| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-07-23 | |||
| 16:53:38 | mriedem | dansmith should probably look at those | |
| 16:53:45 | mriedem | given evacness | |
| 16:53:59 | stephenfin | mriedem: Suits me down to the ground | |
| 16:54:57 | sean-k-mooney | AJaeger: will nova beable to import the job from os-vif? i just kicked off an experimental job to make sure it all works | |
| 16:57:45 | openstackgerrit | Merged openstack/nova master: perform reshaper operations in single transaction https://review.openstack.org/582383 | |
| 16:57:53 | openstackgerrit | Merged openstack/nova master: Refactor _heal_instances_in_cell https://review.openstack.org/577896 | |
| 16:58:34 | AJaeger | sean-k-mooney: that'S how zuul was designed ;) | |
| 16:58:42 | AJaeger | sean-k-mooney: but yeah, I just have done that - thanks | |
| 16:59:12 | AJaeger | thanks, stephenfin | |
| 17:00:16 | AJaeger | sean-k-mooney: it's "check experimental" | |
| 17:00:18 | sean-k-mooney | AJaeger: os-vif has to be whitelisted in the tenant config to suppy job definitons... that said it is in project-config already is so ya i guess this should work. | |
| 17:00:51 | AJaeger | sean-k-mooney: all active repos are in that list ;) | |
| 17:01:36 | sean-k-mooney | AJaeger: ya i was just getting a little confused with how i was configuring my thridparty ci when i was playying with zuulv3 | |
| 17:02:19 | AJaeger | ;) | |
| 17:02:39 | AJaeger | sean-k-mooney: the job is scheduled to run according to zuul.openstack.org, so it's fine. | |
| 17:04:52 | sean-k-mooney | AJaeger: yep i was just checking that myself | |
| 17:05:04 | openstackgerrit | Elod Illes proposed openstack/nova stable/queens: Call generate_image_url only for legacy notification https://review.openstack.org/584969 | |
| 17:06:54 | AJaeger | ok, hope another core will review the job change in 583023 then - so that we can cleanup the infra files. | |
| 17:10:01 | sean-k-mooney | AJaeger: two quick questions by the way. i plan on converting it to a non legacy job. are there good examples of a vannily devstack job i can base it off. and 2 have ye updated the zuul from scratch to cover creating your own base job yet? | |
| 17:11:13 | sean-k-mooney | AJaeger: perhaps those are better asked on the zuul/infra channels instead. | |
| 17:12:25 | AJaeger | sean-k-mooney: for your second question: better ask on #zuul - I'm not involved there. For the devstack job, best ask andreaf on #openstack-qa. I know he converted some but don't know which one is a good example. You might want to search using codesearch, e.g. http://codesearch.openstack.org/?q=parent%3A%20devstack&i=nope&files=&repos= | |
| 17:14:23 | sean-k-mooney | AJaeger: thanks. its too late for rocky but i hope to convert the os-vif legacy job to non legacy and add a linux bridge version in the next month or so | |
| 17:17:29 | kashyap | cfriesen_: Thanks for the quick review | |
| 17:18:14 | AJaeger | sean-k-mooney: IMHO never too late to convert it ;) | |
| 17:22:12 | sean-k-mooney | AJaeger: well what i ment is we are past non-client ff so it wont be on the stable/rocky branch unless we backport it | |
| 17:39:35 | mriedem | stephenfin: heh, sorry, just meant dan has special context on that code | |
| 17:39:41 | mriedem | so he should (also) review it | |
| 17:52:54 | sean-k-mooney | do you ever read code and have to fight the urge to delete it all. | |
| 18:02:52 | AJaeger | sean-k-mooney: ah, yes | |
| 18:05:37 | mriedem | zzzeek: is there any major difference between doing a select query filter with == None vs == null() https://review.openstack.org/#/c/584504/3/nova/objects/instance_mapping.py@156 ? | |
| 18:06:28 | zzzeek | mriedem: shouldn't make a difference | |
| 18:06:32 | mriedem | ok thanks | |
| 18:07:31 | zzzeek | mriedem: null() is helpful in other cases where None might mean something different, like you want to force a SQL expression in some cases. but here == None will do "IS NULL" anyway | |
| 18:10:10 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Use consumer generation in _heal_allocations_for_instance https://review.openstack.org/577905 | |
| 18:13:49 | mriedem | dansmith: i'm +2 on your queued_for_delete migration, i guess we can rebase the series to swap those | |
| 18:13:56 | dansmith | cool | |
| 18:15:16 | mriedem | looks like powervm might have some ci issues http://184.172.12.213/17/560317/30/check/nova-in-tree-pvm/bf5d5bc/logs/stack.sh.txt.gz | |
| 18:15:23 | mriedem | error: command 'powerpc64le-linux-gnu-gcc' failed with exit status 1 | |
| 18:15:37 | mriedem | error: PIC register clobbered by ‘r30’ in ‘asm’ | |
| 18:16:14 | mriedem | powerkvm ci failing on the same https://dal05.objectstorage.softlayer.net/v1/AUTH_3d8e6ecb-f597-448c-8ec2-164e9f710dd6/pkvmci/nova/17/560317/30/check/tempest-dsvm-full-xenial/ac71b2d/devstacklog.txt.gz | |
| 18:16:19 | mriedem | edmondsw: ^ | |
| 18:17:07 | edmondsw | mriedem yep, we're aware, have a fix proposed to requirements to block greenlet 0.4.14 | |
| 18:17:07 | mriedem | https://pypi.org/project/greenlet/0.4.14/ | |
| 18:17:11 | mriedem | ah cool | |
| 18:17:24 | edmondsw | also broken the KVM CI from what I heard | |
| 18:17:29 | edmondsw | broke* | |
| 18:17:35 | mriedem | it did | |
| 18:17:54 | edmondsw | getting a little pushback from the requirements guys, so we'll see how this goes | |
| 18:18:05 | edmondsw | https://review.openstack.org/#/c/584881/ | |
| 18:18:11 | mriedem | yeah i'm looking at it | |
| 18:18:49 | edmondsw | just posted a comment in response | |
| 18:19:48 | mriedem | hmm, so in the before times we would have also blacklisted greenlet 0.4.14 in global-requirements.txt, | |
| 18:19:55 | mriedem | but that's per-project now, | |
| 18:20:09 | mriedem | so i'm wondering if nova can just blacklist 0.4.14 in nova's requirements.txt? | |
| 18:20:16 | mriedem | smcginnis: do you know ^ ? | |
| 18:21:15 | smcginnis | mriedem: Yeah, I believe Nova can just block that now. | |
| 18:27:18 | smcginnis | FWIW, until the gcc issue gets clarified I think we should block it in requirements too. | |
| 18:27:38 | openstackgerrit | Merged openstack/nova master: Rename auth_uri to www_authenticate_uri https://review.openstack.org/576820 | |
| 18:43:52 | openstackgerrit | sean mooney proposed openstack/nova master: fix disk_bus handeling https://review.openstack.org/584999 | |
| 18:55:04 | openstackgerrit | Merged openstack/nova master: Func test for improper cn local DISK_GB reporting https://review.openstack.org/583646 | |
| 18:56:27 | openstackgerrit | Merged openstack/nova master: [placement] disallow additional fields in allocations https://review.openstack.org/583907 | |
| 19:15:28 | mriedem | efried: if you agree with my comments on https://review.openstack.org/#/c/560317/ i'll update those quick | |
| 19:16:08 | efried | looking | |
| 19:17:52 | mriedem | and i'll probably blacklist greenlet 0.4.14 in a change below this so we can get the powerkvm ci passing again | |
| 19:18:07 | efried | mriedem: I assume https://review.openstack.org/#/c/560317/30/nova/virt/libvirt/driver.py@8940 you meant, "I'd make this an info" | |
| 19:18:33 | efried | mriedem: I'm fine if you make those changes, sure. | |
| 19:18:39 | mriedem | yeah heh | |
| 19:18:43 | mriedem | ok will do | |
| 19:18:58 | efried | kashyap: You good with https://review.openstack.org/#/c/560317/ at this point? | |
| 19:19:12 | mriedem | kashyap is likely done for the day | |
| 19:53:06 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add method to get cpu traits https://review.openstack.org/560317 | |
| 19:53:07 | openstackgerrit | Matt Riedemann proposed openstack/nova master: FakeLibvirtFixture: mock get_fs_info https://review.openstack.org/579201 | |
| 19:53:08 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Blacklist greenlet 0.4.14 https://review.openstack.org/585016 | |
| 19:53:56 | mriedem | edmondsw: ^ | |
| 19:55:15 | edmondsw | mriedem +1 | |
| 19:55:35 | edmondsw | mmedvede ^ | |
| 20:06:47 | mnaser | lols here's interesting behaviour | |
| 20:07:05 | mnaser | list doesn't hit cells so vms that were stuck in 'scheduling' are shown when you list the vms in horizon | |
| 20:07:18 | mnaser | but then as horizon tries to 'refresh' them by doing a GET, nova returns a 404 | |
| 20:07:25 | mnaser | so they disappear afterwards | |
| 20:07:46 | dansmith | mnaser: what do you mean by "list doesn't hit cells" ? | |
| 20:08:14 | dansmith | also, GET on a scheduling vm shouldn't return a 404 | |
| 20:08:15 | mnaser | dansmith: well those are instances which seem to have gotten somehow stuck in 'scheduling' status | |
| 20:08:28 | mnaser | ok, let me verify this | |
| 20:08:56 | dansmith | oh, if they're scheduling they're not created in any cell, maybe that's what you mean, but list should still show them from their buildrequest of course | |
| 20:09:04 | dansmith | and GET during/after should all work | |
| 20:09:12 | mnaser | well list seems to show them, but GET gives a 404 afaik | |
| 20:09:36 | mnaser | im double checking the behaviour | |
| 20:09:47 | mnaser | yeah, they show up in list, but GET gives 404 | |
| 20:10:24 | dansmith | something's broken then | |
| 20:10:28 | dansmith | that shouldn't happen | |
| 20:10:43 | mnaser | well i guess the db got in some weird state | |
| 20:10:56 | mnaser | or even if the db is mucked it shouldn't be like this? | |
| 20:11:18 | dansmith | well, depends on the level of mucking I guess, but.. it should never be like this | |
| 20:11:37 | mnaser | so as i understand it, list just grabs stuff from the 'api' db, but the 'get' actually hits the cell | |
| 20:11:43 | dansmith | no | |
| 20:11:58 | dansmith | list grabs stuff from the api and cell databases together | |
| 20:12:11 | mnaser | so it must know the mapping too i guess | |
| 20:12:15 | dansmith | get will pull the mapping from the api db, and if it looks like it is scheduled, will talk to the cell, and if not will return the build request data | |
| 20:12:21 | mnaser | ah okay | |