| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-01-03 | |||
| 16:44:33 | stephenfin | So they need to go in in that order or together | |
| 16:44:47 | jaypipes | dims: ugh. I hate when academics invent some new term like "cloudlet" and act like it's an industry-standard term. | |
| 16:44:48 | ralonsoh_ | stephenfin: perfect | |
| 16:44:53 | stephenfin | ralonsoh_: Should wait and see if I've done it properly first too :) | |
| 16:45:05 | ralonsoh_ | stephenfin: I'll wait for the result | |
| 16:45:27 | dims | jaypipes : y once you get past some interesting stuff in there | |
| 16:45:37 | dims | jaypipes : they have to justify grants :) | |
| 16:45:48 | jaypipes | dims: there's some flat-out incorrect stuff in there, too. :) | |
| 16:46:29 | dims | yep. i was focused on how/what they changed in Nova | |
| 16:47:04 | openstackgerrit | Marcin Juszkiewicz proposed openstack/nova master: libvirt: use 'host-passthrough' as default on AArch64 https://review.openstack.org/530965 | |
| 16:47:18 | hrw | I hope I did that right | |
| 16:48:18 | jaypipes | dims: this picture perfectly represents the coherence of most edge-computing literature. | |
| 16:48:21 | jaypipes | dims: https://i.pinimg.com/originals/d7/cc/31/d7cc31df947e74fc98942bcdedb1de31.png | |
| 16:48:53 | dims | LOLOLOL | |
| 16:49:50 | ildikov | LOL | |
| 16:50:00 | ildikov | jaypipes: I thought cloudlet is an older one | |
| 16:50:52 | jaypipes | cloudlet. it's like saying a banana is a yellow orange. | |
| 16:52:10 | ildikov | you mean a lemon? | |
| 16:53:11 | ildikov | to which we Hungarians would say that it's a little yellow, a little sour, but it's ours... :) | |
| 16:53:39 | mriedem | https://www.youtube.com/watch?v=FT7YZQEHHVE | |
| 16:55:31 | jaypipes | ildikov: :) | |
| 16:59:18 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Don't log a warning for InstanceNotFound in detach_interface https://review.openstack.org/530969 | |
| 17:09:33 | openstackgerrit | Marcin Juszkiewicz proposed openstack/nova master: Use UEFI as the default boot for AArch64 https://review.openstack.org/489951 | |
| 17:09:39 | stephenfin | claudiub: Mind taking a look at this? https://review.openstack.org/#/c/487405/ | |
| 17:10:40 | hrw | stephenfin: thanks for review. updated commit message and fixed that ':' I skipped | |
| 17:11:42 | stephenfin | hrw: Sweet. Still need a bug report though :) | |
| 17:11:49 | hrw | stephenfin: can be done ;d | |
| 17:11:52 | stephenfin | i.e. why this is necessary | |
| 17:11:57 | stephenfin | Cheers (y) | |
| 17:12:23 | hrw | https://bugs.launchpad.net/nova/+bug/1740824 already exists... | |
| 17:12:26 | hrw | will add to commit | |
| 17:12:29 | openstack | Launchpad bug 1740824 in OpenStack Compute (nova) "Make UEFI as the default properties for AArch64" [Undecided,New] | |
| 17:13:38 | stephenfin | hrw: Cool. If you were feeling generous, a little more info in that bug would be appreciated. Why is it essential, etc. | |
| 17:13:47 | hrw | sure | |
| 17:15:13 | openstackgerrit | Marcin Juszkiewicz proposed openstack/nova master: Use UEFI as the default boot for AArch64 https://review.openstack.org/489951 | |
| 17:19:03 | hrw | stephenfin: added comment to the bug | |
| 17:19:22 | hrw | stephenfin: please read and tell me is it enough or should I expand it | |
| 17:19:27 | claudiub | stephenfin: sure, will do. and btw, there are plenty of OVS users on Windows. :) We've added OVS vif plugging support in nova some time ago. I think it was Mitaka. | |
| 17:19:47 | claudiub | or newton | |
| 17:19:51 | hrw | Mitaka... my favourite openstack release. | |
| 17:19:57 | hrw | due to uefi support ;) | |
| 17:20:00 | stephenfin | claudiub: Ah, so I wonder how come we're not using os-vif code for that? | |
| 17:20:11 | claudiub | stephenfin: we are | |
| 17:20:40 | claudiub | stephenfin: https://github.com/openstack/nova/blob/master/nova/virt/hyperv/vif.py#L88 | |
| 17:20:44 | stephenfin | claudiub: I'm confused. How does that work with the code that's there. Surely that's the whole point of that patch? | |
| 17:21:34 | stephenfin | Ahhhh... http://git.openstack.org/cgit/openstack/compute-hyperv/tree/compute_hyperv/nova/vif.py | |
| 17:26:27 | stephenfin | hrw: So to summarize, if you don't set that property, the instance doesn't boot? | |
| 17:26:50 | hrw | stephenfin: unless it is setup to boot kernel+initrd which no one does nowadays | |
| 17:27:44 | stephenfin | hrw: Cool. +2d | |
| 17:27:47 | melwitt | uh, what's happened to tempest job logs? can't filter by log level anymore | |
| 17:28:27 | hrw | stephenfin: cool! thanks | |
| 17:29:39 | hrw | now have to find someone else to take a look ;D | |
| 17:29:41 | claudiub | stephenfin: hm, I think the main point of the patch is to make os-vif a bit more OS agnostic. If you look in the vif_plug_ovs.ovs module, linux_net.device_exists is being used, even though that is also executed on Windows | |
| 17:30:04 | claudiub | and it also adds a Windows specific implementation for device_exists, which is nice. | |
| 17:31:39 | openstackgerrit | Claudiu Belu proposed openstack/nova master: hyper-v: Deprecates support for Windows / Hyper-V Server 2012 https://review.openstack.org/530953 | |
| 17:32:16 | hrw | stephenfin: can you look at https://review.openstack.org/#/c/530965/ one too? It is probably wrong way of doing it and I am open for suggestions | |
| 17:39:09 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: Raise MarkerNotFound if BuildRequestList.get_by_filters doesn't find marker https://review.openstack.org/530980 | |
| 17:45:16 | ralonsoh_ | stephenfin: I don't know why it's failing | |
| 17:45:25 | ralonsoh_ | stephenfin: I can't reproduce the error | |
| 17:48:23 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/ocata: Raise MarkerNotFound if BuildRequestList.get_by_filters doesn't find marker https://review.openstack.org/530981 | |
| 17:49:48 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/newton: Raise MarkerNotFound if BuildRequestList.get_by_filters doesn't find marker https://review.openstack.org/530982 | |
| 17:54:06 | mriedem | stephenfin: https://review.openstack.org/#/c/514309/ is fine except it has to be renamed nova-lvm i guess | |
| 17:59:16 | mriedem | mmedvede: are you powerkvm ci? | |
| 17:59:44 | mriedem | yeah; https://review.openstack.org/#/c/530982/ on stable/newton failed but +1ed the change | |
| 17:59:48 | mmedvede | mriedem: I maintain kvm on power CI, yes | |
| 17:59:58 | mriedem | should we just not run pkvm ci on newton changes? | |
| 18:00:05 | mriedem | that branch is nearly eol | |
| 18:00:28 | mriedem | 2018-01-03 17:51:43.289 | error: pathspec 'stable/newton' did not match any file(s) known to git. | |
| 18:01:20 | mmedvede | mriedem: there is actually a fix for that in the pipeline, let me dig it out | |
| 18:01:39 | mmedvede | it affects all non-zuulv3 third-party CIs | |
| 18:03:02 | mmedvede | mriedem: https://review.openstack.org/#/c/523925/ | |
| 18:05:17 | mmedvede | the CI failing but still +1 the change is due to there being only non-voting job in the set | |
| 18:05:24 | mriedem | i feel like i recently made a similar change to devstack in newton | |
| 18:05:47 | mriedem | ah yes https://review.openstack.org/#/c/519758/ | |
| 18:06:11 | mriedem | https://review.openstack.org/#/c/520095/ | |
| 18:06:11 | mriedem | and apparently it was my devstack-gate patch they are referencing | |
| 18:07:54 | mriedem | sdague: you around for devstack-gate +2? https://review.openstack.org/#/c/523925/ | |
| 18:07:59 | mmedvede | d-g is accumulating lots of cruft due to having to support zuulv3/v2 at the same time :( | |
| 18:09:53 | mmedvede | mriedem: I'll watch the patch, if the fix would not land I'll disable our CI on stable/newton | |
| 18:11:44 | mriedem | ok | |
| 18:28:10 | mriedem | cdent: you want to take a crack at a release note for this? https://review.openstack.org/#/c/448034/ | |
| 18:30:37 | cdent | mriedem: ya, thanks, will look, I don't know much about that one (just did the rebase) but should be able to do it up | |
| 18:31:00 | mriedem | stephenfin: seeing a TypeError in testing the websocket proxy series http://logs.openstack.org/50/530950/1/check/legacy-tempest-dsvm-py35/1dffae7/logs/screen-n-novnc.txt.gz?level=TRACE#_Jan_03_16_09_08_830407 | |
| 18:31:13 | mriedem | in the py35 job | |
| 18:31:24 | cdent | will check with rado first | |
| 18:33:29 | mriedem | huzzah, i've done my job https://review.openstack.org/#/c/345398/ | |
| 18:37:43 | melwitt | mriedem: weird, I wonder how the tempest py35 passed on my DNM patch but not with yours on the console proxy with TLS stuff | |
| 18:38:32 | mriedem | melwitt: because you were configuring one specific job | |
| 18:38:35 | mriedem | which doesn't use py35 | |
| 18:38:41 | melwitt | oh, right. derp | |
| 18:55:07 | openstackgerrit | Chris Dent proposed openstack/nova master: VMware: add support for different firmwares https://review.openstack.org/448034 | |
| 18:56:46 | cdent | I had an amusing conversation with someone outside of openstack today about the complex bits of placement and scheduling and he said what many people have said before: maybe there needs to be an openstack for "give me a vm" and a different openstack for "I am a telco" | |
| 18:57:36 | jaypipes | cdent: bingo. | |
| 18:57:59 | cdent | I thought that might draw you out :) | |
| 19:01:06 | melwitt | hah, I like that. openstack Telco Edition(tm) | |
| 19:01:23 | cdent | now including a super upper ontology | |
| 19:07:10 | ildikov | and it's only the 3rd of January... :) | |
| 19:10:58 | hrw | guys: can you look at https://review.openstack.org/#/c/530965/ one? It is probably wrong way of doing it and I am open for suggestions | |
| 19:23:28 | openstackgerrit | Matt Riedemann proposed openstack/nova master: VMware: add support for different firmwares https://review.openstack.org/448034 | |
| 19:25:28 | stvnoyes | mriedem: Hi Matt, I was talking to ildikov and she suggested I could help with https://review.openstack.org/#/c/266605 , the tempest multiattach test. I see you've been working on it recently. Any objections to me taking it on? | |