| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-05-04 | |||
| 22:14:09 | mriedem | nvm i'll clean it up | |
| 22:28:45 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Verify project id for flavor access calls https://review.openstack.org/435432 | |
| 22:30:56 | mriedem | oomichi: ^ i cleaned that one up | |
| 22:34:44 | oomichi | mriedem: thanks, lets close the bug | |
| 22:36:41 | dansmith | sdague: melwitt: shockingly, I now have logs for that instance | |
| 22:37:27 | melwitt | you mean with sdague's change you do? | |
| 22:38:24 | dansmith | melwitt: checkit, like I thought: http://logs.openstack.org/37/458537/10/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/0601a5e/logs/screen-n-cauth.txt.gz#_May_04_20_47_48_620783 | |
| 22:38:37 | dansmith | melwitt: yeah | |
| 22:39:25 | melwitt | validate_console_port | |
| 22:40:02 | dansmith | dinnuh I say? | |
| 22:40:11 | melwitt | yeah | |
| 22:40:25 | dansmith | I shall flog that tomorrow | |
| 22:41:25 | melwitt | cool. somehow instance must not be targeted to the right MQ even though there's a target_cell call in there | |
| 22:44:16 | dansmith | something. | |
| 22:47:29 | melwitt | strange. not seeing what's wrong there | |
| 22:48:00 | mnaser | i've been flying around the openstack api but this might be a little crazy. is it possible to remove an existing ephermal drive and replace it by cinder volume... on an existing instance? | |
| 22:48:13 | melwitt | the context gets wiped out after the context manager exits, but all of the calls are under it | |
| 22:48:31 | mnaser | so converting to boot from cinder for an existing VM (bdm stuff relevant at all here?) | |
| 23:16:06 | jaypipes | mriedem, dansmith, edleafe, alex_xu: your prompt attention to https://review.openstack.org/#/q/topic:cleanups+project:openstack/os-traits would be appreciated. I'd like to cut a 0.3 os-traits release ASAP because I need to use it in some of the new placement work... | |
| 23:18:37 | dansmith | jaypipes: I -1'ed the fsck out of that last patch | |
| 23:18:38 | dansmith | utter garbage! | |
| 23:20:15 | jaypipes | dansmith: fixed. :) | |
| 23:23:43 | dansmith | jaypipes: that was almost a very embarrassing event | |
| 23:23:50 | jaypipes | :) | |
| 23:29:07 | mriedem | ha | |
| 23:29:13 | mriedem | got you both | |
| 23:30:21 | dansmith | mriedem: luckily I don't care about my ratio on os-traits, so impede progress all you want buddy :) | |
| 23:32:04 | mriedem | i'll always remember | |
| 23:32:17 | mriedem | like your love of pizza and cookies and complete sentences | |
| 23:32:50 | dansmith | heh | |
| 23:33:19 | mriedem | jaypipes: on the last one, | |
| 23:33:27 | mriedem | that regex isn't enforced in the api schema from what i can tell | |
| 23:33:32 | mriedem | https://github.com/openstack/nova/blob/master/nova/api/openstack/placement/handlers/trait.py#L29 | |
| 23:33:50 | mriedem | or does that just mean for custom traits? | |
| 23:34:04 | jaypipes | yeah, that's just custom traits. | |
| 23:34:25 | mriedem | well, https://review.openstack.org/#/c/462759/2/os_traits/hw/cpu/x86.py is a standard trait no? | |
| 23:35:22 | jaypipes | no | |
| 23:35:26 | jaypipes | that's a module. | |
| 23:35:56 | jaypipes | hw.cpu.x86.SSE42 is a standard trait *symbol*, but the trait string is HW_CPU_X86_SSE42 | |
| 23:36:46 | mriedem | but based on the api schema, | |
| 23:36:57 | mriedem | i could create a trait that's named HW_CPU_X86_AES-NI right? | |
| 23:38:12 | mriedem | or are you only allowed to create custom traits via the API? https://github.com/openstack/nova/blob/master/nova/api/openstack/placement/handlers/trait.py#L99 | |
| 23:38:21 | mriedem | and standard traits come from the lib? | |
| 23:44:12 | jaypipes | mriedem: correct. | |
| 23:47:35 | jaypipes | mriedem: meh, fixed that README typo. | |
| 23:51:32 | jaypipes | mriedem: I just want to keep the standard trait naming consistent. | |
| 23:52:45 | mriedem | alright all done | |
| #openstack-nova - 2017-05-05 | |||
| 00:07:15 | openstackgerrit | melanie witt proposed openstack/nova master: Make NovaException format errors fatal for tests https://review.openstack.org/411615 | |
| 00:07:16 | openstackgerrit | melanie witt proposed openstack/nova master: Fix unit test exception KeyErrors https://review.openstack.org/462764 | |
| 00:07:54 | melwitt | mriedem: I revived my old attempt at the exception KeyErrors detection ^ I ordered the fixes after the detection patch for now to demonstrate that it works | |
| 00:13:34 | openstackgerrit | melanie witt proposed openstack/nova master: Fix unit test exception KeyErrors https://review.openstack.org/462764 | |
| 00:15:40 | jaypipes | mriedem: ty sir. | |
| 00:15:58 | jaypipes | mriedem: to cut a release, do I just git tag -a it? | |
| 00:31:33 | mriedem1 | jaypipes: nope | |
| 00:31:40 | mriedem1 | you post to openstack/releases | |
| 00:32:26 | mriedem1 | jaypipes: like this https://github.com/openstack/releases/commit/33ebc95728b40ca6a06c764a51a79dc24e95b555 | |
| 00:32:41 | mriedem1 | so you need a new - version entry | |
| 00:32:45 | mriedem1 | with the latest git hash | |
| 00:32:53 | mriedem1 | highlights are optional | |
| 00:33:33 | mriedem1 | melwitt: ok | |
| 00:33:34 | mriedem1 | nice | |
| 00:33:48 | mriedem1 | melwitt: i'll have to think of some other crazy crap that needs fixing now | |
| 00:39:13 | mriedem1 | melwitt: wait, so, | |
| 00:39:31 | mriedem1 | how does the bottom change not fail if you didn't also fix the things in the top change at the same time? | |
| 00:58:42 | mriedem | Kevin_Zheng: i've got a start on the forum session for listing instances using searchlight https://etherpad.openstack.org/p/BOS-forum-using-searchlight-to-list-instances | |
| 00:58:56 | mriedem | in case you have things you'd like to see covered in that session; the etherpad is a bit basic at the moment | |
| 01:14:03 | mriedem | jaypipes: and this one for you to gaze upon https://etherpad.openstack.org/p/BOS-forum-compute-instance-volume-affinity-hpc | |
| 01:15:54 | jaypipes | mriedem: cool, thx. | |
| 01:16:07 | openstackgerrit | Jay Pipes proposed openstack/nova master: add cache which keeps DB and os_traits in sync https://review.openstack.org/462769 | |
| 01:16:08 | openstackgerrit | Jay Pipes proposed openstack/nova master: integrate trait cache with TraitList and Trait https://review.openstack.org/462770 | |
| 01:16:45 | jaypipes | mriedem, dansmith: ^^ depends on the as-yet-unreleased os-traits 0.3 release. | |
| 01:18:34 | mriedem | jaypipes: ping me when you get that releases patch up | |
| 01:18:38 | mriedem | or have me push it | |
| 01:18:40 | mriedem | whichever | |
| 01:18:46 | mriedem | b/c i can push it real good | |
| 01:19:09 | Kevin_Zheng | mriedem: I do have some information about this, could I update it in the bottom of the etherpad and maybe you can have some idea and update the whole ehterpad？ | |
| 01:19:46 | mriedem | Kevin_Zheng: sure | |
| 01:20:04 | Kevin_Zheng | mriedem: :) | |
| 01:20:22 | mriedem | Kevin_Zheng: i put an "other notes" section at the bottom https://etherpad.openstack.org/p/BOS-forum-using-searchlight-to-list-instances | |
| 01:20:29 | Kevin_Zheng | OK | |
| 01:21:20 | jaypipes | mriedem: https://review.openstack.org/462771 | |
| 01:25:20 | mriedem | jaypipes: beautiful | |
| 01:25:31 | jaypipes | mriedem: thanks honey. | |
| 01:26:13 | sean-k-mooney | jaypipes: does that new release have the network namespace out of interest? | |
| 01:26:20 | jaypipes | sean-k-mooney: yup. | |
| 01:26:27 | sean-k-mooney | cool | |
| 01:26:36 | jaypipes | sean-k-mooney: also... what the heck are you doing up? :) | |
| 01:27:26 | sean-k-mooney | i did not sleep yesterday so i had a 4 hour nap from 8pm to ~midnight my time so im awake again | |
| 01:27:51 | sean-k-mooney | not ideal but oh well | |
| 01:27:53 | jaypipes | ah :) | |
| 01:29:37 | sean-k-mooney | jaypipes: do you know what timezone sdauge is in? | |
| 01:30:01 | jaypipes | EDT | |
| 01:30:03 | jaypipes | sean-k-mooney: ^ | |
| 01:31:00 | sean-k-mooney | ah so he will be offline for the night. its pretty late for you too right | |
| 01:32:04 | jaypipes | sean-k-mooney: 21:30 here, yeah | |
| 01:32:47 | sean-k-mooney | i need to open a minor bug with the new devstack systemd code | |
| 01:33:17 | sean-k-mooney | it almost works across reboots except for https://github.com/openstack-dev/devstack/blob/a292c5068ce8b285afc1ecfd473c91c8789922d3/lib/apache#L231-L232 | |
| 01:33:42 | sean-k-mooney | just need to have systemd create the folder automatically and it will work | |
| 01:34:15 | sean-k-mooney | currently the placement api and keystone are fialing to start because of that | |
| 01:44:50 | melwitt | mriedem: the bottom change should fail, I did that intentionally (temporarily) to show that it will work with the text fixes on top of it | |