| *** tetsuro has joined #openstack-placement | 00:30 | |
| *** yikun has joined #openstack-placement | 00:59 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Transform volume.usage notification https://review.openstack.org/580345 | 01:55 |
|---|---|---|
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Transform libvirt.error notification https://review.openstack.org/484851 | 01:55 |
| *** lei-zh has joined #openstack-placement | 02:05 | |
| openstackgerrit | Tsuyoshi Nagata proposed openstack/nova master: Fix nova can handle device name length more widely for universally. https://review.openstack.org/573066 | 02:35 |
| *** lei-zh has quit IRC | 03:36 | |
| *** lei-zh has joined #openstack-placement | 03:39 | |
| *** takashin has quit IRC | 03:48 | |
| *** takashin has joined #openstack-placement | 03:54 | |
| *** lei-zh has quit IRC | 04:37 | |
| *** lei-zh has joined #openstack-placement | 04:56 | |
| *** e0ne has joined #openstack-placement | 05:34 | |
| *** alex_xu has joined #openstack-placement | 05:47 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Adds view builders for keypairs controller https://review.openstack.org/347289 | 06:02 |
| *** e0ne has quit IRC | 06:16 | |
| openstackgerrit | Tsuyoshi Nagata proposed openstack/nova master: [nova] increasing the number of allowed volumes attached per instance > 26 https://review.openstack.org/567472 | 07:22 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (3) https://review.openstack.org/574104 | 08:04 |
| *** tssurya has joined #openstack-placement | 08:30 | |
| *** takashin has left #openstack-placement | 08:30 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: PoC: Support exclusive traits https://review.openstack.org/596669 | 08:44 |
| *** e0ne has joined #openstack-placement | 08:48 | |
| *** ttsiouts has joined #openstack-placement | 08:52 | |
| openstackgerrit | Brin Zhang proposed openstack/nova-specs master: Resource retrieving: add change-before filter https://review.openstack.org/591976 | 08:55 |
| *** cdent has joined #openstack-placement | 08:59 | |
| openstackgerrit | Surya Seetharaman proposed openstack/nova master: Add scatter-gather-single-cell utility https://review.openstack.org/594947 | 09:07 |
| *** lei-zh has quit IRC | 09:53 | |
| *** ttsiouts has quit IRC | 10:23 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova master: hyper-v: autospec classes before they are instantiated https://review.openstack.org/342211 | 10:30 |
| openstackgerrit | Claudiu Belu proposed openstack/nova master: WIP: replace spec with autospec https://review.openstack.org/557299 | 10:46 |
| *** tetsuro has quit IRC | 10:50 | |
| openstackgerrit | Merged openstack/nova master: api: Remove unnecessary default parameter https://review.openstack.org/564451 | 10:55 |
| *** ttsiouts has joined #openstack-placement | 11:02 | |
| *** nicolasbock has joined #openstack-placement | 11:13 | |
| *** tetsuro has joined #openstack-placement | 11:15 | |
| *** cdent has quit IRC | 11:17 | |
| *** cdent has joined #openstack-placement | 11:18 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: PoC: Support exclusive traits https://review.openstack.org/596669 | 11:29 |
| *** tetsuro has quit IRC | 11:30 | |
| *** jroll has quit IRC | 11:43 | |
| *** jroll has joined #openstack-placement | 11:44 | |
| *** jaypipes has joined #openstack-placement | 11:57 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: [nova] increasing the number of allowed volumes attached per instance > 26 https://review.openstack.org/567472 | 12:01 |
| *** sean-k-mooney has quit IRC | 12:08 | |
| *** sean-k-mooney has joined #openstack-placement | 12:08 | |
| jaypipes | efried: so you rebased the reshaper series on top of a new patch for handling that race condition I uncovered that I said could be done in a followup afterwards? | 12:10 |
| jaypipes | giblet_off: good afternoon. I'm actually closer to +0/+1 on https://review.openstack.org/#/c/565730/ but actually left a -1 because it would be nice to include some of efried's response about group_policy in the spec text. If you could add that at your earliest convenience, I'm +2. thanks! | 12:29 |
| *** mriedem has joined #openstack-placement | 12:55 | |
| openstackgerrit | Surya Seetharaman proposed openstack/nova master: Making instance listing skipping down cells configurable https://review.openstack.org/592428 | 12:56 |
| cdent | edleafe, efried : another iteration on tuning the extracted repo, this time as many small commits: https://github.com/EdLeafe/placement/pull/3 | 13:15 |
| *** ttsiouts has quit IRC | 13:16 | |
| edleafe | cdent: Just reading your emails. Yes, I knew that there was more to be done, but wanted to push what I had before leaving on Friday | 13:16 |
| edleafe | cdent: I did note that there were still 905 appearances of the word 'nova' left in the code :) | 13:18 |
| cdent | edleafe: I'm not sure which/what you are responding to | 13:18 |
| edleafe | cdent: "These changes, and others like them, went only halfway, which was perhaps intentional, as the next step is mechanical too, but wanted to point it out" | 13:19 |
| cdent | oh, yeah. that's not a complaint. merely an annotation | 13:19 |
| cdent | edleafe: is this etherpad showing up properly for you? it's gone wonky for me: https://etherpad.openstack.org/p/placement-extract-stein | 13:23 |
| edleafe | looks ok to me | 13:23 |
| edleafe | what's wonky-looking on your end? | 13:23 |
| cdent | it's empty | 13:24 |
| cdent | I think there must have been some bad caching locally | 13:24 |
| cdent | except that using a different browser doesn't clear it | 13:25 |
| edleafe | Hmmm... just reloaded and it's empty too | 13:25 |
| edleafe | I copied the text just before reloading | 13:25 |
| cdent | good thinking | 13:25 |
| efried | jaypipes: Not in gerrit, no. I put that patch on top of the reshaper API side (that's how I interpreted where cdent wanted it) but then I did the full-stack rebase locally to get more test confidence. | 13:26 |
| efried | jaypipes: iow the series is forked now, not a straight line. Hope that helps :) | 13:27 |
| openstackgerrit | Chris Dent proposed openstack/nova master: DNM: Make nova tests pass with externalized placement https://review.openstack.org/596291 | 13:27 |
| edleafe | cdent: I can't even type on that page, so I think something's borked on the server side | 13:27 |
| cdent | weeeee. I guess paste the test somewhere else for safe keeping? | 13:28 |
| cdent | efried: I'm not sure I expressed an opinion, but as long as everyone knows what's up... | 13:29 |
| efried | alex_xu: https://review.openstack.org/#/c/590041/ okay to merge now? | 13:29 |
| *** ttsiouts has joined #openstack-placement | 13:30 | |
| efried | cdent: Somewhere you said something like "near the bottom of the series". Apparently not in the review, but somewhere. Anyway, it made sense to me at the time, i.e. no sense putting an API-side fix all the way at the top of the client-side series. | 13:30 |
| jaypipes | efried: ack | 13:31 |
| *** tetsuro has joined #openstack-placement | 13:32 | |
| alex_xu | efried: let me check again... | 13:32 |
| efried | alex_xu: Though the commit message is now off. If you're good with the code, I can fix up the commit message real quick and you can fast-approve. | 13:34 |
| alex_xu | efried: ok, just prepare to read the commit message...thanks for the tip, let me check the code | 13:37 |
| openstackgerrit | Chen proposed openstack/nova master: Update volume-attachment API url in policies https://review.openstack.org/596726 | 13:39 |
| efried | cdent: Oh, it was this comment I was thinking of: https://review.openstack.org/#/c/576927/35/placement-api-ref/source/reshaper.inc@45 | 13:40 |
| efried | cdent: About a different fup, which I also put near the bottom of the stack ("next" to the race fix) | 13:40 |
| cdent | I need to stop using bottom and top when referring to anything to do with stacks of code changes | 13:41 |
| cdent | because my brain thinks in the order that rebase lists things, not gerrit | 13:41 |
| efried | o | 13:41 |
| efried | well, still made sense to me to put the fup near the f'up | 13:42 |
| efried | ^ t-shirt ^ | 13:42 |
| openstackgerrit | Konstantinos Samaras-Tsakiris proposed openstack/nova-specs master: Placement model for passthrough devices https://review.openstack.org/591037 | 13:43 |
| cdent | fups should always go near fups, I agree | 13:43 |
| openstackgerrit | Dan Smith proposed openstack/nova master: Make instance_list perform per-cell batching https://review.openstack.org/593131 | 13:44 |
| openstackgerrit | Dan Smith proposed openstack/nova master: Record cell success/failure/timeout in CrossCellLister https://review.openstack.org/594265 | 13:44 |
| openstackgerrit | Dan Smith proposed openstack/nova master: Optimize global marker re-lookup in multi_cell_list https://review.openstack.org/594577 | 13:44 |
| alex_xu | efried: the code LGTM, you can update the commit message | 13:49 |
| efried | ack, stand by | 13:49 |
| openstackgerrit | Eric Fried proposed openstack/nova master: [placement] split gigantor SQL query, add logging https://review.openstack.org/590041 | 13:50 |
| efried | alex_xu: done | 13:50 |
| efried | nova-scheduler meeting in 10 minutes in #openstack-meeting-alt | 13:50 |
| alex_xu | done | 13:51 |
| *** takashin has joined #openstack-placement | 13:57 | |
| *** takashin has quit IRC | 14:13 | |
| openstackgerrit | Konstantinos Samaras-Tsakiris proposed openstack/nova-specs master: Placement model for passthrough devices https://review.openstack.org/591037 | 14:18 |
| *** ttsiouts has quit IRC | 14:31 | |
| *** takashin has joined #openstack-placement | 14:35 | |
| openstackgerrit | Merged openstack/nova master: List instances from all cells explicitly https://review.openstack.org/593717 | 14:35 |
| *** takashin has quit IRC | 14:47 | |
| *** ttsiouts has joined #openstack-placement | 14:51 | |
| efried | edleafe, cdent: How long would "freeze" be for? The whole release, just until we get working code merged, something else...? | 15:00 |
| edleafe | although cdent is leaving soon | 15:00 |
| efried | yuh, just want to get a rough idea. | 15:01 |
| edleafe | efried: it's a few steps: | 15:01 |
| edleafe | 1) get placement in nova to a stable point | 15:01 |
| mriedem | edleafe: clearly this isn't just updating import paths https://github.com/EdLeafe/placement/commit/e984bef8587009378ea430dd1c12ca3e40a3c901#diff-fe3112e0898fbeb2d34ea9db1d28c8c1R98 | 15:01 |
| edleafe | 2) do the extraction | 15:01 |
| edleafe | 3) set up the new placement repo | 15:02 |
| edleafe | 4) run tests in Nova to verify that new repo is working | 15:02 |
| edleafe | 5) Switch nova to use new repo | 15:02 |
| edleafe | 6) Unfreeze new repo; all dev work goes in there moving forward | 15:03 |
| *** tetsuro has quit IRC | 15:03 | |
| edleafe | 7) At some time (no rush) delete the placement stuff from nova | 15:03 |
| edleafe | eof | 15:03 |
| efried | edleafe: Okay, cool. So the important thing I took out of that is that we don't have to wait for the downstream world to have a plan to cut over, or anything. We just need the new repo to be functional with nova. | 15:04 |
| edleafe | efried: yes | 15:04 |
| edleafe | that's the critical first step | 15:05 |
| efried | edleafe: And since y'all have already done that work and proven it, the only thing standing in our way is process. | 15:05 |
| efried | not technical | 15:05 |
| edleafe | then we will focus on how to update the deployment needs | 15:05 |
| cdent | efried: and my code has (in a testy way) demonstrated that, see my email (the (technical) update thread) | 15:05 |
| efried | yes | 15:05 |
| * cdent really leaves | 15:05 | |
| edleafe | I don't *think* it's technical | 15:05 |
| efried | cdent: enjoy | 15:05 |
| edleafe | But that's what step 4 above is concerned with | 15:06 |
| efried | Right, cool, I just wanted to get a sense of whether it was possible that we could restart placement feature dev in the next month or two. Sounds like that would be possible. | 15:06 |
| efried | versus in six months or longer. | 15:06 |
| edleafe | efried: oh, that would totally suck | 15:07 |
| efried | yah, for sure. | 15:07 |
| edleafe | That's why it was important to do this now, when there isn't placement development pressure | 15:07 |
| efried | ++ | 15:07 |
| efried | so | 15:07 |
| edleafe | When we could freeze for a short period | 15:07 |
| efried | I am deliberately staying out of the github business for a couple of reasons. | 15:07 |
| efried | But one is so that I will have fresh eyes and no burnout going into the mondo gerrit review(s), whatever form those will take. | 15:08 |
| efried | I will be committing to reviewing that stuff pronto. | 15:08 |
| efried | Whereupon you need one other core to do so. Which should somewhat alleviate the danger of "languishing" because of folks who have effectively committed to not prioritizing it. | 15:09 |
| efried | I suspect mriedem might take up the other +2 there. | 15:09 |
| edleafe | efried: or -2 :-P | 15:10 |
| efried | for a shiny nickel | 15:10 |
| * edleafe goes to make more coffee | 15:10 | |
| *** e0ne has quit IRC | 15:19 | |
| mriedem | ML reply sent | 15:33 |
| openstackgerrit | sahid proposed openstack/nova master: hardware: fix memory check usage for small/large pages https://review.openstack.org/532168 | 15:34 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: privsep: Handle ENOENT when checking for direct IO support https://review.openstack.org/596815 | 15:34 |
| *** efried has quit IRC | 15:37 | |
| *** ttsiouts has quit IRC | 15:42 | |
| *** efried has joined #openstack-placement | 15:43 | |
| *** ttsiouts has joined #openstack-placement | 15:43 | |
| *** ttsiouts has quit IRC | 15:47 | |
| *** nicolasbock has quit IRC | 15:49 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: privsep: Handle ENOENT when checking for direct IO support https://review.openstack.org/596815 | 15:57 |
| openstackgerrit | Merged openstack/nova master: Fix create_resource_provider docstring https://review.openstack.org/595453 | 16:01 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: privsep: Handle ENOENT when checking for direct IO support https://review.openstack.org/596815 | 16:05 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Validate huge pages https://review.openstack.org/399653 | 16:08 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: tests: Further simplification of test_numa_servers https://review.openstack.org/596832 | 16:08 |
| stephenfin | efried: Weren't you working on something similar to this? https://review.openstack.org/#/c/591037/5/specs/stein/approved/device-placement-model.rst | 16:08 |
| stephenfin | efried: (namely, managing devices in placement) | 16:09 |
| efried | stephenfin: Yes, that's copy/pasted like 80% from my nova-powervm spec, see the reference from L15. | 16:09 |
| stephenfin | efried: Ah, cool cool | 16:09 |
| efried | at least that's how it started. It's moved on a little bit from there. | 16:10 |
| efried | due to excellent discussions with kosamara and sean-k-mooney | 16:10 |
| stephenfin | Figuring out how this fits in with cyborg sounds like fun | 16:10 |
| *** cdent has quit IRC | 16:11 | |
| efried | Indeed. I'm sort of assuming cyborg integration won't really start in earnest until Train. Though the cyborg folks may object to that. | 16:11 |
| efried | stephenfin: But it should be noted that e.g. the namespacing of providers is in place in anticipation of some of that. So that there can be overlap/transition with minimal conflict. | 16:12 |
| sean-k-mooney | efried: is kosamara still refining that proposal | 16:13 |
| efried | sean-k-mooney: He posted a couple new patch sets earlier today. | 16:13 |
| sean-k-mooney | i have not check in on it in a few days | 16:13 |
| efried | sean-k-mooney: I haven't looked to see whether he's filled in the driver attach/detach sections yet; that was the major pending piece I knew of. | 16:14 |
| sean-k-mooney | efried: Device attachment is still todo | 16:14 |
| efried | k | 16:15 |
| sean-k-mooney | i need to write a spec to handel sriov hotplug for neutron sriov interfaces but im debating if it should be part of that spec or not | 16:16 |
| sean-k-mooney | i think it might want to reference it but not depend on it | 16:16 |
| *** cdent has joined #openstack-placement | 16:16 | |
| jaypipes | efried: "effectively committed to not prioritizing it" is insulting, just FYI. Just because something is not a priority for me doesn't mean I'm actively committing to not seeing it happen. | 16:17 |
| jaypipes | efried: in the same way the generic device management isn't my priority right now, but that doesn't mean I'm actively trying to torpedo it. | 16:18 |
| sean-k-mooney | jaypipes: good to know :) i would like to adress it but havent had time focus on it so its kindof dropped down my prority list too. | 16:20 |
| efried | jaypipes: Sorry I offended you. I recognize you have a lot on your plate, and I've seen it take some time to close on the things you *are* prioritizing. So I read "not a priority" as probably meaning that we'll want to rely on others' reviews if we want to close it relatively quickly. Again, I apologize if I worded that poorly. | 16:21 |
| jaypipes | np | 16:22 |
| efried | cdent, edleafe: Is there anything stopping us from doing the paperwork to create the openstack/placement and openstack/placement-specs repositories at this time? | 16:26 |
| cdent | efried: when you create a repo, it's normal to point it at an existing repo to seed it | 16:27 |
| cdent | honest, I'm not here, really :( | 16:27 |
| cdent | so while we still seem to be unclear what that seed is supposed to be, it is hard to start the paperwork | 16:28 |
| efried | oic, we don't just create it blank and then start uploading seed changes to it? | 16:28 |
| cdent | you can do that too, but if you give it a seed, you can have some measure of history (in this case "nova itself") as the base | 16:29 |
| efried | Okay. | 16:30 |
| efried | Well | 16:30 |
| efried | It seems to me like, whatever we decide to do in terms of getting to the final working seed, | 16:31 |
| efried | we've agreed to do the initial seed with the subset of (410?) files straight copied from nova. | 16:31 |
| cdent | that sounds like that's the case, yes | 16:31 |
| efried | So what I'm thinking we could do is cut that right now, so we can get the governance stuff going, and bootstrap the placement-core team. | 16:31 |
| efried | while we're waiting for the ML explosion^W^Wvotes to come in on how we get from there to working. | 16:32 |
| cdent | is my dinner time and I also feel like I'm not really willing to do more work if the expectation is that people will come along in a few more days to say the strategy isn't right | 16:33 |
| cdent | so let's make sure we have agreement on the strategy, and then we can do something | 16:33 |
| cdent | if that's as soon as tomorrow, super awesome | 16:33 |
| efried | ight | 16:34 |
| edleafe | Yeah, I'm on the same page. I don't want to waste my time if it's going to change again | 16:34 |
| *** tssurya has quit IRC | 16:34 | |
| cdent | bon chance everyone | 16:34 |
| efried | ō/ | 16:35 |
| efried | edleafe: oh, ee, do we want to use storyboard? | 16:38 |
| efried | would be hard to argue not, considering it'll be a new thing. | 16:38 |
| edleafe | That's the latest hotness | 16:38 |
| efried | I have no love for it, doesn't seem "ready" yet, but not sure if we'd get away with starting up a new lp project. | 16:39 |
| sean-k-mooney | has anyone figured out how to search for things in storyboard | 16:39 |
| edleafe | efried: it seems solid enough; it's just a completely different mind set when working with it | 16:39 |
| edleafe | sean-k-mooney: I haven't tried yet :) | 16:40 |
| sean-k-mooney | any time i have tried to use it in the past its been. challanging to say the least | 16:40 |
| efried | sean-k-mooney: I haven't tried, but apparently you're supposed to start typing and then wait. | 16:40 |
| efried | the search results show up in a drop-down, or something. | 16:40 |
| efried | no way to get them on an actual page. | 16:40 |
| efried | That was my understanding from a session I attended in Vancouver. | 16:40 |
| sean-k-mooney | ya that alone would be almost enough for me to never use it | 16:41 |
| stephenfin | sean-k-mooney: But it's AJAX. Web 2.0, maaaaaan | 16:41 |
| sean-k-mooney | ajax is old | 16:41 |
| *** e0ne has joined #openstack-placement | 16:41 | |
| sean-k-mooney | stephenfin: the buzz word you want to be semi hipster is reactive | 16:42 |
| sean-k-mooney | react.js is getting mainstream/old too it came out 5 years ago | 16:43 |
| *** openstackgerrit has quit IRC | 17:04 | |
| edleafe | cdent: efried: Because the placement extraction etherpad is borked, I created a copy. Of course, all coloring was lost. https://etherpad.openstack.org/p/placement-extract-stein-copy | 17:09 |
| efried | edleafe: thank you. | 17:36 |
| mriedem | edleafe: it's not solid | 18:08 |
| mriedem | SB i mean | 18:09 |
| mriedem | try using it for real work for a day and you'll see | 18:09 |
| edleafe | After using launchpad for 14 years, I assume it will take more than a day to adjust my brain to the SB workflow | 18:10 |
| *** openstackgerrit has joined #openstack-placement | 18:16 | |
| openstackgerrit | Merged openstack/nova master: [placement] split gigantor SQL query, add logging https://review.openstack.org/590041 | 18:16 |
| * cdent waves goodnight | 18:28 | |
| *** cdent has quit IRC | 18:28 | |
| mriedem | edleafe: not even that | 18:29 |
| mriedem | it's just not intuitive and broken in many ways | 18:29 |
| mriedem | see me in -tc | 18:29 |
| jroll | .v 22 | 18:43 |
| jroll | oops | 18:43 |
| *** e0ne has quit IRC | 19:27 | |
| *** e0ne has joined #openstack-placement | 19:29 | |
| *** e0ne has quit IRC | 19:44 | |
| *** e0ne has joined #openstack-placement | 20:00 | |
| openstackgerrit | Dmitry Sutyagin proposed openstack/nova-specs master: Allow disabling KSM / mem-merge via extra spec https://review.openstack.org/593197 | 20:26 |
| *** nicolasbock has joined #openstack-placement | 20:30 | |
| *** e0ne has quit IRC | 20:37 | |
| openstackgerrit | Merged openstack/nova master: privsep: Handle ENOENT when checking for direct IO support https://review.openstack.org/596815 | 20:41 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Add contributor guide for upgrade status checks https://review.openstack.org/596902 | 21:22 |
| openstackgerrit | sean mooney proposed openstack/nova master: add regression test for bug 1759420 https://review.openstack.org/587615 | 21:39 |
| openstackgerrit | sean mooney proposed openstack/nova master: libvirt: fix disk_bus handling for root disk https://review.openstack.org/584999 | 21:39 |
| openstack | bug 1759420 in OpenStack Compute (nova) "nova does not correctly support HW_DISK_BUS=sata or usb for kvm/qemu" [Medium,In progress] https://launchpad.net/bugs/1759420 - Assigned to sean mooney (sean-k-mooney) | 21:39 |
| openstackgerrit | Patricia Domingues proposed openstack/nova master: load up the volume drivers by checking architecture https://review.openstack.org/541393 | 21:44 |
| openstackgerrit | Patricia Domingues proposed openstack/nova master: load up the volume drivers by checking architecture https://review.openstack.org/541393 | 21:46 |
| *** mriedem has quit IRC | 21:52 | |
| *** nicolasbock has quit IRC | 22:39 | |
| *** nicolasbock has joined #openstack-placement | 22:44 | |
| *** nicolasbock has quit IRC | 23:07 | |
| *** jaypipes has quit IRC | 23:24 | |
| *** jaypipes has joined #openstack-placement | 23:25 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!