| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-10-15 | |||
| 08:25:04 | BlackDex | Hello there. How can i force a specific flavor to be only used on specific compute nodes? I have 2 compute nodes which have ssd local storage and i want to have a flavor which will use that local storage so no volumes, and only run on those two hypervisors. So if i select this flavor, it will only start on those two hypervisors. | |
| 08:25:27 | BlackDex | Can i do this, and how? | |
| 08:42:29 | gibi | BlackDex: make sure that filter scheduler configured to use AggregateInstanceExtraSpecsFilter, create a host aggregate, put the two hosts in the aggregate, add the same metadata to the aggregate and the flavor extra_spec. See https://blog.russellbryant.net/2013/05/21/availability-zones-and-host-aggregates-in-openstack-compute-nova/ (ignore the availability zone as that is not needed for you now) | |
| 08:43:04 | BlackDex | gibi: Oke, thx i will check that out :) | |
| 08:44:14 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Consider nested allocations during allocation cleanup https://review.openstack.org/606050 | |
| 09:04:15 | openstackgerrit | Stephen Finucane proposed openstack/nova master: doc: Rewrite the console doc https://review.openstack.org/606148 | |
| 09:04:15 | openstackgerrit | Stephen Finucane proposed openstack/nova master: doc: Add minimal documentation for RDP consoles https://review.openstack.org/606992 | |
| 09:04:16 | openstackgerrit | Stephen Finucane proposed openstack/nova master: doc: Add minimal documentation for MKS consoles https://review.openstack.org/606993 | |
| 09:04:16 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Deprecate the nova-console service https://review.openstack.org/610075 | |
| 09:04:17 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Deprecate the nova-xvpvncproxy service https://review.openstack.org/610076 | |
| 09:13:00 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Consider nested allocations during allocation cleanup https://review.openstack.org/606050 | |
| 09:13:00 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Reject forced move with nested source allocation https://review.openstack.org/605785 | |
| 09:13:01 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Run negative server moving tests with nested RPs https://review.openstack.org/604125 | |
| 09:13:01 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Handle allocations consuming only from the child RPs https://review.openstack.org/608298 | |
| 09:30:59 | BlackDex | An other question. Is it possible to have multiple local storage disks per instance? | |
| 09:33:30 | BlackDex | kind of multiple local ephemeral disks on one instance | |
| 09:45:30 | gibi | BlackDex: in https://developer.openstack.org/api-ref/compute/#create-server look at the block_device_mapping_v2.source_type attribute. You can do a source_type=blank destination_type=local bdm to add additional local sotrage | |
| 09:45:34 | gibi | storage | |
| 09:45:43 | BlackDex | indeed ;) | |
| 10:26:38 | jaypipes | fried_rice: hey, I'll be boarding a plane during the scheduler meeting today, just FYI | |
| 10:38:00 | s10 | Should https://review.openstack.org/#/c/603044/ be backported to the stable/pike? | |
| 13:00:45 | openstackgerrit | Ivaylo Mitev proposed openstack/nova master: VMware: OVA and StrOpt images as VM templates https://review.openstack.org/609736 | |
| 13:04:19 | openstackgerrit | Merged openstack/nova-specs master: Remove force flag from live-migrate and evacuate https://review.openstack.org/609330 | |
| 13:25:35 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add microversion 2.67 to rest api version history https://review.openstack.org/610349 | |
| 13:47:38 | openstackgerrit | Sean McGinnis proposed openstack/nova master: Increment versioning with pbr instruction https://review.openstack.org/610534 | |
| 13:52:57 | efried | n-sch meeting in 7 minutes in #openstack-meeting-alt | |
| 14:02:36 | mriedem | dansmith: i didn't add a unit test to https://review.openstack.org/#/c/610098/ because the functional test covers it, so a unit test would be redundant. i can add one if you think it's necessary. | |
| 14:04:28 | dansmith | mriedem: well, I think the functional is a looong way away from the actual thing and it'd be really hard to connect the test with that change in even a month's time | |
| 14:04:49 | mriedem | alright | |
| 14:04:55 | dansmith | those kinds of special behaviors seem important to unit test if for no other reason than to document that we have specific behaviors here for a reason | |
| 14:13:28 | artom | mriedem, could I ask you to add https://review.openstack.org/#/c/587439/ to your list of volume/live migration things to look at? | |
| 14:14:53 | artom | (Along with https://review.openstack.org/#/c/609517/ - not sure how stringent you're feeling about the splitting-out-the-method thing) | |
| 14:15:17 | mriedem | i would at least split it out in a follow up | |
| 14:15:26 | mriedem | b/c those live migration methods are a gd mess | |
| 14:17:48 | artom | Oh I definitely agree about that second part | |
| 14:18:04 | artom | It's like the Energizer bunny - keeps going and going and going | |
| 14:18:13 | artom | (Disclaimer: I am not sponsored by Energizer) | |
| 14:46:33 | s10 | Should https://review.openstack.org/#/c/603044/ be backported to the stable/pike? | |
| 14:47:25 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Don't persist RequestSpec.requested_destination https://review.openstack.org/610098 | |
| 14:47:36 | efried | s10: Did you add those reviewers on there? | |
| 14:47:48 | efried | You probably meant to add the stable team. | |
| 14:48:02 | efried | https://review.openstack.org/#/admin/groups/540,members | |
| 14:50:03 | openstackgerrit | Corey Bryant proposed openstack/os-vif master: Change python3.5 job to python3.7 job on Stein+ https://review.openstack.org/610068 | |
| 14:50:16 | s10 | efried: I don't remember, I definitely added myself. Thank you. | |
| 14:51:14 | openstackgerrit | Corey Bryant proposed openstack/osc-placement master: Change python3.5 job to python3.7 job on Stein+ https://review.openstack.org/610074 | |
| 14:52:02 | mriedem | when you cherry pick using the gerrit UI it adds the original reviewers | |
| 14:52:03 | mriedem | i think | |
| 14:52:08 | mriedem | i might be thinking of revert | |
| 14:52:24 | efried | yeah, cherry pick doesn't do that, but revert does. | |
| 14:52:26 | mriedem | you definitely don't need to explicitly add the entire stable core team to a change | |
| 14:52:32 | mriedem | ^ is kind of obnoxious | |
| 14:52:58 | dansmith | extremely | |
| 14:52:59 | s10 | efried: probably i did that. nova-stable-maint adds 17 people. | |
| 14:53:26 | s10 | efried: cherry pick adds only the author of patchset. | |
| 15:01:07 | openstackgerrit | Adam Spiers proposed openstack/nova-specs master: Add spec for libvirt driver launching AMD SEV-encrypted instances https://review.openstack.org/609779 | |
| 15:09:33 | aspiers | artom: thanks again for the feedback - I think patch set 5 of this ^^^^ is much better than 4 was | |
| 15:11:49 | mriedem | dansmith: don't forget https://review.openstack.org/#/c/610088/ :) | |
| 15:13:42 | dansmith | mriedem: oh sorry | |
| 15:14:23 | dansmith | thought I had already done that one last week | |
| 15:16:18 | mriedem | i need some nova core reviews for more upgrade checker docs for people working on the community wide goal https://review.openstack.org/#/c/604486/ | |
| 15:16:22 | mriedem | hopefully easy | |
| 15:16:33 | mriedem | at some point these probably move over to the upgrade check library docs | |
| 15:17:45 | stephenfin | mriedem: Docs? I can take a look | |
| 15:22:11 | openstackgerrit | Adam Spiers proposed openstack/nova-specs master: Add spec for libvirt driver launching AMD SEV-encrypted instances https://review.openstack.org/609779 | |
| 15:30:01 | dansmith | mriedem: are you going to fix stephenfin's literal nits/ | |
| 15:30:09 | dansmith | if so I'll wait and hit it when you do that | |
| 15:30:23 | dansmith | or i can fix it for you and then +W if you're in the middle of something else | |
| 15:32:54 | artom | aspiers, you're welcome - I'll read it over again soon (no firm date promises, sorry!) | |
| 15:35:44 | aspiers | artom: thanks! IIUC it would need to be merged by Oct 25 (s-1 milestone) in order to make it in for Stein, right? | |
| 15:36:06 | aspiers | really hoping that's possible - but I know it's tight ... | |
| 15:37:04 | artom | aspiers, it's doable - you basically have to harass the spec-cores ;) | |
| 15:37:13 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add some more docs for upgrade checkers https://review.openstack.org/604486 | |
| 15:37:16 | aspiers | artom: ah OK ;-) and who are those lucky people? | |
| 15:37:16 | mriedem | dansmith: done | |
| 15:37:18 | melwitt | ||
| 15:37:41 | artom | aspiers, https://review.openstack.org/#/admin/groups/302,members | |
| 15:37:56 | aspiers | yeah, I just realised Gerrit would probably tell me ... | |
| 15:38:05 | artom | For the traits stuff jaypipes would be your man, along with efried, though he's not specs-core (yet?) | |
| 15:39:43 | aspiers | OK cool, well jay, dansmith, johnthetubaguy, and mriedem are already on the reviewer list | |
| 15:40:10 | aspiers | I'm not sure whether it would be fair to spam the *whole* specs-core team ;-) | |
| 15:41:45 | artom | aspiers, it's a fine line between persistent and pest ;) | |
| 15:41:54 | aspiers | indeed X-D | |
| 15:42:01 | artom | Honestly though, it's hard | |
| 15:42:01 | efried | artom, aspiers: What are we talking about? | |
| 15:42:14 | aspiers | efried: https://review.openstack.org/#/c/609779/ | |
| 15:42:16 | efried | ... | |
| 15:42:36 | artom | efried, aspiers has a spec that deals with traits, I figures you might want to look | |
| 15:43:55 | efried | Just looking at the title, is this a thing like: | |
| 15:43:55 | efried | - virt driver sets a "I_CAN_HAZ_SEV" trait | |
| 15:43:55 | efried | - op sets trait:I_CAN_HAZ_SEV=required | |
| 15:43:55 | efried | - scheduler schedules to a host that has I_CAN_HAZ_SEV | |
| 15:43:55 | efried | - virt driver sees the trait and flips a switch in the VM | |
| 15:43:55 | efried | ? | |
| 15:44:21 | artom | efried, pretty much | |
| 15:44:28 | aspiers | yes | |
| 15:44:42 | efried | I see mriedem referred you to the HPET thread. If it's a similar deal, then structure it like that --^ and you should be able to get a reasonably fast approval. | |
| 15:45:06 | aspiers | exactly, except the switch to flip is *slightly* involved due to the extra libvirt XML required | |
| 15:45:29 | aspiers | and also it's not as simple as just looking for a flag in /proc/cpuinfo (but it's not that much more complicated either) | |
| 15:46:10 | mriedem | i haven't reviewed the hpet spec fwiw | |
| 15:46:15 | aspiers | efried: yeah, the recent patch sets of my spec already link to that HPET thread | |