| *** hblixt_ has joined #openstack-nova | 00:00 | |
| *** jerrygb has quit IRC | 00:01 | |
| *** Daisy has quit IRC | 00:01 | |
| openstackgerrit | Tracy Jones proposed openstack/nova: config options: centralize section "wsgi" https://review.openstack.org/252687 | 00:03 |
|---|---|---|
| *** pratikmallya has quit IRC | 00:03 | |
| *** hblixt has quit IRC | 00:03 | |
| *** takashin has joined #openstack-nova | 00:03 | |
| *** yalie1 has quit IRC | 00:03 | |
| *** Sukhdev has quit IRC | 00:05 | |
| *** zhangjn has joined #openstack-nova | 00:05 | |
| *** zhangjn has quit IRC | 00:06 | |
| *** hblixt_ has quit IRC | 00:06 | |
| *** zhangjn has joined #openstack-nova | 00:06 | |
| *** ijw_ has quit IRC | 00:06 | |
| *** mgoddard has joined #openstack-nova | 00:07 | |
| *** otter768 has joined #openstack-nova | 00:08 | |
| *** ijw has joined #openstack-nova | 00:08 | |
| *** diogogmt has quit IRC | 00:09 | |
| *** Sukhdev has joined #openstack-nova | 00:09 | |
| *** Apoorva has joined #openstack-nova | 00:12 | |
| melwitt | dansmith: does this help text for the compute 'auto' option look okay to you? https://review.openstack.org/#/c/239321/ | 00:13 |
| *** salv-orl_ has joined #openstack-nova | 00:18 | |
| *** akshai has quit IRC | 00:18 | |
| *** otter768 has quit IRC | 00:19 | |
| *** alex_xu has quit IRC | 00:19 | |
| *** otter768 has joined #openstack-nova | 00:20 | |
| *** hemna has quit IRC | 00:20 | |
| *** salv-orlando has quit IRC | 00:20 | |
| *** hemna has joined #openstack-nova | 00:20 | |
| *** alex_xu has joined #openstack-nova | 00:21 | |
| *** zenoway has joined #openstack-nova | 00:21 | |
| *** migi has quit IRC | 00:22 | |
| *** vladikr has quit IRC | 00:22 | |
| *** zul has quit IRC | 00:22 | |
| *** mordred has quit IRC | 00:23 | |
| *** zul has joined #openstack-nova | 00:23 | |
| *** migi has joined #openstack-nova | 00:23 | |
| *** vladikr has joined #openstack-nova | 00:23 | |
| *** mordred has joined #openstack-nova | 00:23 | |
| *** EinstCra_ has quit IRC | 00:24 | |
| *** zhangjn has quit IRC | 00:24 | |
| *** otter768 has quit IRC | 00:24 | |
| *** gyee has joined #openstack-nova | 00:25 | |
| *** zenoway has quit IRC | 00:25 | |
| *** ijw has quit IRC | 00:26 | |
| *** mc_nair_ has quit IRC | 00:26 | |
| *** ctrath has quit IRC | 00:26 | |
| *** gokrokve has joined #openstack-nova | 00:28 | |
| *** takedakn has quit IRC | 00:30 | |
| *** tjones has quit IRC | 00:30 | |
| *** mkoderer has quit IRC | 00:30 | |
| *** belmoreira has quit IRC | 00:30 | |
| *** mnestratov has quit IRC | 00:32 | |
| *** ZZelle_ has quit IRC | 00:33 | |
| *** wyang has joined #openstack-nova | 00:36 | |
| *** mgoddard has quit IRC | 00:36 | |
| *** mkoderer has joined #openstack-nova | 00:36 | |
| *** ducttape_ has joined #openstack-nova | 00:38 | |
| *** Sukhdev has quit IRC | 00:40 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Add Hyper-V storage QoS support https://review.openstack.org/175479 | 00:43 |
| *** gokrokve has quit IRC | 00:43 | |
| melwitt | claudiub: hey, I was wondering why this doesn't also update the tests-py3.txt? https://review.openstack.org/#/c/253990/ | 00:43 |
| *** gokrokve has joined #openstack-nova | 00:44 | |
| *** hemna is now known as hemnafk | 00:44 | |
| claudiub | melwitt: it doesn't? weird... might have been a mistake from my part. thanks for pointing it out! will upload a patchset asap. | 00:44 |
| *** tonytan4ever has quit IRC | 00:45 | |
| melwitt | claudiub: okay, cool | 00:45 |
| *** RichardRaseley has quit IRC | 00:45 | |
| *** Sukhdev has joined #openstack-nova | 00:46 | |
| melwitt | claudiub: I noticed this one didn't either, so if we need an update to the blacklist you could upload that too https://review.openstack.org/#/c/238239/ | 00:47 |
| melwitt | as a separate change I mean | 00:47 |
| *** gokrokve has quit IRC | 00:49 | |
| melwitt | that might have been a preliminary fix, I'm not familiar with the series. upon further inspection it looks like the "Fixes..." patches are the ones that do the blacklist removals | 00:49 |
| claudiub | melwitt: yeah, that one is a preliminary fix | 00:49 |
| claudiub | melwitt: it's in a longer patch series, which, at the end, enables ~2k unit tests. | 00:49 |
| melwitt | claudiub: awesome | 00:50 |
| *** ijw has joined #openstack-nova | 00:50 | |
| *** openstackstatus has quit IRC | 00:50 | |
| *** openstack has joined #openstack-nova | 00:52 | |
| claudiub | melwitt: anyways, this is the biggest py3 change I have up, if you're feeling masochistic today. :D https://review.openstack.org/#/c/238323/ | 00:53 |
| *** ducttape_ has quit IRC | 00:53 | |
| melwitt | claudiub: hah, okay | 00:53 |
| *** zhangjn has joined #openstack-nova | 00:55 | |
| *** gokrokve has joined #openstack-nova | 00:58 | |
| *** gongysh_ has joined #openstack-nova | 01:03 | |
| gongysh_ | garyk, hi | 01:03 |
| gongysh_ | alex_xu, hi | 01:03 |
| *** ianbrown_ has quit IRC | 01:03 | |
| *** EinstCrazy has joined #openstack-nova | 01:03 | |
| alex_xu | gongysh_: hi | 01:03 |
| gongysh_ | did you forget my patch yesterday? | 01:04 |
| *** zhangjn has quit IRC | 01:04 | |
| alex_xu | gongysh_: no, too busy last night :) | 01:04 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Fixes network py3 unit tests https://review.openstack.org/253990 | 01:04 |
| *** zhangjn has joined #openstack-nova | 01:06 | |
| *** ijw has quit IRC | 01:13 | |
| *** rk4n has quit IRC | 01:14 | |
| *** Sukhdev has quit IRC | 01:15 | |
| *** chenzeng has quit IRC | 01:15 | |
| *** shihanzhang has joined #openstack-nova | 01:15 | |
| *** c00281451_ has joined #openstack-nova | 01:16 | |
| *** shz has quit IRC | 01:19 | |
| *** diogogmt has joined #openstack-nova | 01:21 | |
| *** krtaylor has quit IRC | 01:22 | |
| *** ducttape_ has joined #openstack-nova | 01:23 | |
| *** rk4n has joined #openstack-nova | 01:23 | |
| *** ljxiash has joined #openstack-nova | 01:24 | |
| *** zhenguo has joined #openstack-nova | 01:24 | |
| *** ljxiash_ has joined #openstack-nova | 01:24 | |
| *** ljxiash__ has joined #openstack-nova | 01:25 | |
| *** ljxias___ has joined #openstack-nova | 01:25 | |
| *** tjones has joined #openstack-nova | 01:25 | |
| *** ljxia____ has joined #openstack-nova | 01:26 | |
| *** ljxia____ has quit IRC | 01:27 | |
| *** ljxiash has quit IRC | 01:28 | |
| *** ljxiash has joined #openstack-nova | 01:28 | |
| *** ljxiash__ has quit IRC | 01:29 | |
| *** ljxiash_ has quit IRC | 01:29 | |
| *** ljxias___ has quit IRC | 01:30 | |
| *** gongysh_ has quit IRC | 01:30 | |
| *** shuquan has joined #openstack-nova | 01:33 | |
| *** krtaylor has joined #openstack-nova | 01:36 | |
| *** zhangjn_ has joined #openstack-nova | 01:36 | |
| *** terryyao has joined #openstack-nova | 01:39 | |
| *** zhangjn has quit IRC | 01:40 | |
| *** alex_xu has quit IRC | 01:41 | |
| *** gjayavelu has quit IRC | 01:42 | |
| *** alex_xu has joined #openstack-nova | 01:43 | |
| *** shuquan has quit IRC | 01:44 | |
| *** ducttape_ has quit IRC | 01:44 | |
| *** jamielennox is now known as jamielennox|away | 01:47 | |
| *** ferest has joined #openstack-nova | 01:47 | |
| *** ducttape_ has joined #openstack-nova | 01:47 | |
| *** cp16net has joined #openstack-nova | 01:49 | |
| *** gokrokve has quit IRC | 01:50 | |
| *** vilobhmm11 has quit IRC | 01:52 | |
| *** vilobhmm11 has joined #openstack-nova | 01:53 | |
| openstackgerrit | Dave McCowan proposed openstack/nova: Check context before returning cached value https://review.openstack.org/254358 | 01:53 |
| *** vilobhmm11 has quit IRC | 01:53 | |
| *** rk4n has quit IRC | 01:53 | |
| *** vilobhmm11 has joined #openstack-nova | 01:53 | |
| *** ferest has quit IRC | 01:53 | |
| *** rk4n has joined #openstack-nova | 01:55 | |
| *** Sukhdev has joined #openstack-nova | 01:56 | |
| *** Sukhdev has quit IRC | 01:59 | |
| *** rk4n has quit IRC | 02:01 | |
| *** Jeffrey4l has joined #openstack-nova | 02:01 | |
| *** armax has quit IRC | 02:02 | |
| *** pratikmallya has joined #openstack-nova | 02:03 | |
| *** rk4n has joined #openstack-nova | 02:04 | |
| *** tjones has left #openstack-nova | 02:10 | |
| *** ferest has joined #openstack-nova | 02:12 | |
| *** ferest has quit IRC | 02:12 | |
| *** ferest has joined #openstack-nova | 02:13 | |
| *** yalie has joined #openstack-nova | 02:13 | |
| tangchen | alex_xu: Hi alex, what is the naming style in nova/compute/instance_actions.py ? I think you told me once before, but I forgot the URL, | 02:13 |
| tangchen | alex_xu_ would you please tell me again ? | 02:14 |
| *** gokrokve has joined #openstack-nova | 02:14 | |
| alex_xu | tangchen: emm...I didn't remember that | 02:14 |
| alex_xu | tangchen: we tell you on the irc or some review on patchset? | 02:15 |
| *** Apoorva has quit IRC | 02:15 | |
| alex_xu | tangchen: looks like many naming style at here | 02:16 |
| tangchen | alex_xu: I think it is review on patch-set. I can find it. If there was not, then it is my mistake, I think. :) | 02:16 |
| *** Apoorva has joined #openstack-nova | 02:16 | |
| *** ducttape_ has quit IRC | 02:16 | |
| *** ducttape_ has joined #openstack-nova | 02:17 | |
| alex_xu | oomichi: do we have naming style for instance_actions? | 02:17 |
| *** mtanino has quit IRC | 02:17 | |
| *** ducttape_ has quit IRC | 02:18 | |
| oomichi | tangchen: what kind of naming style? | 02:18 |
| oomichi | tangchen: in request/response body? | 02:18 |
| tangchen | oomichi: I mean in nova/compute/instance_actions.py, it should be aaa_bbb or aaaBbb ? | 02:19 |
| alex_xu | oomichi: I guess tangchen talk about this https://github.com/openstack/nova/blob/master/nova/compute/instance_actions.py | 02:19 |
| alex_xu | oomichi: that will be the value of API | 02:19 |
| tangchen | oomichi alex_xu: Yes | 02:19 |
| oomichi | tangchen: https://review.openstack.org/#/c/253417/ will help it, I feel | 02:19 |
| oomichi | tangchen: in the response/request body, parameter names should be snake_case | 02:20 |
| *** dave-mccowan has quit IRC | 02:20 | |
| alex_xu | oomichi: that is the value, not field name, but I think we should use same style | 02:20 |
| alex_xu | or say we use same action name as in the API | 02:21 |
| alex_xu | oomichi: https://github.com/openstack/nova/blob/master/doc/api_samples/os-instance-actions/instance-action-get-resp.json#L3 | 02:21 |
| *** ferest has quit IRC | 02:22 | |
| tangchen | alex_xu, oomichi: OK, I see. Thanks. | 02:22 |
| oomichi | alex_xu: yeah, snake_case is nice for that also | 02:23 |
| *** ducttape_ has joined #openstack-nova | 02:23 | |
| oomichi | alex_xu: for the consistency | 02:23 |
| alex_xu | oomichi: ok | 02:23 |
| alex_xu | tangchen: np | 02:23 |
| *** devananda has joined #openstack-nova | 02:27 | |
| *** samueldmq has joined #openstack-nova | 02:27 | |
| *** stackdump has quit IRC | 02:28 | |
| *** gokrokve has quit IRC | 02:30 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: config options: centralize section: "keymgr" https://review.openstack.org/255044 | 02:30 |
| *** gokrokve has joined #openstack-nova | 02:31 | |
| *** Jeffrey4l has quit IRC | 02:31 | |
| *** Jeffrey4l has joined #openstack-nova | 02:32 | |
| *** yarkot has joined #openstack-nova | 02:34 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Expose quiesce/unquiesce API https://review.openstack.org/248989 | 02:34 |
| openstackgerrit | Zhen Qin proposed openstack/nova: Fix the bug that when running nova list --host with invalid hostname (either no such server or there's a typo in server name), it should return an error. https://review.openstack.org/255047 | 02:35 |
| *** ijw has joined #openstack-nova | 02:36 | |
| openstackgerrit | Zhen Qin proposed openstack/nova: Fix the bug that when running nova list --host with invalid hostname (either no such server or there's a typo in server name), it should return an error. https://review.openstack.org/255047 | 02:37 |
| *** rk4n has quit IRC | 02:38 | |
| *** lyanchih has joined #openstack-nova | 02:38 | |
| *** zenoway has joined #openstack-nova | 02:38 | |
| *** gokrokve has quit IRC | 02:38 | |
| *** thorst has joined #openstack-nova | 02:40 | |
| *** Jeffrey4l has quit IRC | 02:40 | |
| *** Jeffrey4l has joined #openstack-nova | 02:42 | |
| *** zenoway has quit IRC | 02:42 | |
| *** ducttape_ has quit IRC | 02:43 | |
| *** zhenq has quit IRC | 02:43 | |
| *** ducttape_ has joined #openstack-nova | 02:46 | |
| *** nelsnelson has quit IRC | 02:49 | |
| *** nelsnels_ has joined #openstack-nova | 02:49 | |
| *** ijw has quit IRC | 02:50 | |
| *** jamielennox|away is now known as jamielennox | 02:50 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/253901 | 02:52 |
| *** baoli has joined #openstack-nova | 02:52 | |
| openstackgerrit | NanZhang proposed openstack/nova: live-migration: Update instance availability_zone https://review.openstack.org/254041 | 02:52 |
| *** diogogmt has quit IRC | 02:53 | |
| *** suro-patz has quit IRC | 02:57 | |
| *** ducttape_ has quit IRC | 02:57 | |
| *** suro-patz has joined #openstack-nova | 02:57 | |
| *** takashin has left #openstack-nova | 02:57 | |
| *** jerrygb has joined #openstack-nova | 02:57 | |
| *** gokrokve has joined #openstack-nova | 02:59 | |
| *** ducttape_ has joined #openstack-nova | 03:01 | |
| *** yamahata has quit IRC | 03:01 | |
| *** greghayn1 has joined #openstack-nova | 03:02 | |
| *** greghayn1 has quit IRC | 03:02 | |
| *** jerrygb has quit IRC | 03:02 | |
| *** gokrokve has quit IRC | 03:04 | |
| *** vilobhmm11 has quit IRC | 03:07 | |
| *** jdurgin1 has quit IRC | 03:07 | |
| *** ducttape_ has quit IRC | 03:09 | |
| *** baoli has quit IRC | 03:09 | |
| *** Apoorva has quit IRC | 03:14 | |
| *** Jeffrey4l__ has joined #openstack-nova | 03:14 | |
| *** ducttape_ has joined #openstack-nova | 03:14 | |
| *** Jeffrey4l has quit IRC | 03:16 | |
| *** jerrygb has joined #openstack-nova | 03:18 | |
| *** zenoway has joined #openstack-nova | 03:20 | |
| *** pratikmallya has quit IRC | 03:21 | |
| *** rook has quit IRC | 03:22 | |
| *** links has joined #openstack-nova | 03:24 | |
| *** thorst has quit IRC | 03:25 | |
| *** zenoway has quit IRC | 03:25 | |
| *** mtanino has joined #openstack-nova | 03:31 | |
| *** unicell has quit IRC | 03:32 | |
| *** terryyao has quit IRC | 03:37 | |
| *** alex_xu_ has joined #openstack-nova | 03:40 | |
| openstackgerrit | lyanchih proposed openstack/nova: NUMATopologyFilter raise exception and not continue filter next node https://review.openstack.org/200630 | 03:40 |
| alex_xu_ | gongysh: the review is done | 03:40 |
| alex_xu_ | gongysh: hopes that helpful | 03:40 |
| *** alex_xu has quit IRC | 03:42 | |
| *** ducttape_ has quit IRC | 03:42 | |
| *** ijw has joined #openstack-nova | 03:43 | |
| *** ijw has quit IRC | 03:43 | |
| *** ijw has joined #openstack-nova | 03:43 | |
| *** thorst has joined #openstack-nova | 03:43 | |
| *** tbh has joined #openstack-nova | 03:44 | |
| *** thorst has quit IRC | 03:44 | |
| *** thorst has joined #openstack-nova | 03:45 | |
| *** jerrygb has quit IRC | 03:48 | |
| *** jerrygb has joined #openstack-nova | 03:48 | |
| *** rook has joined #openstack-nova | 03:48 | |
| openstackgerrit | javeme proposed openstack/nova: Return more accurate error msg for live migration https://review.openstack.org/243105 | 03:48 |
| *** alejandrito has quit IRC | 03:52 | |
| *** jerrygb has quit IRC | 03:52 | |
| *** fawadkhaliq has joined #openstack-nova | 03:53 | |
| *** thorst has quit IRC | 03:54 | |
| *** pratikmallya has joined #openstack-nova | 03:57 | |
| *** zhangjn_ has quit IRC | 03:58 | |
| *** changbl has joined #openstack-nova | 04:00 | |
| *** fawadkhaliq has quit IRC | 04:00 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: pci: adding support to specify a device_type in pci requests https://review.openstack.org/249017 | 04:02 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: pci: changing the claiming and allocation logic for PF/VF assignment https://review.openstack.org/249018 | 04:02 |
| *** links has quit IRC | 04:03 | |
| *** jerrygb has joined #openstack-nova | 04:03 | |
| *** gyee has quit IRC | 04:03 | |
| *** mtanino has quit IRC | 04:03 | |
| *** lyanchih has quit IRC | 04:04 | |
| *** pratikmallya has quit IRC | 04:05 | |
| openstackgerrit | Eric Brown proposed openstack/nova: VMware: Enforce minimum vCenter version of 5.5 https://review.openstack.org/253666 | 04:09 |
| *** aginwala has quit IRC | 04:09 | |
| notmorgan | sdague, mriedem_away: https://review.openstack.org/#/c/253792/ looks good and is passing all checks *except* the xen one that is not publishing logs correctly | 04:12 |
| *** ijw has quit IRC | 04:12 | |
| notmorgan | sdague, mriedem_away: and hyperv | 04:12 |
| notmorgan | sdague, mriedem_away: I'd like to look at getting this landed tomorrow if we can resolve the issues with 3rd party CI stuff (as needed). (cc mordred) | 04:12 |
| *** dims has joined #openstack-nova | 04:13 | |
| *** dims has quit IRC | 04:15 | |
| *** dims has joined #openstack-nova | 04:18 | |
| *** changbl has quit IRC | 04:21 | |
| *** gokrokve has joined #openstack-nova | 04:22 | |
| *** terryyao has joined #openstack-nova | 04:25 | |
| *** klkumar has joined #openstack-nova | 04:28 | |
| *** akshai has joined #openstack-nova | 04:28 | |
| *** zhangjn has joined #openstack-nova | 04:29 | |
| *** dims has quit IRC | 04:29 | |
| *** gokrokve has quit IRC | 04:30 | |
| *** browne has quit IRC | 04:30 | |
| *** changbl has joined #openstack-nova | 04:35 | |
| *** links has joined #openstack-nova | 04:38 | |
| *** vilobhmm11 has joined #openstack-nova | 04:38 | |
| openstackgerrit | Tang Chen proposed openstack/nova: Add a compute API to trigger crash dump in instance. https://review.openstack.org/202605 | 04:45 |
| *** fawadkhaliq has joined #openstack-nova | 04:45 | |
| *** inteq has joined #openstack-nova | 04:46 | |
| *** fawadkhaliq has quit IRC | 04:47 | |
| *** lyanchih has joined #openstack-nova | 04:49 | |
| *** jerrygb has quit IRC | 04:49 | |
| *** jerrygb has joined #openstack-nova | 04:51 | |
| *** jerrygb has quit IRC | 04:51 | |
| *** jerrygb has joined #openstack-nova | 04:51 | |
| *** zhangjn has quit IRC | 04:52 | |
| *** thorst has joined #openstack-nova | 04:52 | |
| *** Jeffrey4l__ has quit IRC | 04:53 | |
| *** Jeffrey4l__ has joined #openstack-nova | 04:58 | |
| *** thorst has quit IRC | 04:58 | |
| *** gokrokve has joined #openstack-nova | 05:01 | |
| *** lyanchih has quit IRC | 05:03 | |
| *** boris-42_ has quit IRC | 05:03 | |
| *** zhenq has joined #openstack-nova | 05:05 | |
| *** zenoway has joined #openstack-nova | 05:07 | |
| *** smoriya_ has joined #openstack-nova | 05:09 | |
| *** smoriya has quit IRC | 05:10 | |
| *** klkumar has quit IRC | 05:11 | |
| *** zenoway has quit IRC | 05:12 | |
| *** ljxiash has quit IRC | 05:14 | |
| *** ljxiash has joined #openstack-nova | 05:14 | |
| *** Sukhdev has joined #openstack-nova | 05:15 | |
| *** klkumar has joined #openstack-nova | 05:15 | |
| *** zhangjn has joined #openstack-nova | 05:15 | |
| *** ljxiash has quit IRC | 05:16 | |
| *** ljxiash has joined #openstack-nova | 05:16 | |
| *** dims has joined #openstack-nova | 05:20 | |
| openstackgerrit | Eric Brown proposed openstack/nova: Fix config option types https://review.openstack.org/168481 | 05:26 |
| *** klkumar has quit IRC | 05:26 | |
| *** rook has quit IRC | 05:26 | |
| *** zhangjn has quit IRC | 05:28 | |
| *** Marga_ has quit IRC | 05:28 | |
| *** gokrokve_ has joined #openstack-nova | 05:28 | |
| *** gokrokve has quit IRC | 05:28 | |
| *** zhangjn has joined #openstack-nova | 05:28 | |
| *** klkumar has joined #openstack-nova | 05:32 | |
| *** fawadkhaliq has joined #openstack-nova | 05:33 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Adds host_capabilities to HostState https://review.openstack.org/254747 | 05:33 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: WIP: scheduler: Adds host_capabilities filter https://review.openstack.org/255073 | 05:33 |
| *** armax has joined #openstack-nova | 05:38 | |
| *** zenoway has joined #openstack-nova | 05:43 | |
| *** yarkot has quit IRC | 05:45 | |
| *** lyanchih has joined #openstack-nova | 05:47 | |
| *** zenoway has quit IRC | 05:47 | |
| *** suro-patz has quit IRC | 05:50 | |
| *** suro-patz has joined #openstack-nova | 05:51 | |
| *** gokrokve_ has quit IRC | 05:51 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Add separated section for extra specs and image properties https://review.openstack.org/253512 | 05:51 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add separated section for configure guest os https://review.openstack.org/253513 | 05:51 |
| *** rook has joined #openstack-nova | 05:52 | |
| *** gokrokve has joined #openstack-nova | 05:52 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: Adds spec for exposing host capabilities https://review.openstack.org/222200 | 05:53 |
| *** suro-patz has quit IRC | 05:53 | |
| gongysh | alex_xu_, hi | 05:55 |
| alex_xu_ | gongysh: hi | 05:56 |
| *** thorst has joined #openstack-nova | 05:56 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Restructure api sample files https://review.openstack.org/255080 | 05:56 |
| *** salv-orl_ has quit IRC | 06:00 | |
| *** rohit has joined #openstack-nova | 06:00 | |
| *** salv-orlando has joined #openstack-nova | 06:00 | |
| *** tbh has quit IRC | 06:01 | |
| *** rohit has quit IRC | 06:02 | |
| *** thorst has quit IRC | 06:04 | |
| *** ljxiash has quit IRC | 06:05 | |
| *** ljxiash has joined #openstack-nova | 06:05 | |
| *** vilobhmm11 has quit IRC | 06:06 | |
| *** shuquan has joined #openstack-nova | 06:07 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/252197 | 06:08 |
| *** ljxiash has quit IRC | 06:08 | |
| *** ljxiash_ has joined #openstack-nova | 06:08 | |
| openstackgerrit | gongysh proposed openstack/nova: Fix the duplicate nics for VMs https://review.openstack.org/252824 | 06:10 |
| *** Marga_ has joined #openstack-nova | 06:12 | |
| *** Marga_ has quit IRC | 06:14 | |
| *** Marga_ has joined #openstack-nova | 06:14 | |
| *** Marga_ has quit IRC | 06:15 | |
| *** Marga_ has joined #openstack-nova | 06:15 | |
| *** jerrygb has quit IRC | 06:17 | |
| *** jerrygb has joined #openstack-nova | 06:17 | |
| *** tbh has joined #openstack-nova | 06:17 | |
| *** salv-orl_ has joined #openstack-nova | 06:17 | |
| *** unicell has joined #openstack-nova | 06:18 | |
| *** zenoway has joined #openstack-nova | 06:18 | |
| *** salv-orlando has quit IRC | 06:20 | |
| *** jerrygb has quit IRC | 06:21 | |
| *** zenoway has quit IRC | 06:23 | |
| *** logan- has quit IRC | 06:24 | |
| *** alex_xu has joined #openstack-nova | 06:25 | |
| *** ijw has joined #openstack-nova | 06:26 | |
| *** aginwala has joined #openstack-nova | 06:28 | |
| *** alex_xu_ has quit IRC | 06:28 | |
| *** ljxiash_ has quit IRC | 06:29 | |
| *** ljxiash has joined #openstack-nova | 06:29 | |
| *** ijw has quit IRC | 06:30 | |
| *** tbh has quit IRC | 06:31 | |
| *** zhenq has quit IRC | 06:31 | |
| *** irenab has joined #openstack-nova | 06:31 | |
| *** jaypipes has joined #openstack-nova | 06:31 | |
| openstackgerrit | Rui Chen proposed openstack/nova-specs: Task states improvement https://review.openstack.org/216578 | 06:31 |
| *** gokrokve has quit IRC | 06:32 | |
| *** ijw has joined #openstack-nova | 06:32 | |
| *** esracelik has joined #openstack-nova | 06:33 | |
| *** markmc has quit IRC | 06:33 | |
| gmann | sdague: alex_xu : oomichi : for api samples dir structure refactor - https://review.openstack.org/#/c/255080/1 | 06:33 |
| *** markmc has joined #openstack-nova | 06:33 | |
| gmann | please have a look and let me know if thats look better | 06:33 |
| *** ljxiash has quit IRC | 06:34 | |
| oomichi | it seems easy to know difference between versions on quick review | 06:35 |
| oomichi | gmann: but it is huge change, and I'd like to take more time for stable steps | 06:35 |
| oomichi | gmann: for reviewing | 06:35 |
| gmann | oomichi: Thanks. please take time. yea stable structure will be nice. | 06:37 |
| *** ljxiash has joined #openstack-nova | 06:37 | |
| *** lpetrut has joined #openstack-nova | 06:37 | |
| *** aginwala has quit IRC | 06:37 | |
| *** ijw_ has joined #openstack-nova | 06:38 | |
| *** ijw has quit IRC | 06:38 | |
| *** spandhe has joined #openstack-nova | 06:41 | |
| *** ijw_ has quit IRC | 06:41 | |
| *** ijw has joined #openstack-nova | 06:42 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Only return latest fault of each instance in db api https://review.openstack.org/223392 | 06:42 |
| *** RuiChen has joined #openstack-nova | 06:43 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Create image for suspended instance booted from volume https://review.openstack.org/223382 | 06:43 |
| *** tbh has joined #openstack-nova | 06:43 | |
| openstackgerrit | Rui Chen proposed openstack/python-novaclient: Fix Resource.__eq__ mismatch semantics of object equal https://review.openstack.org/251020 | 06:44 |
| *** aginwala has joined #openstack-nova | 06:47 | |
| *** armax has quit IRC | 06:48 | |
| *** lpetrut has quit IRC | 06:48 | |
| *** zenoway has joined #openstack-nova | 06:48 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: config options: centralize section: "keymgr" https://review.openstack.org/255044 | 06:51 |
| *** irenab has quit IRC | 06:51 | |
| alex_xu | gmann: thanks, will take a look at | 06:51 |
| gmann | alex_xu: Thanks | 06:52 |
| alex_xu | gmann: after 2.13, we will test both with project_id and without project_id? | 06:53 |
| alex_xu | gmann: or we say before 2.13 is with project_id, and after that is without project_id | 06:54 |
| gmann | alex_xu: yea, current tests with project_id and 2.13 another scenario to tests without project_id | 06:54 |
| gmann | alex_xu: as we discussed yesterday we need to maintain both sample file as removal of project_id is with microversion | 06:55 |
| *** ijw has quit IRC | 06:55 | |
| gmann | alex_xu: so we need to keep with project_id sample files also for <2.13 users | 06:55 |
| alex_xu | gmann: thanks, I see now | 06:55 |
| *** nkrinner has joined #openstack-nova | 06:56 | |
| alex_xu | gmann: so after 2.13, we only test without project_id | 06:56 |
| *** zenoway has quit IRC | 06:56 | |
| *** zenoway has joined #openstack-nova | 06:56 | |
| gmann | alex_xu: with both. sdague has patch up https://review.openstack.org/#/c/254401/ | 06:56 |
| gmann | alex_xu: adding another scneario for 2.13 | 06:57 |
| alex_xu | gmann: ah, ok. so we will have two sets api sample file for 2.13 | 06:57 |
| gmann | alex_xu: i also need to look into that | 06:57 |
| gmann | alex_xu: no, for 2.13 we will have one set which will be without project_id and we will keep current set of sample files (which are with project_id) for <2.13 | 06:58 |
| alex_xu | gmann: ok, I think we got you this time, thanks :) | 06:59 |
| alex_xu | s/we got/i got/ | 06:59 |
| gmann | alex_xu: :) | 06:59 |
| gmann | alex_xu: actually i prefer to maintain only one (without project id) but sdague has valid point to keep old one for <2.12 versions testing | 07:00 |
| *** claudiub has quit IRC | 07:00 | |
| *** jerrygb has joined #openstack-nova | 07:00 | |
| *** pratikmallya has joined #openstack-nova | 07:00 | |
| *** zenoway has quit IRC | 07:01 | |
| gmann | alex_xu: but 2.13 onwards we will have sample files of new versions without project id | 07:01 |
| *** thorst has joined #openstack-nova | 07:01 | |
| alex_xu | gmann: yes, I'm not sure regax validation works for just maintain only one set sample file. | 07:01 |
| gmann | alex_xu: with project_id sample files we need just for <2.13 versions testing | 07:01 |
| *** unicell has quit IRC | 07:01 | |
| alex_xu | gmann: emm...i got you | 07:02 |
| *** gokrokve has joined #openstack-nova | 07:02 | |
| gmann | alex_xu: you mean project_id check in sample files for <2.13 and for 2.13 and onwards do not check ? | 07:03 |
| *** spandhe_ has joined #openstack-nova | 07:03 | |
| alex_xu | gmann: just use regax to check with or without project_id both are valid value. but I think sdague want to test this two different cases | 07:04 |
| alex_xu | but I need dig into more | 07:04 |
| *** unicell has joined #openstack-nova | 07:04 | |
| *** spandhe has quit IRC | 07:04 | |
| *** spandhe_ is now known as spandhe | 07:04 | |
| esracelik | hi all, I am new and have a question. what happens if a group is working on a blueprint at the same time and after every successful merge others on review get merge conflict? | 07:04 |
| *** mpavone has joined #openstack-nova | 07:04 | |
| gmann | alex_xu: i see your point, thats seems nice. | 07:04 |
| gmann | alex_xu: yea i will also look moreb into sdague patch | 07:05 |
| alex_xu | gmann: anyway let me think about, I need dig into more code, then I can get more clear mind | 07:05 |
| gmann | moreb->more | 07:05 |
| *** cfriesen has quit IRC | 07:05 | |
| gmann | alex_xu: great thanks | 07:05 |
| alex_xu | gmann: yeh, cool, thanks | 07:05 |
| *** jerrygb has quit IRC | 07:06 | |
| *** gokrokve has quit IRC | 07:07 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/253901 | 07:08 |
| *** thorst has quit IRC | 07:09 | |
| *** suro-patz has joined #openstack-nova | 07:13 | |
| *** zenoway has joined #openstack-nova | 07:15 | |
| *** mpavone has quit IRC | 07:15 | |
| *** mjura has joined #openstack-nova | 07:16 | |
| *** jaosorior has joined #openstack-nova | 07:19 | |
| *** zenoway has quit IRC | 07:19 | |
| *** jaosorior has quit IRC | 07:19 | |
| *** spandhe has quit IRC | 07:20 | |
| *** zenoway has joined #openstack-nova | 07:20 | |
| *** jaosorior has joined #openstack-nova | 07:20 | |
| *** rcernin has joined #openstack-nova | 07:20 | |
| *** zenoway has quit IRC | 07:24 | |
| *** rook has quit IRC | 07:26 | |
| *** paul-carlton2 has joined #openstack-nova | 07:29 | |
| *** sahid has joined #openstack-nova | 07:30 | |
| *** irenab has joined #openstack-nova | 07:30 | |
| *** yonglihe has quit IRC | 07:31 | |
| *** boris-42_ has joined #openstack-nova | 07:32 | |
| *** scheuran has joined #openstack-nova | 07:32 | |
| paul-carlton2 | Anyone got time to review https://review.openstack.org/#/c/220622 pls | 07:33 |
| *** zhangjn has quit IRC | 07:36 | |
| *** alexschm has joined #openstack-nova | 07:38 | |
| *** meh_di has joined #openstack-nova | 07:39 | |
| *** klkumar has quit IRC | 07:41 | |
| *** klkumar has joined #openstack-nova | 07:44 | |
| *** pratikmallya has quit IRC | 07:45 | |
| *** nkrinner has quit IRC | 07:46 | |
| *** nkrinner has joined #openstack-nova | 07:46 | |
| *** rhagarty_ has joined #openstack-nova | 07:48 | |
| *** rhagarty has quit IRC | 07:49 | |
| *** cdent has joined #openstack-nova | 07:50 | |
| *** lyanchih has quit IRC | 07:52 | |
| *** lyanchih has joined #openstack-nova | 07:52 | |
| *** rook has joined #openstack-nova | 07:52 | |
| *** e0ne has joined #openstack-nova | 07:53 | |
| *** aginwala has quit IRC | 07:53 | |
| *** nkrinner has quit IRC | 07:54 | |
| *** nkrinner has joined #openstack-nova | 07:55 | |
| *** pece has joined #openstack-nova | 07:56 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 07:58 |
| *** moshele has joined #openstack-nova | 08:00 | |
| *** zhangjn has joined #openstack-nova | 08:00 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 08:02 |
| *** matrohon has joined #openstack-nova | 08:03 | |
| *** gokrokve has joined #openstack-nova | 08:03 | |
| *** pece has quit IRC | 08:04 | |
| *** thorst has joined #openstack-nova | 08:06 | |
| *** matrohon has quit IRC | 08:07 | |
| *** gokrokve has quit IRC | 08:08 | |
| *** suro-patz has quit IRC | 08:10 | |
| *** RuiChen has quit IRC | 08:12 | |
| *** RuiChen has joined #openstack-nova | 08:13 | |
| *** aginwala has joined #openstack-nova | 08:14 | |
| openstackgerrit | esracelik proposed openstack/nova: Config options: centralize section "hyperv" https://review.openstack.org/255124 | 08:14 |
| *** thorst has quit IRC | 08:14 | |
| *** liverpooler has joined #openstack-nova | 08:15 | |
| bauzas | good morning Nova | 08:15 |
| *** belmoreira has joined #openstack-nova | 08:15 | |
| *** paul-carlton2 has quit IRC | 08:17 | |
| *** klkumar has quit IRC | 08:17 | |
| *** paul-carlton2 has joined #openstack-nova | 08:19 | |
| *** jichen has joined #openstack-nova | 08:20 | |
| *** amotoki has joined #openstack-nova | 08:22 | |
| *** gjayavelu has joined #openstack-nova | 08:23 | |
| *** ljxiash has quit IRC | 08:23 | |
| *** ljxiash has joined #openstack-nova | 08:23 | |
| *** shuquan_ has joined #openstack-nova | 08:23 | |
| *** markus_z has joined #openstack-nova | 08:23 | |
| cdent | good morning bauzas. I made some changes to the filter scheduler doc that probably could do with your review (I think it was probably you're writing I was updating): https://review.openstack.org/#/c/254394/ | 08:23 |
| *** shuquan has quit IRC | 08:24 | |
| *** shuquan_ is now known as shuquan | 08:24 | |
| bauzas | cdent: yeah, I saw your changes :) | 08:24 |
| bauzas | cdent: btw. welcome to the nova team ! | 08:24 |
| cdent | thanks | 08:24 |
| cdent | happy to be here, etc | 08:24 |
| bauzas | cdent: good to see you again | 08:25 |
| bauzas | cdent: so, CC'ing myself on ^ | 08:25 |
| cdent | we have a strange habit of bumping into each other :) | 08:25 |
| bauzas | hah :) | 08:25 |
| bauzas | cdent: tbh, I'm pretty good with your change, I just feel something needs to be done with that rst file | 08:28 |
| bauzas | cdent: most of the time, people need to review the implementation to know how the filter works | 08:28 |
| bauzas | cdent: I'd love to provide some better explanation of the conditionals etc. directly in the doc | 08:29 |
| *** shuquan has quit IRC | 08:29 | |
| cdent | Yeah, I considered trying to make more substantial changes, but I didn't want to bite off too much in one go. It was simply a case of "I'm in here reading this stuff closely now, may as well fix the typos and such that I find" | 08:29 |
| bauzas | cdent: in general, I love to iterate quickly on docs, so I should +2 the above very quickly unless I'm seeing some big mistake | 08:29 |
| cdent | I'm happy to go back to it again later after I've learned more about the scheduler and give it a real tune up | 08:30 |
| *** smoriya_ has quit IRC | 08:31 | |
| openstackgerrit | Rui Chen proposed openstack/nova-specs: Task states improvement https://review.openstack.org/216578 | 08:32 |
| *** fawadkhaliq has quit IRC | 08:32 | |
| openstackgerrit | lyanchih proposed openstack/nova: [db] Instances named invalid regexp string can't been deleted https://review.openstack.org/255130 | 08:37 |
| *** aginwala has quit IRC | 08:37 | |
| openstackgerrit | dstepanenko proposed openstack/nova: WIP: This is 2nd part of changes according to pci-generate-stats blueprint. https://review.openstack.org/243131 | 08:39 |
| *** PaulMurray has joined #openstack-nova | 08:39 | |
| openstackgerrit | dstepanenko proposed openstack/nova: Moved logic from nova.pci.stats.PciDevStats.consume_requests() to the nova.pci.manager.PciDevTracker._claim_for_instance() method using pci devices instead of pci pools. https://review.openstack.org/243131 | 08:42 |
| openstackgerrit | dstepanenko proposed openstack/nova: WIP: This is 3rd part of changes according to pci-generate-stats https://review.openstack.org/247692 | 08:44 |
| *** jlanoux has joined #openstack-nova | 08:46 | |
| *** matrohon has joined #openstack-nova | 08:46 | |
| *** Sukhdev has quit IRC | 08:47 | |
| *** fawadkhaliq has joined #openstack-nova | 08:48 | |
| *** fawadkhaliq has quit IRC | 08:49 | |
| *** MarcF has joined #openstack-nova | 08:50 | |
| ZZelle | johnthetubaguy, hi | 08:52 |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 08:52 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (virt_type) https://review.openstack.org/252832 | 08:54 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (inject_data) https://review.openstack.org/253439 | 08:54 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (rescue) https://review.openstack.org/251987 | 08:54 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: centralize libvirt options https://review.openstack.org/250907 | 08:54 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: config options help text (use_usb_tablet) https://review.openstack.org/255133 | 08:54 |
| *** rdopiera has joined #openstack-nova | 08:54 | |
| *** jaosorior has quit IRC | 08:55 | |
| *** salv-orl_ has quit IRC | 08:58 | |
| *** paul-carlton2 has quit IRC | 08:59 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: config options help text (use_usb_tablet) https://review.openstack.org/255133 | 08:59 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (virt_type) https://review.openstack.org/252832 | 08:59 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (inject_data) https://review.openstack.org/253439 | 08:59 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: help text for config options (rescue) https://review.openstack.org/251987 | 08:59 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: centralize libvirt options https://review.openstack.org/250907 | 08:59 |
| *** c00281451_ is now known as cheneng | 09:00 | |
| *** cheneng is now known as chenzeng | 09:00 | |
| *** yassine__ has joined #openstack-nova | 09:01 | |
| *** yamahata has joined #openstack-nova | 09:01 | |
| *** wyang has quit IRC | 09:02 | |
| *** jerrygb has joined #openstack-nova | 09:02 | |
| *** sahid has quit IRC | 09:03 | |
| *** gokrokve has joined #openstack-nova | 09:04 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 09:04 |
| alex_xu | sdague: johnthetubaguy oomichi ^ | 09:04 |
| *** MarcF has quit IRC | 09:05 | |
| *** openstackstatus has quit IRC | 09:05 | |
| *** openstack has joined #openstack-nova | 09:07 | |
| *** ssurana has quit IRC | 09:07 | |
| *** jerrygb has quit IRC | 09:08 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: config options: centralize section: "keymgr" https://review.openstack.org/255044 | 09:08 |
| alex_xu | tangchen: will try tomorrow | 09:09 |
| *** gokrokve has quit IRC | 09:09 | |
| tangchen | alex_xu: thanks. :) | 09:10 |
| alex_xu | tangchen: np | 09:10 |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 09:10 |
| *** thorst has joined #openstack-nova | 09:12 | |
| *** ljxiash has quit IRC | 09:12 | |
| *** danpb has joined #openstack-nova | 09:13 | |
| *** ljxiash has joined #openstack-nova | 09:13 | |
| *** ociuhandu has quit IRC | 09:14 | |
| *** yamahata has quit IRC | 09:15 | |
| *** e0ne has quit IRC | 09:15 | |
| *** daemontool has joined #openstack-nova | 09:16 | |
| *** jaosorior has joined #openstack-nova | 09:16 | |
| *** wyang has joined #openstack-nova | 09:17 | |
| *** ljxiash has quit IRC | 09:17 | |
| *** thorst has quit IRC | 09:18 | |
| *** ljxiash has joined #openstack-nova | 09:21 | |
| *** obondarev has quit IRC | 09:23 | |
| openstackgerrit | esracelik proposed openstack/nova: Config options: centralize "configdrive" options https://review.openstack.org/254092 | 09:24 |
| *** ndipanov has joined #openstack-nova | 09:27 | |
| *** terryyao has quit IRC | 09:30 | |
| *** rook has quit IRC | 09:30 | |
| *** garyk1 has joined #openstack-nova | 09:31 | |
| *** openstackgerrit has quit IRC | 09:32 | |
| *** openstackgerrit has joined #openstack-nova | 09:33 | |
| johnthetubaguy | alex_xu: awesome, thanks for getting that up there, would be cool to get PaulMurray and others to take a peak at that too | 09:33 |
| johnthetubaguy | alex_xu: been going through the old v2 API ref docs on the etherpad, I think we missed a few things in the move to v2.1 | 09:33 |
| *** fawadkhaliq has joined #openstack-nova | 09:34 | |
| *** lyanchih has quit IRC | 09:36 | |
| *** lyanchih has joined #openstack-nova | 09:37 | |
| bauzas | alex_xu: you can use that one http://docs.openstack.org/developer/nova/code-review.html#when-a-release-note-is-needed for knowing | 09:37 |
| *** sudipto has joined #openstack-nova | 09:38 | |
| bauzas | alex_xu: tl;dr I think adding a features section is good for me | 09:39 |
| *** Kevin_Zheng has quit IRC | 09:40 | |
| *** matrohon has quit IRC | 09:41 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add project-id and user-id when list server-groups https://review.openstack.org/209018 | 09:41 |
| *** ildikov has quit IRC | 09:42 | |
| *** zhangjn has quit IRC | 09:45 | |
| *** zenoway has joined #openstack-nova | 09:48 | |
| *** zenoway has quit IRC | 09:53 | |
| *** zenoway has joined #openstack-nova | 09:53 | |
| *** lyanchih has quit IRC | 09:53 | |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Added support for new block device format in Hyper-V https://review.openstack.org/246298 | 09:53 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Added support for new block device format in vmops https://review.openstack.org/246299 | 09:54 |
| *** oomichi is now known as oomichi_away | 09:55 | |
| *** rook has joined #openstack-nova | 09:56 | |
| *** lyanchih has joined #openstack-nova | 09:56 | |
| *** zhangjn has joined #openstack-nova | 09:57 | |
| *** salv-orlando has joined #openstack-nova | 09:59 | |
| *** sahid has joined #openstack-nova | 10:00 | |
| *** obondarev has joined #openstack-nova | 10:02 | |
| *** openstackgerrit has quit IRC | 10:02 | |
| *** gcb has quit IRC | 10:02 | |
| *** zhenguo has quit IRC | 10:02 | |
| *** gszasz has joined #openstack-nova | 10:03 | |
| *** openstackgerrit has joined #openstack-nova | 10:03 | |
| *** salv-orlando has quit IRC | 10:03 | |
| openstackgerrit | Tardis Xu proposed openstack/nova: Fix nova-scheduler loads all instance at startup https://review.openstack.org/255178 | 10:05 |
| *** sfinucan has joined #openstack-nova | 10:05 | |
| *** ildikov has joined #openstack-nova | 10:09 | |
| openstackgerrit | Tardis Xu proposed openstack/nova: Fix nova-scheduler loads all instance at startup https://review.openstack.org/255178 | 10:10 |
| *** e0ne has joined #openstack-nova | 10:10 | |
| *** jamielennox is now known as jamielennox|away | 10:13 | |
| *** ljxiash has quit IRC | 10:13 | |
| *** ljxiash has joined #openstack-nova | 10:14 | |
| *** garyk1 has quit IRC | 10:14 | |
| *** garyk1 has joined #openstack-nova | 10:14 | |
| *** sileht has quit IRC | 10:16 | |
| *** pece has joined #openstack-nova | 10:16 | |
| markus_z | BobBall: Are you the right person when I have questions about the check-citrix-xenserver CI? | 10:18 |
| *** thorst has joined #openstack-nova | 10:18 | |
| *** ljxiash has quit IRC | 10:18 | |
| *** sileht has joined #openstack-nova | 10:19 | |
| BobBall | markus_z: uhhhh uhhh uhhhh... well uhhhh that depends. | 10:19 |
| *** sileht has quit IRC | 10:19 | |
| BobBall | Can you tell me the question then I'll let you know if I'm the right person or if you should write to /dev/null? :) | 10:20 |
| markus_z | BobBall: Let's give it a try :) http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/32/252832/4/18724/logs/screen-n-cpu.txt.gz | 10:20 |
| *** sileht has joined #openstack-nova | 10:20 | |
| *** aix has joined #openstack-nova | 10:20 | |
| markus_z | BobBall: oh maann, nevermind, I found the issue | 10:20 |
| BobBall | oh good | 10:21 |
| BobBall | lol | 10:21 |
| markus_z | right when I wanted to start complaining ;) | 10:21 |
| *** rushiagr_away is now known as rushiagr | 10:21 | |
| BobBall | how does a change in libvirt cause a failure like that for the xenserver cI?? | 10:22 |
| markus_z | I'm not 100% sure yet. I'm shuffeling the config options and nova-conductor has an issue with that. | 10:24 |
| markus_z | http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/32/252832/4/18724/logs/screen-n-cond.txt.gz | 10:24 |
| *** thorst has quit IRC | 10:24 | |
| markus_z | timestamp: 2015-12-09 09:25:51.578 | 10:24 |
| *** dims has quit IRC | 10:25 | |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: libvirt: Add ppc to supported arch for NUMA https://review.openstack.org/245765 | 10:25 |
| *** alexschm has quit IRC | 10:26 | |
| PaulMurray | johnthetubaguy, scroll back on my client has gone made - was there something in docs you wanted me to look at? | 10:27 |
| johnthetubaguy | PaulMurray: ah, yes, let me take a quick look | 10:27 |
| johnthetubaguy | PaulMurray: https://review.openstack.org/255122 | 10:27 |
| PaulMurray | johnthetubaguy, thanks - looking | 10:28 |
| *** alexschm has joined #openstack-nova | 10:29 | |
| *** ianbrown has joined #openstack-nova | 10:30 | |
| *** PaulMurray has quit IRC | 10:30 | |
| openstackgerrit | Tardis Xu proposed openstack/nova: Fix nova-scheduler loads all instance at startup https://review.openstack.org/255178 | 10:30 |
| *** lyanchih has quit IRC | 10:30 | |
| *** PaulMurray has joined #openstack-nova | 10:30 | |
| openstackgerrit | xiexs proposed openstack/nova: WIP: Add sriov info for live migration https://review.openstack.org/255193 | 10:31 |
| *** alexschm_ has joined #openstack-nova | 10:31 | |
| *** alexschm has quit IRC | 10:34 | |
| *** AJaeger has joined #openstack-nova | 10:37 | |
| *** aix has quit IRC | 10:37 | |
| AJaeger | nova cores, during translation of nova for Liberty, we noticed quite a few inconsistencies. I've created a change to fix some of that at https://review.openstack.org/#/c/225923/ | 10:38 |
| *** rk4n has joined #openstack-nova | 10:38 | |
| AJaeger | that change has not seen a review for over a month... | 10:38 |
| AJaeger | Is that something you like to have or should I abandon it since there's no interest? | 10:38 |
| AJaeger | johnthetubaguy, could you look at it, please? | 10:39 |
| johnthetubaguy | AJaeger: I can, if we attach a high priority bug to that, it will get more attention | 10:39 |
| *** jistr has joined #openstack-nova | 10:39 | |
| AJaeger | johnthetubaguy: shall I file a bug myself? | 10:40 |
| johnthetubaguy | AJaeger: that would be great, thank you | 10:40 |
| AJaeger | johnthetubaguy: Just want to know how to best continue with it... | 10:40 |
| AJaeger | on my way... | 10:40 |
| *** rk4n has quit IRC | 10:40 | |
| *** rk4n_ has joined #openstack-nova | 10:40 | |
| johnthetubaguy | AJaeger: I wonder if we want to make Floating always have a capital F? | 10:41 |
| markus_z | johnthetubaguy: Would the "trivial bug list" in the priorities tracking etherpad be useful for such changes from AJaeger ? | 10:41 |
| johnthetubaguy | markus_z: good point, it would | 10:42 |
| AJaeger | johnthetubaguy: https://bugs.launchpad.net/nova/+bug/1524276 | 10:42 |
| openstack | Launchpad bug 1524276 in OpenStack Compute (nova) "Inconsistent IP usage in help strings" [Undecided,New] | 10:42 |
| AJaeger | Will either of you add it to the etherpad - or shall I? IN that case, I need a link to it... | 10:43 |
| * AJaeger updates the review to mention the bug report. | 10:43 | |
| markus_z | AJaeger: It's already added | 10:43 |
| openstackgerrit | dstepanenko proposed openstack/nova: Modified the nova.objects.ComputeNode object to load on-demand the pci_device_pools field by a subquery instead of pulling from the compute_nodes.pci_stats field in the database. https://review.openstack.org/247692 | 10:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/nova: Fix capitalization of IP https://review.openstack.org/225923 | 10:44 |
| markus_z | AJaeger: It's this etherpad: https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking Just ping me when you have such a change. | 10:44 |
| AJaeger | thanks, markus_z ! | 10:45 |
| openstackgerrit | dstepanenko proposed openstack/nova: WIP: This is 4th patch of changes according to pci-generate-stats blueprint. https://review.openstack.org/250757 | 10:45 |
| *** PaulMurray has quit IRC | 10:46 | |
| *** PaulMurray has joined #openstack-nova | 10:47 | |
| johnthetubaguy | AJaeger: got a question on that patch, wondering about Floating vs floating? | 10:47 |
| openstackgerrit | dstepanenko proposed openstack/nova: Introduced PciDevicePoolList get_all method that gets pci_device_pools from the database using PciDevice table https://review.openstack.org/250757 | 10:48 |
| AJaeger | johnthetubaguy: I commented. Documentation team uses lowercase | 10:48 |
| AJaeger | johnthetubaguy: You marked one change where it's the first word of the string, there we could uppercase it, since a sentence starts with uppercase. | 10:49 |
| johnthetubaguy | AJaeger: yeah, there are those API error code things | 10:49 |
| johnthetubaguy | AJaeger: good to know thats the choice that was made | 10:49 |
| AJaeger | yep | 10:50 |
| AJaeger | johnthetubaguy: thanks for the review | 10:50 |
| AJaeger | markus_z: thanks to you as well... | 10:51 |
| markus_z | AJaeger: you're welcome | 10:52 |
| openstackgerrit | jichenjc proposed openstack/nova: Change some wording on server_concepts.rst https://review.openstack.org/248063 | 10:54 |
| *** rdopiera has quit IRC | 10:55 | |
| *** yassine__ has quit IRC | 10:56 | |
| *** fawadkhaliq has quit IRC | 10:57 | |
| *** fawadkhaliq has joined #openstack-nova | 10:57 | |
| *** rdopiera has joined #openstack-nova | 10:59 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: trivial: Add additional logs for NUMA scheduling https://review.openstack.org/254783 | 11:01 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Make use of 'InstanceNUMACell.cpu_policy' field https://review.openstack.org/202653 | 11:01 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'hw:cpu_threads_policy=isolate' scheduling https://review.openstack.org/202655 | 11:01 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'cpu_policy' and 'cpu_thread_policy' fields https://review.openstack.org/202648 | 11:01 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'hw:cpu_thread_policy=require' scheduling https://review.openstack.org/202651 | 11:01 |
| alex_xu | johnthetubaguy: yeh, will catch live-migration guys when they wake up | 11:01 |
| alex_xu | bauzas: got it, thanks | 11:01 |
| alex_xu | johnthetubaguy: which bits we missed? | 11:01 |
| johnthetubaguy | alex_xu: put them on the etherpad, os-limits has an extra tenant param, and os-volume_attachments (and possible os-volumes_boot, but that seems odd) http://developer.openstack.org/api-ref-compute-v2-ext.html#os-volume_attachments | 11:02 |
| *** jyuso1 has joined #openstack-nova | 11:02 | |
| johnthetubaguy | alex_xu: I am just preping the patches for those now | 11:02 |
| *** fawadkhaliq has quit IRC | 11:03 | |
| alex_xu | johnthetubaguy: ok, thanks for the patches | 11:03 |
| *** jerrygb has joined #openstack-nova | 11:04 | |
| johnthetubaguy | I need to update the topic to match, but will do that in a bit | 11:04 |
| alex_xu | johnthetubaguy: the tenant_id param? | 11:04 |
| johnthetubaguy | yeah https://review.openstack.org/255195 | 11:05 |
| *** kmartin has quit IRC | 11:05 | |
| alex_xu | johnthetubaguy: the 'tenant_id' means the project_id in the url I think | 11:05 |
| johnthetubaguy | alex_xu: there are two in this case, also has a query string parameter | 11:05 |
| *** kmartin has joined #openstack-nova | 11:05 | |
| johnthetubaguy | alex_xu: http://developer.openstack.org/api-ref-compute-v2-ext.html#os-used-limits-for-admins | 11:06 |
| johnthetubaguy | alex_xu: we totally need to add some redirects to stop these old pages still working... I think bauzas knows how that magic works | 11:06 |
| alex_xu | johnthetubaguy: yea, the old api ref still in the memu | 11:08 |
| johnthetubaguy | alex_xu: it not in the menu any more, its just old pages never get deleted, for reasons I don't really understand | 11:09 |
| *** jerrygb has quit IRC | 11:09 | |
| *** salv-orlando has joined #openstack-nova | 11:10 | |
| alex_xu | johnthetubaguy: emm..ture, i'm curious where did you find the old ref entry | 11:12 |
| alex_xu | johnthetubaguy: I think I can file a bug, then see doc team whether can help us | 11:12 |
| johnthetubaguy | alex_xu: I think bauzas knows how to add that redirect, we hit the same issue with deverf | 11:13 |
| johnthetubaguy | devref | 11:13 |
| alex_xu | johnthetubaguy: ok | 11:13 |
| *** aix has joined #openstack-nova | 11:14 | |
| *** aysyd has joined #openstack-nova | 11:14 | |
| *** lucas-dinner is now known as lucasagomes | 11:16 | |
| *** aysyd has quit IRC | 11:16 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce freeze filesystems https://review.openstack.org/249140 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: check graphic/serial consoles at pre_live_migration step https://review.openstack.org/254709 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce thaw filesystems https://review.openstack.org/249141 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: replace direct libvirt's call AbortJobBlock https://review.openstack.org/249138 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: return a root block device https://review.openstack.org/249139 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce "snapshot" to Guest's object https://review.openstack.org/249353 | 11:17 |
| *** markmc has quit IRC | 11:17 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce block_stats to return device stats https://review.openstack.org/251288 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce "pause" to Guest's object https://review.openstack.org/249834 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: wrapper list_guests to Host's object https://review.openstack.org/249835 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce "shutdown" to Guest's object https://review.openstack.org/249833 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce get_job_info to Guest's object https://review.openstack.org/250740 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: configure migrate max downtime https://review.openstack.org/250739 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce "abort_job" to Guest's object https://review.openstack.org/250738 | 11:17 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce "migrate" to Guest's object https://review.openstack.org/250365 | 11:17 |
| *** aysyd has joined #openstack-nova | 11:18 | |
| *** markmc has joined #openstack-nova | 11:19 | |
| sdague | alex_xu: that spec update is awesome, and really a model for what I'd like api changing specs to look like | 11:23 |
| *** zhangjn has quit IRC | 11:23 | |
| *** thorst has joined #openstack-nova | 11:23 | |
| alex_xu | sdague: :) | 11:23 |
| *** EinstCrazy has quit IRC | 11:24 | |
| noqa_v_qoovnie | hello | 11:25 |
| noqa_v_qoovnie | i've got a question about the nature of nova | 11:25 |
| noqa_v_qoovnie | and aggregation groups and AZones | 11:25 |
| noqa_v_qoovnie | first of all, is there a way to create AZ w/o aggregation group at first? | 11:26 |
| *** links has quit IRC | 11:26 | |
| *** thorst has quit IRC | 11:28 | |
| *** klkumar has joined #openstack-nova | 11:28 | |
| *** moshele has quit IRC | 11:30 | |
| *** paul-carlton2 has joined #openstack-nova | 11:31 | |
| *** eglynn has joined #openstack-nova | 11:32 | |
| *** ljxiash has joined #openstack-nova | 11:32 | |
| *** moshele has joined #openstack-nova | 11:33 | |
| gongysh | alex_xu, hi | 11:33 |
| gongysh | alex_xu, https://review.openstack.org/#/c/252824/ | 11:34 |
| gongysh | I have a response to your comment | 11:35 |
| *** gongysh has quit IRC | 11:35 | |
| *** smatzek has joined #openstack-nova | 11:36 | |
| *** dims has joined #openstack-nova | 11:37 | |
| *** kashyap_ has joined #openstack-nova | 11:40 | |
| *** PaulMurray has quit IRC | 11:42 | |
| openstackgerrit | Ryan Moore proposed openstack/nova: Add better help to rdp options https://review.openstack.org/253556 | 11:42 |
| openstackgerrit | Ryan Moore proposed openstack/nova: config options: centralize section: "rdp" https://review.openstack.org/252389 | 11:42 |
| *** rook has quit IRC | 11:44 | |
| lxsli | Anyone want to +A https://review.openstack.org/#/c/234659/? Has 2x+2 and +1 Jenkins | 11:44 |
| *** rushiagr is now known as rushiagr_away | 11:48 | |
| *** ndipanov has quit IRC | 11:48 | |
| *** ndipanov has joined #openstack-nova | 11:49 | |
| *** kmartin has quit IRC | 11:49 | |
| sdague | BobBall: is this a real xenserver fail on https://review.openstack.org/#/c/254828/3 ? | 11:49 |
| *** rushiagr_away is now known as rushiagr | 11:49 | |
| *** zhangjn has joined #openstack-nova | 11:49 | |
| sdague | lxsli: but it has a bunch of patches under it that aren't approved | 11:50 |
| *** gokrokve has joined #openstack-nova | 11:50 | |
| sdague | so hitting +A is somewhat pointless | 11:50 |
| *** vladikr has quit IRC | 11:50 | |
| lxsli | ah shoot, thanks for looking | 11:51 |
| sdague | sahid: I responded to comments on https://review.openstack.org/#/c/254214/4 - want to take a look? | 11:51 |
| *** EinstCrazy has joined #openstack-nova | 11:51 | |
| sdague | and the one above it | 11:51 |
| *** rushiagr is now known as rushiagr_away | 11:52 | |
| *** igordcard_ has joined #openstack-nova | 11:54 | |
| *** igordcard_ has quit IRC | 11:55 | |
| *** gokrokve has quit IRC | 11:55 | |
| *** yassine__ has joined #openstack-nova | 12:00 | |
| *** ljxiash has quit IRC | 12:01 | |
| *** rotbeard has joined #openstack-nova | 12:02 | |
| *** daemontool has quit IRC | 12:04 | |
| *** cdelatte has quit IRC | 12:05 | |
| *** takedakn has joined #openstack-nova | 12:11 | |
| *** igordcard_ has joined #openstack-nova | 12:12 | |
| *** igordcard_ has joined #openstack-nova | 12:14 | |
| BobBall | sdague: Presume you mean libvirt+xen fail? :) | 12:15 |
| BobBall | No - looks like a race condition to me | 12:15 |
| BobBall | tempest.thirdparty.boto.test_ec2_instance_run.InstanceRunTest.test_run_stop_terminate_instance_with_tags was the failure | 12:15 |
| *** links has joined #openstack-nova | 12:16 | |
| sdague | BobBall: ok, whew | 12:17 |
| *** salv-orl_ has joined #openstack-nova | 12:17 | |
| *** jyuso1 has quit IRC | 12:19 | |
| *** jyuso1 has joined #openstack-nova | 12:19 | |
| *** salv-orlando has quit IRC | 12:20 | |
| *** smoriya has joined #openstack-nova | 12:24 | |
| esracelik | markus_z: hi markus | 12:24 |
| markus_z | esracelik: Hi Esra | 12:24 |
| esracelik | markus_z: all unit tests are ok for my commits but I am getting jenkins failures from review.openstack.org do you any ideas? https://review.openstack.org/#/c/254092/ | 12:25 |
| esracelik | markus_z: and https://review.openstack.org/#/c/255124/ | 12:25 |
| markus_z | esracelik: I'm checking, one sec | 12:26 |
| *** thorst has joined #openstack-nova | 12:27 | |
| *** sudipto has quit IRC | 12:28 | |
| *** matrohon has joined #openstack-nova | 12:29 | |
| *** migi_ has joined #openstack-nova | 12:30 | |
| *** houming has joined #openstack-nova | 12:31 | |
| *** migi_ has quit IRC | 12:31 | |
| *** rushiagr_away is now known as rushiagr | 12:32 | |
| *** meh_di has left #openstack-nova | 12:32 | |
| *** thorst has quit IRC | 12:33 | |
| *** gcb has joined #openstack-nova | 12:33 | |
| markus_z | esracelik: It looks like messages from Rabbit got lost: http://logs.openstack.org/92/254092/4/check/gate-grenade-dsvm/b251f9b/logs/new/screen-n-cond.txt.gz?level=ERROR | 12:34 |
| *** matrohon has quit IRC | 12:34 | |
| markus_z | esracelik: I'm not sure why. | 12:34 |
| markus_z | esracelik: I'm checking if such message came up before at http://logstash.openstack.org | 12:35 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 12:35 |
| *** matrohon has joined #openstack-nova | 12:36 | |
| esracelik | markus_z: oh ok. anything I can do? | 12:36 |
| markus_z | esracelik: I recomment reading this ML post when you have some time: http://openstack.markmail.org/thread/piwfpukf6yemj2ra | 12:36 |
| *** thorst has joined #openstack-nova | 12:36 | |
| markus_z | esracelik: I'm still checking what the real issue is. | 12:36 |
| esracelik | markus_z: ok I am reading it now | 12:37 |
| *** fawadkhaliq has joined #openstack-nova | 12:38 | |
| kashyap_ | win close | 12:39 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 12:40 |
| *** kashyap_ has left #openstack-nova | 12:40 | |
| *** alejandrito has joined #openstack-nova | 12:40 | |
| *** ducttape_ has joined #openstack-nova | 12:40 | |
| *** Daisy has joined #openstack-nova | 12:41 | |
| *** jaosorior has quit IRC | 12:42 | |
| *** doug-fis_ is now known as doug-fish | 12:42 | |
| *** jaosorior has joined #openstack-nova | 12:42 | |
| *** terryyao has joined #openstack-nova | 12:43 | |
| *** boris-42_ has quit IRC | 12:43 | |
| *** salv-orl_ has quit IRC | 12:44 | |
| *** atuvenie has joined #openstack-nova | 12:45 | |
| *** thorst_ has joined #openstack-nova | 12:45 | |
| markus_z | Is anyone aware of issues with logstash? Executing a simple query doesn't end in any result but with a loading symbol without an end. | 12:46 |
| markus_z | esracelik: I assume your patch is fine, but the gate has issue. You will see a peak here: http://grafana.openstack.org/dashboard/db/tempest-failure-rate | 12:47 |
| *** cdelatte has joined #openstack-nova | 12:48 | |
| markus_z | esracelik: I interpret the peak as an issue. | 12:48 |
| *** takedakn1 has joined #openstack-nova | 12:48 | |
| *** thorst has quit IRC | 12:49 | |
| *** salv-orlando has joined #openstack-nova | 12:49 | |
| *** takedakn has quit IRC | 12:49 | |
| *** takedakn1 is now known as takedakn | 12:49 | |
| markus_z | esracelik: I usually go to the #openstack-infra channel and try to figure out if that observation is correct and known. | 12:49 |
| *** claudiub has joined #openstack-nova | 12:50 | |
| *** terryyao has quit IRC | 12:50 | |
| esracelik | markus_z: oh ok, and if I goes normal I comment recheck, right? | 12:50 |
| esracelik | *it goes normal | 12:51 |
| *** terryyao has joined #openstack-nova | 12:51 | |
| markus_z | esracelik: "recheck no bug" if it is an exceptional thing. | 12:51 |
| markus_z | esracelik: sometimes "elastic recheck" shows you that you might hit a known bug, then "recheck bug #nnnn" | 12:52 |
| markus_z | esracelik: The link about debugging the gate includes also explanation about how to write an "elastic recheck" query. | 12:53 |
| *** lyanchih has joined #openstack-nova | 12:53 | |
| *** jerrygb has joined #openstack-nova | 12:53 | |
| openstackgerrit | Zhihai Song proposed openstack/python-novaclient: Change the logic for the client to retrive resources https://review.openstack.org/246881 | 12:53 |
| notmorgan | sdague: is a followup patch to address your comments sufficient? | 12:54 |
| esracelik | markus_z: ok thanks, btw after every successful merge in nova/conf, we will get conflict because of nova/conf/_init_.py file. So will we be committing another patch resolving that conflict? | 12:54 |
| *** bauzas has quit IRC | 12:55 | |
| notmorgan | sdague: because i dislike rolling a new patch for relatively minor nits at the cost of re-review/full battery of tests again on a more significant changeset. | 12:55 |
| *** tbh has quit IRC | 12:55 | |
| *** amotoki_ has joined #openstack-nova | 12:56 | |
| markus_z | esracelik: yeah, I've noticed it this morning too. I thought git could resolve that. I don't know yet how to solve this :( | 12:57 |
| *** ducttape_ has quit IRC | 12:58 | |
| *** amotoki__ has joined #openstack-nova | 12:59 | |
| esracelik | markus_z: i guess only __init__.py causes the conflict, maybe for now we should do register_opts somewhere else | 12:59 |
| *** amotoki has quit IRC | 12:59 | |
| *** takedakn has quit IRC | 12:59 | |
| *** amotoki__ is now known as amotoki___ | 12:59 | |
| markus_z | esracelik: When I did the rebase this morning my local git complaint about the import statement at the beginning of the file. | 13:00 |
| openstackgerrit | javeme proposed openstack/python-novaclient: remove the default arguments "[]" https://review.openstack.org/254171 | 13:00 |
| markus_z | esracelik: I'm in a short meeting, right be back | 13:00 |
| *** garyk1 has quit IRC | 13:00 | |
| *** amotoki___ is now known as amotoki__ | 13:02 | |
| sdague | notmorgan: it should update the existing patch, we typically like to get this right the first time, and it will make it easier for additional folks to review | 13:02 |
| *** amotoki_ has quit IRC | 13:02 | |
| *** amotoki__ is now known as amotoki | 13:02 | |
| *** paul-carlton2 has quit IRC | 13:03 | |
| notmorgan | sdague: i would like to point out the decorator comment is really far into bikeshed land | 13:03 |
| notmorgan | sdague: imo | 13:03 |
| sdague | notmorgan: the old code used a decorator | 13:03 |
| moshele | ndipanov: can you review my answer to you comment in https://review.openstack.org/#/c/227160/ ? let me know if you agree | 13:04 |
| notmorgan | sdague: cargo cult programming is not a good reason to keep with a pattern. | 13:04 |
| sdague | so this is about making it a more minimal and reviewable change because it looks more like the old code | 13:04 |
| sdague | notmorgan: consistency is useful when you have 15,000 unit tests | 13:04 |
| esracelik | markus_z: maybe we can import directly from nova.conf.<newmodule> and call register_opts afterwards. and add a TODO message to merge these lines in __init__.py at the end | 13:04 |
| *** Daisy has quit IRC | 13:05 | |
| notmorgan | sdague: i'm not refusting but i also want to point out that causing another cycle of review/test is expensive for minor comments [it's why in keystone we often do followup patches *before* the original one is merged] | 13:05 |
| *** doug-fish has quit IRC | 13:06 | |
| sdague | right, so that's mostly not the culture here | 13:06 |
| *** ildikov has quit IRC | 13:07 | |
| ndipanov | moshele, well I'm not sure we want to do validation really | 13:07 |
| *** atuvenie_ has joined #openstack-nova | 13:08 | |
| ndipanov | just not stacktrace if the address is not correct | 13:08 |
| ndipanov | the check was basically there before your patch | 13:08 |
| ndipanov | and you completely remove it | 13:08 |
| ndipanov | so my proposal is just to keep it around vf case | 13:09 |
| moshele | ndipanov: I can add it, but I think that is was optimization to skip it normal pci, as the assumption was the PF with no VF is normal PCI | 13:10 |
| *** atuvenie has quit IRC | 13:11 | |
| moshele | ndipanov: so you just want to add len(fun_cap.device_addrs) != 0 to the vf section? | 13:12 |
| *** terryyao has quit IRC | 13:13 | |
| ndipanov | moshele, yeah | 13:13 |
| moshele | ndipanov: ok will do | 13:13 |
| ndipanov | moshele, as a reviewer I wonder why is this removed when it still seems like a vlaid thing to check for | 13:13 |
| *** claudiub_ has joined #openstack-nova | 13:14 | |
| *** claudiub has quit IRC | 13:14 | |
| johnthetubaguy | ndipanov: I was just checking out the API docs, do you know about this API at all? http://developer.openstack.org/api-ref-compute-v2-ext.html#os-volume_boot | 13:16 |
| esracelik | markus_z: i will be away for half an hour | 13:16 |
| ndipanov | johnthetubaguy, well | 13:17 |
| ndipanov | it's not an API as such | 13:17 |
| ndipanov | it's an extension | 13:17 |
| ndipanov | remember that that was once a way to communicate what is supported to the user | 13:17 |
| johnthetubaguy | ndipanov: right, just extends the server create call I am assuming? | 13:17 |
| ndipanov | It may not even do that... but we can check easily | 13:18 |
| johnthetubaguy | the tests are a bit confusing around that I guess: https://github.com/openstack/nova/blob/1635086529c43edcb69966c48c41b5fca2d1b26c/nova/tests/unit/api/openstack/compute/test_volumes.py#L176 | 13:18 |
| sdague | but it also creates a new resource? we could have extensions in the list that don't have new resources. | 13:18 |
| johnthetubaguy | sdague: yeah, trying to track that down | 13:19 |
| johnthetubaguy | ndipanov: https://github.com/openstack/nova/blob/3ef150322ce1e377c98b54777ac745281c6c5ece/nova/api/openstack/compute/volumes.py#L555 is where I was looking | 13:19 |
| ndipanov | so there it is | 13:20 |
| ndipanov | it does not add anything to the boot request though | 13:20 |
| johnthetubaguy | sdague: been checking what was in the v2.0-ext doc, but wasn't in the v2.1 doc, spotted two extra extensions | 13:20 |
| sdague | johnthetubaguy: gotcha | 13:20 |
| ndipanov | so we should probably support the resource | 13:20 |
| sdague | yeh, it would be nice to remove in a future microversion, as it seems mostly confusing to have | 13:21 |
| openstackgerrit | Morgan Fainberg proposed openstack/nova: Remove deprecated neutron auth options https://review.openstack.org/253792 | 13:21 |
| notmorgan | sdague: ^ | 13:21 |
| johnthetubaguy | sdague: ndipanov: yeah, I am thinking I might just "forget" to document this one | 13:22 |
| ndipanov | nice | 13:22 |
| johnthetubaguy | I spotted this one was missing, but this one seems OK: http://developer.openstack.org/api-ref-compute-v2-ext.html#os-volume_attachments | 13:22 |
| *** ildikov has joined #openstack-nova | 13:22 | |
| johnthetubaguy | didn't realise we had two ways to attach a volume, post boot, until I looked into these docs | 13:23 |
| ZZelle | johnthetubaguy, hi | 13:23 |
| *** doug-fish has joined #openstack-nova | 13:23 | |
| johnthetubaguy | ZZelle: hi | 13:23 |
| ZZelle | johnthetubaguy, which core devs can review https://review.openstack.org/186716 (vmware related change)? | 13:24 |
| sdague | johnthetubaguy - we require a reno for conf option deprecation, are we also doing one for the actual drop? | 13:25 |
| johnthetubaguy | ZZelle: if it has been reviewed by the vmware subteam, feel free to add it on the list in here: https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking | 13:25 |
| sdague | I'm still not entirely clear on that front | 13:25 |
| johnthetubaguy | sdague: I guess we should | 13:25 |
| ZZelle | johnthetubaguy, ok, thanks | 13:25 |
| sdague | notmorgan: you want to stick a reno on https://review.openstack.org/253792 ? That's something I missed on the last review, then I think we're good. | 13:26 |
| *** terryyao has joined #openstack-nova | 13:26 | |
| johnthetubaguy | sdague: we should probably update those code review guidelines so thats clearer: http://docs.openstack.org/developer/nova/code-review.html#when-a-release-note-is-needed | 13:26 |
| notmorgan | sdague: sure. will do shortly | 13:26 |
| *** fawadkhaliq has quit IRC | 13:27 | |
| ZZelle | garyk, hi | 13:28 |
| *** edmondsw has joined #openstack-nova | 13:36 | |
| *** takedakn has joined #openstack-nova | 13:37 | |
| *** terryyao has quit IRC | 13:38 | |
| ndipanov | johnthetubaguy, might be wort checking what novaclient does | 13:38 |
| ndipanov | johnthetubaguy, ok looks like it does not use th reasource so... | 13:38 |
| ndipanov | johnthetubaguy, not related: it would be really good to get reviews on this one https://review.openstack.org/#/c/229574/ | 13:39 |
| ndipanov | and the next one: https://review.openstack.org/#/c/229575 | 13:39 |
| ndipanov | it's a prerequisit for a feature which is done basically | 13:40 |
| ndipanov | and unless you review it johnthetubaguy no one else will | 13:40 |
| johnthetubaguy | ndipanov: that stuff is on my radar, just focusing on the API doc sprint today | 13:40 |
| *** baoli has joined #openstack-nova | 13:40 | |
| *** jyuso1 has quit IRC | 13:41 | |
| ndipanov | johnthetubaguy, thanks | 13:41 |
| *** baoli_ has joined #openstack-nova | 13:41 | |
| *** diogogmt has joined #openstack-nova | 13:41 | |
| openstackgerrit | Morgan Fainberg proposed openstack/nova: Remove deprecated neutron auth options https://review.openstack.org/253792 | 13:42 |
| notmorgan | sdague: ^ reno added | 13:42 |
| *** ljxiash has joined #openstack-nova | 13:44 | |
| *** eharney has joined #openstack-nova | 13:44 | |
| *** smatzek has quit IRC | 13:44 | |
| markus_z | sdague: what would be the downside when we add a reno when we drop a deprecated conf option? potentially too much noise in the overall release notes? | 13:44 |
| pkholkin | dansmith: Hello Dan, are you here? Could you please share your opinion about this small patch https://review.openstack.org/#/c/250807/1, thanks | 13:45 |
| *** baoli has quit IRC | 13:45 | |
| sdague | markus_z: honestly, it seems reasonable to add it both at deprecate and remove, but we're still figuring out "when" at this point | 13:46 |
| sdague | upgradeimpact was always so nebulous | 13:46 |
| *** mgoddard has joined #openstack-nova | 13:49 | |
| sdague | notmorgan: +2 | 13:49 |
| pkholkin | johnthetubaguy: Hello John! Could you please return to our 'migration' enginefacade patch https://review.openstack.org/#/c/243496/10 ? We collect several opinions in 7-th patchset, as we understand there were no negative ones, just improvements and smth like this. We updated our patch, please leave your comments, thanks! | 13:49 |
| sdague | johnthetubaguy: you want to hit https://review.openstack.org/253792 when you can? | 13:49 |
| notmorgan | sdague: yeah it might be | 13:49 |
| *** snikitin has joined #openstack-nova | 13:50 | |
| notmorgan | sdague: and. thnx | 13:50 |
| *** gongysh has joined #openstack-nova | 13:50 | |
| notmorgan | mordred: see sdague's +2 on the neutron conf opts | 13:50 |
| sdague | notmorgan: I've nearly worked my way backwards through the glance options to get it ready to put that service catalog lookup patch on top of it https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:glance_image_config,n,z | 13:51 |
| *** yamahata has joined #openstack-nova | 13:51 | |
| sdague | that's actually working on xen with the new calls where you can pass urls around | 13:52 |
| markus_z | sdague: FWIW, I have to admit that I still have a problem in understanding what exactly the "N" is, when upgrading to "N+1". My first understanding was N=kilo|liberty|mitaka. But in the code we ensure that we can upgrade from commit to commit, at least that's my impression. Not sure if that's the problem you described with "when at this point". | 13:53 |
| mordred | sdague: woot! | 13:53 |
| mordred | sdague: so, I haven't pushed it up | 13:53 |
| mordred | sdague: but I have a local patch on top of yours that takes the _next_ step and stops using urls too | 13:54 |
| sdague | markus_z: we need both | 13:54 |
| mordred | sdague: mainly just so I could poke at the codebase and see what's needed | 13:54 |
| esracelik | markus_z: i am back | 13:54 |
| sdague | mordred: on top of which mine? | 13:54 |
| mordred | sdague: but I'll wait for yours to land and stuff before I start down that road | 13:54 |
| *** ducttape_ has joined #openstack-nova | 13:54 | |
| mordred | sdague: "pass a url not a tuple" | 13:54 |
| sdague | mordred: well that's 7 patch series now :) | 13:55 |
| mordred | sdague: sure. :) | 13:55 |
| sdague | you want to stick it up somewhere so I can see what it looks like? | 13:55 |
| sdague | we can hopefully get this whole stack into shape | 13:55 |
| markus_z | esracelik: could you write your suggestion at the end of this etherpad https://etherpad.openstack.org/p/config-options ? | 13:56 |
| markus_z | esracelik: I guess I think what you want to do, just want to be sure | 13:57 |
| *** vladikr has joined #openstack-nova | 13:57 | |
| *** eglynn has quit IRC | 13:58 | |
| markus_z | esracelik: I still don't figure out why git sees a merge problem. Maybe it has too little context for now. | 13:58 |
| esracelik | markus_z: ok, I should think about it more. then if I find a good solution i will share, ok? | 14:00 |
| *** wuhg has joined #openstack-nova | 14:00 | |
| *** logan- has joined #openstack-nova | 14:01 | |
| johnthetubaguy | notmorgan: sdague: not sure about that reno note, the prelude appears in a very odd place. There are some nodes on that here: http://docs.openstack.org/developer/nova/code-review.html#when-a-release-note-is-needed | 14:02 |
| *** yamahata has quit IRC | 14:02 | |
| markus_z | esracelik: yes, sounds good. I'm doing the reviews now. | 14:02 |
| johnthetubaguy | notmorgan: sdague: I am +2 otherwise | 14:02 |
| johnthetubaguy | pkholkin: I think I got the decorator order mixed up in my head, but are you sure it is the correct way around? | 14:03 |
| esracelik | markus_z: ok thanks | 14:03 |
| notmorgan | so uh | 14:03 |
| notmorgan | johnthetubaguy: remove the prelude? | 14:03 |
| *** mdrabe has joined #openstack-nova | 14:03 | |
| johnthetubaguy | notmorgan: yeah, sorry, just drop that bit | 14:03 |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Corrects the API propose for live-migration progress https://review.openstack.org/255122 | 14:04 |
| *** ccarmack has joined #openstack-nova | 14:04 | |
| openstackgerrit | Morgan Fainberg proposed openstack/nova: Remove deprecated neutron auth options https://review.openstack.org/253792 | 14:04 |
| johnthetubaguy | notmorgan: they all appear at the very top of the release notes, in commit order, I think | 14:04 |
| notmorgan | johnthetubaguy: yeah really not familiar with reno | 14:04 |
| notmorgan | johnthetubaguy: so... following recommendations | 14:05 |
| *** rlrossit has joined #openstack-nova | 14:05 | |
| sdague | johnthetubaguy: oh, we're dropping preludes? | 14:05 |
| sdague | johnthetubaguy: reno's documentation says you should use it | 14:05 |
| sdague | http://docs.openstack.org/developer/reno/usage.html#creating-new-release-notes | 14:06 |
| sdague | so maybe we should get that updated if it's not going to be the case | 14:06 |
| *** gcb has quit IRC | 14:06 | |
| johnthetubaguy | sdague: just not using them till the end, is the current plan, at least that was my take away from a discussion with bauzas | 14:06 |
| sdague | "till the end" ? | 14:06 |
| dhellmann | yeah, it's not really expected that each note has its own prelude | 14:06 |
| sdague | dhellmann: ok, that's not super clear | 14:07 |
| dhellmann | the prelude section is to add text describing the release generally | 14:07 |
| johnthetubaguy | sdague: I am thinking, we could add a release summary there, eventually | 14:07 |
| johnthetubaguy | ah, yeah, what dhellmann was saying | 14:07 |
| openstackgerrit | Thomas Maddox proposed openstack/nova: virt: allow for direct mounting of LocalBlockImages https://review.openstack.org/215311 | 14:07 |
| sdague | maybe some guidance with "this is not expected to be in every release note, but only in the final summary notes of the release" | 14:07 |
| sdague | or something | 14:08 |
| dhellmann | sdague : this is described under http://docs.openstack.org/developer/reno/usage.html#editing-a-release-note but if you have suggestions for clarifying that I'll definitely take them -- I want this stuff to be obvious | 14:08 |
| sdague | dhellmann: yeh, where is the source doc for that? | 14:08 |
| *** ducttape_ has quit IRC | 14:08 | |
| sdague | maybe I can suggest wording | 14:08 |
| *** jerrygb_ has joined #openstack-nova | 14:08 | |
| dhellmann | http://git.openstack.org/cgit/openstack/reno/tree/doc/source/usage.rst#n48 and http://git.openstack.org/cgit/openstack/reno/tree/reno/create.py#n22 | 14:09 |
| *** houming has quit IRC | 14:09 | |
| dhellmann | the latter more directly says that not every note needs a prelude, so maybe we just need to sync those up | 14:09 |
| sdague | ok, I'll circle around | 14:09 |
| sdague | in the middle of this test refactor, only 24 fails left.... | 14:10 |
| *** ljxiash has quit IRC | 14:10 | |
| *** haomaiwang has joined #openstack-nova | 14:10 | |
| dhellmann | k | 14:10 |
| alex_xu | mdbooth: pkoniszewski, hi, I updated the progress spec, appreciate you can take a look at, https://review.openstack.org/#/c/255122/ thanks in advance! | 14:10 |
| * alex_xu didn't found other live-migration team members online, just add them to the reviewer directly | 14:11 | |
| pkholkin | johnthetubaguy: are you afraid that the order of the decorators is wrong and cause some problems? | 14:11 |
| *** gongysh has quit IRC | 14:11 | |
| johnthetubaguy | pkholkin: yes, added comments in the review | 14:11 |
| pkholkin | does some private function helps? | 14:11 |
| *** jerrygb has quit IRC | 14:11 | |
| *** pece has quit IRC | 14:11 | |
| johnthetubaguy | pkholkin: yes, that would help | 14:11 |
| *** cdent has quit IRC | 14:12 | |
| *** atuvenie_ has quit IRC | 14:12 | |
| johnthetubaguy | pkholkin: the problem is we need the remote decorator first, I think, then only when it reaches the conductor should the mutation of the context be applied, its possible it actually works OK, because the first mutation is ignored, just doesn't seem correct | 14:12 |
| *** pratikmallya has joined #openstack-nova | 14:12 | |
| *** ccarmack has quit IRC | 14:12 | |
| *** ljxiash has joined #openstack-nova | 14:13 | |
| *** lykinsbd has joined #openstack-nova | 14:13 | |
| *** ccarmack has joined #openstack-nova | 14:13 | |
| *** bauzas has joined #openstack-nova | 14:13 | |
| johnthetubaguy | pkholkin: also, it seems to require an object version bump in its current form, as it changes the object hash, but I suspect if you move the decorator, that might not need a version bump | 14:13 |
| pkholkin | johnthetubaguy: one more question, will it be right if we swap decorators? as I understand it will be ok but not clear to others | 14:14 |
| *** ijuwang has joined #openstack-nova | 14:14 | |
| pkholkin | johnthetubaguy: hmm, yes, private function will not cause a new verson | 14:15 |
| johnthetubaguy | better readability and avoiding a version bump seem like a win here | 14:15 |
| openstackgerrit | Thomas Maddox proposed openstack/nova: virt: allow for direct mounting of LocalBlockImages https://review.openstack.org/215311 | 14:15 |
| johnthetubaguy | pkholkin: I wonder if we actually want a context manager instead of a decorator? | 14:15 |
| * bauzas back from an Internet issue since 1130UTC | 14:16 | |
| *** houming has joined #openstack-nova | 14:17 | |
| *** cdent has joined #openstack-nova | 14:18 | |
| *** takedakn has quit IRC | 14:19 | |
| *** breitz has quit IRC | 14:19 | |
| *** breitz has joined #openstack-nova | 14:19 | |
| pkholkin | johnthetubaguy: we think that decorator is good for readability, we tried to do smth like this in 4-th patchset https://review.openstack.org/#/c/243496/4/nova/objects/migration.py | 14:20 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: virt: Allow multiple volume attachments https://review.openstack.org/193135 | 14:20 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: comp: Allow multiple volume attachments https://review.openstack.org/193134 | 14:20 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: WIP: REST API changes to support multi-attach https://review.openstack.org/153038 | 14:20 |
| pkholkin | but still decorator is more beautiful imho | 14:20 |
| *** AJaeger has quit IRC | 14:21 | |
| johnthetubaguy | pkholkin: decorator is good if it was not a remoted method is also via a decorator that drops the context changes | 14:22 |
| *** gokrokve has joined #openstack-nova | 14:22 | |
| *** pratikmallya has quit IRC | 14:22 | |
| *** jichen has quit IRC | 14:22 | |
| pkholkin | johnthetubaguy: yes( private method will help us | 14:23 |
| johnthetubaguy | pkholkin: I quite like that pattern, because its clear, although I don't like the method names much: https://review.openstack.org/#/c/243496/4/nova/objects/migration.py | 14:23 |
| *** yalie has quit IRC | 14:23 | |
| johnthetubaguy | pkholkin | 14:23 |
| johnthetubaguy | pkholkin: yeah, happy to try the private method, to see how that looks | 14:23 |
| johnthetubaguy | pkholkin: thats probably easier to read | 14:23 |
| *** ctrath has joined #openstack-nova | 14:23 | |
| pkholkin | johnthetubaguy: ok, let's try) | 14:23 |
| *** diogogmt has quit IRC | 14:24 | |
| openstackgerrit | Andrew Bogott proposed openstack/nova: Prevent scheduling of instances on full hosts. https://review.openstack.org/242251 | 14:24 |
| *** mriedem_away is now known as mriedem | 14:25 | |
| *** diogogmt has joined #openstack-nova | 14:26 | |
| *** ildikov has quit IRC | 14:28 | |
| *** bklei has joined #openstack-nova | 14:28 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 14:29 |
| *** ildikov has joined #openstack-nova | 14:30 | |
| pkholkin | johnthetubaguy: here it is ^ | 14:31 |
| *** mtanino has joined #openstack-nova | 14:31 | |
| pkholkin | one sec more please | 14:32 |
| johnthetubaguy | pkholkin: sorry, I know I suggested that, but looking at it, added one more idea | 14:35 |
| *** gjayavelu_ has joined #openstack-nova | 14:35 | |
| *** ildikov has quit IRC | 14:35 | |
| johnthetubaguy | pkholkin: let me know if that looks better? | 14:35 |
| pkholkin | johnthetubaguy: in this case it will looks better but we have several places when we have several db calls | 14:38 |
| pkholkin | I don't think we can adopt this concept everywhere | 14:38 |
| *** diogogmt has quit IRC | 14:38 | |
| *** dave-mccowan has joined #openstack-nova | 14:38 | |
| johnthetubaguy | pkholkin: true | 14:39 |
| johnthetubaguy | pkholkin: don't feel too bad mixing those two forms though | 14:39 |
| *** gongysh has joined #openstack-nova | 14:40 | |
| *** akshai_ has joined #openstack-nova | 14:40 | |
| *** atuvenie has joined #openstack-nova | 14:41 | |
| *** pradk has joined #openstack-nova | 14:42 | |
| *** pradk has quit IRC | 14:43 | |
| *** fawadkhaliq has joined #openstack-nova | 14:43 | |
| *** akshai has quit IRC | 14:44 | |
| *** lykinsbd has quit IRC | 14:44 | |
| mriedem | ttx: i assumed that nova stable/liberty release would be 12.1.0 because of dependency updates, is that not true? | 14:44 |
| *** pradk has joined #openstack-nova | 14:44 | |
| mriedem | i thought that's how we've handled versions in the past, at least on clients/libraries | 14:44 |
| *** mc_nair has joined #openstack-nova | 14:45 | |
| mriedem | it's also how we've versioned stable releases before | 14:45 |
| openstackgerrit | dstepanenko proposed openstack/nova: Modified the nova.objects.ComputeNode object to load on-demand the pci_device_pools field by a subquery instead of pulling from the compute_nodes.pci_stats field in the database. https://review.openstack.org/247692 | 14:45 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: DO NOT MERGE Ceph for live-migration job https://review.openstack.org/252283 | 14:45 |
| *** pratikmallya has joined #openstack-nova | 14:45 | |
| *** lucasagomes is now known as lucas-hungry | 14:46 | |
| *** pratikma_ has joined #openstack-nova | 14:46 | |
| *** mrkz has joined #openstack-nova | 14:47 | |
| sdague | johnthetubaguy / alex_xu hmmm, we've got api samples tests that actually specify differences if they are on the old vs. new code stack | 14:47 |
| sdague | which is bad | 14:47 |
| *** lykinsbd has joined #openstack-nova | 14:47 | |
| *** esracelik has quit IRC | 14:47 | |
| johnthetubaguy | sdague: I think we have a few cases were that is valid, sadly | 14:48 |
| johnthetubaguy | sdague: we released v2.1 with some extensions missing, if I remember correctly | 14:48 |
| pkholkin | johnthetubaguy: In general I like to decorate small functions olnly with db calls, but e.g. will it be ok here? https://github.com/openstack/nova/blob/master/nova/objects/compute_node.py#L395-L417 | 14:48 |
| pkholkin | what do you think | 14:48 |
| sdague | johnthetubaguy: hmmm... ok | 14:48 |
| *** atuvenie has quit IRC | 14:48 | |
| johnthetubaguy | pkholkin: that sounds like a good place for the _impl pattern | 14:49 |
| johnthetubaguy | pkholkin: I am OK with that, I think | 14:49 |
| *** gokrokve has quit IRC | 14:49 | |
| johnthetubaguy | sdague: it could be one of the bad ones though, where we don't get turn all the extensions on for some of the v2.0 tests? | 14:50 |
| *** pratikmallya has quit IRC | 14:50 | |
| sdague | used_limits is currently one that I'm hitting | 14:50 |
| sdague | I'm trying to mildly unspaghetify api_samples | 14:50 |
| *** burt has joined #openstack-nova | 14:51 | |
| johnthetubaguy | sdague: oh, odd, I wouldn't have expected that to be different | 14:51 |
| *** gjayavelu_ has quit IRC | 14:52 | |
| sdague | yeh, in the v2 legacy front there is rate information | 14:52 |
| sdague | which is not true in v1 | 14:53 |
| sdague | v2.1 or 2 on 2.1 | 14:53 |
| *** ducttape_ has joined #openstack-nova | 14:54 | |
| openstackgerrit | Dina Belova proposed openstack/nova: == WIP == Integrate OSprofiler and Nova https://review.openstack.org/254703 | 14:55 |
| *** mgoddard has quit IRC | 14:57 | |
| openstackgerrit | Tomi Juvonen proposed openstack/nova: Add host_status attribute for servers/detail and servers/{server_id} https://review.openstack.org/241562 | 14:57 |
| ttx | mriedem: what kind of dependency updates do we have ? I seem to remember someone saying stable release updates should only be .z... Maybe lifeless or dhellmann | 14:58 |
| johnthetubaguy | sdague: ouch | 14:59 |
| johnthetubaguy | sdague: :( | 14:59 |
| ttx | mriedem: for libs for example we only do .z, even with dep updates ? | 14:59 |
| mriedem | ttx: we've done .y releases for libs and clients because of dependency updates | 15:00 |
| mriedem | to leave room for bug fixes on .z | 15:00 |
| mriedem | on stable branches | 15:00 |
| mriedem | in fact a lot of the gate wedges on dep installs in stable was because we didn't update .y enough | 15:00 |
| mriedem | so we had no wiggle room for patches on .z on stable | 15:00 |
| ttx | mriedem: on stable branches we did simple y bumps when we changed branches, leaving room for only .z, no ? | 15:00 |
| mriedem | ttx: ideally yeah | 15:01 |
| mriedem | between juno and kilo that wasn't always the case with some projeects | 15:01 |
| mriedem | so juno and kilo would be sharing the same .y version | 15:01 |
| ttx | so we can't do y bumps for stable branches in libs | 15:01 |
| *** sneti has joined #openstack-nova | 15:01 | |
| *** smatzek has joined #openstack-nova | 15:01 | |
| mriedem | sure, not libs | 15:01 |
| mriedem | but nova isn't a lib | 15:01 |
| openstackgerrit | Moshe Levi proposed openstack/nova: libvirt: report pci Type-PF type even when VFs are disabled https://review.openstack.org/227160 | 15:02 |
| mriedem | if dependency updates for nova stable/liberty don't qualify as a .y bump, then what does since we don't backport features? | 15:02 |
| *** ljxiash_ has joined #openstack-nova | 15:02 | |
| *** ndipanov has quit IRC | 15:02 | |
| *** eharney has quit IRC | 15:02 | |
| *** ndipanov has joined #openstack-nova | 15:02 | |
| ttx | mriedem: nothing would, the stable branch would be only .z. Not saying that's what we should do (since we have room for .y since we bumped x), just trying to be consistent with what we do for libs | 15:03 |
| *** dustins|gone is now known as dustins | 15:03 | |
| mriedem | ttx: that seems weird to me. stable server releases in the past were always a .y bump | 15:03 |
| ttx | let's see if dhellmann or lifeless see an issue with bumping .y on stable branch updates for things that bumped x on stable branch boundary | 15:03 |
| mriedem | i'm not sure why decentralizaing stable server releases makes that any different | 15:03 |
| *** eharney has joined #openstack-nova | 15:03 | |
| ttx | mriedem: no, they always were .z | 15:04 |
| *** cfriesen has joined #openstack-nova | 15:04 | |
| ttx | mriedem: 2015.1.0, 2015.1.1 | 15:04 |
| mriedem | oh sh*t, you're right | 15:04 |
| mriedem | we never had a .y as 0 before i guess | 15:04 |
| *** claudiub_ has quit IRC | 15:04 | |
| mriedem | y was always 1 or 2 for first or second half release | 15:04 |
| ttx | mriedem: yes | 15:04 |
| mriedem | ok, yeah, but that was pre-semver | 15:05 |
| *** ljxiash has quit IRC | 15:05 | |
| mriedem | now we have unherded cats with semver | 15:05 |
| notmorgan | ~. | 15:05 |
| ttx | so I'm just returning the question -- i'm not sure why decentralizaing stable server releases makes that any different | 15:05 |
| ttx | (I could live with it, but I seem to remember someone having a good reason why we shouldn't do precisely that | 15:06 |
| *** xyang1 has joined #openstack-nova | 15:06 | |
| jroll | the "bump major version every cycle" means we *can* do a .y bump if we would like | 15:06 |
| *** mdrabe has quit IRC | 15:06 | |
| jroll | I don't think "decentralizaing stable server releases" does change anything | 15:06 |
| pkholkin | johnthetubaguy: I think it will be better to use _impl pattern everytime | 15:07 |
| moshele | ndipanov: I addressed your comment in https://review.openstack.org/#/c/227160/ , you can review again | 15:07 |
| *** irenab has quit IRC | 15:07 | |
| johnthetubaguy | pkholkin: why is that, just for consistency? | 15:07 |
| *** chinmaya has joined #openstack-nova | 15:07 | |
| ttx | mriedem: also we already had a number of liberty stable point releases and those were all .z bumps, so there is precendent | 15:07 |
| *** pumaranikar has joined #openstack-nova | 15:07 | |
| *** pumaranikar has quit IRC | 15:07 | |
| mriedem | ttx: yeah, but screw those guys :) the projects are independent now right? | 15:08 |
| *** pumaranikar has joined #openstack-nova | 15:08 | |
| *** pushkaru has joined #openstack-nova | 15:08 | |
| mriedem | i'd argue those .z bumps were wrong and should have been .y | 15:08 |
| *** pushkaru has quit IRC | 15:08 | |
| *** jeffrey4l has joined #openstack-nova | 15:09 | |
| *** pumaranikar has quit IRC | 15:09 | |
| *** wanghao has joined #openstack-nova | 15:09 | |
| pkholkin | johnthetubaguy: yes, also our new decorator always wants self and context parameters, for db calls self is not always needed, so we can't use the only decorator for both cases | 15:09 |
| ttx | mriedem: I'm raising the question on #openstack-release, will get back to you | 15:09 |
| *** pumaranikar has joined #openstack-nova | 15:09 | |
| *** cdelatte has quit IRC | 15:09 | |
| sdague | migi: I approved https://review.openstack.org/#/c/254398/ - but there is also a comment in there about a uuid sentinal class that might make this clearer as it expands through the test code | 15:10 |
| *** cdelatte has joined #openstack-nova | 15:10 | |
| sdague | let me know what you think | 15:10 |
| johnthetubaguy | pkholkin: hmm... the decorator is written wrongly I think | 15:10 |
| sdague | mriedem: aren't you assuming the nova version is semver then? | 15:11 |
| pkholkin | johnthetubaguy: when we write it we look at object functions where cls and context are always the first arguments | 15:11 |
| dansmith | sdague: migi: I'm taking a crack and that sentinel class right now because it sounds fun | 15:11 |
| mriedem | sdague: yes, i thought that was the idea | 15:11 |
| sdague | mriedem: what are the rules for X besides time based? | 15:12 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 15:12 |
| *** delattec has joined #openstack-nova | 15:12 | |
| sdague | dansmith: coolio | 15:12 |
| mriedem | sdague: well normally would be backwards incompat changes - which we shouldn't do in the api, but we certainly drop function and options every 6 months | 15:13 |
| mriedem | those kinds of things don't get back to stable though | 15:13 |
| sdague | ok | 15:13 |
| mriedem | in the past dep updates in clients/libs were always a Y bump | 15:13 |
| johnthetubaguy | pkholkin: yes, added a note on that old patch set, of how to avoid that | 15:14 |
| mriedem | and bug fixes/small patches are a Z | 15:14 |
| johnthetubaguy | pkholkin: at least I think that would work | 15:14 |
| pkholkin | johnthetubaguy: thanks, looking | 15:14 |
| *** cdelatte has quit IRC | 15:15 | |
| *** derekh has joined #openstack-nova | 15:18 | |
| *** ssurana has joined #openstack-nova | 15:20 | |
| *** sudipto has joined #openstack-nova | 15:20 | |
| *** ildikov has joined #openstack-nova | 15:20 | |
| *** noqa_v_qoovnie has quit IRC | 15:22 | |
| *** ijuwang has quit IRC | 15:23 | |
| *** diogogmt has joined #openstack-nova | 15:25 | |
| *** inteq has quit IRC | 15:25 | |
| *** Daisy has joined #openstack-nova | 15:25 | |
| sdague | dansmith: something is up with grenade jobs and conductor - http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-cond.txt.gz#_2015-12-09_12_44_35_396 | 15:26 |
| sdague | post upgrade it's failing to connect to rabbit | 15:26 |
| dansmith | um | 15:27 |
| dansmith | only conductor? | 15:27 |
| sdague | well, it's where things blow up | 15:27 |
| sdague | because of the floating ip dealocation | 15:27 |
| sdague | of the resources that span the upgrade | 15:27 |
| *** mdrabe has joined #openstack-nova | 15:28 | |
| sdague | o.m 3.1 | 15:28 |
| sudipto | mriedem, could you please help on my query above? | 15:29 |
| sudipto | Hi, I am facing a strange issue with Jenkins...my review here: http://logs.openstack.org/65/245765 reports a failure in jenkins: http://paste.openstack.org/show/481323/ but my changes are not even there in the same line that jenkins reports ?? This works fine on my local environment...so a bit clueless... | 15:29 |
| *** Daisy has quit IRC | 15:29 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Add uuidsentinel test module https://review.openstack.org/255321 | 15:30 |
| *** gokrokve has joined #openstack-nova | 15:30 | |
| *** ljxiash_ has quit IRC | 15:30 | |
| *** ociuhandu has joined #openstack-nova | 15:30 | |
| dansmith | sdague: see this? http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-api.txt.gz?level=TRACE | 15:30 |
| *** houming has quit IRC | 15:31 | |
| sdague | dansmith: oh, hey, I didn't see that one | 15:31 |
| dansmith | sdague: that's later, so I think the thing you see there in conductor is just mid-upgrade | 15:32 |
| sdague | I guess I jumped directly to the network | 15:32 |
| sdague | dansmith: hmmm... this is the single node case | 15:32 |
| sdague | there is no mid upgrade | 15:32 |
| dansmith | sdague: by mid-upgrade I meant whilst restarting things.. do we not restart rabbit? | 15:32 |
| sdague | we do not restart rabbit | 15:33 |
| dansmith | okay | 15:33 |
| dansmith | well, I still think that API thing is more related to the failure, right? | 15:33 |
| dansmith | sdague: and conductor is clearly connected to rabbit after the reconnect error | 15:33 |
| dansmith | sdague: it handles a couple requests | 15:33 |
| sdague | dansmith: ok, so any idea why this race is here? This seems weird | 15:34 |
| *** gokrokve has quit IRC | 15:35 | |
| sdague | oh, sorry, this is multinode | 15:35 |
| *** thorst_ has quit IRC | 15:35 | |
| dansmith | ..I was going to say, it's doing backports, so it better be multinode :) | 15:35 |
| sdague | do we have an incompatibility with the network node? | 15:35 |
| dansmith | sdague: you're asking about the failure in the api log? | 15:35 |
| mriedem | sdague: i thought we knew the failure was https://review.openstack.org/#/c/255245/ | 15:35 |
| mriedem | http://logs.openstack.org/11/246211/19/check/gate-grenade-dsvm/06a815e/logs/new/screen-n-api.txt.gz?level=TRACE | 15:36 |
| mriedem | wtf, where is http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-api.txt.gz?level=TRACE coming from? | 15:36 |
| *** sneti has quit IRC | 15:37 | |
| openstackgerrit | gongysh proposed openstack/nova: Fix the duplicate nics for VMs https://review.openstack.org/252824 | 15:37 |
| *** sneti has joined #openstack-nova | 15:37 | |
| snikitin | johnthetubaguy, pkholkin: John, unfortunately it doesn't work :( getcallargs can't work with args. just with kwargs. https://github.com/openstack/nova/blob/master/nova/safe_utils.py#L28-L31 | 15:37 |
| sdague | mriedem: right, the second issue is the one that I'm curious about | 15:37 |
| dansmith | mriedem: sdague I don't see how this is a race, it seems like it must be something clearly wrong because we're deref'ing a name that doesn't exist | 15:38 |
| johnthetubaguy | snikitin: I think thats talking about the function you pass in (i.e. the decorated function), not the function using it? | 15:38 |
| sdague | dansmith: well, it's a race in that it's not happening all the tim | 15:38 |
| alaski | snikitin: it works with args, as long as the method being called has named them. *args doesn't work | 15:38 |
| dansmith | sdague: okay, I just wonder how that is possible | 15:39 |
| sdague | http://tinyurl.com/oxsyeyy | 15:39 |
| *** READ10 has joined #openstack-nova | 15:39 | |
| *** belmoreira has quit IRC | 15:39 | |
| sdague | just after we tackled the other fail rate, we're spiking here | 15:39 |
| dansmith | and you're sure it's not all the time? | 15:39 |
| dansmith | the slope of that curve seems high enough to be "all the time" | 15:40 |
| sdague | it looks like it's landing in around 50% failure rate | 15:40 |
| dansmith | which means either the old or new network node is broken 100% of the time right? | 15:40 |
| openstackgerrit | dstepanenko proposed openstack/nova: Duplicated the logic of nova.pci.stats.PciDeviceStats support_requests and apply_requests method into the nova.objects.PciDevicePoolList object. https://review.openstack.org/242087 | 15:40 |
| sdague | yeh, probably | 15:41 |
| sdague | let me figure out in logstash the last time we got a pass | 15:41 |
| sdague | mriedem: is logstash responding reasonably for you? | 15:41 |
| *** lucas-hungry is now known as lucasagomes | 15:42 | |
| mriedem | sdague: nope | 15:43 |
| *** gongysh has quit IRC | 15:43 | |
| *** thangp has joined #openstack-nova | 15:44 | |
| *** Sukhdev has joined #openstack-nova | 15:44 | |
| *** gokrokve has joined #openstack-nova | 15:45 | |
| dansmith | I don't see anything obvious that merged in the last couple days | 15:47 |
| bauzas | dansmith: are we still okay to provide that one https://review.openstack.org/#/c/242602/ ? | 15:50 |
| bauzas | dansmith: I can make it a reality, hence my question | 15:51 |
| *** jeffrey4l has quit IRC | 15:51 | |
| *** alexschm_ has quit IRC | 15:51 | |
| *** jeffrey4l has joined #openstack-nova | 15:51 | |
| bauzas | that's good thing anyway, but I wanted to make sure | 15:51 |
| dansmith | bauzas: I think so, but mriedem and alaski were discussing it most recently, so I'd ask them | 15:51 |
| bauzas | okay, 'd love to hear thoughts on that | 15:52 |
| bauzas | mriedem: thoughts? I can see you +1ing only | 15:52 |
| *** ljxiash has joined #openstack-nova | 15:52 | |
| *** zhangjn has quit IRC | 15:52 | |
| mriedem | bauzas: dansmith: alaski: so.... | 15:53 |
| bauzas | mriedem: https://review.openstack.org/#/c/242603/13 seems correct to me too, so I feel it's fine to merge both | 15:53 |
| bauzas | unless I'm missing something | 15:53 |
| *** rubasov has quit IRC | 15:53 | |
| mriedem | right now i'm not sure we have anything consuming that | 15:53 |
| mriedem | since i think we'll probably just drop https://review.openstack.org/#/c/242649/ | 15:53 |
| mriedem | i brought it up the other day b/c of https://review.openstack.org/#/c/254428/ | 15:53 |
| bauzas | fair to say but it helps us anyway | 15:54 |
| mriedem | and my concern about how today the bdm uniqueness is based on volume_id + device_name, | 15:54 |
| mriedem | but ^ breaks down when boot from snapshot/blank/image | 15:54 |
| bauzas | having only ways to count BDMs using device_name just makes me sad | 15:54 |
| mriedem | where nova creates the volume and then sets the volume_id later | 15:54 |
| alaski | bauzas: I think the db migration for adding the uuid is good. I would need to look at the later patches again | 15:55 |
| *** liverpooler has quit IRC | 15:55 | |
| bauzas | mriedem: I see your point, I just feel that we would get benefits from BDMs having uuids maybe in the next releases | 15:55 |
| openstackgerrit | dstepanenko proposed openstack/nova: Moved logic from nova.pci.stats.PciDevStats.consume_requests() to the nova.pci.manager.PciDevTracker._claim_for_instance() method using pci devices instead of pci pools. https://review.openstack.org/243131 | 15:55 |
| bauzas | mriedem: just because it helps our problems | 15:55 |
| mriedem | bauzas: well, it's a future requirement | 15:56 |
| *** daemontool has joined #openstack-nova | 15:56 | |
| *** irenab has joined #openstack-nova | 15:56 | |
| bauzas | mriedem: so IIUC your concern is that we wouldn't use the uuid right in that cycle, correct? | 15:56 |
| mriedem | so i'm hesitant to put that in when nothing is going to use it right now | 15:56 |
| bauzas | I see | 15:56 |
| *** ljxiash has quit IRC | 15:56 | |
| mriedem | right, i kind of feel we should add it when we plan on using it | 15:56 |
| mriedem | otherwise it's technical debt | 15:57 |
| bauzas | mriedem: but I'm considering the upgrade stratefy | 15:57 |
| bauzas | strategy even | 15:57 |
| pkholkin | alaski, johnthetubaguy: why 'self' or 'cls' is not removed from argnames? next zip operation works incorrectly because it thinks that 'self' value should be used for first remaining_argname | 15:57 |
| mriedem | bauzas: how so? | 15:57 |
| pkholkin | and our result is shifted | 15:57 |
| *** haomaiwang has quit IRC | 15:57 | |
| bauzas | mriedem: say that we would want to have uuids in the next releases, it would mean a migration path right? something like we usually do with the object (+nova-manage kind of stuff) | 15:58 |
| alaski | bauzas: I agree with mriedem that the series should be in good shape before going ahead with the db migration. But I think this is something we need to do | 15:58 |
| *** haomaiwang has joined #openstack-nova | 15:58 | |
| *** gokrokve has quit IRC | 15:58 | |
| bauzas | mriedem: that would probably require some magic for using legacy computes | 15:58 |
| mriedem | alaski: well i think we're giving up on fixing that cells v1 bdm bug right? | 15:58 |
| *** haomaiwang has quit IRC | 15:59 | |
| mriedem | that's what people said in the nova meeting a few weeks ago | 15:59 |
| bauzas | mriedem: by merging the uuid stuff now, we could prevent some upgrade backwards path in a next release, nope? | 15:59 |
| mriedem | no new fixes for latent cells bugs exposed by new tests | 15:59 |
| bauzas | right, that's what we said | 15:59 |
| *** erlon has joined #openstack-nova | 15:59 | |
| *** smurke has joined #openstack-nova | 15:59 | |
| bauzas | only fix what regressed, not try to boil the ocean of uncovered bugs | 15:59 |
| *** haomaiwa_ has joined #openstack-nova | 15:59 | |
| alaski | yeah | 15:59 |
| *** gokrokve has joined #openstack-nova | 15:59 | |
| *** tjones has joined #openstack-nova | 15:59 | |
| *** haomaiwa_ has quit IRC | 16:00 | |
| alaski | but there was a ceph bug for similar reasons right? it still seems useful to me | 16:00 |
| bauzas | mriedem: that said, I got your concern about some debt we would pay | 16:00 |
| *** haomaiwang has joined #openstack-nova | 16:00 | |
| alaski | rather, a race with deleting instances | 16:00 |
| *** haomaiwang has quit IRC | 16:01 | |
| mriedem | alaski: that's https://review.openstack.org/#/c/254428/ | 16:01 |
| mriedem | alaski: which i'm hoping fixes that bug | 16:01 |
| *** haomaiwa_ has joined #openstack-nova | 16:01 | |
| pkholkin | alaski: could you please check 'getcallargs', please look the message above | 16:01 |
| mriedem | alaski: my thinking was along https://review.openstack.org/#/c/254428/2/nova/compute/manager.py and how it's using the volume_id to delete the volume that nova created, | 16:02 |
| mriedem | alaski: the problem was the volume_id wasn't set in the race failure | 16:02 |
| mriedem | alaski: so i was thinking along the lines of we could get these things by uuid, but that wouldn't help this case since nova still doesn't have the volume_id set on the bdm at that point | 16:02 |
| alaski | mriedem: okay, I see your point | 16:02 |
| mriedem | alaski: so the real solution is to stop creating volumes in nova :) | 16:03 |
| alaski | \o/ | 16:03 |
| mriedem | or to not allow delete while doing so :) | 16:03 |
| mriedem | but neither of those are going to happen | 16:03 |
| *** zz_john5223 is now known as john5223 | 16:03 | |
| mriedem | ndipanov: btw, can you check this out - should fix a race bug in the gate https://review.openstack.org/#/c/254428/ | 16:03 |
| alaski | I do think there's still value in assigning a uuid to bdms, however the priority is much lower if we're not going to address cells | 16:04 |
| ndipanov | yeah I am just looking at that mriedem and really want to say that this does not fix the fundamental race | 16:04 |
| *** pumaranikar has quit IRC | 16:04 | |
| ndipanov | the race of all races! | 16:04 |
| ndipanov | just kidding | 16:04 |
| ndipanov | but it does not fix the race | 16:04 |
| *** pumaranikar has joined #openstack-nova | 16:05 | |
| *** pradk has quit IRC | 16:05 | |
| mriedem | ndipanov: well, from the log digging i did in one of the ceph job gate failures, with notes in the bug report, it should close the gap | 16:05 |
| mriedem | *narrow the gap | 16:05 |
| ndipanov | narrow yes | 16:05 |
| mriedem | b/c in the logs i was looking at, we had created the volume and were attaching by the time the delete came in | 16:05 |
| mriedem | but we hadn't yet saved off bdm.volume_id | 16:05 |
| mriedem | so when we deleted the instance, that was None and we orphaned the volume | 16:05 |
| mriedem | which then tempest puked on b/c it tried deleting the volume snapshot before the related volume was gone | 16:06 |
| mriedem | this is still racy for sure | 16:06 |
| mriedem | tempest might fire off the snapshot delete while that dependent volume is still being deleted, idk, but that would fail | 16:06 |
| *** kmartin has joined #openstack-nova | 16:07 | |
| alaski | pkholkin: looking now | 16:07 |
| bauzas | mriedem: fwiw, I got your concern about approving something not really planned yet, I'm holding off | 16:07 |
| mriedem | jgriffith: does cinder have an API to find dependent volumes for a given snapshot? | 16:07 |
| pkholkin | alaski: I think we should add this in IF args = args[1:] | 16:07 |
| *** vishwanathj has quit IRC | 16:08 | |
| *** nkrinner has quit IRC | 16:08 | |
| *** vishwanathj has joined #openstack-nova | 16:08 | |
| *** tonytan4ever has joined #openstack-nova | 16:09 | |
| *** artom has joined #openstack-nova | 16:09 | |
| *** pradk has joined #openstack-nova | 16:10 | |
| sdague | dansmith: hmmm... I'm kind of stumped on this fail because API is throwing an exception, which should be from over the wire | 16:12 |
| edleafe | markus_z: around? | 16:12 |
| *** mjura has quit IRC | 16:13 | |
| *** chinmaya has quit IRC | 16:14 | |
| edleafe | bauzas: did you see the discussion I had yesterday with belmoreira? | 16:16 |
| dansmith | sdague: well, I'm not so sure about that | 16:17 |
| dansmith | sdague: the rpc bit I mean | 16:17 |
| dansmith | sdague: I know that there is one line of o.m in there, but it's also calling to "LocalManager", and there doesn't seem to be enough in the stack trace to support going over the wire | 16:18 |
| *** jerrygb_ has quit IRC | 16:18 | |
| *** kfarr has joined #openstack-nova | 16:18 | |
| sdague | could oslo.messaging be doing a short circuit? | 16:19 |
| dansmith | and no corresponding fail in the log of another service | 16:19 |
| sdague | yeh, I know it's not seeming to hit the other side of the wire here | 16:19 |
| dansmith | o.m doesn't know what things connect to what, so I don't think so | 16:20 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: DO NOT MERGE Ceph for live-migration job https://review.openstack.org/252283 | 16:20 |
| dansmith | I think this is just the api service using something in network-api that doesn't involve an rpc call | 16:20 |
| *** lyanchih has quit IRC | 16:20 | |
| alaski | pkholkin: working on a test case to verify this | 16:21 |
| sdague | dansmith: ok, but this is a pretty new fail | 16:21 |
| *** ljxiash has joined #openstack-nova | 16:21 | |
| sdague | I guess it times with the constraints update - https://review.openstack.org/#/c/254571/2/upper-constraints.txt,cm | 16:21 |
| *** diogogmt has quit IRC | 16:22 | |
| sdague | which unfortunately doesn't run any multinode jobs | 16:22 |
| dansmith | hmm, not sure what in constraints/reqs would have caused this | 16:22 |
| dansmith | but then again, I got nuthin anyway | 16:22 |
| bauzas | edleafe: no, and my IRC bouncer had some issues since my EU morning :( | 16:22 |
| mriedem | dansmith: it allows running with newer oslo | 16:22 |
| markus_z | edleafe: yepp, present | 16:22 |
| mriedem | the dvsm jobs are otherwise constrained by what's in that u-c file | 16:22 |
| mriedem | but updating the u-c file isn't tested on lots of things (multinode/upgrade, ironic, ceilometer) | 16:23 |
| pkholkin | alaski: you can try any of this one https://review.openstack.org/#/c/243496/10/nova/tests/unit/db/test_db_api.py but remove 'self' and 'context' from the corresponding decorator as John advised | 16:23 |
| *** daemontool_ has joined #openstack-nova | 16:23 | |
| mriedem | the u-c change did go through grenade though | 16:23 |
| sdague | mriedem: single node | 16:23 |
| *** sahid has quit IRC | 16:23 | |
| dansmith | what all was in that update? | 16:24 |
| mriedem | oslo.messaging 3.1.0 https://review.openstack.org/#/c/254571/2/upper-constraints.txt,cm | 16:24 |
| mriedem | versioned objects | 16:24 |
| mriedem | mistralclient - i always knew that was going to break nova somehow | 16:24 |
| pkholkin | alaski: http://xsnippet.org/361239/ | 16:25 |
| *** suro-patz has joined #openstack-nova | 16:25 | |
| sdague | an amqp bump | 16:26 |
| edleafe | bauzas: the _init_instance_info of HostManager loads the instances for each compute node on startup | 16:26 |
| edleafe | bauzas: this includes deleted instances | 16:26 |
| alaski | pkholkin: http://xsnippet.org/361240/ passes for me | 16:26 |
| edleafe | bauzas: belmoreira had a case where they had *lots* of deleted instances, and it was taking forever | 16:27 |
| *** hemnafk is now known as hemna | 16:27 | |
| bauzas | edleafe: mmm | 16:27 |
| pkholkin | alaski: I want test with 'self' :) | 16:27 |
| bauzas | edleafe: I see, so you plan to fix that ? | 16:27 |
| *** suro-patz has quit IRC | 16:28 | |
| *** daemontool has quit IRC | 16:28 | |
| edleafe | bauzas: my question is: is there any reason we can't add a filter to not retrieve deleted instances? IOW, do any filters require deleted instance info? | 16:28 |
| bauzas | edleafe: that sounds weird to me IMHO | 16:28 |
| edleafe | bauzas: yes, unless it would cause other issues | 16:28 |
| alaski | pkholkin: https://github.com/openstack/nova/blob/master/nova/tests/unit/test_safeutils.py got you covered :) | 16:28 |
| edleafe | bauzas: what sounds weird? | 16:28 |
| bauzas | edleafe: I don't see a point where we would need to know states of deleted stuff, it's the past while we want to schedule vs. the present | 16:28 |
| sdague | at least in the case I'm looking at the node where this vm is running is on the same one as the api server | 16:28 |
| ndipanov | mriedem, tl;dr I am strongly -1 on this | 16:29 |
| ndipanov | it defies common sense | 16:29 |
| sdague | dansmith: is there some fast path here that just openned up? | 16:29 |
| openstackgerrit | dstepanenko proposed openstack/nova: Moved logic from nova.pci.stats.PciDeviceStats.consume_requests() to the nova.pci.manager.PciDevTracker._claim_for_instance() method using pci devices instead of pci pools. https://review.openstack.org/243131 | 16:29 |
| bauzas | edleafe: I need to review the filters using the instance info, but AFAIK, they're only needing to know the running instances, right ? | 16:29 |
| mriedem | ndipanov: how so? | 16:29 |
| alaski | pkholkin: I may not be understanding yet. if you have a test case that fails that would help | 16:29 |
| dansmith | sdague: I dunno what you mean.. there is no RPC going on here | 16:29 |
| ndipanov | mriedem, we found a bug | 16:29 |
| dansmith | sdague: and hasn't been for a couple years at least | 16:30 |
| ndipanov | with a test | 16:30 |
| edleafe | bauzas: my only concern was anti-affinity, if a deleted instance was restored | 16:30 |
| ndipanov | so instead of fixing it | 16:30 |
| bauzas | edleafe: if so, you should just fix the call to only get the running ones, and if someone wants to provide another in-tree filter using deleted instances, then we could discuss on that by the review change | 16:30 |
| ndipanov | we hack the code to keep the test that we know is failing running | 16:30 |
| sdague | dansmith: for floating ips disassociate? | 16:30 |
| dansmith | sdague: correct | 16:30 |
| sdague | huh | 16:30 |
| sdague | how does that work with multinode? | 16:30 |
| *** ijw has joined #openstack-nova | 16:30 | |
| bauzas | edleafe: sec, I'm trying to remember when I thought about a possible VM resurrection | 16:31 |
| bauzas | edleafe: AFAIK, it's a dead-end when you delete | 16:31 |
| bauzas | edleafe: but I remember some edge case that I need to verify | 16:31 |
| edleafe | bauzas: ok, thanks. I wasn't sure about that either | 16:31 |
| dansmith | sdague: I'm saying I'm tracing through the stack that it's barfing and we're not in the middle of an rpc call | 16:32 |
| alaski | ndipanov: I don't think the test is buggy. we have plenty of users who would do exactly the same thing, delete a server while it's building | 16:32 |
| ndipanov | mriedem, 2 answers are - fix the bug - we care! or turn the test off - we don't give a f*k. everything else is | 16:32 |
| *** ijw has quit IRC | 16:32 | |
| dansmith | sdague: it has to make a call to the appropriate network at some point, but not in that stack trace | 16:32 |
| ndipanov | alaski, my point exactly | 16:32 |
| ndipanov | the test is fine | 16:32 |
| ndipanov | the code is buggy | 16:32 |
| sdague | dansmith: right, I see that, but I don't understand | 16:32 |
| ndipanov | either fix the code | 16:32 |
| jgriffith | mriedem: oddly enough no | 16:32 |
| bauzas | edleafe: there is a conf opt defaulted to 0 that helps that, but I need to doublecheck | 16:32 |
| *** jaosorior has quit IRC | 16:32 | |
| ndipanov | or remove the test since we don't care enough | 16:32 |
| *** ijw has joined #openstack-nova | 16:33 | |
| ndipanov | don't hack to code to keep the test less honest | 16:33 |
| jgriffith | mriedem: that would be a reasonable enhancement to the show command IMHO, and I *thought* somebody had proposed that already | 16:33 |
| dansmith | sdague: the line in the stack trace from o.m is purely the expected_exceptions wrapper, which we're using even in the direct call because we're calling straight to the manager | 16:33 |
| jgriffith | mriedem: I'll look and see if it's hung up in review somewhere | 16:33 |
| jgriffith | mriedem: but for now, the answer is nope | 16:33 |
| sdague | dansmith: ok, so the issue is the manager never initialized correctly? | 16:33 |
| dansmith | sdague: well, that's what I was looking for, but I don't see it, hang on | 16:34 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 16:34 |
| dansmith | sdague: this is why we're not making an rpc call: https://github.com/openstack/nova/blob/master/nova/network/floating_ips.py#L453-L456 | 16:35 |
| dansmith | sdague: so either self.l3driver isn't initialized when it should be, or interface is truthy when it shouldn't be | 16:36 |
| alaski | ndipanov: I get your point about using volume_id being a bit hacky, but the save() isn't intended as a hack | 16:36 |
| bauzas | edleafe: yeah I remember now, it's the soft-delete possibility of the API | 16:36 |
| mriedem | ndipanov: i've replied to https://review.openstack.org/#/c/254428/ | 16:37 |
| pkholkin | alaski: please try any of these new tests from this patch https://review.openstack.org/#/c/243496/13/nova/tests/unit/db/test_db_api.py, values incorrectly get for 'context' and 'value' in our new decorator http://xsnippet.org/361242/ | 16:37 |
| mriedem | ndipanov: i can update the comments in block_device to say it reduces the window but doesn't eliminate it | 16:37 |
| mriedem | ndipanov: and i can check for bdm.volume_id in _shutdown_instance before trying to delete (and log a warning about a race) | 16:37 |
| ndipanov | mriedem, and also check and log if volume_id is missing | 16:38 |
| ndipanov | yeah | 16:38 |
| mriedem | ndipanov: yeah, that's fine | 16:38 |
| ndipanov | it would still be sad about it | 16:38 |
| mriedem | ndipanov: but otherwise i'm not sure what your opposition is as this being a partial at least fix | 16:38 |
| *** yassine__ has quit IRC | 16:38 | |
| mriedem | from the logs, this is clearly a race that happened | 16:38 |
| ndipanov | but we'd at least be honest in saying - look we are obviously using duck tape here | 16:38 |
| mriedem | as noted here https://bugs.launchpad.net/cinder/+bug/1464259/comments/13 | 16:38 |
| openstack | Launchpad bug 1464259 in OpenStack Compute (nova) "Volumes tests fails often with rbd backend" [High,In progress] - Assigned to Matt Riedemann (mriedem) | 16:38 |
| ndipanov | but black is a good color | 16:38 |
| sdague | dansmith: there was an o.vo update in that patch, would that have affected the truthiness of interface | 16:39 |
| mriedem | ndipanov: sure i'm on board with a check and warning | 16:39 |
| sdague | here - https://github.com/openstack/nova/blob/master/nova/network/floating_ips.py#L435 | 16:39 |
| dansmith | sdague: not that I can think of, that'd be a pretty major thing.. either way, interface *should* be set, right/ | 16:39 |
| *** haomaiwa_ has quit IRC | 16:40 | |
| bauzas | edleafe: yeah, I remember now, it's basically the conf opt reclaim_instance_interval defaulted to 0 | 16:40 |
| edleafe | bauzas: exactly - if you undelete a soft-deleted instance, it might break anti-affinity | 16:40 |
| sdague | dansmith: right, I suppose | 16:40 |
| dansmith | sdague: also, we're not setting l3_lib in the config, which means it defaults to the linux net one, which means l3driver should be set | 16:40 |
| *** haomaiwang has joined #openstack-nova | 16:40 | |
| sdague | I guess the real issue is why isn't l3driver loaded | 16:40 |
| bauzas | edleafe: but that just holds off the *real* deletion, it just marks the instance as SOFT_DELETED without putting deleted=uuid in the instances DB table | 16:40 |
| pkholkin | alaski: do you understand our case? maybe we do smth wrong | 16:41 |
| bauzas | edleafe: so if you get the list of instances, you should still get soft deleted as considered active | 16:41 |
| bauzas | edleafe: if filtering on deleted != 0 | 16:41 |
| dansmith | sdague: it's not just not loaded, it's not set on the manager instance, which can't really happen, AFAIK | 16:41 |
| bauzas | well, deleted == 0 rater | 16:41 |
| dansmith | sdague: can't really escape this: https://github.com/openstack/nova/blob/master/nova/network/manager.py#L282-L282 | 16:41 |
| sdague | dansmith: well it clearly can happen | 16:41 |
| pkholkin | dansmith: hi Dan, could you please leave your comments about this small patch https://review.openstack.org/#/c/250807/1 ? | 16:42 |
| alaski | pkholkin: yeah, I see that you're hitting an issue here but I'm not sure why yet. I'm going to pull this down and trace it | 16:42 |
| dansmith | pkholkin: I already did | 16:42 |
| edleafe | bauzas: right, because the instance files are still on the compute node | 16:42 |
| pkholkin | dansmith: oh sorry, thanks, looking | 16:42 |
| bauzas | edleafe: zactly, so that's not a real case of resurrecting a VM | 16:42 |
| dansmith | pkholkin: I don't know why you're pinging me to look at that though | 16:42 |
| pkholkin | alaski: hmm, ok thanks | 16:43 |
| bauzas | edleafe: which means that when you filter by deleted, you should be sure that what has been deleted is the past, and there is zero way to resurrect that | 16:43 |
| pkholkin | dansmith: afaik you are good in databases) | 16:43 |
| dansmith | pkholkin: afaik you are very wrong about that :) | 16:43 |
| edleafe | bauzas: makes sense. Thanks! | 16:43 |
| bauzas | edleafe: speaking of https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L377-L380 (which is the problem I guess) | 16:43 |
| pkholkin | dansmith: :) | 16:44 |
| mriedem | ndipanov: so are you good with my response? if so i'll get started on those updates. | 16:44 |
| edleafe | bauzas: correct - that's where we need to add the deleted filter | 16:44 |
| bauzas | edleafe: zactly | 16:44 |
| bauzas | edleafe: and like I said, if you do that, it's safe because those deleted instances can't be resurrected to be clear | 16:45 |
| sdague | dansmith: unfortunately the FloatingIP mixin is assuming that attribute, even if it's not mixed in with manager | 16:45 |
| dansmith | sdague: the only thing I can think of is if something changed recently that is causing us to not super() somewhere | 16:45 |
| sdague | https://github.com/openstack/nova/blob/b96d4ad15e158be35b315b75a5f3c91e13361e75/nova/network/floating_ips.py#L694 | 16:45 |
| sdague | if you build a local manager, you bypass it | 16:45 |
| dansmith | sdague: right, all those manager mixins for network work that way | 16:45 |
| pkholkin | dansmith: as I understand that code was added similar to some other place, is it true that we could have two equal rows in db? | 16:45 |
| edleafe | bauzas: I don't know if belmoreira entered the bug yet, but he manually changed the filter to exclude deleted and saw a huge improvement | 16:46 |
| pkholkin | maybe we should add some sort to fetch the same row everytime | 16:46 |
| bauzas | edleafe: bug me when the patch is there, I'll prio that | 16:46 |
| dansmith | sdague: yeah, I guess so | 16:46 |
| dansmith | sdague: so what changed? | 16:46 |
| sdague | dansmith: no idea | 16:47 |
| edleafe | bauzas: thx! | 16:47 |
| dansmith | pkholkin: I don't know, as I said in my review comments, and further supported by me explaining that I'm not a db guru | 16:47 |
| dansmith | pkholkin: ask jaypipes | 16:47 |
| dansmith | sdague: all of that code that I chased down looked pretty old | 16:47 |
| sdague | dansmith: but api manager seems to always do it this way | 16:47 |
| pkholkin | dansmith: ok thanks | 16:48 |
| dansmith | sdague: ....api manager... ? | 16:48 |
| dansmith | sdague: the floating_manager in network.api you mean? | 16:48 |
| dansmith | sdague: any chance this is also hostname related such that we used to make an rpc call and now we're not? | 16:50 |
| dansmith | I mean, hostname should be the same between n-api and n-net, so I don't see how this wouldn't have always been called, | 16:50 |
| dansmith | but that would explain it | 16:50 |
| dansmith | sdague: can you look to see if all the fails are non-rax or something? | 16:50 |
| *** earlephilhower has joined #openstack-nova | 16:50 | |
| *** gokrokve has quit IRC | 16:51 | |
| *** klkumar has quit IRC | 16:51 | |
| sdague | dansmith: well, logstash is down, so pattern analysis is out the window right now | 16:51 |
| sdague | this fail is hp cloud | 16:51 |
| dansmith | okay, hmm | 16:52 |
| dansmith | that seems like the most likely thing that could have caused us to take a different path in the last 24 hours | 16:52 |
| dansmith | because I don't see any other related changes | 16:52 |
| mriedem | sdague: there is a change up to fix logstash | 16:53 |
| *** paul-carlton2 has joined #openstack-nova | 16:53 | |
| *** rushiagr is now known as rushiagr_away | 16:53 | |
| sdague | dansmith: I'm also somewhat suspect of - http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-net.txt.gz | 16:53 |
| dansmith | sdague: why? | 16:54 |
| sdague | so, that's the new network node coming up | 16:54 |
| mriedem | sdague: back up to date now http://status.openstack.org/elastic-recheck/gate.html | 16:54 |
| mriedem | let's try kibana | 16:54 |
| *** rushiagr_away is now known as rushiagr | 16:54 | |
| *** gszasz has quit IRC | 16:55 | |
| mriedem | woohoo some results! | 16:55 |
| mriedem | http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22AttributeError:%20'LocalManager'%20object%20has%20no%20attribute%20'l3driver'%5C%22%20AND%20tags:%5C%22screen-n-api.txt%5C%22 | 16:55 |
| mriedem | definitely started in the last 24 hours | 16:56 |
| dansmith | sdague: ah, failed to update its service record | 16:56 |
| sdague | dansmith: is it possible the new network node never checks in after the restart | 16:56 |
| sdague | yeh | 16:56 |
| sdague | because oslo.messaging went bonkers | 16:56 |
| *** thorst has joined #openstack-nova | 16:56 | |
| dansmith | sdague: well, I dunno about that.. but the request timed out at least | 16:57 |
| *** ljxiash has quit IRC | 16:57 | |
| *** xymind has left #openstack-nova | 16:57 | |
| dansmith | sdague: the service stayed up for three more minutes, which means it probably updated itself again later | 16:57 |
| sdague | http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22closed%20the%20connection.%20Check%20login%20credentials:%20Socket%20closed%5C%22 | 16:57 |
| sdague | dansmith: maybe, though don't we send a message when we do that | 16:57 |
| dansmith | sdague: don't send a message when we do what? | 16:57 |
| sdague | sorry, log a message if we've updated? | 16:58 |
| dansmith | I see, yes | 16:58 |
| dansmith | but we'd continue to see traces here like we do for compute if it failed | 16:58 |
| sdague | so that's the huge new pile of oslo.messaging fails | 16:58 |
| cdent | bauzas: you still around? Wonder about your "hosts" comment on my filter_scheduler changes. Does it apply anywhere I'm using "filter hosts" or just the place you commented? | 16:58 |
| mriedem | https://github.com/openstack/oslo.messaging/compare/3.0.0...3.1.0 | 16:58 |
| *** sudipto_ has joined #openstack-nova | 16:59 | |
| mriedem | https://github.com/openstack/oslo.messaging/commit/daddb82788918296f8b34d6cdeb40d01620fb183 | 16:59 |
| mriedem | sdague: ^ | 16:59 |
| dims | sileht : around? ^^ | 16:59 |
| sileht | yes | 16:59 |
| *** unicell1 has joined #openstack-nova | 16:59 | |
| *** suro-patz has joined #openstack-nova | 17:00 | |
| sileht | doesn't this intraduce an issue ? | 17:00 |
| dansmith | sdague: mriedem how hard is it to get some runs with 3.0.0 again? | 17:00 |
| dims | sileht : any ideas? | 17:00 |
| mriedem | dansmith: probably hard given we don't control that in nova anymore, it's in u-c | 17:00 |
| *** ssurana has quit IRC | 17:00 | |
| *** haomaiwang has quit IRC | 17:01 | |
| dansmith | sdague: the way that crash is logged kinda looks like it takes out the service.. what's confusing is that we have two logs later during shutdown | 17:01 |
| *** unicell has quit IRC | 17:01 | |
| dansmith | sdague: but that could definitely have paralyzed the service | 17:01 |
| mriedem | do we have a bug in nova for this yet? | 17:01 |
| *** links has quit IRC | 17:01 | |
| dansmith | sdague: but it still doesn't explain how n-api got where it is, because if the service is down, it won't do the l3 bits very explicitly | 17:01 |
| *** haomaiwang has joined #openstack-nova | 17:01 | |
| sileht | dims, what the issue ? | 17:01 |
| dims | mriedem : sdague : explain a bit please? | 17:02 |
| *** sudipto has quit IRC | 17:02 | |
| dansmith | dims: we don't know yet | 17:02 |
| openstackgerrit | Ed Leafe proposed openstack/nova: WIP: Remove the opts.py from nova/scheduler https://review.openstack.org/255383 | 17:02 |
| dims | "so that's the huge new pile of oslo.messaging fails" | 17:03 |
| dansmith | sileht: we're wondering if the above linked change is causing us to fail during init_host on a node | 17:03 |
| sdague | dansmith: actually, no it totally explains it | 17:03 |
| mriedem | here is the bug https://bugs.launchpad.net/nova/+bug/1524418 | 17:03 |
| openstack | Launchpad bug 1524418 in OpenStack Compute (nova) "gate-grenade-dsvm-multinode fails with "AttributeError: 'LocalManager' object has no attribute 'l3driver'"" [Undecided,New] | 17:03 |
| sdague | https://github.com/openstack/nova/blob/b96d4ad15e158be35b315b75a5f3c91e13361e75/nova/network/floating_ips.py#L439 | 17:03 |
| sdague | https://github.com/openstack/nova/blob/b96d4ad15e158be35b315b75a5f3c91e13361e75/nova/network/floating_ips.py#L448 | 17:03 |
| dansmith | sdague: those are 404 for me | 17:03 |
| sdague | which then takes you into https://github.com/openstack/nova/blob/b96d4ad15e158be35b315b75a5f3c91e13361e75/nova/network/floating_ips.py#L455 | 17:03 |
| dims | me too | 17:03 |
| sdague | oh, dammit | 17:03 |
| sdague | local patch | 17:03 |
| sdague | one sec | 17:03 |
| sdague | https://github.com/openstack/nova/blob/b5890b3c36613919338f83c4f59225f424c99cb1/nova/network/floating_ips.py#L439 | 17:04 |
| sdague | if service returns none | 17:04 |
| sdague | then we hard code connect host = self.host | 17:04 |
| sdague | which forces us down the local path | 17:04 |
| dansmith | right, so we set interface to none | 17:04 |
| dansmith | which doesn't run the l3 code | 17:04 |
| dansmith | that's what I was saying | 17:05 |
| dansmith | see https://github.com/openstack/nova/blob/master/nova/network/floating_ips.py#L481-L484 | 17:05 |
| openstackgerrit | Mark Doffman proposed openstack/nova: Make os-instance-action read deleted instances. https://review.openstack.org/253087 | 17:05 |
| dansmith | it's really close, but it shouldn't be happening I think | 17:05 |
| sdague | right, but running down the local path always fails? | 17:05 |
| sdague | I don't actually see how using LocalManager can work | 17:05 |
| dansmith | it fails to run the l3 bits, but it shouldn't be running the l3 bits | 17:05 |
| sdague | given that it never loads l3driver | 17:06 |
| dansmith | oh wait | 17:06 |
| dansmith | if multihost=false | 17:06 |
| dansmith | then we will run it local | 17:06 |
| *** Marga_ has quit IRC | 17:06 | |
| dansmith | because L450-451 | 17:06 |
| sdague | oh, right, and we've nulled out interface in the multihost true | 17:07 |
| dansmith | network.multi-host that is | 17:07 |
| sdague | yeh | 17:07 |
| dansmith | we can't null it out in the non-multihost case and still work, but we also fail here in api if not | 17:07 |
| dansmith | so I'm guessing maybe we're multihost=true in the config, | 17:07 |
| dansmith | but not on the network | 17:08 |
| *** diogogmt has joined #openstack-nova | 17:08 | |
| dansmith | and if that's the case and n-net is down, then we hit this path? | 17:08 |
| dansmith | I wonder if we can just do "or CONF.multihost" up there | 17:08 |
| *** thangp has quit IRC | 17:08 | |
| *** daemontool_ has quit IRC | 17:08 | |
| dansmith | CONF.multihost means every network is multihost It hink | 17:08 |
| mriedem | where is multi_host set on the network object? shouldn't that be based on the config to begin with? | 17:09 |
| sdague | we're multihost true in the config | 17:09 |
| dansmith | mriedem: I think you can have some networks multihost and some not, if multihost=false | 17:09 |
| dansmith | although I'm sure you end up with lots of fails like this if you do | 17:10 |
| pkholkin | alaski: could you please ping me when you research this problem with args, it is needed for that patch | 17:10 |
| mriedem | https://github.com/openstack/nova/blob/master/nova/network/manager.py#L1246 | 17:10 |
| sdague | dansmith: ok, are you saying that network.multihost is coming back wrong? | 17:10 |
| mriedem | look for this? https://github.com/openstack/nova/blob/master/nova/network/manager.py#L1266 | 17:11 |
| alaski | pkholkin: yep. I see why it's failing for you, but now I have to go understand why it worked elsewhere | 17:11 |
| dansmith | sdague: I'm saying I think that's a nullable property and might not be set | 17:11 |
| dansmith | and it defaults false | 17:11 |
| dansmith | https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/models.py#L780-L780 | 17:11 |
| *** Marga_ has joined #openstack-nova | 17:11 | |
| *** rcernin has quit IRC | 17:11 | |
| *** zhenq has joined #openstack-nova | 17:12 | |
| *** Marga_ has quit IRC | 17:12 | |
| dansmith | just to be clear here, | 17:12 |
| dansmith | I think we have two bugs | 17:12 |
| dansmith | one is o.m crashing us and two is us handling it improperly in this network stuff | 17:12 |
| *** jlanoux has quit IRC | 17:13 | |
| sdague | so the difference between the good runs and the bad runs is n-net having the o.m stacktrace in them | 17:13 |
| dansmith | https://github.com/openstack/nova/blob/master/nova/network/manager.py#L1369-L1369 | 17:13 |
| dansmith | yeah | 17:13 |
| *** diogogmt_ has joined #openstack-nova | 17:13 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Get list of disks to copy early to avoid multiple DB hits https://review.openstack.org/252506 | 17:13 |
| sdague | yeh, I could accept we have > 1 bugs here | 17:13 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Allow block live migration of an instance with attached volumes https://review.openstack.org/227278 | 17:13 |
| *** diogogmt has quit IRC | 17:13 | |
| *** diogogmt_ is now known as diogogmt | 17:13 | |
| mriedem | from this: http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-api.txt.gz#_2015-12-09_12_44_41_250 | 17:13 |
| mriedem | it looks like the network object does have multi_host=True | 17:13 |
| *** aginwala has joined #openstack-nova | 17:13 | |
| dansmith | I think this network stuff hasn't changed in a looong time, but we're only hitting it if n-net fails to start | 17:14 |
| mriedem | i have a hard time parsing serialized json in the logs though | 17:14 |
| sdague | dansmith: I would agree | 17:14 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Enable live migration with iso9660 config drive format https://review.openstack.org/234659 | 17:14 |
| sileht | about https://github.com/openstack/oslo.messaging/commit/daddb82788918296f8b34d6cdeb40d01620fb183, (if it doesn't introduce a bug) it can change how the replies message are ordered on the queues when the rabbitmq connections is fully used | 17:14 |
| dansmith | mriedem: that's in info_cache | 17:14 |
| dansmith | sdague: we don't have a db dump after this run, right? | 17:15 |
| sdague | dansmith: we do not | 17:15 |
| *** mgoddard has joined #openstack-nova | 17:15 | |
| sdague | I can look at doing that dump | 17:15 |
| dansmith | sdague: dumps are "taken" not "done", but .. whatever | 17:16 |
| sdague | :) | 17:16 |
| sdague | anyway, making the tools take such a dump, they don't today | 17:16 |
| *** BobBall is now known as BobBall_AWOL | 17:16 | |
| dansmith | yeah, that'd be nice | 17:16 |
| sdague | but the root trigger is the hugely increased failure rate by o.m | 17:17 |
| sdague | which is probably triggered on the multinode jobs because there are more services talking to it (multiple nodes)? | 17:17 |
| sdague | though doesn't seem like that many more connectings | 17:17 |
| sdague | connections | 17:17 |
| dims | sileht : take a look at this job? http://logs.openstack.org/92/253192/6/check/gate-grenade-dsvm/76ad561/logs/new/screen-n-cond.txt.gz | 17:18 |
| sdague | sileht: so semantically that means the clients need to retry for themselves? | 17:18 |
| sileht | sdague, nothing change for the client | 17:18 |
| *** jistr has quit IRC | 17:19 | |
| sileht | sdague, I have just moved a piece of code to an upper layer to be able to release an unused tcp connection | 17:19 |
| dims | sileht : i see lots of ERROR reports http://logs.openstack.org/92/253192/6/check/gate-grenade-dsvm/76ad561/logs/rabbitmq/rabbit@devstack-trusty-rax-iad-6449917.txt.gz | 17:19 |
| mriedem | dims: that's a single node grenade job so those could be when we're upgrading services | 17:20 |
| sdague | mriedem: nope | 17:20 |
| *** ianbrown has quit IRC | 17:20 | |
| sdague | it's single node, all services are offline for upgrade | 17:20 |
| sdague | then restarted | 17:20 |
| sdague | it has the same n-net fail as well | 17:21 |
| dims | mriedem : is there another set of logs you are looking at? | 17:21 |
| mriedem | http://logs.openstack.org/38/249138/9/check/gate-grenade-dsvm-multinode/5b991dc/logs/new/screen-n-api.txt.gz?level=TRACE | 17:21 |
| *** Apoorva has joined #openstack-nova | 17:21 | |
| mriedem | dims: ^ | 17:21 |
| sdague | mriedem: right, but dims fail is actually the same fail, but on a single node environment | 17:22 |
| dansmith | can't we just go back to 3.0.0 and see if we stop seeing this? | 17:22 |
| sdague | so I think the issue is about disconnecting and reconnecting to o.m | 17:22 |
| dims | sileht : so can you look in the set i pointed to please? | 17:22 |
| dims | i have to drop off back in a bit | 17:22 |
| sileht | dims, I'm in | 17:22 |
| *** ianbrown has joined #openstack-nova | 17:23 | |
| sdague | dansmith: it's a statistical fail, so maybe | 17:23 |
| dims | thanks! if i don't go now i'll have to be hungry in a place i can't speak/read the language :) | 17:23 |
| alaski | pkholkin: got it. I commented on the review | 17:23 |
| sdague | but it's going to require a bunch of data | 17:23 |
| pkholkin | alaski: the same for me :) thanks) | 17:23 |
| pkholkin | alaski: oh, ok) | 17:23 |
| jaypipes | dansmith: just looked at that bw_usage PK patch from pkholkin... there is actually a data corruption bug there, but that patch itself only removes a misleading code comment. the real fix for the data corruption bug would be first to ensure an orderby() is set on the SQLAlchemy query object returned on the line that calls first() on the query object and add a nova-manage bw_usage_cleanup (or similar) command that would clean up bad records in the D | 17:23 |
| jaypipes | B. not super-important because the surface area of the bug is very small (race condition but only exposed for the duration of the bandwidth collection at any given time) | 17:23 |
| *** dims has quit IRC | 17:24 | |
| pkholkin | alaski: where are you comments)? | 17:24 |
| * jaypipes needs to run now :( | 17:25 | |
| alaski | pkholkin: heh, didn't save them yet | 17:25 |
| *** smoriya has quit IRC | 17:25 | |
| jaypipes | so does pkholkin :) | 17:25 |
| sileht | sdague, that looks weird rabbitmq said "connection_closed_abruptly" | 17:25 |
| dansmith | jaypipes: yeah, I don't really care, he was asking me and I didn't notice, so.. cool | 17:25 |
| dansmith | s/notice/know/ | 17:25 |
| jaypipes | dansmith: ok... | 17:25 |
| *** jaypipes has quit IRC | 17:25 | |
| pkholkin | alaski: ok thanks, I will look | 17:26 |
| *** unicell1 has quit IRC | 17:27 | |
| openstackgerrit | dstepanenko proposed openstack/nova: Modified the nova.objects.ComputeNode object to load on-demand the pci_device_pools field by a subquery instead of pulling from the compute_nodes.pci_stats field in the database. Added get_by_compute_node method to PciDevicePoolList which get all the pci https://review.openstack.org/247692 | 17:27 |
| sileht | sdague, also on nova-network we see a MessagingTimeout, reply is received by nova-network but too late | 17:28 |
| sileht | 3 minutes later ... | 17:28 |
| sileht | that huge | 17:28 |
| *** Sukhdev has quit IRC | 17:29 | |
| mriedem | dansmith: we might want to drop this to debug level http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22Automatically%20selected%20compute%20RPC%20version%5C%22 | 17:30 |
| *** salv-orlando has quit IRC | 17:30 | |
| *** thangp has joined #openstack-nova | 17:30 | |
| mriedem | 1.3 million hits in 24 hours | 17:30 |
| dansmith | I have to run a call, back later | 17:30 |
| *** gokrokve has joined #openstack-nova | 17:30 | |
| *** claudiub has joined #openstack-nova | 17:31 | |
| sdague | http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22Timed%20out%20waiting%20for%20a%20reply%20to%20message%20ID%5C%22%20AND%20tags:screen-n-net.txt | 17:32 |
| sdague | yeh, this is clearly a spike in n-net not checking in, nearly always a failure, and only hit on grenade jobs | 17:32 |
| sdague | sileht: is the o.m code assuming graceful shutdown of services, so they clean stuff up? | 17:33 |
| sileht | sdague, if you call stop that wait yes | 17:34 |
| sdague | because we're forcably killing services, as there was an o.service bug that was not guarunteeing services go down | 17:34 |
| *** mgoddard has quit IRC | 17:34 | |
| sileht | sdague, like SIGINT ? | 17:34 |
| sdague | which is probably the "connection_closed_abruptly" | 17:34 |
| sdague | well a double SIGTERM | 17:34 |
| sdague | but same effect | 17:34 |
| sdague | mriedem: did you register a base bug for this yet? | 17:35 |
| mriedem | https://bugs.launchpad.net/oslo.messaging/+bug/1524418 | 17:35 |
| openstack | Launchpad bug 1524418 in oslo.messaging "gate-grenade-dsvm-multinode fails with "AttributeError: 'LocalManager' object has no attribute 'l3driver'"" [Undecided,New] | 17:35 |
| mriedem | i sure did | 17:35 |
| sileht | ok that explain the "connection_closed_abruptly" | 17:35 |
| *** suro-patz has quit IRC | 17:36 | |
| claudiub | alaski: hi. I've replied to your comments on the expose host capabilities spec. I've also added a comment with a quick final patch POC for it, as you requested. https://review.openstack.org/#/c/222200 | 17:36 |
| *** ssurana has joined #openstack-nova | 17:36 | |
| *** armax_ has joined #openstack-nova | 17:38 | |
| *** hemna is now known as hemnafk | 17:38 | |
| *** matrohon has quit IRC | 17:39 | |
| sileht | sdague, is rabbit restarted during upgrade ? | 17:39 |
| openstackgerrit | Chris Dent proposed openstack/nova: Improve Filter Scheduler doc clarity https://review.openstack.org/254394 | 17:39 |
| * sileht guess not | 17:39 | |
| *** sfinucan has quit IRC | 17:40 | |
| sdague | sileht: no | 17:40 |
| dansmith | mriedem: well, the reason I left it at info was because I want people that are going to try it for mitaka to be able to go and verify that it looks right in the logs | 17:41 |
| sdague | dansmith: right, it just seems like it emits a lot | 17:41 |
| dansmith | mriedem: making it debug won't make it appear any less often, right? it only happens on service start (and now SIGHUP) so not really a big deal I think | 17:41 |
| mriedem | 1.3 million per day is a lot | 17:41 |
| mriedem | well, that's for upstream CI | 17:41 |
| sdague | dansmith: http://logs.openstack.org/92/253192/6/check/gate-grenade-dsvm/76ad561/logs/new/screen-n-api.txt.gz#_2015-12-09_16_50_42_905 | 17:41 |
| dansmith | it'll happen any time any code creates an rpc object.. maybe that's happening somewhere dynamically? | 17:42 |
| sdague | that seems a little excessive :) | 17:42 |
| dansmith | sdague: ah, that's because they fork first | 17:42 |
| dansmith | I guess conductor will do that too | 17:42 |
| sdague | there are 392 iterations of that line in n-api | 17:42 |
| sdague | that's way more than one per child | 17:43 |
| dansmith | hmm | 17:43 |
| dansmith | well, you see where it's logged from | 17:43 |
| sdague | I mean, after being told compute api is 4.5 for the 390th time, I think I've got it :) | 17:43 |
| *** changbl has quit IRC | 17:43 | |
| * dansmith rolls his eyes | 17:43 | |
| dansmith | I don't think making it debug solves the problem and it lowers the visibility of the thing, so what are your ideas? | 17:44 |
| mriedem | can we set a flag? | 17:44 |
| mriedem | 'we logged this already, don't do it again' | 17:44 |
| dansmith | that won't solve it for the fork case | 17:44 |
| dansmith | we could just use warnings, | 17:44 |
| dansmith | but it will still go once per child | 17:44 |
| sdague | once per child would be much better | 17:45 |
| *** nic has joined #openstack-nova | 17:45 | |
| dansmith | if you're okay with once per child | 17:45 |
| dansmith | okay | 17:45 |
| sdague | it seems like we should have once per child, plus once again if it changed for some reason | 17:45 |
| sdague | so cache the value that was emitted | 17:45 |
| dansmith | sure | 17:45 |
| sdague | also, when did we lose our pid in the logs? | 17:46 |
| sdague | it would be more clear if it the pid was in these log items to understand it was once per worker | 17:47 |
| sdague | http://logs.openstack.org/92/253192/6/check/gate-grenade-dsvm/76ad561/logs/new/screen-n-api.txt.gz#_2015-12-09_16_50_42_210 - it's in periodic, not sure why it's not here | 17:47 |
| *** ndipanov has quit IRC | 17:47 | |
| mriedem | sdague: very weird | 17:48 |
| mriedem | oslo.log change? | 17:48 |
| *** spandhe has joined #openstack-nova | 17:48 | |
| dansmith | you know, every one of those logs was a db hit to look up the minimum version | 17:49 |
| *** spandhe has quit IRC | 17:49 | |
| *** yonglihe has joined #openstack-nova | 17:49 | |
| dansmith | which makes me wonder what we're doing in n-api to recreate that object each time | 17:49 |
| *** salv-orlando has joined #openstack-nova | 17:49 | |
| sdague | dansmith: yeh, I don't know | 17:49 |
| danpb | does anyone know if there's a cinder backend that just stores volumes in plain files in a directory ? | 17:49 |
| mriedem | jgriffith: ^ | 17:50 |
| * danpb wants to setup a minimally invasive cinder that doesn't touch block devices or network storage at all | 17:50 | |
| mriedem | scottda: ^ | 17:50 |
| *** pratikma_ has quit IRC | 17:51 | |
| *** rk4n_ has quit IRC | 17:51 | |
| jgriffith | danpb: I had a test driver that did that... let me see if it's still there | 17:51 |
| jgriffith | danpb: I think it was abandoned/removed due to lack of interest :) | 17:52 |
| danpb | NFS looks like the closest i can get to that | 17:52 |
| danpb | i guess i could just setup a local NFS server | 17:52 |
| jgriffith | danpb: hmm... well, depending on what you're trying to do maybe | 17:52 |
| danpb | because 'mount' is clever enough to just do bind mounts if it sees the NFS server is local | 17:52 |
| *** salv-orlando has quit IRC | 17:52 | |
| danpb | jgriffith: basically i'm setting up devstack inside a container | 17:52 |
| openstackgerrit | Dan Smith proposed openstack/nova: Cut down the automatic RPC version messages to once per version https://review.openstack.org/255404 | 17:52 |
| danpb | jgriffith: so it'll have no ability to use block devices at all | 17:53 |
| jgriffith | danpb: well, that's not entirely true... but up to you | 17:53 |
| openstackgerrit | dstepanenko proposed openstack/nova: Added support_requests method to PciDevicePoolList https://review.openstack.org/242087 | 17:53 |
| sdague | danpb: last time I poked at that, even if you sort that out, the iscsi handling was a blocker | 17:53 |
| jgriffith | danpb: it is possible to plumb in things to use other drivers... but probably more effort than you want to put forth :) | 17:53 |
| jgriffith | sdague: yes, the iscsi part is the trick... it's possible but it's rather ugly and a PITA | 17:54 |
| jgriffith | sdague: danpb and frankly in my setup it wasn't very reliable | 17:54 |
| danpb | sdague: i would think i could use the NFS volumebackend in nova so it wouldn't need iscsi | 17:55 |
| *** tonytan4ever has quit IRC | 17:55 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Don't load deleted instances https://review.openstack.org/255405 | 17:55 |
| jgriffith | danpb: anyway, based on what you're saying the file driver I had wouldn't be much use for you anyway | 17:55 |
| mriedem | dansmith: sdague: opened a bug for those chatty auto-version service logs https://bugs.launchpad.net/nova/+bug/1524444 - not sure what we want to do yet, but we should look into it at some point | 17:55 |
| openstack | Launchpad bug 1524444 in OpenStack Compute (nova) "auto-select rpc version info logging is too chatty" [Undecided,New] | 17:55 |
| danpb | (i guess that does assume cinder lets you fully turn off iscsi when not needed which may not be true) | 17:55 |
| sileht | sdague, dansmith, mriedem https://review.openstack.org/#/c/255403/ will reduce the excessive and useless AMQPDestinationNotFound logging | 17:55 |
| dansmith | mriedem: I just pushed a bug | 17:55 |
| dansmith | mriedem: er, a patch | 17:56 |
| *** lpetrut has quit IRC | 17:56 | |
| *** bauzas has quit IRC | 17:56 | |
| dansmith | mriedem: https://review.openstack.org/255404 | 17:56 |
| jgriffith | danpb: it does not | 17:56 |
| mriedem | dansmith: ok can we link to the bug then please? | 17:56 |
| danpb | maybe i'll just disable cinder for short term until i can investigate it more | 17:56 |
| jgriffith | danpb: although there is a local driver now... | 17:56 |
| jgriffith | danpb: that might actually do what you want | 17:56 |
| jgriffith | danpb: https://github.com/openstack/cinder/blob/master/cinder/volume/drivers/block_device.py | 17:56 |
| jgriffith | danpb: FTR I hate that driver and want it to die a horrible death, but it's there | 17:57 |
| openstackgerrit | Dan Smith proposed openstack/nova: Cut down the automatic RPC version messages to once per version https://review.openstack.org/255404 | 17:57 |
| danpb | jgriffith: hehe, ok | 17:57 |
| jgriffith | danpb: it would do what you want though I think | 17:57 |
| dansmith | mriedem: I don't understand why we need double accounting for something like this but... done ^ | 17:57 |
| jgriffith | danpb: assuming anybody has tested/updated it in the last 2 years | 17:57 |
| *** smurke has quit IRC | 17:57 | |
| paul-carlton2 | mriedem, would you have time to look at https://review.openstack.org/#/c/220622 | 17:59 |
| *** hemnafk is now known as hemna | 17:59 | |
| *** markus_z has quit IRC | 17:59 | |
| paul-carlton2 | If we can get some reviews this should be ready to merge now | 18:00 |
| sileht | sdague, dansmith, mriedem it's highly possible that this AMQPDestinationNotFound is unrelated to the issue it was already exists before as LOG.info() message and just due to the nova-XXX serivces restart | 18:00 |
| *** haomaiwang has quit IRC | 18:01 | |
| sileht | sdague, dansmith mriedem only the last occurence is an real error because it come from a newly started service (n-cpu) | 18:01 |
| *** haomaiwang has joined #openstack-nova | 18:01 | |
| *** changbl has joined #openstack-nova | 18:02 | |
| *** moshele has quit IRC | 18:02 | |
| sileht | sdague, dansmith mriedem the valid lost reply is due to rabbitmq that close the connection because it miss a heartbeat | 18:03 |
| openstackgerrit | Dan Smith proposed openstack/nova: DNM: Figure out where we create ComputeAPI instances https://review.openstack.org/255410 | 18:03 |
| *** rushiagr is now known as rushiagr_away | 18:05 | |
| lifeless | ttx: hi; whats up? | 18:06 |
| *** browne has joined #openstack-nova | 18:06 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove the opts.py from nova/scheduler https://review.openstack.org/255383 | 18:06 |
| *** rushiagr_away is now known as rushiagr | 18:07 | |
| *** aginwala has quit IRC | 18:14 | |
| *** nic has quit IRC | 18:15 | |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: DNM: Making project_id optional in v2.1 urls https://review.openstack.org/233076 | 18:16 |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: Generate api sample data without project id's to support v2.13 https://review.openstack.org/254484 | 18:16 |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id https://review.openstack.org/254485 | 18:16 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/253901 | 18:16 |
| *** unicell has joined #openstack-nova | 18:17 | |
| *** markvoelker has quit IRC | 18:18 | |
| *** rushiagr is now known as rushiagr_away | 18:18 | |
| *** pratikmallya has joined #openstack-nova | 18:18 | |
| *** tonytan4ever has joined #openstack-nova | 18:20 | |
| *** smurke has joined #openstack-nova | 18:21 | |
| *** markvoelker has joined #openstack-nova | 18:21 | |
| *** jinxing has quit IRC | 18:23 | |
| paul-carlton2 | tonyb, Thanks for completing https://review.openstack.org/#/c/220622, I've tested it just need reviews now | 18:23 |
| rlrossit | dansmith: Is https://review.openstack.org/#/c/255404/ going to clean up the dirtying of the version variable in the service object? | 18:24 |
| rlrossit | (https://github.com/openstack/nova/blob/master/nova/objects/service.py#L268-L270 for reference) | 18:24 |
| *** spandhe has joined #openstack-nova | 18:24 | |
| *** ccarmack has quit IRC | 18:25 | |
| rlrossit | well it looks like that will just clean up the logging, not anything else... | 18:25 |
| *** ducttape_ has quit IRC | 18:26 | |
| *** vilobhmm11 has joined #openstack-nova | 18:28 | |
| *** edmondk has quit IRC | 18:29 | |
| *** vilobhmm11 has quit IRC | 18:29 | |
| *** rk4n has joined #openstack-nova | 18:30 | |
| *** kfarr has quit IRC | 18:31 | |
| *** paul-carlton2 has quit IRC | 18:33 | |
| *** rk4n has quit IRC | 18:33 | |
| dansmith | rlrossit: we need the version to be dirty every time | 18:35 |
| dansmith | it's kindof the point | 18:35 |
| *** suro-patz has joined #openstack-nova | 18:37 | |
| *** e0ne has quit IRC | 18:41 | |
| sileht | sdague, dansmith mriedem I think https://review.openstack.org/255435 will fix the oslo.messaging side of https://bugs.launchpad.net/nova/+bug/1524418 | 18:41 |
| openstack | Launchpad bug 1524418 in oslo.messaging "gate-grenade-dsvm-multinode fails with "AttributeError: 'LocalManager' object has no attribute 'l3driver'"" [Undecided,In progress] - Assigned to Mehdi Abaakouk (sileht) | 18:41 |
| *** burgerk has joined #openstack-nova | 18:41 | |
| dansmith | sileht: okay | 18:41 |
| *** aginwala has joined #openstack-nova | 18:42 | |
| mriedem | sileht: reasoning? | 18:42 |
| mriedem | we just weren't waiting long enough? | 18:42 |
| mriedem | sileht: i guess the assumptions made in https://review.openstack.org/#/c/252265/ were really about rabbit dying | 18:43 |
| sileht | when rabbitmq break the connection because of missing heartbeat before we retry to send replies during 60 seconds this change set the default to 5 second | 18:43 |
| mriedem | in our upgrade, rabbit doesn't die, | 18:43 |
| *** aix has quit IRC | 18:43 | |
| sileht | mriedem, but the rabbit log said it closes the connection because it misses heartbeat, and then oslo.messaging don't send the reply because the connection take more than 5 seconds to be reeastablisehd | 18:44 |
| sileht | mriedem, so the 'missing heartbeat' issue is here for a long time bug mitigated by the reply retry meachisnm | 18:45 |
| *** lpetrut has joined #openstack-nova | 18:45 | |
| sileht | bbl | 18:45 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove the opts.py from nova/scheduler https://review.openstack.org/255383 | 18:45 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Move nova/opts.py to nova/conf/ https://review.openstack.org/255436 | 18:45 |
| *** Marga_ has joined #openstack-nova | 18:46 | |
| openstackgerrit | Zhen Qin proposed openstack/nova: Fix the bug that when running nova list --host with invalid hostname (either no such server or there's a typo in server name), it should return an error. https://review.openstack.org/255047 | 18:47 |
| *** Marga__ has joined #openstack-nova | 18:49 | |
| *** rdopiera has quit IRC | 18:49 | |
| *** ducttape_ has joined #openstack-nova | 18:50 | |
| *** Marga_ has quit IRC | 18:50 | |
| rlrossit | dansmith: wait... why do we need it to be dirty all the time? don't we only want to dirty it when it changed from the last time? Or is the Service object kinda like a singleton thing where we only make a new one when we get a new version? | 18:50 |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: Remove hardcoding namespaces for monitors https://review.openstack.org/255438 | 18:50 |
| *** Marga__ has quit IRC | 18:50 | |
| *** Marga_ has joined #openstack-nova | 18:51 | |
| dansmith | rlrossit: it's only dirty if we don't query a service object from the DB, which is the case where we need to make sure an update or create puts the current version in there | 18:52 |
| *** aginwala has quit IRC | 18:54 | |
| rlrossit | oh so is the noop workaround in https://github.com/openstack/nova/blob/master/nova/objects/service.py#L267-L271 only valid when it's the first save? other than that the changes get reset so version isn't dirty anymore? | 18:55 |
| rlrossit | dansmith: ^ | 18:55 |
| dansmith | rlrossit: right | 18:55 |
| *** gokrokve has quit IRC | 18:57 | |
| rlrossit | dansmith: that noop workaround made me think this happened more than just once :) | 18:57 |
| *** danpb has quit IRC | 18:57 | |
| dansmith | well, the thing is, we almost only ever call save() when we really have something to change | 18:58 |
| dansmith | and if we're going to save, we might as well send our version to make sure it's right.. it's free | 18:58 |
| *** ctrath has quit IRC | 18:59 | |
| dansmith | so that noop is there the keep the save() semantic of the other objects, but I'm not sure it ever really gets exercised | 18:59 |
| *** bauzas has joined #openstack-nova | 18:59 | |
| *** gokrokve has joined #openstack-nova | 18:59 | |
| *** gokrokve has quit IRC | 18:59 | |
| rlrossit | dansmith: well it better be used, or else https://review.openstack.org/#/c/245678/17 isn't needed :) | 18:59 |
| dansmith | um, what? | 19:00 |
| *** david-ly_ has joined #openstack-nova | 19:00 | |
| *** ducttape_ has quit IRC | 19:00 | |
| *** haomaiwang has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-nova | 19:01 | |
| *** ducttape_ has joined #openstack-nova | 19:01 | |
| rlrossit | save() better be used, or else we'll never emit a notification (not sure if you mean it being exercised as the if check or the save()) | 19:02 |
| *** zenoway has quit IRC | 19:02 | |
| dansmith | no, save is used of course | 19:03 |
| *** moshele has joined #openstack-nova | 19:03 | |
| dansmith | what I meant was, | 19:03 |
| *** mgoddard has joined #openstack-nova | 19:03 | |
| dansmith | I'm not sure it's ever called with nothing changed (other than version, obligatorily) | 19:03 |
| *** david-ly_ is now known as david-lyle_ | 19:03 | |
| *** moshele has quit IRC | 19:03 | |
| rlrossit | oooohhhhh... gotcha | 19:03 |
| *** moshele has joined #openstack-nova | 19:03 | |
| * rlrossit shuts up and goes back to the corner | 19:03 | |
| *** david-lyle has quit IRC | 19:03 | |
| dansmith | o | 19:04 |
| dansmith | m | 19:04 |
| dansmith | g | 19:04 |
| * dansmith has leftover tacos for lunch | 19:04 | |
| *** fawadkhaliq has quit IRC | 19:04 | |
| *** gokrokve has joined #openstack-nova | 19:04 | |
| *** david-lyle_ is now known as david-lyle | 19:04 | |
| rlrossit | dansmith: don't overeat. It turns your brain to mush | 19:05 |
| *** ctrath has joined #openstack-nova | 19:05 | |
| rlrossit | I've been stuffing my face with food for the last 2 hours | 19:05 |
| dansmith | it's december, I'm allowed. | 19:05 |
| *** salv-orlando has joined #openstack-nova | 19:05 | |
| *** gokrokve has quit IRC | 19:05 | |
| rlrossit | I haven't heard mriedem on here in a while, he's probably in a food coma in his office | 19:06 |
| *** gokrokve has joined #openstack-nova | 19:06 | |
| mriedem | parsing ndipanov's comments on my change | 19:07 |
| *** rotbeard has quit IRC | 19:07 | |
| rlrossit | mriedem: when you're done, go reward yourself with a cookie... or like 12 | 19:07 |
| mriedem | already did that | 19:08 |
| rlrossit | well there's still like a billion left | 19:08 |
| *** doug-fish has quit IRC | 19:08 | |
| *** paul-carlton_ has joined #openstack-nova | 19:08 | |
| *** doug-fish has joined #openstack-nova | 19:09 | |
| *** Apoorva has quit IRC | 19:09 | |
| *** derekh has quit IRC | 19:10 | |
| *** gokrokve has quit IRC | 19:10 | |
| mriedem | alaski: unless ndipanov is talking about tasks, which isn't a thing yet, do you understand the locking he's looking for in https://review.openstack.org/#/c/254428/ ? | 19:10 |
| *** aginwala has joined #openstack-nova | 19:10 | |
| *** doug-fish has quit IRC | 19:14 | |
| *** gyee has joined #openstack-nova | 19:14 | |
| *** zenoway has joined #openstack-nova | 19:14 | |
| *** jichen has joined #openstack-nova | 19:14 | |
| *** e0ne has joined #openstack-nova | 19:15 | |
| *** lykinsbd has quit IRC | 19:15 | |
| *** vilobhmm11 has joined #openstack-nova | 19:16 | |
| *** vilobhmm11 has quit IRC | 19:16 | |
| *** READ10 has quit IRC | 19:16 | |
| *** ijw has quit IRC | 19:19 | |
| alaski | mriedem: I would guess the same as you did, a synchronization decorator on it | 19:19 |
| mriedem | ok, so that's already there | 19:20 |
| *** garyk has quit IRC | 19:20 | |
| mriedem | and it doesn't help | 19:20 |
| mriedem | because that's just per-process right? | 19:20 |
| mriedem | rlrossit: i do need coffee and there is none in the food room, that's a major failure on the planning part | 19:20 |
| rlrossit | there's egg nog | 19:21 |
| mriedem | gross | 19:21 |
| rlrossit | get noggin' | 19:21 |
| mriedem | i will not consume something with the word 'nog' in it | 19:21 |
| alaski | mriedem: I thought it was per host through a lock file | 19:21 |
| mriedem | s/nog/phlegm/ | 19:21 |
| *** devananda has quit IRC | 19:22 | |
| *** garyk has joined #openstack-nova | 19:22 | |
| *** doug-fish has joined #openstack-nova | 19:23 | |
| mriedem | alaski: https://github.com/openstack/oslo.concurrency/blob/master/oslo_concurrency/lockutils.py#L183 | 19:23 |
| *** doug-fish has quit IRC | 19:23 | |
| mriedem | external=False by default | 19:23 |
| mriedem | so we only lock per process | 19:23 |
| openstackgerrit | Tomi Juvonen proposed openstack/nova: Add host_status attribute for servers/detail and servers/{server_id} https://review.openstack.org/241562 | 19:23 |
| *** doug-fish has joined #openstack-nova | 19:23 | |
| *** Apoorva has joined #openstack-nova | 19:24 | |
| alaski | mriedem: how many processes are running here? | 19:24 |
| mriedem | hmm, i guess we do just have one n-cpu worker | 19:25 |
| alaski | oh, I think the problem may be that delete causes the build to bail early and release the lock | 19:27 |
| *** wanghao has quit IRC | 19:27 | |
| gjayavelu | johnthetubaguy: Hi, would this require a spec? https://blueprints.launchpad.net/nova/+spec/vmware-resource-pool | 19:27 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: virt: Allow multiple volume attachments https://review.openstack.org/193135 | 19:27 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: WIP: REST API changes to support multi-attach https://review.openstack.org/153038 | 19:27 |
| alaski | mriedem: though the state can be anything at that point | 19:27 |
| alaski | mriedem: compute/api updates the task_state of the instance causing the early break | 19:28 |
| mriedem | alaski: yeah like here http://logs.openstack.org/20/218120/3/check/gate-tempest-dsvm-full-ceph/2349f2d/logs/screen-n-cpu.txt.gz#_2015-12-07_05_51_15_251 | 19:28 |
| mriedem | aborting claim | 19:28 |
| *** angular_mike__ has quit IRC | 19:29 | |
| alaski | yeah, a few lines down is http://logs.openstack.org/20/218120/3/check/gate-tempest-dsvm-full-ceph/2349f2d/logs/screen-n-cpu.txt.gz#_2015-12-07_05_51_15_280 | 19:29 |
| mriedem | yup | 19:30 |
| *** vladikr has quit IRC | 19:31 | |
| *** scheuran has quit IRC | 19:31 | |
| *** unicell1 has joined #openstack-nova | 19:32 | |
| alaski | unless we exit that lock in a good state it isn't really helping there | 19:32 |
| *** armax_ is now known as armax | 19:32 | |
| *** unicell has quit IRC | 19:34 | |
| *** devananda has joined #openstack-nova | 19:34 | |
| mriedem | god i have a hard time following nova.compute.api._delete | 19:34 |
| *** _ducttape_ has joined #openstack-nova | 19:34 | |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: virt: Allow multiple volume attachments https://review.openstack.org/193135 | 19:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: comp: Allow multiple volume attachments https://review.openstack.org/193134 | 19:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: obj: Allow multiple volume attachments https://review.openstack.org/193133 | 19:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: db: Allow multiple volume attachments https://review.openstack.org/153033 | 19:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: WIP: REST API changes to support multi-attach https://review.openstack.org/153038 | 19:35 |
| *** angular_mike__ has joined #openstack-nova | 19:36 | |
| *** salv-orlando has quit IRC | 19:36 | |
| alaski | mriedem: looking at the build path on the compute, block devices aren't cleaned up like network resources are. possibly because we don't have the reference to the block device in some cases | 19:36 |
| mriedem | alaski: yeah we definitely get here https://github.com/openstack/nova/blob/c4ca1abb4a49bf0bce765acd3ce906bd117ce9b7/nova/compute/manager.py#L1954 | 19:37 |
| *** ducttape_ has quit IRC | 19:38 | |
| mriedem | and we can't hold up the delete request in the compute api | 19:38 |
| alaski | even if we could I'm not sure we should | 19:38 |
| alaski | we're able to clean up network resources without getting the network info back, but we can't do the same for volumes | 19:39 |
| alaski | if only we had a volume id before getting to this point :) | 19:40 |
| mriedem | alaski: well we do the volume cleanup here, but not when rolling back the build https://github.com/openstack/nova/blob/c4ca1abb4a49bf0bce765acd3ce906bd117ce9b7/nova/compute/manager.py#L2307-L2315 | 19:40 |
| mriedem | only when doing the delete | 19:40 |
| mriedem | and that was the part that was failing in the gate b/c the bdm was missing the volume_id | 19:41 |
| mriedem | jgriffith: i guess w/o a cinder api to list volumes for a given snapshot, we could list volumes and filter on snapshot_id in the client | 19:42 |
| mriedem | jgriffith: just going back to tempest doing a cleaner teardown | 19:42 |
| mriedem | because even with the 'fix' in nova, there is still a chance for a race in tempest | 19:42 |
| mriedem | alaski: so we might have a bdm.volume_id by the time the build is aborted | 19:43 |
| alaski | okay, I was just looking to see if there was a gap where we could fail and not have that | 19:43 |
| mriedem | maybe we need reservations | 19:43 |
| mriedem | in cinder | 19:44 |
| *** vladikr has joined #openstack-nova | 19:44 | |
| *** briancurtin has quit IRC | 19:46 | |
| *** briancurtin has joined #openstack-nova | 19:46 | |
| alaski | mriedem: the volume is allocated in https://github.com/openstack/nova/blob/c4ca1abb4a49bf0bce765acd3ce906bd117ce9b7/nova/compute/manager.py#L2144 right? | 19:46 |
| mriedem | alaski: yeah | 19:47 |
| mriedem | right here https://github.com/openstack/nova/blob/c4ca1abb4a49bf0bce765acd3ce906bd117ce9b7/nova/compute/manager.py#L1731 | 19:47 |
| *** mgoddard has quit IRC | 19:47 | |
| alaski | as long as block_device_mapping is correct after that return then it should get cleaned up at https://github.com/openstack/nova/blob/c4ca1abb4a49bf0bce765acd3ce906bd117ce9b7/nova/compute/manager.py#L2188 in the event of failure | 19:47 |
| mriedem | alaski: that's what my fix is doing | 19:48 |
| mriedem | well, it's at least saving the bdm.volume_id as soon as we have it back from cinder | 19:48 |
| mriedem | which is right after volume.create comes back | 19:48 |
| mriedem | so w/o reservations or something in the cinder api, i'm not sure how else to handle this | 19:49 |
| alaski | okay. but the delete can't fail the build during that part I think | 19:49 |
| mriedem | not sure what to do with locks | 19:49 |
| alaski | unless there's an instance update in there | 19:49 |
| spandhe | Hi folks, is it possible to create logical interfaces using —nic option in nova? Neutron port supports multiple fixed_ips from same subnet. But there is no way to use that functionality from nova boot as far as I know. Any options? | 19:49 |
| alaski | mriedem: I'm not sure there's a gap left after your change | 19:49 |
| mriedem | alaski: i'm not sure i'm following 'but the delete can't fail the build during that part' | 19:50 |
| *** Sukhdev has joined #openstack-nova | 19:50 | |
| mriedem | brb, coffeee | 19:50 |
| *** moshele has quit IRC | 19:50 | |
| *** gokrokve has joined #openstack-nova | 19:51 | |
| alaski | mriedem: the failure has been a delete setting task_state to deleting, which causes instance updates during the build process to raise an exception. As long as there's no instance update between asking cinder to create a volume and saving the volume_id we should have the volume_id available when the build does fail on the next instance update | 19:51 |
| *** lucasagomes is now known as lucas-dinner | 19:51 | |
| *** gokrokve has quit IRC | 19:52 | |
| *** gokrokve has joined #openstack-nova | 19:52 | |
| *** gokrokve has quit IRC | 19:53 | |
| *** mgoddard has joined #openstack-nova | 19:54 | |
| *** kevinbenton has quit IRC | 19:54 | |
| *** sudipto_ has quit IRC | 19:54 | |
| *** baoli_ has quit IRC | 19:54 | |
| *** omnipresent has joined #openstack-nova | 19:54 | |
| spandhe | melwitt: hey, is it possible to create logical interfaces using —nic option in nova? Neutron port supports multiple fixed_ips from same subnet. But there is no way to use that functionality from nova boot as far as I know. Any options? | 19:55 |
| *** john5223 is now known as zz_john5223 | 19:56 | |
| *** zenoway_ has joined #openstack-nova | 19:57 | |
| *** zenoway has quit IRC | 19:57 | |
| *** david-lyle has quit IRC | 19:58 | |
| *** ociuhandu has quit IRC | 19:58 | |
| *** zenoway has joined #openstack-nova | 19:58 | |
| *** david-lyle has joined #openstack-nova | 19:58 | |
| *** tonytan4ever has quit IRC | 19:59 | |
| *** tjones has quit IRC | 19:59 | |
| *** omnipresent has quit IRC | 19:59 | |
| *** aginwala has quit IRC | 20:01 | |
| *** haomaiwang has quit IRC | 20:01 | |
| *** ctrath has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-nova | 20:01 | |
| *** zenoway_ has quit IRC | 20:01 | |
| *** ZZelle_ has joined #openstack-nova | 20:05 | |
| *** hemna is now known as hemnalunch | 20:06 | |
| *** aginwala has joined #openstack-nova | 20:07 | |
| mriedem | alaski: you mean this? http://logs.openstack.org/20/218120/3/check/gate-tempest-dsvm-full-ceph/2349f2d/logs/screen-n-cpu.txt.gz#_2015-12-07_05_51_15_280 | 20:07 |
| *** zhenq has quit IRC | 20:08 | |
| mriedem | alaski: even if an instance update fails, the block_device code is updating the bdm table, not the instances table, so i'm not sure if those matter together | 20:08 |
| alaski | mriedem: right. since that can't happen, afaict, between calling cinder to get the volume and actually getting it we should be good | 20:08 |
| *** cdent has quit IRC | 20:08 | |
| alaski | I think with the save you added, that there's not a point of failure where we won't have the volume id | 20:10 |
| *** salv-orlando has joined #openstack-nova | 20:10 | |
| mriedem | now if the instance delete comes in while we're waiting for cinder to respond on the vol create request, then we could still race and not have bdm.volume_id set | 20:10 |
| alaski | excluding cinder failures | 20:10 |
| mriedem | which is what i think ndipanov was asking we log as a warning in _shutdown_instances | 20:10 |
| mriedem | realistically, the only way to avoid that is to create the volumes in the api | 20:10 |
| mriedem | before building the instance | 20:10 |
| mriedem | which we actually talked about when we were talking about removing boot from volume, back when we were talking about the AZ stuff being a gd mess | 20:11 |
| alaski | but the delete shouldn't get past the lock that build_and_run is holding | 20:11 |
| *** Marga_ has quit IRC | 20:11 | |
| alaski | until the build fails | 20:11 |
| alaski | at which point volume_id should be set | 20:11 |
| *** Marga_ has joined #openstack-nova | 20:12 | |
| alaski | creating the volumes in the api, or conductor, would be better than doing it on the compute | 20:12 |
| mriedem | yeah, because the only thing we need to do on the compute is the attach | 20:13 |
| *** kevinbenton_ has joined #openstack-nova | 20:13 | |
| mriedem | and the block_device code just gets the volume by id from cinder when doing that anyway | 20:13 |
| mriedem | compute can still wait for the volume to be available if needed | 20:13 |
| mriedem | but if we created the volume early, we'd at least have the id for cleanup later | 20:13 |
| mriedem | same with failing in the compute API if the AZ is wrong (doesn't work in cinder i mean) | 20:14 |
| mriedem | rather than NoValidHost in the compute | 20:14 |
| *** kevinbenton_ has quit IRC | 20:14 | |
| *** kevinbenton has joined #openstack-nova | 20:14 | |
| mriedem | so, i'm going to rev this to address nikola's comment about logging a warning | 20:14 |
| mriedem | anything longer term like moving vol create into the api or conductor is going to be a bp i think | 20:15 |
| alaski | okay. I'm still not seeing the gap he's seeing, but if it does happen a warning would be good | 20:15 |
| mriedem | yeah, and a warning will be trackable in logstash | 20:15 |
| dansmith | alaski: the flavor migration patch isn't on this list: https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking | 20:15 |
| *** rfolco has joined #openstack-nova | 20:16 | |
| alaski | hmm, I may have added it to the other one by mistake | 20:16 |
| jgriffith | mriedem: so the one wonky way around it is snapshot list which includes parent-volume ids in the list | 20:16 |
| dansmith | alaski: the spec one? | 20:16 |
| alaski | yeah | 20:17 |
| mriedem | jgriffith: ok, i think we can try to cross that bridge in tempest if the nova change doesn't help | 20:17 |
| * dansmith doesn't like that there are two | 20:17 | |
| *** xyang1 has quit IRC | 20:17 | |
| *** xyang1 has joined #openstack-nova | 20:17 | |
| alaski | are you also long lived, and fight with swords? | 20:18 |
| alaski | dansmith: added them | 20:18 |
| dansmith | alaski: was that intended for me? | 20:18 |
| *** tonytan4ever has joined #openstack-nova | 20:18 | |
| mriedem | heh, i was wondering about that too | 20:18 |
| alaski | dansmith: yeah, it was a highlander joke | 20:19 |
| alaski | "there can be only one" | 20:19 |
| mriedem | prone to beheadings | 20:19 |
| dansmith | I've never watched it | 20:19 |
| mriedem | better add it to the list | 20:19 |
| *** aginwala has quit IRC | 20:19 | |
| mriedem | btw, i watched commando again this last weekend | 20:19 |
| mriedem | classic | 20:19 |
| mriedem | https://www.youtube.com/watch?v=B9aiYqSRfCM | 20:20 |
| jgriffith | mriedem: LOL.. "let off some steam Bennett" | 20:21 |
| *** rfolco has quit IRC | 20:21 | |
| *** lykinsbd has joined #openstack-nova | 20:23 | |
| mriedem | alaski: i think this is actually where the UnexpectedDeletingTaskStateError comes from? https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L2102 | 20:24 |
| *** gokrokve has joined #openstack-nova | 20:24 | |
| openstackgerrit | Brian Haley proposed openstack/nova: Disable IPv6 on bridge devices https://review.openstack.org/198054 | 20:24 |
| *** aginwala has joined #openstack-nova | 20:24 | |
| *** armax has quit IRC | 20:25 | |
| alaski | mriedem: it could be line 2009 or 2904 as well | 20:25 |
| mriedem | yeah | 20:26 |
| *** gokrokve has quit IRC | 20:27 | |
| *** gokrokve has joined #openstack-nova | 20:28 | |
| *** RichardRaseley has joined #openstack-nova | 20:35 | |
| *** dedery has joined #openstack-nova | 20:39 | |
| dedery | mgoddard: ping | 20:39 |
| *** ctrath has joined #openstack-nova | 20:39 | |
| mgoddard | dedery: hi | 20:40 |
| *** belmoreira has joined #openstack-nova | 20:41 | |
| *** Marga_ has quit IRC | 20:42 | |
| *** gokrokve has quit IRC | 20:43 | |
| *** derekh has joined #openstack-nova | 20:45 | |
| *** derekh has quit IRC | 20:45 | |
| *** vladikr has quit IRC | 20:46 | |
| *** ssurana has quit IRC | 20:47 | |
| *** dims has joined #openstack-nova | 20:49 | |
| *** baoli has joined #openstack-nova | 20:54 | |
| *** krtaylor has quit IRC | 20:56 | |
| *** Apoorva has quit IRC | 20:57 | |
| *** ssurana has joined #openstack-nova | 20:57 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Move nova/opts.py to nova/conf/ https://review.openstack.org/255436 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Remove the opts.py from nova/scheduler https://review.openstack.org/255383 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (1) https://review.openstack.org/255512 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (2) https://review.openstack.org/255513 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (3) https://review.openstack.org/255514 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (4) https://review.openstack.org/255515 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (5) https://review.openstack.org/255516 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (6) https://review.openstack.org/255517 | 20:58 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add better help text to scheduler options (7) https://review.openstack.org/255518 | 20:58 |
| *** vladikr has joined #openstack-nova | 20:58 | |
| *** dims has quit IRC | 20:58 | |
| *** bklei_ has joined #openstack-nova | 20:58 | |
| edleafe | ^^ for bauzas :) | 20:59 |
| *** Marga_ has joined #openstack-nova | 20:59 | |
| bauzas | heh | 20:59 |
| *** Daisy has joined #openstack-nova | 21:00 | |
| openstackgerrit | Sean Dague proposed openstack/nova: WIP: refactor API samples for clarity https://review.openstack.org/255519 | 21:00 |
| *** dims has joined #openstack-nova | 21:00 | |
| *** haomaiwang has quit IRC | 21:01 | |
| *** dims has quit IRC | 21:01 | |
| *** 17WAA61QI has joined #openstack-nova | 21:01 | |
| *** bklei has quit IRC | 21:02 | |
| *** Marga_ has quit IRC | 21:03 | |
| *** Daisy has quit IRC | 21:04 | |
| *** baoli has quit IRC | 21:05 | |
| *** baoli has joined #openstack-nova | 21:06 | |
| *** raildo is now known as raildo-afk | 21:06 | |
| *** hemnalunch is now known as hemna | 21:08 | |
| *** lpetrut has quit IRC | 21:09 | |
| *** wverdugo500 has quit IRC | 21:09 | |
| *** w_verdugo has quit IRC | 21:10 | |
| *** aysyd has quit IRC | 21:11 | |
| *** thorst has quit IRC | 21:12 | |
| *** thorst has joined #openstack-nova | 21:13 | |
| *** bklei_ has quit IRC | 21:15 | |
| *** Marga_ has joined #openstack-nova | 21:15 | |
| *** bklei has joined #openstack-nova | 21:15 | |
| *** cdent has joined #openstack-nova | 21:15 | |
| *** erlon has quit IRC | 21:16 | |
| *** w_verdugo has joined #openstack-nova | 21:17 | |
| *** smatzek has quit IRC | 21:17 | |
| *** baoli has quit IRC | 21:17 | |
| *** baoli has joined #openstack-nova | 21:17 | |
| *** _ducttape_ has quit IRC | 21:19 | |
| *** eharney has quit IRC | 21:19 | |
| *** ducttape_ has joined #openstack-nova | 21:19 | |
| *** Sukhdev has quit IRC | 21:20 | |
| *** thorst has quit IRC | 21:21 | |
| *** paul-carlton_ has quit IRC | 21:22 | |
| *** Marga_ has quit IRC | 21:23 | |
| *** zhenq has joined #openstack-nova | 21:23 | |
| *** thangp has quit IRC | 21:24 | |
| *** belmoreira has quit IRC | 21:26 | |
| *** aginwala has quit IRC | 21:26 | |
| *** yarkot has joined #openstack-nova | 21:26 | |
| *** lpetrut has joined #openstack-nova | 21:27 | |
| *** vladikr has quit IRC | 21:28 | |
| *** baoli has quit IRC | 21:29 | |
| *** baoli has joined #openstack-nova | 21:30 | |
| *** lpetrut has quit IRC | 21:32 | |
| *** e0ne has quit IRC | 21:32 | |
| *** ianbrown has quit IRC | 21:33 | |
| *** mgoddard has quit IRC | 21:35 | |
| *** armax has joined #openstack-nova | 21:35 | |
| *** Redcavalier has quit IRC | 21:36 | |
| smcginnis | We're getting a grenade failure in Cinder on a global requirements update patch. | 21:38 |
| smcginnis | Failure is with nova. | 21:38 |
| smcginnis | http://logs.openstack.org/53/253753/9/check/gate-grenade-dsvm/31df984/logs/grenade.sh.txt.gz | 21:38 |
| smcginnis | <class 'oslo_messaging.exceptions.MessagingTimeout'> (HTTP 500) (Request-ID: req-12ddd755-a3a7-4130-b73f-4783f4d0c2fd) | 21:38 |
| mriedem | smcginnis: known issue | 21:38 |
| *** Apoorva has joined #openstack-nova | 21:39 | |
| smcginnis | Anyone know if this is a kn... .ok. :) | 21:39 |
| mriedem | http://status.openstack.org/elastic-recheck/index.html#1524418 | 21:39 |
| mriedem | top failure | 21:39 |
| smcginnis | mriedem: Thank you sir. | 21:39 |
| smcginnis | mriedem: Fix in the pipeline yet, or still being worked? | 21:40 |
| mriedem | should have a g-r update soonish that blocks oslo.messaging 3.1.0 | 21:40 |
| mriedem | just merge | 21:40 |
| mriedem | *just merged in the reqs repo | 21:40 |
| sdague | dansmith / auggy: because you might have opinions here - in the samples, it seems like we should replace the inconsistent use of %(host)s / openstack.example.com with actual %(versioned_compute_endpoint) / %(compute_endpoint) and such | 21:40 |
| smcginnis | Perfect, thanks. | 21:40 |
| mriedem | so expecting https://review.openstack.org/#/c/253901/ to be updtaed soon | 21:40 |
| *** baoli has quit IRC | 21:41 | |
| *** baoli has joined #openstack-nova | 21:42 | |
| auggy | sdague: that would definitely be a more clear label | 21:42 |
| auggy | clearer? more clear? hi i can't english today | 21:43 |
| claudiub | alaski: hi. I've replied to your comments on the expose host capabilities spec. I've also added a comment with a quick final patch POC for it, as you requested. https://review.openstack.org/#/c/222200 | 21:46 |
| *** baoli has quit IRC | 21:47 | |
| *** baoli has joined #openstack-nova | 21:48 | |
| alaski | claudiub: great. I'll take a look | 21:48 |
| alaski | might be tomorrow though | 21:48 |
| sdague | yeh, I'm staring at all this all over again, and if we just make those changes, it makes all the optional project id stuff a lot simpler | 21:49 |
| *** Marga_ has joined #openstack-nova | 21:51 | |
| *** zz_john5223 is now known as john5223 | 21:52 | |
| *** tonytan4ever has quit IRC | 21:56 | |
| *** yarkot has quit IRC | 21:56 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make sure bdm.volume_id is set after auto-creating volumes https://review.openstack.org/254428 | 21:57 |
| mriedem | alaski: ^ latest update, smallish delta | 21:57 |
| mriedem | see if that still floats your boat (if you have time today) | 21:57 |
| mriedem | i can make the diff in the compute manager smaller | 21:58 |
| *** baoli has quit IRC | 21:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/253901 | 21:59 |
| mriedem | smcginnis: ^ there we go | 21:59 |
| *** baoli has joined #openstack-nova | 22:00 | |
| *** Marga_ has quit IRC | 22:00 | |
| openstackgerrit | jichenjc proposed openstack/nova: TrivialFix: remove 'deleted' flag https://review.openstack.org/255562 | 22:00 |
| *** 17WAA61QI has quit IRC | 22:01 | |
| *** yarkot has joined #openstack-nova | 22:01 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make sure bdm.volume_id is set after auto-creating volumes https://review.openstack.org/254428 | 22:01 |
| *** haomaiwang has joined #openstack-nova | 22:01 | |
| notmorgan | sdague: going to look at your project_id stuff soon and the glance thing | 22:02 |
| notmorgan | sdague: FYI | 22:02 |
| sdague | notmorgan: cool | 22:02 |
| alaski | mriedem: works for me | 22:02 |
| *** pratikma_ has joined #openstack-nova | 22:02 | |
| mriedem | alaski: gotta update it quick | 22:02 |
| notmorgan | sdague: :) trying to resolve the las tof the Keystoneauth conversion in nova first | 22:02 |
| notmorgan | sdague: but the code is doing someting i don't understand yet (the catalog lookup for cinder is the sticking point) | 22:03 |
| *** pratikma_ has quit IRC | 22:03 | |
| notmorgan | sdague: i am hopeful to have it resolved todya. | 22:03 |
| notmorgan | or tomorrow | 22:03 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make sure bdm.volume_id is set after auto-creating volumes https://review.openstack.org/254428 | 22:03 |
| alaski | mriedem: I'm going to wait on ndipanov before voting | 22:03 |
| mriedem | yeah that's fair | 22:03 |
| *** nic has joined #openstack-nova | 22:03 | |
| *** pratikma_ has joined #openstack-nova | 22:03 | |
| *** pratikmallya has quit IRC | 22:03 | |
| notmorgan | that would be the followup to the options deprecation/removal | 22:03 |
| *** edmondsw has quit IRC | 22:04 | |
| *** bauzas has quit IRC | 22:05 | |
| *** bauzas has joined #openstack-nova | 22:05 | |
| *** jichen has quit IRC | 22:07 | |
| *** baoli has quit IRC | 22:08 | |
| *** baoli has joined #openstack-nova | 22:08 | |
| *** smurke has quit IRC | 22:10 | |
| *** aginwala has joined #openstack-nova | 22:11 | |
| *** baoli has quit IRC | 22:12 | |
| *** baoli has joined #openstack-nova | 22:16 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Fix use of safeutils.getcallargs https://review.openstack.org/255572 | 22:17 |
| *** sneti has quit IRC | 22:17 | |
| auggy | sdague: ok cool, i have this patch with the .tpl file changes and the associated template changes - https://review.openstack.org/#/c/254485/ | 22:19 |
| *** thorst has joined #openstack-nova | 22:19 | |
| auggy | so whenever your stuff is ready i can compare to what your stuff is doing and make any relevant changes | 22:20 |
| *** dave-mccowan has quit IRC | 22:21 | |
| *** thorst has quit IRC | 22:22 | |
| *** thorst has joined #openstack-nova | 22:22 | |
| *** baoli has quit IRC | 22:23 | |
| *** Marga_ has joined #openstack-nova | 22:24 | |
| *** baoli has joined #openstack-nova | 22:24 | |
| *** thorst has quit IRC | 22:24 | |
| *** thorst has joined #openstack-nova | 22:24 | |
| *** alejandrito has quit IRC | 22:26 | |
| *** lykinsbd has quit IRC | 22:27 | |
| *** bklei has quit IRC | 22:28 | |
| *** Marga_ has quit IRC | 22:28 | |
| *** bklei has joined #openstack-nova | 22:29 | |
| *** thorst has quit IRC | 22:29 | |
| *** bklei has quit IRC | 22:33 | |
| *** david-lyle has quit IRC | 22:34 | |
| *** baoli has quit IRC | 22:35 | |
| artom | When running Nova unit tests, what policy.json does it use? The system one in /etc/nova? | 22:36 |
| *** Sukhdev has joined #openstack-nova | 22:36 | |
| *** baoli has joined #openstack-nova | 22:36 | |
| artom | Follow up question, I've made changes to policy.json and need it reflected in tests. | 22:36 |
| *** betherly has quit IRC | 22:36 | |
| *** yarkot has quit IRC | 22:37 | |
| *** w_verdugo has quit IRC | 22:38 | |
| *** briancurtin has quit IRC | 22:38 | |
| *** david-ly_ has joined #openstack-nova | 22:39 | |
| *** w_verdugo has joined #openstack-nova | 22:39 | |
| *** patrickeast has quit IRC | 22:40 | |
| *** pratikmallya has joined #openstack-nova | 22:43 | |
| mriedem | artom: i think this is what you care about https://github.com/openstack/nova/blob/master/nova/test.py#L235 | 22:44 |
| *** andrewbogott has quit IRC | 22:44 | |
| mriedem | https://github.com/openstack/nova/blob/master/nova/tests/unit/policy_fixture.py | 22:44 |
| mriedem | it should use the 'real' default policy, but can be faked out if needed | 22:44 |
| mriedem | see https://github.com/openstack/nova/commit/c64cba9ce61d564bab69413b679d85051e1d3f36 | 22:45 |
| *** david-ly_ is now known as david-lyle | 22:45 | |
| artom | mriedem, thanks, but I don't think I made myself clear | 22:46 |
| artom | I've added a method to the compute api, and need to allow it in policy.json | 22:46 |
| *** pratikma_ has quit IRC | 22:46 | |
| artom | I've made the change in the git repo's policy.json file, but that change hasn't affected /etc/nova/policy.json | 22:47 |
| artom | I could just add the damn line manually, but this feels like something there's a system in place for | 22:47 |
| *** baoli has quit IRC | 22:47 | |
| *** pratikmallya has quit IRC | 22:47 | |
| *** baoli has joined #openstack-nova | 22:48 | |
| openstackgerrit | Tracy Jones proposed openstack/nova: Add better help to the guestfs option https://review.openstack.org/255578 | 22:48 |
| artom | python setup.py develop' or whatever' | 22:49 |
| openstackgerrit | Tracy Jones proposed openstack/nova: config options: centralize section "guestfs" https://review.openstack.org/248963 | 22:49 |
| mriedem | artom: do you have the change posted? i guess i'm not sure what you're trying to test and where. would be easier with code posted. i have to head out anyway | 22:49 |
| mriedem | i'd look for recent changes to policy.json and look at how those changes did their testing | 22:50 |
| *** mriedem is now known as mriedem_away | 22:51 | |
| *** harlowja has quit IRC | 22:51 | |
| *** harlowja has joined #openstack-nova | 22:51 | |
| *** ssurana1 has joined #openstack-nova | 22:52 | |
| *** rlrossit has quit IRC | 22:53 | |
| *** Marga_ has joined #openstack-nova | 22:55 | |
| *** ssurana has quit IRC | 22:55 | |
| *** mdrabe has quit IRC | 22:55 | |
| *** nic has quit IRC | 22:57 | |
| *** wuhg has quit IRC | 22:58 | |
| *** ducttape_ has quit IRC | 22:59 | |
| *** baoli has quit IRC | 22:59 | |
| *** boris-42_ has joined #openstack-nova | 22:59 | |
| *** Marga_ has quit IRC | 22:59 | |
| *** akshai_ has quit IRC | 22:59 | |
| *** baoli has joined #openstack-nova | 22:59 | |
| *** tonytan4ever has joined #openstack-nova | 23:00 | |
| *** dedery has quit IRC | 23:00 | |
| *** burgerk has quit IRC | 23:00 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** artom has quit IRC | 23:01 | |
| *** haomaiwang has joined #openstack-nova | 23:01 | |
| *** akshai has joined #openstack-nova | 23:01 | |
| *** pixelbeat has joined #openstack-nova | 23:02 | |
| *** gokrokve has joined #openstack-nova | 23:04 | |
| *** artom has joined #openstack-nova | 23:06 | |
| *** doug-fish has quit IRC | 23:06 | |
| suro-patz | melwitt: review request https://review.openstack.org/#/c/243825/ | 23:06 |
| *** dave-mccowan has joined #openstack-nova | 23:07 | |
| *** nic has joined #openstack-nova | 23:08 | |
| *** gokrokve has quit IRC | 23:08 | |
| *** dustins has quit IRC | 23:09 | |
| *** baoli has quit IRC | 23:09 | |
| *** baoli_ has joined #openstack-nova | 23:10 | |
| *** baoli_ has quit IRC | 23:10 | |
| *** baoli_ has joined #openstack-nova | 23:10 | |
| *** Marga_ has joined #openstack-nova | 23:11 | |
| *** Marga_ has quit IRC | 23:12 | |
| *** Marga_ has joined #openstack-nova | 23:13 | |
| *** Marga_ has quit IRC | 23:13 | |
| *** dustins has joined #openstack-nova | 23:14 | |
| *** Marga_ has joined #openstack-nova | 23:14 | |
| *** ctrath has quit IRC | 23:17 | |
| *** claudiub has quit IRC | 23:17 | |
| *** ctrath has joined #openstack-nova | 23:17 | |
| *** pumaranikar has quit IRC | 23:17 | |
| *** pushkaru has joined #openstack-nova | 23:18 | |
| *** baoli_ has quit IRC | 23:19 | |
| openstackgerrit | David Edery proposed openstack/nova: ports & networks gather should validate existance https://review.openstack.org/252565 | 23:21 |
| openstackgerrit | Eric Brown proposed openstack/nova: VMware: speed up start of n-cpu https://review.openstack.org/255586 | 23:22 |
| *** mrkz has quit IRC | 23:22 | |
| *** ctrath has quit IRC | 23:23 | |
| *** baoli has joined #openstack-nova | 23:25 | |
| *** xyang1 has quit IRC | 23:25 | |
| *** thorst has joined #openstack-nova | 23:27 | |
| *** briancurtin has joined #openstack-nova | 23:28 | |
| *** burt has quit IRC | 23:29 | |
| *** pushkaru has quit IRC | 23:30 | |
| *** haomaiwang has quit IRC | 23:32 | |
| *** haomaiwang has joined #openstack-nova | 23:32 | |
| openstackgerrit | Andrew Bogott proposed openstack/nova: Prevent scheduling of instances on full hosts. https://review.openstack.org/242251 | 23:34 |
| *** thorst has quit IRC | 23:34 | |
| *** jlk has joined #openstack-nova | 23:35 | |
| *** baoli has quit IRC | 23:35 | |
| jlk | Hey all, I'm looking to debug the routines nova uses to auth to keystone, and I'm hoping for an easy place to toss a pdb or epdb in that I can then use to walk through the auth. Can somebody help me with that a bit? | 23:35 |
| *** baoli has joined #openstack-nova | 23:36 | |
| *** ianbrown has joined #openstack-nova | 23:36 | |
| *** ianbrown has quit IRC | 23:36 | |
| *** haomaiwang has quit IRC | 23:37 | |
| *** rcernin has joined #openstack-nova | 23:37 | |
| *** tjones1 has joined #openstack-nova | 23:39 | |
| *** pradk has quit IRC | 23:40 | |
| *** cn28h has quit IRC | 23:40 | |
| *** cn28h has joined #openstack-nova | 23:41 | |
| *** takedakn has joined #openstack-nova | 23:41 | |
| *** andrewbogott has joined #openstack-nova | 23:41 | |
| *** claudiub has joined #openstack-nova | 23:42 | |
| artom | mriedem_away, thanks - I wanted to pass tests before posting, but if I really can't figure it out I'll just post and ask for reviewer help. | 23:44 |
| *** takashin has joined #openstack-nova | 23:45 | |
| *** betherly has joined #openstack-nova | 23:48 | |
| *** baoli has quit IRC | 23:50 | |
| *** ljxiash has joined #openstack-nova | 23:50 | |
| *** patrickeast has joined #openstack-nova | 23:51 | |
| jlk | n/m. | 23:52 |
| *** jlk has left #openstack-nova | 23:52 | |
| *** bklei has joined #openstack-nova | 23:53 | |
| *** cdent has quit IRC | 23:54 | |
| *** ljxiash has quit IRC | 23:55 | |
| *** ianbrown has joined #openstack-nova | 23:55 | |
| *** Sukhdev has quit IRC | 23:56 | |
| *** ZZelle_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!