| *** tetsuro has joined #openstack-placement | 00:25 | |
| *** tetsuro has quit IRC | 00:30 | |
| *** tetsuro has joined #openstack-placement | 00:32 | |
| *** tetsuro has quit IRC | 00:35 | |
| *** tetsuro_ has joined #openstack-placement | 00:35 | |
| *** takashin has joined #openstack-placement | 00:38 | |
| *** tetsuro has joined #openstack-placement | 01:20 | |
| *** tetsuro_ has quit IRC | 01:20 | |
| *** mriedem has quit IRC | 01:23 | |
| *** tetsuro has quit IRC | 01:35 | |
| *** tetsuro_ has joined #openstack-placement | 01:35 | |
| *** tetsuro has joined #openstack-placement | 02:01 | |
| *** tetsuro_ has quit IRC | 02:01 | |
| *** tetsuro has quit IRC | 03:22 | |
| *** tetsuro_ has joined #openstack-placement | 03:22 | |
| *** tetsuro has joined #openstack-placement | 03:26 | |
| *** tetsuro_ has quit IRC | 03:26 | |
| *** takashin has quit IRC | 03:28 | |
| *** takashin has joined #openstack-placement | 03:49 | |
| *** tetsuro has quit IRC | 04:05 | |
| *** e0ne has joined #openstack-placement | 06:24 | |
| *** e0ne has quit IRC | 06:35 | |
| *** e0ne has joined #openstack-placement | 07:12 | |
| *** e0ne has quit IRC | 07:16 | |
| *** e0ne has joined #openstack-placement | 07:24 | |
| *** e0ne has quit IRC | 07:32 | |
| *** takashin has quit IRC | 07:35 | |
| *** takashin has joined #openstack-placement | 07:38 | |
| *** e0ne has joined #openstack-placement | 07:42 | |
| *** helenafm has joined #openstack-placement | 08:31 | |
| *** e0ne has quit IRC | 08:35 | |
| *** e0ne has joined #openstack-placement | 08:37 | |
| *** e0ne has quit IRC | 08:38 | |
| *** e0ne has joined #openstack-placement | 08:46 | |
| *** e0ne has quit IRC | 08:48 | |
| *** takashin has left #openstack-placement | 08:52 | |
| *** e0ne has joined #openstack-placement | 08:54 | |
| *** e0ne has quit IRC | 08:56 | |
| *** tetsuro has joined #openstack-placement | 09:00 | |
| *** tetsuro has quit IRC | 09:05 | |
| *** tetsuro has joined #openstack-placement | 09:05 | |
| *** tetsuro has quit IRC | 09:16 | |
| *** tssurya has joined #openstack-placement | 09:17 | |
| *** tetsuro has joined #openstack-placement | 09:17 | |
| *** e0ne has joined #openstack-placement | 09:25 | |
| *** e0ne has quit IRC | 09:26 | |
| *** rubasov has joined #openstack-placement | 09:27 | |
| *** cdent has joined #openstack-placement | 09:35 | |
| *** tetsuro has quit IRC | 09:39 | |
| *** stephenfin_ is now known as stephenfin | 10:01 | |
| *** belmoreira has joined #openstack-placement | 10:12 | |
| *** purplerbot has joined #openstack-placement | 10:41 | |
| *** tetsuro has joined #openstack-placement | 11:33 | |
| *** tetsuro has quit IRC | 11:46 | |
| *** mriedem has joined #openstack-placement | 13:04 | |
| *** openstackgerrit has quit IRC | 13:37 | |
| *** e0ne has joined #openstack-placement | 13:56 | |
| *** mriedem is now known as mriedem_snow | 14:10 | |
| *** e0ne has quit IRC | 14:44 | |
| *** e0ne has joined #openstack-placement | 15:09 | |
| *** e0ne has quit IRC | 15:33 | |
| *** e0ne has joined #openstack-placement | 15:41 | |
| *** mriedem_snow is now known as mriedem | 15:52 | |
| cdent | placement folk (jaypipes, efried, edleafe, mriedem, gibi, tetsuro, others): When we do finally have placement out on its own, do people have any strong feelings about having a separate specs repo? I kinda think that "in the main docs" is actually mo-beta | 16:35 |
|---|---|---|
| jaypipes | cdent: ++ for not having a separate specs repo | 16:35 |
| efried | tbc, propose specs in openstack/placement under a specs/ subdirectory rather than creating an openstack/placement-specs project similar to nova-specs ? | 16:36 |
| efried | cdent: ^ | 16:37 |
| cdent | efried: pretty much, yes | 16:37 |
| efried | Meaning, among other things, that there's not a separate "spec core" group? | 16:37 |
| cdent | yes | 16:37 |
| efried | ++ | 16:37 |
| mriedem | wfm | 16:37 |
| cdent | gibi, who is allowed to veto anything because he finds all the bugs, will scupper this plan for us | 16:38 |
| gibi | specs/ in the openstack/placement repo works for me too | 16:39 |
| gibi | :) | 16:40 |
| *** helenafm has quit IRC | 16:40 | |
| cdent | great | 16:40 |
| edleafe | Yeah, I can't think of a good reason for a separate repo | 16:41 |
| *** e0ne has quit IRC | 17:27 | |
| *** gryf has quit IRC | 17:34 | |
| *** e0ne has joined #openstack-placement | 17:43 | |
| *** mriedem is now known as mriedem_afk | 17:45 | |
| *** tssurya has quit IRC | 18:09 | |
| *** gryf has joined #openstack-placement | 18:35 | |
| *** alex_xu has quit IRC | 19:04 | |
| *** stephenfin has quit IRC | 19:04 | |
| *** jroll has quit IRC | 19:04 | |
| *** belmoreira has quit IRC | 19:04 | |
| *** stephenfin has joined #openstack-placement | 19:06 | |
| *** jroll has joined #openstack-placement | 19:06 | |
| *** alex_xu has joined #openstack-placement | 19:06 | |
| *** mriedem_afk is now known as mriedem | 20:05 | |
| *** e0ne has quit IRC | 20:25 | |
| *** e0ne has joined #openstack-placement | 20:48 | |
| *** e0ne has quit IRC | 21:28 | |
| *** krasmussen has joined #openstack-placement | 21:33 | |
| krasmussen | Is the ip capacity supposed to be accurate in placement in Pike? It seems to be WAY off in my cloud. | 21:35 |
| krasmussen | ^ Just wondering before I go down a rabbit hole. | 21:35 |
| krasmussen | I guess I should be saying ip "inventory" opposed to capacity. | 21:35 |
| efried | krasmussen: "IP"? | 21:46 |
| cdent | krasmussen: which resource class of inventory are you looking at, and how are you counting capacity? | 21:46 |
| krasmussen | Example call: curl -s -H "Content-type: application/json" -H "X-Auth-Token: $TOKEN" -H "Openstack-Api-Version: placement 1.1" -X GET https://${host}:${port}/resource_providers/${resource}/inventories | 21:47 |
| krasmussen | where resource is the id for the `Neutron segment id ` named resources | 21:48 |
| cdent | krasmussen: as far as I'm aware, in pike neutron is only reporting segments, not doing anything to consume them | 21:48 |
| krasmussen | It does not match even close to the subnet availability from a `openstack ip availability show ${network}` | 21:49 |
| krasmussen | Ahh well that makes sense. Whats weird is it shows a few as "reserved" and I wasn't sure if that was consumed or reserved so it can't be consumed. | 21:50 |
| cdent | when looking at inventories 'reserved' means "a count of stuff that is set aside for use by the system, not for workloads" | 21:50 |
| krasmussen | Sweet thanks :) | 21:51 |
| cdent | there's stuff in progress this cycle (stein) for using placement + neutron to do minimum bandwidth handling, but I'm not aware of any segment management, yet | 21:51 |
| krasmussen | OK, thats what I was looking to implement as rumor had it there may have been some kind of mapping between HV and segments within placement and I was digging around. | 21:52 |
| krasmussen | Appreciate the info guys :) | 21:53 |
| cdent | if you come back and implement some, it will be worth it ;) | 21:53 |
| * efried didn't even know neutron was reporting resources to placement in pike | 21:55 | |
| cdent | neuron had a formal client for placement before nova did | 21:58 |
| krasmussen | Yeah, our old Liberty cloud we had a bunch of local patches to handle ip subnet capacity / availability aware scheduling and we moved to pike and have been trying to avoid as many local patches as possible so I was in the nova channel the other day looking for opinions on a BP for a scheduler filter that did this and was told that it should be done with placement instead and some of the work may already | 21:59 |
| krasmussen | be done for host -> segment mapping hense why I was asking :P I'm leaning towards a local patch for the short term but hopefully will be able to help implement something later. | 21:59 |
| cdent | talking to and using placement is really easy. If it already has the info you want in it, it might give you a huge boost | 22:02 |
| * cdent tries to think of some examples | 22:02 | |
| cdent | All the stuff I've got is written as gabbi tests: | 22:02 |
| cdent | https://github.com/cdent/placecat | 22:02 |
| cdent | https://github.com/cdent/placement-disk-models | 22:02 |
| cdent | https://github.com/cdent/etcd-compute | 22:03 |
| cdent | efried: do you have a link to you and edleafe's presentation? | 22:04 |
| efried | ... | 22:04 |
| cdent | i have it in a blog post somewhere, but there are so many blog posts... | 22:04 |
| efried | https://www.openstack.org/videos/summits/vancouver-2018/placement-present-and-future-in-nova-and-beyond | 22:05 |
| cdent | there you krasmussen more than you ever really wanted | 22:05 |
| cdent | https://anticdent.org/tag/placement.html | 22:05 |
| krasmussen | I actually watched that video earlier today. | 22:06 |
| efried | woot. I'm the dude without the beard. edleafe is the garden gnome. cdent is the mop of black hair in the front row who gets credit for the actual content. | 22:10 |
| cdent | damn, now I will never be clear of that garden gnome image | 22:11 |
| efried | oh, come on, I can't be the only one who sees it. | 22:12 |
| krasmussen | I was at that summit too but never bothered to see if I was in any of the videos to point myself out to you guys. | 22:12 |
| cdent | I guess I just never thought of it. Ed looks like Ed. | 22:13 |
| efried | Do you get Buddy Holly ever? | 22:16 |
| efried | I think I just have a weird perspective. When people remind me of people, it's less because there's a dramatic physical resemblance than because there's an overall impression, or a particular feature. For example, I can't dissociate mriedem from neil patrick harris in my head. | 22:19 |
| cdent | i can see that | 22:19 |
| cdent | and I can see the buddy holly thing too | 22:19 |
| efried | side by side they really look nothing alike. But there's a certain Doogie-ness about mriedem | 22:19 |
| cdent | yeah | 22:19 |
| mriedem | i'm kevin bacon mfers | 22:19 |
| mriedem | not doogie howser | 22:20 |
| efried | yeah, you keep saying that. | 22:20 |
| mriedem | i will cut you | 22:20 |
| cdent | I do that kind of thing too, but only if I spend a lot of physical time with people. And since I avoid video... | 22:20 |
| efried | interesting bit of trivia: I have an extremely low bacon number. | 22:20 |
| efried | Possibly as low as two at this point. | 22:20 |
| efried | okay, it's interesting to me. | 22:21 |
| * cdent is still thinking about matt doogie or matt bacon | 22:22 | |
| edleafe | efried: you want garden gnome? I'll give ya garden gnome: https://photos.app.goo.gl/YJf8wtXnAVkAcabu6 | 22:23 |
| efried | edleafe: THANK you. | 22:24 |
| cdent | the brace of pheasants make that | 22:25 |
| edleafe | That was in a garden of a 500-year-old house outside Sudbury, England | 22:27 |
| edleafe | Can't get any more garden-gnomey than that | 22:27 |
| efried | Yes, I can get to Bacon in one hop via Billy Bob Thornton, woot. | 22:27 |
| cdent | you and bbt roll? | 22:29 |
| cdent | (seems like the best version of the story I can construct) | 22:29 |
| efried | I was an extra in Parkland | 22:31 |
| efried | You can even see me in the preview https://www.youtube.com/watch?v=V5-Bg7ZrGAg around 1:40, top right, cowboy hat, unmistakeable chin. | 22:31 |
| efried | before that I still had a pretty good bacon number. I was in a TV series as a kid with Naomie Harris (the latest Bond Moneypenny). | 22:32 |
| cdent | all because of the chin, presumably? | 22:34 |
| efried | Well, the Parkland gig for sure because of the chin. | 22:35 |
| efried | It's a pretty good story, remind me to tell it over dinner some day. | 22:35 |
| cdent | you must have had a very good shave that day | 22:35 |
| cdent | sounds like an excellent plan | 22:35 |
| cdent | it's a good note to end my day on (it's 22:30) | 22:36 |
| * cdent bows out | 22:36 | |
| efried | o/ | 22:36 |
| *** cdent has quit IRC | 22:36 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!