| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-07-12 | |||
| 17:55:54 | melwitt | oh yeah, queens. 17.0.5 is queens, sorry | |
| 17:56:02 | dansmith | okay | |
| 18:03:16 | melwitt | this is a weird one that I've noticed myself when trying to trace requests during debugging, periodic tasks sharing request_id with actual requests. the weird thing is I swear this did not use to happen in the past, but AFAIK nothing has changed around the periodic task context use https://bugs.launchpad.net/nova/+bug/1773102 | |
| 18:03:17 | openstack | Launchpad bug 1773102 in OpenStack Compute (nova) "Abnormal request id in logs" [Undecided,New] | |
| 18:03:49 | melwitt | I don't know why this has started happening in the last couple of cycles (not sure when it started) | |
| 18:04:48 | melwitt | maybe cell targeting is messing with it somehow | |
| 18:05:01 | efried | melwitt: Nothing to do with the global request ID work? | |
| 18:05:30 | melwitt | efried: I don't think so. the global request ID is a separate request ID IIUC | |
| 18:19:36 | artom | *sigh* | |
| 18:19:47 | artom | How do I file a bug in Storyboard? Is it just a Story? | |
| 18:24:36 | melwitt | artom: I don't know, but what are you looking to file a bug for? I didn't think we use storyboard in nova | |
| 18:24:44 | artom | melwitt, osc | |
| 18:24:51 | melwitt | ah | |
| 18:25:14 | artom | Am I an old man now? Grumbling about change for change's sake? | |
| 18:25:43 | arvindn05 | mriedem: thanks for the review on https://review.openstack.org/#/c/581920/ - Update glance documentation for trait support. I've addressed the comments. I asked the glance core folks to take a look as well, they should put in their comments/approvals early next week | |
| 18:29:33 | mriedem | artom: yes it's a story | |
| 18:29:51 | openstackgerrit | melanie witt proposed openstack/nova master: DNM: Don't overwrite greenthread-local context during target_cell https://review.openstack.org/582332 | |
| 18:30:18 | artom | mriedem, thanks! | |
| 18:39:12 | mriedem | oopsydoodle http://logs.openstack.org/04/528704/33/check/ironic-tempest-dsvm-ipa-partition-pxe_ipmitool-tinyipa-python3/ec5520f/logs/screen-n-cpu.txt.gz#_Jul_12_18_10_59_143344 | |
| 18:41:11 | melwitt | hm, looks like that's coming from ironicclient | |
| 18:43:23 | openstackgerrit | Merged openstack/nova master: Revert "docs: Disable smartquotes" https://review.openstack.org/578841 | |
| 18:48:01 | melwitt | from distutils.version import StrictVersion and the compare operator for StrictVersion is the thing that accesses the 'version' attribute | |
| 18:48:27 | melwitt | so something about statements like this blows up https://github.com/openstack/python-ironicclient/blob/master/ironicclient/common/http.py#L161 | |
| 18:49:07 | openstackgerrit | Merged openstack/nova master: Mention osc-placement for managing traits in docs https://review.openstack.org/582173 | |
| 18:54:43 | mriedem | i can't get hits on that in logstash, not sure if the ironic job logs are indexed | |
| 18:55:30 | efried | Is the gate particularly slow at the moment? The check queue seems unusually deep and not moving very fast. | |
| 18:56:08 | mriedem | http://grafana.openstack.org/dashboard/db/zuul-status | |
| 18:56:35 | mriedem | looks like executors have dropped down | |
| 18:59:19 | efried | you must be seeing something I'm not. That dashboard looks pretty healthy to me. | |
| 19:00:00 | efried | or at least pretty stable for the past 8h. | |
| 19:00:23 | efried | 'cept for the spike in "Node Requests", whatever that is. | |
| 19:09:19 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Avoid up-call in get_instance_availability_zone https://review.openstack.org/582342 | |
| 19:09:22 | mriedem | fix for the cross_az_attach=False up-call issue with cells v2 ^ | |
| 19:09:38 | mriedem | kashyap: ^ is that commit message more to your liking? | |
| 19:12:08 | melwitt | really wish we still had mirroring of release tags to github *tear* | |
| 19:14:43 | mriedem | we do, but sometimes you have to put them in yourself | |
| 19:15:11 | mriedem | https://github.com/openstack/nova/tree/17.0.5 | |
| 19:15:21 | mriedem | i don't know why they don't show up in the dropdown | |
| 19:15:35 | melwitt | oh, weird. I didn't know that hack | |
| 19:16:57 | melwitt | yass | |
| 19:33:24 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Make reshape-provider-tree reflect reality https://review.openstack.org/582350 | |
| 20:01:08 | mriedem | TheJulia: just a few small things in the ironic rescue patch https://review.openstack.org/#/c/416487/ - i'd like to see the ironicclient 2.2.0 version negotiation stuff split out as a dependent change in the series if possible, | |
| 20:01:18 | mriedem | but i'm not married to that if it's more pain than it's worth | |
| 20:02:59 | TheJulia | I guess my real concern with retooling that is that we're less likely to get any traction on negotiation changes, and we'll end up shooting ourselves... again... with a static pin update. | |
| 20:03:46 | TheJulia | then again, I guess it could be before, but then what is the point if we're not actually upping the version in that single patch. | |
| 20:07:00 | mriedem | the point is it's laying the groundwork for a feature that's going to use it | |
| 20:07:18 | mriedem | again, i'm not a hard stop on splitting it out, just seems we could/should | |
| 20:07:32 | mriedem | also, seeing issues in n-cpu logs with StrictVersion usage in there | |
| 20:07:54 | mriedem | http://logs.openstack.org/04/528704/33/check/ironic-tempest-dsvm-ipa-partition-pxe_ipmitool-tinyipa-python3/ec5520f/logs/screen-n-cpu.txt.gz#_Jul_12_18_10_59_143344 | |
| 20:09:01 | mriedem | we must be passing in StrictVersion(None) at some point | |
| 20:10:22 | TheJulia | Interesting | |
| 20:10:48 | mriedem | anywho, gotta drop for a bit | |
| 20:10:50 | TheJulia | Great point on the logging, I think. It has been a long time since I looked at this patch | |
| 20:11:13 | TheJulia | oh! | |
| 20:11:17 | TheJulia | i know how :\ | |
| 20:21:54 | openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Compute: Handle reshaped provider trees https://review.openstack.org/576236 | |
| 20:30:32 | openstackgerrit | Eric Fried proposed openstack/nova master: WIP: Compute: Handle reshaped provider trees https://review.openstack.org/576236 | |
| 20:31:11 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Make reshape-provider-tree reflect reality https://review.openstack.org/582350 | |
| 20:32:21 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Make reshape-provider-tree spec reflect reality https://review.openstack.org/582350 | |
| 20:43:39 | melwitt | has anyone gotten this error trying to run functional tests? tox.ConfigError: ConfigError: substitution key 'posargs' not found | |
| 20:54:14 | efried | melwitt: Paste me your [testenv:functional] paragraph from tox.ini | |
| 20:54:45 | melwitt | I just pulled it from master to see if it changed anything and it didn't. I wonder if my tox version is too old or something | |
| 20:55:04 | efried | that would have been my first guess, yeah. | |
| 20:55:31 | efried | melwitt: It's near the top of my todo list to bump our min required tox version to take advantage of the basepython fix anyway... | |
| 20:57:23 | efried | melwitt: What is your tox version? | |
| 20:57:35 | melwitt | 2.3.1 | |
| 20:57:41 | efried | there ya go. | |
| 20:57:46 | efried | https://github.com/tox-dev/tox/issues/279 | |
| 20:57:48 | melwitt | from whatever's for xenial64 | |
| 20:57:56 | openstackgerrit | Jay Pipes proposed openstack/nova master: do not assume 1 consumer in AllocList.delete_all() https://review.openstack.org/582382 | |
| 20:57:57 | openstackgerrit | Jay Pipes proposed openstack/nova master: perform reshaper operations in single transaction https://review.openstack.org/582383 | |
| 20:58:52 | melwitt | thanx efried | |
| 20:58:54 | efried | melwitt: Be interesting to see if it goes away immediately by upgrading to 2.3.2 | |
| 20:59:01 | efried | rather than jumping straight to latest or whatever. | |
| 20:59:11 | efried | make sure it's the same bug and all. | |
| 20:59:12 | efried | but whatevs. | |
| 20:59:13 | melwitt | yeah, lemme see | |
| 21:01:30 | melwitt | yup, 2.3.2 works | |
| 21:01:31 | melwitt | yey | |
| 21:01:58 | efried | sweet | |
| 21:02:10 | efried | I'm going to work on that patch to bump the minimum to 3.1.1 and see if it flies. | |
| 21:09:35 | melwitt | 2.3.2 forever | |
| 21:11:17 | melwitt | just hassling efried | |
| 21:15:50 | efried | TheJulia: Slightly lower down on my to-do list was adding os-ironic-api-version kwargs to more methods in ironicclient. That's not what you're working on, is it? | |
| 21:16:51 | TheJulia | efried: my brain kind of segfaulted upon all of the everyone implement x api features and bump their major versions discussion at the ptg.. and has not revisited versioning since. | |
| 21:16:54 | TheJulia | err | |
| 21:16:58 | TheJulia | s/ptg/summit/ | |
| 21:17:22 | efried | TheJulia: ooo, that's also not at all what I'm referring to. | |
| 21:18:10 | TheJulia | oh, I know what your referring to, I'm not sure it is really actually needed for most usage patterns, at least that is what I was thinking about when I added it to a few more | |
| 21:18:44 | TheJulia | I'm just stating, my brain broke later on, and has not revisited the subject. | |
| 21:18:55 | efried | TheJulia: Okay. IIRC it was only needed in a small handful of nova use cases, like half a dozen. | |
| 21:19:08 | TheJulia | I thought it was... 3 or 4 | |
| 21:19:37 | openstackgerrit | Eric Fried proposed openstack/nova master: Use tox 3.1.1 and basepython fix https://review.openstack.org/582392 | |
| 21:19:40 | TheJulia | anyway, back to splitting it apart which actually makes sense given the fix necessary | |
| 21:19:42 | efried | melwitt: ^ | |
| 21:20:23 | efried | stephenfin: https://review.openstack.org/582392 | |
| 21:32:57 | openstackgerrit | melanie witt proposed openstack/nova master: DNM: Don't overwrite greenthread-local context in target_cell https://review.openstack.org/582332 | |
| 22:05:33 | openstackgerrit | melanie witt proposed openstack/nova master: Add functional regression test for bug 1778305 https://review.openstack.org/582407 | |
| 22:05:34 | openstack | bug 1778305 in OpenStack Compute (nova) "Nova may erronously look up service version of a deleted service, when hostname have been reused" [Undecided,New] https://launchpad.net/bugs/1778305 | |
| 22:05:35 | openstackgerrit | melanie witt proposed openstack/nova master: Don't generate service UUID for deleted services https://review.openstack.org/582408 | |