| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-07-23 | |||
| 14:17:06 | stephenfin | (y) | |
| 14:18:15 | stephenfin | In that case, I've two references to config options in that second bit that should probably do with the ':oslo.config:option:' role, but as a nice-to-have | |
| 14:19:05 | dansmith | can you barf up the rst for me to link to that schedulers section? | |
| 14:23:51 | dansmith | hmm, maybe i just use the full link? I thought there was a shortcut | |
| 14:24:38 | stephenfin | To the section I added? | |
| 14:25:35 | stephenfin | You need to add an anchor before the title and then you can use the :ref: role, assuming it's within the same doc build | |
| 14:25:56 | stephenfin | so add e.g. '.. _hypervisor-specific-considerations:' | |
| 14:26:13 | stephenfin | then reference wherever you want with ':ref:`hypervisor-specific-considerations`' | |
| 14:27:01 | stephenfin | or https://docs.openstack.org/nova/latest/admin/configuration/schedulers.html#hypervisor-specific-considerations if it's from a reno or something | |
| 14:27:05 | dansmith | ack was just piecing that together from mah grepping | |
| 14:27:09 | dansmith | I have to re-learn this every damn time | |
| 14:28:32 | dansmith | stephenfin: er, I don't see any other :ref: usage in the renos.. are renos in the same build as the other docs? | |
| 14:28:52 | stephenfin | You've to use a URL there | |
| 14:28:59 | openstackgerrit | Eric Fried proposed openstack/nova master: DNM: experiment with GreenThread passthrough https://review.opendev.org/672299 | |
| 14:29:01 | stephenfin | because they're a different build | |
| 14:29:07 | dansmith | right okay, but the anchor makes it #able yeah? | |
| 14:29:11 | stephenfin | Yup | |
| 14:29:24 | dansmith | stephenfin: aren't you glad I'm saving you time by doing this myself? :P | |
| 14:29:38 | stephenfin | Very :) | |
| 14:29:43 | dansmith | haha | |
| 14:29:50 | stephenfin | Here's another example, fwiw https://docs.openstack.org/nova/latest/admin/configuration/schedulers.html#allocation-ratios | |
| 14:29:58 | stephenfin | That should have an anchor in the rST | |
| 14:30:10 | dansmith | that's the one I was copying yeah | |
| 14:30:51 | stephenfin | though interestingly that has no anchor | |
| 14:31:15 | dansmith | it does | |
| 14:31:17 | dansmith | oh no, | |
| 14:31:22 | dansmith | that one was auto from the heading I think | |
| 14:31:25 | dansmith | I found another one though | |
| 14:31:39 | dansmith | capabilities-as-traits is what I copied | |
| 14:31:49 | stephenfin | Ah yeah, the auto from the heading behavior isn't the default, I thought | |
| 14:31:57 | stephenfin | Might be a thing we do in openstackdocstheme | |
| 14:32:09 | stephenfin | Or not, idk. Doesn't matter. capabilities-as-traits is good | |
| 14:34:54 | openstackgerrit | Eric Fried proposed openstack/nova master: Use OpenStack SDK for placement https://review.opendev.org/656023 | |
| 14:37:09 | sean-k-mooney1 | stephenfin: the headings are automaticaly referenceable via the css selector | |
| 14:37:23 | sean-k-mooney1 | its thte content list or what ever we use that arnt | |
| 14:39:08 | dansmith | stephenfin: I'm going to address my other nit I just commented as well | |
| 14:42:06 | dansmith | and your comment on the config option | |
| 14:43:04 | stephenfin | coolness | |
| 14:48:54 | aspiers | sean-k-mooney: I'm about to review your https://review.opendev.org/#/c/670189/ - thanks for working on that | |
| 14:49:52 | openstackgerrit | Dan Smith proposed openstack/nova master: Remove deprecated CPU, RAM, disk claiming in resource tracker https://review.opendev.org/551026 | |
| 14:49:52 | openstackgerrit | Dan Smith proposed openstack/nova master: Remove deprecated Core/Ram/DiskFilter https://review.opendev.org/672065 | |
| 14:49:53 | openstackgerrit | Dan Smith proposed openstack/nova master: Remove 'nova.virt.driver.ComputeDriver.estimate_instance_overhead' https://review.opendev.org/672106 | |
| 14:50:59 | sean-k-mooney | aspiers: nice to see you around again. and i need it for my own stuff too so its not entily un selfish :) | |
| 14:56:27 | aspiers | sean-k-mooney: :) | |
| 14:56:44 | aspiers | sean-k-mooney: I am not sure why I thought that defaulting to q35 would be OK :-/ | |
| 15:02:17 | mriedem | ralonsoh: jangutter: efried: stephenfin: i'm late but have some concerns in https://review.opendev.org/#/c/641670/ that should be addressed | |
| 15:03:04 | efried | mriedem: should I unapprove the bp? | |
| 15:03:19 | mriedem | efried: probably not right now | |
| 15:03:23 | efried | k | |
| 15:03:58 | efried | mriedem: if you're in a spec reviewing mood, this one could use some love https://review.opendev.org/#/c/608696/ | |
| 15:04:18 | ralonsoh | mriedem, I'll check your comments today | |
| 15:05:22 | mriedem | efried: i'm not really | |
| 15:05:27 | mriedem | i just saw that one was merged and short | |
| 15:10:31 | openstackgerrit | sean mooney proposed openstack/nova master: libvirt: delegate ovs plug to os-vif https://review.opendev.org/602432 | |
| 15:14:31 | stephenfin | efried: It looks like we've gone from zero to two implementations of the cpu-resources spec in less than a week \o/ | |
| 15:15:14 | stephenfin | I'm working on comparing and likely merging the two, but I'd like your thoughts on this particular difference of implementation when you can https://review.opendev.org/#/c/672223/ and https://review.opendev.org/#/c/671801/ | |
| 15:16:03 | stephenfin | They're the two patches to convert legacy 'hw:cpu_(thread_)policy' extra specs to PCPU requests and they've taken very different approaches. I can't decide which one I prefer | |
| 15:20:33 | stephenfin | TIL '(x or y) in foo' is a valid Python statement | |
| 15:30:19 | stephenfin | Actually, it does (thank ralonsoh) | |
| 15:30:23 | stephenfin | >>> (0 or None) in [0] | |
| 15:30:24 | stephenfin | False | |
| 15:33:26 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Convert nova-next to a zuul v3 job https://review.opendev.org/670196 | |
| 15:38:33 | mriedem | is there a follow up needed for surya's spec here https://review.opendev.org/#/c/636132/ ? before digging into the code series i was going to go over the spec again but don't really want to if it's a mess that needs an update. | |
| 15:44:21 | stephenfin | sean-k-mooney: Question: do we want to allow users to request 'resources:VCPU' or 'resources:PCPU' in their flavor extra specs? | |
| 15:44:50 | stephenfin | as opposed to 'hw:cpu_policy=dedicated' and whatever the new mask option to allow instances with both types of CPU ends up being | |
| 15:44:59 | stephenfin | or in addition to, I should say | |
| 15:46:33 | openstackgerrit | Merged openstack/nova master: Translatable output strings in heal allocation https://review.opendev.org/668925 | |
| 15:59:21 | efried | stephenfin: you may be expecting too much of me wrt those PCPU impls, but I'll do what I can. | |
| 15:59:42 | stephenfin | That's all I can ask (y) | |
| 16:00:08 | stephenfin | They both "work", I'm more interested in which approach is preferable | |
| 16:01:13 | efried | but we should definitely *support* (and encourage as the primary usage) flavor-ese lingo translated by request_filter to placement-ese. | |
| 16:01:13 | efried | stephenfin: also, to answer "do we want to allow users to request 'resources:VCPU' or 'resources:PCPU' in their flavor extra specs?" -- I think we need to *allow* it, because downstream thingies are going to want to override the meanings of the RCs for things like frequency/priority (per ptg direction) -- | |
| 16:02:00 | stephenfin | Aye, so allow it unless the 'hw:cpu_policy' extra spec/image meta prop is present | |
| 16:02:05 | stephenfin | In which case we spit the dummy out? | |
| 16:18:09 | mriedem | efried: dansmith: melwitt: you were all +2 on the spec for https://review.opendev.org/#/c/645611/ so i might be off base on my comments, | |
| 16:18:19 | mriedem | nothing major but i'm not sure why the driver even needs to implement this | |
| 16:18:54 | dansmith | mriedem: what do you mean? | |
| 16:19:12 | dansmith | why we're plumbing this to the driver instead of just telling the compute about the change? | |
| 16:20:29 | mriedem | all the virt driver does is set the values on the instance and save them to the db, | |
| 16:20:38 | mriedem | which is what stop/start_instance in the compute manager does for normal stop/start, | |
| 16:20:52 | mriedem | so why not just do the db updates in the compute manager code rather than muck with driver plumbing | |
| 16:20:56 | mriedem | just seems weird to me, | |
| 16:21:08 | mriedem | if it's "we don't want non-ironic drivers to do this" then add a driver capability flag | |
| 16:21:12 | dansmith | mriedem: compute start/stop will call the driver and tell it what to do right? | |
| 16:21:35 | mriedem | to actually power on / off a guest yeah | |
| 16:21:37 | mriedem | this is just db muckery | |
| 16:21:45 | dansmith | the point of this is to decide what needs to happen when some external change happens to the instance, just like how libvirt gets direct events | |
| 16:22:00 | dansmith | the point being, ironic needs to know this should be swallowed and the db updated, | |
| 16:22:03 | mriedem | none of that logic is in the actual driver | |
| 16:22:10 | dansmith | but that's not necessarily what should happen for vmware | |
| 16:22:29 | openstackgerrit | Merged openstack/nova master: Update api-ref location https://review.opendev.org/672077 | |
| 16:22:39 | mriedem | ok i guess "it's just a simple abstraction" is the answer then | |
| 16:22:57 | dansmith | remember the first version of this *did* just call start/top, | |
| 16:23:00 | mriedem | i don't expect any other drivers to ever implement this, but i guess it's generic enough that they could | |
| 16:23:08 | dansmith | but had a "don't actually call the virt driver=True" if we're ironic | |
| 16:23:28 | dansmith | so the change to delgate "what should I do with this external change?" to the driver was to make that cleaner | |
| 16:23:33 | mriedem | yeah i agree that was gross | |
| 16:23:46 | dansmith | didn't we discuss all this during spec review? | |
| 16:23:50 | dansmith | I thought you were involved with that | |
| 16:23:54 | mriedem | not really | |