| *** rbrndt has quit IRC | 00:00 | |
| *** ijw has quit IRC | 00:02 | |
| *** rlandy|rover is now known as rlandy|rover|bbl | 00:06 | |
| *** salv-orlando has quit IRC | 00:12 | |
| *** salv-orlando has joined #openstack-infra | 00:13 | |
| jeblair | EmilienM: i've looked at 522667 and related changes and i also don't know why that isn't working | 00:16 |
|---|---|---|
| jeblair | infra-root: i'd like to restart zuulv3 now with the repl installed to try to debug that further ^ | 00:16 |
| *** salv-orlando has quit IRC | 00:17 | |
| mordred | jeblair: ++ | 00:17 |
| jeblair | restarting now | 00:18 |
| openstackgerrit | K Jonathan Harker proposed openstack/diskimage-builder master: zypper: fix package removal https://review.openstack.org/517658 | 00:20 |
| openstackgerrit | K Jonathan Harker proposed openstack/diskimage-builder master: Install dpkg on yum-based systems in testing https://review.openstack.org/523251 | 00:20 |
| EmilienM | jeblair: good to know | 00:23 |
| jeblair | post-restart re-enqueue is finishing up now. i'll dig into the problem first thing tomorrow | 00:24 |
| *** ijw has joined #openstack-infra | 00:30 | |
| *** hongbin has quit IRC | 00:33 | |
| *** stakeda has joined #openstack-infra | 00:33 | |
| *** jkilpatr has quit IRC | 00:36 | |
| *** raissa has quit IRC | 00:43 | |
| *** raissa has joined #openstack-infra | 00:44 | |
| mordred | AJaeger: whence you awaken, I've pushedup a few more revs to updated-pti patches andhave made some notes in the etherpad | 00:44 |
| *** jascott1 has quit IRC | 00:44 | |
| *** jkilpatr has joined #openstack-infra | 00:45 | |
| *** Kevin_Zheng has joined #openstack-infra | 00:46 | |
| clarkb | pabelanger: dmsimard how are we supposed to explicitly enable epel in jobs now? | 00:47 |
| clarkb | pabelanger: dmsimard specifically http://logs.openstack.org/51/523251/1/check/dib-dsvm-functests-python2-centos-7/443cb4c/job-output.txt.gz#_2017-11-28_00_33_49_002799 needs to execute with epel enabled | 00:47 |
| clarkb | yum -y install --enablerepo=epel ? | 00:48 |
| pabelanger | yah, it is installed by disabled | 00:48 |
| dmsimard | There's an arg to do it automatically | 00:49 |
| dmsimard | iirc openstack-ansible does it | 00:49 |
| clarkb | it broke dib functional tests because dpkg is in epel | 00:49 |
| * clarkb will be updating jesusaur's patch in a sec | 00:49 | |
| openstackgerrit | Clark Boylan proposed openstack/diskimage-builder master: Install dpkg on yum-based systems in testing https://review.openstack.org/523251 | 00:50 |
| *** jkilpatr has quit IRC | 00:50 | |
| dmsimard | clarkb: http://docs.ansible.com/ansible/latest/yum_module.html see enablerepo | 00:51 |
| clarkb | I don't see email about the change to our images. Did we send anything out? | 00:51 |
| clarkb | (I know other projects have been affected too) | 00:51 |
| *** dklyle has quit IRC | 00:54 | |
| clarkb | I'm rebooting logstash-worker03 as it is out to lunch | 00:58 |
| clarkb | I think similar to some of the other hosts we've seen rencelty | 00:58 |
| pabelanger | clarkb: epel has been removed for a few months, maybe it was centos-release-openstack-ocata that is now missing | 00:59 |
| pabelanger | or the tests have been broken that long | 00:59 |
| *** david-lyle has joined #openstack-infra | 01:01 | |
| clarkb | pabelanger: I doubt they've been broken for very long since ianw and others have been doing dib work. Could be something else | 01:01 |
| *** cuongnv has joined #openstack-infra | 01:01 | |
| ianw | it might have been broken for a few days | 01:02 |
| *** jascott1 has joined #openstack-infra | 01:03 | |
| *** jkilpatr has joined #openstack-infra | 01:03 | |
| clarkb | I think it was manilla also had problems because no erlang or rabbitmq in the base repos | 01:03 |
| clarkb | that seems more likely to be the rdo repo removal though | 01:03 |
| *** s-shiono has joined #openstack-infra | 01:06 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/zuul-jobs master: Move to dictionary list of projects zuul._projects (take 2) https://review.openstack.org/518815 | 01:09 |
| ianw | jeblair: now "tested" ... will comment | 01:09 |
| *** jkilpatr has quit IRC | 01:09 | |
| jeblair | ianw: cool, thanks. i'll try to catch up on that tomorrow | 01:10 |
| *** jaypipes has quit IRC | 01:10 | |
| *** jkilpatr has joined #openstack-infra | 01:11 | |
| clarkb | logstash-worker03 is back up and running workers | 01:11 |
| clarkb | I think starting tomorrow we can work to add in new files to indexing and possibly go back to 10 days of indexed logs | 01:12 |
| clarkb | will have to look at disk usage | 01:12 |
| *** salv-orlando has joined #openstack-infra | 01:14 | |
| *** salv-orlando has quit IRC | 01:19 | |
| *** liujiong has joined #openstack-infra | 01:23 | |
| *** armax has quit IRC | 01:27 | |
| *** Guest888 has joined #openstack-infra | 01:29 | |
| *** aeng has quit IRC | 01:33 | |
| *** dave-mccowan has joined #openstack-infra | 01:36 | |
| *** jkilpatr has quit IRC | 01:38 | |
| *** s-shiono_ has joined #openstack-infra | 01:39 | |
| *** Guest888 has quit IRC | 01:41 | |
| *** s-shiono has quit IRC | 01:43 | |
| *** esberglu has joined #openstack-infra | 01:46 | |
| *** sdague has quit IRC | 01:47 | |
| *** mriedem has joined #openstack-infra | 01:48 | |
| *** aeng has joined #openstack-infra | 01:49 | |
| *** esberglu has quit IRC | 01:50 | |
| *** hughhalf has quit IRC | 01:51 | |
| *** hughhalf has joined #openstack-infra | 01:53 | |
| *** yolanda has quit IRC | 01:54 | |
| *** Goneri has quit IRC | 01:57 | |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Added get unpublished events endpoint https://review.openstack.org/523274 | 02:02 |
| *** camunoz has quit IRC | 02:02 | |
| *** ianychoi has quit IRC | 02:03 | |
| *** annp has joined #openstack-infra | 02:10 | |
| *** salv-orlando has joined #openstack-infra | 02:15 | |
| *** zhurong has joined #openstack-infra | 02:17 | |
| *** yolanda has joined #openstack-infra | 02:18 | |
| *** salv-orlando has quit IRC | 02:21 | |
| *** Apoorva_ has joined #openstack-infra | 02:21 | |
| *** yamamoto has joined #openstack-infra | 02:22 | |
| *** yolanda has quit IRC | 02:24 | |
| *** Apoorva has quit IRC | 02:25 | |
| *** Apoorva_ has quit IRC | 02:25 | |
| *** hongbin has joined #openstack-infra | 02:27 | |
| *** gcb has joined #openstack-infra | 02:27 | |
| *** wangqian has joined #openstack-infra | 02:31 | |
| *** esberglu has joined #openstack-infra | 02:36 | |
| *** esberglu has quit IRC | 02:36 | |
| *** esberglu has joined #openstack-infra | 02:37 | |
| *** mat128 has joined #openstack-infra | 02:38 | |
| wangqian | Hello everyone | 02:39 |
| wangqian | I downloaded the diskimage-builder project. | 02:39 |
| wangqian | I want to build a image locally instead of connecting the network. | 02:39 |
| wangqian | What should I do ？ | 02:39 |
| *** erlon has quit IRC | 02:41 | |
| *** esberglu has quit IRC | 02:41 | |
| *** yolanda has joined #openstack-infra | 02:41 | |
| *** ramishra has joined #openstack-infra | 02:45 | |
| ianw | wangqian: your best bet is to look in project-config | 02:47 |
| ianw | tools/build-image.sh | 02:48 |
| *** iyamahat has quit IRC | 02:53 | |
| *** yamahata has quit IRC | 02:53 | |
| wangqian | @ianw I found only the tox_install.sh file in the Tools Directory | 02:54 |
| ianw | wangqian: you want to clone git://git.openstack.org/openstack-infra/project-config and look in there | 02:55 |
| *** sam_wan has joined #openstack-infra | 02:55 | |
| *** gyee_ has quit IRC | 02:55 | |
| *** rlandy|rover|bbl is now known as rlandy|rover | 02:57 | |
| wangqian | ianw: Thx I will read it | 02:57 |
| *** ijw has quit IRC | 02:59 | |
| *** dhinesh has quit IRC | 03:01 | |
| *** ijw has joined #openstack-infra | 03:02 | |
| dmsimard | ianw: this pattern of micro roles drives me nuts :P | 03:10 |
| dmsimard | one role to install the thing, one role to configure the thing and one role to run the thing | 03:10 |
| dmsimard | I agree on your sentiment that if unbound isn't actually installed and pre-configured as currently done in the nodepool base image it likely wouldn't work | 03:11 |
| ianw | dmsimard: well i don't really disagree, and if we started everything fresh we wouldn't do it like this ... but yeah, as i said, there's really no point in even installing it, it won't be configured | 03:11 |
| *** dave-mccowan has quit IRC | 03:12 | |
| dmsimard | I could say the same about the directory creation that was there before | 03:12 |
| dmsimard | If the directory isn't there you have other problems | 03:12 |
| *** dave-mccowan has joined #openstack-infra | 03:12 | |
| ianw | it would be fine as an assert | 03:13 |
| dmsimard | We discussed moving the unbound installation/configuration outside of nodepool-builder but it seems like some providers are not expected to provide working DNS resolution out of the box | 03:13 |
| ianw | right, that was how we ended up like this with it split | 03:13 |
| dmsimard | So DNS needs to be pre-configured to some extent if we're interested in installing it later | 03:13 |
| ianw | i probably should have called it "dynamic-" which might have made it even clearer what's going on | 03:13 |
| dmsimard | It's out of scope to me to do that, at least for now | 03:14 |
| dmsimard | ianw: I think weighing the balance of what we move out the image and keep inside is an interesting topic | 03:14 |
| *** coolsvap has joined #openstack-infra | 03:15 | |
| dmsimard | Moving something out of the image means we're going to be doing it >10k times a day so it definitely has a cost | 03:15 |
| dmsimard | But at the same time, (although this role is in openstack-zuul-jobs, I'm speaking in general) we'd like zuul-jobs to be generic enough to be widely used outside of OpenStack | 03:15 |
| dmsimard | yet there's a bunch of assumptions in there, some assumes pip and/or tox is installed, or even that you're running on ubuntu, etc | 03:16 |
| *** dave-mcc_ has joined #openstack-infra | 03:16 | |
| ianw | right, it's always been a bit of tension | 03:16 |
| ianw | dib makes for a terrible, terrible ansible/puppet | 03:17 |
| *** salv-orlando has joined #openstack-infra | 03:17 | |
| dmsimard | ianw: For me, the configure-unbound role could be a generic "dns-cache" role that would live in zuul-jobs. Users want a dns-cache, they include it, it works with sane defaults out of the box. | 03:17 |
| ianw | pabelanger has had some stabs at making an ansible element, so we're not writing dodgy custom script elements | 03:17 |
| dmsimard | ianw: oh, actually, you know, that's not a bad idea at all | 03:18 |
| ianw | obviously we used to run puppet on the images in dib as part of daily builds, but have excised that | 03:18 |
| pabelanger | I did, I should rebase that and get simple-ansible going again | 03:18 |
| dmsimard | pabelanger, ianw: some of the roles from zuul-jobs/openstack-zuul-jobs could run on the image | 03:18 |
| dmsimard | at build-time | 03:18 |
| *** dave-mccowan has quit IRC | 03:18 | |
| pabelanger | Yup, we talked about that maybe 18months ago at midcycle in germany | 03:19 |
| ianw | yes ... but that was essentially what we did with puppet too. the push is to make the images dumb and the ansible smart ... so yeah. there's no fast rule, i think it's up to us to use good taste when evaluating options | 03:19 |
| dmsimard | Well, it's not about ansible, or puppet, or bash or DIB | 03:19 |
| pabelanger | https://review.openstack.org/385608/ | 03:19 |
| pabelanger | I'll rebase it tomorrow, and test agian | 03:20 |
| pabelanger | again* | 03:20 |
| clarkb | in the dns case for me that is pretty clearly an image thing because image should have working dns | 03:20 |
| clarkb | (at least imo) | 03:20 |
| dmsimard | It's about running something >10k times a day when you could embed it in the image -- even if it costs 5 seconds that's 13 hours worth of computing | 03:20 |
| clarkb | its independwnt of running jobs, you want dns to work on boot | 03:20 |
| pabelanger | maybe even get https://github.com/ansible/ansible/issues/12728 working with chroot | 03:20 |
| clarkb | same with the firewall | 03:20 |
| dmsimard | clarkb: what's stopping us from just putting the 4 nameservers in /etc/resolv.conf ? | 03:20 |
| clarkb | dmsimard: we put one there today | 03:21 |
| clarkb | whats wrong with that? | 03:21 |
| dmsimard | clarkb: but you missed the part of the discussion where I suggested we ran the role on the image | 03:21 |
| dmsimard | clarkb: so that there's no delta and there's also no assumption | 03:21 |
| clarkb | ? | 03:21 |
| clarkb | delta or assumption? | 03:21 |
| clarkb | this us purely about make an image that has base working functionality | 03:21 |
| clarkb | so ssh, dns, firewall | 03:22 |
| clarkb | (we've had to tweak things with haveged too for ssh host key generation) | 03:22 |
| clarkb | basically those things belong in the image imo | 03:23 |
| pabelanger | and xenial intropy pump | 03:23 |
| *** ijw has quit IRC | 03:23 | |
| dmsimard | clarkb: right, but what I'm trying to say is that right now this is embedded in the images and this makes some things weird or broken in zuul-jobs when you're trying to consume it from outside OpenStack nodepool images. | 03:23 |
| *** salv-orlando has quit IRC | 03:23 | |
| clarkb | why? | 03:23 |
| dmsimard | clarkb: so pretend we had a role in zuul-jobs instead that'd do dns, ssh and firewall, and then we ran those roles in the images | 03:23 |
| clarkb | why would dns do that? | 03:23 |
| clarkb | computers assume working dns for the most part tpday so I'd be surprised if something conflicted with that | 03:24 |
| clarkb | I'd also say anything in zuul-jobs that requires a specific dns setup is buggy | 03:25 |
| dmsimard | The discussion spawned from ianw's comment in https://review.openstack.org/#/c/523228/ | 03:25 |
| clarkb | things should work with any reasonable internet facing resolver | 03:25 |
| dmsimard | There is likely a use case for a generic role that would set up a local dns cache in zuul-jobs, which we happen to be doing half in nodepool, half in openstack-zuul-jobs | 03:26 |
| clarkb | ah ok so the real issue here is the late "binding" of resolverz | 03:26 |
| dmsimard | This local dns cache role in zuul-jobs could hypothetically run in the image and this would also mean that users outside of OpenStack could leverage that | 03:26 |
| clarkb | we could bake that into the image instead if we like | 03:27 |
| dmsimard | and we could take the same approach for other components, make them re-usable | 03:27 |
| dmsimard | *exactly* | 03:27 |
| dmsimard | bake that into the image, yes, but by running a generic role :D | 03:27 |
| clarkb | the problem is it has to run post boot | 03:27 |
| dmsimard | oh ? | 03:27 |
| clarkb | so bake into the image means bake in a script/systemd unit/whatever that does it | 03:27 |
| clarkb | yes the whole point us it cant know what resolvers to use until networking is configured | 03:28 |
| clarkb | so you cant hardocde in the image | 03:28 |
| clarkb | but you could have the image configure itself post boot | 03:28 |
| dmsimard | well right now we already hammer the four nameservers in right ? | 03:28 |
| clarkb | no just two iirc | 03:28 |
| dmsimard | the configure-unbound role did take a different approach and only configure v4 or v6 | 03:28 |
| dmsimard | nope, I looked at it today :D | 03:29 |
| clarkb | because it should be ipv4 xor ipv6 | 03:29 |
| dmsimard | http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/nodepool-base/finalise.d/89-unbound#n29 | 03:29 |
| clarkb | but ehat do we do late? | 03:29 |
| clarkb | because thats overridden | 03:30 |
| pabelanger | I want to say we drop in both ipv4 / ipv6 in DIB, and if ipv6 is missing, unbound fails back to ipv4 | 03:30 |
| dmsimard | clarkb: configure-unbound checks if the default route is ipv6 | 03:30 |
| dmsimard | and configures accordingly | 03:30 |
| dmsimard | but yeah like pabelanger mentions, it doesn't seem to matter to unbound | 03:30 |
| dmsimard | ¯\_(ツ)_/¯ | 03:30 |
| clarkb | well it round robins them | 03:30 |
| clarkb | iirc it just loops around them | 03:30 |
| clarkb | until one answers | 03:31 |
| clarkb | problem is when ipv4 is nated | 03:31 |
| clarkb | and the failures are more faily | 03:31 |
| dmsimard | there's actually a setting in unbound to enable v4 and v6 | 03:31 |
| dmsimard | we don't seem to be touching those | 03:32 |
| dmsimard | not sure if we should | 03:32 |
| *** sree has joined #openstack-infra | 03:32 | |
| clarkb | dmsimard: thats on the listen side or the resolving side? | 03:32 |
| clarkb | but anyways the problem being addressed is you cant set the config until instamce is booted | 03:32 |
| clarkb | that ckuld be encoded into a boot time script if we want | 03:33 |
| dmsimard | clarkb: looks like both? http://paste.openstack.org/raw/627529/ | 03:33 |
| clarkb | but you cant hardcode the image and expect it to be happy in all cases (ipv6 with natted ipv4being the known problem case) | 03:33 |
| dmsimard | that's fair | 03:34 |
| pabelanger | yah, a lot of that was done in v2 ready-script for nodepool | 03:34 |
| *** ijw has joined #openstack-infra | 03:34 | |
| pabelanger | so, make sense we need to do it in pre-run in zuulv3 | 03:34 |
| *** armax has joined #openstack-infra | 03:34 | |
| clarkb | byt also that wouldnt be zuul-jobs right? | 03:34 |
| clarkb | so I'm not sure I see where there is a conflict with reusability | 03:35 |
| dmsimard | My goal here is to take as much out of the image as we can and make it re-usable by other zuul operators/users, while keeping the ability to embed things inside the image (by running the generic roles at build time) because, like I said, running something >10k times a day has a cost | 03:35 |
| clarkb | zuul-jobs should likely just expect dns to work | 03:35 |
| dmsimard | clarkb: I see a use case for a generic role that sets up dns caching | 03:35 |
| dmsimard | that people could include in base if they case | 03:35 |
| dmsimard | s/case/care/ | 03:35 |
| clarkb | ya I can see that, however I really do think there are fundamental bits you want in the image and that is one of them (working ssh another) | 03:36 |
| clarkb | what is concern with others using the unbound setup as it exists in dib today? | 03:37 |
| clarkb | you cam set your own resolvers | 03:37 |
| dmsimard | it doesn't really exist in dib actually | 03:37 |
| clarkb | maybe we need to document that better? | 03:37 |
| dmsimard | it exists as a project-config nodepool element | 03:37 |
| dmsimard | so they'd need to fork it etc | 03:37 |
| clarkb | that is a dib element | 03:37 |
| dmsimard | it's not provided out of the box if I install dib or nodepool | 03:38 |
| clarkb | we can move it into dib if dib thinks it is appropriate | 03:38 |
| clarkb | it does path style element lookups too so not even an issue to have it in both places during a transition iirc | 03:38 |
| dmsimard | Yeah for me the lines between dib/zuul(-jobs)/ansible are pretty blurry sometimes :) | 03:38 |
| dmsimard | but I was arguably never a fan of dib (sorry) | 03:39 |
| *** links has joined #openstack-infra | 03:39 | |
| clarkb | thats an orthogonal issue | 03:39 |
| * dmsimard nods | 03:39 | |
| clarkb | if you want to make image building easier using the tools we use thats the way to address it | 03:39 |
| *** rlandy|rover has quit IRC | 03:41 | |
| clarkb | fwiw a lot of people seem to think its dib or X rather than dib with X for various values of X like ansible, docker, libguestfs | 03:42 |
| clarkb | but in many cases they are complimentary | 03:42 |
| clarkb | its just a matter of hooking in $tool at appropriate run step or executing it as part of an element | 03:43 |
| *** sree has quit IRC | 03:43 | |
| *** dhinesh has joined #openstack-infra | 03:44 | |
| dmsimard | Don't get me wrong, I know what dib is, what it does and I use it -- I've just been doing too much Ansible and when you have a hammer.. everything is a nail :) | 03:44 |
| clarkb | right my point us nothing stops you from using ansible with dib | 03:44 |
| clarkb | we did dib with puppet for forever | 03:44 |
| dmsimard | but dib elements are basically ansible roles :P | 03:44 |
| dmsimard | in bash | 03:45 |
| clarkb | no they are not | 03:45 |
| clarkb | nothing requires they be bash | 03:45 |
| dmsimard | ok, fair | 03:45 |
| clarkb | they are run parts executed collections of scripts | 03:45 |
| clarkb | which can be ansible | 03:45 |
| clarkb | or puppet | 03:45 |
| dmsimard | which is launched by bash? | 03:45 |
| clarkb | no its python | 03:45 |
| clarkb | or more specifically whatever run parts you use | 03:45 |
| dmsimard | oh? I thought it would be an element that'd do like puppet apply | 03:45 |
| clarkb | dib happens to ship one written in python | 03:46 |
| *** dhinesh_ has joined #openstack-infra | 03:46 | |
| clarkb | dmsimard: the only requirement is that the script start witha digit and be executable iirc | 03:46 |
| clarkb | could be comliled C or Go or rust | 03:47 |
| clarkb | or java even | 03:47 |
| dmsimard | Anyway, I'm not going to ditch DIB tomorrow or anything :) | 03:48 |
| *** dhinesh has quit IRC | 03:48 | |
| dmsimard | wrapping something up and signing off o/ | 03:48 |
| clarkb | not expecting you too just want to make it clear you could have dib set up to do ansible with dib handling the image conversion at the end | 03:48 |
| pabelanger | the approach in https://review.openstack.org/385608/ was you'd run ansible outside the chroot, but you could do just the same by installing ansible in to virtualenv inside chroot, then delete on cleanup.d phase | 03:48 |
| clarkb | bash not required | 03:48 |
| pabelanger | puppet, was harder to do that | 03:48 |
| *** sree has joined #openstack-infra | 03:49 | |
| clarkb | bcause like I said everyone seems to think its dib or X but they can complement each other imo | 03:49 |
| dmsimard | clarkb: I think I'm not knowledgeable enough of what DIB does under the hood, because I can only see "elements" being ansible roles that would be sandwhiched between tasks that create the image, chroot, etc. Ansible even has a chroot connection plugin so it'd be 100% transparent | 03:50 |
| dmsimard | the image/chroot would be a host from the perspective of ansible | 03:50 |
| clarkb | yup thats what pabelanger did iirc | 03:50 |
| pabelanger | dmsimard: review 385608 :) | 03:50 |
| *** yamamoto has quit IRC | 03:51 | |
| dmsimard | more than a year old | 03:51 |
| dmsimard | let's get that in :P | 03:51 |
| clarkb | dmsimard: I think the point would be elemtns are a way if collecting useful bits without caring what the implementation is | 03:51 |
| clarkb | so I can write something that puppets or salts and you can ansible and we dont care | 03:51 |
| clarkb | we just tell dib to runthe element that does what we nees | 03:52 |
| pabelanger | dmsimard: yes, we've had this discussion for 18months or so. Now that zuulv3 is live, and we are using ansible more, I think we can loop back to the discussion again | 03:52 |
| pabelanger | I'll rebase patch in the morning | 03:52 |
| pabelanger | and maybe get ianw to review :) | 03:52 |
| *** yamamoto has joined #openstack-infra | 03:53 | |
| *** yamamoto has quit IRC | 03:53 | |
| ianw | let's get the dib gate working first :) | 03:53 |
| dmsimard | clarkb: btw, what I am describing is the approach from ansible-container https://github.com/ansible/ansible-container | 03:53 |
| dmsimard | which has ansible build docker images (and optionally orchestrate them) | 03:54 |
| pabelanger | yah, I don't want to run ansible-container | 03:54 |
| dmsimard | out of roles/playbooks instead of dockerfiles | 03:54 |
| dmsimard | pabelanger: well it doesn't do anything else than docker (and maybe OCI?) anyway | 03:54 |
| *** yamamoto has joined #openstack-infra | 03:54 | |
| * dmsimard goes back to fight flask in ara unit tests | 03:55 | |
| *** yamamoto has quit IRC | 03:55 | |
| pabelanger | FWIW: it it is much easier to build docker images with DIB and simple-playbook element, the only part I couldn't crack was Docker json file that is added to a tarball | 03:55 |
| pabelanger | someday I will figure that out | 03:55 |
| *** ihrachys has quit IRC | 03:58 | |
| *** sree has quit IRC | 03:59 | |
| *** yamamoto has joined #openstack-infra | 04:05 | |
| *** ijw has quit IRC | 04:08 | |
| *** ykarel has joined #openstack-infra | 04:09 | |
| *** dhinesh_ has quit IRC | 04:11 | |
| *** udesale has joined #openstack-infra | 04:14 | |
| *** salv-orlando has joined #openstack-infra | 04:19 | |
| *** sam_wan has quit IRC | 04:20 | |
| *** raissa_ has joined #openstack-infra | 04:23 | |
| *** kiennt26 has quit IRC | 04:24 | |
| *** dbecker has quit IRC | 04:25 | |
| *** salv-orlando has quit IRC | 04:25 | |
| *** raissa has quit IRC | 04:25 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Use EPEL for debootstrap on centos https://review.openstack.org/523251 | 04:26 |
| ianw | if that one doesn't win the "most confusing confluence of bugs" of the day award, not sure what will | 04:27 |
| *** zhurong has quit IRC | 04:30 | |
| *** gouthamr has quit IRC | 04:31 | |
| *** hongbin has quit IRC | 04:32 | |
| *** psachin has joined #openstack-infra | 04:37 | |
| *** dbecker has joined #openstack-infra | 04:38 | |
| *** rosmaita has quit IRC | 04:39 | |
| *** dave-mcc_ has quit IRC | 04:42 | |
| *** yamamoto has quit IRC | 04:43 | |
| *** sree has joined #openstack-infra | 04:46 | |
| *** dhajare has joined #openstack-infra | 04:51 | |
| *** pgadiya has joined #openstack-infra | 04:51 | |
| *** pgadiya has quit IRC | 04:52 | |
| *** yamamoto has joined #openstack-infra | 04:54 | |
| *** yamamoto has quit IRC | 04:54 | |
| *** yamamoto has joined #openstack-infra | 04:54 | |
| *** raissa_ has quit IRC | 05:06 | |
| *** raissa has joined #openstack-infra | 05:06 | |
| *** raissa has quit IRC | 05:07 | |
| *** janki has joined #openstack-infra | 05:07 | |
| *** raissa has joined #openstack-infra | 05:07 | |
| *** raissa has quit IRC | 05:08 | |
| *** raissa has joined #openstack-infra | 05:08 | |
| *** raissa has quit IRC | 05:08 | |
| *** threestrands has quit IRC | 05:10 | |
| *** threestrands has joined #openstack-infra | 05:10 | |
| *** claudiub|2 has joined #openstack-infra | 05:11 | |
| *** threestrands has quit IRC | 05:12 | |
| *** threestrands has joined #openstack-infra | 05:12 | |
| *** pgadiya has joined #openstack-infra | 05:13 | |
| *** pgadiya has quit IRC | 05:13 | |
| *** dhinesh has joined #openstack-infra | 05:20 | |
| *** salv-orlando has joined #openstack-infra | 05:20 | |
| *** salv-orlando has quit IRC | 05:27 | |
| *** slaweq has joined #openstack-infra | 05:31 | |
| *** sam_wan has joined #openstack-infra | 05:32 | |
| *** jascott1 has quit IRC | 05:35 | |
| *** slaweq has quit IRC | 05:35 | |
| *** jascott1 has joined #openstack-infra | 05:36 | |
| *** esberglu has joined #openstack-infra | 05:37 | |
| *** esberglu has quit IRC | 05:41 | |
| *** dsariel has joined #openstack-infra | 05:44 | |
| *** cshastri has joined #openstack-infra | 05:53 | |
| Jeffrey4l | is it OK to add a squid cache proxy in the infra env? | 05:58 |
| *** armax has quit IRC | 05:58 | |
| Jeffrey4l | then we can download files through that proxy, which could cache files. | 05:59 |
| Jeffrey4l | it is better than reverse proxy | 05:59 |
| *** armax has joined #openstack-infra | 06:01 | |
| *** claudiub|2 has quit IRC | 06:01 | |
| *** armax has quit IRC | 06:02 | |
| *** pcaruana has joined #openstack-infra | 06:06 | |
| *** pcaruana has quit IRC | 06:06 | |
| *** zhurong has joined #openstack-infra | 06:06 | |
| *** kiennt26 has joined #openstack-infra | 06:10 | |
| *** huanxie has joined #openstack-infra | 06:12 | |
| *** sam_wan has quit IRC | 06:13 | |
| *** esberglu has joined #openstack-infra | 06:15 | |
| *** kiennt26 has quit IRC | 06:18 | |
| *** esberglu has quit IRC | 06:20 | |
| *** salv-orlando has joined #openstack-infra | 06:22 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: mqtt: add basic reporter https://review.openstack.org/518279 | 06:27 |
| *** salv-orlando has quit IRC | 06:28 | |
| *** mat128 has quit IRC | 06:28 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Use EPEL for debootstrap on centos https://review.openstack.org/523251 | 06:31 |
| AJaeger | mordred: looked at octavia again - and commented on etherpad. We miss an important part: The autogen support that we got from pbr ;( | 06:32 |
| *** armaan has joined #openstack-infra | 06:36 | |
| *** tojuvone has quit IRC | 06:39 | |
| *** armaan_ has joined #openstack-infra | 06:40 | |
| *** armaan has quit IRC | 06:43 | |
| *** xinliang has quit IRC | 06:46 | |
| *** jtomasek has joined #openstack-infra | 06:48 | |
| *** jtomasek has quit IRC | 06:48 | |
| *** jtomasek has joined #openstack-infra | 06:49 | |
| *** salv-orlando has joined #openstack-infra | 06:49 | |
| *** salv-orlando has quit IRC | 06:50 | |
| *** salv-orlando has joined #openstack-infra | 06:50 | |
| *** e0ne has joined #openstack-infra | 06:52 | |
| *** aeng has quit IRC | 06:55 | |
| ianw | Jeffrey4l: we have not added a generic proxy to the mirrors. i do not think it would be practical to figure out how to limit it's scope; we don't have a private internal network between ci hosts to limit access from, for example. | 06:55 |
| Jeffrey4l | ianw, not know the current deployment arch. but isn't the reverse proxy run on each nodepool? or a central place? | 06:57 |
| Jeffrey4l | now kolla download lots of packages from many place. add reverse proxy is complicated. | 06:58 |
| ianw | Jeffrey4l: we have one mirror host in each region we run CI in. ansible sets up hosts to use their region-local mirror early in the base jobs | 06:58 |
| Jeffrey4l | how about add a squid cache proxy on each region? | 06:59 |
| snapiri | pabelanger: any chance you could give me a quick review for https://review.openstack.org/#/c/520320/ ? | 06:59 |
| ianw | well we could not add it as a completely public proxy | 07:00 |
| *** xinliang has joined #openstack-infra | 07:00 | |
| *** xinliang has quit IRC | 07:00 | |
| *** xinliang has joined #openstack-infra | 07:00 | |
| *** mriedem has quit IRC | 07:00 | |
| Jeffrey4l | ianw, so the real issue is public? how about restrict the access address? | 07:01 |
| ianw | in theory, we could firewall it to allowed hosts, but i have no idea how you'd figure out the allowed hosts and instrument that through all the deployment | 07:01 |
| ianw | it would be quite a project | 07:04 |
| *** threestrands has quit IRC | 07:04 | |
| Jeffrey4l | ianw, how about, dynamic add/remove the rules when lunch a vm? the we can get the ip address. | 07:04 |
| *** zhurong has quit IRC | 07:05 | |
| ianw | that's very easy to say, but not very easy to do :) | 07:05 |
| Jeffrey4l | aha, yes. but it will save lots of network bandwith. | 07:06 |
| Jeffrey4l | now lots of kolla jobs failed due to network issue. | 07:06 |
| Jeffrey4l | anyway, i will try to add more reverse proxy now. | 07:06 |
| *** namnh has joined #openstack-infra | 07:06 | |
| openstackgerrit | Jose Luis Franco proposed openstack-infra/tripleo-ci master: WIP: Upgrade UC and OC using tripleo-upgrade role https://review.openstack.org/515643 | 07:09 |
| *** yamahata has joined #openstack-infra | 07:11 | |
| *** slaweq has joined #openstack-infra | 07:17 | |
| *** iyamahat has joined #openstack-infra | 07:17 | |
| *** iyamahat_ has joined #openstack-infra | 07:18 | |
| *** spectr has joined #openstack-infra | 07:18 | |
| *** spectr has quit IRC | 07:21 | |
| *** iyamahat has quit IRC | 07:22 | |
| *** slaweq has quit IRC | 07:25 | |
| *** sree_ has joined #openstack-infra | 07:28 | |
| *** claudiub|2 has joined #openstack-infra | 07:28 | |
| *** sree_ is now known as Guest79701 | 07:28 | |
| *** danpawlik has quit IRC | 07:29 | |
| *** sree has quit IRC | 07:31 | |
| *** ykarel is now known as ykarel|lunch | 07:34 | |
| *** makowals has joined #openstack-infra | 07:35 | |
| mordred | AJaeger: ah! yes | 07:39 |
| *** dhinesh has quit IRC | 07:40 | |
| mordred | AJaeger: I remember discussing there being a more native sphinx tool for this - but I don't remember the details ... investigating | 07:41 |
| *** hashar has joined #openstack-infra | 07:42 | |
| *** slaweq has joined #openstack-infra | 07:44 | |
| mordred | ianw: "Use EPEL for debootstrap on centos" is the best commit message ever | 07:45 |
| AJaeger | mordred: seeing the change to openstackdocstheme: This will hit many repos that use the same tox_install ;( | 07:50 |
| AJaeger | config-core, a couple of cleanups on project-config for review sometime, please: https://review.openstack.org/522894 https://review.openstack.org/522895 https://review.openstack.org/522888 - and an unrelated one https://review.openstack.org/#/c/513926 | 07:51 |
| *** rcernin has quit IRC | 07:53 | |
| *** d0ugal has joined #openstack-infra | 07:55 | |
| *** hoangcx has joined #openstack-infra | 07:56 | |
| hoangcx | frickler: Hi, Can you please bless this one in project-config https://review.openstack.org/#/c/522995/ . Thanks a lot | 07:58 |
| *** andreas_s has joined #openstack-infra | 07:58 | |
| *** e0ne has quit IRC | 08:01 | |
| *** kiennt26 has joined #openstack-infra | 08:01 | |
| *** kjackal has quit IRC | 08:06 | |
| *** shardy has joined #openstack-infra | 08:08 | |
| *** andreas_s has quit IRC | 08:08 | |
| *** iyamahat_ has quit IRC | 08:16 | |
| *** iyamahat_ has joined #openstack-infra | 08:16 | |
| *** kjackal has joined #openstack-infra | 08:17 | |
| *** hoangcx has quit IRC | 08:18 | |
| *** hoangcx has joined #openstack-infra | 08:18 | |
| *** ralonsoh has joined #openstack-infra | 08:18 | |
| mordred | AJaeger: boo. well, maybe it's worth figuring out why that file didn't work then | 08:19 |
| mordred | AJaeger: woot! I think I've got a solution for the octavia autoindex thing | 08:20 |
| *** amoralej|off is now known as amoralej | 08:21 | |
| *** florianf has joined #openstack-infra | 08:22 | |
| *** danpawlik has joined #openstack-infra | 08:22 | |
| *** danpawlik has quit IRC | 08:23 | |
| *** iyamahat has joined #openstack-infra | 08:23 | |
| *** danpawlik has joined #openstack-infra | 08:24 | |
| mordred | AJaeger: https://review.openstack.org/521211 Change how path munging works for sphinx docs | 08:24 |
| *** slaweq_ has joined #openstack-infra | 08:24 | |
| *** iyamahat_ has quit IRC | 08:26 | |
| *** danpawlik has quit IRC | 08:26 | |
| *** pcaruana has joined #openstack-infra | 08:28 | |
| *** danpawlik has joined #openstack-infra | 08:29 | |
| mordred | AJaeger: that's working for me locally - if it works up in the gate, I think we should extract the code into a sphinx plugin | 08:29 |
| *** slaweq_ has quit IRC | 08:29 | |
| *** dhajare has quit IRC | 08:30 | |
| *** alexchadin has joined #openstack-infra | 08:31 | |
| *** Hal has joined #openstack-infra | 08:36 | |
| mordred | AJaeger: https://review.openstack.org/521729 seems to be working now | 08:36 |
| odyssey4me | dmsimard re: running our plays directly by zuul... maybe, but to do that we'd need to focus our implementation to run a range of ansible versions, rather than use new features when we can... which makes things a little different | 08:38 |
| odyssey4me | also, it'll be important because zuul may just switch to a new version and our stable branches may require code churn as a result of that... so it's not ideal | 08:39 |
| odyssey4me | so for now we're keeping the execution binary of the tests independent of zuul | 08:39 |
| mordred | odyssey4me: ++ | 08:39 |
| *** ykarel|lunch is now known as ykarel | 08:40 | |
| mordred | odyssey4me: there has been a bit of discussion around providing user-selectable ansible versions - but that's a fairy-tale for now :) | 08:40 |
| odyssey4me | yeah, that would change things for us - but that would be quite a things to maintain I would guess | 08:41 |
| frickler | hoangcx: I'm failing to find logs for the periodic jobs, are you sure that these are really running? | 08:41 |
| odyssey4me | zuul would effectively have to maintain being able to use somewhere around 4 versions of ansible for tests execution at any one time, maybe even 6 if you take minors into account | 08:42 |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config master: Remove legacy jobs from oaktree repos https://review.openstack.org/523247 | 08:43 |
| odyssey4me | it would also mean having to hold back on using new features due to old versions not being able to work wiith them, although I suppose it would be possible to shim/plugin things to work around that | 08:43 |
| *** wewe0901 has joined #openstack-infra | 08:45 | |
| *** iyamahat has quit IRC | 08:46 | |
| *** iyamahat has joined #openstack-infra | 08:46 | |
| *** ccamacho has joined #openstack-infra | 08:46 | |
| openstackgerrit | Ian Y. Choi proposed openstack-infra/system-config master: Clean up translate.o.o before vhost https://review.openstack.org/523350 | 08:49 |
| mordred | odyssey4me: figuring out the maint burden is the hardest bit - otoh, it's not a *huge* surface area if we play our cards right. there are definitely bigger fish to fry first though | 08:50 |
| *** jascott1 has quit IRC | 08:51 | |
| *** jascott1 has joined #openstack-infra | 08:51 | |
| *** gmann is now known as gmann_afk | 08:53 | |
| *** ianychoi has joined #openstack-infra | 08:53 | |
| *** gridinv has quit IRC | 08:53 | |
| *** jascott1 has quit IRC | 08:55 | |
| *** dhajare has joined #openstack-infra | 08:56 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Refactor provider config to driver module https://review.openstack.org/488384 | 08:56 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement a static driver for Nodepool https://review.openstack.org/468624 | 08:56 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement an OpenContainer driver https://review.openstack.org/468753 | 08:57 |
| odyssey4me | mordred yeah, agreed - it would certainly make things more interesting for us - we could cut out a ton of machinery | 08:58 |
| odyssey4me | but right now we're also targeting higher value things | 08:58 |
| mordred | ++ | 08:59 |
| *** priteau has joined #openstack-infra | 09:01 | |
| *** yamamoto has quit IRC | 09:04 | |
| *** yamamoto has joined #openstack-infra | 09:05 | |
| AJaeger | mordred: what timezone are you in currently? | 09:06 |
| AJaeger | mordred: https://review.openstack.org/520363 fails ;( | 09:07 |
| *** jpich has joined #openstack-infra | 09:07 | |
| mordred | AJaeger: US/Central ... but for some reason I was lucky enough to wake up at 1am (yay!) | 09:08 |
| AJaeger | mordred: Oh no ;( | 09:08 |
| AJaeger | mordred: go back to sleep ;) | 09:08 |
| * AJaeger can't dig into jobs right now either... | 09:08 | |
| *** gridinv has joined #openstack-infra | 09:08 | |
| *** iyamahat has quit IRC | 09:09 | |
| *** iyamahat has joined #openstack-infra | 09:09 | |
| *** jpena|off is now known as jpena | 09:09 | |
| mordred | AJaeger: oh. I think I see the issue with 520363 | 09:10 |
| *** yamamoto has quit IRC | 09:11 | |
| *** mnencia has quit IRC | 09:13 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/openstack-zuul-jobs master: Switch build-openstack-sphinx-docs to build-sphinx-docs https://review.openstack.org/521145 | 09:14 |
| mordred | AJaeger: ^^ triggered a recheck on https://review.openstack.org/#/c/520363 | 09:14 |
| mordred | AJaeger: and yes - sleep sounds like a great idea | 09:14 |
| openstackgerrit | Merged openstack-infra/project-config master: Cleanup voting entries in projects.yaml https://review.openstack.org/522894 | 09:15 |
| *** iyamahat has quit IRC | 09:15 | |
| *** iyamahat_ has joined #openstack-infra | 09:15 | |
| AJaeger | mordred: will review later today, thanks | 09:19 |
| *** lucas-afk is now known as lucasagomes | 09:19 | |
| *** iyamahat_ has quit IRC | 09:20 | |
| *** zhurong has joined #openstack-infra | 09:21 | |
| *** dbecker has quit IRC | 09:21 | |
| *** sbezverk has quit IRC | 09:22 | |
| AJaeger | mordred: yeah, 520363 works now. | 09:22 |
| ianychoi | config-core: Would you review and approve https://review.openstack.org/#/c/523350/ ? This was the old configuration before using vhost with translate01.o.o and it seems that this is the main blocker for an ongoing patch from I18n PTL now: https://review.openstack.org/#/c/506795/ | 09:22 |
| mordred | AJaeger: \o/ | 09:23 |
| *** mnencia has joined #openstack-infra | 09:23 | |
| *** dbecker has joined #openstack-infra | 09:23 | |
| AJaeger | ianychoi: config-core does not have +2 on system-config - only infra-root has ^ | 09:24 |
| *** pgadiya has joined #openstack-infra | 09:24 | |
| *** pgadiya has quit IRC | 09:24 | |
| ianychoi | AJaeger, Aha... I thought that they take core both project-config & system-config. Thanks! | 09:24 |
| ianychoi | s/core/care | 09:24 |
| mordred | ianychoi: https://review.openstack.org/#/c/523350/ lgtm - ianw still around and feel like a +A? | 09:25 |
| ianw | sure :) | 09:27 |
| mordred | ianw: yay! if you feel like a second easy one ... https://review.openstack.org/#/c/523247/ | 09:28 |
| * mordred hands ianw a pie | 09:28 | |
| AJaeger | frickler, ianw https://review.openstack.org/#/c/513926 as well, please | 09:29 |
| * AJaeger holds up a pie as well to hand out ;) | 09:29 | |
| AJaeger | and a couple of cleanups on project-config for review sometime, please: https://review.openstack.org/522895 and https://review.openstack.org/522888 | 09:29 |
| AJaeger | and https://review.openstack.org/522185 as well - the repo is retired now, let's formalize it | 09:30 |
| *** salv-orlando has quit IRC | 09:33 | |
| mordred | AJaeger: ooh. i like https://review.openstack.org/522888 :) | 09:33 |
| *** martinkopec has joined #openstack-infra | 09:33 | |
| *** salv-orlando has joined #openstack-infra | 09:33 | |
| * ianw was in the study to do some time on the elliptical machine and has now run out of excuses :) | 09:34 | |
| openstackgerrit | Rui Chen proposed openstack-infra/nodepool feature/zuulv3: Fix nodepool alien-list issue https://review.openstack.org/522495 | 09:35 |
| openstackgerrit | Merged openstack-infra/project-config master: remove non-voting from experimental jobs https://review.openstack.org/522895 | 09:36 |
| AJaeger | ianw: regarding 522888, I used codesearch to check for them, hope I didn't screw up ;) | 09:36 |
| *** e0ne has joined #openstack-infra | 09:36 | |
| AJaeger | thanks | 09:36 |
| AJaeger | any takers for 523247? | 09:37 |
| priteau | Hello. I noticed that there were no project release announcements on openstack-announce since last November: http://lists.openstack.org/pipermail/openstack-announce/2016-November/date.html | 09:38 |
| priteau | Is it not done anymore? I would be surprised that no new client was released for a year | 09:38 |
| *** salv-orlando has quit IRC | 09:38 | |
| AJaeger | priteau: better ask on #openstack-release channel - and yes that was changed... | 09:39 |
| priteau | oh, wrong channel, thanks | 09:39 |
| AJaeger | priteau: but read http://lists.openstack.org/pipermail/openstack-announce/2016-November/001966.html first | 09:40 |
| AJaeger | that part of the list you pasted... | 09:40 |
| priteau | Thanks! I didn't see it in the middle of all the announcements. I guess that was the issue… | 09:40 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove ceilometerclient from infra https://review.openstack.org/522185 | 09:41 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove unused jenkins/scripts scripts https://review.openstack.org/522888 | 09:41 |
| *** ociuhandu has joined #openstack-infra | 09:43 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Remove old python-sdist job https://review.openstack.org/513926 | 09:44 |
| *** ociuhandu has quit IRC | 09:46 | |
| *** derekh has joined #openstack-infra | 09:46 | |
| *** openstackgerrit has quit IRC | 09:48 | |
| *** electrofelix has joined #openstack-infra | 09:50 | |
| *** openstackgerrit has joined #openstack-infra | 09:51 | |
| openstackgerrit | Merged openstack-infra/project-config master: Remove legacy jobs from oaktree repos https://review.openstack.org/523247 | 09:51 |
| *** Hal has quit IRC | 09:52 | |
| openstackgerrit | Merged openstack-infra/system-config master: Clean up translate.o.o before vhost https://review.openstack.org/523350 | 09:52 |
| *** erlon has joined #openstack-infra | 09:55 | |
| *** pblaho has joined #openstack-infra | 09:55 | |
| *** rcernin has joined #openstack-infra | 09:55 | |
| *** makowals has quit IRC | 09:57 | |
| openstackgerrit | Duong Ha-Quang proposed openstack-infra/openstack-zuul-jobs master: Remove legacy jobs in neutron-vpnaas https://review.openstack.org/523368 | 09:59 |
| openstackgerrit | Duong Ha-Quang proposed openstack-infra/project-config master: Remove legacy jobs in neutron-vpnaas https://review.openstack.org/523369 | 09:59 |
| *** Hal has joined #openstack-infra | 10:00 | |
| *** pblaho has quit IRC | 10:03 | |
| ykarel | mordred, hi | 10:03 |
| mordred | hi ykarel | 10:04 |
| ykarel | mordred, is this the channel for query for openstacksdk | 10:04 |
| *** bhavik1 has joined #openstack-infra | 10:04 | |
| AJaeger | ykarel: check https://governance.openstack.org/tc/reference/projects/index.html | 10:05 |
| AJaeger | should point you to #openstack-shade | 10:05 |
| *** sdatko has quit IRC | 10:06 | |
| ykarel | AJaeger, Ok will continute there | 10:06 |
| *** yamahata has quit IRC | 10:07 | |
| *** namnh has quit IRC | 10:07 | |
| *** cuongnv has quit IRC | 10:10 | |
| *** ganso has joined #openstack-infra | 10:10 | |
| *** sambetts|afk is now known as sambetts | 10:11 | |
| *** pblaho has joined #openstack-infra | 10:12 | |
| openstackgerrit | Michał Dulko proposed openstack-infra/project-config master: Add release-notes-jobs to kuryr-kubernetes https://review.openstack.org/523376 | 10:13 |
| *** yamamoto has joined #openstack-infra | 10:14 | |
| openstackgerrit | Goutham Pratapa proposed openstack-infra/project-config master: Create Kingbird-Dashboard Project. https://review.openstack.org/523377 | 10:16 |
| *** yamamoto has quit IRC | 10:19 | |
| *** zhurong has quit IRC | 10:19 | |
| *** sdague has joined #openstack-infra | 10:22 | |
| openstackgerrit | Merged openstack-infra/irc-meetings master: Adjust oslo meeting time https://review.openstack.org/522740 | 10:22 |
| openstackgerrit | Merged openstack-infra/irc-meetings master: Update Designate's meeting time https://review.openstack.org/523177 | 10:24 |
| *** alexchadin has quit IRC | 10:28 | |
| *** gcb has quit IRC | 10:28 | |
| *** s-shiono_ has quit IRC | 10:29 | |
| *** Hal has quit IRC | 10:30 | |
| *** tosky has joined #openstack-infra | 10:31 | |
| *** Hal has joined #openstack-infra | 10:32 | |
| *** huanxie has quit IRC | 10:36 | |
| *** slaweq_ has joined #openstack-infra | 10:37 | |
| *** slaweq_ has quit IRC | 10:42 | |
| *** bhavik1 has quit IRC | 10:43 | |
| *** haleyb has quit IRC | 10:45 | |
| *** kiennt26 has quit IRC | 10:45 | |
| *** esberglu has joined #openstack-infra | 10:46 | |
| *** alexchadin has joined #openstack-infra | 10:51 | |
| *** esberglu has quit IRC | 10:51 | |
| *** udesale has quit IRC | 10:55 | |
| *** ldnunes has joined #openstack-infra | 10:55 | |
| amotoki | AJaeger: Question on updated-pti tox_install.sh patches: do you have further steps on tox_install.sh? | 10:59 |
| *** salv-orlando has joined #openstack-infra | 11:01 | |
| *** annp has quit IRC | 11:03 | |
| amotoki | AJaeger: some comments are not correct: for example, you say 'neutron' in horizon plugin repos. it is good if you have further steps and they will be clean up later. | 11:03 |
| *** jkilpatr has joined #openstack-infra | 11:07 | |
| *** liujiong has quit IRC | 11:07 | |
| *** salv-orlando has quit IRC | 11:07 | |
| *** salv-orlando has joined #openstack-infra | 11:08 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack-infra/tripleo-ci master: Make sure we collect var/lib/heat-config directory. https://review.openstack.org/523388 | 11:08 |
| *** makowals has joined #openstack-infra | 11:12 | |
| *** salv-orlando has quit IRC | 11:13 | |
| *** salv-orlando has joined #openstack-infra | 11:14 | |
| amotoki | AJaeger: I think you also want to do this https://review.openstack.org/#/c/523390/ | 11:19 |
| amotoki | AJaeger: I just would like to know more plans. thanks. | 11:19 |
| *** makowals has quit IRC | 11:21 | |
| *** makowals has joined #openstack-infra | 11:23 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Refactor provider config to driver module https://review.openstack.org/488384 | 11:24 |
| *** wewe0901 has quit IRC | 11:25 | |
| *** jkilpatr has quit IRC | 11:27 | |
| *** adarazs|ruck is now known as adarazs|ruck|lnc | 11:30 | |
| *** jkilpatr has joined #openstack-infra | 11:38 | |
| *** dhajare has quit IRC | 11:40 | |
| *** wolverineav has joined #openstack-infra | 11:48 | |
| AJaeger | amotoki: good catch with trove, thanks. We don't have further changes out there. horizon and neutron are the same class of "problems" | 11:51 |
| amotoki | AJaeger: yes, they are classified into a same problem, but you changed 'horizon' into 'neutron' in most patches. is it intended? | 11:52 |
| *** efried has quit IRC | 11:54 | |
| amotoki | AJaeger: FYI: tox_install.sh with "os:horizon" were created by the requirement team + (me [I copied what was created by the req team). I thought that was the new and recommended way at that time, but it turns out not a right way now. | 11:54 |
| *** Guest79701 has quit IRC | 11:54 | |
| amotoki | AJaeger: I would like to see a document what is the recommended way. It is really a chaos world without no guidance. | 11:56 |
| amotoki | AJaeger: if the infra team would like to have a single interface, some guidance would be really appreciated. | 11:56 |
| AJaeger | amotoki: I just copied a file over ;( | 11:58 |
| amotoki | I refactors install_project() in networking-projects several months ago. perhaps you copied it over. | 11:58 |
| AJaeger | amotoki: mordred is working on updating the doc jobs and we found that we need to consolidate for this. Best discuss with him later | 11:58 |
| amotoki | AJaeger: okay. thanks for sharing the context. | 11:59 |
| AJaeger | amotoki: http://lists.openstack.org/pipermail/openstack-dev/2017-November/124815.html is some more info about this | 11:59 |
| *** shardy is now known as shardy_afk | 11:59 | |
| amotoki | AJaeger: another projects which uses tox_install.sh is python-openstackclient/osc-lib (perhaps mordred is aware of them) | 12:00 |
| AJaeger | amotoki, mordred : It might indeed be a good way to have *one* master source for tox_install.sh (or two as needed) | 12:00 |
| *** e0ne has quit IRC | 12:01 | |
| *** psachin has quit IRC | 12:01 | |
| *** sree has joined #openstack-infra | 12:02 | |
| *** pblaho has quit IRC | 12:03 | |
| *** d0ugal has quit IRC | 12:04 | |
| *** e0ne has joined #openstack-infra | 12:05 | |
| *** jkilpatr has quit IRC | 12:06 | |
| *** jkilpatr has joined #openstack-infra | 12:07 | |
| *** efried has joined #openstack-infra | 12:07 | |
| *** sree has quit IRC | 12:07 | |
| *** d0ugal has joined #openstack-infra | 12:07 | |
| *** armaan_ has quit IRC | 12:08 | |
| *** armaan has joined #openstack-infra | 12:09 | |
| *** yamamoto has joined #openstack-infra | 12:18 | |
| *** pblaho has joined #openstack-infra | 12:20 | |
| *** yamamoto has quit IRC | 12:22 | |
| *** sree has joined #openstack-infra | 12:23 | |
| *** udesale has joined #openstack-infra | 12:23 | |
| *** sree has quit IRC | 12:27 | |
| *** raissa has joined #openstack-infra | 12:30 | |
| *** ramishra has quit IRC | 12:33 | |
| *** esberglu has joined #openstack-infra | 12:34 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement a static driver for Nodepool https://review.openstack.org/468624 | 12:35 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement an OpenContainer driver https://review.openstack.org/468753 | 12:36 |
| *** katkapilatova has joined #openstack-infra | 12:37 | |
| *** rossella_s has quit IRC | 12:38 | |
| *** esberglu has quit IRC | 12:39 | |
| *** rossella_s has joined #openstack-infra | 12:41 | |
| *** dhajare has joined #openstack-infra | 12:42 | |
| *** katkapilatova has quit IRC | 12:43 | |
| *** rossella_s has quit IRC | 12:46 | |
| *** armaan has quit IRC | 12:46 | |
| *** armaan has joined #openstack-infra | 12:46 | |
| *** bobh has joined #openstack-infra | 12:46 | |
| *** rossella_s has joined #openstack-infra | 12:47 | |
| *** jtomasek_ has joined #openstack-infra | 12:49 | |
| *** ldnunes has quit IRC | 12:50 | |
| *** adarazs|ruck|lnc is now known as adarazs|ruck | 12:51 | |
| *** jtomasek has quit IRC | 12:52 | |
| *** sboyron has joined #openstack-infra | 12:52 | |
| *** shardy_afk is now known as shardy | 12:57 | |
| *** smatzek has joined #openstack-infra | 13:02 | |
| *** amoralej is now known as amoralej|lunch | 13:02 | |
| *** ldnunes has joined #openstack-infra | 13:02 | |
| *** dtantsur|afk is now known as dtantsur | 13:02 | |
| *** jpena is now known as jpena|lunch | 13:02 | |
| *** bhavik1 has joined #openstack-infra | 13:03 | |
| *** sree has joined #openstack-infra | 13:04 | |
| *** wangqian has quit IRC | 13:05 | |
| *** dave-mccowan has joined #openstack-infra | 13:06 | |
| *** sree has quit IRC | 13:08 | |
| *** rhallisey has joined #openstack-infra | 13:09 | |
| *** slaweq_ has joined #openstack-infra | 13:10 | |
| *** ijw has joined #openstack-infra | 13:11 | |
| *** dprince has joined #openstack-infra | 13:11 | |
| *** rosmaita has joined #openstack-infra | 13:12 | |
| *** slaweq_ has quit IRC | 13:14 | |
| *** lucasagomes is now known as lucas-hungry | 13:14 | |
| *** ijw has quit IRC | 13:15 | |
| *** links has quit IRC | 13:16 | |
| *** yolanda has quit IRC | 13:16 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack-infra/tripleo-ci master: Make sure we collect var/lib/heat-config directory. https://review.openstack.org/523388 | 13:17 |
| *** trown|outtypewww is now known as trown | 13:20 | |
| *** Wei_Liu has quit IRC | 13:20 | |
| *** yolanda has joined #openstack-infra | 13:21 | |
| *** edmondsw has joined #openstack-infra | 13:22 | |
| *** jascott1 has joined #openstack-infra | 13:23 | |
| *** rossella_s has quit IRC | 13:24 | |
| *** sree has joined #openstack-infra | 13:24 | |
| *** rossella_s has joined #openstack-infra | 13:27 | |
| *** bobh has quit IRC | 13:27 | |
| *** jascott1 has quit IRC | 13:28 | |
| *** sree has quit IRC | 13:29 | |
| *** esberglu has joined #openstack-infra | 13:29 | |
| *** esberglu has quit IRC | 13:29 | |
| *** esberglu has joined #openstack-infra | 13:29 | |
| *** esberglu has quit IRC | 13:29 | |
| *** esberglu has joined #openstack-infra | 13:30 | |
| *** dhajare has quit IRC | 13:30 | |
| *** sree has joined #openstack-infra | 13:30 | |
| *** jaypipes has joined #openstack-infra | 13:31 | |
| *** mriedem has joined #openstack-infra | 13:32 | |
| *** mat128 has joined #openstack-infra | 13:33 | |
| *** rlandy has joined #openstack-infra | 13:33 | |
| *** bhavik1 has quit IRC | 13:33 | |
| *** rlandy is now known as rlandy|rover | 13:35 | |
| *** sree has quit IRC | 13:35 | |
| *** esberglu has quit IRC | 13:35 | |
| *** fultonj has joined #openstack-infra | 13:41 | |
| *** kgiusti has joined #openstack-infra | 13:41 | |
| *** sree has joined #openstack-infra | 13:43 | |
| *** HeOS has quit IRC | 13:43 | |
| *** ramishra has joined #openstack-infra | 13:44 | |
| *** hashar has quit IRC | 13:46 | |
| *** hashar has joined #openstack-infra | 13:47 | |
| *** fultonj has quit IRC | 13:48 | |
| *** ykarel is now known as ykarel|away | 13:54 | |
| *** salv-orlando has quit IRC | 13:57 | |
| *** haleyb has joined #openstack-infra | 13:58 | |
| *** mihalis68 has joined #openstack-infra | 14:00 | |
| *** esberglu has joined #openstack-infra | 14:00 | |
| *** yamahata has joined #openstack-infra | 14:00 | |
| *** ykarel|away has quit IRC | 14:06 | |
| *** thorst has joined #openstack-infra | 14:06 | |
| *** lucas-hungry is now known as lucasagomes | 14:07 | |
| *** bobh has joined #openstack-infra | 14:08 | |
| *** jpena|lunch is now known as jpena|off | 14:08 | |
| *** amoralej|lunch is now known as amoralej | 14:08 | |
| *** amoralej is now known as amoralej|off | 14:10 | |
| openstackgerrit | David Moreau Simard proposed openstack-infra/openstack-zuul-jobs master: General improvements to configure-unbound role https://review.openstack.org/523228 | 14:13 |
| *** pcaruana has quit IRC | 14:13 | |
| *** rossella_s has quit IRC | 14:14 | |
| *** pcaruana has joined #openstack-infra | 14:17 | |
| *** Goneri has joined #openstack-infra | 14:18 | |
| *** mat128 has quit IRC | 14:18 | |
| *** rossella_s has joined #openstack-infra | 14:18 | |
| *** mat128 has joined #openstack-infra | 14:20 | |
| *** yolanda has quit IRC | 14:21 | |
| *** yamamoto has joined #openstack-infra | 14:21 | |
| *** armaan has quit IRC | 14:22 | |
| ttx | fungi: could you add me as initial member to the meta-sig-core Gerrit group at https://review.openstack.org/#/admin/groups/1850,members ? It's used as a core reviewer group for openstack/governance-sigs | 14:25 |
| ttx | I'll add Melvin | 14:26 |
| *** janki has quit IRC | 14:29 | |
| *** smatzek has quit IRC | 14:30 | |
| *** smatzek has joined #openstack-infra | 14:31 | |
| *** slaweq has quit IRC | 14:31 | |
| openstackgerrit | Thierry Carrez proposed openstack-infra/project-config master: Set up doc-building jobs for governance-sigs repo https://review.openstack.org/523423 | 14:31 |
| *** slaweq has joined #openstack-infra | 14:32 | |
| *** alexchadin has quit IRC | 14:32 | |
| *** yamamoto has quit IRC | 14:33 | |
| *** eharney has quit IRC | 14:34 | |
| *** smatzek has quit IRC | 14:35 | |
| *** yolanda has joined #openstack-infra | 14:35 | |
| *** slaweq has quit IRC | 14:36 | |
| *** cshastri has quit IRC | 14:37 | |
| *** sree_ has joined #openstack-infra | 14:39 | |
| *** sree_ is now known as Guest97530 | 14:39 | |
| *** sree has quit IRC | 14:40 | |
| *** armax has joined #openstack-infra | 14:40 | |
| *** kgiusti has quit IRC | 14:42 | |
| *** rossella_s has quit IRC | 14:44 | |
| *** katkapilatova has joined #openstack-infra | 14:45 | |
| fungi | ttx: done | 14:45 |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: Added get unpublished events endpoint https://review.openstack.org/523274 | 14:45 |
| *** rossella_s has joined #openstack-infra | 14:47 | |
| *** camunoz has joined #openstack-infra | 14:50 | |
| *** salv-orlando has joined #openstack-infra | 14:51 | |
| *** salv-orl_ has joined #openstack-infra | 14:56 | |
| *** kgiusti has joined #openstack-infra | 14:56 | |
| *** ihrachys has joined #openstack-infra | 14:58 | |
| *** salv-orlando has quit IRC | 14:59 | |
| *** yamamoto has joined #openstack-infra | 15:00 | |
| openstackgerrit | Alex Kavanagh proposed openstack-infra/openstack-zuul-jobs master: Add python35 charm jobs template for new python3-only charms https://review.openstack.org/523433 | 15:01 |
| openstackgerrit | Alex Kavanagh proposed openstack-infra/project-config master: Modify charm-ceph-mon to make py27 a voting job https://review.openstack.org/523098 | 15:01 |
| *** hongbin has joined #openstack-infra | 15:02 | |
| *** rossella_s has quit IRC | 15:02 | |
| openstackgerrit | Alex Kavanagh proposed openstack-infra/project-config master: Modify charm-ceph-mon to make it an python35 only for gate https://review.openstack.org/523098 | 15:03 |
| *** smatzek has joined #openstack-infra | 15:03 | |
| *** udesale has quit IRC | 15:03 | |
| *** Hal has quit IRC | 15:04 | |
| *** liusheng has quit IRC | 15:07 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Create ovb-ha-ipv6 and deprecate fs024 job https://review.openstack.org/522618 | 15:09 |
| openstackgerrit | Sam Betts proposed openstack-infra/os-loganalyze master: Repair htmlify-log.py command https://review.openstack.org/383812 | 15:10 |
| *** marst has joined #openstack-infra | 15:10 | |
| *** yamamoto has quit IRC | 15:10 | |
| *** rbrndt has joined #openstack-infra | 15:11 | |
| mordred | amotoki: the *ultimate* goal is to stop needing tox_install.sh at all - but we're still a few steps away from being able to do that... | 15:16 |
| mordred | amotoki: you make a very good point though - I think we should make a larger picture plan and document it | 15:16 |
| AJaeger | mordred: just analyzed the octavia problem and updated etherpad with some comments | 15:17 |
| mordred | AJaeger: awesome, thanks | 15:17 |
| *** ramishra has quit IRC | 15:17 | |
| mordred | AJaeger: also - the issue ykarel came up with last night is ... rather troubling | 15:18 |
| mordred | AJaeger: oh, no, nevermind - it's localized | 15:18 |
| * mordred will keep from spouting too much trouble | 15:18 | |
| *** ramishra has joined #openstack-infra | 15:19 | |
| AJaeger | ;) | 15:19 |
| *** clayton has quit IRC | 15:20 | |
| AJaeger | who reviews bashate and hacking? https://review.openstack.org/520670 and https://review.openstack.org/520669 ifix the releasenotes and need some love, please | 15:20 |
| *** Hal has joined #openstack-infra | 15:21 | |
| *** clayton has joined #openstack-infra | 15:21 | |
| fungi | AJaeger: bashate and hacking are qa deliverables, if that helps | 15:23 |
| AJaeger | fungi, thanks - will try there as well. Saw some infra folks on reviewer list... | 15:23 |
| mordred | AJaeger: I have +2 on at least hacking ... | 15:24 |
| openstackgerrit | Merged openstack-infra/tripleo-ci master: Add novajoin to collect-logs.yml list https://review.openstack.org/521591 | 15:24 |
| *** coolsvap has quit IRC | 15:24 | |
| *** hashar is now known as hasharAway | 15:25 | |
| mordred | AJaeger: hacking done | 15:26 |
| AJaeger | thanks | 15:26 |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Updated get summit by id JSON https://review.openstack.org/523441 | 15:26 |
| jeblair | AJaeger: +2 bashate | 15:27 |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: Updated get summit by id JSON https://review.openstack.org/523441 | 15:27 |
| AJaeger | thanks, jeblair | 15:27 |
| *** d0ugal has quit IRC | 15:28 | |
| *** janki has joined #openstack-infra | 15:28 | |
| *** armaan has joined #openstack-infra | 15:30 | |
| mordred | AJaeger: just in case it comes up again, the 'troubling' issue ykarel found last night is that openstacksdk has "basepython = {env:OPENSTACKSDK_TOX_PYTHON:python2}" set in tox.ini ... which causes py35 tox envs to not actually use python3 | 15:31 |
| mordred | AJaeger: mostly mentioning it in case you happen to see people reporting issues somewhere that tests in the gate are not running the same as they do in local testing | 15:31 |
| openstackgerrit | Merged openstack-dev/hacking master: Remove setting of version/release from releasenotes https://review.openstack.org/520670 | 15:32 |
| AJaeger | mordred: ok | 15:33 |
| AJaeger | who can +2A the same change for subunit2sql, please? | 15:34 |
| *** ramishra has quit IRC | 15:34 | |
| *** HeOS has joined #openstack-infra | 15:38 | |
| *** slaweq has joined #openstack-infra | 15:38 | |
| *** trown is now known as trown|brb | 15:40 | |
| *** d0ugal has joined #openstack-infra | 15:42 | |
| *** slaweq has quit IRC | 15:43 | |
| *** dklyle has joined #openstack-infra | 15:43 | |
| *** yamahata__ has joined #openstack-infra | 15:44 | |
| *** gouthamr has joined #openstack-infra | 15:45 | |
| *** isq_ has quit IRC | 15:45 | |
| *** slaweq has joined #openstack-infra | 15:45 | |
| *** yamamoto has joined #openstack-infra | 15:46 | |
| *** ijw has joined #openstack-infra | 15:46 | |
| *** david-lyle has quit IRC | 15:46 | |
| *** nicolasbock has joined #openstack-infra | 15:49 | |
| *** slaweq has quit IRC | 15:50 | |
| *** ijw has quit IRC | 15:50 | |
| *** trown|brb is now known as trown | 15:51 | |
| *** Hal has quit IRC | 15:52 | |
| *** jamesmcarthur has joined #openstack-infra | 15:52 | |
| *** yamamoto has quit IRC | 15:52 | |
| AJaeger | mordred: regarding https://review.openstack.org/522369 - seems I was to fast to handle doc/requirements in requirements, this bites us now. should we disable requirements check briefly? | 15:52 |
| ttx | fungi: thx | 15:53 |
| mordred | AJaeger: no- I think you were correct - I think we should add that requirement to g-r - I'll get a patch up for that in just a few | 15:56 |
| *** masuberu has joined #openstack-infra | 16:01 | |
| AJaeger | mordred: ok | 16:02 |
| *** dhajare has joined #openstack-infra | 16:04 | |
| *** masber has quit IRC | 16:05 | |
| *** sbezverk has joined #openstack-infra | 16:07 | |
| smcginnis | Anyone remember - we had an issue with library projects failing to install themselves with LIBS_FROM_GIT due to checking that the new version didn't match what was in upper-constraints. | 16:07 |
| smcginnis | I see another failure: http://logs.openstack.org/50/520650/1/gate/legacy-tempest-dsvm-neutron-src/06f9eb3/logs/devstacklog.txt.gz#_2017-11-28_15_03_58_534 | 16:07 |
| *** raissa has quit IRC | 16:07 | |
| *** masuberu has quit IRC | 16:08 | |
| *** raissa has joined #openstack-infra | 16:08 | |
| *** dhinesh has joined #openstack-infra | 16:08 | |
| smcginnis | But it actually does look like it was installed in this case: http://logs.openstack.org/50/520650/1/gate/legacy-tempest-dsvm-neutron-src/06f9eb3/logs/devstacklog.txt.gz#_2017-11-28_14_53_26_169 | 16:08 |
| *** masuberu has joined #openstack-infra | 16:08 | |
| clarkb | smcginnis: there was a change merged to devstack that broke the check, I believe the fix went in though | 16:08 |
| clarkb | smcginnis: it did a test -z instead of a test -n so check was inverted | 16:09 |
| smcginnis | clarkb: Oh? Do you know when that merged? | 16:09 |
| clarkb | let me find it | 16:09 |
| smcginnis | This just failed within probably 30 minutes ago. | 16:09 |
| clarkb | https://review.openstack.org/#/c/522077/ | 16:09 |
| clarkb | november 22 | 16:09 |
| openstackgerrit | Alex Kavanagh proposed openstack-infra/openstack-zuul-jobs master: Add python35 charm jobs template for new python3-only charms https://review.openstack.org/523433 | 16:10 |
| smcginnis | clarkb: Hmm, last week. But I just hit this today. | 16:10 |
| *** Hal has joined #openstack-infra | 16:10 | |
| clarkb | smcginnis: I think the issue there is glance_store vs glance-store | 16:11 |
| clarkb | oh | 16:11 |
| clarkb | wait did pip just update? | 16:11 |
| smcginnis | Ruh roh. | 16:11 |
| clarkb | no it is still 9.0.1 | 16:11 |
| clarkb | so ya I think maybe glance_store vs glance-store? | 16:12 |
| smcginnis | clarkb: Ah, could be on to something there. | 16:12 |
| smcginnis | "Successfully installed glance-store hacking-0.12.0 requests-mock-1.3.0" | 16:12 |
| smcginnis | But it is indeed looking for glance_store, not glance-store. | 16:12 |
| *** dhinesh has quit IRC | 16:13 | |
| frickler | clarkb: smcginnis: https://review.openstack.org/522423 should be the fix for that | 16:13 |
| fungi | glance_store seems to be the canonical name | 16:14 |
| fungi | strange that pip list converts _ to - | 16:14 |
| *** eharney has joined #openstack-infra | 16:14 | |
| smcginnis | frickler, fungi, clarkb: Thanks, that does appear to be the case. I'll wait for that fix to merge before I recheck. | 16:14 |
| smcginnis | Is there some background sync that needs to happen before I recheck, or will that be ready after it merges? | 16:15 |
| clarkb | should be ready once it merges | 16:16 |
| smcginnis | Great, I will hold tight for now. | 16:16 |
| fungi | should be able to depends-on it too | 16:16 |
| clarkb | I will review it shortly | 16:16 |
| fungi | i'm not a devstack-core, but lgtm | 16:17 |
| *** bobh has quit IRC | 16:24 | |
| *** salv-orlando has joined #openstack-infra | 16:28 | |
| *** salv-orl_ has quit IRC | 16:31 | |
| *** yamamoto has joined #openstack-infra | 16:31 | |
| *** yamamoto has quit IRC | 16:36 | |
| *** caphrim007 has joined #openstack-infra | 16:37 | |
| SamYaple | looking for some love on this bindep patch, definetely needs to be merged before we cut the bindep tag as it adjust some logic on my previous patch https://review.openstack.org/#/c/517105/ | 16:39 |
| *** salv-orlando has quit IRC | 16:42 | |
| *** salv-orlando has joined #openstack-infra | 16:42 | |
| *** [HeOS] has joined #openstack-infra | 16:42 | |
| *** slaweq has joined #openstack-infra | 16:43 | |
| *** HeOS has quit IRC | 16:44 | |
| *** slaweq_ has joined #openstack-infra | 16:45 | |
| *** salv-orlando has quit IRC | 16:46 | |
| *** slaweq has quit IRC | 16:48 | |
| *** jamesmcarthur has quit IRC | 16:52 | |
| openstackgerrit | David Moreau Simard proposed openstack-infra/openstack-zuul-jobs master: General improvements to configure-unbound role https://review.openstack.org/523228 | 16:52 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Add command_socket setting to executor section https://review.openstack.org/523465 | 16:52 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Add command socket support to zuul-scheduler https://review.openstack.org/523466 | 16:52 |
| *** jamesmcarthur has joined #openstack-infra | 16:53 | |
| AJaeger | mordred: did you see the -1 on https://review.openstack.org/521558 ? | 16:54 |
| mordred | AJaeger: I did not- but now have, and I agree with it | 16:59 |
| clarkb | infra-root reviews on https://review.openstack.org/#/c/522399/3 much appreciated as I think that is the last piece to get us back to logstash state pre zuulv3 | 17:00 |
| *** armaan has quit IRC | 17:00 | |
| *** trown is now known as trown|lunch | 17:00 | |
| *** armaan has joined #openstack-infra | 17:00 | |
| clarkb | SamYaple: re bindep change, is that behavior that should be explicitly tested? | 17:01 |
| SamYaple | clarkb: yes, but we dont really have any infrastructure in place to test it | 17:02 |
| SamYaple | we basically just do syntax checking | 17:02 |
| clarkb | SamYaple: bindep has unittests doesn't it? | 17:02 |
| clarkb | so you'd check if the package is excluded when in a group that doesn't match? | 17:03 |
| SamYaple | clarkb: its been about a month since i touched the patch, give me a few to look remember what i found | 17:03 |
| SamYaple | i explictly rememebr it not being that simple for some reason | 17:04 |
| *** katkapilatova has left #openstack-infra | 17:04 | |
| SamYaple | which is why i started downthe functional path, but i cant for remember what that reason was at this moment | 17:04 |
| *** liusheng has joined #openstack-infra | 17:04 | |
| clarkb | SamYaple: in this case I think you can just call _evaluate() directly and check the results with different inputs | 17:06 |
| SamYaple | clarkb: i think youre right and i may have just had tunnel vision | 17:07 |
| SamYaple | im not seeing a reason i cant do that | 17:08 |
| SamYaple | will get some tests in for it | 17:08 |
| *** ijw has joined #openstack-infra | 17:09 | |
| *** rossella_s has joined #openstack-infra | 17:09 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Revert "Reduce elasticsearch retention to 7 days" https://review.openstack.org/523474 | 17:11 |
| clarkb | I'm sure some of it is just slower week with holiday but our disk usage has dropped significantly on the ES servers. I think because we've excluded those very large files that were OOMing logstash workers | 17:11 |
| clarkb | pabelanger: dmsimard ^ so I think we can go back to 10 days | 17:11 |
| *** jamesmcarthur has quit IRC | 17:11 | |
| clarkb | (and just keep an eye on it) | 17:11 |
| *** ccamacho has quit IRC | 17:12 | |
| pabelanger | +2 | 17:13 |
| *** guinm is now known as mguiney | 17:14 | |
| *** yamamoto has joined #openstack-infra | 17:16 | |
| AJaeger | argh, what are they doing with bindep here? https://review.openstack.org/#/c/523455/ fungi, care to comment? | 17:19 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove legacy jobs from python-cinderclient https://review.openstack.org/521779 | 17:20 |
| fungi | AJaeger: i'll leave a comment. i have a feeling they don't know they can define arbitrary profiles | 17:21 |
| *** jamesmcarthur has joined #openstack-infra | 17:21 | |
| *** yamamoto has quit IRC | 17:21 | |
| AJaeger | fungi: yep... | 17:21 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Create ovb-ha-ipv6 and deprecate fs024 job https://review.openstack.org/522618 | 17:22 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Create ovb-ha-ipv6 and deprecate fs024 job https://review.openstack.org/522618 | 17:22 |
| clarkb | jeblair: so the rough make a storyboard board process is stories for every board item then make board and attach stories to board cards? | 17:23 |
| *** jamesmcarthur has quit IRC | 17:25 | |
| jeblair | clarkb: boards can have stories, tasks, or both | 17:25 |
| *** jascott1 has joined #openstack-infra | 17:26 | |
| jeblair | clarkb: https://storyboard.openstack.org/#!/board/53 is all stories | 17:26 |
| jeblair | clarkb: https://storyboard.openstack.org/#!/board/41 is all tasks | 17:26 |
| clarkb | looks like the cards call that out too "task in story foo" for the task case | 17:27 |
| jeblair | ya | 17:27 |
| jeblair | when you click on em | 17:27 |
| dsariel | hi! I want to run on my local machine all steps that are performed by upstream gate. E.g http://logs.openstack.org/72/512572/7/check/manila-tempest-minimal-dsvm-cephfs-native-centos-7/50f16bc/logs/devstack-early.txt.gz | 17:27 |
| jeblair | clarkb: tasks can have assignees and statuses. stories can have tags. | 17:27 |
| jeblair | clarkb: task statuses roll up into stories, so you can have a story that's active or not depending on its task statuses | 17:28 |
| jeblair | clarkb: but story tags don't roll down onto tasks | 17:28 |
| clarkb | dsariel: http://logs.openstack.org/72/512572/7/check/manila-tempest-minimal-dsvm-cephfs-native-centos-7/50f16bc/logs/reproduce.sh may be of help until that job stops being a legacy job | 17:28 |
| dsariel | is there somewhere a list of steps to be taken? | 17:28 |
| *** camunoz has quit IRC | 17:28 | |
| jeblair | clarkb: that's the thing i keep bumping into | 17:28 |
| *** dhinesh_ has joined #openstack-infra | 17:29 | |
| clarkb | jeblair: interesting. at first glance stories are probably what I want to use beacuse I want ot make a more explicit "this is work infra is doing" with a "help" wanted section | 17:29 |
| clarkb | jeblair: so that it is easier to people driving by to see where they might help out | 17:29 |
| clarkb | (and I think task may be too granular for that) | 17:29 |
| clarkb | I'll fiddle with it | 17:29 |
| jeblair | clarkb: yeah, stories + tags will probably work for that. | 17:29 |
| jeblair | clarkb: board 53 is entirely automatic, driven by storyboard queries, because it's only stories + tags | 17:29 |
| dsariel | clarkb, thanks a lot. I'll try it | 17:29 |
| openstackgerrit | David Moreau Simard proposed openstack-infra/openstack-zuul-jobs master: Add support for configuring TTLs in configure-unbound https://review.openstack.org/523178 | 17:29 |
| jeblair | clarkb: board 41 is driven by a script in cron, because it maps tags to tasks locally since storyboard doesn't do that. | 17:30 |
| *** jascott1 has quit IRC | 17:30 | |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Fixed postEvent and publishEvent Endpoints https://review.openstack.org/523484 | 17:30 |
| clarkb | jeblair: the queries are something you inject into the board for column data? | 17:30 |
| jeblair | clarkb: i think what you describe is most like board 53. the only thing that makes difficult is showing that something is in progress vs available for someone to work on. | 17:30 |
| clarkb | *column data population | 17:30 |
| jeblair | clarkb: yep. you can have manual columns ("lanes") or automatic ones. you can create the automatic ones through the web interface | 17:31 |
| clarkb | jeblair: awesome, thanks that helps | 17:31 |
| jeblair | clarkb: board 41 uses manual columns (with a bot to automatically manage them). board 53 uses automatic columns. | 17:31 |
| openstackgerrit | Merged openstack-infra/openstackid-resources master: Fixed postEvent and publishEvent Endpoints https://review.openstack.org/523484 | 17:31 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Create ovb-ha-ipv6 and deprecate fs024 job https://review.openstack.org/522618 | 17:33 |
| *** Hal has quit IRC | 17:33 | |
| jeblair | EmilienM: i think i have reproduced the issue in 522667 locally | 17:34 |
| jeblair | debugging now | 17:34 |
| EmilienM | cool | 17:35 |
| *** shardy has quit IRC | 17:37 | |
| *** florianf is now known as florianf|afk | 17:37 | |
| sshnaidm | jeblair, EmilienM, mwhahaha, in continue of discussion with pabelanger, jeblair and other infra folks, I created 2 stories in storyboard for zuul project, which will help us to move tripleo jobs to native for zuulv3: https://storyboard.openstack.org/#!/story/2001353 and https://storyboard.openstack.org/#!/story/2001354 | 17:37 |
| sshnaidm | jeblair, pabelanger please tell me if something more is needed or you want me test it | 17:38 |
| *** Apoorva has joined #openstack-infra | 17:40 | |
| *** Apoorva has quit IRC | 17:40 | |
| EmilienM | sshnaidm: sounds good | 17:41 |
| *** Apoorva has joined #openstack-infra | 17:41 | |
| openstackgerrit | Tin Lam proposed openstack-infra/project-config master: Remove all remaining single node jobs https://review.openstack.org/523487 | 17:41 |
| openstackgerrit | Sergii Golovatiuk proposed openstack-infra/tripleo-ci master: Introduce TRIPLEO_HEAT_TEMPLATES_ROOT https://review.openstack.org/523488 | 17:42 |
| *** dhinesh_ has quit IRC | 17:42 | |
| *** dhinesh has joined #openstack-infra | 17:42 | |
| openstackgerrit | Tin Lam proposed openstack-infra/openstack-zuul-jobs master: Remove all legacy zuul jobs for openstack-helm https://review.openstack.org/523489 | 17:43 |
| *** dhinesh has quit IRC | 17:44 | |
| *** dhinesh has joined #openstack-infra | 17:44 | |
| *** jamesmcarthur has joined #openstack-infra | 17:45 | |
| openstackgerrit | Tin Lam proposed openstack-infra/openstack-zuul-jobs master: Remove all legacy zuul jobs for openstack-helm https://review.openstack.org/523489 | 17:45 |
| *** jascott1 has joined #openstack-infra | 17:46 | |
| *** martinkopec has quit IRC | 17:46 | |
| *** yamahata has quit IRC | 17:47 | |
| *** camunoz has joined #openstack-infra | 17:47 | |
| *** dtantsur is now known as dtantsur|afk | 17:50 | |
| *** sambetts is now known as sambetts|afk | 17:50 | |
| *** raissa has quit IRC | 17:51 | |
| *** raissa has joined #openstack-infra | 17:52 | |
| *** pcaruana has quit IRC | 17:53 | |
| EmilienM | http://codesearch.openstack.org/ is down for me or very slow | 17:56 |
| *** pblaho has quit IRC | 17:56 | |
| *** thorst has quit IRC | 17:57 | |
| *** thorst has joined #openstack-infra | 17:58 | |
| *** mihalis68 has quit IRC | 17:59 | |
| *** sshnaidm is now known as sshnaidm|off | 17:59 | |
| *** jamesdenton has joined #openstack-infra | 18:00 | |
| *** trown|lunch is now known as trown | 18:01 | |
| clarkb | could be next victim of the live migrations | 18:01 |
| clarkb | "live" migrations | 18:01 |
| *** yamamoto has joined #openstack-infra | 18:01 | |
| *** tushar_ has joined #openstack-infra | 18:02 | |
| tushar_ | Hi All | 18:02 |
| *** thorst has quit IRC | 18:02 | |
| tushar_ | I am facing issue in PCI passthrough on ubuntu 16.04 | 18:02 |
| *** tosky has quit IRC | 18:03 | |
| *** derekh has quit IRC | 18:04 | |
| jeblair | EmilienM: i believe if you drop the 3.x jobs from the puppet-openstack-module-unit-jobs project-template on the stable/newton branch of puppet-openstack-integration, you'll get the behavior you want. | 18:06 |
| tushar_ | error message - error: Failed to attach device from hostdev.xml error: internal error unable to execute QEMU command 'device_add': Device 'pci-assign' could not be initialized | 18:06 |
| *** yamamoto has quit IRC | 18:06 | |
| tushar_ | any idea about this error? | 18:06 |
| jeblair | tushar_: you probably want the #openstack channel, this channel is for discussing the developer infrastructure for the project | 18:07 |
| jeblair | EmilienM: does that work as a solution, or would that affect too many projects? (ie, it would affect more than puppet-tripleo, it would affect all the projects using that template) | 18:08 |
| tushar_ | jeblair : thanks . I am facing this issue in my third party CI, I thought openstack-infra might be using PCI passthrough while execute zuul check jobs | 18:09 |
| *** Swami has joined #openstack-infra | 18:09 | |
| tushar_ | I will connect in other group | 18:10 |
| clarkb | EmilienM: I'm attempting to ssh to that server and not getting to it so ya think it may be the live migration thing again. I will reboot it shortly | 18:10 |
| *** thorst has joined #openstack-infra | 18:10 | |
| fungi | tushar_: nope, we just use vanilla virtual machine instances from public cloud providers. see https://docs.openstack.org/infra/manual/testing.html for an explanation | 18:10 |
| tushar_ | fungi : okk, thanks. | 18:11 |
| fungi | tushar_: there's also a #openstack-third-party-ci channel you may want to try | 18:11 |
| *** ralonsoh has quit IRC | 18:11 | |
| clarkb | it pings but doesn't ssh which seems to be a common occurence | 18:11 |
| tushar_ | fungi : yeh , will connect there | 18:12 |
| jeblair | EmilienM: (fwiw, i think there may still be a bug here, i'm considering ways to fix it -- just wanted to unblock you asap) | 18:12 |
| dhellmann | is codesearch.openstack.org down? | 18:12 |
| clarkb | #status log rebooting non responsive codesearch.openstack.org. It pings but does not http(s) or ssh. Probably another live migration gone bad | 18:12 |
| openstackstatus | clarkb: finished logging | 18:12 |
| clarkb | dhellmann: yes see ^ | 18:12 |
| dhellmann | heh | 18:12 |
| dhellmann | thanks, clarkb | 18:12 |
| fungi | clarkb: today was a slew of xen advisories, so i'm guessing that's why all the quiet live migration at rax | 18:13 |
| fungi | also several qemu advisories... not a great day for openstack cloud operators, i'm betting | 18:13 |
| clarkb | server is up now, but not hound /me figures that out | 18:14 |
| fungi | clarkb: also remember that hound takes a while to (re)start | 18:14 |
| *** thorst has quit IRC | 18:14 | |
| fungi | which is why we don't auto-restart it for config updates/repo additions | 18:14 |
| clarkb | it appears to be starting up now | 18:15 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Add command_socket setting to executor section https://review.openstack.org/523465 | 18:16 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Add command socket support to zuul-scheduler https://review.openstack.org/523466 | 18:16 |
| *** dsariel has quit IRC | 18:16 | |
| *** yamamoto has joined #openstack-infra | 18:18 | |
| *** yamamoto has quit IRC | 18:18 | |
| clarkb | EmilienM: dhellmann codesearch appears back now | 18:19 |
| dhellmann | clarkb : thanks! | 18:19 |
| clarkb | infra-root fyi when you have servers that ping in rax but do not ssh they appear to be doing "live" migrations there that don't always work properly. A hard reboot of the server seems to get things going again when thsi happens | 18:20 |
| fungi | clarkb: have you looked for open tickets too? | 18:22 |
| clarkb | fungi: I didn't this time because in the past there haven't been any | 18:22 |
| *** lucasagomes is now known as lucas-afk | 18:23 | |
| *** yamahata has joined #openstack-infra | 18:23 | |
| fungi | ahh, there's an open one from 3 hours ago... | 18:23 |
| clarkb | fungi: like when review.o.o was hit. I think from rax's perspective the migration was either successful or in progress but the server was down for an extended period of time from a functionality standpoint and there wasnt tickets then | 18:23 |
| fungi | Rackspace is migrating the cloud servers listed below from their current location in our DFW datacenter to another location in the DFW datacenter. [...] codesearch.openstack.org (3f74b152-eea6-46c5-8fad-e8fd9c142a6d) | 18:23 |
| fungi | i also closed out one from yesterday for... | 18:24 |
| fungi | logstash-worker04.openstack.org | 18:24 |
| fungi | and there's one for stackalytics.openstack.org... i'll check it now | 18:24 |
| fungi | that last one seems fine at least | 18:25 |
| clarkb | ya I think for the most part the live migration do end up being relatively live | 18:25 |
| clarkb | not sure what causes some of these to fail | 18:25 |
| fungi | i wouldn't be surprised if they're not technically "hung" but rather in scheduling limbo | 18:25 |
| fungi | and the sync is taking forever or live cutover hasn't happened yet at least | 18:26 |
| fungi | but for whatever reason they the original instance ends up paused or stopped | 18:27 |
| fungi | s/they/then/ | 18:27 |
| fungi | i'll go ahead and close out the codesearch ticket | 18:27 |
| fungi | ahh, yep there's the sentence i missed: "You are receiving this message because our automatic live-migration tool failed." | 18:28 |
| *** tushar_ has quit IRC | 18:28 | |
| *** dhinesh has quit IRC | 18:29 | |
| *** ansmith has joined #openstack-infra | 18:31 | |
| pabelanger | wonder what the failure rate is on live migrations | 18:31 |
| *** armaan has quit IRC | 18:32 | |
| fungi | based on our observations, somewhere around 5%? | 18:32 |
| *** bobh has joined #openstack-infra | 18:32 | |
| fungi | maybe as high as 10%? | 18:32 |
| *** armaan has joined #openstack-infra | 18:32 | |
| clarkb | heads up infra meeting in just under half an hour | 18:33 |
| fungi | but we're probably also not running the typical workload compared to most rax customers | 18:33 |
| *** dhill_ has quit IRC | 18:33 | |
| *** thorst has joined #openstack-infra | 18:34 | |
| *** rbrndt has quit IRC | 18:35 | |
| fungi | doubling back to a topic from last week... here's how i eventually coerced pyyaml into emitting indented block sequences that would meet the expectations of yamllint: http://paste.openstack.org/show/627624/ | 18:37 |
| fungi | guess i should propose dumper/emitter params upstream to make that an option | 18:37 |
| *** thorst has quit IRC | 18:37 | |
| *** thorst has joined #openstack-infra | 18:38 | |
| ansmith | fungi: hello, regarding https://review.openstack.org/#/c/523455/, would very much like to have single bindep.txt | 18:39 |
| *** dhill_ has joined #openstack-infra | 18:39 | |
| fungi | ansmith: awesome... i expect you can just merge those lists but tag the divergent lines with your use-specific profile names instead of having a separate file for each | 18:40 |
| ansmith | fungi: the issue we have is how to set the bindep_profile for each driver tests | 18:40 |
| ansmith | fungi: seems to be fixed as 'test' in the bindep role | 18:40 |
| fungi | ansmith: cool, so that sounds like something we should make more extensible in the ansible role | 18:41 |
| fungi | so that you can pass a profile parameter rather than always getting the "test" profile | 18:41 |
| ansmith | fungi: that would be great, that is why we split out to files as passing bindep_file works | 18:41 |
| fungi | pabelanger: ^ happen to know the best way to plumb that in as an optional parameter but have it continue to default to the current value? | 18:42 |
| *** adisky_ has quit IRC | 18:43 | |
| fungi | i'm poking around in http://git.openstack.org/cgit/openstack-infra/zuul-jobs/tree/roles/bindep/ in hopes it'll be fairly intuitive | 18:44 |
| *** jamesmcarthur has quit IRC | 18:45 | |
| fungi | looks like it's already at least mostly set up for that | 18:45 |
| fungi | we have a bindep_profile defaulting to test in defaults/main.yaml | 18:45 |
| pabelanger | fungi: we'd likely need to make a change to use something like {{ bindep_profile | default(omit) }}, then update our bindep_profile variable in base playbooks to be test | 18:45 |
| pabelanger | otherwise, we could try passing bindep_profile: "" to the role | 18:46 |
| fungi | pabelanger: well... i see that the bindep_run task passes {{ bindep_profile }} | 18:46 |
| pabelanger | yah, jobs should be able to override it from test | 18:47 |
| fungi | so i guess what we're missing is plumbing for that in our shared jobs | 18:47 |
| pabelanger | fungi: ansmith: which job are you currently? | 18:47 |
| fungi | pabelanger: https://review.openstack.org/523455 | 18:48 |
| pabelanger | http://git.openstack.org/cgit/openstack-infra/project-config/tree/playbooks/puppet-tarball/pre.yaml#n3 is what we do for puppet-tarball jobs, to use the 'build' profile | 18:48 |
| fungi | pabelanger: so openstack-tox0py27 is being derived | 18:48 |
| fungi | er, openstack-tox-py27 and openstack-tox-py35 i guess | 18:49 |
| ansmith | parent is openstack-tox-py* | 18:49 |
| pabelanger | looking | 18:49 |
| pabelanger | yah, Isee the issue | 18:50 |
| pabelanger | we hardcode bindep_profile to test in unittests/pre.yaml | 18:51 |
| pabelanger | we'll have to stop doing that | 18:51 |
| pabelanger | I think we can just remove it | 18:51 |
| fungi | pabelanger: we can just drop it there since the bindep role already has test as a defauly | 18:51 |
| fungi | er, yep | 18:51 |
| pabelanger | and default to what the role provides | 18:51 |
| pabelanger | yah | 18:51 |
| jeblair | clarkb: i may be slightly late for meeting | 18:52 |
| ansmith | fungi, pabelanger: thanks for checking it out, would it be hard to fix? | 18:53 |
| *** dhinesh has joined #openstack-infra | 18:53 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Remove bindep_profile from unittests/pre.yaml https://review.openstack.org/523501 | 18:53 |
| pabelanger | ansmith: fungi: ^not hard at all | 18:54 |
| ansmith | pabelanger: perfect, thanks! | 18:54 |
| clarkb | jeblair: ok, maybe we'll talk about the not zuulv3 thing first | 18:54 |
| fungi | ansmith: you should even be able to depends-on that from your change | 18:54 |
| ansmith | fungi: yes, thanks | 18:56 |
| *** eumel8 has joined #openstack-infra | 18:57 | |
| openstackgerrit | Graham Hayes proposed openstack-infra/openstack-zuul-jobs master: Remove legacy Designate jobs content https://review.openstack.org/523502 | 18:57 |
| openstackgerrit | Graham Hayes proposed openstack-infra/project-config master: Remove legacy designate jobs https://review.openstack.org/523503 | 18:57 |
| clarkb | ok infra meeting now | 19:00 |
| clarkb | join us in #openstack-meeting | 19:00 |
| *** armaan has quit IRC | 19:01 | |
| *** armaan has joined #openstack-infra | 19:02 | |
| *** jamesmcarthur has joined #openstack-infra | 19:03 | |
| *** janki has quit IRC | 19:05 | |
| *** bobh has quit IRC | 19:05 | |
| *** jamesmcarthur has quit IRC | 19:07 | |
| *** rtjure has joined #openstack-infra | 19:13 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/zuul-jobs master: DO NOT MERGE: testing the bindep_profile fallback https://review.openstack.org/523506 | 19:13 |
| eumel8 | hello infra-team! Zanata team and I18n worked again on simplification of puppet-zanata. Maybe you have time for review: https://review.openstack.org/#/c/506795/ | 19:16 |
| clarkb | eumel8: yup I just brought it up in the infra meeting and will look to review it today | 19:16 |
| clarkb | eumel8: ianw and I will likely be your volunteers for the infra side of the upgrade if you want to ping us for when more items come up | 19:17 |
| *** salv-orlando has joined #openstack-infra | 19:17 | |
| *** yamamoto has joined #openstack-infra | 19:18 | |
| *** rtjure has quit IRC | 19:18 | |
| *** claudiub|2 has quit IRC | 19:20 | |
| jeblair | dmsimard: there's a topic in the infra meeting of interest to you (zuul-jobs third-party ci) | 19:20 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove legacy-cinder-tox-* jobs from Cinder https://review.openstack.org/521655 | 19:20 |
| dmsimard | damn it I forgot to fix my calendar event timezone | 19:21 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/openstack-zuul-jobs master: Remove legacy-cinder-tox-* jobs https://review.openstack.org/521656 | 19:22 |
| *** jamesmcarthur has joined #openstack-infra | 19:23 | |
| flaper87 | hey folks, would it be possible to set an autohold on the kubernetes job in this patch? https://review.openstack.org/#/c/521951/ I'm trying to debug what's going on with the network in that deployment and it's kind of difficult to do it just using logs | 19:23 |
| flaper87 | I have a deployment in my local dev environment that works :/ | 19:24 |
| eumel8 | thx clarkb | 19:25 |
| *** electrofelix has quit IRC | 19:25 | |
| *** jamesmcarthur has quit IRC | 19:27 | |
| *** yamamoto has quit IRC | 19:28 | |
| *** tosky has joined #openstack-infra | 19:29 | |
| *** priteau has quit IRC | 19:30 | |
| *** Goneri has quit IRC | 19:31 | |
| *** jamesmcarthur has joined #openstack-infra | 19:32 | |
| *** ijw has quit IRC | 19:33 | |
| *** rbrndt has joined #openstack-infra | 19:34 | |
| *** sdague has quit IRC | 19:37 | |
| *** links has joined #openstack-infra | 19:38 | |
| *** links has quit IRC | 19:42 | |
| fungi | flaper87: we can set zuul to hold the next N nodes for failing runs of ansible-role-k8s-keystone-kubernetes-centos across the system. perhaps with suitably large N that's sufficient? | 19:42 |
| flaper87 | fungi: that would be great. How long would it hold those nodes for? | 19:43 |
| fungi | until one of us clears the hold manually | 19:44 |
| flaper87 | fungi: cool, let's do it. I think N can be 1 for now. I'm installing everything in a single node and I'm the only one running these jobs | 19:44 |
| flaper87 | fungi: I guess you need my ssh key, right? https://github.com/flaper87.keys | 19:45 |
| openstackgerrit | Ihar Hrachyshka proposed openstack-infra/project-config master: neutron: don't execute legacy in-tree tempest jobs for master https://review.openstack.org/523514 | 19:45 |
| clarkb | looks like its a single node job? | 19:45 |
| clarkb | not sure how networking would be drastically funny in that cse | 19:45 |
| flaper87 | fungi: can we do 1 for ansible-role-k8s-keystone-kubernetes-centos and one for ansible-role-k8s-keystone-openshift-centos | 19:45 |
| fungi | flaper87: autohold set with --count=1 | 19:45 |
| fungi | i can get you access to it once the job completes | 19:45 |
| fungi | just let me know | 19:46 |
| flaper87 | fungi: cool, let me recheck it | 19:46 |
| clarkb | but reading logs my hunch is firewall | 19:46 |
| clarkb | flaper87: ^ | 19:46 |
| clarkb | dial tcp 10.0.1.46:2379: getsockopt: connection refused" says firewall to me | 19:46 |
| flaper87 | clarkb: that's my hunch too, need to understand exactly what the heck it is doing and get the right rule | 19:46 |
| clarkb | flaper87: we bloc keverything but ssh to the "external" ip by default | 19:47 |
| flaper87 | clarkb: a-ha, mmh, that could be the issue. Although, kubespray sets the firewall rules for k8s. Maybe not all of them | 19:48 |
| flaper87 | http://logs.openstack.org/51/521951/20/check/ansible-role-k8s-keystone-kubernetes-centos/7288a26/primary/logs/iptables.txt.gz | 19:48 |
| flaper87 | that's the iptable rules | 19:48 |
| * flaper87 has the feeling this will end up being something obvious and dumb | 19:50 | |
| fungi | isn't it usually? | 19:52 |
| *** gouthamr_ has joined #openstack-infra | 19:53 | |
| *** bobh has joined #openstack-infra | 19:54 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Remove bindep_profile from unittests/pre.yaml https://review.openstack.org/523501 | 19:54 |
| flaper87 | it's basically my life, yes! Stumbling from one dumb issue to the other | 19:54 |
| *** gouthamr has quit IRC | 19:56 | |
| *** tlaxkit has joined #openstack-infra | 19:57 | |
| pabelanger | jeblair: dmsimard: If you'd like to re-review: https://review.openstack.org/513506/ so we can move forward with removing fetch-zuul-cloner from base jobs. Thanks! | 19:58 |
| mtreinish | clarkb: was there ever a patched pushed up for the git 2.15.0 issues with git-review? | 19:58 |
| mtreinish | clarkb: because I'm starting to get annoyed having to deal with the errors every time I use git review | 19:58 |
| clarkb | mtreinish: yes update git review | 19:59 |
| *** bobh has quit IRC | 19:59 | |
| clarkb | mtreinish: we made a release that should have it working again (works for me locally at least) | 20:00 |
| mtreinish | ah, ok. I guess I was just out of date | 20:01 |
| dmsimard | pabelanger: it's a good observation, there's actually nothing preventing jobs from setting whatever DNS resolvers they like | 20:01 |
| mtreinish | just installed 2.16.0 I'll give it a try now | 20:01 |
| dmsimard | pabelanger: with or without the configure-unbound role fwiw | 20:01 |
| clarkb | pabelanger: my biggest concern is that knowingly breaking dns for eg github should be an opt in rather than a global thing or opt out. I mean I totally break dns for things at home but I opt into that :) | 20:01 |
| dmsimard | echo "8.8.8.8" >/etc/resolv.conf | 20:01 |
| clarkb | dmsimard: pabelanger also another data point is that docker will default to 8.8.8.8 if the hosts resolv.conf is for localhost | 20:02 |
| clarkb | dmsimard: pabelanger any cahnce these are dns lookup failures happening due to overrides like ^ | 20:02 |
| pabelanger | dmsimard: clarkb: right, but in the past, jobs that have changed unbound settings, tend to have more DNS failures then others. And really didn't understand why we setup unbound in the first place. | 20:02 |
| clarkb | pabelanger: yup | 20:02 |
| clarkb | pabelanger: for example bypassing unbound entirely because docker | 20:02 |
| dmsimard | pabelanger: well there's a saying that goes with "great power comes great responsability" ? that comes with the power of shooting yourself in the foot | 20:03 |
| fungi | ansmith: 523501 merged a few minutes ago, so you should be all set to configure bindep_profile from your job definitions now | 20:03 |
| dmsimard | we're defaulting to sane configurations, if they knowingly change that, they're accountable | 20:03 |
| clarkb | dmsimard: pabelanger considering the failurse we've seen are for containery projects it may be good to just double check really quickly that our dns is being used in the first place before we try to optimize it | 20:03 |
| *** jamesmcarthur has quit IRC | 20:03 | |
| *** e0ne has quit IRC | 20:03 | |
| ansmith | fungi: trying it out now, thanks again | 20:03 |
| *** jamesmcarthur has joined #openstack-infra | 20:04 | |
| pabelanger | right, I agree it should be opt-in, just keep in mind could mean more education on why it is a good / bad thing to break github.com :) | 20:04 |
| dmsimard | I don't expect that enforcing a higher minimum TTL will break github, and if it does, it will probably be at a reduced frequency than it is occurring today | 20:05 |
| dmsimard | I mean, we can look but I don't think they change frontend IPs 50 times a day | 20:05 |
| pabelanger | I'd likely suggest min-ttl of 300 to start | 20:05 |
| clarkb | dmsimard: they've communicated that they may simply by setting the ttl to 20 seconds | 20:05 |
| pabelanger | then double to 600, if still rate limited | 20:06 |
| clarkb | pabelanger: do we know that we are rate limited? or that these jobs are even resolving against unbound? | 20:06 |
| dmsimard | clarkb: the TTL is a suggestion, most public resolvers (surely google included) have minimum TTLs | 20:06 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove all remaining single node jobs https://review.openstack.org/523487 | 20:06 |
| clarkb | (the addition of unbound logs should help figure out the second thing, not sure if it will help with the first) | 20:06 |
| pabelanger | clarkb: most of the failures are from tripleo, when I last checked they are still correctly using unbound | 20:06 |
| clarkb | pabelanger: even via the docker containers? | 20:07 |
| pabelanger | clarkb: does google or opendns have an API to check ratelimit? | 20:07 |
| clarkb | pabelanger: I don't know of any api to check ratelimits on public dns servers | 20:07 |
| pabelanger | clarkb: I don't know that, things I have seen are not in containers | 20:07 |
| dmsimard | clarkb: so anyway, what I'm saying is that even if the github.com TTL is 1 second, and they change IPs -- and they kill the servers that owned those IPs right away, it'll knock out like half the planet using github because their resolvers haven't caught up yet | 20:07 |
| clarkb | pabelanger: ok, the specific case I am thinking of is docker will override your dns resolver config if your host's resolver is localhost. If failures happen outside of containers then this is likely not an issue | 20:08 |
| clarkb | dmsimard: ya so its common to do things like double ttl or whatever | 20:08 |
| clarkb | dmsimard: but double 40 seconds or whatever the ttl is is still much shorter than 300 | 20:09 |
| *** bobh has joined #openstack-infra | 20:09 | |
| pabelanger | clarkb: Yah, I remember you saying that about docker, but thought it was fixed | 20:09 |
| clarkb | (and I totally do terrible things to dns at home so I get it, its a thing people like to do. I'm just not sure its appropriate globally on our test infrastructure) | 20:09 |
| pabelanger | yah, that make sense | 20:10 |
| ianw | clarkb: when time, might like to look @ 512245 which does the automatic -eol tag detection workspaces under zuulv2. i am just tesitng in 523509 it doesn't break zuulv3 (as it touches the tag detection), otherwise am +2 | 20:10 |
| dmsimard | clarkb: I'd be super comfortable testing out 900 minimum TTL, remember that the jobs are running on ephemeral nodes and they don't have stale cache at the beginning -- their first lookup will be cached for 15 minutes and it's probably very OK | 20:10 |
| clarkb | (turns out you can use your dns server as a bad adblocker >_>, run http server with 1x1 transparent gif at ip address. have dns server lookup known bad names in table if match send them to that ip address) | 20:10 |
| *** armaan has quit IRC | 20:10 | |
| dhellmann | is gerrit down? I can't load a web view or clone with "git review -d" | 20:11 |
| clarkb | dhellmann: https just loaded for me | 20:11 |
| smcginnis | dhellmann: Was just going to ask too. Hung loading for me. | 20:11 |
| dhellmann | https://review.openstack.org/#/c/523233/ | 20:11 |
| dmsimard | dhellmann: up for me | 20:11 |
| dhellmann | it doesn't seem to be specific to that patch | 20:11 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove legacy designate jobs https://review.openstack.org/523503 | 20:12 |
| dhellmann | ah, now it loaded | 20:12 |
| fungi | dhellmann: smcginnis: i'm flushing gerrit's caches to confirm whether removing the old zuul account's voting permissions has properly taken effect | 20:12 |
| openstackgerrit | Merged openstack-infra/project-config master: Set up doc-building jobs for governance-sigs repo https://review.openstack.org/523423 | 20:12 |
| clarkb | java melody says we've been doing some garbage collecting today | 20:12 |
| clarkb | fungi: aha that may be it | 20:12 |
| dhellmann | whee | 20:12 |
| clarkb | everything has to hit the db instead of cache | 20:12 |
| *** HenryG has quit IRC | 20:12 | |
| fungi | usually it's relatively non-impacting, but perhaps confluence of other factors | 20:12 |
| dhellmann | seems likely | 20:12 |
| fungi | it should finish here in a moment | 20:12 |
| smcginnis | My review did finally load. Just took an inordinate amount of time. | 20:12 |
| fungi | takes 3-4 minutes to flush these days | 20:12 |
| fungi | especially now that we don't restart gerrit weekly ;) | 20:13 |
| *** HenryG has joined #openstack-infra | 20:13 | |
| openstackgerrit | Merged openstack-infra/project-config master: periodic py jobs for vmware-nsx ocata and pike https://review.openstack.org/521538 | 20:13 |
| fungi | it just completed | 20:14 |
| *** bobh has quit IRC | 20:14 | |
| *** ldnunes has quit IRC | 20:14 | |
| fungi | infra-root: recent gerrit behavior change... removing voting permission from an account does not seem to hide its prior votes in open changes any longer | 20:15 |
| clarkb | huh | 20:15 |
| fungi | i may have to script something against the api to delete jenkins verify -1 votes from open changes after all | 20:15 |
| fungi | unless i've missed somewhere else it was also getting permission | 20:16 |
| openstackgerrit | David Moreau Simard proposed openstack-infra/project-config master: Add documentation build and publish jobs for python-tripleoclient https://review.openstack.org/523523 | 20:16 |
| openstackgerrit | Merged openstack-infra/nodepool feature/zuulv3: Fix broken use of pre-existing cloud images https://review.openstack.org/498050 | 20:16 |
| efried | dhellmann Noticed this: https://wiki.openstack.org/wiki/Bugs#Status doesn't seem to reflect reality for Fix Committed & Fix Released anymore. But not sure what it should actually say. | 20:16 |
| clarkb | dmsimard: pabelanger also we should round robin between providers right? seems odd we'd generate enough traffic to cause one or both of them to rate limit us | 20:17 |
| dmsimard | clarkb: I don't know unbound enough to know whether it actually round-robins or not | 20:17 |
| openstackgerrit | Merged openstack-infra/project-config master: Remove legacy jobs from Bandit https://review.openstack.org/522504 | 20:17 |
| openstackgerrit | Merged openstack-infra/project-config master: Add release-notes-jobs to kuryr-kubernetes https://review.openstack.org/523376 | 20:17 |
| clarkb | dmsimard: iirc reading the logs from the last time we debugged it it does round robin them | 20:17 |
| clarkb | dmsimard: so each request that goes out to the upstream server is handed to the least recently used upstream server I think | 20:17 |
| clarkb | dmsimard: and it will mix ipv4 and ipv6 | 20:18 |
| dhellmann | efried : wow, that's old old stuff. I don't see a relevant page in https://docs.openstack.org/project-team-guide/ to replace it though | 20:18 |
| pabelanger | I am not sure, did we land the debug patch to unbound to increase logs? | 20:18 |
| efried | dhellmann I got there via https://launchpad.net/~nova-bugs FWIW | 20:19 |
| clarkb | pabelanger: dmsimard I've got to grab lunch now but can help with ^ if not yet done | 20:19 |
| pabelanger | clarkb: dmsimard: I haven't actually debugged a github.com failure, aside from it having a TTL of 30 | 20:19 |
| dmsimard | clarkb, pabelanger: that was https://review.openstack.org/#/c/521500/ | 20:19 |
| dhellmann | efried : make sure to wear a filter mask when you go into those dusty corners! :-) | 20:19 |
| clarkb | that should also tell us if the unbound server is being asked for those names at all | 20:19 |
| dhellmann | efried : seriously, we mark bugs as "fix released" when a commit merges now because that's a "closed" state in launchpad | 20:19 |
| *** eharney has quit IRC | 20:19 | |
| dhellmann | and the release tools leave a comment about the actual release that includes the fix | 20:20 |
| *** bobh has joined #openstack-infra | 20:20 | |
| efried | dhellmann I only just found out there was actual documentation for this stuff, as opposed to purely tribal knowledge. | 20:20 |
| dhellmann | I don't think we use fix committed at all | 20:20 |
| efried | dhellmann Okay, that's about what I thought. Want me to update that page accordingly? | 20:20 |
| dhellmann | efried : I'm not sure I even realized that wiki page existed | 20:20 |
| fungi | dhellmann: efried: also as a parallel for storyboard, the closed states are merged and invalid (no distinction between committed and released there) | 20:21 |
| dhellmann | efried : how do you feel about moving that info over to the project-team-guide instead? | 20:21 |
| *** Goneri has joined #openstack-infra | 20:21 | |
| *** pcaruana has joined #openstack-infra | 20:21 | |
| openstackgerrit | sebastian marcet proposed openstack-infra/openstackid-resources master: Refactoring/Update of logic for SummitEvent Update https://review.openstack.org/523526 | 20:22 |
| efried | dhellmann I'm not opposed to that. The LP links would need to be updated eventually as well. I'll need a jumpstart on how to do said move. There's a project called openstack/project-team-guide - I assume that's where you're talking about. Where in there do we go? | 20:23 |
| dhellmann | efried : you'd want to make a new file with a name like doc/source/bugs.rst for the content and add it to the toctree in doc/source/index.rst | 20:23 |
| dhellmann | we can replace the body of the wiki page with a link to the new published location | 20:24 |
| efried | dhellmann Got it. And bleh, porting from wiki to rst, fun. Okay, I'll see what I can do. | 20:24 |
| dhellmann | that avoids having to hunt down all existing references to the old location | 20:24 |
| dhellmann | you can leave out the list of links to all of the project but trackers | 20:24 |
| dhellmann | efried : the list-table directive is your friend for reformatting those tables | 20:24 |
| dhellmann | efried : and thanks, I appreciate you going to the extra trouble | 20:25 |
| *** bobh has quit IRC | 20:26 | |
| openstackgerrit | David Moreau Simard proposed openstack-infra/project-config master: Add documentation build and publish jobs for python-tripleoclient https://review.openstack.org/523523 | 20:26 |
| *** iyamahat has joined #openstack-infra | 20:26 | |
| efried | dhellmann Thank me if I don't give up partway through :$ | 20:26 |
| dmsimard | AJaeger: bad copy pasta, we don't really need to filter, fixed ^ | 20:26 |
| dhellmann | efried : if you run into trouble or out of time put up a partial patch and I'll pitch in | 20:26 |
| efried | dhellmann Cool | 20:26 |
| *** bobh has joined #openstack-infra | 20:26 | |
| dhellmann | efried : if you copy over all of the content, even if the formatting is wrong, we can tag-team | 20:27 |
| efried | dhellmann Roger that. | 20:27 |
| *** esberglu has quit IRC | 20:27 | |
| efried | dhellmann Where does this content render to? | 20:28 |
| dhellmann | https://docs.openstack.org/project-team-guide/ | 20:28 |
| efried | dhellmann Thanks. Gonna go read that and get a sense for style/layout/organization first... | 20:28 |
| fungi | 4339 changes currently match 'is:open label:verified=-1,jenkins' | 20:28 |
| dhellmann | efried : ++ | 20:28 |
| *** bobh has quit IRC | 20:32 | |
| spotz | Hey all if anyone could help push this along would appreciate it https://review.openstack.org/521972 | 20:32 |
| *** jpich has quit IRC | 20:33 | |
| fungi | infra-root: ^ system config change in support of the reinvigoration of the diversity working group | 20:35 |
| fungi | (also a very easy review!) | 20:35 |
| *** dhajare has quit IRC | 20:38 | |
| *** bobh has joined #openstack-infra | 20:38 | |
| *** jamesmcarthur has quit IRC | 20:38 | |
| jkilpatr | so why does this change to our tox.ini cause retry limit exceeded on linters? https://review.openstack.org/#/c/506351/9/tox.ini Can't seem to figure it out. | 20:39 |
| jkilpatr | it looks like what's failing is some sort of log upload task, I don't see how that's related to the change. | 20:39 |
| *** jamesmcarthur has joined #openstack-infra | 20:40 | |
| *** esberglu has joined #openstack-infra | 20:40 | |
| *** bobh has quit IRC | 20:43 | |
| *** wolverineav has quit IRC | 20:43 | |
| fungi | ERROR: unknown environment 'linters' | 20:44 |
| *** wolverineav has joined #openstack-infra | 20:44 | |
| fungi | jkilpatr: http://logs.openstack.org/51/506351/9/check/openstack-tox-linters/9e2a3a7/job-output.txt.gz#_2017-11-28_13_07_31_857271 | 20:44 |
| fungi | and since it's hitting that in a "pre" task zuul retries the job instead of failing it the first time | 20:45 |
| *** wolverineav has quit IRC | 20:45 | |
| fungi | because failures in "pre" are sometimes due to flaky instances/networks/providers | 20:45 |
| *** wolverineav has joined #openstack-infra | 20:46 | |
| *** wolverin_ has joined #openstack-infra | 20:48 | |
| *** wolverineav has quit IRC | 20:48 | |
| pabelanger | spotz: fungi: +3 | 20:49 |
| *** priteau has joined #openstack-infra | 20:50 | |
| flaper87 | fungi: seems to have failed already | 20:52 |
| flaper87 | fungi: ansible-role-k8s-keystone-kubernetes-centos | 20:52 |
| spotz | pabelanger: :) | 20:52 |
| flaper87 | fungi: https://github.com/flaper87.keys | 20:52 |
| *** edmondsw_ has joined #openstack-infra | 20:52 | |
| flaper87 | fungi: http://logs.openstack.org/51/521951/21/check/ansible-role-k8s-keystone-kubernetes-centos/d959691/ <- not sure if you need this | 20:52 |
| openstackgerrit | Mikhail S Medvedev proposed openstack-infra/devstack-gate master: Checkout from -eol tag on zuul v2.x https://review.openstack.org/521936 | 20:53 |
| *** bobh has joined #openstack-infra | 20:53 | |
| *** dhajare has joined #openstack-infra | 20:54 | |
| *** edmondsw has quit IRC | 20:56 | |
| fungi | flaper87: ssh zuul@104.130.199.66 and see if it lets you in now | 20:57 |
| *** dprince has quit IRC | 20:57 | |
| flaper87 | fungi: I'm in, thanks! I'll report back again tomorrow and let y'all know | 20:57 |
| fungi | flaper87: thanks, just lmk when you're done and i'll delete it | 20:58 |
| * fungi disappears to find some food... bbiaw | 20:58 | |
| *** bobh has quit IRC | 20:58 | |
| *** bobh has joined #openstack-infra | 21:01 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Create ovb-ha-ipv6 and deprecate fs024 job https://review.openstack.org/522618 | 21:03 |
| *** thorst has quit IRC | 21:03 | |
| *** bobh has quit IRC | 21:06 | |
| *** dhajare has quit IRC | 21:06 | |
| *** bobh has joined #openstack-infra | 21:10 | |
| openstackgerrit | Merged openstack-infra/system-config master: Add #openstack-diversity to meetbot and statusbot channels https://review.openstack.org/521972 | 21:10 |
| *** reed has quit IRC | 21:14 | |
| *** reed has joined #openstack-infra | 21:14 | |
| *** bobh has quit IRC | 21:15 | |
| *** armax has quit IRC | 21:17 | |
| *** gouthamr_ is now known as gouthamr | 21:21 | |
| *** pcaruana has quit IRC | 21:21 | |
| *** smatzek has quit IRC | 21:22 | |
| *** bobh has joined #openstack-infra | 21:24 | |
| *** jamesmcarthur has quit IRC | 21:25 | |
| *** kgiusti has left #openstack-infra | 21:26 | |
| *** jamesmcarthur has joined #openstack-infra | 21:26 | |
| clarkb | dmsimard: pabelanger I've approved the unbound logging verbosity increase change. FWIW when we debugged previous unbound/dns issues we just flipped that switch in specific jobs. | 21:27 |
| *** edmondsw_ is now known as edmondsw | 21:27 | |
| clarkb | The size of the file wasn't very large so should be fine to do globally, just more time to turnaround the switch because new images must be built | 21:27 |
| *** jamesmcarthur has quit IRC | 21:28 | |
| *** bobh has quit IRC | 21:29 | |
| *** sboyron has quit IRC | 21:30 | |
| *** thorst has joined #openstack-infra | 21:30 | |
| *** sboyron has joined #openstack-infra | 21:31 | |
| *** rbrndt has quit IRC | 21:31 | |
| *** kjackal has quit IRC | 21:31 | |
| openstackgerrit | Merged openstack-infra/project-config master: Set unbound logging to debug https://review.openstack.org/521500 | 21:32 |
| *** rcernin has quit IRC | 21:33 | |
| jeblair | EmilienM: if you have a sec to answer my question from earlier, it would be helpful (even though i'm almost done fixing the bug that made this behave in an unexpected way) | 21:34 |
| EmilienM | jeblair: I think I missed the ping, let me read again | 21:35 |
| jeblair | EmilienM: http://eavesdrop.openstack.org/irclogs/%23openstack-infra/%23openstack-infra.2017-11-28.log.html#t2017-11-28T18:06:10 and 2 lines that follow | 21:35 |
| EmilienM | jeblair: it doesn't work for us | 21:36 |
| EmilienM | jeblair: because we want to run puppet3 jobs on stable/newton for puppet projects but on puppet-tripleo no | 21:36 |
| EmilienM | the layout is different | 21:37 |
| EmilienM | such are the requirements | 21:37 |
| jeblair | EmilienM: cool, thanks, that helps. | 21:37 |
| jeblair | EmilienM: i'm writing a commit message for a fix now. maybe we can get it in for tomorrow. | 21:37 |
| EmilienM | jeblair: well, it's complex I know :D | 21:37 |
| EmilienM | maybe my use case is stupid | 21:37 |
| EmilienM | but we could do it before zuul v3 | 21:38 |
| *** bobh has joined #openstack-infra | 21:38 | |
| jeblair | EmilienM: nope, i think it makes a lot of sense and this is definitely not doing the thing you or i or pabelanger expect it to do. :) | 21:39 |
| *** ansmith has quit IRC | 21:39 | |
| EmilienM | ok | 21:39 |
| *** openstackstatus has quit IRC | 21:42 | |
| *** openstack has joined #openstack-infra | 21:44 | |
| *** ChanServ sets mode: +o openstack | 21:44 | |
| *** bobh has quit IRC | 21:44 | |
| *** sdague has joined #openstack-infra | 21:44 | |
| *** erlon has quit IRC | 21:44 | |
| *** bobh has joined #openstack-infra | 21:48 | |
| *** hamzy has quit IRC | 21:50 | |
| *** bobh has quit IRC | 21:53 | |
| *** slaweq_ has quit IRC | 21:55 | |
| *** slaweq has joined #openstack-infra | 21:56 | |
| *** jamesmcarthur has joined #openstack-infra | 21:56 | |
| *** slaweq_ has joined #openstack-infra | 21:56 | |
| *** thorst has quit IRC | 21:56 | |
| *** slaweq_ has quit IRC | 21:57 | |
| *** slaweq_ has joined #openstack-infra | 21:57 | |
| clarkb | ianw: sambetts|afk I am reading https://review.openstack.org/#/c/512245/3/functions.sh and isn't that mostly a noop change since the only path updating is the setup_project path which doesn't run anymore under zuulv3? I guess this is for enabling third party ci? | 21:57 |
| *** bobh has joined #openstack-infra | 21:58 | |
| *** jamesmca_ has joined #openstack-infra | 21:59 | |
| *** slaweq has quit IRC | 22:00 | |
| *** ijw has joined #openstack-infra | 22:01 | |
| *** trown is now known as trown|outtypewww | 22:01 | |
| *** armax has joined #openstack-infra | 22:02 | |
| *** bobh has quit IRC | 22:03 | |
| *** bmjen has quit IRC | 22:03 | |
| *** bmjen has joined #openstack-infra | 22:03 | |
| *** slaweq_ has quit IRC | 22:03 | |
| *** slaweq has joined #openstack-infra | 22:04 | |
| *** jamesmcarthur has quit IRC | 22:04 | |
| *** slaweq has quit IRC | 22:04 | |
| *** oneswig has joined #openstack-infra | 22:05 | |
| oneswig | Hello infra - our meeting had a meetbot restart, now it's not doing endmeeting. | 22:05 |
| oneswig | Do I need to do anything? | 22:05 |
| oneswig | Channel #openstack-meeting | 22:06 |
| *** hemna_ has joined #openstack-infra | 22:06 | |
| openstackgerrit | Julia Kreger proposed openstack-infra/shade master: De-client-ify baremetal node_set_provision_state https://review.openstack.org/499773 | 22:07 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Combine branch templates and pipeline branch matchers https://review.openstack.org/523544 | 22:07 |
| jeblair | EmilienM, pabelanger: ^ that should address the puppet-tripleo template issue | 22:07 |
| *** ijw has quit IRC | 22:07 | |
| EmilienM | jeblair: thanks | 22:09 |
| *** bobh has joined #openstack-infra | 22:09 | |
| *** ijw has joined #openstack-infra | 22:09 | |
| clarkb | oneswig: if the meetbot restarts then there is no longer a running meeting to end | 22:10 |
| *** rbrndt has joined #openstack-infra | 22:10 | |
| EmilienM | jeblair: it looks like our case was complex but somehow valid | 22:10 |
| oneswig | clarkb: thanks - I guess that means I have nothing to do. | 22:10 |
| *** mat128 has quit IRC | 22:13 | |
| *** slaweq has joined #openstack-infra | 22:14 | |
| *** bobh has quit IRC | 22:14 | |
| *** ijw has quit IRC | 22:15 | |
| clarkb | sambetts|afk: ianw I went ahead and approved the cahnge as I'm fairly certain it is a noop for us and if it makes third party ci happier then yay | 22:22 |
| ianw | clarkb: yep, it's for the v2 path | 22:22 |
| *** oneswig has quit IRC | 22:22 | |
| *** rcernin has joined #openstack-infra | 22:23 | |
| *** priteau has quit IRC | 22:23 | |
| ianw | i did a test with it under our v3, since it touches the eol tag detection, and it seems to work, i think i commeneted | 22:23 |
| *** jamesmca_ has quit IRC | 22:25 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci master: Fix sub_nodes and sub_nodes_private with multiple nodes https://review.openstack.org/523226 | 22:25 |
| *** jamesmcarthur has joined #openstack-infra | 22:25 | |
| *** rhallisey has quit IRC | 22:26 | |
| *** thorst has joined #openstack-infra | 22:27 | |
| *** jamesmcarthur has quit IRC | 22:28 | |
| * clarkb bugs infra-root one more time for reviews on https://review.openstack.org/522399 and https://review.openstack.org/523474 before diving into the storyboards | 22:28 | |
| *** jtomasek_ has quit IRC | 22:29 | |
| ianw | clarkb: will we never have a digit in there? like maybe a datestamp or something? | 22:30 |
| clarkb | ianw: ya the idea is to collapse datestamps or iteration numbers as logically its all the same content to be learned on | 22:31 |
| *** thorst has quit IRC | 22:31 | |
| ianw | ohhh, [^ | 22:34 |
| *** jaypipes has quit IRC | 22:35 | |
| *** jamesmcarthur has joined #openstack-infra | 22:36 | |
| *** edmondsw has quit IRC | 22:36 | |
| ianw | clarkb: i'm wondering if logrotate names things differently on at least fedora ... never considered that before :) | 22:37 |
| jeblair | clarkb: on 523474 -- could we stick a note in there, at least as a review comment, indicating why it's okay to expand back to 10 days? | 22:37 |
| clarkb | ianw: ya its possible the regex there could be much more robust for all the varieties of log rotations/iterations | 22:37 |
| clarkb | jeblair: I can | 22:37 |
| jeblair | clarkb: i've +2d that -- maybe you could go ahead and +W it with the note? | 22:37 |
| *** thorst has joined #openstack-infra | 22:38 | |
| *** jamesmcarthur has quit IRC | 22:38 | |
| efried | dhellmann wotcher: https://review.openstack.org/523551 Port wiki bugs pages (1 of ?) | 22:38 |
| *** edmondsw has joined #openstack-infra | 22:39 | |
| efried | (dhellmann What channel gets patchbot notifications for project-team-guide ?) | 22:39 |
| ianw | clarkb: ok, so it's dateext in /etc/logrotate.conf which fedora has but debian doesn't | 22:39 |
| dhellmann | efried : I'm not sure if any channels do | 22:39 |
| dhellmann | efried : oh, it looks like #openstack-tc does | 22:40 |
| efried | k, if there's a more appropriate place for us to discuss, let me know. | 22:40 |
| efried | okay, joining... | 22:40 |
| clarkb | jeblair: just posted, how does that look? | 22:40 |
| ianw | clarkb: while we don't need to support arbitrary strings, it's probably wise to at least handle dateext/!dateext | 22:40 |
| *** thorst has quit IRC | 22:42 | |
| clarkb | ianw: dateext is something like nova.log.2017-11-28(maybetimestampheretoo?) | 22:42 |
| ianw | clarkb: no it's just - and numbers "Archive old versions of log files adding a date extension like YYYYMMDD instead of simply adding a number. " | 22:43 |
| ianw | /var/log/cron-20171105 | 22:43 |
| ianw | /var/log/cron-20140720.gz | 22:43 |
| *** edmondsw has quit IRC | 22:43 | |
| clarkb | ianw: thanks I can work on a followup change to handle dateext | 22:44 |
| ianw | also like "dnf.rpm.log-20171105" | 22:44 |
| openstackgerrit | Monty Taylor proposed openstack-infra/project-config master: Shift shade IRC activity to openstack-sdks https://review.openstack.org/523556 | 22:45 |
| *** jamesmcarthur has joined #openstack-infra | 22:46 | |
| mordred | fungi: in https://docs.openstack.org/infra/system-config/irc.html#renaming-an-irc-channel .. the topic line has a double-# ... is that intentional? | 22:47 |
| *** ansmith has joined #openstack-infra | 22:47 | |
| *** bobh has joined #openstack-infra | 22:50 | |
| fungi | mordred: your guess is as good as mine | 22:51 |
| fungi | i think ## channels are special in some way (transient?) so might have been copy/paste from when someone was testing out the commands | 22:52 |
| mordred | yah - that's sort of what I was thinking | 22:52 |
| *** bobh has quit IRC | 22:55 | |
| *** marst has quit IRC | 22:56 | |
| ianw | clarkb: heh, i was just looking at 506795 too. should we merge and monitor? my only thought was we could split it up a bit to make it slightly less risky, but if we're babysitting fine | 22:58 |
| jeblair | clarkb: thanks! | 22:59 |
| openstackgerrit | Clark Boylan proposed openstack-infra/puppet-zuul master: Ensure ara is updated on executors https://review.openstack.org/516740 | 22:59 |
| clarkb | lets see if lint and stuff is happier with this iteration | 22:59 |
| *** sboyron has quit IRC | 23:00 | |
| *** rbrndt has quit IRC | 23:01 | |
| clarkb | http://logs.openstack.org/99/522399/3/gate/legacy-openstackci-beaker-ubuntu-trusty/ef58c00/job-output.txt.gz#_2017-11-28_22_50_58_158621 | 23:02 |
| clarkb | at this point is puppet/ruby trolling us? | 23:02 |
| *** ijw has joined #openstack-infra | 23:03 | |
| clarkb | ianw: I think self.re_remove_suffix = re.compile(r'((\.|-)[^a-zA-Z]+)?(\.gz)?$') handles the dateext case? | 23:04 |
| cmurphy | https://github.com/fog/fog-aws/commit/5cb6edaa0a0c828b59b587a371e3c349c15903a6 | 23:04 |
| ianw | clarkb: ruby was removed from bindep fallback just recently. i have no idea if that affects this | 23:04 |
| clarkb | ianw: I don't think it does, its just ruby is fianlly ditching 1.9 support upstream in libs looks like | 23:04 |
| clarkb | we've had to fix a few of these recently | 23:04 |
| clarkb | I'll have a patch up shortly | 23:05 |
| *** rbrndt has joined #openstack-infra | 23:05 | |
| *** jamesmcarthur has quit IRC | 23:05 | |
| *** wolverin_ has quit IRC | 23:05 | |
| *** ijw has quit IRC | 23:08 | |
| ianw | clarkb: regex lgtm, i can't fool it http://paste.openstack.org/show/627648/ | 23:09 |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config master: Remove an extra # from irc renaming instructions https://review.openstack.org/523564 | 23:09 |
| mordred | fungi: ^^ it was extraneous | 23:10 |
| openstackgerrit | Clark Boylan proposed openstack-infra/puppet-openstack_infra_spec_helper master: Cap fog-aws < 2.0.0 for ruby 1.9 support https://review.openstack.org/523565 | 23:10 |
| clarkb | cmurphy: ianw ^ I think that should fix it | 23:10 |
| openstackgerrit | Clark Boylan proposed openstack-infra/puppet-log_processor master: Improve crm114 file collapsing for dateext https://review.openstack.org/523566 | 23:12 |
| fungi | thanks mordred. suspicion confirmed! | 23:12 |
| clarkb | ianw: ^ did it in a followup so that I can recheck the previous change once openstack infra spec helper works again | 23:12 |
| *** lbragstad has quit IRC | 23:17 | |
| *** edmondsw has joined #openstack-infra | 23:18 | |
| jeblair | are we renaming an irc channel? | 23:18 |
| jeblair | mordred: ? | 23:18 |
| fungi | jeblair: thread on the -dev ml, #openstack-shade is being folded into #openstack-sdks | 23:19 |
| *** jklare has quit IRC | 23:19 | |
| jeblair | maybe worth a #status log ? | 23:19 |
| *** calbers has quit IRC | 23:19 | |
| *** bobh has joined #openstack-infra | 23:20 | |
| *** jklare has joined #openstack-infra | 23:21 | |
| *** calbers has joined #openstack-infra | 23:23 | |
| *** edmondsw has quit IRC | 23:23 | |
| *** camunoz has quit IRC | 23:23 | |
| mordred | jeblair: ++ | 23:24 |
| *** lbragstad has joined #openstack-infra | 23:24 | |
| *** bobh has quit IRC | 23:24 | |
| jeblair | sounds like we're planning a rename, not executing one; i misunderstood some comments. | 23:25 |
| *** pbourke has quit IRC | 23:25 | |
| mordred | jeblair: well, it's not completely a rename as much as it's retiring one channel in favor of a different channel (although the renaming docs were the ones used) | 23:26 |
| *** jamesmcarthur has joined #openstack-infra | 23:26 | |
| mordred | #status log #openstack-shade was retired, redirect put in place and users directed to join #openstack-sdks | 23:26 |
| openstackstatus | mordred: finished logging | 23:26 |
| jeblair | oh it *did* happen | 23:26 |
| jeblair | so i was very right to be confused | 23:27 |
| *** pbourke has joined #openstack-infra | 23:27 | |
| jeblair | mordred: thanks for the log. for the rare instances where we have do to something manually instead of by code review, i'd like us to continue the practice of making sure everyone's aware of those actions. | 23:28 |
| fungi | agreed, i find that habit very helpful | 23:30 |
| clarkb | ++ | 23:30 |
| *** ganso has quit IRC | 23:30 | |
| mordred | jeblair: ++ agree | 23:30 |
| mordred | jeblair: I was actually a little weirded out that I couldn't do it via code review :) | 23:30 |
| clarkb | jeblair: when you say that you can use tags to build boards of stories, is the process to change out tags to move things between lanes? | 23:31 |
| jeblair | clarkb: that's how 53 works, yeah | 23:31 |
| *** tosky has quit IRC | 23:31 | |
| openstackgerrit | Merged openstack-infra/puppet-openstack_infra_spec_helper master: Cap fog-aws < 2.0.0 for ruby 1.9 support https://review.openstack.org/523565 | 23:32 |
| fungi | clarkb: basically, automatic worklists and board lanes are built from queries, so you change whatever needs to be changed to alter the query that task or story shows up in | 23:34 |
| *** tosky has joined #openstack-infra | 23:34 | |
| *** bobh has joined #openstack-infra | 23:34 | |
| jeblair | yeah, doesn't have to be just tags, can be more complex | 23:34 |
| fungi | automatic worklists and boards are basically just a fancy query result display | 23:35 |
| clarkb | thanks. Looks like it may also support a click drag and drop workflow maybe? | 23:35 |
| Zara | just the manual ones, not the automatic ones | 23:35 |
| *** Guest888 has joined #openstack-infra | 23:36 | |
| Zara | since the drag and drop wouldn't be governed by rules. we vaguely discussed ways to make it compatible but it seemed tricky | 23:36 |
| jeblair | clarkb: so each lane is either automatic or manual. you can mix the two types of lanes in a board. | 23:36 |
| jeblair | (board 41 is all manual, board 53 is all automatic) | 23:36 |
| *** salv-orl_ has joined #openstack-infra | 23:37 | |
| *** tosky has quit IRC | 23:38 | |
| fungi | cosmetic, but if we can decorate automatic and manual lists/lanes with specific additional style classes we could then add subtle css clues to better differentiate them in the webui i suppose | 23:39 |
| *** bobh has quit IRC | 23:39 | |
| fungi | not sure whether the api queries backing those pages include sufficient metadata to differentiate | 23:40 |
| *** salv-orlando has quit IRC | 23:40 | |
| *** slaweq has quit IRC | 23:40 | |
| Zara | yeah, I've wanted something like that for a while, haven't got around to it yet (also I wasn't sure how to convey the info with the style class; we can make the cards not draggable but people might still wonder why or think it's a bug) | 23:40 |
| Zara | I still need to poke in the logs for the recent auth issues | 23:41 |
| *** slaweq has joined #openstack-infra | 23:41 | |
| fungi | if it's got a fuschia border that's an automatic list, mauve means manual ;) | 23:42 |
| Zara | :D we do have different css for automatic lanes already, iirc it is actually the bakcground colour that changes | 23:43 |
| fungi | oh, fair enough. so subtle i hadn't even noticed! | 23:43 |
| Zara | I'm checking now | 23:44 |
| clarkb | thats interesting I am trying to create a new story and the save chagnes button is greyed out | 23:44 |
| Zara | (I think the idea was that that way you could tell which are which in a mixed board but it only helps if you know what hte options are already) | 23:44 |
| fungi | Zara: confirmed, manual lists have a slightly lighter grey backgruond in our theme, slightly darker means automagic | 23:45 |
| Zara | heh cool I'm not imagining it | 23:45 |
| *** slaweq has quit IRC | 23:45 | |
| fungi | since i never see them side-by-side, i'd never spotted that | 23:45 |
| Zara | I think it's there because it kept hitting me when testing automatic things and I whined to SotK | 23:45 |
| *** thorst has joined #openstack-infra | 23:46 | |
| Zara | rather than it being intended as a foolproof ui to end users | 23:46 |
| clarkb | I seem to be unable to add tasks to openstack-infra/puppet-zanata? | 23:46 |
| Zara | fungi: so there's defo a lot of scope for improving that | 23:46 |
| Zara | clarkb: how was the project selected? | 23:46 |
| clarkb | Zara: I'm using the top bar create new to create a new story | 23:47 |
| clarkb | below the top bar is a search for system-config | 23:47 |
| clarkb | is that context taken into account and I can only add a task against system-config? | 23:47 |
| clarkb | because changing the project to system-config makes the box work | 23:47 |
| clarkb | s/box/button/ | 23:47 |
| Zara | could you get a screenshot? there's a bug where if you type the project name rather than pick it from the dropdown, it doesn't work (or if you pick it from the dropdown and then type it) | 23:48 |
| Zara | but I'm not sure if this is the same thing | 23:48 |
| Zara | you do have to pick a project and make at least one task for it, though the task is automatically populated | 23:48 |
| fungi | clarkb: once i fill in a title and description for the story and some text naming the task, the save changes button is no longer locked out for me | 23:48 |
| fungi | created https://storyboard.openstack.org/#!/story/2001361 with an initial task for openstack-infra/puppet-zanata | 23:50 |
| clarkb | https://imgur.com/a/n1WOI | 23:50 |
| *** thorst has quit IRC | 23:50 | |
| clarkb | Zara: ^ | 23:51 |
| clarkb | fungi: all the fields are populated. I am wondering if its only going to allow me to make a story with a system-config task beacuse that is part of the larger context? | 23:52 |
| *** Guest888 has quit IRC | 23:52 | |
| mordred | clarkb: heh. openstack-tox-pypy job doesn't work - because pypy itself isnt' installed on the node | 23:52 |
| Zara | stories are only related to projects via tasks | 23:52 |
| Zara | that looks like it might be the project issue, could you try deleting the project and reselecting it from the dropdown that appears when you start retyping it? | 23:52 |
| Zara | it shouldn't be preventing you from using that project | 23:53 |
| jeblair | oh yeah there's a quirk of some of these fields -- | 23:53 |
| Zara | yeah :( | 23:53 |
| jeblair | you can't just type it in, you may need to actually select from one of the options that pops up after you type it in | 23:53 |
| clarkb | hrm that seems to fix it | 23:54 |
| clarkb | but you also have to switch the text value | 23:54 |
| clarkb | so if I typed system-config then select it that won't work | 23:54 |
| clarkb | have to go from else to system-config again | 23:54 |
| jeblair | clarkb: agree, i just reproduced the same thing | 23:54 |
| fungi | ahh, yep, i was lazy and selected the project name out of the typeahead drop-down after i got as far as openstack-infra/puppet-z | 23:55 |
| clarkb | if I very carefully do these things it seems to be working | 23:55 |
| fungi | that does seem a smidge fiddly | 23:55 |
| mordred | clarkb, jeblair, pabelanger: I'm not 100% sure the 'right' way to handle this ... we do revoke-sudo in unittests pre ... so the pre playbook of tox-pypy can't install the pypy package | 23:55 |
| Zara | https://storyboard.openstack.org/#!/story/2001272 | 23:56 |
| Zara | is at least one story about it | 23:56 |
| *** lbragstad has quit IRC | 23:56 | |
| clarkb | mordred: isn't the bindep step before revoke-sudo though? | 23:56 |
| clarkb | mordred: as part of a more base job? | 23:56 |
| mordred | and while we could just say "put pypy into your bindep file if you want to use tox-pypy" - that seems weird | 23:57 |
| fungi | mordred: add pypy to bindep.txt, maybe as a custom profile? | 23:57 |
| clarkb | Zara: and now http://paste.openstack.org/show/627650/ after clicking save changes twice | 23:57 |
| jeblair | i agree it's weird with no profile. with a custom profile, won't we run into the same problem? (needing unittests pre or above to perform the install of that profile?) | 23:57 |
| *** rbrndt has quit IRC | 23:58 | |
| clarkb | mordred: fwiw I think we should be putting python3-dev-all in bindep too | 23:58 |
| clarkb | and so on | 23:58 |
| mordred | jeblair: yah - we could have bindep in unittests pre add tox_envlist to the list of bindep profiles if it's there- but that seems like a layer violation | 23:58 |
| fungi | jeblair: today pabelanger fixed the unittests base definition to allow you to pass in a custom profile list | 23:58 |
| Zara | oh yeah it probably tried to create a story with the same attributes as the first one and then failed. it should've made the story now though. | 23:58 |
| Zara | just means there won't be a duplicate | 23:58 |
| fungi | so that the bindep role will use it but still default to the generic "test" profile | 23:58 |
| Zara | yeah https://storyboard.openstack.org/#!/story/2001362 seems to be up there | 23:59 |
| jeblair | fungi, mordred: oh, hrm that may work then? | 23:59 |
| mordred | fungi: so we could maybe have tox-pypy set bindep_profile to ['test', 'pypy'] | 23:59 |
| clarkb | Zara: but it didn't add the puppet-zanata task | 23:59 |
| clarkb | I can add it now, but weird | 23:59 |
| fungi | mordred: yeah, i don't know for sure what the syntax is, may need to just be a space-spearated string of profile names | 23:59 |
| jeblair | clarkb: lesson: click everything, once. :) | 23:59 |
| mordred | kk. I'll go explore that | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!