| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-06-21 | |||
| 17:57:33 | mriedem | ildikov: it's on the list - my only tab open right now | |
| 17:57:35 | melwitt | the gate seems really angry right now. getting weird novnc tempest failures and "no route to host" | |
| 17:58:28 | ildikov | mriedem: I'm guessing that's a good sign, plz don't close it until you reviewed :) | |
| 17:59:12 | ildikov | mriedem: added a few tests to the swap patch | |
| 18:00:00 | ildikov | mriedem: should I/we involve more people from the Nova team to ensure we're progressing fast enough? | |
| 18:00:53 | mriedem | ildikov: more nova people don't exist | |
| 18:01:21 | ildikov | mriedem: hmm, then what now? | |
| 18:01:26 | ildikov | mriedem: should we clone you? | |
| 18:01:31 | mriedem | i sugggest a rebranding in the governance repo to recruit new nova people | |
| 18:03:08 | ildikov | that'll need some coaching and mentoring | |
| 18:04:04 | ildikov | I'm constantly trying to recruit, will need to do a better job... | |
| 18:09:18 | mriedem | ildikov: i was just joking | |
| 18:09:46 | mriedem | melwitt: jaypipes: dansmith: $.05 to whoever knows the answer to: what is the type of the object returned from driver.block_device_info_get_mapping() | |
| 18:09:57 | ildikov | mriedem: well, every joke has some truth in it :) | |
| 18:10:16 | melwitt | mriedem: I'm gonna guess a dict | |
| 18:10:17 | jaypipes | mriedem: no clue. | |
| 18:10:29 | mriedem | melwitt: bzzt | |
| 18:10:40 | ildikov | mriedem: and I would really love to have the new attach stuff landed in Pike :) | |
| 18:10:42 | melwitt | dangit. | |
| 18:10:46 | mriedem | hint: it's some type of list | |
| 18:11:02 | mriedem | don't let the singular in the name fool you | |
| 18:11:11 | melwitt | BlockDeviceMappingList | |
| 18:11:21 | melwitt | hah | |
| 18:11:53 | mriedem | is that your final answer? | |
| 18:12:14 | melwitt | sure | |
| 18:12:21 | mriedem | WRONG | |
| 18:12:25 | mriedem | BZZT | |
| 18:12:31 | melwitt | gah | |
| 18:12:44 | mriedem | it's a list of nova.virt.block_device.DriverBlockDevice objects | |
| 18:12:49 | mriedem | s/objects/subclasses/ | |
| 18:12:52 | mriedem | which are wrappers of, | |
| 18:12:54 | mriedem | you guessed it, | |
| 18:13:02 | mriedem | nova.objects.BlockDeviceMapping | |
| 18:13:05 | melwitt | lol | |
| 18:13:16 | melwitt | um, yeah. would never have guessed that | |
| 18:13:29 | mriedem | mdbooth would | |
| 18:13:31 | mriedem | he dreams about it | |
| 18:13:47 | mriedem | that and haggas | |
| 18:13:55 | mriedem | *haggis | |
| 18:13:57 | melwitt | lol! | |
| 18:28:37 | mriedem | we don't have a ListOfUUIDsField in o.vo?! | |
| 18:36:50 | jaypipes | mriedem: nope. | |
| 18:50:09 | openstackgerrit | Ildiko Vancsa proposed openstack/nova master: Add Cinder v3 detach to swap_volume https://review.openstack.org/456971 | |
| 19:02:58 | mriedem | ildikov: ok here you go https://review.openstack.org/#/c/463987/ | |
| 19:03:04 | mriedem | stvnoyes already skedaddled for the day | |
| 19:03:17 | mriedem | i have some bad news | |
| 19:03:49 | mriedem | ^ is only implemented for the libvirt driver, but needs to also be handled for xen and hyperv | |
| 19:04:14 | ildikov | mriedem: thanks for the review | |
| 19:04:30 | ildikov | mriedem: can the other hypervisor be covered in follow-up patches? | |
| 19:05:42 | ildikov | mriedem: also if you feel like adding a zillion comments to the swap patch that's very welcomed as well ^^ | |
| 19:05:56 | ildikov | mriedem: just to keep me busy while you're partying in China :) | |
| 19:09:41 | mriedem | I suppose the other virt drivers, or all virt drivers, can be done in a series of changes since none of this does anything with the new flow until we start attaching with the new style at the very end | |
| 19:10:20 | ildikov | ok, that was my thinking too | |
| 19:28:39 | openstackgerrit | Merged openstack/nova master: hypervisor_hostname must match get_available_nodes https://review.openstack.org/475926 | |
| 19:34:08 | openstackgerrit | Merged openstack/nova master: Provide a hint when _verify_response fails https://review.openstack.org/475512 | |
| 19:55:31 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix redundant BDM lookups during rebuild https://review.openstack.org/476258 | |
| 19:59:29 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Explain why API services are filtered out of GET /os-services https://review.openstack.org/476189 | |
| 20:00:08 | mriedem | dansmith: https://review.openstack.org/#/c/475067/ won't pass mitaka without us doing some things | |
| 20:00:14 | mriedem | some dirty deeds | |
| 20:08:22 | sdague | mriedem: honestly, just make all the mitaka tempest jobs non-voting | |
| 20:08:35 | sdague | and trust people read the results | |
| 20:10:26 | mriedem | i'm checking with lbragstad if the roles api actually changed it's response code at some point | |
| 20:17:14 | cfriesen_ | mriedem: do you think it would make sense to submit https://review.openstack.org/#/c/468203 for review before https://review.openstack.org/#/c/461456 ? I'm a bit worried about the second one getting blocked and preventing the first from being considered. | |
| 20:18:47 | mriedem | cfriesen_: i don't have any idea about that code | |
| 20:22:53 | dansmith | mriedem: oh, this is that thing, right.. I just blindly rechecked it | |
| 20:23:18 | mriedem | yeah trying to figure out if keystone changed their api since mitaka | |
| 20:23:49 | dansmith | tests not passing indicate that it should be taken out back and shot, IMHO | |
| 20:24:26 | mriedem | keystone? | |
| 20:25:10 | dansmith | no mitaka | |
| 20:25:26 | mriedem | oh, yeah, well it's already after it's eol date | |
| 20:25:32 | dansmith | right | |
| 20:29:48 | cfriesen_ | mriedem: dansmith: given the -1 to https://review.openstack.org/#/c/300616/, should I submit a patch to disallow swap-to-noswap resizes? | |
| 20:31:16 | mriedem | i don't have sufficient context on that | |
| 20:31:25 | cfriesen_ | bug is https://bugs.launchpad.net/nova/+bug/1552777 | |
| 20:31:25 | openstack | Launchpad bug 1552777 in OpenStack Compute (nova) "resizing from flavor with swap to one without swap puts instance into Error status" [Medium,In progress] - Assigned to Kam Nasim (knasim-wrs) | |
| 20:31:33 | mriedem | let me rephrase | |
| 20:31:48 | mriedem | i don't have nor want to have at this current point in time sufficient context on that | |
| 20:31:57 | cfriesen_ | kay | |
| 20:32:04 | mriedem | nor will "cleaned up BDMs" probably ever happen | |
| 20:32:21 | mriedem | so holding it hostage for something that won't get done is rough | |
| 20:32:24 | cfriesen_ | indeed, hence the proposal to go with mdbooth's suggestion to just block it | |
| 20:32:44 | mriedem | oh, well, i don't know what that looks like | |
| 20:32:55 | mriedem | but maybe chat with mdbooth about it | |
| 20:33:00 | cfriesen_ | can do | |
| 20:49:47 | jaypipes | mriedem, dansmith: wow. just... wow. http://paste.openstack.org/show/613333/ | |
| 20:56:40 | openstackgerrit | Rick Bartra proposed openstack/nova master: Add policy granularity to the Flavors API https://review.openstack.org/449288 | |
| 20:56:51 | jaypipes | mriedem, dansmith: ludicrously annoying levels of useless abstractions. :( | |
| 20:57:08 | dansmith | jaypipes: are you blaming me or just venting to me? | |
| 20:57:25 | jaypipes | dansmith: venting. | |
| 20:57:33 | dansmith | jaypipes: roger. carry on. | |
| 20:59:02 | mriedem | that's making sure the host manager calls the scheduler driver method right? | |
| 20:59:37 | mriedem | i would say that abstraction is less annoying than the bdm turducken i quizzed everyone with earlier today | |
| 20:59:45 | dansmith | heh | |
| 21:01:40 | edleafe | jaypipes: that was to reflect the version change from 4.3 to 4.4 - see the following test | |
| 21:01:55 | openstackgerrit | Mathieu Gagné proposed openstack/nova master: Add ability to signal and perform online volume size change https://review.openstack.org/454322 | |
| 21:03:33 | jaypipes | edleafe: I know. it's just annoying. | |
| 21:04:08 | edleafe | jaypipes: the last time I made a version change I got dinged for *not* writing a test like that | |
| 21:06:40 | mriedem | dansmith: i might need a hangout with you tomorrow to clarify plans for paging over compute nodes with a uuid marker | |
| 21:06:43 | mriedem | i know we talked about it once, | |
| 21:06:51 | mriedem | but i've already forgotten what we talked about | |