| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-04-24 | |||
| 06:30:51 | openstackgerrit | Merged openstack/nova master: Record the host info in EventReporter https://review.openstack.org/556746 | |
| 06:31:33 | kashyap | stephenfin: IOW: | |
| 06:31:36 | kashyap | [quote] | |
| 06:31:36 | kashyap | The advertized NEXT_MIN_{LIBVIRT,QEMU} versions during | |
| 06:31:37 | kashyap | "Pike" release were set to: libvirt 1.3.1 and QEMU 2.5.0 -- but they | |
| 06:31:37 | kashyap | weren't actually bumped for "Queens". So let's actually bump them for | |
| 06:31:40 | kashyap | the "Rocky" release. | |
| 06:31:42 | kashyap | [/quote] | |
| 06:31:45 | kashyap | (From the commit message) | |
| 07:10:58 | alex_xu | jaypipes, ericyoung, we miss this case in this thread http://lists.openstack.org/pipermail/openstack-dev/2018-April/129763.html | |
| 07:15:10 | kashyap | stephenfin: And for "Stein", we've picked the NEXT_MIN versions here (this is merged): https://review.openstack.org/#/c/558171/ | |
| 07:53:27 | ygl | anyone here can help me with my issue please | |
| 07:54:01 | ygl | i have created a vm using a heat stack. when I delete that stack, I want a user script to be trigerred automatically in the VM before its deltion | |
| 07:54:16 | ygl | is that possible in openstack as of now ? | |
| 08:06:54 | bauzas | ygl: looks like it's not a Nova feature | |
| 08:07:36 | bauzas | ygl: AFAIK | |
| 08:16:02 | stephenfin | kashyap: Ah, so it was a different commit. That's what caught me out :) | |
| 08:16:18 | kashyap | stephenfin: Maybe I should've mentioned it | |
| 08:16:26 | kashyap | stephenfin: Because it requires discussion on the Operators / Dev list | |
| 08:17:39 | stephenfin | kashyap: Ah, it's fine. +Wd now | |
| 08:18:01 | kashyap | stephenfin: Most excellent. I can build other clean-up patches on top of it | |
| 08:18:04 | kashyap | Thanks | |
| 08:18:21 | stephenfin | kashyap: Cool. Feel free to add me as a reviewer to those when you do | |
| 08:18:38 | openstackgerrit | sahid proposed openstack/nova master: hardware: only take into account small pages https://review.openstack.org/532168 | |
| 08:18:45 | kashyap | Definitely. But before that I'm trying to get this out of the way: https://bugs.launchpad.net/nova/+bug/1766208 | |
| 08:18:45 | openstack | Launchpad bug 1766208 in OpenStack Compute (nova) "Lift the restriction on choices for `cpu_model_extra_flags` config attribute" [Undecided,New] - Assigned to Kashyap Chamarthy (kashyapc) | |
| 08:25:01 | openstackgerrit | sahid proposed openstack/nova master: remove IVS plug/unplug as they're moved to separate plugin https://review.openstack.org/534371 | |
| 09:19:59 | Kevin_Zheng | alex_xu Thanks for the review, but Error instances could also have Ip, so I guess we should not skip cell0? | |
| 09:28:54 | alex_xu | Kevin_Zheng: cell0 means the instance didn't get right scheduling, so they shouldn't have IP I think | |
| 09:29:25 | Kevin_Zheng | let me think | |
| 09:29:32 | alex_xu | even if they go to a cell and fail, the IP should be released? | |
| 09:29:34 | Kevin_Zheng | error instance will still be cell1? | |
| 09:29:46 | alex_xu | Kevin_Zheng: yes | |
| 09:29:55 | Kevin_Zheng | No, Ip will still be on instances records | |
| 09:30:04 | alex_xu | Kevin_Zheng: one of case I'm sure is the pre-created ported | |
| 09:30:55 | Kevin_Zheng | then you should comment erlier :) | |
| 09:30:58 | alex_xu | Kevin_Zheng: if the instance turn to error after booting, they shouldn't be in cell0, and yes they has IP | |
| 09:31:10 | Kevin_Zheng | yeah, I got that | |
| 09:31:14 | alex_xu | Kevin_Zheng: yea, but i'm not -1 :) | |
| 09:32:00 | alex_xu | Kevin_Zheng: if that makes sense, I still think they should be separated patch | |
| 09:40:41 | Kevin_Zheng | alex_xu I will consider it :) | |
| 09:50:59 | openstackgerrit | XiaojueGuan proposed openstack/nova master: change complex if clause to nice or clause https://review.openstack.org/563872 | |
| 09:53:43 | openstackgerrit | Merged openstack/nova master: libvirt: Bump MIN_{LIBVIRT,QEMU}_VERSION for "Rocky" https://review.openstack.org/558783 | |
| 10:17:21 | openstackgerrit | Merged openstack/nova master: Drop MIN_LIBVIRT_SET_ADMIN_PASSWD https://review.openstack.org/562312 | |
| 10:17:28 | openstackgerrit | Merged openstack/nova master: Remove vestigial system_metadata param from info_from_instance() https://review.openstack.org/561724 | |
| 10:56:45 | openstackgerrit | Merged openstack/nova master: Improve performance when list instances with IP filter https://review.openstack.org/539469 | |
| 11:23:33 | openstackgerrit | Naichuan Sun proposed openstack/nova master: xenapi(N-R-P): Add API to support vgpu resource provider create https://review.openstack.org/520313 | |
| 12:08:40 | jaypipes | alex_xu: will respond on the ML thread, ok? | |
| 12:12:05 | kashyap | Goddamnit | |
| 12:12:09 | kashyap | The Gerrit UI | |
| 12:12:33 | kashyap | It _keeps_ jumping around. | |
| 12:13:37 | kashyap | No wonder I still like the blazing fast e-mail workflow. /me whips up the 'ncurses'-based 'Gertty' tool to see if it has gotten a bit less crashy | |
| 12:13:41 | sean-k-mooney | kashyap: matt found a workaround for that. in your preferences set render mode to slow | |
| 12:13:57 | kashyap | sean-k-mooney: Oh really? /me goes to look | |
| 12:13:59 | jaypipes | kashyap: I know. It's infuriating. | |
| 12:14:11 | kashyap | This just makes the throw the laptop out the fucking window | |
| 12:14:19 | jaypipes | sean-k-mooney: I have that set in my prefs and still happens occasionally. | |
| 12:14:28 | cdent | sean-k-mooney, kashyap the slow render mode helps some, but not all of it | |
| 12:14:32 | cdent | jinx | |
| 12:14:47 | sean-k-mooney | jaypipes: ya it seams to depend on gerrit version | |
| 12:15:19 | sean-k-mooney | there is a new gerrit ui based on javascitp on the latest realease called polygerrit | |
| 12:15:24 | kashyap | sean-k-mooney: Okay, so in Diff Preferences --> Render --> "Slow", yes? | |
| 12:15:31 | sean-k-mooney | kashyap: yes | |
| 12:16:07 | kashyap | sean-k-mooney: Any other preferences that you "can't live without"? | |
| 12:16:19 | kashyap | Maybe I'm "missing out" something | |
| 12:16:25 | kashyap | cdent: Noted | |
| 12:16:25 | sean-k-mooney | kashyap: not that come to mind | |
| 12:16:31 | kashyap | Cool; thanks | |
| 12:18:13 | sean-k-mooney | kashyap: if we ever update the openstack gerrit to support polygerrit ui then that should solve this permently but im sure that has its own bugs | |
| 12:18:18 | jaypipes | kashyap: oh, I'm pretty sure you're missing out on lots of things :P | |
| 12:18:50 | kashyap | jaypipes: I just maintain my sanity (from Gerrit) by spending a portion of time on mailing list based projects :P | |
| 12:18:56 | kashyap | sean-k-mooney: What does "polygerrit" do? | |
| 12:19:42 | jaypipes | kashyap: it's gerrit that simply repeats stuff back to you. | |
| 12:19:51 | jaypipes | kashyap: oh wait, no, that's PollyGerrit. | |
| 12:20:20 | sean-k-mooney | kashyap: its a complete reimplementaiton of the entire gerrit ui in plomer.js and html5 so no more java/gwt | |
| 12:20:35 | kashyap | Unrelated "taste" question -- is it bad taste to write a small clarifying comment in a unit test? | |
| 12:20:38 | kashyap | E.g.: | |
| 12:20:40 | kashyap | + # For 'host-model', it is now valid to use 'extra_flags'; | |
| 12:20:43 | kashyap | + # so warning should not be thrown | |
| 12:20:45 | kashyap | + self.assertFalse(mock_warn.called) | |
| 12:20:54 | jaypipes | kashyap: that is good taste, IMHO. | |
| 12:20:55 | kashyap | There are multiple unit tests, but the "asserts" won't speak for themselves, if you're not reading code carefully around | |
| 12:21:19 | kashyap | jaypipes: Thanks. It is subtle, and this helps. So I'll add it in | |
| 12:21:42 | kashyap | sean-k-mooney: My single biggest feature request is: if I can reply from Gerrit mail, that'll solve "everything" for me with Gerrit | |
| 12:21:46 | sean-k-mooney | kashyap: the only time i would consider that bad taste is if it is done instead of splitting a test that is too complex. adding comments to explain clearly what your testing is good | |
| 12:21:58 | kashyap | As long as I don't have to deal with the clickety-clicky, RSI-inducing GUI | |
| 12:22:33 | kashyap | sean-k-mooney: Good point; I did split the test. I'm lifting the restriction we added here: https://bugs.launchpad.net/nova/+bug/1766208 | |
| 12:22:33 | openstack | Launchpad bug 1766208 in OpenStack Compute (nova) "Lift the restriction on choices for `cpu_model_extra_flags` config attribute" [Undecided,New] - Assigned to Kashyap Chamarthy (kashyapc) | |
| 12:22:42 | sean-k-mooney | kashyap: if openstack had an email only interface it would never work on it again. but if email via gerrit work then that would be cool | |
| 12:23:02 | kashyap | In the earlier patch, for 'host-model' it was invalid to specify 'extra_flags'. But now we allow it (as it is valid libvirt form) | |
| 12:23:23 | kashyap | sean-k-mooney: s/"it would"/I would/ ? | |
| 12:23:37 | sean-k-mooney | ah the restiction we added for backporting | |
| 12:24:54 | sean-k-mooney | ya i think being able to extend host-model makes sense. it was just one extra vairible that we did not want to backport originally so i dont think that bug is controverial | |
| 12:25:46 | kashyap | sean-k-mooney: Yeah, now I'm lifting that | |
| 12:26:18 | sean-k-mooney | kashyap: it is technically a feature not a bug but even a specless blueprint seams to heavy for that so bug makes sense | |
| 12:26:18 | kashyap | git diff --cached | |
| 12:26:22 | kashyap | Wrong window | |
| 12:26:30 | kashyap | sean-k-mooney: Yep | |
| 12:26:40 | sean-k-mooney | kashyap: password: :) | |
| 12:26:48 | kashyap | Heh | |