks | 22:24 |
| jeblair | clarkb: correct, we aren't, because i really want to get some ovh servers spun up | 22:24 |
| jeblair | clarkb: but we might be able to with a little more thought | 22:24 |
| Shrews | reed: we still need to release a new shade that has that change | 22:24 |
| Shrews | reed: coming soon | 22:24 |
| mordred | yup. I agree - we should definitely learn from this and make things better | 22:24 |
| clarkb | mordred: there are reasons to wait for a server without a public IP | 22:24 |
| reed | calebb, see what Shrews says ^^ | 22:24 |
| clarkb | mordred: one of them would be to not serialize multiple actions that happen post boot | 22:25 |
| clarkb | mordred: like attaching a voliume and floating ip | 22:25 |
| jeblair | clarkb: (for example, we managed to make shade compatible with nodepool's rate limiting, we may be able to make it compatible with its caching) | 22:25 |
| calebb | Shrews: thanks | 22:25 |
| jeblair | clarkb: yeah, it should be an option | 22:25 |
| clarkb | mordred: as is that code pushes us to serialize and I am saying just split it up | 22:25 |
| jeblair | clarkb: but since nodepool wants to do the _most common thing possible_ shade should make it happen in one call | 22:25 |
| mordred | clarkb: ok. I can see that. the other side though is that I don't want the simple case to take two api calls | 22:25 |
| clarkb | mordred: but it is | 22:25 |
| jeblair | clarkb: (which it does satisfactorily for most users, just not nodepool because of the volume) | 22:25 |
| mordred | clarkb: so, I'm happy to provide an API that can slide it however | 22:25 |
| clarkb | lets back up | 22:25 |
| reed | are mordred's commit messages always so entertaining? | 22:25 |
| clarkb | the code in nodepool right now is multiple calls | 22:26 |
| *** nijaba has joined #openstack-infra | 22:26 | |
| *** nijaba has joined #openstack-infra | 22:26 | |
| Shrews | reed: always | 22:26 |
| clarkb | So, expose that in nodepool the same way | 22:26 |
| clarkb | then if and when we fix the problems here change it | 22:26 |
| clarkb | jeblair: I get that but it isn't currently working that way | 22:27 |
| mordred | clarkb: the code in nodepool is currently doing two things | 22:27 |
| clarkb | jeblair: so I am confused why we are acting as if it were the case | 22:27 |
| clarkb | if and when that changes change nodepool | 22:27 |
| mordred | clarkb: the first is that it's requesting a server be created and to not wait for it | 22:27 |
| mordred | which is what nodepool has always done | 22:27 |
| mordred | we then let nodepool poll for the server to be active | 22:27 |
| mordred | then we have nodepool request an IP for the server | 22:27 |
| mordred | this is what nodepool has always done | 22:27 |
| *** ZZelle_ has quit IRC | 22:28 | |
| clarkb | mordred: except that you have serialized it with waiting for the server | 22:28 |
| mordred | we've encapsulated the two things that are not "poll for the server to be active" behind shade calls | 22:28 |
| clarkb | which means any other async actions have to wait | 22:28 |
| mordred | clarkb: we're not waiting for the server | 22:28 |
| clarkb | which was not the case before | 22:28 |
| mordred | clarkb: wait=False | 22:28 |
| clarkb | mordred: after you wait | 22:28 |
| clarkb | its literally after the wait for resource call | 22:28 |
| clarkb | I am saying don't tie those two actions together | 22:28 |
| clarkb | so that calling code can be smarter about async actions | 22:29 |
| mordred | you are making absolutely no sense to me | 22:29 |
| clarkb | then if/when shade is smarter we can change nodepool | 22:29 |
| clarkb | mordred: https://review.openstack.org/#/c/226751/27/nodepool/provider_manager.py line 283 | 22:29 |
| mordred | yes. | 22:29 |
| clarkb | mordred: bubble that up one stack frame | 22:29 |
| mordred | why? | 22:29 |
| mordred | this is a thread that is for the task of launching this server | 22:30 |
| mordred | it's not blocking anything else | 22:30 |
| clarkb | one reason is testability, its easier to test coherent units, the other is readability, you hav ea function that is not doing what it says it does and the last thing is to give a greater degree of control to the calling code | 22:30 |
| mordred | we don't have to yield to things | 22:30 |
| clarkb | this code is hella confusing as is | 22:30 |
| clarkb | as evidenced by the massive amount of churn | 22:30 |
| mordred | we don't need a greater degree of control to the calling code | 22:30 |
| clarkb | so make it less so | 22:30 |
| mordred | the calling code | 22:31 |
| clarkb | mordred: you are essentially saying that you cannot wait for a server without also attaching a floating IP | 22:32 |
| clarkb | that removes control from the calling code | 22:32 |
| mordred | yes | 22:32 |
| clarkb | this is fine if the api was different but its not | 22:32 |
| jeblair | nodepool will never want a server without an ip address | 22:32 |
| clarkb | the api as exposed is these are discrete things | 22:32 |
| mordred | becaus we never want the other hting IN NODEPOOL | 22:32 |
| jeblair | there is no api | 22:32 |
| clarkb | jeblair: shade is an api | 22:32 |
| mordred | shade already has this split | 22:32 |
| jeblair | clarkb: we're talking about nodepool | 22:32 |
| mordred | you're talking about the consumption code in nodepool | 22:32 |
| clarkb | nevermind its not worth this | 22:32 |
| clarkb | it was a small suggestion that got exploded | 22:33 |
| *** xyang1 has quit IRC | 22:33 | |
| clarkb | my gripe is that the function does not do what it says it does. This is confusing. Confusion in this code hsa led to many bugs and revisions in this change. If we don't want to change it thats fine | 22:33 |
| jeblair | k, i just installed that shade and nodepool on my dsvm, running it now | 22:33 |
| mordred | sorry - I missed the small part - it sounded like (and still does, because maybe I'm just missing it) "completely rework the API" | 22:33 |
| *** nijaba has quit IRC | 22:33 | |
| clarkb | mordred: no literally bubble that up one level | 22:33 |
| clarkb | mordred: so that the code is less consfusing and easier to read an test and debug | 22:34 |
| mordred | clarkb: let me make a non-related follow up patch that does what I think you're saying | 22:34 |
| mordred | just so that we can be on the same page | 22:34 |
| clarkb | or maybe rename the function you are calling | 22:34 |
| clarkb | so that in the caller it makes sense that "this will attach a floating IP for you" | 22:34 |
| clarkb | because right now that is not apparently at all | 22:34 |
| mordred | I _think_ I know what you're saying - but it'll be easier to figure that out with a patch | 22:34 |
| *** signed8bit has quit IRC | 22:35 | |
| *** nijaba has joined #openstack-infra | 22:36 | |
| jeblair | clarkb, mordred: waitForServer() is called 3 times in nodepool. i would rather not add 3 calls to "andAlsoEnsureItHasAnIP()", so i favor changing the name to "waitForServerToBeActiveAndHaveAnIP" if that satisfies everyone :) | 22:36 |
| *** geoffarn_ has quit IRC | 22:37 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Split adding a network https://review.openstack.org/230188 | 22:37 |
| jeblair | (actual names may vary, use as directed) | 22:37 |
| clarkb | the reason I think this is important is we glossed over floating IPs and it broke this | 22:37 |
| mordred | clarkb: I tink that is what you were saying - is it? | 22:37 |
| clarkb | and we are glossing over them again | 22:37 |
| *** geoffarnold has joined #openstack-infra | 22:37 | |
| *** dimtruck is now known as zz_dimtruck | 22:38 | |
| clarkb | mordred: ya that or jeblairs suggestion | 22:38 |
| clarkb | mordred: because the change as written now is just like before in that you don't really know what the floating Ip is attached | 22:38 |
| anteaya | jeblair: you do have great names | 22:38 |
| mordred | clarkb: right. but I prefer that. I WANT the floating ip creation to be hidden and handled by the library becuase it's a stupid thing to need to do | 22:39 |
| mordred | clarkb: HOWEVER | 22:39 |
| jeblair | clarkb: i don't agree with your analysis of the errors we are fixing, but i don't feel it's worth analyzing. i don't think that helps or hinders your valid point about having code be clear. | 22:39 |
| mordred | clarkb: nodepool has a really specific caching need that we needed account for in shade | 22:39 |
| jeblair | clarkb: (rather, your analysis of the underlying cause of the errors) | 22:39 |
| jeblair | clarkb, mordred: i think my dsvm instance is building a snapshot image... | 22:41 |
| jeblair | clarkb, mordred: at least, it no longer seems to be in a loop | 22:41 |
| mordred | jeblair: that's exciting! | 22:41 |
| jeblair | oh yeah, it's doing node prep now. | 22:41 |
| clarkb | jeblair: it should be able to boot a dib image real quickly too | 22:41 |
| clarkb | since I assume the dib image already exists on disk | 22:41 |
| *** tlian has quit IRC | 22:42 | |
| jeblair | clarkb, mordred: new error in the etherpad https://etherpad.openstack.org/p/nodepool-shade | 22:42 |
| clarkb | mordred: sure I think caching is orthogonal. I just think that if nodepool needs to be responsible it should be clear in nodepool that that is happening | 22:42 |
| clarkb | mordred: then if/when nodepool does not need to be responsible we care less | 22:43 |
| jeblair | clarkb: i can agree with that | 22:43 |
| clarkb | jeblair: thats odd since I assume name was the key used back when we directly consumed nova client | 22:43 |
| mordred | jeblair: line 300 in provider manager ... one sec | 22:43 |
| dansmith | fungi: clarkb: we have a new gate blocker I think | 22:43 |
| clarkb | dansmith: the Pillow libjpeg thing? | 22:44 |
| dansmith | clarkb: yeah, already know? | 22:44 |
| clarkb | dansmith: ya there is a global requirements patch up and a change to the infra image building to install the needed dev headers | 22:44 |
| dansmith | clarkb: right on, thanks | 22:44 |
| jeblair | anteaya found it like an hour ago :) | 22:44 |
| dansmith | jeblair: well, thanks for deflating my balloon! | 22:44 |
| clarkb | dansmith: so hopefully will be all better soon | 22:44 |
| jeblair | dansmith: too slow! :) | 22:44 |
| dansmith | clarkb: sweet | 22:45 |
| dansmith | jeblair: yeah, more coffee or cocaine I guess | 22:45 |
| clarkb | dansmith: in this part of the world I believe meth and coffee are preferred | 22:45 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 22:45 |
| dansmith | clarkb: fair point, yeah | 22:45 |
| *** doug-fish has quit IRC | 22:45 | |
| *** sambetts has quit IRC | 22:45 | |
| *** dprince has quit IRC | 22:46 | |
| *** doug-fish has joined #openstack-infra | 22:46 | |
| *** dramalho has joined #openstack-infra | 22:46 | |
| *** Goneri has quit IRC | 22:47 | |
| mordred | oh! wait - no - that's a bad patch | 22:47 |
| jeblair | mordred: too late already running! | 22:47 |
| mordred | jeblair: neat! | 22:48 |
| clarkb | mordred: image_name appears t be the correct name based on the old code | 22:48 |
| jeblair | mordred: yeah, it's failing unit tests | 22:48 |
| *** geoffarnold is now known as geoffarnoldX | 22:48 | |
| mordred | yup. gimme a sec - I was too quick on the draw | 22:48 |
| *** sambetts has joined #openstack-infra | 22:48 | |
| jeblair | mordred: (i think that was one of the things i fixed yesterday, so you must _really_ want that to be the other thing) | 22:48 |
| *** DinaBelova has quit IRC | 22:49 | |
| *** hashar has joined #openstack-infra | 22:50 | |
| *** ddieterly has quit IRC | 22:50 | |
| mordred | jeblair: I really want many things | 22:50 |
| *** alazarev has quit IRC | 22:51 | |
| *** SergeyLukjanov has quit IRC | 22:51 | |
| *** mriedem has joined #openstack-infra | 22:51 | |
| *** SergeyLukjanov has joined #openstack-infra | 22:51 | |
| *** ybathia has quit IRC | 22:51 | |
| *** DinaBelova has joined #openstack-infra | 22:52 | |
| *** signed8bit has joined #openstack-infra | 22:52 | |
| *** alazarev has joined #openstack-infra | 22:52 | |
| openstackgerrit | Dmitry Borodaenko proposed openstack-infra/project-config: per-project -core and -release groups for Fuel https://review.openstack.org/230195 | 22:52 |
| anteaya | replied to the infra ml about Pillow | 22:53 |
| *** FallenPegasus has joined #openstack-infra | 22:53 | |
| clarkb | anteaya: me too :) | 22:53 |
| anteaya | ha ha ha | 22:54 |
| anteaya | jinx | 22:54 |
| anteaya | bet your response was better than mine | 22:54 |
| anteaya | I'm for asking for this patch to wait until after the mass rename: https://review.openstack.org/#/c/230195/ | 22:54 |
| anteaya | any counter arguments? | 22:54 |
| anteaya | it creates 15 new fuel acl files | 22:55 |
| jhesketh | Morning | 22:55 |
| anteaya | presumably with new gerrit groups at about *2 per acl file | 22:55 |
| anteaya | morning jhesketh | 22:55 |
| *** vmbrasseur is now known as vmb_out | 22:56 | |
| timrc | Nothing like a little Pillow talk. | 22:56 |
| anteaya | ooooohhhhhh | 22:56 |
| anteaya | kevinbenton: had the best pictures in -neutron earlier | 22:56 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Plumb wait and timout down to add_auto_ip https://review.openstack.org/230178 | 22:58 |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Pass parameters correctly for image snapshots https://review.openstack.org/230196 | 22:58 |
| *** amuller has quit IRC | 22:58 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 22:58 |
| *** signed8bit has quit IRC | 22:58 | |
| mordred | jeblair: ok. I verified that against the novaclient source | 22:58 |
| mordred | Shrews: ^^ we're missing test coverage on image snapshot calls | 22:59 |
| *** pradk has quit IRC | 22:59 | |
| *** geoffarnold has joined #openstack-infra | 22:59 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:59 | |
| *** geoffarnoldX has quit IRC | 22:59 | |
| clarkb | mordred: I am not sure that latest patch is right | 23:00 |
| clarkb | mordred: I think it is image_name not name | 23:00 |
| clarkb | I guess the tests will tell us | 23:01 |
| mordred | clarkb: it is image_name to novaclient | 23:01 |
| mordred | it's name to shade | 23:01 |
| clarkb | mordred: then why did name to shade break previously? | 23:01 |
| mordred | because shade was passing name on to the client | 23:01 |
| mordred | rather than passing image_name | 23:01 |
| clarkb | oh we need that shade patch | 23:01 |
| mordred | yeah | 23:01 |
| mordred | it was a shade bug, not a nodepool bug | 23:01 |
| clarkb | I see | 23:02 |
| anteaya | jhesketh: feel like reviewing a dependency addition for images (which clarkb is currently building with the new dependency)? https://review.openstack.org/#/c/230175/ | 23:02 |
| doug-fish | fungi: are you avail to chat about http://lists.openstack.org/pipermail/openstack-dev/2015-October/076020.html | 23:02 |
| anteaya | doug-fish: he is ensuring his home is hurricane prepared | 23:02 |
| anteaya | doug-fish: can anyone else help? | 23:02 |
| doug-fish | maybe | 23:02 |
| doug-fish | is what he says in the email really true? | 23:02 |
| clarkb | doug-fish: yes | 23:02 |
| *** stevemar has quit IRC | 23:02 | |
| anteaya | I haven't known fungi to ever lie about anything | 23:03 |
| doug-fish | I know it's true for jobs that run after approval, but is that what happens during a review? | 23:03 |
| *** dizquierdo has joined #openstack-infra | 23:03 | |
| doug-fish | :-) | 23:03 |
| clarkb | zuul is interested in testing the state of a review if it would merge regardless of approval state | 23:03 |
| doug-fish | I was thinking more mistake/wrong context | 23:03 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/nodepool: Update nodepool to use secure.conf file https://review.openstack.org/189762 | 23:04 |
| doug-fish | wow, I can't believe I've had the wrong model in my head all of this time without noticing! | 23:04 |
| *** tjones has quit IRC | 23:04 | |
| mordred | doug-fish: zuul creates a virtual queue of changes and tests them as if each will pass, stacking them on top of each other | 23:04 |
| anteaya | doug-fish: fungi is fairly ocd about being accurate | 23:04 |
| doug-fish | I'd have *sworn* some patches had to be rebased after a fix to a test | 23:04 |
| mordred | doug-fish: if one fails, it ejects it from the queue and re-enqueues everytihng | 23:04 |
| mordred | doug-fish: nope. it's the reason we wrote zuul | 23:04 |
| doug-fish | well thanks for the clarification. | 23:05 |
| mordred | sure thing! | 23:05 |
| mordred | doug-fish: we have a lovely animated slide if you want to see it | 23:05 |
| doug-fish | mordred: sure | 23:05 |
| mordred | doug-fish: http://docs.openstack.org/infra/publications/zuul/#%2818%29 | 23:05 |
| doug-fish | thanks | 23:05 |
| mordred | (click in the browser window to advance the animation) | 23:05 |
| *** chlong has quit IRC | 23:07 | |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: update grenade branch logic for mitaka https://review.openstack.org/229363 | 23:08 |
| ruagair | Does anyone know if Storyboard has gone through any schema changes since April? | 23:08 |
| ruagair | (the database schema) | 23:08 |
| *** asettle has joined #openstack-infra | 23:08 | |
| ruagair | In particular dropping the "username" column from the storyboard.users table? | 23:08 |
| *** lawrancejing has joined #openstack-infra | 23:08 | |
| *** hashar has quit IRC | 23:08 | |
| ruagair | morning asettle :-D | 23:08 |
| clarkb | ruagair check the git log? | 23:09 |
| * ruagair prods mordred and jeblair for the above comments | 23:09 | |
| jhesketh | anteaya: lgtm | 23:09 |
| mordred | ruagair: no idea, sorry | 23:09 |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: DNM: test devstack work around https://review.openstack.org/230108 | 23:09 |
| ruagair | thanks mordred, clarkb. I'll wander through the git log. | 23:10 |
| *** mriedem has quit IRC | 23:10 | |
| *** mriedem has joined #openstack-infra | 23:10 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Builders no longer use nodedb for diskimages https://review.openstack.org/222398 | 23:11 |
| *** dkranz has joined #openstack-infra | 23:11 | |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: DNM: test oslo.service backports https://review.openstack.org/230101 | 23:11 |
| anteaya | jhesketh: thank you | 23:12 |
| *** SumitNaiksatam has quit IRC | 23:14 | |
| *** SumitNaiksatam has joined #openstack-infra | 23:14 | |
| *** SumitNaiksatam has quit IRC | 23:15 | |
| *** mriedem has quit IRC | 23:15 | |
| openstackgerrit | Merged openstack-infra/system-config: Add Pillow C deps to thick slaves https://review.openstack.org/230175 | 23:15 |
| asselin | ruagair, Zara and SotK might konw | 23:16 |
| ruagair | Thanks asselin | 23:16 |
| *** yamamoto has joined #openstack-infra | 23:17 | |
| ruagair | I'm wading through mordred 's phab-migration script and number of fields referenced in storybook.users no longer appear to be present (if they ever were), so i'm seeking context for those changes. | 23:17 |
| mordred | ruagair: ah - nod | 23:18 |
| mordred | I promise that script worked at a point in the past | 23:18 |
| *** lawrancejing has quit IRC | 23:18 | |
| *** shashank_hegde has joined #openstack-infra | 23:19 | |
| ruagair | Oh, I can see that it did. It's only failing on missing fields like "username" which appear to have no current equivalent. | 23:19 |
| * ruagair will submit a patch once he's got it sorted. | 23:19 | |
| mordred | ruagair: http://paste.openstack.org/show/475129/ is the table structure from the last dump I had | 23:20 |
| mordred | fwiw | 23:20 |
| mordred | prolly not helpful | 23:20 |
| ruagair | Should be helpful. Thanks mordred | 23:20 |
| mordred | because historical state of a db is ... well... | 23:20 |
| mordred | you knoea | 23:20 |
| ruagair | Yup. | 23:20 |
| ruagair | I'm pretty certain I'm just going to have to make hacks til it works but I like a little historical context :-D | 23:21 |
| ruagair | Unless jeblair removed the username field when the dump was sanitised? | 23:21 |
| *** ashleighfarnham has quit IRC | 23:22 | |
| *** mwagner_ has joined #openstack-infra | 23:22 | |
| *** fbo has quit IRC | 23:22 | |
| jeblair | ruagair: i don't think so; i think i only excluded some tables with auth tokens | 23:24 |
| jeblair | ruagair: i can run a 'show create table' for you in prod if you like -- let me know what table | 23:25 |
| *** mriedem has joined #openstack-infra | 23:25 | |
| jeblair | mordred, clarkb: running shade+nodepool on dsvm now | 23:25 |
| ruagair | Just wanted to check that storyboard.users is missing "username" in production as it is in my copy, jeblair. | 23:28 |
| *** dingyichen has joined #openstack-infra | 23:29 | |
| jeblair | ruagair: looks like it: http://paste.openstack.org/show/475130/ | 23:30 |
| *** markvoelker has quit IRC | 23:30 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/nodepool: Update nodepool to use secure.conf file https://review.openstack.org/189762 | 23:30 |
| jeblair | mordred: new error in etherpad https://etherpad.openstack.org/p/nodepool-shade | 23:31 |
| jeblair | mordred: i have to eod. feel free to use that host if you want. | 23:31 |
| openstackgerrit | Derek Higgins proposed openstack/diskimage-builder: Ignore ./proc when deleting python object files https://review.openstack.org/230202 | 23:31 |
| mordred | jeblair: ok. will do | 23:31 |
| mordred | at least it's a new error | 23:32 |
| *** johnsom has quit IRC | 23:32 | |
| *** dizquierdo has quit IRC | 23:32 | |
| anteaya | jeblair: enjoy evening | 23:32 |
| *** johnsom has joined #openstack-infra | 23:32 | |
| ruagair | Awesome, thanks jeblair. That confirms a schema change since mordred wrote the stript and I'll have to adapt it. Thanks! | 23:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool: Use shade for all OpenStack interactions https://review.openstack.org/226751 | 23:34 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/nodepool: Update nodepool to use secure.conf file https://review.openstack.org/189762 | 23:35 |
| mordred | jeblair: fwiw, something on the node will not let me ssh in | 23:36 |
| *** FallenPegasus has quit IRC | 23:37 | |
| *** tjones has joined #openstack-infra | 23:38 | |
| clarkb | mordred: you hvae an account on it | 23:38 |
| clarkb | mordred: I am able to log in | 23:38 |
| *** baoli has quit IRC | 23:39 | |
| *** gyee has joined #openstack-infra | 23:39 | |
| clarkb | I am not seeing anything in hte auth.log for you | 23:40 |
| *** geoffarnold has quit IRC | 23:41 | |
| *** geoffarnold has joined #openstack-infra | 23:41 | |
| *** yamamoto has quit IRC | 23:42 | |
| *** FallenPegasus has joined #openstack-infra | 23:43 | |
| *** hemna is now known as hemnafk | 23:44 | |
| *** achanda has quit IRC | 23:47 | |
| *** ybathia has joined #openstack-infra | 23:48 | |
| *** Daisy has joined #openstack-infra | 23:50 | |
| mordred | Timeout, server 23.253.110.184 not responding. | 23:51 |
| mordred | clarkb: I get that | 23:51 |
| mordred | clarkb: oh! weird. I'm in now | 23:51 |
| clarkb | mordred: thats not the same host I see | 23:51 |
| *** ddieterly has joined #openstack-infra | 23:52 | |
| clarkb | mordred: https://jenkins07.openstack.org/job/gate-dsvm-nodepool/26/console is what I ssh'd to | 23:52 |
| *** bltavares has quit IRC | 23:52 | |
| mordred | clarkb: 23.253.110.184 is the one jeblair was on | 23:52 |
| mordred | I think I'm going to need to eat food before I sort out re-running the nodepool tests though | 23:53 |
| clarkb | oh I was looking at the one running your latest patch | 23:53 |
| mordred | my brain is heading downhill | 23:53 |
| mordred | clarkb: : | 23:53 |
| mordred | clarkb: :) | 23:53 |
| mordred | clarkb: too many patches | 23:53 |
| *** achanda has joined #openstack-infra | 23:54 | |
| *** bltavares has joined #openstack-infra | 23:54 | |
| *** Goneri has joined #openstack-infra | 23:56 | |
| *** dims has joined #openstack-infra | 23:56 | |
| *** bnemec has joined #openstack-infra | 23:57 | |
| *** bltavares has quit IRC | 23:58 | |
| *** hdd has quit IRC | 23:58 | |
| *** dims has quit IRC | 23:58 | |
| *** dims_ has quit IRC | 23:58 | |
| *** dims has joined #openstack-infra | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
