| *** jamesmcarthur has quit IRC | 00:46 | |
| *** tosky has quit IRC | 00:52 | |
| *** jamesmcarthur has joined #openstack-tc | 01:01 | |
| *** jamesmcarthur has quit IRC | 01:23 | |
| *** tetsuro has joined #openstack-tc | 01:24 | |
| openstackgerrit | zhongshengping proposed openstack/election master: Add Shengping Zhong candidacy for Puppet OpenStack PTL https://review.opendev.org/715109 | 01:40 |
|---|---|---|
| *** diablo_rojo has quit IRC | 02:01 | |
| *** diablo_rojo has joined #openstack-tc | 03:10 | |
| *** jamesmcarthur has joined #openstack-tc | 03:18 | |
| *** jamesmcarthur has quit IRC | 03:22 | |
| *** jamesmcarthur has joined #openstack-tc | 03:27 | |
| *** jamesmcarthur has quit IRC | 03:33 | |
| *** jamesmcarthur has joined #openstack-tc | 03:34 | |
| *** tetsuro has quit IRC | 03:46 | |
| *** tetsuro has joined #openstack-tc | 03:49 | |
| *** jamesmcarthur has quit IRC | 04:23 | |
| *** jamesmcarthur has joined #openstack-tc | 04:26 | |
| *** jamesmcarthur has quit IRC | 04:34 | |
| *** jamesmcarthur has joined #openstack-tc | 04:34 | |
| *** jamesmcarthur has quit IRC | 04:43 | |
| *** jamesmcarthur has joined #openstack-tc | 04:44 | |
| *** tonyb[m] has quit IRC | 04:49 | |
| *** tonyb[m] has joined #openstack-tc | 04:49 | |
| *** evrardjp has quit IRC | 05:36 | |
| *** evrardjp has joined #openstack-tc | 05:36 | |
| *** diablo_rojo has quit IRC | 05:39 | |
| *** jamesmcarthur has quit IRC | 05:55 | |
| *** jamesmcarthur has joined #openstack-tc | 05:57 | |
| *** jamesmcarthur has quit IRC | 06:02 | |
| *** jamesmcarthur has joined #openstack-tc | 06:07 | |
| *** jamesmcarthur has quit IRC | 06:30 | |
| *** dklyle has quit IRC | 06:53 | |
| *** evrardjp has quit IRC | 07:57 | |
| *** evrardjp has joined #openstack-tc | 08:01 | |
| *** slaweq has joined #openstack-tc | 08:01 | |
| *** dmellado has quit IRC | 08:06 | |
| *** dmellado has joined #openstack-tc | 08:10 | |
| *** rpittau|afk is now known as rpittau | 08:26 | |
| *** e0ne has joined #openstack-tc | 08:51 | |
| *** tosky has joined #openstack-tc | 09:00 | |
| *** e0ne_ has joined #openstack-tc | 09:25 | |
| *** e0ne has quit IRC | 09:25 | |
| *** e0ne_ has quit IRC | 09:50 | |
| *** e0ne has joined #openstack-tc | 09:51 | |
| openstackgerrit | Lucian Petrut proposed openstack/election master: Add Lucian Petrut candidacy for Winstackers https://review.opendev.org/715148 | 10:06 |
| *** tetsuro has quit IRC | 10:34 | |
| *** e0ne has quit IRC | 10:59 | |
| *** e0ne has joined #openstack-tc | 10:59 | |
| *** e0ne has quit IRC | 11:00 | |
| *** e0ne has joined #openstack-tc | 11:00 | |
| *** rpittau is now known as rpittau|bbl | 11:45 | |
| *** e0ne_ has joined #openstack-tc | 11:46 | |
| *** e0ne has quit IRC | 11:47 | |
| *** e0ne has joined #openstack-tc | 11:50 | |
| *** e0ne_ has quit IRC | 11:50 | |
| *** e0ne has quit IRC | 11:53 | |
| *** e0ne has joined #openstack-tc | 11:54 | |
| *** e0ne has quit IRC | 12:14 | |
| *** e0ne has joined #openstack-tc | 12:14 | |
| *** lpetrut has joined #openstack-tc | 12:18 | |
| *** e0ne has quit IRC | 12:28 | |
| *** e0ne has joined #openstack-tc | 12:28 | |
| *** ijolliffe has joined #openstack-tc | 12:34 | |
| *** rpittau|bbl is now known as rpittau | 13:10 | |
| njohnston | o/ | 13:26 |
| openstackgerrit | Ivan Kolodyazhny proposed openstack/election master: Add Ivan Kolodyazhny candidacy for Horizon PTL https://review.opendev.org/715204 | 13:39 |
| *** e0ne_ has joined #openstack-tc | 13:41 | |
| *** e0ne has quit IRC | 13:41 | |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Add Mohammed Naser to OSA https://review.opendev.org/715218 | 14:20 |
| *** corvus has quit IRC | 14:30 | |
| *** corvus has joined #openstack-tc | 14:31 | |
| gmann | o/ | 14:36 |
| *** jamesmcarthur has joined #openstack-tc | 14:51 | |
| *** dklyle has joined #openstack-tc | 14:53 | |
| evrardjp | o/ | 14:58 |
| *** jamesmcarthur has quit IRC | 14:59 | |
| *** jamesmcarthur has joined #openstack-tc | 15:00 | |
| *** jamesmcarthur_ has joined #openstack-tc | 15:02 | |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Add Mohamemd Naser to TC https://review.opendev.org/715237 | 15:03 |
| ttx | o/ | 15:03 |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Drop Python 3.5 jobs https://review.opendev.org/715239 | 15:05 |
| mnaser | election officials, things are failing bc ^ | 15:05 |
| smcginnis | And https://review.opendev.org/715088 | 15:05 |
| mnaser | smcginnis: i have no idea how the py35 jobs didnt run for your change by the way | 15:05 |
| mnaser | i think maybe it ignores doc/source changes | 15:06 |
| jungleboyj | o/ | 15:06 |
| smcginnis | Yeah, looks like it. | 15:06 |
| *** jamesmcarthur has quit IRC | 15:07 | |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Generate title lines based on length https://review.opendev.org/715088 | 15:10 |
| mnaser | AH CRAP | 15:10 |
| mnaser | i mis-rebased | 15:10 |
| smcginnis | That really wasn't needed. :) | 15:11 |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Generate title lines based on length https://review.opendev.org/715088 | 15:11 |
| mnaser | smcginnis: i was going to rebase my own change | 15:11 |
| mnaser | on top of that one | 15:11 |
| openstackgerrit | Mohammed Naser proposed openstack/election master: Drop Python 3.5 jobs https://review.opendev.org/715239 | 15:12 |
| mnaser | that's better | 15:12 |
| *** KeithMnemonic has quit IRC | 15:13 | |
| jungleboyj | mnaser: Question for you. | 15:16 |
| mnaser | hi :) | 15:16 |
| jungleboyj | Trying to help TheJulia get the Baremetal Whitepaper wrapped up. | 15:17 |
| openstackgerrit | Sean McGinnis proposed openstack/election master: Use doc-only requirements https://review.opendev.org/715242 | 15:17 |
| smcginnis | mnaser: And that should make it less of an issue going forward. ^ | 15:17 |
| jungleboyj | https://docs.google.com/document/d/1BmB2JL_oG3lWXId_NXT9KWcBJjqgtnbmixIcNsfGooA/edit# | 15:17 |
| jungleboyj | There is a case study in there from Vexxhost. | 15:17 |
| jungleboyj | Wondering if you would be willing to look at that and maybe improve on it a bit so that it isn't just a copy of what had been previously done. | 15:18 |
| jungleboyj | Would have more weight to it with your input. | 15:18 |
| mnaser | jungleboyj: sure | 15:18 |
| mnaser | any timeline? | 15:18 |
| jungleboyj | We met to discuss things this morning. We were going to go off and work on it over the next week. If you could have input in the next week or two that would be awesome! | 15:19 |
| mnaser | okay cool ill try to have something ready by next week and ping | 15:20 |
| jungleboyj | mnaser: Awesome! Thank you so much! | 15:20 |
| mnaser | so we briefly talked about potentially shipping containers with dockerfile's inside repos | 15:21 |
| mnaser | for those who are curious | 15:22 |
| mnaser | https://review.opendev.org/#/c/713975/ | 15:22 |
| mnaser | that's all we needed to have a fully functional keystone image | 15:22 |
| njohnston | mnaser: How is configuration injected? | 15:24 |
| mnaser | njohnston: that's up to the user/consumer of it to use | 15:24 |
| mnaser | they can mount /etc/keystone | 15:25 |
| mnaser | they can use the oslo.config environment vars feature | 15:25 |
| evrardjp | again, keystone is easy, waiting for nova :) | 15:25 |
| mnaser | nova will be just as easy too. again, we're not solving the "how do i deploy those" but the "here's an image" problem | 15:26 |
| mnaser | i mean if folks want to see nova, i'll gladly do nova next | 15:26 |
| evrardjp | awesome! | 15:27 |
| njohnston | one of the things that people are likely to be interested in is 'I get both a server and an agent from this repo, how do I do that easily and test both' so nova is a good use case for that | 15:28 |
| mnaser | multistage builds and different targets | 15:28 |
| mnaser | we can do what zuul does | 15:28 |
| mnaser | https://opendev.org/zuul/zuul/src/branch/master/Dockerfile | 15:28 |
| mnaser | there's a bit more jazz in there but | 15:29 |
| mnaser | L60-L70 is what we would do for things like nova | 15:29 |
| njohnston | yep, that makes sense to me, just trying to anticipate what the FAQs will be | 15:29 |
| mnaser | yeah im not as much trying to find a way of just doing "docker run openstack/keystone" but instead enabling people to potentially do that | 15:31 |
| openstackgerrit | Merged openstack/election master: Generate title lines based on length https://review.opendev.org/715088 | 15:32 |
| * mnaser feels like our office hour activity has significantly decreased | 15:33 | |
| ttx | mnaser: it usually does before elections | 15:35 |
| ttx | Also I've found the COVID-19 situation to make people less adventurous, so not a great time to propose or discuss crazy changes | 15:36 |
| njohnston | ttx: Very true | 15:37 |
| ttx | I can't blame people to want to have a few certainties in such an uncertain world | 15:38 |
| jroll | I'm not sure if it's less adventurous, or that it's just difficult to think about long-term things in this situation | 15:40 |
| ttx | jroll: yes, probably anythign longterm | 15:46 |
| ttx | which is why I put the brakes on my crazy ideas :) | 15:47 |
| ttx | THEY WILL BE BACK THOUGH | 15:48 |
| evrardjp | just put them in the repo | 15:52 |
| evrardjp | and regularily ping on the ML | 15:52 |
| evrardjp | so that they don't die | 15:52 |
| evrardjp | haha | 15:52 |
| evrardjp | until there is an idea that lands on how to refactor ideas to drop them though | 15:52 |
| * evrardjp has something like that somewhere. | 15:53 | |
| ttx | evrardjp: That reminds me, we discussed reformatting the monthly meeting, a while ago | 15:54 |
| jungleboyj | Sorry, I got distracted. Our church administrator needed help buying a computer so she could work remotely. | 15:55 |
| ttx | I guess that's more a topic for the Victoria chair | 15:55 |
| mnaser | personally, i've come to terms that the whole covid-19 is just reality of how we live these days | 15:56 |
| mnaser | it's uncomfortable having people still need to go to our datacenters but some things must go on | 15:57 |
| jungleboyj | mnaser: The image build for Keystone makes sense. Nice that it isn't that complicated. | 15:57 |
| jungleboyj | mnaser: Are most of your people able to be home though? | 15:57 |
| knikolla | mnaser: will the container images run with a root user? | 15:58 |
| knikolla | also o/ | 15:58 |
| mnaser | jungleboyj: everyone is home, except for folks that need to go into datacenters and do work (which we're trying to minimize as much as we can) | 15:58 |
| jungleboyj | Cool. That is what Lenovo has done as well. | 15:58 |
| mnaser | knikolla: hola. no, the uwsgi base image actually runs as nobody/nogroup | 15:58 |
| mnaser | btw, this is all cool because zuul and it's speculative stuff works for images too | 15:59 |
| mnaser | so i built this thing on an image that hasn't merged yet :) | 15:59 |
| mnaser | see https://review.opendev.org/#/c/713953/ and https://review.opendev.org/#/c/713972/ | 16:00 |
| *** jamesmcarthur_ has quit IRC | 16:00 | |
| knikolla | mnaser: really cool stuff! | 16:00 |
| fungi | behold the power of zuul! | 16:00 |
| mnaser | with something like this, our devstack jobs would be flying | 16:00 |
| mnaser | build image only for project under test, rest is just pulled and quickly setup | 16:01 |
| *** jamesmcarthur has joined #openstack-tc | 16:01 | |
| knikolla | hmmm... if you could cache an image for the services built from master, and use those in integration testing. amazing. | 16:01 |
| fungi | projectS under test | 16:01 |
| mnaser | ^ yes, that :) | 16:01 |
| fungi | that way you get proper testing in dependent pipelines and with depends-on to changes in other repos | 16:01 |
| mnaser | and knikolla - yep -- that's the whole point of it, and then those images are promoted as well | 16:02 |
| fungi | knikolla: right, we've been using that heavily in zuul already | 16:02 |
| fungi | (for zuul's testing of zuul) | 16:02 |
| knikolla | finally some excitement under quarantine days | 16:02 |
| mnaser | i think it might increase the viability of something like devstack | 16:06 |
| mnaser | making it similar to minikube | 16:06 |
| clarkb | mnaser: it should be noted that a significant cost of devstack is configuring the resources with osc | 16:06 |
| clarkb | not the installation | 16:06 |
| clarkb | (this is why I always bring this up) | 16:07 |
| clarkb | its like 50% osc | 16:07 |
| knikolla | also, how are these container images different from kolla built ones | 16:07 |
| mnaser | clarkb: i think this means we probably should explore avenues of configuring things in a much smarter way | 16:07 |
| knikolla | i think tripleo, etc are using the kolla images | 16:07 |
| mnaser | clarkb: say like, when you have containers, keystone can have a "discovery" driver for docker and it would automatically create endpoints that way | 16:07 |
| clarkb | mnaser: yes I agree, that is why I always bring this up. And I've even rewritten a chunk of devstack to use sdk so that tokens can be cached and to have single python process startup time | 16:08 |
| clarkb | mnaser: the feedback was "people won't understand this" | 16:08 |
| clarkb | which is fair | 16:08 |
| clarkb | mnaser: please please please don't do that | 16:08 |
| mnaser | clarkb: happily listening as to why there's 3 pleases' in there for that :P | 16:08 |
| clarkb | I've just discovered that being done by another project and mounting the docker socket into containers seems to be completely backwardds | 16:08 |
| clarkb | mnaser: because the way everyone does that is to give containers the docker socket and yolo | 16:09 |
| fungi | knikolla: apparently the main difference is these are much lighter-weight, and not multi-distro | 16:09 |
| mnaser | knikolla: kolla does things by actually building what is the equivilant of almost a virtual machine | 16:09 |
| clarkb | I went from super interested in a project to completely uninterested in investigating it further after about 30 seconds of reading their docker compose file | 16:09 |
| mnaser | clarkb: interesting, but then we can look at other approaches like maybe introducing some very basic service discovery using etcd (given that its an actual openstack required service and it's deployed with devstack by default too) | 16:09 |
| mnaser | the idea is revisiting the way we've historically done things | 16:10 |
| clarkb | mnaser: sure if you implement it in the services themselves it would be fine | 16:10 |
| clarkb | mnaser: note its not just endpoints though (thats a tiny fraction of the setup) | 16:10 |
| clarkb | its all the tenants and users and networks and routers and subnets and images and ssh keys and flavors and I'm probably forgetting a bunch of it | 16:11 |
| mordred | clarkb: I'm hopefully that in victoria I can fix the osc startup cost | 16:11 |
| clarkb | if people actually want devstack to be fast addressing ^ is where I'd start not container images | 16:11 |
| mordred | I have an idea - I just need to go down the rabbithole | 16:11 |
| knikolla | mordred and his crazy ideas | 16:11 |
| knikolla | :) | 16:12 |
| mordred | yeah - I think container images is orthogonally useful - I don't think the primary benefit is "we can update devstack to use it and stop doing X amount of workk in devstack" | 16:12 |
| mordred | however, if we have images that are zuul-dependency aware with all of the speculative job magic, then it stands to reason that experimenting with re-using those instead of direct pip installs in devstack could also be worthwhile | 16:13 |
| clarkb | ya I think the images can be useful. I don't think they are useful for making devstack faster | 16:13 |
| fungi | but when you have a nice shiny container hammer, every performance problem looks like a nail? ;) | 16:14 |
| mordred | I agree. although I think that if we have them - reworkign devstack to use them _might_ make some amount of things simplier. it might not - it'll definitely be worth exploring | 16:14 |
| clarkb | one thing they will theoretically make easier is grabbing and running code that is tested in the gate | 16:15 |
| mordred | also - it would be a great way to validate that the container content contains the proper contents | 16:15 |
| mordred | yup | 16:15 |
| mordred | that is the reason I think it's a thing we should do | 16:15 |
| clarkb | and that is an excellent reason to do it | 16:15 |
| clarkb | I just don't want people to think it will make devstack take 10 minutes instead of 25 | 16:15 |
| clarkb | fixing osc does ^ | 16:15 |
| mordred | yeah. fixing osc in the next cycle is going to help devstack speed a lot | 16:16 |
| mnaser | that's really useful insight clarkb. i remember that thread back then | 16:16 |
| mnaser | but yeah, i think a "devstack always works" and "some random thing just broke everything" would be reduced | 16:17 |
| knikolla | especially with regards to distro support | 16:19 |
| clarkb | knikolla: though we've recently disovered in infra land that that problem has been muddied a bit with docker standins that aren't quite docker | 16:20 |
| clarkb | overall it should simplify the problem space but it doesn't go away | 16:20 |
| knikolla | you mean podman? | 16:20 |
| clarkb | knikolla: ya | 16:21 |
| fungi | the docker which shall not be named | 16:21 |
| clarkb | turns out podman and docker handle mounts differently. And in a way that affected the infra team and is likely to affect swift and cinder in devstack | 16:21 |
| fungi | red hat: no really, it's docker... | 16:21 |
| clarkb | you can work around it, but you still have distro specific problems | 16:21 |
| fungi | infra: wait, this says "z-ray" | 16:21 |
| fungi | red hat: z is just as good. in fact is better, z is two better than x | 16:22 |
| clarkb | what happened to y? | 16:22 |
| fungi | the last sucker bought the y-ray eyes | 16:22 |
| *** jamesmcarthur has quit IRC | 16:33 | |
| *** jamesmcarthur has joined #openstack-tc | 16:34 | |
| *** jamesmcarthur_ has joined #openstack-tc | 16:36 | |
| *** jamesmcarthur has quit IRC | 16:37 | |
| *** jamesmcarthur_ has quit IRC | 16:46 | |
| *** jamesmcarthur has joined #openstack-tc | 16:47 | |
| *** jamesmcarthur has quit IRC | 16:53 | |
| *** e0ne_ has quit IRC | 17:12 | |
| *** jamesmcarthur has joined #openstack-tc | 17:12 | |
| *** rpittau is now known as rpittau|afk | 17:12 | |
| *** e0ne has joined #openstack-tc | 17:12 | |
| *** lpetrut has quit IRC | 17:21 | |
| *** evrardjp has quit IRC | 17:49 | |
| *** evrardjp has joined #openstack-tc | 17:49 | |
| *** openstack has quit IRC | 17:49 | |
| *** openstack has joined #openstack-tc | 17:51 | |
| *** ChanServ sets mode: +o openstack | 17:51 | |
| *** jamesmcarthur has quit IRC | 18:20 | |
| *** jamesmcarthur has joined #openstack-tc | 18:21 | |
| *** jamesmcarthur has quit IRC | 18:24 | |
| *** jamesmcarthur has joined #openstack-tc | 18:24 | |
| *** jamesmcarthur has quit IRC | 18:26 | |
| *** jamesmcarthur has joined #openstack-tc | 18:27 | |
| *** jamesmcarthur has quit IRC | 18:36 | |
| *** jamesmcarthur has joined #openstack-tc | 18:37 | |
| *** diablo_rojo has joined #openstack-tc | 18:47 | |
| *** jamesmcarthur has quit IRC | 18:50 | |
| *** jamesmcarthur has joined #openstack-tc | 18:50 | |
| *** e0ne has joined #openstack-tc | 19:18 | |
| *** e0ne has quit IRC | 19:39 | |
| *** jamesmcarthur has quit IRC | 20:09 | |
| *** jamesmcarthur has joined #openstack-tc | 20:15 | |
| *** e0ne has joined #openstack-tc | 20:20 | |
| *** e0ne has quit IRC | 20:29 | |
| *** e0ne has joined #openstack-tc | 20:29 | |
| *** e0ne has quit IRC | 20:42 | |
| *** jamesmcarthur has quit IRC | 20:45 | |
| jungleboyj | diablo_rojo: Thanks for starting the virtual mid-cycle etherpad. Good to collect that and hopefully can use some of it as we start brainstorming the PTG. | 20:51 |
| diablo_rojo | jungleboyj, yep, that was my thought. Also just good to have as a general resource for the community though too. | 20:52 |
| diablo_rojo | Happy to help :) | 20:52 |
| jungleboyj | Yeah. I was skeptical of the virtual cinder Mid-Cycles but they have actually worked well. | 20:52 |
| *** ijolliffe has quit IRC | 20:57 | |
| diablo_rojo | jungleboyj, it will never be the same as an in person, but hopefully by collecting all the data we can still make them pretty effective. | 21:01 |
| jungleboyj | Yeah, well, it appears it is something we are going to have to get good at. :-) | 21:02 |
| jungleboyj | The Red Hat Virtual Summit is going to be interesting. | 21:02 |
| *** jamesmcarthur has joined #openstack-tc | 21:08 | |
| mnaser | diablo_rojo: mind checking out https://review.opendev.org/#/c/715239/ and https://review.opendev.org/#/c/715242/1 ? :) | 21:11 |
| diablo_rojo | mnaser, sure! | 21:12 |
| * diablo_rojo opens tabs | 21:13 | |
| fungi | i usually only open one can of tab at a time | 21:13 |
| diablo_rojo | I honestly don't think I have ever consumed tab. | 21:13 |
| smcginnis | Is that still around? | 21:14 |
| smcginnis | Sure enough, I can even get a case here tomorrow from Amazon. | 21:14 |
| fungi | not actually a fan in reality | 21:18 |
| fungi | if i'm going to drink soda, it's usually moxie | 21:18 |
| smcginnis | Me neither. | 21:18 |
| fungi | or reed's extra ginger brew | 21:19 |
| *** jamesmcarthur has quit IRC | 21:30 | |
| *** jamesmcarthur has joined #openstack-tc | 21:33 | |
| *** jamesmcarthur has quit IRC | 21:59 | |
| clarkb | I'm drinking fever tree ginger beer | 22:00 |
| diablo_rojo | Fever Tree is good stuff. | 22:01 |
| diablo_rojo | I usually use it to mix cocktails. | 22:01 |
| *** jamesmcarthur has joined #openstack-tc | 22:01 | |
| jungleboyj | smcginnis: Is Tab considered an essential item? | 22:38 |
| fungi | for time travellers from 1985, sure | 22:39 |
| *** jamesmcarthur has quit IRC | 22:46 | |
| jungleboyj | Oh how I would love to go back to 1985 right now. :-) | 22:47 |
| fungi | you'll need a much bigger pocket if you're going to put your carphone in it | 22:50 |
| *** slaweq has quit IRC | 22:59 | |
| *** jamesmcarthur has joined #openstack-tc | 23:02 | |
| *** slaweq has joined #openstack-tc | 23:10 | |
| *** slaweq has quit IRC | 23:15 | |
| *** jamesmcarthur has quit IRC | 23:32 | |
| *** tetsuro has joined #openstack-tc | 23:33 | |
| *** jamesmcarthur has joined #openstack-tc | 23:33 | |
| *** tetsuro has quit IRC | 23:35 | |
| *** tetsuro has joined #openstack-tc | 23:37 | |
| *** jamesmcarthur has quit IRC | 23:38 | |
| *** tetsuro has quit IRC | 23:47 | |
| *** jamesmcarthur has joined #openstack-tc | 23:53 | |
| *** tosky has quit IRC | 23:53 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!