| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-02-01 | |||
| 14:30:26 | bauzas | you can even modify an existing relnote from a stable branch, which was something not working for Newton :p | |
| 14:30:40 | bauzas | (if you remember the prelude issue we had) | |
| 14:35:21 | edleafe | gibi: Responded to your comments on https://review.openstack.org/#/c/539254/ | |
| 14:37:22 | bauzas | artom: mriedem_afk: oh man, just noticed a long chat on https://review.openstack.org/#/c/539266 | |
| 14:37:38 | bauzas | I need to digest all the comments for working on a new rev | |
| 14:42:04 | gibi | edleafe: thanks. you are right, let's fix just the real warnings first. | |
| 14:42:24 | gibi | edleafe: I guess the same reasoning holds for https://review.openstack.org/#/c/531479 as well | |
| 14:42:38 | openstackgerrit | Radoslav Gerganov proposed openstack/nova master: test_compute_mgr: fix couple of unit tests https://review.openstack.org/539973 | |
| 14:42:49 | edleafe | gibi: yeah, I just saw the comments on the other one first | |
| 14:42:58 | edleafe | gibi: should I file a TODO bug? | |
| 14:43:10 | edleafe | with the caveat that it won't be done until Rocky? | |
| 14:43:20 | gibi | edleafe: yeah, why not. It can be a low hanging fruit for some newcommer | |
| 14:43:46 | edleafe | ok, will do | |
| 14:45:18 | gibi | edleafe: thanks. I +Wd both patches | |
| 14:46:46 | edleafe | gibi: cool. Writing the bug now | |
| 14:47:47 | artom | bauzas, most were just awkward English nits | |
| 14:48:04 | bauzas | 60 comments, man | |
| 14:48:28 | bauzas | my whole Gerrit web page is needing a long scroll :p | |
| 14:56:39 | openstackgerrit | Marcin Juszkiewicz proposed openstack/nova master: Make sure that we have USB Host controller for USB tablet https://review.openstack.org/538003 | |
| 14:56:49 | hrw | pep8 in comments this time | |
| 14:58:26 | mriedem_afk | artom: i hope you're just being extra harsh because it's me and sylvain on https://review.openstack.org/#/c/539266 | |
| 14:58:42 | mriedem | at some point, any docs is better than perfect docs | |
| 14:59:38 | stephenfin | mriedem: I respectfully ``disagree`` :) | |
| 15:00:22 | mriedem | stephenfin: heh you would | |
| 15:00:22 | bauzas | if I can pass a new rev, no problem | |
| 15:00:28 | stephenfin | bauzas: Nice doc. Fancy writing up the real-time doc that I've been meaning to get around to for ages, while you're at it? | |
| 15:00:35 | stephenfin | ;) | |
| 15:00:36 | kashyap | mriedem: That's a good principle: "for documentation, _anything_ (that is not misinfo) is better than nothing" | |
| 15:00:42 | kashyap | But I'm with stephenfin on writing | |
| 15:01:25 | mriedem | https://docs.openstack.org/nova/latest/contributor/documentation.html#review-guidelines | |
| 15:01:42 | kashyap | mriedem: Aside: I hope you saw my comments in the scrollback on your question about: http://logs.openstack.org/66/539266/2/check/build-openstack-sphinx-docs/f7a1712/html/admin/virtual-gpu.html#how-to-know-a-gpu-type | |
| 15:01:47 | kashyap | ? | |
| 15:02:09 | bauzas | mriedem: stephenfin: artom: honestly, I'm fine with comments, given I'm not a English expert | |
| 15:02:24 | bauzas | it's just I don't want to have that change merged after Queens | |
| 15:02:55 | mriedem | kashyap: i just did now - i had just removed the part of the docs about the libvirt api to find types | |
| 15:02:58 | kashyap | bauzas: Yeah, FWIW, I also welcome all nit-picks from native speakers and take no offense. | |
| 15:03:38 | kashyap | mriedem: Cool. The official API is: virNodeDeviceGetXMLDesc() -- have to look up the Python binding name. | |
| 15:03:49 | kashyap | (To use in place of that command in the doc.) | |
| 15:04:11 | mriedem | kashyap: so i guess "Note that the current libvirt API doesn’t provide a way to access that information." was wrong | |
| 15:04:16 | kashyap | Yes | |
| 15:04:58 | edleafe | gibi: Done: https://bugs.launchpad.net/nova/+bug/1746747 | |
| 15:04:59 | openstack | Launchpad bug 1746747 in OpenStack Compute (nova) "Clean up UUID usage in tests" [Low,Confirmed] | |
| 15:05:16 | mriedem | bauzas: maybe you want to see if "virsh nodedev-dumpxml" works for you | |
| 15:05:27 | edleafe | gibi: would be interested in your opinion on a related patch: https://review.openstack.org/#/c/539323/ | |
| 15:06:39 | openstackgerrit | Matt Riedemann proposed openstack/nova master: doc: mark the max microversions for queens https://review.openstack.org/539978 | |
| 15:07:42 | Spazmotic | Was hoping to get more code work done before Korea but looks like that's not going to happen | |
| 15:07:44 | gibi | edleafe: thanks for the bug | |
| 15:07:47 | Spazmotic | Current commits will have to sit for a couple of weeks | |
| 15:07:57 | gibi | edleafe: opened the related patch | |
| 15:12:31 | artom | bauzas, mriedem, English is easier to wrap my head around than code, so nits come faster and easier | |
| 15:12:45 | artom | For what it's worth, group projects back in uni were torture for me | |
| 15:12:48 | bauzas | artom: no worries ;) | |
| 15:12:54 | artom | I had to force myself not to rewrite others prose | |
| 15:13:04 | artom | In my defense, some were terrible writers | |
| 15:13:08 | bauzas | once French language will rule the world, I'll bite all of you, folks | |
| 15:13:24 | artom | Sauf que je parle Français aussi | |
| 15:13:35 | artom | Donc tu vas juste mordre mriedem | |
| 15:13:53 | bauzas | artom: attention, je serai intransigeant sur la grammaire | |
| 15:14:31 | artom | Tu va me lésser aucune chanse? | |
| 15:15:02 | bauzas | quoted for posterity | |
| 15:15:33 | mriedem | hrw: comments inline | |
| 15:15:39 | mriedem | hrw: there is a libvirt version thing you have to deal with | |
| 15:17:13 | mriedem | stephenfin: something to think about for rocky, but we should look at bumping minimum required libvirt/qemu again, haven't done that since ocata i think | |
| 15:17:25 | hrw | 1.2.9? sounds like version from year when i440fx got created | |
| 15:17:47 | stephenfin | Good call. I think kashyap did that investigation last time, but I can pick up the mantle if he doesn't want to | |
| 15:17:54 | stephenfin | I'll add it to the PTG agenda as a reminder | |
| 15:18:05 | mriedem | i just did | |
| 15:18:07 | mriedem | L178 | |
| 15:18:15 | stephenfin | Heh, good stuff :) | |
| 15:18:53 | mriedem | i guess the last bump was pike https://review.openstack.org/#/c/432700/ | |
| 15:19:14 | hrw | mriedem: I checked. ps2mouse (kernel driver) can be compiled on other archs. is it useful? no idea. | |
| 15:20:16 | hrw | mriedem: the idea of 'lets just add usb host' was also based on sahid's comment. | |
| 15:20:34 | mriedem | hrw: the only time that _get_guest_pointer_model returns anything is if we use usbtablet | |
| 15:20:48 | mriedem | so seems if you just moved that into the conditional block "if pointer_model:" it would be good enough | |
| 15:20:53 | hrw | 10:03 <sahid> but pointer model can have different value yes | |
| 15:21:32 | hrw | mriedem: and you already said to me that sahid dealt with pointer_model in past so I assume that he knows what he is saying | |
| 15:22:23 | mriedem | i only see sahid's comments in the patch, not irc | |
| 15:22:25 | mriedem | so https://review.openstack.org/#/c/538003/2/nova/virt/libvirt/driver.py@5036 | |
| 15:23:08 | hrw | mriedem: and that's what I did in 5th patch | |
| 15:24:21 | Spazmotic | Hehe my commit went from working on someone elses with large overhaul.. to my possibility of a XCP version check.. to now just a single line to eat an exception. | |
| 15:24:29 | Spazmotic | The power of code review. | |
| 15:27:10 | Spazmotic | Speaking of code reduction, I resubmitted my commit with the changes you suggested bauzas, if you get a chance sometime this week to take a look. You're still on the review so no rush :) | |
| 15:27:21 | Spazmotic | Have a good night novaers.. time for me to hit the hay | |
| 15:28:49 | hrw | I do wonder how much time it will take when I submit patch which adds usb-keyboard ;D | |
| 15:29:42 | mriedem | mgoddard: johnthetubaguy: how much of this stuff is required for queens? https://review.openstack.org/#/q/topic:bug/1746209+(status:open+OR+status:merged) | |
| 15:29:48 | mriedem | i see the ironic driver traits patches are merged | |
| 15:29:51 | hrw | hm. | |
| 15:30:16 | hrw | mriedem: I think that I will have to rewrite that patch again anyway. | |
| 15:31:31 | hrw | mriedem: I just realized that as I need a keyboard in VM (usb one to make sure it works on all archs) I may need to add a check "if video requested and not x86 then add usb host controller and usb keyboard" | |
| 15:32:10 | hrw | mriedem: and then it does not really matter is it usbtablet or ps2mouse or other mambojambo | |
| 15:33:55 | mgoddard_ | mriedem: none of that is required for queens. https://review.openstack.org/539506 is a speculative bugfix, so might be nice to get that one in | |
| 15:36:05 | jroll | +1 to that | |
| 15:38:28 | gibi | edleafe: see my suggestion inline https://review.openstack.org/#/c/539323/1/nova/cmd/manage.py@115 | |
| 15:39:13 | jroll | mriedem: agree with mgoddard_ there, though they all look fairly safe to me, I'd just land them in queens personally | |
| 15:41:20 | kashyap | mriedem: stephenfin: Back here. About the bumping mi libvirt / QEMU, we can work it out. Should also send out an email to operators list as a heads-up | |
| 15:42:11 | stephenfin | kashyap: I'm assuming we use min(version in Ubuntu 16.04, version in RHEL 7.4, etc.) ? | |
| 15:42:24 | stephenfin | Or is it little more nuanced than that? :) | |
| 15:43:11 | hrw | stephenfin: ubuntu 16.04 or canonical cloud archive for 16.04? | |
| 15:43:22 | kashyap | stephenfin: Not that straightforward :-) | |
| 15:43:46 | kashyap | stephenfin: But yes, the lowest common denominator across distros upstream cares about | |