| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-10-24 | |||
| 22:04:14 | efried | melwitt: yeah, something like that | |
| 22:05:07 | efried | where realistically based on past experience I don't expect 2 or 4 or 5 to happen, and anything discussed in 3 will get completely reversed as soon as certain people are paying attention. | |
| 22:05:21 | efried | which is why I haven't been super motivated to start the process. | |
| 22:05:38 | efried | also, efried doesn't have to be the one to initiate 1 | |
| 22:05:46 | efried | so if you have something you want to air, I encourage you to do that. | |
| 22:09:23 | melwitt | efried: understood, thanks. I'll let my team know and encourage them to start ML threads if they want to kick off some discussion | |
| 22:10:43 | efried | Cool. I guess it wouldn't hurt for me to send a primer email to that effect as well. | |
| 22:11:08 | mriedem | i promise i won't certain people it up | |
| 22:11:40 | mriedem | my reply to all post-ptg decisions will be "lol lgtm!" | |
| 22:20:58 | openstackgerrit | Merged openstack/nova stable/train: Revert "vif: Resolve a TODO and update another" https://review.opendev.org/690126 | |
| 22:21:04 | openstackgerrit | Merged openstack/nova stable/train: Add functional regression test for bug 1849409 https://review.opendev.org/690710 | |
| 22:21:04 | openstack | bug 1849409 in OpenStack Compute (nova) train "openstack server list --deleted --limit -1 hangs" [High,In progress] https://launchpad.net/bugs/1849409 - Assigned to Matt Riedemann (mriedem) | |
| 22:29:02 | efried | melwitt: http://lists.openstack.org/pipermail/openstack-discuss/2019-October/010363.html | |
| 22:29:39 | efried | mriedem: a naked +1 will suffice. | |
| 22:29:51 | melwitt | efried: cool, thank you | |
| 22:33:37 | mriedem | naked what now? | |
| 22:35:18 | mriedem | melwitt: speaking of your teammates, lee would probably be grateful if you took a look at this simple bug fix https://review.opendev.org/#/c/682622/ | |
| 22:35:51 | melwitt | mriedem: already reviewing it brah! | |
| 22:36:35 | mriedem | i do respect you bro | |
| 22:36:45 | melwitt | I'm glad | |
| 22:37:36 | mriedem | that was a reference to the bro down that happened in here the other day, may have been early | |
| 22:37:39 | melwitt | hah, of course that was a reference | |
| 22:38:00 | mriedem | there was a lot of territory markings going on | |
| 22:38:21 | dansmith | still reeks of piss in here | |
| 22:39:15 | mriedem | dansmith: so i've got your cache_images notification test done, just trying to figure out how to get those samples hacked into the docs | |
| 22:39:42 | dansmith | mriedem: good work, mr. iedeman | |
| 22:39:50 | openstackgerrit | Merged openstack/nova stable/train: Fix listing deleted servers with a marker https://review.opendev.org/690711 | |
| 22:39:55 | melwitt | https://www.youtube.com/watch?v=B8dMXyJJwUA | |
| 22:40:14 | dansmith | melwitt: that was pasted peri-bro-down | |
| 22:40:17 | dansmith | you must've missed it | |
| 22:40:22 | melwitt | I did miss it | |
| 22:40:35 | dansmith | but it looked JUST like that, so you're caught yp | |
| 22:41:46 | melwitt | perfect | |
| 22:44:33 | mriedem | aha | |
| 22:52:33 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add notification sample test for aggregate.cache_images.start|end https://review.opendev.org/691129 | |
| 22:52:34 | mriedem | dansmith: there you go ^ | |
| 23:31:31 | openstackgerrit | melanie witt proposed openstack/nova master: Add new policy rule for viewing host status UNKNOWN https://review.opendev.org/679181 | |
| #openstack-nova - 2019-10-25 | |||
| 04:11:38 | VedaAnnayappa | hi | |
| 04:11:52 | VedaAnnayappa | I need help with tempest test bed. | |
| 04:17:35 | VedaAnnayappa | is there any impact on nova testcases in py37? | |
| 05:36:10 | openstackgerrit | Merged openstack/nova master: Add regression test for bug 1824435 https://review.opendev.org/688205 | |
| 05:36:10 | openstack | bug 1824435 in OpenStack Compute (nova) "fill_virtual_interface_list migration fails on second attempt" [Medium,In progress] https://launchpad.net/bugs/1824435 - Assigned to melanie witt (melwitt) | |
| 10:28:33 | openstackgerrit | Merged openstack/nova master: libvirt: Ignore volume exceptions during post_live_migration https://review.opendev.org/682622 | |
| 10:33:16 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: libvirt: Ignore volume exceptions during post_live_migration https://review.opendev.org/691281 | |
| 10:33:40 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/stein: libvirt: Ignore volume exceptions during post_live_migration https://review.opendev.org/691282 | |
| 10:34:07 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/rocky: libvirt: Ignore volume exceptions during post_live_migration https://review.opendev.org/691283 | |
| 10:34:25 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/queens: libvirt: Ignore volume exceptions during post_live_migration https://review.opendev.org/691284 | |
| 11:22:49 | sean-k-mooney | aspiers: are you around for a quick question about amd cpus and openstack | |
| 11:24:29 | sean-k-mooney | aspiers: mainly have you tried to use cpu pinning and have you turned on SMT/hyperthreading when useing openstack? if so did it all just work fine? im ignoring the slighly dumb numa topology of eypc gen 1. | |
| 11:26:35 | sean-k-mooney | aspiers: i have no reason to belive that nova will need any modifcation to work with amd eypc cpu for normal vms but im closing a ticket downstream saying i am happy to rely on the testing done by libvirt and qemu in this area. | |
| 13:34:16 | mriedem | gibi: do you think you'll be able to get to these today? https://review.opendev.org/#/q/topic:bp/support-move-ops-with-qos-ports-ussuri+status:open | |
| 13:35:02 | ygk_12345 | hi all | |
| 13:35:23 | ygk_12345 | can someone point me to any resource in understanding the code for nova wsgi.py file ? | |
| 13:35:35 | gibi | mriedem: if I don't have to solve the rollback of the parent_ifname ugliness there then yes I can fix your comments to clean up the patches | |
| 13:36:05 | mriedem | yeah i'm not asking for that | |
| 13:36:26 | gibi | mriedem: OK, then I will go and fix those patches up | |
| 13:45:27 | mriedem | ygk_12345: http://lists.openstack.org/pipermail/openstack-discuss/2019-October/010307.html may be helpful | |
| 13:49:15 | openstackgerrit | Merged openstack/nova master: [Trivial] Add missing ws between words https://review.opendev.org/689921 | |
| 13:57:22 | mriedem | lyarwood: very weird that your stein backport fails pep8 https://review.opendev.org/#/c/691282/ for an unused variable but it didn't on master or train, | |
| 13:57:28 | mriedem | something must have changed since train with flake8 rules | |
| 13:59:09 | mriedem | my guess would be something due to https://github.com/openstack/nova/commit/3e65f778bdb51e2cf0693f4e9b1b4539ea5c294c#diff-b91f3d5bd63fcd17221b267e851608e8 | |
| 14:01:50 | lyarwood | mriedem: what the flying friday | |
| 14:01:59 | mriedem | pycodestyle doesn't check for F841 | |
| 14:02:02 | mriedem | https://pycodestyle.readthedocs.io/en/latest/intro.html#error-codes | |
| 14:03:51 | mriedem | stephenfin: ^ | |
| 14:04:54 | sean-k-mooney | mriedem: stephenfin is currently in japan watching the rugby world cup | |
| 14:05:01 | mriedem | yeah i know he's not around | |
| 14:06:13 | bbobrov | dansmith: hi, i will be fixing amd sev code | |
| 14:10:34 | mriedem | lyarwood: if you're able it would be good to move forward on these rocky backports today https://review.opendev.org/#/q/status:open+project:openstack/nova+branch:stable/rocky+label:Code-Review=2 | |
| 14:10:50 | mriedem | because those are also in queens https://review.opendev.org/#/q/status:open+project:openstack/nova+branch:stable/queens | |
| 14:13:56 | lyarwood | mriedem: yup stuck in calls for a while but I'll take a look once I'm out | |
| 14:17:10 | dansmith | mriedem: thanks for that notification sample patch | |
| 14:17:30 | dansmith | mriedem: so onto the more progress-related notifications, I think we could do one of two things: | |
| 14:17:58 | dansmith | 1. Have the computes all emit a notification when they've downloaded an image, which won't include the aggregate | |
| 14:18:25 | dansmith | 2. Have the conductor emit one notification per compute after it reports what it did | |
| 14:18:59 | dansmith | 3. Have conductor emit a progress notification as it is doing it with numbers that can be turned into percentage (like 30% complete or N/M complete) | |
| 14:19:26 | dansmith | I can see benefits and drawbacks to all of those approaches | |
| 14:19:57 | dansmith | 1-2 don't let you predict completion, 3 doesn't given you much info, like which ones failed | |
| 14:20:55 | openstackgerrit | Eric Fried proposed openstack/os-traits master: Stop testing with py2 https://review.opendev.org/690997 | |
| 14:22:06 | openstackgerrit | Eric Fried proposed openstack/os-resource-classes master: Stop testing with py2 https://review.opendev.org/690996 | |
| 14:24:52 | openstackgerrit | Eric Fried proposed openstack/os-vif master: Drop python2 support and testing https://review.opendev.org/691364 | |
| 14:28:09 | mriedem | dansmith: couldn't 2 and 3 be combined? | |
| 14:28:28 | openstackgerrit | Eric Fried proposed openstack/python-novaclient master: Stop supporting and testing python2 https://review.opendev.org/691365 | |
| 14:28:36 | dansmith | mriedem: ah, so per compute, but with progress? sure | |
| 14:28:48 | mriedem | also, i think you can do 3 without 3 using the existing notification, | |
| 14:29:02 | mriedem | so you'd listen for aggregate.cache_images.start which passes the list of hosts being processed already, | |
| 14:29:21 | mriedem | then you listen for per-compute notifications for that aggregate, and since you know how many are in the aggregate from the first notification, you can do your own progress tracking | |
| 14:29:42 | mriedem | you know you're done when you get aggregate.cache_images.end | |
| 14:29:51 | mriedem | so i think i'd just go with 2 | |
| 14:30:09 | dansmith | that is assuming you don't have multiple operations going at once and/or that you don't have overlapping aggregates, both of which I think are not reasonable | |
| 14:30:11 | mriedem | 1 isn't great if the compute is down | |
| 14:30:14 | dansmith | not reasonable assumptions I mean | |
| 14:30:39 | mriedem | the initial notification has the aggregate name in it though | |
| 14:30:54 | mriedem | so even if you have hosts that overlap aggregates i don't think that's a problem from a tracking pov | |
| 14:31:19 | dansmith | you have to have aggregate in each notification to link all that up | |
| 14:31:20 | dansmith | either way, | |
| 14:31:35 | dansmith | 2 with a total and complete count makes the most sense to me I think | |
| 14:31:50 | mriedem | having the aggregate in the per-compute notification is fine i think | |
| 14:31:53 | dansmith | I wasn't sure if once per compute was too much, but sounds like you're thinking that way | |
| 14:32:06 | mriedem | you'd have the aggregate, the compute, maybe the image(s) and the status | |