| *** rama_y has quit IRC | 00:01 | |
| *** mlavalle has quit IRC | 00:02 | |
| mriedem | can't believe how much an email confirming that nova is going to be at the PTG can spiral out of control | 00:03 |
|---|---|---|
| mriedem | i guess that's my fault | 00:03 |
| openstackgerrit | melanie witt proposed openstack/nova: Always use python2.7 for functional tests https://review.openstack.org/385207 | 00:04 |
| openstackgerrit | STEW TY proposed openstack/nova: Transform instance.unshelve notifications https://review.openstack.org/382565 | 00:04 |
| *** McNinja has quit IRC | 00:04 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove _set_up_controller() from attach tests https://review.openstack.org/385211 | 00:06 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove redundant req setting https://review.openstack.org/385212 | 00:06 |
| melwitt | tox -efunctional fails with python 3.x, I put up a patch ^ | 00:06 |
| *** esberglu has quit IRC | 00:06 | |
| *** esberglu has joined #openstack-nova | 00:07 | |
| *** hamzy has joined #openstack-nova | 00:08 | |
| *** gjayavelu has quit IRC | 00:09 | |
| *** esberglu has quit IRC | 00:12 | |
| diana_clarke | melwitt: Oh! Is that the solution for the "unichr() arg not in range(0x00000) (narrow Python build)" error I keep getting? | 00:13 |
| diana_clarke | melwitt: I suppose I could check myself ;) I'll take that patch for a spin. | 00:14 |
| *** andrewbogott has quit IRC | 00:14 | |
| melwitt | diana_clarke: the error I was getting was "TypeError: a bytes-like object is required, not 'str'" so it may not be related to what you're seeing | 00:14 |
| *** andrewbogott has joined #openstack-nova | 00:15 | |
| *** gouthamr has joined #openstack-nova | 00:16 | |
| *** McNinja has joined #openstack-nova | 00:17 | |
| *** stewie925 has quit IRC | 00:19 | |
| diana_clarke | melwitt: Looks like a different issue from mine, sadly. The error I'm seeing is related to gabbi. I haven't looked all that hard at it (instead I just comment out the placement tests). | 00:19 |
| *** sdake_ is now known as sdake | 00:22 | |
| *** gouthamr has quit IRC | 00:23 | |
| *** tdurakov has quit IRC | 00:23 | |
| *** ediardo has quit IRC | 00:23 | |
| *** ediardo has joined #openstack-nova | 00:23 | |
| *** tdurakov has joined #openstack-nova | 00:23 | |
| *** gouthamr has joined #openstack-nova | 00:23 | |
| *** toan has quit IRC | 00:24 | |
| *** JayF has quit IRC | 00:24 | |
| *** fungi has quit IRC | 00:26 | |
| *** efried has quit IRC | 00:26 | |
| openstackgerrit | Shoham Peller proposed openstack/nova-specs: Cleanup running shelved-offloaded instances https://review.openstack.org/373371 | 00:27 |
| *** markvoelker has quit IRC | 00:28 | |
| *** ducttape_ has quit IRC | 00:28 | |
| *** esberglu has joined #openstack-nova | 00:29 | |
| *** gyee has quit IRC | 00:29 | |
| *** toan has joined #openstack-nova | 00:29 | |
| *** fungi has joined #openstack-nova | 00:30 | |
| *** esberglu has quit IRC | 00:31 | |
| *** esberglu has joined #openstack-nova | 00:31 | |
| *** efried has joined #openstack-nova | 00:32 | |
| *** rmcall has quit IRC | 00:34 | |
| *** rmcall has joined #openstack-nova | 00:35 | |
| *** esberglu has quit IRC | 00:36 | |
| *** ssurana has quit IRC | 00:36 | |
| *** markvoelker has joined #openstack-nova | 00:36 | |
| *** JayF has joined #openstack-nova | 00:38 | |
| *** rmcall has quit IRC | 00:38 | |
| mriedem | melwitt: is it all functional tests or only some? | 00:39 |
| *** rmcall has joined #openstack-nova | 00:39 | |
| mriedem | looks like all b/c it can't import one? | 00:39 |
| *** rmcall has quit IRC | 00:39 | |
| *** mtanino has quit IRC | 00:40 | |
| melwitt | mriedem: it's several api sample tests but it looks to be all emanating from the ServersSampleBase | 00:40 |
| *** mgagne has quit IRC | 00:40 | |
| melwitt | so if the line "user_data = base64.b64encode(user_data_contents)" could be fixed then that might solve it | 00:41 |
| *** ducttape_ has joined #openstack-nova | 00:42 | |
| melwitt | I'm not sure how it's always using python 2.7 in the gate. in my xenial VM I'm getting python 3.x as the default | 00:42 |
| openstackgerrit | Li Yingjun proposed openstack/nova-specs: Flavor notifications spec https://review.openstack.org/321336 | 00:43 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: doc: add dev policy about no new metrics monitors https://review.openstack.org/385219 | 00:43 |
| *** yingjun has joined #openstack-nova | 00:44 | |
| *** nic has quit IRC | 00:45 | |
| *** mgagne has joined #openstack-nova | 00:45 | |
| *** hoangcx has joined #openstack-nova | 00:45 | |
| melwitt | we could fix the python 3.x problem but it'll probably be whack-a-mole if anything else breaks it in the future until we run them in the gate with python 3.x too | 00:45 |
| *** sdake has quit IRC | 00:46 | |
| *** sdague has quit IRC | 00:46 | |
| *** haplo37 has quit IRC | 00:47 | |
| *** baoli has quit IRC | 00:49 | |
| *** amotoki has joined #openstack-nova | 00:49 | |
| *** baoli has joined #openstack-nova | 00:50 | |
| *** mvk has quit IRC | 00:50 | |
| *** amotoki has quit IRC | 00:54 | |
| *** mgoddard has joined #openstack-nova | 00:54 | |
| *** haplo37 has joined #openstack-nova | 00:54 | |
| *** jamielennox is now known as jamielennox|away | 00:56 | |
| *** gjayavelu has joined #openstack-nova | 00:56 | |
| mriedem | melwitt: yeah until we gate functional tests on py3 we should just restrict to py27 | 00:56 |
| mriedem | the gate vms are using custom built images i think | 00:57 |
| mriedem | that infra creates | 00:57 |
| mriedem | don't quote me on that though | 00:57 |
| *** baoli has quit IRC | 00:59 | |
| *** tuanluong has joined #openstack-nova | 01:00 | |
| melwitt | sounds good enough to me. I briefly tried to figure out how it defaults to python 2.7 from the console log without success | 01:01 |
| *** sdake has joined #openstack-nova | 01:01 | |
| *** ducnc has joined #openstack-nova | 01:01 | |
| *** esberglu has joined #openstack-nova | 01:02 | |
| *** baoli has joined #openstack-nova | 01:02 | |
| *** jamielennox|away is now known as jamielennox | 01:02 | |
| *** phuongnh has joined #openstack-nova | 01:03 | |
| *** McNinja has quit IRC | 01:03 | |
| yingjun | mriedem, https://review.openstack.org/#/c/321336/ updated, please have a look at that again if you have time, thanks! | 01:04 |
| *** thorst has joined #openstack-nova | 01:04 | |
| *** McNinja has joined #openstack-nova | 01:05 | |
| mriedem | yingjun: thanks, +2 | 01:05 |
| *** xinliang has quit IRC | 01:08 | |
| *** rmcall has joined #openstack-nova | 01:15 | |
| *** mgoddard has quit IRC | 01:17 | |
| *** baoli has quit IRC | 01:20 | |
| *** baoli has joined #openstack-nova | 01:21 | |
| mriedem | dansmith: https://review.openstack.org/#/c/385225/ there you go | 01:21 |
| *** thorst has quit IRC | 01:22 | |
| *** mriedem has quit IRC | 01:22 | |
| *** gjayavelu has quit IRC | 01:22 | |
| *** artom has quit IRC | 01:24 | |
| *** esberglu has quit IRC | 01:24 | |
| *** hieulq_ has quit IRC | 01:25 | |
| *** phuongnh has quit IRC | 01:25 | |
| *** xinliang has joined #openstack-nova | 01:25 | |
| *** tuanluong has quit IRC | 01:25 | |
| *** hoangcx has quit IRC | 01:25 | |
| *** phuongnh has joined #openstack-nova | 01:26 | |
| *** tuanluong has joined #openstack-nova | 01:29 | |
| *** esberglu has joined #openstack-nova | 01:30 | |
| *** artom has joined #openstack-nova | 01:31 | |
| *** hieulq_ has joined #openstack-nova | 01:31 | |
| *** esberglu has quit IRC | 01:31 | |
| *** esberglu has joined #openstack-nova | 01:32 | |
| *** esberglu has quit IRC | 01:33 | |
| *** Kevin_Zheng has joined #openstack-nova | 01:33 | |
| *** trisq has joined #openstack-nova | 01:34 | |
| *** esberglu has joined #openstack-nova | 01:36 | |
| *** yuntongjin has joined #openstack-nova | 01:37 | |
| *** ljxiash has joined #openstack-nova | 01:38 | |
| *** ljxiash has quit IRC | 01:38 | |
| *** esberglu has quit IRC | 01:39 | |
| *** ducttape_ has quit IRC | 01:40 | |
| *** tangchen has joined #openstack-nova | 01:40 | |
| *** zhangyufei has joined #openstack-nova | 01:42 | |
| *** ducttape_ has joined #openstack-nova | 01:44 | |
| *** ducttape_ has quit IRC | 01:44 | |
| *** ducttape_ has joined #openstack-nova | 01:44 | |
| *** markmcclain has joined #openstack-nova | 01:47 | |
| *** markmcclain has quit IRC | 01:51 | |
| *** zhangyufei has quit IRC | 01:53 | |
| *** liusheng has joined #openstack-nova | 01:54 | |
| *** thorst has joined #openstack-nova | 01:58 | |
| *** McNinja has quit IRC | 01:59 | |
| *** Guest50630 has quit IRC | 02:03 | |
| *** gjayavelu has joined #openstack-nova | 02:03 | |
| *** thorst has quit IRC | 02:03 | |
| *** baoli has quit IRC | 02:05 | |
| *** zigo has joined #openstack-nova | 02:05 | |
| *** zigo is now known as Guest59427 | 02:06 | |
| *** McNinja has joined #openstack-nova | 02:07 | |
| *** zhangyufei has joined #openstack-nova | 02:08 | |
| *** yulong|away is now known as liuyulong | 02:10 | |
| *** unicell has quit IRC | 02:12 | |
| *** mtanino has joined #openstack-nova | 02:16 | |
| *** ducttape_ has quit IRC | 02:19 | |
| *** yamahata has quit IRC | 02:20 | |
| *** browne has quit IRC | 02:22 | |
| *** gcb has joined #openstack-nova | 02:22 | |
| *** ducttape_ has joined #openstack-nova | 02:23 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Add support for tag instances when boot https://review.openstack.org/366469 | 02:23 |
| *** gjayavelu has quit IRC | 02:25 | |
| *** huanxie has quit IRC | 02:26 | |
| *** esberglu has joined #openstack-nova | 02:27 | |
| *** huanxie has joined #openstack-nova | 02:27 | |
| *** McNinja has quit IRC | 02:31 | |
| *** esberglu has quit IRC | 02:32 | |
| openstackgerrit | Hieu LE proposed openstack/nova: Config options: improve libvirt help text (2) https://review.openstack.org/316550 | 02:32 |
| *** McNinja has joined #openstack-nova | 02:33 | |
| *** ducttape_ has quit IRC | 02:39 | |
| *** McNinja has quit IRC | 02:40 | |
| *** McNinja has joined #openstack-nova | 02:42 | |
| *** ducttape_ has joined #openstack-nova | 02:44 | |
| *** ducttape_ has quit IRC | 02:45 | |
| *** ducttape_ has joined #openstack-nova | 02:49 | |
| *** amotoki has joined #openstack-nova | 02:51 | |
| *** amotoki has quit IRC | 02:51 | |
| *** amotoki has joined #openstack-nova | 02:51 | |
| *** kaisers_ has joined #openstack-nova | 02:51 | |
| *** tjones- has joined #openstack-nova | 02:52 | |
| *** tjones- has left #openstack-nova | 02:52 | |
| *** gongysh_ has joined #openstack-nova | 02:53 | |
| *** rama_y has joined #openstack-nova | 02:53 | |
| *** ducttape_ has quit IRC | 02:55 | |
| *** kaisers_ has quit IRC | 02:56 | |
| *** tjones- has joined #openstack-nova | 02:56 | |
| *** tjones- has left #openstack-nova | 02:57 | |
| *** sdake has quit IRC | 02:57 | |
| *** ssurana has joined #openstack-nova | 02:58 | |
| *** amotoki has quit IRC | 02:59 | |
| *** ducttape_ has joined #openstack-nova | 02:59 | |
| *** baoli has joined #openstack-nova | 02:59 | |
| *** baoli has quit IRC | 03:05 | |
| *** jdillaman1 has quit IRC | 03:09 | |
| *** ducttape_ has quit IRC | 03:09 | |
| *** gouthamr has quit IRC | 03:10 | |
| *** ducttape_ has joined #openstack-nova | 03:11 | |
| *** odyssey4me has quit IRC | 03:14 | |
| *** odyssey4me has joined #openstack-nova | 03:15 | |
| *** sdake has joined #openstack-nova | 03:18 | |
| *** ducttape_ has quit IRC | 03:19 | |
| *** jamielennox is now known as jamielennox|away | 03:20 | |
| *** jdillaman1 has joined #openstack-nova | 03:21 | |
| *** hoangcx has joined #openstack-nova | 03:21 | |
| *** esberglu has joined #openstack-nova | 03:21 | |
| *** gjayavelu has joined #openstack-nova | 03:23 | |
| *** amotoki has joined #openstack-nova | 03:24 | |
| *** yuntongjin has quit IRC | 03:25 | |
| *** ducttape_ has joined #openstack-nova | 03:25 | |
| *** TravT has joined #openstack-nova | 03:27 | |
| *** eliqiao has quit IRC | 03:27 | |
| *** TravT has quit IRC | 03:27 | |
| *** TravT has joined #openstack-nova | 03:28 | |
| *** unicell has joined #openstack-nova | 03:29 | |
| *** unicell has quit IRC | 03:29 | |
| *** unicell has joined #openstack-nova | 03:30 | |
| *** unicell has quit IRC | 03:30 | |
| *** esberglu has quit IRC | 03:32 | |
| *** TravT_ has joined #openstack-nova | 03:32 | |
| *** TravT has quit IRC | 03:33 | |
| *** armax has quit IRC | 03:33 | |
| *** TravT_ has quit IRC | 03:33 | |
| *** armax has joined #openstack-nova | 03:33 | |
| *** TravT has joined #openstack-nova | 03:34 | |
| *** armax has quit IRC | 03:34 | |
| *** TravT has quit IRC | 03:34 | |
| *** TravT has joined #openstack-nova | 03:35 | |
| *** jamielennox|away is now known as jamielennox | 03:38 | |
| *** baoli has joined #openstack-nova | 03:38 | |
| openstackgerrit | pawnesh kumar proposed openstack/nova: Fix some PEP8 issues and OpenStack Licensing https://review.openstack.org/383924 | 03:38 |
| *** esberglu has joined #openstack-nova | 03:39 | |
| *** TravT has quit IRC | 03:39 | |
| *** TravT has joined #openstack-nova | 03:40 | |
| *** ayoung has quit IRC | 03:40 | |
| *** amotoki has quit IRC | 03:40 | |
| *** esberglu has quit IRC | 03:40 | |
| *** TravT has quit IRC | 03:44 | |
| *** gongysh has quit IRC | 03:44 | |
| *** anush_ has joined #openstack-nova | 03:45 | |
| *** tjones- has joined #openstack-nova | 03:47 | |
| *** TravT has joined #openstack-nova | 03:49 | |
| *** ducttape_ has quit IRC | 03:50 | |
| *** TravT has quit IRC | 03:50 | |
| *** TravT has joined #openstack-nova | 03:50 | |
| *** haplo37 has quit IRC | 03:51 | |
| *** amotoki has joined #openstack-nova | 03:53 | |
| *** baoli has quit IRC | 03:53 | |
| *** baoli has joined #openstack-nova | 03:54 | |
| *** TravT_ has joined #openstack-nova | 03:54 | |
| *** baoli has quit IRC | 03:54 | |
| *** takashin has joined #openstack-nova | 03:54 | |
| *** TravT_ has quit IRC | 03:55 | |
| *** TravT has quit IRC | 03:55 | |
| *** anush_ has quit IRC | 03:55 | |
| *** TravT has joined #openstack-nova | 03:55 | |
| *** baoli has joined #openstack-nova | 03:55 | |
| *** gongysh has joined #openstack-nova | 03:55 | |
| *** haplo37 has joined #openstack-nova | 03:57 | |
| *** psachin has joined #openstack-nova | 03:58 | |
| *** ducttape_ has joined #openstack-nova | 03:58 | |
| *** TravT has quit IRC | 03:58 | |
| *** amotoki has quit IRC | 03:59 | |
| *** TravT has joined #openstack-nova | 03:59 | |
| *** takashin has quit IRC | 03:59 | |
| *** TravT has quit IRC | 04:03 | |
| *** TravT has joined #openstack-nova | 04:04 | |
| *** dikonoor has joined #openstack-nova | 04:04 | |
| *** Jeffrey4l_ has quit IRC | 04:05 | |
| *** ducttape_ has joined #openstack-nova | 04:08 | |
| *** TravT has quit IRC | 04:08 | |
| *** TravT has joined #openstack-nova | 04:08 | |
| *** links has joined #openstack-nova | 04:10 | |
| openstackgerrit | pawnesh kumar proposed openstack/nova: Fix some PEP8 issues and OpenStack Licensing https://review.openstack.org/383924 | 04:12 |
| *** TravT_ has joined #openstack-nova | 04:13 | |
| *** ducttape_ has quit IRC | 04:13 | |
| *** TravT has quit IRC | 04:13 | |
| *** jistr has quit IRC | 04:15 | |
| *** jistr has joined #openstack-nova | 04:16 | |
| *** kaisers_ has joined #openstack-nova | 04:17 | |
| *** TravT has joined #openstack-nova | 04:18 | |
| *** TravT_ has quit IRC | 04:18 | |
| *** alex_xu has quit IRC | 04:18 | |
| *** zhangyufei has quit IRC | 04:20 | |
| *** alex_xu has joined #openstack-nova | 04:20 | |
| *** baoli has quit IRC | 04:20 | |
| *** rama_y has quit IRC | 04:21 | |
| *** GB21 has joined #openstack-nova | 04:21 | |
| *** rama_y has joined #openstack-nova | 04:22 | |
| *** TravT has quit IRC | 04:22 | |
| *** TravT has joined #openstack-nova | 04:23 | |
| *** wasmum has quit IRC | 04:23 | |
| *** mdrabe has quit IRC | 04:24 | |
| *** mdrabe has joined #openstack-nova | 04:25 | |
| *** Jeffrey4l has joined #openstack-nova | 04:26 | |
| *** yuntongjin has joined #openstack-nova | 04:27 | |
| *** TravT has quit IRC | 04:27 | |
| *** TravT has joined #openstack-nova | 04:27 | |
| *** dikonoor has quit IRC | 04:30 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Ironic Driver: override get_serial_console() https://review.openstack.org/328157 | 04:30 |
| *** vishwanathj has quit IRC | 04:32 | |
| *** TravT has quit IRC | 04:32 | |
| *** ducttape_ has joined #openstack-nova | 04:32 | |
| *** TravT has joined #openstack-nova | 04:32 | |
| *** mdrabe has quit IRC | 04:33 | |
| *** raunak has joined #openstack-nova | 04:33 | |
| *** wasmum has joined #openstack-nova | 04:35 | |
| *** ducttape_ has quit IRC | 04:36 | |
| *** TravT has quit IRC | 04:37 | |
| *** mtanino has quit IRC | 04:39 | |
| *** agireud has quit IRC | 04:42 | |
| *** TravT has joined #openstack-nova | 04:42 | |
| *** rama_y has quit IRC | 04:45 | |
| *** claudiub has joined #openstack-nova | 04:47 | |
| *** TravT_ has joined #openstack-nova | 04:47 | |
| *** TravT has quit IRC | 04:47 | |
| *** unicell has joined #openstack-nova | 04:51 | |
| *** agireud has joined #openstack-nova | 04:51 | |
| *** TravT_ has quit IRC | 04:51 | |
| *** TravT has joined #openstack-nova | 04:51 | |
| *** karthiks has joined #openstack-nova | 04:54 | |
| *** agireud has quit IRC | 04:56 | |
| *** TravT_ has joined #openstack-nova | 04:56 | |
| *** TravT has quit IRC | 04:56 | |
| *** prateek has joined #openstack-nova | 04:56 | |
| *** dikonoor has joined #openstack-nova | 04:56 | |
| *** TravT_ has quit IRC | 05:01 | |
| *** prateek has quit IRC | 05:01 | |
| *** maticue has quit IRC | 05:01 | |
| *** salv-orlando has joined #openstack-nova | 05:01 | |
| *** TravT has joined #openstack-nova | 05:01 | |
| *** jaosorior has joined #openstack-nova | 05:02 | |
| *** pbandark has joined #openstack-nova | 05:03 | |
| *** ratailor has joined #openstack-nova | 05:03 | |
| *** rama_y has joined #openstack-nova | 05:05 | |
| *** TravT has quit IRC | 05:05 | |
| *** TravT has joined #openstack-nova | 05:06 | |
| *** prateek has joined #openstack-nova | 05:07 | |
| *** wasmum has quit IRC | 05:09 | |
| *** Jeffrey4l has quit IRC | 05:09 | |
| *** alex_xu has quit IRC | 05:09 | |
| *** TravT_ has joined #openstack-nova | 05:10 | |
| *** TravT has quit IRC | 05:11 | |
| *** alex_xu has joined #openstack-nova | 05:11 | |
| *** salv-orlando has quit IRC | 05:11 | |
| *** Jeffrey4l has joined #openstack-nova | 05:11 | |
| *** Jeffrey4l has quit IRC | 05:12 | |
| *** Jeffrey4l has joined #openstack-nova | 05:13 | |
| *** Jeffrey4l has quit IRC | 05:13 | |
| *** Jeffrey4l_ has joined #openstack-nova | 05:13 | |
| *** lpetrut has joined #openstack-nova | 05:14 | |
| *** TravT_ has quit IRC | 05:15 | |
| *** TravT has joined #openstack-nova | 05:15 | |
| *** rama_y has quit IRC | 05:18 | |
| *** tanee_away is now known as tanee | 05:19 | |
| *** amotoki has joined #openstack-nova | 05:19 | |
| *** wasmum has joined #openstack-nova | 05:19 | |
| *** dikonoor has quit IRC | 05:19 | |
| *** TravT has quit IRC | 05:20 | |
| *** TravT_ has joined #openstack-nova | 05:20 | |
| *** TravT_ has quit IRC | 05:24 | |
| *** salv-orlando has joined #openstack-nova | 05:25 | |
| *** TravT has joined #openstack-nova | 05:25 | |
| *** TravT_ has joined #openstack-nova | 05:29 | |
| *** TravT has quit IRC | 05:30 | |
| *** GB21 has quit IRC | 05:30 | |
| *** ducttape_ has joined #openstack-nova | 05:33 | |
| *** TravT_ has quit IRC | 05:34 | |
| *** TravT has joined #openstack-nova | 05:34 | |
| *** gabor_antal_ has joined #openstack-nova | 05:35 | |
| *** pjm6_ has joined #openstack-nova | 05:35 | |
| *** kylek3h_ has joined #openstack-nova | 05:36 | |
| *** ekhugen has quit IRC | 05:37 | |
| *** gabor_antal has quit IRC | 05:37 | |
| *** ekhugen has joined #openstack-nova | 05:37 | |
| *** ducttape_ has quit IRC | 05:38 | |
| *** kylek3h has quit IRC | 05:38 | |
| *** raunak has quit IRC | 05:39 | |
| *** TravT has quit IRC | 05:39 | |
| *** pjm6 has quit IRC | 05:39 | |
| *** TravT has joined #openstack-nova | 05:39 | |
| *** Jeffrey4l_ has quit IRC | 05:41 | |
| *** nkrinner_afk is now known as nkrinner | 05:42 | |
| *** TravT has quit IRC | 05:44 | |
| *** GB21 has joined #openstack-nova | 05:44 | |
| *** TravT has joined #openstack-nova | 05:44 | |
| *** dikonoor has joined #openstack-nova | 05:47 | |
| *** amotoki has quit IRC | 05:48 | |
| *** TravT has quit IRC | 05:48 | |
| *** TravT has joined #openstack-nova | 05:48 | |
| *** yuntongjin has quit IRC | 05:49 | |
| *** lpetrut has quit IRC | 05:51 | |
| *** Jeffrey4l_ has joined #openstack-nova | 05:52 | |
| *** TravT_ has joined #openstack-nova | 05:54 | |
| *** links has quit IRC | 05:54 | |
| *** TravT has quit IRC | 05:54 | |
| *** cybojanek has quit IRC | 05:55 | |
| *** Jeffrey4l_ has quit IRC | 05:57 | |
| *** TravT_ has quit IRC | 05:58 | |
| *** Jeffrey4l_ has joined #openstack-nova | 05:58 | |
| *** TravT has joined #openstack-nova | 05:58 | |
| *** unicell1 has joined #openstack-nova | 06:02 | |
| *** unicell has quit IRC | 06:02 | |
| *** TravT has quit IRC | 06:03 | |
| *** TravT has joined #openstack-nova | 06:03 | |
| *** Jeffrey4l_ has quit IRC | 06:06 | |
| *** TravT has quit IRC | 06:07 | |
| *** sahid has joined #openstack-nova | 06:07 | |
| *** TravT has joined #openstack-nova | 06:08 | |
| *** campee_ has joined #openstack-nova | 06:09 | |
| *** links has joined #openstack-nova | 06:10 | |
| *** mjura has joined #openstack-nova | 06:11 | |
| *** campee has quit IRC | 06:11 | |
| *** TravT has quit IRC | 06:12 | |
| *** zhangyufei has joined #openstack-nova | 06:16 | |
| *** gjayavelu1 has joined #openstack-nova | 06:16 | |
| *** ssurana has quit IRC | 06:17 | |
| *** gjayavelu has quit IRC | 06:17 | |
| *** Jeffrey4l_ has joined #openstack-nova | 06:19 | |
| *** Jeffrey4l_ has quit IRC | 06:23 | |
| *** Jeffrey4l_ has joined #openstack-nova | 06:23 | |
| *** andreas_s has joined #openstack-nova | 06:29 | |
| *** zenoway has joined #openstack-nova | 06:33 | |
| *** rkrum has quit IRC | 06:33 | |
| *** ducttape_ has joined #openstack-nova | 06:34 | |
| *** haplo37 has quit IRC | 06:34 | |
| *** s1061123_ has quit IRC | 06:34 | |
| *** amotoki has joined #openstack-nova | 06:34 | |
| *** rcernin has joined #openstack-nova | 06:36 | |
| *** ducttape_ has quit IRC | 06:38 | |
| *** cristicalin has joined #openstack-nova | 06:38 | |
| *** s1061123 has joined #openstack-nova | 06:41 | |
| *** hshiina has joined #openstack-nova | 06:41 | |
| *** haplo37 has joined #openstack-nova | 06:44 | |
| *** yuntongjin has joined #openstack-nova | 06:53 | |
| *** unicell1 has quit IRC | 06:55 | |
| *** unicell has joined #openstack-nova | 06:55 | |
| *** zenoway has quit IRC | 06:55 | |
| *** zenoway has joined #openstack-nova | 07:01 | |
| *** pilgrimstack has joined #openstack-nova | 07:03 | |
| *** nherciu has joined #openstack-nova | 07:04 | |
| *** tesseract has joined #openstack-nova | 07:06 | |
| *** zenoway has quit IRC | 07:06 | |
| *** tesseract is now known as Guest24890 | 07:06 | |
| *** wasmum has quit IRC | 07:10 | |
| *** zenoway has joined #openstack-nova | 07:10 | |
| *** cristicalin has quit IRC | 07:11 | |
| *** matrohon has joined #openstack-nova | 07:12 | |
| *** coolsvap has joined #openstack-nova | 07:13 | |
| *** wasmum has joined #openstack-nova | 07:15 | |
| *** markus_z has joined #openstack-nova | 07:16 | |
| *** kaisers_ has quit IRC | 07:21 | |
| *** sdake has quit IRC | 07:22 | |
| *** kaisers_ has joined #openstack-nova | 07:22 | |
| *** gjayavelu1 has quit IRC | 07:23 | |
| *** kaisers_ has quit IRC | 07:26 | |
| *** gongysh has quit IRC | 07:26 | |
| *** dimtruck is now known as zz_dimtruck | 07:32 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Add notifications for server groups https://review.openstack.org/375316 | 07:32 |
| *** Guest59427 is now known as zigo | 07:33 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Add support for tag instances when boot https://review.openstack.org/366469 | 07:33 |
| *** ducttape_ has joined #openstack-nova | 07:35 | |
| *** acabot has joined #openstack-nova | 07:37 | |
| *** lpetrut has joined #openstack-nova | 07:38 | |
| *** ducttape_ has quit IRC | 07:39 | |
| openstackgerrit | melanie witt proposed openstack/nova: Add quota related tables to the api database. https://review.openstack.org/331810 | 07:40 |
| *** mkoderer has joined #openstack-nova | 07:41 | |
| *** rmart04 has joined #openstack-nova | 07:43 | |
| *** rcernin has quit IRC | 07:45 | |
| *** rcernin has joined #openstack-nova | 07:49 | |
| *** gongysh has joined #openstack-nova | 07:49 | |
| *** alexchadin has joined #openstack-nova | 07:55 | |
| *** salv-orlando has quit IRC | 07:59 | |
| *** ssbarnea has joined #openstack-nova | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-nova | 08:00 | |
| *** mnasiadka has joined #openstack-nova | 08:05 | |
| *** gibi has joined #openstack-nova | 08:08 | |
| *** TravT has joined #openstack-nova | 08:08 | |
| *** ihrachys has joined #openstack-nova | 08:12 | |
| *** TravT has quit IRC | 08:13 | |
| *** TravT has joined #openstack-nova | 08:13 | |
| *** alexchad_ has joined #openstack-nova | 08:15 | |
| *** ihrachys has quit IRC | 08:17 | |
| *** TravT has quit IRC | 08:18 | |
| *** alexchadin has quit IRC | 08:18 | |
| *** TravT has joined #openstack-nova | 08:18 | |
| *** yassine has joined #openstack-nova | 08:21 | |
| *** yassine is now known as Guest81681 | 08:22 | |
| *** TravT has quit IRC | 08:22 | |
| *** TravT has joined #openstack-nova | 08:23 | |
| *** pradk has quit IRC | 08:27 | |
| *** yuntongjin has quit IRC | 08:27 | |
| *** TravT has quit IRC | 08:28 | |
| *** TravT_ has joined #openstack-nova | 08:28 | |
| *** TravT_ has quit IRC | 08:32 | |
| *** TravT has joined #openstack-nova | 08:32 | |
| *** amotoki has quit IRC | 08:33 | |
| *** ihrachys has joined #openstack-nova | 08:34 | |
| *** ducttape_ has joined #openstack-nova | 08:35 | |
| *** TravT has quit IRC | 08:37 | |
| *** TravT has joined #openstack-nova | 08:37 | |
| *** dikonoo has joined #openstack-nova | 08:38 | |
| *** ihrachys has quit IRC | 08:39 | |
| *** ducttape_ has quit IRC | 08:40 | |
| *** TravT has quit IRC | 08:42 | |
| *** TravT_ has joined #openstack-nova | 08:42 | |
| *** dikonoor has quit IRC | 08:42 | |
| *** BobBall has quit IRC | 08:43 | |
| *** TravT_ has quit IRC | 08:46 | |
| *** TravT has joined #openstack-nova | 08:47 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add Hierarchy info https://review.openstack.org/383726 | 08:47 |
| *** derekh has joined #openstack-nova | 08:47 | |
| *** mkoderer has quit IRC | 08:49 | |
| *** mkoderer has joined #openstack-nova | 08:50 | |
| *** TravT has quit IRC | 08:51 | |
| *** yuntongjin has joined #openstack-nova | 08:51 | |
| *** TravT has joined #openstack-nova | 08:51 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add child_hard_limits column https://review.openstack.org/383727 | 08:54 |
| *** wasmum has quit IRC | 08:54 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Nested Quota: Set default values to subproject https://review.openstack.org/383728 | 08:54 |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add nested quota support https://review.openstack.org/383729 | 08:54 |
| *** TravT has quit IRC | 08:56 | |
| *** TravT has joined #openstack-nova | 08:56 | |
| *** wasmum has joined #openstack-nova | 08:58 | |
| *** TravT has quit IRC | 09:01 | |
| *** TravT_ has joined #openstack-nova | 09:01 | |
| *** alexchad_ has quit IRC | 09:01 | |
| *** alexchadin has joined #openstack-nova | 09:02 | |
| *** hshiina has quit IRC | 09:03 | |
| *** alexchad_ has joined #openstack-nova | 09:05 | |
| *** alexchad_ is now known as alexchadin_ | 09:05 | |
| *** hieulq_ has quit IRC | 09:05 | |
| *** hieulq_ has joined #openstack-nova | 09:06 | |
| *** TravT_ has quit IRC | 09:06 | |
| *** TravT has joined #openstack-nova | 09:06 | |
| *** alexchadin has quit IRC | 09:06 | |
| *** TravT has quit IRC | 09:10 | |
| *** kaisers_ has joined #openstack-nova | 09:10 | |
| *** TravT_ has joined #openstack-nova | 09:10 | |
| *** kaisers_ has quit IRC | 09:15 | |
| *** TravT_ has quit IRC | 09:15 | |
| *** TravT has joined #openstack-nova | 09:15 | |
| *** ihrachys has joined #openstack-nova | 09:16 | |
| *** mvk has joined #openstack-nova | 09:17 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add _rebuild_instance helper method https://review.openstack.org/376004 | 09:19 |
| *** TravT has quit IRC | 09:20 | |
| *** TravT_ has joined #openstack-nova | 09:20 | |
| *** matrohon has quit IRC | 09:23 | |
| *** TravT_ has quit IRC | 09:25 | |
| *** TravT has joined #openstack-nova | 09:25 | |
| *** code-R has joined #openstack-nova | 09:29 | |
| *** TravT has quit IRC | 09:29 | |
| *** salv-orlando has joined #openstack-nova | 09:30 | |
| *** TravT has joined #openstack-nova | 09:30 | |
| openstackgerrit | sahid proposed openstack/nova: object: add new field cpu_emulator_threads_policy https://review.openstack.org/385022 | 09:32 |
| openstackgerrit | sahid proposed openstack/nova: virt: add numa constraint for emulator threads policy https://review.openstack.org/385023 | 09:32 |
| openstackgerrit | sahid proposed openstack/nova: virt: update overhead to take into account vCPUs https://review.openstack.org/385024 | 09:32 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: return overhead if isolate emulator threads requested https://review.openstack.org/385364 | 09:32 |
| *** TravT has quit IRC | 09:34 | |
| *** salv-orlando has quit IRC | 09:34 | |
| *** TravT has joined #openstack-nova | 09:34 | |
| bauzas | p!spy | 09:35 |
| *** ducttape_ has joined #openstack-nova | 09:36 | |
| *** haplo37 has quit IRC | 09:37 | |
| *** TravT has quit IRC | 09:39 | |
| *** TravT has joined #openstack-nova | 09:39 | |
| *** jaosorior has quit IRC | 09:39 | |
| *** jaosorior has joined #openstack-nova | 09:40 | |
| *** McNinja has quit IRC | 09:40 | |
| *** ducttape_ has quit IRC | 09:41 | |
| *** yingjun has quit IRC | 09:42 | |
| *** McNinja has joined #openstack-nova | 09:43 | |
| *** sree has joined #openstack-nova | 09:43 | |
| *** TravT has quit IRC | 09:44 | |
| *** TravT has joined #openstack-nova | 09:44 | |
| *** pilgrimstack has quit IRC | 09:44 | |
| *** pjm6_ is now known as pjm6 | 09:46 | |
| *** ralonsoh has joined #openstack-nova | 09:46 | |
| *** alexchadin_ is now known as alexchadin | 09:46 | |
| *** haplo37 has joined #openstack-nova | 09:47 | |
| *** pjm6 has quit IRC | 09:47 | |
| *** GnomeKing has joined #openstack-nova | 09:48 | |
| *** GnomeKing is now known as BobBall | 09:48 | |
| *** TravT has quit IRC | 09:48 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: virtlogd: use virtlogd for char devices https://review.openstack.org/323765 | 09:48 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: remove py26 compat code in "get_console_output" https://review.openstack.org/334480 | 09:48 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: read rotated "console.log" files https://review.openstack.org/385366 | 09:48 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Merge all console device creation into one method https://review.openstack.org/385367 | 09:48 |
| *** TravT has joined #openstack-nova | 09:49 | |
| *** mvk has quit IRC | 09:50 | |
| *** cdent has joined #openstack-nova | 09:51 | |
| *** yuntongjin has quit IRC | 09:53 | |
| *** TravT has quit IRC | 09:53 | |
| *** TravT has joined #openstack-nova | 09:54 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add PCIWeigher https://review.openstack.org/379524 | 09:54 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: hardware: Flatten functions https://review.openstack.org/379624 | 09:54 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Prefer non-PCI host nodes for non-PCI instances https://review.openstack.org/379625 | 09:54 |
| *** pjm6 has joined #openstack-nova | 09:55 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: doc: Update libvirt-numa guide https://review.openstack.org/374280 | 09:57 |
| *** TravT has quit IRC | 09:58 | |
| *** TravT_ has joined #openstack-nova | 09:58 | |
| openstackgerrit | Hieu LE proposed openstack/nova: Config options: improve libvirt help text (2) https://review.openstack.org/316550 | 09:59 |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add Hierarchy info https://review.openstack.org/383726 | 10:03 |
| *** TravT_ has quit IRC | 10:03 | |
| *** TravT has joined #openstack-nova | 10:03 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add child_hard_limits column https://review.openstack.org/383727 | 10:04 |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Nested Quota: Set default values to subproject https://review.openstack.org/383728 | 10:04 |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add nested quota support https://review.openstack.org/383729 | 10:04 |
| *** gszasz has joined #openstack-nova | 10:05 | |
| *** McNinja has quit IRC | 10:06 | |
| *** McNinja has joined #openstack-nova | 10:07 | |
| *** hoangcx has quit IRC | 10:08 | |
| *** TravT has quit IRC | 10:08 | |
| *** TravT has joined #openstack-nova | 10:08 | |
| *** eglynn has joined #openstack-nova | 10:10 | |
| *** smcginnis has quit IRC | 10:10 | |
| *** smcginnis_ is now known as smcginnis | 10:10 | |
| *** TravT has quit IRC | 10:12 | |
| *** smcginnis has quit IRC | 10:13 | |
| *** smcginnis has joined #openstack-nova | 10:14 | |
| *** salv-orlando has joined #openstack-nova | 10:15 | |
| *** bhagyashri_s has joined #openstack-nova | 10:16 | |
| *** bhagyashri_s has quit IRC | 10:16 | |
| *** mgoddard has joined #openstack-nova | 10:16 | |
| *** bhagyashri_s has joined #openstack-nova | 10:17 | |
| *** TravT has joined #openstack-nova | 10:17 | |
| *** esberglu has joined #openstack-nova | 10:19 | |
| *** oanson has joined #openstack-nova | 10:21 | |
| *** TravT_ has joined #openstack-nova | 10:22 | |
| *** alexchadin has quit IRC | 10:23 | |
| *** esberglu has quit IRC | 10:24 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.create notification https://review.openstack.org/384621 | 10:25 |
| *** mvk has joined #openstack-nova | 10:25 | |
| *** TravT has quit IRC | 10:26 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Add {get_,set_}aggregates to objects.ResourceProvider https://review.openstack.org/355263 | 10:26 |
| openstackgerrit | Chris Dent proposed openstack/nova: Implement get and set aggregates in the placement API https://review.openstack.org/357993 | 10:26 |
| openstackgerrit | Chris Dent proposed openstack/nova: [placement] add a placement_aggregates table to api_db https://review.openstack.org/362863 | 10:26 |
| *** TravT_ has quit IRC | 10:26 | |
| *** TravT has joined #openstack-nova | 10:27 | |
| *** dosaboy_ is now known as dosaboy | 10:28 | |
| openstackgerrit | sahid proposed openstack/nova: object: add new field cpu_emulator_threads_policy https://review.openstack.org/385022 | 10:28 |
| openstackgerrit | sahid proposed openstack/nova: virt: add numa constraint for emulator threads policy https://review.openstack.org/385023 | 10:28 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: return overhead if isolate emulator threads requested https://review.openstack.org/385364 | 10:28 |
| openstackgerrit | sahid proposed openstack/nova: virt: update overhead to take into account vCPUs https://review.openstack.org/385024 | 10:28 |
| *** pilgrimstack has joined #openstack-nova | 10:29 | |
| *** ftersin has quit IRC | 10:30 | |
| *** gongysh_ has quit IRC | 10:32 | |
| *** TravT_ has joined #openstack-nova | 10:32 | |
| *** TravT has quit IRC | 10:32 | |
| *** trisq has quit IRC | 10:33 | |
| johnthetubaguy | bauzas: alex_xu: I see disagreement on this one: https://review.openstack.org/#/c/379655/ | 10:34 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Use more specific asserts in tests https://review.openstack.org/363043 | 10:35 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: hacking: Use assertIs(Not), assert(True|False) https://review.openstack.org/363685 | 10:35 |
| bauzas | johnthetubaguy: I'll remove my -1 but I'll need to review again | 10:35 |
| johnthetubaguy | I am tempted by request the extra negative test you are looking for | 10:36 |
| bauzas | johnthetubaguy: yeah, that's why I put my -1 | 10:36 |
| bauzas | but that's something we can merge later | 10:36 |
| bauzas | I'll rather put a +1 asking for a negative test | 10:36 |
| *** takashin has joined #openstack-nova | 10:37 | |
| *** ducttape_ has joined #openstack-nova | 10:37 | |
| *** TravT_ has quit IRC | 10:37 | |
| johnthetubaguy | bauzas: OK | 10:38 |
| *** TravT has joined #openstack-nova | 10:41 | |
| *** ducttape_ has quit IRC | 10:41 | |
| tojuvone | johnthebaguy: I got to think of "uuid" and had some more radical ideas as last comment on: https://review.openstack.org/310510 | 10:42 |
| *** code-R_ has joined #openstack-nova | 10:43 | |
| johnthetubaguy | tojuvone: you mean tell the users what host_uuid there instance is on, so they can look it up outside of Nova? | 10:44 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova-specs: resource-providers: Scheduler filters in DB https://review.openstack.org/300178 | 10:44 |
| tojuvone | johnthebaguy: yes, if go with the most radical approach | 10:45 |
| johnthetubaguy | tojuvone: we do already return a hostId, its unique to each tenant, but it should do the trick | 10:45 |
| *** avolkov has quit IRC | 10:46 | |
| *** TravT has quit IRC | 10:46 | |
| *** TravT_ has joined #openstack-nova | 10:46 | |
| *** code-R has quit IRC | 10:46 | |
| bauzas | johnthetubaguy: alaski: edleafe: dansmith: just re-proposed https://review.openstack.org/300178 for using the resource providers in the scheduler tree, and will propose myself for the implementation | 10:46 |
| bauzas | jay and mriedem being off | 10:47 |
| bauzas | johnthetubaguy: alaski: edleafe: dansmith: implementation patches should come up soon | 10:47 |
| * bauzas bbiab | 10:48 | |
| *** huanxie has quit IRC | 10:49 | |
| *** avolkov has joined #openstack-nova | 10:49 | |
| *** takashin has quit IRC | 10:50 | |
| tojuvone | johnthebaguy: ok. Surely it is that does it make sense to have some clear link from Nova or not. | 10:50 |
| *** TravT_ has quit IRC | 10:50 | |
| *** TravT has joined #openstack-nova | 10:51 | |
| *** ftersin has joined #openstack-nova | 10:52 | |
| *** ihrachys has quit IRC | 10:53 | |
| johnthetubaguy | tojuvone: sorry, not sure I get your question? | 10:55 |
| *** TravT has quit IRC | 10:56 | |
| johnthetubaguy | tojuvone: what I would do here, is prototype the system using the hostId in the server response | 10:56 |
| johnthetubaguy | tojuvone: then we will know what bits don't work right for the user or the system, and we could talk about adding those concrete items | 10:56 |
| johnthetubaguy | tojuvone: the root problem here is Nova is too big already, trying to do too many things, so we don't want to expand its scope where we don't have to | 10:57 |
| *** ssbarnea has quit IRC | 10:57 | |
| *** takashin has joined #openstack-nova | 10:58 | |
| tojuvone | johnthetubaguy: yes, I got that we do not want much change to Nova. So then thought just how to have the link | 10:59 |
| *** dikonoo has quit IRC | 10:59 | |
| *** haplo37_ has quit IRC | 10:59 | |
| tojuvone | johnthebaguy: And then thought that actually the host_uuid wold be the ting or as you said hostid | 10:59 |
| *** TravT has joined #openstack-nova | 11:00 | |
| *** haplo37_ has joined #openstack-nova | 11:01 | |
| tojuvone | johnthebaguy: one could then form what ever url something like: http://mythingy/<tenant_id>/<hostid> | 11:02 |
| johnthetubaguy | tojuvone: yeah, that could work, and mean no changes in Nova, I think? | 11:04 |
| tojuvone | johnthebaguy: So yes, as this would clearly be the thing, there should be this mythingy/prototype first | 11:04 |
| *** TravT has quit IRC | 11:05 | |
| johnthetubaguy | I suspect we might still need the URL, but would be good to have that prototype to see if it really is needed | 11:05 |
| tojuvone | johnthebaguy: If looking where this goes, I would agree no changes to Nova | 11:05 |
| *** TravT has joined #openstack-nova | 11:05 | |
| openstackgerrit | veena proposed openstack/nova: Adding hugepage and NUMA support check for aarch64 https://review.openstack.org/372304 | 11:06 |
| *** sdague has joined #openstack-nova | 11:06 | |
| johnthetubaguy | tojuvone: you will need to share a key with Nova, to ensure the HostIds can match up on both sides, I suspect, but thats probably OK | 11:06 |
| tojuvone | Yes, url just might have the problem of being tenant specific. | 11:08 |
| *** sree has quit IRC | 11:08 | |
| *** TravT has quit IRC | 11:10 | |
| *** TravT_ has joined #openstack-nova | 11:10 | |
| *** sree has joined #openstack-nova | 11:11 | |
| johnthetubaguy | tojuvone: you can convert inside the service, and you will have the tenant-id from the keystone auth, so it should be OK | 11:13 |
| *** pilgrimstack has quit IRC | 11:13 | |
| *** amotoki has joined #openstack-nova | 11:14 | |
| *** TravT_ has quit IRC | 11:14 | |
| tojuvone | johnthebaguy: ok, so the url could be generic without tenant_id | 11:14 |
| *** TravT has joined #openstack-nova | 11:15 | |
| *** bvanhav has joined #openstack-nova | 11:16 | |
| tojuvone | I could start with prototype, but soon might need to gather some people with common interest. Surely this might be much more interesting than anything else :) | 11:17 |
| johnthetubaguy | tojuvone: its worth trying the large operators group, they might have some good ideas / people | 11:18 |
| *** rtheis has joined #openstack-nova | 11:18 | |
| johnthetubaguy | tojuvone: basically, most folks don't want users to find out where each others VMs are, so either way, it will have to be a tenant specific URL | 11:19 |
| *** mkoderer has quit IRC | 11:19 | |
| *** TravT has quit IRC | 11:19 | |
| *** TravT has joined #openstack-nova | 11:19 | |
| tojuvone | johnthebaguy: Yes, clearly tenant specific and without knowledge of real physical host for other than admin access to "hostid" | 11:20 |
| *** mkoderer has joined #openstack-nova | 11:20 | |
| tojuvone | johnthebaguy: Abandon spec, or something to keep the conversation, if the url will still come later? I guess abandoned spec can anyhow be referred later anyhow. | 11:22 |
| *** salv-orlando has quit IRC | 11:23 | |
| *** TravT has quit IRC | 11:24 | |
| *** TravT_ has joined #openstack-nova | 11:24 | |
| *** TravT_ has quit IRC | 11:29 | |
| *** TravT has joined #openstack-nova | 11:29 | |
| *** tuanluong has quit IRC | 11:32 | |
| *** phuongnh has quit IRC | 11:33 | |
| *** TravT has quit IRC | 11:34 | |
| *** TravT_ has joined #openstack-nova | 11:34 | |
| *** sdague has quit IRC | 11:37 | |
| *** sdague has joined #openstack-nova | 11:37 | |
| *** alexchadin has joined #openstack-nova | 11:38 | |
| *** TravT_ has quit IRC | 11:38 | |
| tdurakov | markus_z: hi,are you around? | 11:40 |
| markus_z | tdurakov: yep, I'm here | 11:40 |
| tdurakov | markus_z: walking through bp/libvirt-virtlogd patches | 11:41 |
| tdurakov | markus_z: missed that first, https://review.openstack.org/#/c/334480/17/nova/virt/libvirt/driver.py - line 2632 | 11:42 |
| tdurakov | markus_z: what's the purpose of that double check that file exists? | 11:42 |
| *** TravT_ has joined #openstack-nova | 11:43 | |
| markus_z | tdurakov: whops, that's not needed | 11:44 |
| tdurakov | markus_z: missed that first, only noticed it on second patch... | 11:45 |
| markus_z | I can remove that in a follow up patch. It doesn't hurt right now. Is that OK with you? | 11:46 |
| tdurakov | markus_z: let me check all chain first | 11:46 |
| *** thorst has joined #openstack-nova | 11:47 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add PCIWeigher https://review.openstack.org/379524 | 11:47 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: hardware: Flatten functions https://review.openstack.org/379624 | 11:47 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Prefer non-PCI host nodes for non-PCI instances https://review.openstack.org/379625 | 11:47 |
| markus_z | sure, thanks for reviewing. | 11:47 |
| *** takashin has quit IRC | 11:48 | |
| *** TravT_ has quit IRC | 11:48 | |
| *** TravT has joined #openstack-nova | 11:48 | |
| *** BobBall has quit IRC | 11:48 | |
| *** salv-orlando has joined #openstack-nova | 11:51 | |
| *** TravT has quit IRC | 11:53 | |
| *** TravT has joined #openstack-nova | 11:53 | |
| *** dikonoo has joined #openstack-nova | 11:55 | |
| *** TravT has quit IRC | 11:58 | |
| *** TravT has joined #openstack-nova | 11:58 | |
| *** jaypipes has joined #openstack-nova | 11:59 | |
| jaypipes | morning Novas. | 11:59 |
| *** tblakes has joined #openstack-nova | 12:00 | |
| *** TravT has quit IRC | 12:02 | |
| *** sree_ has joined #openstack-nova | 12:04 | |
| *** sree_ is now known as Guest28741 | 12:04 | |
| *** zhangyufei has quit IRC | 12:05 | |
| *** sree has quit IRC | 12:06 | |
| *** zhangyufei has joined #openstack-nova | 12:06 | |
| *** ralonsoh has quit IRC | 12:06 | |
| *** ralonsoh_ has joined #openstack-nova | 12:06 | |
| *** matrohon has joined #openstack-nova | 12:07 | |
| *** edmondsw has joined #openstack-nova | 12:07 | |
| *** TravT has joined #openstack-nova | 12:07 | |
| *** pradk has joined #openstack-nova | 12:08 | |
| *** esberglu has joined #openstack-nova | 12:08 | |
| *** ducttape_ has joined #openstack-nova | 12:09 | |
| pkholkin | morning Jay! | 12:09 |
| *** TravT has quit IRC | 12:12 | |
| *** TravT has joined #openstack-nova | 12:12 | |
| *** TravT has quit IRC | 12:13 | |
| *** esberglu has quit IRC | 12:13 | |
| bauzas | jaypipes: morning | 12:14 |
| bauzas | jaypipes: in case you haven't seen it yet, https://review.openstack.org/#/c/300178/ | 12:15 |
| bauzas | jaypipes: I'm just beginning to implement it | 12:15 |
| *** takashin has joined #openstack-nova | 12:16 | |
| bauzas | jaypipes: do we have an etherpad like we did for Newton ? https://etherpad.openstack.org/p/placement-next | 12:17 |
| markus_z | anyone encountered an issue with os-vif objects in unit tests? http://paste.openstack.org/show/585448/ | 12:18 |
| markus_z | I tried on current master and also recreated my tox env. | 12:18 |
| *** pilgrimstack has joined #openstack-nova | 12:18 | |
| *** eharney has joined #openstack-nova | 12:18 | |
| markus_z | I also updated the os-vif package although it was already the most current 1.2.1 | 12:21 |
| *** sdake has joined #openstack-nova | 12:21 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Remove deprecated ``use_glance_v1`` https://review.openstack.org/384991 | 12:22 |
| *** sdake has quit IRC | 12:22 | |
| *** maticue has joined #openstack-nova | 12:22 | |
| *** pilgrimstack has quit IRC | 12:22 | |
| *** pilgrimstack has joined #openstack-nova | 12:22 | |
| *** sdake has joined #openstack-nova | 12:23 | |
| *** TravT has joined #openstack-nova | 12:27 | |
| *** tblakes_ has joined #openstack-nova | 12:30 | |
| *** tblakes_ has quit IRC | 12:30 | |
| *** tblakes has quit IRC | 12:30 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: Remove sata bus for virtuozzo hypervisor https://review.openstack.org/282398 | 12:30 |
| *** tblakes has joined #openstack-nova | 12:30 | |
| *** mgoddard has quit IRC | 12:31 | |
| *** TravT has quit IRC | 12:31 | |
| *** TravT has joined #openstack-nova | 12:31 | |
| *** annegentle has joined #openstack-nova | 12:32 | |
| *** TravT has quit IRC | 12:35 | |
| *** TravT has joined #openstack-nova | 12:36 | |
| *** lamt has quit IRC | 12:37 | |
| *** prithiv has joined #openstack-nova | 12:38 | |
| prithiv | can someone review this please https://blueprints.launchpad.net/nova/+spec/modify-default-cpu-mode | 12:38 |
| *** pbandark has quit IRC | 12:39 | |
| *** prateek has quit IRC | 12:39 | |
| *** ducttape_ has quit IRC | 12:39 | |
| *** TravT has quit IRC | 12:40 | |
| *** TravT has joined #openstack-nova | 12:41 | |
| *** links has quit IRC | 12:42 | |
| *** alexchadin has quit IRC | 12:42 | |
| *** alexchadin has joined #openstack-nova | 12:42 | |
| *** TravT has quit IRC | 12:45 | |
| *** TravT has joined #openstack-nova | 12:45 | |
| *** jheroux has joined #openstack-nova | 12:46 | |
| *** TravT has quit IRC | 12:46 | |
| *** TravT has joined #openstack-nova | 12:47 | |
| *** kaisers_ has joined #openstack-nova | 12:48 | |
| *** hshiina has joined #openstack-nova | 12:48 | |
| *** mriedem has joined #openstack-nova | 12:49 | |
| *** haplo37 has quit IRC | 12:49 | |
| alex_xu | nova-api meeting is in 10 mins at #openstack-meeting-4 | 12:50 |
| *** TravT has quit IRC | 12:51 | |
| *** mah has joined #openstack-nova | 12:52 | |
| *** TravT has joined #openstack-nova | 12:52 | |
| *** haplo37 has joined #openstack-nova | 12:52 | |
| *** gibi has quit IRC | 12:52 | |
| *** gibi has joined #openstack-nova | 12:52 | |
| mah | Hi all, How to activate the Enhancement Performance Awareness property in openstack ? and what I need to do that ? is there a guide for it ? | 12:52 |
| *** GB21 has quit IRC | 12:53 | |
| *** kaisers_ has quit IRC | 12:53 | |
| *** salv-orlando has quit IRC | 12:53 | |
| mriedem | dansmith: thanks for the stellar yelp review on my newton release patch | 12:54 |
| bauzas | mriedem: you were off so no way to hit you but I just added https://review.openstack.org/#/c/300178/2 | 12:54 |
| mriedem | ok | 12:55 |
| *** dave-mccowan has joined #openstack-nova | 12:57 | |
| *** octopusZ1ang has joined #openstack-nova | 12:57 | |
| *** TravT has quit IRC | 12:57 | |
| *** TravT has joined #openstack-nova | 12:58 | |
| *** zhangyufei has quit IRC | 12:58 | |
| *** gmann_ has joined #openstack-nova | 12:59 | |
| *** sree has joined #openstack-nova | 13:00 | |
| bauzas | mriedem: alaski: in case I want to write a functional test for verifying that the RequestSpec object is deleted when we call an instance delete, I'm a bit torn about adding it either in nova.tests.functional.test_servers or in nova.tests.functional.wsgi.test_servers | 13:01 |
| alex_xu | sdague: johnthetubaguy are you around for api meeting? | 13:01 |
| bauzas | mriedem: alaski: given I don't really need to test every API version, I think it should be okay if I would add in the latter, right? | 13:01 |
| mriedem | bauzas: just run it against 2.latest | 13:02 |
| sdague | alex_xu: o/ | 13:02 |
| bauzas | mriedem: yup, but in the wsgi module, right? | 13:02 |
| openstackgerrit | Alexis Lee proposed openstack/nova: Mutable compute options: reserves, ratios https://review.openstack.org/349561 | 13:02 |
| mriedem | bauzas: i'm not sure | 13:02 |
| alaski | bauzas: yeah, I would go with the wsgi one | 13:02 |
| mriedem | *i'm not sure what the difference is | 13:02 |
| *** TravT has quit IRC | 13:02 | |
| bauzas | mriedem: because I'm a bit confused about the difference | 13:02 |
| bauzas | yeah that | 13:02 |
| *** TravT has joined #openstack-nova | 13:03 | |
| bauzas | I think test_servers (not wsgi) is for testing the changes between 2 microversions | 13:03 |
| alaski | functional/test_servers appears to be about testing various microversions | 13:03 |
| alaski | yeah | 13:03 |
| *** baoli has joined #openstack-nova | 13:03 | |
| bauzas | okay | 13:03 |
| bauzas | cool then | 13:03 |
| bauzas | alaski: thanks | 13:03 |
| *** ratailor has quit IRC | 13:03 | |
| alaski | wsgi/test_servers is just testing behaviour on latest version. but not API behavior, but what happens in response to an API call | 13:04 |
| *** gongysh_ has joined #openstack-nova | 13:04 | |
| *** Guest28741 has quit IRC | 13:04 | |
| *** baoli_ has joined #openstack-nova | 13:04 | |
| bauzas | okay I see | 13:05 |
| *** andymaier has joined #openstack-nova | 13:05 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.create notification https://review.openstack.org/384621 | 13:06 |
| *** TravT has quit IRC | 13:07 | |
| *** mtanino has joined #openstack-nova | 13:08 | |
| *** esberglu_ has joined #openstack-nova | 13:08 | |
| *** baoli has quit IRC | 13:08 | |
| mah | Hi all, How to activate the Enhancement Performance Awareness property in openstack ? and what I need to do that ? is there a guide for it ? | 13:08 |
| *** prithiv has quit IRC | 13:09 | |
| bauzas | mah: I don't know what is that | 13:09 |
| *** gouthamr has joined #openstack-nova | 13:09 | |
| bauzas | mah: and we don't have properties, we have services and options :) | 13:09 |
| bauzas | mah: oh that ? https://wiki.openstack.org/wiki/Enhanced-platform-awareness-pcie | 13:10 |
| *** sree has quit IRC | 13:10 | |
| *** lamt has joined #openstack-nova | 13:11 | |
| *** mnasiadka has quit IRC | 13:11 | |
| bauzas | the above looks like a blueprint description to me | 13:11 |
| *** TravT has joined #openstack-nova | 13:12 | |
| *** alexchadin has quit IRC | 13:12 | |
| *** cdelatte has joined #openstack-nova | 13:13 | |
| *** rcernin has quit IRC | 13:13 | |
| *** rcernin has joined #openstack-nova | 13:13 | |
| *** mdrabe has joined #openstack-nova | 13:14 | |
| *** vishwanathj has joined #openstack-nova | 13:15 | |
| *** TravT has quit IRC | 13:16 | |
| *** sree has joined #openstack-nova | 13:17 | |
| lbeliveau | mah: this is been in nova for a while, have a look at http://docs.openstack.org/draft/networking-guide/config-sriov.html | 13:17 |
| *** mdrabe_ has joined #openstack-nova | 13:17 | |
| *** mtanino has quit IRC | 13:17 | |
| mah | Thanks | 13:18 |
| *** mdrabe_ has quit IRC | 13:18 | |
| *** mdrabe has quit IRC | 13:18 | |
| *** jamesdenton has joined #openstack-nova | 13:19 | |
| *** sree has quit IRC | 13:23 | |
| *** READ10 has joined #openstack-nova | 13:24 | |
| *** tongli has joined #openstack-nova | 13:24 | |
| *** TravT has joined #openstack-nova | 13:24 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: nova-manage: Use consistent prints for errors https://review.openstack.org/289309 | 13:25 |
| *** ducttape_ has joined #openstack-nova | 13:25 | |
| *** pilgrimstack has quit IRC | 13:27 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: Remove sata bus for virtuozzo hypervisor https://review.openstack.org/282398 | 13:28 |
| *** sree has joined #openstack-nova | 13:29 | |
| dansmith | mriedem: heh, it was an ebay review, but you're welcome | 13:29 |
| *** TravT_ has joined #openstack-nova | 13:30 | |
| *** TravT has quit IRC | 13:30 | |
| *** ducttape_ has quit IRC | 13:30 | |
| *** mordred has quit IRC | 13:30 | |
| *** phschwartz has quit IRC | 13:30 | |
| *** xyang1 has joined #openstack-nova | 13:31 | |
| wznoinsk | mriedem: hi, the 'free space' issue in Intel NFV CI was solved this morning and rechecks posted, it was late European time so nobody really on the hook to monitor it, thanks again for poking | 13:32 |
| *** TravT_ has quit IRC | 13:34 | |
| *** sree has quit IRC | 13:35 | |
| *** sree has joined #openstack-nova | 13:35 | |
| *** mordred has joined #openstack-nova | 13:35 | |
| mriedem | wznoinsk: thanks | 13:35 |
| *** phschwartz has joined #openstack-nova | 13:36 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: deprecate ``periodic_enable`` https://review.openstack.org/385478 | 13:36 |
| *** TravT has joined #openstack-nova | 13:39 | |
| *** Oku_OS is now known as Oku_OS-away | 13:41 | |
| *** Oku_OS-away is now known as Oku_OS | 13:42 | |
| *** annegentle has quit IRC | 13:42 | |
| *** TravT has quit IRC | 13:43 | |
| *** TravT has joined #openstack-nova | 13:44 | |
| *** cleong has joined #openstack-nova | 13:44 | |
| dansmith | derekh: hey did you see and/or try the fix for the archiving thing? | 13:44 |
| danpawlik | Hello, if you have some free time pls review https://review.openstack.org/#/c/314928/ | 13:46 |
| openstackgerrit | Sreeram Vancheeswaran proposed openstack/nova-specs: Adding the specification for the blueprint https://blueprints.launchpad.net/nova/+spec/dpm-driver https://review.openstack.org/385484 | 13:47 |
| *** pilgrimstack has joined #openstack-nova | 13:48 | |
| *** burt has joined #openstack-nova | 13:48 | |
| *** TravT has quit IRC | 13:48 | |
| mriedem | snikitin: mdbooth: put a comment/question in https://review.openstack.org/#/c/357884/ about the disk id - could we use a bdm uuid for that as a unique identifier? | 13:49 |
| mriedem | so we at least return something unique? | 13:49 |
| mriedem | if we're not going to return the device name / mountpoint or virtual device tag | 13:50 |
| *** esberglu_ has quit IRC | 13:51 | |
| mdbooth | mriedem: This is for admin diagnostics though, right? | 13:51 |
| mriedem | os-diagnostics is an admin api by default yes | 13:51 |
| *** ducttape_ has joined #openstack-nova | 13:51 | |
| mdbooth | Is an admin ever going to want to correlate this info over a long time period? | 13:51 |
| mdbooth | Or are they just going to look at a misbehaving disk and say: your second disk is going crazy, stop that. | 13:52 |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Remove few TODO's from cells.py https://review.openstack.org/385140 | 13:52 |
| mriedem | mdbooth: given we don't know who uses this api or how, i can't really answer that | 13:52 |
| mdbooth | mriedem: That's the real problem :/ | 13:52 |
| mriedem | but there was discussion in that spec at one point about returning an id | 13:52 |
| johnthetubaguy | sdague: macsz: is there more code we should be deleting in here? https://review.openstack.org/#/c/384991/3 | 13:52 |
| mdbooth | mriedem: Right. When I proposed that I thought it was for users. | 13:53 |
| mriedem | an admin could take the bdm uuid and look up bdm details in the db if needed | 13:53 |
| mdbooth | Users might want that and/or device tags. | 13:53 |
| mdbooth | Admins, though... | 13:53 |
| sdague | johnthetubaguy: maybe, I'd rather get rid of that which is th user exposure, and do follow up deletes later | 13:53 |
| mdbooth | It's hard to design something if you don't know what it's for imho. | 13:53 |
| *** jaosorior has quit IRC | 13:53 | |
| johnthetubaguy | sdague: yeah, I think I am with you on that, good to kill the *use* of it quickly | 13:54 |
| *** pcaruana has joined #openstack-nova | 13:54 | |
| mdbooth | mriedem: I also understood that a concern of using bdm id was that it would require a trip to the db, whereas pulling info out of the hypervisor is 100% local to the compute host. | 13:54 |
| mriedem | true | 13:55 |
| mriedem | hell maybe we should just deprecate the api altogether | 13:55 |
| mdbooth | If it's for an admin to poke around with manually they can probably work it out from context. | 13:56 |
| mdbooth | If it's for automating something, we should really understand what. | 13:56 |
| *** mgoddard has joined #openstack-nova | 13:56 | |
| mriedem | i'll query the ops list | 13:56 |
| dansmith | melwitt: since you pushed this last, do you want to fix up these nits? https://review.openstack.org/#/c/314928/20 | 13:57 |
| *** ljxiash has joined #openstack-nova | 13:58 | |
| jaypipes | bauzas: I had asked cdent to work on integrating the scheduler with calls to the placement REST API. can you chat with cdent about what you have already started please? | 13:58 |
| *** nherciu has quit IRC | 13:59 | |
| *** anush_ has joined #openstack-nova | 14:00 | |
| *** kfarr_ has joined #openstack-nova | 14:02 | |
| *** gmann_ has quit IRC | 14:02 | |
| *** pilgrimstack has quit IRC | 14:03 | |
| lyarwood | cdent / mriedem ; if you have a second can you take a look at https://bugs.launchpad.net/nova/+bug/1632723 - regarding a missed WebOb min version bump in Newton. Does this warrent a global requirements bump? | 14:03 |
| openstack | Launchpad bug 1632723 in OpenStack Compute (nova) "New WebOb minimum version requirement of >=1.6.1" [Undecided,New] | 14:03 |
| * cdent looks | 14:04 | |
| lyarwood | cdent: annnnnnnnnd I didn't explain why it now requires 1.6.1 sorry, https://github.com/Pylons/webob/commit/87c8749a57c1ff2442db2d74d9fb86935b7b201e is the reason. | 14:04 |
| * lyarwood adds a comment | 14:04 | |
| mriedem | lyarwood: if it fails w/o 1.6.1 then g-r min should be bumped | 14:06 |
| mriedem | it passes upstream because of upper-constraints and we don't test lower bounds | 14:06 |
| mriedem | File "nova/api/openstack/placement/util.py", line 131 doesn't match up for me though | 14:07 |
| lyarwood | pretty sure that was master and not stable/newton | 14:07 |
| * lyarwood checks | 14:07 | |
| mriedem | i think you have the wrong change | 14:08 |
| mriedem | actually you have the right one | 14:09 |
| mriedem | it just moved later | 14:09 |
| mriedem | https://review.openstack.org/#/c/352573/6/nova/api/openstack/placement/util.py@41 | 14:09 |
| lyarwood | mriedem: phew, and that trace lines up for me on master at the moment. | 14:09 |
| kashyap | mdbooth: Randomly chiming in, looking at your comment on that spec mriedem, speaking of machine types. At FOSDEM earlier this year one of the QEMU devs brought up the idea of exposing machine types to Nova users | 14:09 |
| mdbooth | kashyap: I still agree we should do that, btw. | 14:10 |
| mriedem | lyarwood: yeah so we need at least webob>=1.6.0 for https://github.com/Pylons/webob/commit/87c8749a57c1ff2442db2d74d9fb86935b7b201e | 14:10 |
| mriedem | lyarwood: but we're screwed on stable/newton for that now | 14:10 |
| *** anush_ has quit IRC | 14:10 | |
| mriedem | we aren't going to raise the minimum webob in stable/newton | 14:10 |
| mdbooth | kashyap: Not this api, though. Metadata api would probably be the place for that? | 14:10 |
| lyarwood | mriedem: maybe a release note then? | 14:10 |
| mriedem | lyarwood: or a patch that handles both versions | 14:11 |
| wznoinsk | tdurakov: | 14:11 |
| wznoinsk | tdurakov: ping | 14:11 |
| mriedem | before and after 1.6.0 | 14:11 |
| kashyap | mdbooth: Sure, I don't disagree with your comment there. (It just reminded me of this topic, which probably deserves an independent discussion.) | 14:11 |
| tdurakov | wznoinsk: hi | 14:11 |
| lyarwood | mriedem: kk, I'll take a look at that | 14:11 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: Straighten driver.cleanup & destroy interfaces https://review.openstack.org/385492 | 14:11 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: Cleanup instance directory on revert resize https://review.openstack.org/385493 | 14:11 |
| lyarwood | mriedem: thanks | 14:11 |
| kashyap | mdbooth: Do you know top off your head, what else kind of QEMU-specific details we expose in metadata API? | 14:11 |
| mriedem | lyarwood: well it's a good thing the placement api is optional i guess :) | 14:11 |
| *** ihrachys has joined #openstack-nova | 14:11 | |
| mdbooth | kashyap: None afaik. | 14:12 |
| *** READ10 has quit IRC | 14:12 | |
| kashyap | Also good reminder about `virt-what`, speaking of which, `systemd` alread usurped it :-) | 14:12 |
| kashyap | [stack@f24vm devstack]$ systemd-detect-virt | 14:12 |
| kashyap | qemu | 14:12 |
| cdent | mriedem, lyarwood: Did something happen incorrectly here, or is it just a slip through the cracks as a result of not testing lower bounds? | 14:12 |
| mdbooth | It's for guest config info, stuff that would normally go on a config disk. | 14:12 |
| wznoinsk | tdurakov: hi, we've enabled live-migration api tests for multinode job: http://intel-openstack-ci-logs.ovh/65/323765/25/check/tempest-dsvm-multinode-ovsdpdk-nfv-networking-xenial/50834ef/logs/testr_results.html.gz | 14:13 |
| cdent | I had assumed the upper-constraints was the "packaging target"? | 14:13 |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: Don't parse PCI whitelist eveytime neutron ports are created https://review.openstack.org/337236 | 14:13 |
| tdurakov | wznoinsk: cool! | 14:13 |
| lyarwood | cdent: I think it is, however for some reason downstream we were only carrying 1.4.1 and hit this | 14:13 |
| wznoinsk | tdurakov: I'm wondering what state of live migration testing you want to achieve before the summit, we'd like to follow and do the same for ovsdpdk with nfv | 14:13 |
| cdent | lyarwood: so I don't need to be filled with crushing guilt (at least not for this)? | 14:14 |
| kashyap | mdbooth: Got it. | 14:14 |
| lyarwood | cdent: ha no I shouldn't think so ;) | 14:14 |
| tdurakov | wznoinsk: it's focus on upgrades right now, and me trying to reenable ceph | 14:14 |
| cdent | :) | 14:14 |
| mriedem | cdent: well, generally should check the minimum in g-r when passing new kwargs to an external library | 14:14 |
| mriedem | cdent: it failed on the review side as well | 14:14 |
| mriedem | cdent: so you should feel at least a tiny bit guilty is what i'm saying :) | 14:15 |
| bauzas | jaypipes: cdent: why should we call the placement API from the scheduler ? | 14:15 |
| tdurakov | wznoinsk: upgrades requires back and forth live-migration testing during upgrade process(grenade) | 14:15 |
| jaypipes | bauzas: yes... | 14:15 |
| cdent | mriedem: yes sir, will do | 14:15 |
| bauzas | jaypipes: cdent: if you see https://review.openstack.org/#/c/300178/2 I think we should just check the objects directly | 14:15 |
| *** READ10 has joined #openstack-nova | 14:15 | |
| mriedem | lyarwood: for master https://review.openstack.org/#/c/385495/ | 14:15 |
| bauzas | cdent: if you agree, I'd prefer to work on implementing https://review.openstack.org/#/c/300178/2 | 14:16 |
| jaypipes | bauzas: that was written before we thought we'd have a placement REST API done and ready for querying. | 14:16 |
| cdent | bauzas: I think we should definitely _not_ query the objects, as that violates a boundary we're hoping to establish | 14:16 |
| *** mnestratov has joined #openstack-nova | 14:17 | |
| wznoinsk | tdurakov: so in short https://review.openstack.org/#/c/379638/ would be enabled for grenade jobs, is there a value in a normal (non-grenaded) runs? | 14:17 |
| bauzas | cdent: jaypipes: my thought is that the current scheduler is the same project that the placement API | 14:17 |
| wznoinsk | tdurakov: s/,/? | 14:17 |
| lyarwood | mriedem: thanks, I was just about to git review the very same ;) | 14:17 |
| lyarwood | mriedem: want me to look into stable/newton still? | 14:17 |
| mriedem | bauzas: everything in nova should go through the placement REST API | 14:17 |
| tdurakov | wznoinsk: I think, no | 14:17 |
| mriedem | lyarwood: yes | 14:17 |
| bauzas | cdent: jaypipes: so I don't understand why we need HTTP calls for something that can be done either directly or by RPC | 14:17 |
| mriedem | pelase | 14:17 |
| lyarwood | mriedem: np | 14:17 |
| derekh | dansmith: I saw the fix, will find a recent package to upgrade too and give it a go | 14:18 |
| mriedem | bauzas: we went over that in oregon | 14:18 |
| mriedem | we aren't using rpc for comms with the placement service | 14:18 |
| dansmith | derekh: cool thanks | 14:18 |
| mriedem | or leaking into access the api db for placement things | 14:18 |
| dansmith | what mriedem said | 14:18 |
| tdurakov | wznoinsk: it's for upgrades purposes only. e.g. test backward rpc api compatibility | 14:18 |
| *** anush_ has joined #openstack-nova | 14:18 | |
| jaypipes | bauzas: the Nova scheduler may at some point in the future be the placement API, but we're a ways off from that. | 14:18 |
| jaypipes | bauzas: remember that the end idea is to have a broken out scheduler that is usable by more than just Nova | 14:18 |
| bauzas | so that means each instance action will call the placement API with HTTP | 14:19 |
| wznoinsk | tdurakov: as we try not to throw grenades in here ourselves... would there be anything else about LM you'd like to see tested (by 3rd party CI) ? | 14:19 |
| jaypipes | bauzas: boot, migrate and live-migrate will, yes | 14:19 |
| bauzas | resize, etc. but yeah | 14:19 |
| wznoinsk | tdurakov: in next 2-3 weeks timeframe for start | 14:19 |
| jaypipes | resize == migrate | 14:19 |
| bauzas | yeah I know | 14:19 |
| *** mnestratov|2 has quit IRC | 14:20 | |
| bauzas | rebuild too | 14:20 |
| bauzas | anyway | 14:20 |
| bauzas | okay, if that's a consensus, I can modify the spec then | 14:20 |
| *** palexster has quit IRC | 14:20 | |
| bauzas | and I'll be modifying that to rather call the placement API instead of directly using the objects | 14:20 |
| bauzas | cdent: jaypipes: WFY ? | 14:20 |
| mriedem | yes please | 14:20 |
| dansmith | that has been the plan | 14:21 |
| dansmith | so, yes :) | 14:21 |
| edleafe | bauzas: yes | 14:21 |
| tdurakov | wznoinsk: I think it would be ok for now, and be ready for special nfv features and live-migration compatibility, do you use sr-iov? | 14:21 |
| jaypipes | bauzas: I've asked cdent to push code that demonstrates that -- but before that occurs he needs to push a patch with a proposed REST API modification to GET /resource-providers that will allow filtering based on available inventory capacity | 14:21 |
| cdent | bauzas: at a high level, yes, but I suspect the general thrust of the spec is a little bit off since we're still keeping around the other filters, what we want to do is call the filters with fewer candidates compute nodes | 14:21 |
| jaypipes | bauzas: which is why I asked you to please coordinate closely with cdent :) | 14:21 |
| *** burgerk has joined #openstack-nova | 14:22 | |
| jaypipes | bauzas: I'd appreciate your insight and help in crafting that REST API modification. | 14:22 |
| wznoinsk | tdurakov: we will, not for the ovsdpdk-nfv job tho (at least for start), I guess that's after the summit topic? | 14:22 |
| bauzas | jaypipes: tbc, reviewing is good to me, but I'd love to write a few stuff | 14:22 |
| jaypipes | bauzas: specifically, we will need to convert the Nova RequestSpec object into the payload of that GET request. understood? | 14:22 |
| tdurakov | wznoinsk: yes | 14:22 |
| mriedem | lyarwood: so i guess for the newton patch, you'd basically check pkg_resources for the webob version and if webob<1.6.0 just don't pass that json_formatter kwarg | 14:22 |
| mriedem | lyarwood: it's ugly but we've done that before | 14:22 |
| bauzas | jaypipes: wait, what? | 14:22 |
| dansmith | bauzas: we discussed that in portland too | 14:23 |
| dansmith | bauzas: reqspec -> GET /placement | 14:23 |
| lyarwood | mriedem: ack thanks | 14:23 |
| jaypipes | bauzas: the call from the Nova scheduler to the placement REST API will need to have its payload constructed by looking at the existing Nova RequestSpec object (things like request_spec.flavor.memory_mb etc) | 14:23 |
| bauzas | jaypipes: oh that okay | 14:23 |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: ephemeral disk support for virtuozzo containers https://review.openstack.org/312488 | 14:24 |
| jaypipes | bauzas: that's what I would love to have you and cdent work closely on, since you know the ReqeuestSpec object inside out. | 14:24 |
| *** takashin has quit IRC | 14:24 | |
| bauzas | I was confused, I thought you were wanting to disable the Spec object | 14:24 |
| *** salv-orlando has joined #openstack-nova | 14:24 | |
| *** takashin has joined #openstack-nova | 14:24 | |
| jaypipes | bauzas: so if for instance, cdent can push patches modding the placement REST API and you could push patches that *call* that API from the scheduler to limit the number of compute hosts queried from the Nova DB, that would be great | 14:25 |
| cdent | that could work well, we'll need to do some brain share on what the api should be, but it's a good division of labor | 14:25 |
| jaypipes | bauzas: no, not at all. just need to transform some of the information in the req spec into the JSON payload for the GET /resource-providers call | 14:25 |
| jaypipes | bauzas: that all make sense now? :) | 14:26 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Require WebOb>=1.6.0 https://review.openstack.org/385499 | 14:26 |
| bauzas | jaypipes: okay, it's far more clear now | 14:26 |
| cdent | jaypipes: are we keen on a GET with a body (is fine with me, I just know some people don't like it) | 14:26 |
| jaypipes | cdent: I'd love you, bauzas, edleafe and alex_xu to hash out the options there. | 14:26 |
| jaypipes | cdent: I'll go along with whatever you think is best as a group. | 14:26 |
| cdent | sounds good to me | 14:27 |
| *** pcaruana has quit IRC | 14:27 | |
| *** palexster has joined #openstack-nova | 14:27 | |
| bauzas | jaypipes: I think that https://review.openstack.org/#/c/300178/2 is actually very focusing on the client side, so I'm good | 14:27 |
| bauzas | jaypipes: except the fact that it uses objects and that is currently being rewritten of course | 14:28 |
| cdent | bauzas: I have that in my list for looking at very soon, should I wait for the next revision before doing so? | 14:28 |
| *** leonaldo has joined #openstack-nova | 14:28 | |
| *** nherciu has joined #openstack-nova | 14:28 | |
| bauzas | cdent: the next rev will litterally be in a couple of minutes hopefully :) | 14:28 |
| *** andrewbogott has quit IRC | 14:28 | |
| *** andrewbogott has joined #openstack-nova | 14:28 | |
| *** stewie925 has joined #openstack-nova | 14:28 | |
| cdent | cool | 14:28 |
| mriedem | lyarwood: mdbooth: would appreciate you both reviewing this when you get a chance https://review.openstack.org/#/c/314928/ | 14:29 |
| edleafe | cdent: bauzas: jaypipes: personally, GET with a body is fine by me, but we should check with API WG guidelines | 14:29 |
| *** salv-orlando has quit IRC | 14:29 | |
| mdbooth | mriedem: I've been pretty involved with this one for a bit iirc | 14:29 |
| mriedem | yeah i see | 14:29 |
| lyarwood | mriedem: yup will do | 14:30 |
| mriedem | just wanted to make sure you guys are still ok since there is a +2 on it now | 14:30 |
| mdbooth | I was +1 until melwitt noticed something I missed :) | 14:30 |
| mriedem | she'll do that | 14:30 |
| *** bswartz has quit IRC | 14:30 | |
| bauzas | edleafe: that sounds an implementation detail to me :) | 14:30 |
| edleafe | bauzas: well, cdent asked :) | 14:30 |
| *** esberglu has joined #openstack-nova | 14:30 | |
| *** tanvirt has joined #openstack-nova | 14:31 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova-specs: Add IBM System z (S/390) PR/SM hypervisor in DPM mode as nova driver https://review.openstack.org/385484 | 14:31 |
| sfinucan | dhellmann: odd question for you | 14:33 |
| *** esberglu has quit IRC | 14:34 | |
| jaypipes | sfinucan: even question for you | 14:34 |
| dansmith | mriedem: argh that is so annoying when she does that :P | 14:34 |
| *** leonaldo has quit IRC | 14:34 | |
| *** esberglu has joined #openstack-nova | 14:34 | |
| sfinucan | dhellmann: I want to define an entry point where the command has an underscore in it. Is that possible? | 14:35 |
| sfinucan | ...figure you're the best gent to ask here | 14:35 |
| sfinucan | jaypipes: It's too e̶a̶r̶l̶y̶ late for that :P | 14:35 |
| sfinucan | https://i.ytimg.com/vi/6-F2HKLzB6c/hqdefault.jpg | 14:35 |
| *** baoli_ has quit IRC | 14:36 | |
| edleafe | cdent: bauzas: jaypipes: hmmm... couldn't find anything in the API WG docs, but after reading up on the HTTP spec, GET with a body looks like a bad idea | 14:36 |
| *** rfolco has joined #openstack-nova | 14:36 | |
| bauzas | edleafe: well | 14:36 |
| dansmith | I support method equality and you should too | 14:37 |
| *** kaisers_ has joined #openstack-nova | 14:37 | |
| cdent | edleafe: in official-land POST is probably a better choice | 14:38 |
| cdent | and is a reasonable choice in this case given the semantics we're trying to create | 14:38 |
| ftersin | mdbooth: hi. sahid blocked my series https://review.openstack.org/#/c/334609/ until https://review.openstack.org/#/c/346263/ which is blocked by you particularly. could you take a look at my answer to you there? | 14:39 |
| edleafe | cdent: yeah, I agree | 14:39 |
| edleafe | cdent: the thing I didn't consider is that caches will ignore a GET body | 14:39 |
| edleafe | they cache on URI only | 14:39 |
| stewie925 | gibi: hello gibi, would like to request your review on https://review.openstack.org/#/c/382565/5 when you get the chance? thank you | 14:39 |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add nested quota support https://review.openstack.org/383729 | 14:40 |
| *** sree has quit IRC | 14:41 | |
| * edleafe needs to cache some caffeine | 14:41 | |
| *** mlavalle has joined #openstack-nova | 14:41 | |
| *** kaisers_ has quit IRC | 14:42 | |
| mdbooth | ftersin: I see you added a change to _test_finish_migration. That test is awful, btw: it really doesn't test anything. | 14:42 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova-specs: Add IBM System z (S/390) PR/SM hypervisor in DPM mode as nova driver https://review.openstack.org/385484 | 14:43 |
| *** nherciu has quit IRC | 14:43 | |
| *** gongysh_ has quit IRC | 14:44 | |
| mdbooth | ftersin: Meh, I'll cave. I'd prefer a better test, but this one passes the bar. | 14:44 |
| openstackgerrit | Francesco Santoro proposed openstack/nova: [WIP] os-vif: add os-vif plugin for vhostuser_fp https://review.openstack.org/385061 | 14:44 |
| *** gongysh_ has joined #openstack-nova | 14:45 | |
| ftersin | mdbooth: it tests that bdi is passed to _create_image. What would you like instead? And why? | 14:45 |
| mdbooth | ftersin: I've just noticed, btw, that I completed missed that you were testing that first time out. Sorry about that, I'm blind. | 14:46 |
| tanvirt | mdbooth: Could we talk about the libvirt storage BP tommorow instead? My meetings end at 2:00pm (US Central Time) today, but I now realize thats past end of day for UK time | 14:49 |
| *** gongysh_ has quit IRC | 14:49 | |
| *** Swami has joined #openstack-nova | 14:50 | |
| mdbooth | tanvirt: Sure. Are you free 9am your time tomorrow? | 14:50 |
| *** xinliang has quit IRC | 14:50 | |
| tanvirt | mdbooth: Yes, that works for me. | 14:50 |
| mdbooth | Cool, ping me tomorrow | 14:51 |
| *** TravT has joined #openstack-nova | 14:51 | |
| *** annegentle has joined #openstack-nova | 14:51 | |
| *** gjayavelu has joined #openstack-nova | 14:54 | |
| *** Swami_ has joined #openstack-nova | 14:57 | |
| bauzas | jaypipes: cdent: dansmith: do we have any REST resource the scheduler could call out that would return a list of resource providers matching ( inventories - allocations - <proposed> > 0) for the specific resource classes that we mention ? | 14:57 |
| bauzas | jaypipes: cdent: dansmith: I mean, is this something we already spec'd ? | 14:57 |
| bauzas | because I'm updating the spec on the scheduler-side, and I want that :) | 14:57 |
| *** Swami__ has joined #openstack-nova | 14:58 | |
| edleafe | bauzas: for a single resource? Wouldn't it be better to accept multiple resources, and return those that match all? | 14:58 |
| bauzas | edleafe: for the specific resource *classes* | 14:59 |
| cdent | bauzas: tha sounds like tighter detail than we were thinking. | 14:59 |
| dansmith | bauzas: do you mean a dry run of "where might you put me if I made this request for real" ? | 14:59 |
| openstackgerrit | Hironori Shiina proposed openstack/nova: Ironic: Support boot from Cinder volume https://review.openstack.org/215385 | 14:59 |
| ftersin | mdbooth: but why do you think that this test is useless? we must verify that bdi is passed, mustn't we? do you want this to be verified other way? | 14:59 |
| edleafe | bauzas: so you want to query all providers that have at least X amount of RAM? | 14:59 |
| bauzas | dansmith: no I mean, "hey placement, please give me the providers that have enough room for accepting X of that and Y of this" | 14:59 |
| dansmith | bauzas: well, that's what I just said I think | 14:59 |
| cdent | the idea that i recall is: send request spec, get back list of resource providers | 15:00 |
| *** pbandark has joined #openstack-nova | 15:00 | |
| dansmith | bauzas: but I don't think we're there yet.. that's what we're talking about writing next right? | 15:00 |
| *** _ducttape_ has joined #openstack-nova | 15:00 | |
| edleafe | bauzas: yes, my point was on single match vs. matching a full request | 15:00 |
| mdbooth | ftersin: I think it comes under a general class of 'diff test', which essentially tests that the code is as written. | 15:00 |
| cdent | and there is as yet no clear spec for that | 15:00 |
| *** yamahata has joined #openstack-nova | 15:00 | |
| *** Swami has quit IRC | 15:00 | |
| bauzas | dansmith: well, if we want to narrow the list of compute nodes in the scheduler select_dests() call, there are 2 ways | 15:00 |
| bauzas | dansmith: either asking for that, or doing that in the scheduler code | 15:00 |
| cdent | but the POST /resource_providers with a json body (or somewhere else) that we were talking about above is the general idea | 15:01 |
| bauzas | cdent: which would return ? | 15:01 |
| dansmith | bauzas: the whole point of this effort is to ask placement for a list of hosts to start with, and then do scheduler stuff on it | 15:01 |
| *** TravT has quit IRC | 15:01 | |
| edleafe | cdent: exactly | 15:01 |
| cdent | and then that liast gets passed into the existing filtering loop | 15:01 |
| mdbooth | ftersin: It's fine, though. I really think your change is good and required, and it's absolutely not worth holding up for this imho. | 15:01 |
| dansmith | bauzas: instead of just listing all hosts from the db.. right? | 15:01 |
| cdent | what's returned is a list of compute node uuids | 15:01 |
| cdent | what dansmith says | 15:01 |
| dansmith | right, that ^ | 15:01 |
| cdent | jinx | 15:01 |
| mdbooth | ftersin: It passes the mdbooth 'are we better with or without this' test with flying colours :) | 15:02 |
| bauzas | dansmith: right, but I'd love to get a list of resource providers that would be able to fit my request, and they I would pass those corresponding nodes to the other filters | 15:02 |
| jaypipes | bauzas: yeah, that's the exact REST API we need :) | 15:02 |
| *** mjura has quit IRC | 15:02 | |
| *** Swami_ has quit IRC | 15:02 | |
| edleafe | bauzas: that's what we've been saying :) | 15:02 |
| dansmith | bauzas: that's what we're saying we need | 15:02 |
| jaypipes | bauzas: thus, me asking you all to think it through and debate the options :) | 15:02 |
| edleafe | jinx | 15:02 |
| bauzas | okay, seems like we're all in violent agreement then | 15:02 |
| *** geektoor has joined #openstack-nova | 15:03 | |
| edleafe | For compute nodes, there will be multiple requirements passed. For some future RP supoorted by the placement engine, it may only be one | 15:03 |
| *** TravT has joined #openstack-nova | 15:03 | |
| *** xinliang has joined #openstack-nova | 15:03 | |
| bauzas | I'll amend https://review.openstack.org/#/c/300178/2/specs/ocata/approved/resource-providers-scheduler-db-filters.rst to mention a specific call that would return the list of compute node uuids, the rest is magic for that spec | 15:03 |
| *** ducttape_ has quit IRC | 15:03 | |
| bauzas | jaypipes: dansmith: cdent: WFY ? ^ | 15:03 |
| bauzas | jaypipes: well, I'm fine if that HTTP call doesn't commit things in DB | 15:04 |
| jaypipes | bauzas: just say "returns resource provider records" not just UUIDs, but yep, works for me. | 15:04 |
| jaypipes | bauzas: no GET /XXX call should ever commit anything in a DB :) | 15:04 |
| bauzas | jaypipes: I'm totally fine with a GET, but I saw cdent and edleafe discussing about POST :) | 15:05 |
| cdent | bauzas: yeah, wfm, we can iterate on some of the details if we need too | 15:05 |
| cdent | getting the basic outline down is going to be a great start | 15:05 |
| *** psachin has quit IRC | 15:05 | |
| *** sdake has quit IRC | 15:05 | |
| *** rodrigods has quit IRC | 15:05 | |
| *** rodrigods has joined #openstack-nova | 15:05 | |
| *** sdake has joined #openstack-nova | 15:06 | |
| *** mnestratov has quit IRC | 15:06 | |
| *** krtaylor has joined #openstack-nova | 15:06 | |
| edleafe | cdent: exactly. Move forward, adjust as needed | 15:06 |
| bauzas | cdent: jaypipes: dansmith: if you will, I can write that other spec that would describe that specific GET :) | 15:07 |
| bauzas | and put that as a dep | 15:07 |
| bauzas | for the other spec I'm just amending | 15:07 |
| dansmith | bauzas: if we will what? | 15:07 |
| cdent | let's wait until we've had a chance to digest the current spec? | 15:07 |
| bauzas | dansmith: oops :) | 15:07 |
| cdent | and then we can discuss what can or should happen? | 15:07 |
| *** mnestratov has joined #openstack-nova | 15:07 | |
| *** markmcclain has joined #openstack-nova | 15:07 | |
| jaypipes | bauzas: up to you, I'm fine with that but based on experience from last cycle, would certainly prefer to see code over detailed specs. | 15:07 |
| bauzas | jaypipes: that's doable :) | 15:08 |
| bauzas | cdent: doesn't matter you if I'm writing some POC about a specific GET action on the /resource_providers resource that would return the list of matching RPs ? | 15:09 |
| cdent | as long as it is easy to change things as we are learning, if it turns out we need to change, then that's fine with me | 15:09 |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: Obtain network_info from api instead of info_cache https://review.openstack.org/385086 | 15:10 |
| bauzas | cdent: I'm not seeing that REST action as a big thing to modify | 15:11 |
| cdent | ✔ | 15:11 |
| bauzas | cool, I have homework ! | 15:11 |
| *** prateek has joined #openstack-nova | 15:12 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: Obtain network_info from api instead of info_cache https://review.openstack.org/385086 | 15:12 |
| *** eantyshev has quit IRC | 15:13 | |
| *** mnestratov has quit IRC | 15:15 | |
| *** pradk has quit IRC | 15:16 | |
| *** eantyshev has joined #openstack-nova | 15:17 | |
| *** rcernin has quit IRC | 15:19 | |
| *** mdrabe has joined #openstack-nova | 15:20 | |
| *** salv-orlando has joined #openstack-nova | 15:21 | |
| *** mtanino has joined #openstack-nova | 15:22 | |
| wznoinsk | tdurakov: we do test nfv enabled flavors in our jobs with live-migration, would you know whether there's any scenario LM test being thought about? | 15:23 |
| *** annegentle has quit IRC | 15:24 | |
| *** jamesdenton has quit IRC | 15:24 | |
| *** Swami__ has quit IRC | 15:27 | |
| *** rmart04 has quit IRC | 15:27 | |
| *** Swami__ has joined #openstack-nova | 15:27 | |
| *** rama_y has joined #openstack-nova | 15:27 | |
| markus_z | sdague: I think https://github.com/openstack/nova/commit/735f710 introduced (or revealed) an issue with os-vif. I get this issue: http://paste.openstack.org/show/585448/ | 15:28 |
| markus_z | sdague: But I'm not able to find that in logstash and I don't know why. | 15:29 |
| *** andreas_s has quit IRC | 15:29 | |
| *** hieulq has joined #openstack-nova | 15:30 | |
| *** esberglu has quit IRC | 15:30 | |
| *** esberglu has joined #openstack-nova | 15:31 | |
| markus_z | After applying nasty hack http://paste.openstack.org/show/585475/ the test passes with error message http://paste.openstack.org/show/585480/ | 15:31 |
| *** _ducttape_ has quit IRC | 15:31 | |
| *** esberglu_ has joined #openstack-nova | 15:32 | |
| *** ducttape_ has joined #openstack-nova | 15:32 | |
| *** baoli has joined #openstack-nova | 15:33 | |
| markus_z | I have to go for today, I'll investigate more tomorrow. | 15:33 |
| *** markus_z has quit IRC | 15:33 | |
| *** kylek3h_ is now known as kylek3h | 15:33 | |
| *** annegentle has joined #openstack-nova | 15:34 | |
| *** rfolco has quit IRC | 15:34 | |
| *** afazekas has quit IRC | 15:35 | |
| *** esberglu has quit IRC | 15:35 | |
| *** pradk has joined #openstack-nova | 15:36 | |
| *** esberglu_ has left #openstack-nova | 15:36 | |
| *** GB21 has joined #openstack-nova | 15:36 | |
| *** mordred has quit IRC | 15:36 | |
| *** nkrinner is now known as nkrinner_afk | 15:37 | |
| *** afazekas has joined #openstack-nova | 15:38 | |
| *** links has joined #openstack-nova | 15:39 | |
| *** kfarr_ has quit IRC | 15:40 | |
| *** mordred has joined #openstack-nova | 15:41 | |
| *** davidlenwell_ has joined #openstack-nova | 15:41 | |
| *** mnaser has quit IRC | 15:43 | |
| *** hieulq has quit IRC | 15:43 | |
| *** matrohon has quit IRC | 15:44 | |
| *** hieulq has joined #openstack-nova | 15:44 | |
| *** adam_g` has joined #openstack-nova | 15:45 | |
| *** davidlenwell_ has quit IRC | 15:45 | |
| *** david-lyle has joined #openstack-nova | 15:46 | |
| openstackgerrit | Sujitha proposed openstack/nova: Change host for osapi and metadata services in db. https://review.openstack.org/384724 | 15:46 |
| *** takashin has left #openstack-nova | 15:47 | |
| melwitt | dansmith: sure, I'll fix up the things on https://review.openstack.org/#/c/314928/ | 15:47 |
| *** mdrabe has quit IRC | 15:49 | |
| *** mdrabe has joined #openstack-nova | 15:49 | |
| *** baoli has quit IRC | 15:50 | |
| *** pbandark has quit IRC | 15:50 | |
| *** david-lyle has quit IRC | 15:52 | |
| *** mkoderer has quit IRC | 15:52 | |
| *** mnaser has joined #openstack-nova | 15:53 | |
| *** code-R_ has quit IRC | 15:54 | |
| *** ljxiash has quit IRC | 15:55 | |
| ftersin | mdbooth: got it, thx | 15:56 |
| bauzas | jaypipes: one point, in case you haven't yet reviewed the spec update I wrote, I think we could probably leave some opportunities for the operators to not ask for *all* the resource classes, but rather disable/enable those | 15:58 |
| bauzas | your point on that? | 15:58 |
| *** annegentle has quit IRC | 15:58 | |
| bauzas | jaypipes: the problem I see is for example with operators not wanting to check CoreFilter for example | 15:58 |
| *** TravT_ has joined #openstack-nova | 15:59 | |
| *** TravT_ has quit IRC | 16:00 | |
| *** TravT has quit IRC | 16:00 | |
| cdent | bauzas: aren't we just worried, at this point, with the interaction between the existing scheduler and the placement api and the placement api has no concept (at this time) of filters? | 16:01 |
| * cdent goes to look at the revised spec | 16:01 | |
| bauzas | cdent: well, good point | 16:01 |
| *** jamesdenton has joined #openstack-nova | 16:01 | |
| bauzas | cdent: the problem I see is that people provide their own ways | 16:01 |
| bauzas | cdent: that said, I'm thinking of something else we discussed during the midcycle, about having weighers | 16:02 |
| bauzas | in case people provide a weigher for the VCPU class as 0, then it disables it | 16:02 |
| bauzas | that could be the way to have like custom filters with custom resource classes and weighers | 16:03 |
| *** prateek has quit IRC | 16:03 | |
| openstackgerrit | melanie witt proposed openstack/nova: Fix issue with not removing rbd rescue disk https://review.openstack.org/314928 | 16:04 |
| *** thorst is now known as thorst_afk | 16:04 | |
| *** david-lyle has joined #openstack-nova | 16:06 | |
| *** annegentle has joined #openstack-nova | 16:06 | |
| jaypipes | bauzas: I'm not really interested in supporting the "use case" of an operator disabling the core filter. In the case of the placement API, the operator can simply set the allocation_ratio on the resource classes they want to provide unlimited access for to some high value like 1000.0 | 16:06 |
| bauzas | jaypipes: that sounds acceptable to me, I think dansmith provided another point during the midcycle about having weighers but that looks related to allocation ratios | 16:07 |
| jaypipes | bauzas: in other words, I'm not interested in maintaining the use-case-via-config-option that currently exists. I would prefer to have the operator simply set the data in the placement API as needed. | 16:07 |
| *** davidlenwell_ has joined #openstack-nova | 16:07 | |
| bauzas | jaypipes: so I think there are 2 things | 16:07 |
| bauzas | jaypipes: 1/ we shouldn't possibly accept that in the placement API which I agree | 16:08 |
| bauzas | jaypipes: 2/ we could still leave room for the scheduler to have "disabled"' metrics, and that would turn that into asking for 1000.0 ratio on the allocation request for example | 16:09 |
| bauzas | oh wait | 16:09 |
| bauzas | the ratios are on the inventories, right, not on the allocations | 16:09 |
| bauzas | jaypipes: ^ | 16:09 |
| *** kfarr_ has joined #openstack-nova | 16:09 | |
| * bauzas quickly checking the API | 16:09 | |
| jaypipes | bauzas: correct, allocation_ratios are on the inventory records. | 16:10 |
| bauzas | so, lemme think at the usecase | 16:10 |
| bauzas | as an operator, I'd like to not check the VCPU, so I'd put the inventory ratio related to the vcpu resource class to 0.0 | 16:11 |
| bauzas | jaypipes: that could work ? ^ | 16:11 |
| bauzas | heh no | 16:11 |
| jaypipes | bauzas: no, you'd put it to 10000.0 or something. | 16:11 |
| jaypipes | no 0.0 :) We still want basic math to work :) | 16:12 |
| bauzas | yeah, you're right | 16:12 |
| bauzas | so we would never fence on the vcpu metrics | 16:12 |
| *** rama_y has quit IRC | 16:12 | |
| bauzas | okay, that sounds acceptable | 16:12 |
| cdent | (shades of infinite resource classes :) ) | 16:12 |
| jaypipes | cdent: don't start with me, young man. | 16:12 |
| bauzas | 50 shades of resource classes, heh ? | 16:12 |
| * cdent pokes the greater florida bear | 16:12 | |
| jaypipes | cdent: she's peacefully sleeping at my feet. as is usually the case. | 16:13 |
| bauzas | jaypipes: okay, what we discussed over simplifies the spec | 16:13 |
| * jaypipes goes back to reviews | 16:13 | |
| edleafe | So I'm confused: why can't an operator disable a filter via a config option? In that case, we simply don't add that to the WHERE. | 16:13 |
| bauzas | coolness | 16:13 |
| bauzas | edleafe: no, it's ever simplier | 16:13 |
| bauzas | even* | 16:13 |
| *** slaweq_ has joined #openstack-nova | 16:14 | |
| bauzas | edleafe: say an operator doesn't want to filter on vcpus, he will update the inventory.allocation_ratio for vcpus to 10000.0 | 16:14 |
| edleafe | bauzas: which is totally non-intuitive | 16:14 |
| bauzas | edleafe: for the transition, we can still ask him to leave CoreFilter, but we tell that once we have ways to set the ratio for the inventory, we will deprecate those old filters | 16:15 |
| *** GB21 has quit IRC | 16:15 | |
| bauzas | edleafe: well, we could do that automatically | 16:15 |
| bauzas | edleafe: but that's another spec | 16:15 |
| edleafe | bauzas: it seems *much* simpler to say that if I don't want to filter on X, I remove the X filter from the config option | 16:16 |
| bauzas | edleafe: the thing is, we would still support the filter disablement by Ocata | 16:16 |
| jaypipes | edleafe: and that means you will need to have a sync'd config option in both the placement API *and* the individual nova-compute's nova.conf. | 16:16 |
| edleafe | Then have the placement only add WHERE clauses for those things that are enabled | 16:16 |
| jaypipes | edleafe: and we'll be back to the problem we originally had with a single allocation_ratio set in config options on the scheduler. | 16:16 |
| bauzas | jaypipes: I think edleafe wants some opt that would turn into us setting 100000.0 to X | 16:17 |
| edleafe | No | 16:17 |
| jaypipes | bauzas: see my comment above. | 16:17 |
| *** danpawlik has quit IRC | 16:17 | |
| edleafe | Setting the allocation ratio to an absurdly high number is only a way of faking not having the corresponding WHERE clause in the query | 16:17 |
| *** PaulMurray has joined #openstack-nova | 16:17 | |
| cdent | I think we're getting ahead of ourselves | 16:18 |
| edleafe | jaypipes: I don't see how that follows | 16:18 |
| edleafe | cdent: perhaps, but it's an important design distinction | 16:18 |
| cdent | none of the existing behavior after getting the initial list changes | 16:18 |
| bauzas | edleafe: jaypipes: one other way would be to ask in the GET request each class we want to disable | 16:18 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: Cleanup instance directory on revert resize https://review.openstack.org/385493 | 16:18 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: Straighten driver.cleanup & destroy interfaces https://review.openstack.org/385492 | 16:18 |
| bauzas | cdent: it *could* be changing and that's the problem | 16:18 |
| *** raunak has joined #openstack-nova | 16:19 | |
| bauzas | cdent: even if people disable DiskFilter, we could have the scheduler avoiding hosts based on disk if we don't ask the placement API to not verify it | 16:19 |
| edleafe | cdent: understood | 16:19 |
| *** raunak has quit IRC | 16:19 | |
| jaypipes | bottom line for me is that we have the ability to use the existing API to handle this "use case". Let's just use it and not overcomplicate things. | 16:19 |
| jaypipes | with additional configuration options. | 16:19 |
| bauzas | jaypipes: I'm definitely -2 on adding conf opts | 16:20 |
| edleafe | jaypipes: my point exactly. Setting absurd allocation ratios is complicating things | 16:20 |
| jaypipes | bauzas: you realize that's what "some operators disable the CoreFilter" is, though, right? | 16:20 |
| edleafe | bauzas: adding? Nope. I'm saying using the existing conf opt | 16:20 |
| PaulMurray | mriedem, hi - you asked me about "detach boot volumes": https://review.openstack.org/#/c/340874/ | 16:21 |
| jaypipes | edleafe: it's not an absurd allocation ratio. | 16:21 |
| *** Apoorva has joined #openstack-nova | 16:21 | |
| *** raunak has joined #openstack-nova | 16:21 | |
| edleafe | bauzas: and not requiring operators to jump through another hoop | 16:21 |
| jaypipes | edleafe: you want to see in th eplacement API the actual overconsumption of a particular resource. | 16:21 |
| PaulMurray | mriedem, paul_carlton is going to take it over | 16:21 |
| bauzas | jaypipes: yeah I know, but that's here for a very long time | 16:21 |
| jaypipes | edleafe: you just want to indicate that you, as an operator, are willing to accept a large overconsumption. | 16:21 |
| edleafe | jaypipes: it's not an allocation ratio. You are not setting overcommitment | 16:21 |
| *** ociuhandu has joined #openstack-nova | 16:21 | |
| edleafe | jaypipes: you are using something in a quirky way to achieve something completely different | 16:22 |
| jaypipes | edleafe: erm, sorry, not following that last one. could you elaborate how the allocation_ratio is not an overcommitment? | 16:22 |
| edleafe | You're not saying "overcommit this much". You're saying "ignore this thing" | 16:23 |
| cdent | can someone help me to understand: what does it mean to "ignore vcpu"? Don't we have to know about vcpu to know what we can use a compute node? | 16:23 |
| *** kevinz has quit IRC | 16:23 | |
| edleafe | cdent: perhaps it's an optimization trick that accepts retries. I don't know the real thinkning behind it | 16:24 |
| *** esberglu has joined #openstack-nova | 16:24 | |
| jaypipes | edleafe: no, you're still using the thing. you're saying "I don't care how much you consume of this thing" | 16:24 |
| bauzas | cdent: the logic is that people don't want to have an host being avoided because it's cpu-bounded | 16:25 |
| bauzas | cdent: but I agree, those people could use the alloc ratio to put that bar way higher | 16:25 |
| cdent | bauzas: okay, that means they want that node to have an insanely high allocation ratio! | 16:25 |
| edleafe | jaypipes: so why set it to 10,000? Or 100,000? Or... | 16:25 |
| bauzas | yup | 16:25 |
| *** rama_y has joined #openstack-nova | 16:25 | |
| jaypipes | this is kind of the thing I didn't want to get bogged down in, really. I don't see too much of a reason to even discuss this right now. | 16:25 |
| edleafe | cdent: infinite is not a ratio | 16:25 |
| bauzas | cdent: I don't think I disagree, I just care about babysitting ops that would have done that | 16:25 |
| *** kaisers_ has joined #openstack-nova | 16:26 | |
| edleafe | jaypipes: for ops, it's a matter of saying "turn this off" vs. "set this to a rediculously high number" | 16:26 |
| * cdent gives edleafe a cookie for the infinite callback | 16:26 | |
| bauzas | jaypipes: well, there is obviously no problem when we're talking about vcpus, but what about NUMA resources when we'll begin to count those | 16:26 |
| bauzas | like the number of cells | 16:27 |
| *** nic has joined #openstack-nova | 16:27 | |
| edleafe | jaypipes: this feels like a hack that will work with our initial design, but complicate things long-term, especially for ops | 16:28 |
| bauzas | anyway | 16:28 |
| bauzas | sounds there is a workaround that we can communicate | 16:28 |
| edleafe | bauzas: thank you. It's a workaround | 16:28 |
| bauzas | edleafe: here is the thing, I propose to leave us not to care with that for the moment, write a firm document that would explain how to do that, and leave operators ranting in case they *really* want that | 16:29 |
| *** nic has quit IRC | 16:29 | |
| *** davidlenwell_ has quit IRC | 16:29 | |
| bauzas | edleafe: because at the moment, I'm struggling with providing real usecases | 16:29 |
| bauzas | edleafe: so I don't want to overcomplicate things for stuff that probably operators wouldn't care | 16:29 |
| bauzas | and in case they care about, they will be back to us very soon | 16:30 |
| * edleafe has to go on a call now. Back in 30 | 16:30 | |
| bauzas | edleafe: jaypipes actually convinced me there is no much of a big deal here, given there is a way to workaround that | 16:30 |
| *** kaisers_ has quit IRC | 16:30 | |
| *** baoli has joined #openstack-nova | 16:31 | |
| mdbooth | dansmith: Those whitespace nits are harsh! They're all fixes immediately adjacent to code which is being legitimately changed. | 16:34 |
| *** baoli_ has joined #openstack-nova | 16:34 | |
| mdbooth | It's not as if they're result in additional merge conflicts. | 16:34 |
| *** baoli_ has quit IRC | 16:35 | |
| dansmith | mdbooth: um what? | 16:35 |
| mdbooth | https://review.openstack.org/#/c/314928/ | 16:35 |
| *** baoli_ has joined #openstack-nova | 16:35 | |
| dansmith | mdbooth: I can't tell if you're being serious or sarcastic, but .. okay | 16:36 |
| *** kevinz has joined #openstack-nova | 16:36 | |
| *** baoli has quit IRC | 16:36 | |
| mdbooth | dansmith: Serious :) It's not a big deal, but I wouldn't have called those out. | 16:36 |
| *** rama_y has quit IRC | 16:37 | |
| dansmith | mdbooth: oh, I would have.. they completely screw with git blame for no reason | 16:37 |
| mdbooth | Lots of things do that anyway... | 16:37 |
| mdbooth | And it's not as it we wouldn't fix them at all | 16:37 |
| kashyap | mdbooth: FWIW, I've seen reviewers complain in upstream QEMU, too, to not make unrelated white-space changes. I used to wonder why | 16:37 |
| mdbooth | which is the only way not to screw with blame | 16:38 |
| kashyap | Then, someone explained the 'git blame' rationale | 16:38 |
| *** slaweq_ has quit IRC | 16:38 | |
| dansmith | mdbooth: seriously it's pretty much the standard thing here to not tolerate random changes in a patch that aren't related | 16:38 |
| dansmith | mdbooth: if you want to do another patch to refactor or cleanup, then that's cool, but mixing that with functional changes is terrible | 16:38 |
| mdbooth | Yeah, and in general I agree. But whitespace fixes are in general ok, and these are immediately adjacent. | 16:39 |
| dansmith | I totally disagree | 16:39 |
| mdbooth | So they're not going to screw with context or anything, because we already changed that. | 16:39 |
| mdbooth | Anyway, not a big deal. I just thought it was harsh. | 16:39 |
| dansmith | they are in this case.. that whole instance line looks like it's part of the change, but has nothing to do with it | 16:39 |
| sneti | johnthetubaguy, I'm working on this bug: https://bugs.launchpad.net/nova/+bug/1581203. There is a patch submitted for this: https://review.openstack.org/#/c/315757/. I see your comment there. But there is already a test to check this policy in nova/tests/unit/test_policy.py. Do we need to add any other test? | 16:39 |
| openstack | Launchpad bug 1581203 in OpenStack Compute (nova) "Default policy allows unrestricted CRUD on os-server-tags" [Medium,Confirmed] - Assigned to Sujitha (sujitha-neti) | 16:39 |
| *** rama_y has joined #openstack-nova | 16:39 | |
| *** sahid has quit IRC | 16:39 | |
| *** nic has joined #openstack-nova | 16:40 | |
| *** Swami__ has quit IRC | 16:41 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova-specs: resource-providers: Scheduler filters in DB https://review.openstack.org/300178 | 16:42 |
| *** esberglu has quit IRC | 16:42 | |
| mriedem | mdbooth: fwiw i agree with dansmith, i prefer changes to be as small as possible to fix the thing, it makes it easier to review, and style cleanups can be done later if someone is really bothered by them | 16:43 |
| mdbooth | mriedem: Oh, me too. I just think dansmith is mean :) | 16:43 |
| bauzas | jaypipes: cdent: removed all occurrences of disabling specific resource classes and mentioning a future spec (that I'm going to write tonight) about getting the list of RPs that match the request in https://review.openstack.org/300178 | 16:43 |
| johnthetubaguy | sneti: I just don't think that admin_or_owner is actually any different to allow any for that API, I could be wrong | 16:43 |
| johnthetubaguy | sneti: need a test to prove that one way or the other | 16:43 |
| dansmith | mdbooth: because I +0'd a patch calling them out? | 16:43 |
| * bauzas becomes hungry, be back later | 16:43 | |
| johnthetubaguy | sneti: now we have the defaults in code, we can test that | 16:44 |
| cdent | bauzas: will look shortly, thanks! | 16:44 |
| johnthetubaguy | sneti: I mean we can test the default | 16:44 |
| mriedem | mdbooth: he didn't -1 it | 16:44 |
| *** slaweq_ has joined #openstack-nova | 16:44 | |
| mriedem | but it also keeps out a +1 or +2 in some cases when there are a lot of them | 16:45 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: WIP: Remove deprecated options from compute https://review.openstack.org/385129 | 16:45 |
| mriedem | i've done that before, basically, clean all of this up and then i'm +2 | 16:45 |
| *** links has quit IRC | 16:45 | |
| *** paul-carlton2 has joined #openstack-nova | 16:46 | |
| *** annegentle has quit IRC | 16:48 | |
| *** paul-carlton2 has quit IRC | 16:51 | |
| sneti | johnthetubaguy, you mean testing the default rule for each action? | 16:51 |
| johnthetubaguy | sneti: I just added that to talk about at the summit actually | 16:51 |
| johnthetubaguy | sneti: but in this patch, yeah, I think test the default rule, and make sure it rejects your request in the way you expect | 16:51 |
| *** ssurana has joined #openstack-nova | 16:52 | |
| *** derekh has quit IRC | 16:52 | |
| sneti | johnthetubaguy, ok..are there any tests for other actions checking the default rule. I think I didn't find any. But I will check again. | 16:53 |
| johnthetubaguy | sneti: probably there are non yet | 16:54 |
| johnthetubaguy | sneti: I think if you don't set the policy you would get the default, but I am not 100% sure where we are with that | 16:54 |
| *** unicell1 has joined #openstack-nova | 16:54 | |
| johnthetubaguy | sneti: basically, I was looking for a test that would have failed before that patch, and goes green when you add the current path | 16:55 |
| alaski | cells meeting in 5 minutes in #openstack-meeting-3 | 16:56 |
| *** unicell has quit IRC | 16:56 | |
| sneti | thanks johnthetubaguy..got it..will try to add test.. | 16:57 |
| *** rcernin has joined #openstack-nova | 16:57 | |
| johnthetubaguy | sneti: no problem, glad I could help | 16:57 |
| *** ssurana has quit IRC | 16:58 | |
| *** slaweq_ has quit IRC | 16:59 | |
| *** kfarr_ has quit IRC | 16:59 | |
| *** ssurana has joined #openstack-nova | 16:59 | |
| *** Guest24890 has quit IRC | 17:00 | |
| *** haplo37_ has quit IRC | 17:02 | |
| openstackgerrit | Emilien Macchi proposed openstack/nova: Document experimental pipeline in Nova CI https://review.openstack.org/381838 | 17:02 |
| *** ralonsoh_ has quit IRC | 17:02 | |
| openstackgerrit | Mike Perez proposed openstack/nova: Remove deprecated nova-all binary https://review.openstack.org/385564 | 17:03 |
| *** jeblair_ is now known as jeblair | 17:03 | |
| *** thingee has joined #openstack-nova | 17:04 | |
| *** paul-carlton2 has joined #openstack-nova | 17:04 | |
| stewie925 | hello guys, would like your input on an 'out of range' issue in https://review.openstack.org/#/c/382565/ | 17:04 |
| *** unicell1 has quit IRC | 17:04 | |
| *** haplo37_ has joined #openstack-nova | 17:04 | |
| stewie925 | I posted the error log in http://paste.openstack.org/show/585243/ | 17:04 |
| *** unicell has joined #openstack-nova | 17:04 | |
| *** claudiub has quit IRC | 17:05 | |
| *** unicell has quit IRC | 17:05 | |
| *** PaulMurray has quit IRC | 17:05 | |
| *** claudiub has joined #openstack-nova | 17:06 | |
| stewie925 | this is for a nova notification for 'instance.unshelve' | 17:06 |
| mriedem | stewie925: do these enums exist? action=fields.NotificationAction.UNSHELVE, | 17:07 |
| mriedem | phase=fields.NotificationPhase.END | 17:07 |
| mriedem | hmm, i guess it does | 17:07 |
| stewie925 | yes mriedem - I added the UNSHELVE under the nova.objects.fields.py | 17:07 |
| openstackgerrit | Chris Dent proposed openstack/nova: Add {get_,set_}aggregates to objects.ResourceProvider https://review.openstack.org/355263 | 17:08 |
| openstackgerrit | Chris Dent proposed openstack/nova: Implement get and set aggregates in the placement API https://review.openstack.org/357993 | 17:08 |
| openstackgerrit | Chris Dent proposed openstack/nova: [placement] add a placement_aggregates table to api_db https://review.openstack.org/362863 | 17:08 |
| *** ducnc has quit IRC | 17:08 | |
| *** baoli_ has quit IRC | 17:08 | |
| cdent | jaypipes: ^ fixed your unique constraint suggestion | 17:08 |
| jaypipes | dansmith: hey, regarding the StringField thing, would you prefer I change those back to ResourceClassField and do the coerce() validation in a future patch that adds the plumbing to create custom resource classes in the API? | 17:08 |
| *** ducnc has joined #openstack-nova | 17:08 | |
| jaypipes | cdent: danke | 17:08 |
| jaypipes | cdent: +2d | 17:09 |
| cdent | gracias | 17:09 |
| dansmith | jaypipes: I dunno. I kinda feel like that's probably the best plan.. but I hadn't thought of it until he pointed it out, so.. | 17:09 |
| mriedem | stewie925: found i | 17:11 |
| *** zenoway has quit IRC | 17:11 | |
| mriedem | *found it | 17:11 |
| mriedem | https://review.openstack.org/#/c/382565/5/nova/tests/unit/compute/test_shelve.py@297 | 17:11 |
| *** zul has quit IRC | 17:11 | |
| stewie925 | mriedem: thank you so much! | 17:11 |
| jaypipes | dansmith: I'll change em back to ResourceClassField and also update Usage's VERSION | 17:12 |
| stewie925 | it would have taken me an eternity to figure it out | 17:12 |
| *** baoli has joined #openstack-nova | 17:12 | |
| *** TravT has joined #openstack-nova | 17:12 | |
| *** hieulq has quit IRC | 17:12 | |
| *** zul has joined #openstack-nova | 17:12 | |
| *** ftersin has quit IRC | 17:13 | |
| dansmith | jaypipes: ack | 17:14 |
| *** thorst_afk is now known as thorst | 17:14 | |
| *** eglynn has quit IRC | 17:15 | |
| *** prateek has joined #openstack-nova | 17:17 | |
| *** baoli has quit IRC | 17:19 | |
| *** prateek has quit IRC | 17:22 | |
| *** prateek has joined #openstack-nova | 17:22 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Correct bug in microversion headers in placement https://review.openstack.org/384806 | 17:23 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 17:23 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add get_all_by_root_provider_uuid() https://review.openstack.org/377215 | 17:23 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 17:23 |
| *** karlamrhein has quit IRC | 17:24 | |
| *** karlamrhein has joined #openstack-nova | 17:27 | |
| cdent | jaypipes: what about over the wire backward compatibility, not database backwards compatibility on nested rps? | 17:27 |
| *** gszasz has quit IRC | 17:28 | |
| cdent | (and all the my comments related to that) | 17:28 |
| openstackgerrit | Karen Bradshaw proposed openstack/nova: API Ref: update server_id params https://review.openstack.org/384664 | 17:29 |
| *** oanson has quit IRC | 17:29 | |
| *** baoli has joined #openstack-nova | 17:32 | |
| *** unicell has joined #openstack-nova | 17:32 | |
| *** hshiina has quit IRC | 17:33 | |
| *** zz_dimtruck is now known as dimtruck | 17:36 | |
| *** baoli has quit IRC | 17:37 | |
| *** lpetrut has quit IRC | 17:37 | |
| *** nherciu has joined #openstack-nova | 17:38 | |
| *** mag009 has quit IRC | 17:39 | |
| jaypipes | cdent: it is wire-backwards-compatible. | 17:50 |
| jaypipes | cdent: the obj_make_compatible() does that, though we currently don't send ResourceProvider objects over the wire anyway. | 17:51 |
| cdent | jaypipes: on the second point: yeah, that's what i was saying: doesn't matter, maybe we can make that official | 17:51 |
| cdent | on the first point I guess I missed the obj_make_compatible changes? | 17:52 |
| * cdent looks closer | 17:52 | |
| cdent | oh, is that built in? | 17:52 |
| cdent | okay, that's the source of my confusion, I thought we had to manage that explicitly. | 17:54 |
| jaypipes | cdent: no, sorry, you are totes correct... I was mistakenly thinking of the obj_make_compat() I added in this patch: https://review.openstack.org/#/c/379656/10/nova/objects/resource_provider.py | 17:54 |
| artom | Actually, maybe someone in here knows - in Neutron, does clearing a port's binding levels result in the port appearing as 'active: false' in the API? Specifically, https://github.com/openstack/neutron/blob/master/neutron/plugins/ml2/db.py#L127 | 17:55 |
| jaypipes | cdent: I will fix up the nested resource providers patch. good catch. | 17:55 |
| cdent | jaypipes: now you're confusing me. | 17:55 |
| * cdent finds link | 17:55 | |
| stewie925 | mriedem: I implemented the change(s) and rerun the tests - this time got a different error - log in http://paste.openstack.org/show/585501/ | 17:55 |
| mriedem | your mocks are in the wrong order | 17:56 |
| jaypipes | mriedem: you mocking me? you mocking me? | 17:56 |
| cdent | jaypipes: I think I'm clear now | 17:56 |
| *** rcernin has quit IRC | 17:56 | |
| mriedem | jaypipes: only in jacksonville | 17:57 |
| *** paul-carlton2 has quit IRC | 17:57 | |
| stewie925 | lol | 17:58 |
| stewie925 | let me switch the mocking order | 17:59 |
| jaypipes | mriedem: :) | 17:59 |
| *** dimtruck is now known as zz_dimtruck | 18:01 | |
| *** zz_dimtruck is now known as dimtruck | 18:01 | |
| stewie925 | mriedem: oh wow, it worked with the order switched - thank you again, guru! | 18:01 |
| *** oanson has joined #openstack-nova | 18:02 | |
| *** rfolco has joined #openstack-nova | 18:03 | |
| *** dikonoo has quit IRC | 18:05 | |
| *** tanvirt has quit IRC | 18:08 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Correct bug in microversion headers in placement https://review.openstack.org/384806 | 18:08 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 18:08 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add get_all_by_root_provider_uuid() https://review.openstack.org/377215 | 18:08 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 18:08 |
| jaypipes | cdent: k, fixed. | 18:08 |
| cdent | rocking, will look soon, looking at the filtering spec at the mo | 18:09 |
| *** dimtruck is now known as zz_dimtruck | 18:11 | |
| *** prateek has quit IRC | 18:13 | |
| *** kaisers_ has joined #openstack-nova | 18:15 | |
| *** lpetrut has joined #openstack-nova | 18:15 | |
| *** ihrachys has quit IRC | 18:15 | |
| openstackgerrit | STEW TY proposed openstack/nova: Transform instance.unshelve notifications https://review.openstack.org/382565 | 18:16 |
| cdent | bauzas: left some comments on the spec review, good start | 18:18 |
| *** armax has joined #openstack-nova | 18:18 | |
| *** paul-carlton2 has joined #openstack-nova | 18:19 | |
| *** prateek has joined #openstack-nova | 18:20 | |
| *** annegentle has joined #openstack-nova | 18:21 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: WIP: discoverable API endpoints https://review.openstack.org/363584 | 18:21 |
| *** rama_y has quit IRC | 18:22 | |
| mriedem | jroll: oh so close https://review.openstack.org/#/c/328157/ | 18:22 |
| *** mvk has quit IRC | 18:28 | |
| *** andymaier has quit IRC | 18:29 | |
| *** READ10 has quit IRC | 18:34 | |
| *** ociuhandu has quit IRC | 18:37 | |
| *** Guest81681 has quit IRC | 18:41 | |
| *** rama_y has joined #openstack-nova | 18:43 | |
| *** stewie925 has quit IRC | 18:45 | |
| *** rfolco has quit IRC | 18:49 | |
| *** raunak has quit IRC | 18:49 | |
| *** paul-carlton2 has quit IRC | 18:50 | |
| *** salv-orlando has quit IRC | 18:51 | |
| *** owalsh has joined #openstack-nova | 18:52 | |
| *** owalsh has quit IRC | 18:52 | |
| *** owalsh has joined #openstack-nova | 18:53 | |
| *** penick has joined #openstack-nova | 18:54 | |
| *** yamahata has quit IRC | 18:54 | |
| *** ociuhandu has joined #openstack-nova | 18:56 | |
| *** raunak has joined #openstack-nova | 18:58 | |
| *** Kevin_Zheng has quit IRC | 18:59 | |
| *** Kevin_Zheng has joined #openstack-nova | 19:02 | |
| *** mmedvede has quit IRC | 19:02 | |
| *** TravT_ has joined #openstack-nova | 19:06 | |
| *** TravT_ has quit IRC | 19:06 | |
| *** TravT has quit IRC | 19:06 | |
| *** TravT has joined #openstack-nova | 19:06 | |
| *** baoli_ has joined #openstack-nova | 19:07 | |
| *** penick has quit IRC | 19:11 | |
| *** kaisers_ has quit IRC | 19:11 | |
| *** kaisers_ has joined #openstack-nova | 19:12 | |
| *** penick has joined #openstack-nova | 19:16 | |
| *** rfolco has joined #openstack-nova | 19:18 | |
| *** oanson has quit IRC | 19:19 | |
| cdent | bauzas, jaypipes: some wild hair rambling about the api side of winnowing compute nodes, for sake of brainstorming and discussion: https://etherpad.openstack.org/p/placement-request-providers | 19:21 |
| bauzas | cdent: I was just writing the spec... :) | 19:22 |
| bauzas | cdent: your point about GET vs. POST doesn't seem really needed | 19:22 |
| bauzas | cdent: honestly, the QS lenght should be fine | 19:23 |
| bauzas | length even | 19:23 |
| cdent | It's not about QS lenght | 19:23 |
| cdent | it's about comprehensibility | 19:23 |
| cdent | and what it expresses | 19:23 |
| cdent | if we can pull of the correct expression in a query string, that's great | 19:23 |
| bauzas | well, python requests module takes a dict as the QS | 19:23 |
| bauzas | so I think it's not really a problem | 19:24 |
| bauzas | about comprehensibility | 19:24 |
| bauzas | rather than just a POST body | 19:24 |
| cdent | it is if we want to document the api | 19:24 |
| cdent | and we ever want people to use the api from places other than code we right | 19:24 |
| cdent | oops | 19:24 |
| cdent | write | 19:24 |
| cdent | an API isn't just a tool for code to interact with | 19:24 |
| cdent | it's for humans to dig at | 19:24 |
| cdent | if they want to | 19:25 |
| bauzas | yeah, but honeslty, I think it's fine | 19:25 |
| bauzas | we can still see how to accept a payload for GET | 19:25 |
| bauzas | which is acceptable | 19:25 |
| bauzas | if we still support a QS | 19:25 |
| cdent | what's more important, I think, is being clear about what we want the query string to express and keeping it constrained and sensible | 19:25 |
| bauzas | I agree | 19:26 |
| mriedem | johnthetubaguy: alex_xu: i've got some concerns about this one https://review.openstack.org/#/c/384261/1 | 19:27 |
| *** mvk has joined #openstack-nova | 19:28 | |
| bauzas | cdent: honestly, if you look at http://developer.openstack.org/api-ref-compute-v2.1.html#createServer then I think it's not a problem to have a long list of optional or not parameters | 19:28 |
| cdent | uh, bauzas, that's completely incomprehensible | 19:28 |
| mriedem | gah use http://developer.openstack.org/api-ref/compute/ | 19:29 |
| mriedem | http://developer.openstack.org/api-ref/compute/?expanded=create-server-detail#id7 | 19:29 |
| mriedem | the horrific beauty of extensions | 19:29 |
| bauzas | yeah what mriedem said, my bad and apologies to sdague and others :) | 19:29 |
| cdent | the docs look great | 19:29 |
| *** rfolco has quit IRC | 19:30 | |
| cdent | but I can't get a whole picture of anything without scrolling multiple folds | 19:30 |
| bauzas | anyway, my point is, if we say /resource_providers/request as an action | 19:30 |
| bauzas | and we're accepting a list of resource requests, I think we're fine | 19:31 |
| cdent | no | 19:31 |
| cdent | we cannot have action endpoints in the placement API | 19:32 |
| bauzas | cdent: one question tho, do we have an API for getting the list of resource classes we support yet ? or is it related to custom resource classes ? | 19:32 |
| bauzas | cdent: why? | 19:32 |
| mriedem | bauzas: we don't yet | 19:32 |
| mriedem | that's part of dynamic resource classes | 19:32 |
| bauzas | mriedem: s/dynamic/custom but yeah | 19:32 |
| bauzas | I remember | 19:32 |
| cdent | because a) that's horrible URI design, b) violates planned agi-wg guidelines, c) is bad resource design | 19:32 |
| bauzas | I was confused because I reviewed the series | 19:32 |
| mriedem | jpenick's super special resource classes | 19:33 |
| mriedem | is the name of the API | 19:33 |
| cdent | and yeah, you can't list resource classes, yet, but soon | 19:33 |
| bauzas | cdent: so we would get another REST resource ? | 19:33 |
| cdent | no, have you read the etherpad I just posted? | 19:34 |
| bauzas | I missed that point | 19:34 |
| bauzas | re-reading | 19:34 |
| cdent | since we have a resource which means list of resource providers | 19:34 |
| cdent | at /resource_providers | 19:34 |
| cdent | then we would want to use that resource with an effective filter | 19:34 |
| bauzas | I see | 19:35 |
| bauzas | FWIW fine by me | 19:35 |
| *** yamahata has joined #openstack-nova | 19:35 | |
| bauzas | cdent: the real problem is that we need some way for the user to discover which parameters he can pass in | 19:36 |
| *** ducnc has quit IRC | 19:36 | |
| bauzas | cdent: which should be done by the custom resource classes patches | 19:36 |
| *** bvanhav has quit IRC | 19:36 | |
| bauzas | cdent: but then, IMHO I think it's fine to leave the user providing the list of resource classes he wants the RP to be filtered on | 19:37 |
| *** pradk- has joined #openstack-nova | 19:37 | |
| bauzas | all those resource classes are discrete numbers, so I think we're fine | 19:37 |
| cdent | those numbers are never exposed anywhere | 19:37 |
| *** cfriesen_ has quit IRC | 19:38 | |
| cdent | so I'm not sure what numbers has to do with anything? | 19:39 |
| *** ducnc has joined #openstack-nova | 19:39 | |
| *** penick has quit IRC | 19:39 | |
| *** pradk has quit IRC | 19:40 | |
| openstackgerrit | Jin Li proposed openstack/nova: Transform instance.soft_delete notifications https://review.openstack.org/383970 | 19:42 |
| bauzas | cdent: sorry, missed your ping | 19:43 |
| bauzas | cdent: if you want to build a request, you wanna know the resource classes | 19:43 |
| bauzas | cdent: so the first point would be to look at the classes, and then write the request for asking a slice of each | 19:44 |
| cdent | sure, but you'll use resource class names for that | 19:45 |
| cdent | edleafe: I responded to your comment on the spec, I think we're in vigorous agreement and there's been a passing of ships in the night or whatever | 19:45 |
| *** salv-orlando has joined #openstack-nova | 19:47 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Changed the name of the standard resource classes https://review.openstack.org/382669 | 19:48 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: change resource class to a StringField https://review.openstack.org/379656 | 19:48 |
| jaypipes | dansmith: k, corrected the StringField back to ResourceClassField ^^ | 19:48 |
| *** penick has joined #openstack-nova | 19:48 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Correct bug in microversion headers in placement https://review.openstack.org/384806 | 19:49 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 19:49 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add get_all_by_root_provider_uuid() https://review.openstack.org/377215 | 19:49 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 19:49 |
| *** bswartz has joined #openstack-nova | 19:49 | |
| *** yamahata has quit IRC | 19:51 | |
| *** lpetrut has quit IRC | 19:53 | |
| bauzas | cdent: the real problem I see with your approach is that if we accept a POST, it could be confusing given we're not persisting something | 19:53 |
| bauzas | while we really want to *get* a list of RPs :) | 19:54 |
| bauzas | conceptually, the HTTP method is getting far more importance to me than the payload | 19:54 |
| cdent | bauzas: right, that's why I say in that document (and on the comment on the spec) that I've changed my mind: I think GET is the way to go | 19:54 |
| cdent | in fact I think so much that GET is the way to go that we should use it even it means that the api is not as expressive as some think it should or could be | 19:55 |
| bauzas | my point indeed | 19:55 |
| mriedem | you know what's not fun? creating a port and floating IP in neutron and associating that to an instance via neutron apis | 19:56 |
| *** lpetrut has joined #openstack-nova | 19:56 | |
| bauzas | mriedem: what's fun is a cat over a roomba | 19:56 |
| mriedem | that's definitely easier to understand | 19:56 |
| bauzas | well, that depends on the model of the cat | 19:57 |
| mriedem | ffs, well i guess i need to dust off the manual | 19:57 |
| mriedem | but the complexity scares me to replace https://review.openstack.org/#/c/384261/ with a neutron driven workflow | 19:58 |
| bauzas | mriedem: btw. are we done with pre-allocating the port in the API ? | 19:58 |
| bauzas | I missed progress on that one | 19:58 |
| mriedem | pre-allocating the port in the api? | 19:58 |
| mriedem | in conductor? | 19:58 |
| bauzas | yup, AFAIR, we discussed about creating the port far earlier | 19:59 |
| bauzas | yeah, sorry the conductor | 19:59 |
| *** gjayavelu has quit IRC | 19:59 | |
| edleafe | cdent: bauzas: the problem I see is that if there is any caching going on, GET bodies are ignored. So if we use GET, then QS is way better than a body | 19:59 |
| mriedem | bauzas: i don't think so | 20:00 |
| bauzas | edleafe: you can't really say that | 20:00 |
| cdent | edleafe: yes, we've resolved that QS has to happen if GET, I think | 20:00 |
| dansmith | edleafe: how would it make sense to cache the GET to the placement api? | 20:00 |
| edleafe | bauzas: it's in the HTTP spec | 20:00 |
| cdent | dansmith: caching proxies | 20:00 |
| dansmith | edleafe: like, if you do that you're screwed anyway right? | 20:00 |
| bauzas | either way, I don't care :) | 20:00 |
| dansmith | cdent: yeah, I know what he means, but... that makes no sense right? | 20:01 |
| bauzas | a QS or a payload is the same to me | 20:01 |
| edleafe | dansmith: I'm not saying it makes sense; I'm saying that if it's there, it would be a problem when using a body | 20:01 |
| bauzas | that's just how we're passing it | 20:01 |
| cdent | of course not, but we don't build cache busting into our apis, and who knows what will get in the middle | 20:01 |
| bauzas | but the format is the same | 20:01 |
| dansmith | edleafe: if you put a cache in front of placement, you're completely screwed anyway | 20:01 |
| edleafe | dansmith: and not with a QS | 20:01 |
| dansmith | edleafe: in fact, them not persisting the body would probably *help* | 20:01 |
| bauzas | either way, I'm just creating that spec | 20:02 |
| edleafe | dansmith: I'm wearing my API WG hat here, not my placement API hat | 20:02 |
| bauzas | folks, feel free to comment it once it's up | 20:02 |
| dansmith | edleafe: yeah, but I mean.. that makes no sense.. it's like putting a cache in front of a random number generator, or caching the result of a locking function | 20:02 |
| bauzas | we're needing yet another BP btw. :) | 20:02 |
| cdent | dansmith: thou shalt not produce HTTP which suffers in the face of caches or proxies in the middle lest the be a sinner | 20:02 |
| dansmith | cdent: if that's the case, you better not use GET at all | 20:03 |
| cdent | for none can know if there be things in the middle | 20:03 |
| edleafe | dansmith: I agree it would be stupid for placement, but that may not be true in all GETs OpenStack-wide | 20:03 |
| cdent | dansmith: no, actually, we should just use cache busting headers by default | 20:03 |
| cdent | and etags | 20:03 |
| cdent | and all the normal webby stuff that people out there in the rest of the world do :) | 20:03 |
| cdent | but there's only so much time in the day, sadly | 20:04 |
| dansmith | cdent: are you saying defeating the cache with etags, but assuming that body won't be even *passed* for GETs through a proxy? | 20:04 |
| cdent | i'm speaking generally about cache handling being an issue for any GET request if there happens to be a caching proxy | 20:05 |
| cdent | ed is more focussed on the fact that if a GET request has a body you can't rely on the proxy to pass that body | 20:05 |
| dansmith | and I'm assuming that we're going to actually discuss the real problem and not generalities | 20:05 |
| penick | I suppose one could argue there’d be value in caching the output of /servers in the event one’s tenant has an awful lot of servers, but.. Egh. | 20:05 |
| dansmith | penick: we're talking specifically about placement here | 20:06 |
| cdent | dansmith: there is no real problem because we're not going to do GETs with bodies | 20:06 |
| cdent | (in the request) | 20:06 |
| penick | Ah. That would be crazypants. | 20:06 |
| cdent | so we're just yanking it at the moment | 20:06 |
| dansmith | cdent: well, I was getting to the point of saying I think we're going to really hate that long term | 20:06 |
| *** Apoorva has quit IRC | 20:07 | |
| *** davidlenwell_ has joined #openstack-nova | 20:07 | |
| cdent | dansmith: see https://etherpad.openstack.org/p/placement-request-providers if you haven't already, I've tried to write down some of the pros and cons and concerns about the various options | 20:07 |
| cdent | and one of them is "we're going to really hate that long term" | 20:07 |
| cdent | or are we? | 20:07 |
| cdent | and hoping for clarification from people who have thought about it | 20:07 |
| cdent | initially I was convinced that we need to do a POST with a body | 20:07 |
| cdent | to avoid hatin' | 20:08 |
| cdent | but in writing the etherpad and talking around with a few http people, changed my mind a bit | 20:08 |
| dansmith | I think POST makes more sense | 20:08 |
| edleafe | cdent: POST with body sound correct for purists, but GET with qs just feels better | 20:08 |
| dansmith | dealing with multiple parameters of the name name is always confusing | 20:08 |
| bauzas | dansmith: POST for getting a list of RPs seems conceptually wrong to me | 20:09 |
| cdent | edleafe: yes, exactly | 20:09 |
| cdent | edleafe: or rather, yes, GET feels better, and in fact GET is more correct for purists (because it is in fact GET semantics) but POST feels like a nice way to hide a complex request query | 20:09 |
| bauzas | dansmith: besides the fact we're not committing anything | 20:10 |
| dansmith | bauzas: I know, it's a little weird | 20:10 |
| edleafe | cdent: yeah, that | 20:10 |
| cdent | it's perfectly fine for a POST to not change things on the server, POST is the "nothing else fits" method | 20:10 |
| dansmith | bauzas: but imagine when we start wanting to represent a grouped set of "this is my preference, and if not that, then that is my preference" wherethis and that are complex sets of resources, traits, etc | 20:10 |
| *** sean-k-mooney is now known as sean-k-mooneyAFK | 20:10 | |
| mriedem | thanks red hat doc writers https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux_OpenStack_Platform/6/html/Administration_Guide/IPAddressing.html | 20:10 |
| dansmith | we can put all of that in the query string for sure, but it becomes a nightmare | 20:10 |
| cdent | dansmith: yes, and the discussion is around "do we even want to allow that or should we just make it simple from the outset and _constrain_ things" | 20:11 |
| cdent | s/the discussion/a discussion/ | 20:11 |
| dansmith | we can base64 encode some json kinda like kibana does, but that's also super annoying | 20:11 |
| edleafe | cdent: maybe in common use, but POST is definitely not that. https://tools.ietf.org/html/rfc2616#section-9.5 | 20:11 |
| alaski | mriedem: want to hit https://review.openstack.org/#/c/331810/7 again? | 20:11 |
| cdent | edleafe: at least link to the right reference: https://tools.ietf.org/html/rfc7231#section-4.3.3 | 20:11 |
| jaypipes | dansmith: yeah -1 on the elasticsearch ?q=gunk mechanism. | 20:11 |
| dansmith | cdent: well, this is my point about in the future, when we stop assuming only academic examples, we're going to be constrained if we have an overly simple query string, IMHO | 20:11 |
| mriedem | alaski: done | 20:12 |
| melwitt | so we're considering POST because gigantic query possibility? I didn't realize GET with a body isn't something reliable | 20:12 |
| cdent | dansmith: yes, but there's an argument to be made that being constrained is a good thing. lack of constraint has gotten us into messes all over the plae | 20:12 |
| cdent | (edleafe not the first bullet point) | 20:13 |
| cdent | note! | 20:13 |
| jaypipes | melwitt: my preference would be GET with a body, frankly. Makes things cleaner and gets around querystring limitations. | 20:13 |
| edleafe | cdent: my link was in response to" POST is the "nothing else fits" method | 20:13 |
| bauzas | dansmith: I agree, that could become messy | 20:13 |
| dansmith | jaypipes: I'd like that too, standards be damned | 20:13 |
| dansmith | query strings suck | 20:13 |
| cdent | edleafe: right and I was just reacting to say that that RFC is out of date | 20:13 |
| dansmith | jaypipes: and if we're going to not do get with a body, then post | 20:13 |
| bauzas | dansmith: jaypipes: let's be pragmatic then | 20:13 |
| bauzas | ooooor | 20:14 |
| melwitt | jaypipes: +1. I didn't know it's against a standard | 20:14 |
| bauzas | say that we only accept one parameter that would serialize the request :p | 20:14 |
| dansmith | bauzas: that's what I said above, like kibana | 20:14 |
| edleafe | melwitt: it's reliable if every piece of the puzzle behaves that way | 20:14 |
| cdent | jaypipes: which querystring limitations do you mean? | 20:14 |
| jaypipes | melwitt: not entirely sure that it is against a standard. more that the standard doesn't prescribe an answer? | 20:14 |
| cdent | length is no longer true | 20:14 |
| melwitt | edleafe: I see. so things aren't guaranteed to forward body with GET | 20:14 |
| bauzas | dansmith: I missed that, my brain is occupied writing the spec | 20:14 |
| *** gouthamr has quit IRC | 20:14 | |
| *** lpetrut has quit IRC | 20:14 | |
| bauzas | dansmith: but I actually like that | 20:14 |
| jaypipes | cdent: length is dependent on the containing web server typically, yes? | 20:15 |
| bauzas | dansmith: we could accept both | 20:15 |
| edleafe | melwitt: zactly | 20:15 |
| bauzas | jaypipes: AFAIK, yes | 20:15 |
| dansmith | edleafe: you could also argue that we be pragmatic, do the thing that makes the most sense here, and when people deploy this at scale enough to hit these problems, we work out a solution | 20:15 |
| bauzas | dansmith: we could accept a list of resource classes as a QS, but we could also accept a specific complex parameter like "request" that would serialize a more complex object | 20:15 |
| cdent | s/no longer true/no longer true in any practical sense/ | 20:15 |
| cdent | please not ambiguity | 20:16 |
| bauzas | that would /take a serialized object/ | 20:16 |
| cdent | one way | 20:16 |
| dansmith | bauzas: or we just be pragmatic until our pragmatism becomes a problem and look at the best solution when we see what actual problems there are | 20:16 |
| edleafe | dansmith: so, again with my API WG hat, we recommend one thing because it's consistent with standards, and then watch groups like Nova do their own thing? | 20:16 |
| *** gjayavelu has joined #openstack-nova | 20:16 | |
| dansmith | edleafe: do whatever you want with your hat | 20:17 |
| edleafe | dansmith: that's what we've been trying to avoid | 20:17 |
| dansmith | so, jaypipes, bauzas: my vote is GET with body and/or POST with body, deal with the fallout if there is fallout later | 20:17 |
| dansmith | happy to articulate that on a spec, | 20:17 |
| dansmith | but I don't really want to rathole on this forever | 20:17 |
| bauzas | :) | 20:18 |
| melwitt | looks like dropbox ran into the same thing and used POST https://blogs.dropbox.com/developers/2015/03/limitations-of-the-get-method-in-http/ | 20:18 |
| jaypipes | dansmith: precisely my feelings. | 20:18 |
| *** cfriesen_ has joined #openstack-nova | 20:18 | |
| bauzas | okay, I promise I'll be short | 20:18 |
| cdent | melwitt: post was the original plan | 20:18 |
| *** davidlenwell_ has quit IRC | 20:18 | |
| dansmith | melwitt: dropbox is of course a little more constrained than us, because they have to be visible from anywhere | 20:18 |
| bauzas | then it would necessarly be another resource | 20:18 |
| dansmith | melwitt: this only affects interaction between services that use placement and placement | 20:19 |
| bauzas | like /resource_providers_request | 20:19 |
| *** yamahata has joined #openstack-nova | 20:19 | |
| bauzas | and we would POST /request | 20:19 |
| cdent | bauzas: why? | 20:19 |
| jaypipes | dansmith: and they have to be usable via a web browser, which we don't. | 20:19 |
| dansmith | jaypipes: right | 20:19 |
| bauzas | because I hate saying we would POST /resource_providers for getting those | 20:19 |
| melwitt | dansmith: yeah. just googling around to see what other options might exist. I found a thing called REPORT that some random blog says might work | 20:19 |
| dansmith | jaypipes: from any remote network to their network, and without control over all the clients | 20:19 |
| cdent | jaypipes, dansmith It would be nice if we wrote out apis so they could be used from anything because that's what http apis are _for_ | 20:19 |
| bauzas | I would POST /request or GET /rps | 20:20 |
| dansmith | melwitt: yup. just sayin'. | 20:20 |
| dansmith | cdent: I don't agree that dropbox's public API has the same requirements as placement. Sorry, I just don't :) | 20:20 |
| cdent | bauzas: POST /resource_providers?some_query_flag \n\nserialize object in body | 20:20 |
| bauzas | oh gosh no | 20:20 |
| cdent | dansmith: I see that, I'm merely expressing an opinion | 20:20 |
| cdent | bauzas: we have a resource that lists resource providers | 20:20 |
| cdent | we should use it | 20:20 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Remove nova/openstack/* from .coveragerc https://review.openstack.org/385611 | 20:21 |
| bauzas | cdent: I think I don't disagree unless I misunderstood | 20:21 |
| bauzas | heh, I'm political \o/ | 20:21 |
| *** cfriesen__ has joined #openstack-nova | 20:22 | |
| *** cfriesen_ has quit IRC | 20:22 | |
| bauzas | cdent: either way, I'll go with the simpliest and leave the gods of standardization just put a trick on me | 20:22 |
| melwitt | bauzas: agree on if we choose POST, the URL should make it obvious it's a read-only resource somehow, like you said /requests or /queries | 20:23 |
| dansmith | also agree | 20:23 |
| cdent | I think that was the plan all along, I hadn't realized linking to that thing would cause such a stir, I had hoped people would just read it and comment | 20:23 |
| cdent | that ^ was in response to bauzas | 20:23 |
| *** flwang has joined #openstack-nova | 20:23 | |
| jaypipes | cdent: I would be fine using GET /r-p with a request body and a Vary: * header to tell caching proxies not to bother. | 20:25 |
| *** nwbi has joined #openstack-nova | 20:27 | |
| *** cdelatte has quit IRC | 20:27 | |
| nwbi | hi | 20:27 |
| nwbi | is there any way to cleanup stale tap ports on nova compute ? | 20:27 |
| cdent | jaypipes: the caching proxy bit was a red herring raised by me and my desired FIX ALL THE HTTP, which is out of scope and I should leave it there. The bigger deal is what edleafe was saying about reverse proxies not necessarily passing the body, and I'm not sure if any amount of headering will deal with that. edleafe you have more info? | 20:28 |
| jaypipes | nwbi: please see /topic. For usage questions, best to ask on the openstack@ mailing list or the #openstack channel. | 20:29 |
| *** annegentle has quit IRC | 20:29 | |
| nwbi | jaypipes, thnx for info | 20:29 |
| jaypipes | cdent: yeah, that makes sense re: the reverse proxying. if a SSH terminating proxy doesn't forward the GET request body that would be bad, mkay. | 20:30 |
| jaypipes | cdent: but then again, do we know of any reverse proxies that do that? | 20:30 |
| *** annegentle has joined #openstack-nova | 20:30 | |
| *** danpawlik has joined #openstack-nova | 20:30 | |
| nwbi | jaypipes, i thought this issue has something to do with nova instance deletion , thats why i asked | 20:30 |
| danpawlik | Hello, if you can pls review: https://review.openstack.org/#/c/314928/ | 20:31 |
| jaypipes | cdent: I'd prfer to just do what dansmith says and just do the GET with body and if there is any fallout, consider moving it to POST. | 20:31 |
| edleafe | cdent: no, except for what googling revealed. The gist was that since GET doesn't define semantics for a body, proxies are not expected to forward it | 20:31 |
| jaypipes | nwbi: it's cool :) | 20:31 |
| dansmith | jaypipes: or if there is fallout do anything else.. when there is fallout, and when we know where the hangups are | 20:31 |
| *** mtreinish has quit IRC | 20:31 | |
| edleafe | jaypipes: I've added this to the API WG agenda for tomorrow. | 20:31 |
| jaypipes | dansmith: ya | 20:31 |
| *** owalsh has quit IRC | 20:31 | |
| *** mtreinish has joined #openstack-nova | 20:32 | |
| *** nwbi has quit IRC | 20:34 | |
| *** davidlenwell_ has joined #openstack-nova | 20:34 | |
| *** owalsh has joined #openstack-nova | 20:34 | |
| *** agireud has joined #openstack-nova | 20:35 | |
| *** suresh12 has joined #openstack-nova | 20:37 | |
| *** owalsh has quit IRC | 20:38 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Add {get_,set_}aggregates to objects.ResourceProvider https://review.openstack.org/355263 | 20:38 |
| openstackgerrit | Chris Dent proposed openstack/nova: Implement get and set aggregates in the placement API https://review.openstack.org/357993 | 20:38 |
| openstackgerrit | Chris Dent proposed openstack/nova: [placement] add a placement_aggregates table to api_db https://review.openstack.org/362863 | 20:38 |
| cdent | jaypipes, edleafe: that ^ is fixing the functional test failures | 20:39 |
| *** gouthamr has joined #openstack-nova | 20:39 | |
| *** nherciu has quit IRC | 20:42 | |
| *** salv-orl_ has joined #openstack-nova | 20:43 | |
| *** salv-orlando has quit IRC | 20:46 | |
| *** slaweq_ has joined #openstack-nova | 20:46 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove redundant req setting https://review.openstack.org/385212 | 20:47 |
| *** baoli_ has quit IRC | 20:48 | |
| mriedem | sdague: https://review.openstack.org/#/c/385484/ for your viewing pleasure | 20:48 |
| *** Apoorva has joined #openstack-nova | 20:48 | |
| *** baoli has joined #openstack-nova | 20:49 | |
| sdague | mriedem: looking | 20:49 |
| mriedem | zvm v3 | 20:49 |
| mriedem | zVMv3: The Reckoning | 20:50 |
| mriedem | zVMv3: The Revenge of Z | 20:50 |
| sdague | heh | 20:50 |
| sdague | I think I need to go crack that bottle of bourbon right now | 20:50 |
| *** gjayavelu has quit IRC | 20:50 | |
| cdent | sdague: do you have a preferred bourbon? | 20:52 |
| sdague | cdent: honestly, not really. But I picked up a bottle from a new local distillery the other day - http://www.denningspointdistillery.com/ | 20:53 |
| sdague | and it's been quite nice | 20:53 |
| cdent | nice, the idea of artisnal spirits in has made me quite happy. I'm not a huge fan of beer so getting some of the same localness in booze is cool | 20:54 |
| * cdent heads to the liquor cabinet | 20:55 | |
| cdent | good night | 20:55 |
| *** cdent has quit IRC | 20:55 | |
| *** penick has quit IRC | 20:56 | |
| mriedem | can we include stats gathering in the 'no more metrics gathering' bucket? https://review.openstack.org/#/c/342001/4/specs/newton/approved/nova-api-sriov-pcipt-stats.rst | 20:58 |
| mriedem | stats are metrics right? | 20:58 |
| *** penick has joined #openstack-nova | 21:01 | |
| *** cleong has quit IRC | 21:02 | |
| *** thorst has quit IRC | 21:05 | |
| *** mtreinish has quit IRC | 21:05 | |
| *** salv-orl_ has quit IRC | 21:07 | |
| *** anush_ has quit IRC | 21:08 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova-specs: resource-providers: Scheduler filters in DB https://review.openstack.org/300178 | 21:08 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova-specs: Narrow the list of ResourceProviders https://review.openstack.org/385618 | 21:08 |
| *** burgerk_ has joined #openstack-nova | 21:08 | |
| *** esberglu has joined #openstack-nova | 21:09 | |
| *** thorst has joined #openstack-nova | 21:11 | |
| auggy | mriedem I always assumed they were synonymous | 21:11 |
| bauzas | cdent: jaypipes: mriedem: dansmith: edleafe: https://review.openstack.org/#/c/385618/ for your pleasure | 21:12 |
| edleafe | bauzas: added | 21:12 |
| *** mtreinish has joined #openstack-nova | 21:12 | |
| bauzas | cdent: jaypipes: mriedem: dansmith: edleafe: I'll write a DNM by tomorrow morning | 21:12 |
| bauzas | (fortunately) | 21:12 |
| *** Lehtu has joined #openstack-nova | 21:13 | |
| *** campee has joined #openstack-nova | 21:13 | |
| *** gerhardq1x has joined #openstack-nova | 21:13 | |
| *** salv-orlando has joined #openstack-nova | 21:13 | |
| *** gus_ has joined #openstack-nova | 21:14 | |
| *** thorst has quit IRC | 21:16 | |
| *** alex_xu_ has joined #openstack-nova | 21:16 | |
| *** zigo_ has joined #openstack-nova | 21:17 | |
| *** odyssey4me_ has joined #openstack-nova | 21:17 | |
| *** mgagne_ has joined #openstack-nova | 21:17 | |
| *** palexster1 has joined #openstack-nova | 21:17 | |
| *** burgerk__ has joined #openstack-nova | 21:18 | |
| *** McNinja_ has joined #openstack-nova | 21:18 | |
| *** haplo37- has joined #openstack-nova | 21:18 | |
| *** odyssey4me has quit IRC | 21:18 | |
| *** odyssey4me_ is now known as odyssey4me | 21:18 | |
| *** hieulq__ has joined #openstack-nova | 21:18 | |
| *** afazekas_ has joined #openstack-nova | 21:19 | |
| *** bswartz has quit IRC | 21:19 | |
| *** mordred has quit IRC | 21:19 | |
| *** afazekas has quit IRC | 21:19 | |
| *** ducttape_ has quit IRC | 21:19 | |
| *** krtaylor has quit IRC | 21:19 | |
| *** burgerk has quit IRC | 21:19 | |
| *** haplo37 has quit IRC | 21:19 | |
| *** jheroux has quit IRC | 21:19 | |
| *** bhagyashri_s has quit IRC | 21:19 | |
| *** McNinja has quit IRC | 21:19 | |
| *** wasmum has quit IRC | 21:19 | |
| *** campee_ has quit IRC | 21:19 | |
| *** alex_xu has quit IRC | 21:19 | |
| *** zigo has quit IRC | 21:19 | |
| *** mgagne has quit IRC | 21:19 | |
| *** thingee has quit IRC | 21:19 | |
| *** anush has quit IRC | 21:19 | |
| *** aunnam has quit IRC | 21:19 | |
| *** raj_singh has quit IRC | 21:19 | |
| *** gerhardqux has quit IRC | 21:19 | |
| *** Lehtu_ has quit IRC | 21:19 | |
| *** ababich has quit IRC | 21:19 | |
| *** knikolla has quit IRC | 21:19 | |
| *** gus has quit IRC | 21:19 | |
| *** haplo37- is now known as haplo37 | 21:19 | |
| *** McNinja_ is now known as McNinja | 21:19 | |
| *** knikolla has joined #openstack-nova | 21:19 | |
| *** raj_singh has joined #openstack-nova | 21:19 | |
| *** rmcall has quit IRC | 21:20 | |
| *** palexster has quit IRC | 21:20 | |
| *** hieulq_ has quit IRC | 21:20 | |
| *** burgerk_ has quit IRC | 21:20 | |
| *** tjones- has left #openstack-nova | 21:21 | |
| *** anush_ has joined #openstack-nova | 21:21 | |
| *** mordred has joined #openstack-nova | 21:21 | |
| *** annegentle has quit IRC | 21:21 | |
| *** markmcclain has quit IRC | 21:21 | |
| *** annegentle has joined #openstack-nova | 21:22 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove redundant req setting https://review.openstack.org/385212 | 21:23 |
| *** _ducttape_ has joined #openstack-nova | 21:23 | |
| *** _ducttape_ has quit IRC | 21:23 | |
| *** pkholkin has quit IRC | 21:23 | |
| *** ujjain has quit IRC | 21:23 | |
| *** StevenK has quit IRC | 21:23 | |
| *** clif_h has quit IRC | 21:23 | |
| *** frickler has quit IRC | 21:23 | |
| *** adreznec has quit IRC | 21:23 | |
| *** kamtamtun has quit IRC | 21:23 | |
| *** tomhambleton has quit IRC | 21:23 | |
| *** macsz has quit IRC | 21:23 | |
| *** ujjain has joined #openstack-nova | 21:23 | |
| *** ujjain has quit IRC | 21:23 | |
| *** ujjain has joined #openstack-nova | 21:23 | |
| *** ababich has joined #openstack-nova | 21:23 | |
| *** anush_ has quit IRC | 21:24 | |
| *** kevinz has quit IRC | 21:24 | |
| *** phschwartz has quit IRC | 21:24 | |
| *** xyang1 has quit IRC | 21:24 | |
| *** dave-mccowan has quit IRC | 21:24 | |
| *** artom has quit IRC | 21:24 | |
| *** krtaylor has joined #openstack-nova | 21:25 | |
| *** bhagyashri_s has joined #openstack-nova | 21:25 | |
| *** thingee has joined #openstack-nova | 21:25 | |
| jaypipes | looks like freenode ate Taco Bell today. | 21:25 |
| mriedem | dorito fingers? | 21:26 |
| jaypipes | yeah. | 21:26 |
| *** phschwartz has joined #openstack-nova | 21:26 | |
| mriedem | or dick fingers? | 21:26 |
| jaypipes | heh | 21:26 |
| *** markmcclain has joined #openstack-nova | 21:26 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Changed the name of the standard resource classes https://review.openstack.org/382669 | 21:27 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: change resource class to a StringField https://review.openstack.org/379656 | 21:27 |
| *** _ducttape_ has joined #openstack-nova | 21:28 | |
| *** pkholkin has joined #openstack-nova | 21:28 | |
| *** clif_h has joined #openstack-nova | 21:28 | |
| *** StevenK has joined #openstack-nova | 21:28 | |
| *** frickler has joined #openstack-nova | 21:28 | |
| *** adreznec has joined #openstack-nova | 21:28 | |
| *** kamtamtun has joined #openstack-nova | 21:28 | |
| *** tomhambleton has joined #openstack-nova | 21:28 | |
| *** macsz has joined #openstack-nova | 21:28 | |
| *** macsz has quit IRC | 21:28 | |
| *** wasmum has joined #openstack-nova | 21:28 | |
| *** wasmum has quit IRC | 21:28 | |
| *** wasmum has joined #openstack-nova | 21:28 | |
| *** zigo_ is now known as zigo | 21:28 | |
| *** rtheis has quit IRC | 21:29 | |
| *** davidlenwell_ has quit IRC | 21:31 | |
| *** tblakes_ has joined #openstack-nova | 21:32 | |
| *** macsz has joined #openstack-nova | 21:33 | |
| *** Jeffrey4l__ has joined #openstack-nova | 21:34 | |
| *** tblakes has quit IRC | 21:35 | |
| *** tblakes_ is now known as tblakes | 21:35 | |
| *** Jeffrey4l_ has quit IRC | 21:35 | |
| *** mtreinish has quit IRC | 21:36 | |
| *** zz_dimtruck is now known as dimtruck | 21:37 | |
| *** macsz is now known as macsz_ | 21:37 | |
| *** pradk- is now known as pradk | 21:39 | |
| *** portdirect has joined #openstack-nova | 21:40 | |
| *** edmondsw has quit IRC | 21:41 | |
| *** tblakes has quit IRC | 21:42 | |
| *** mtreinish has joined #openstack-nova | 21:42 | |
| *** davidlenwell_ has joined #openstack-nova | 21:44 | |
| *** thorst has joined #openstack-nova | 21:45 | |
| *** tongli has quit IRC | 21:45 | |
| *** dimtruck is now known as zz_dimtruck | 21:47 | |
| *** gjayavelu has joined #openstack-nova | 21:48 | |
| *** mdrabe has quit IRC | 21:49 | |
| *** thorst has quit IRC | 21:49 | |
| *** mdrabe has joined #openstack-nova | 21:49 | |
| *** burgerk__ has quit IRC | 21:52 | |
| mriedem | easy bug fix for someone https://bugs.launchpad.net/python-novaclient/+bug/1632866 | 21:53 |
| openstack | Launchpad bug 1632866 in python-novaclient "nova server-group-create fails with AttributeError if policy isn't specified" [Undecided,Confirmed] | 21:53 |
| *** thorst has joined #openstack-nova | 21:55 | |
| *** thorst has quit IRC | 22:00 | |
| *** mdrabe has quit IRC | 22:01 | |
| *** burt has quit IRC | 22:01 | |
| *** danpawlik has left #openstack-nova | 22:02 | |
| *** prateek has quit IRC | 22:03 | |
| *** TravT_ has joined #openstack-nova | 22:06 | |
| *** TravT has quit IRC | 22:08 | |
| *** mtreinish has quit IRC | 22:08 | |
| *** Sukhdev has joined #openstack-nova | 22:10 | |
| *** jamesdenton has quit IRC | 22:12 | |
| mriedem | sdague: alex_xu_: weren't we going to deprecate os-certificates in ocata? | 22:13 |
| mriedem | and cloudpipe? | 22:13 |
| *** TravT_ has quit IRC | 22:14 | |
| *** edmondsw has joined #openstack-nova | 22:16 | |
| *** markmcclain has quit IRC | 22:18 | |
| *** mtreinish has joined #openstack-nova | 22:19 | |
| *** markmcclain has joined #openstack-nova | 22:19 | |
| *** unicell has quit IRC | 22:20 | |
| *** lamt has quit IRC | 22:22 | |
| *** slaweq_ has quit IRC | 22:23 | |
| *** bswartz has joined #openstack-nova | 22:24 | |
| openstackgerrit | Pushkar Umaranikar proposed openstack/nova: Add debug logs to output host weights for weighers https://review.openstack.org/385636 | 22:24 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove deprecated command options https://review.openstack.org/385637 | 22:24 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove unused helper volume methods in v2 shell https://review.openstack.org/385638 | 22:24 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Update docs for instructions on deprecating commands https://review.openstack.org/385639 | 22:24 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove deprecated commands https://review.openstack.org/385637 | 22:25 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Update docs for instructions on deprecating commands https://review.openstack.org/385639 | 22:25 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove unused helper volume methods in v2 shell https://review.openstack.org/385638 | 22:25 |
| *** mgoddard has quit IRC | 22:25 | |
| *** smatzek has joined #openstack-nova | 22:27 | |
| *** salv-orlando has quit IRC | 22:30 | |
| *** esberglu has quit IRC | 22:31 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove redundant req setting https://review.openstack.org/385212 | 22:31 |
| *** penick has quit IRC | 22:31 | |
| *** annegentle has quit IRC | 22:31 | |
| mriedem | you know, i don't think i've heard 'radar love' in about 15 years | 22:32 |
| *** ducnc has quit IRC | 22:32 | |
| *** ducnc has joined #openstack-nova | 22:32 | |
| *** esberglu has joined #openstack-nova | 22:35 | |
| *** esberglu has quit IRC | 22:35 | |
| jroll | and you put it in my head in two seconds. gw | 22:37 |
| auggy | if the song was "radar level" it would be a full palindrome | 22:38 |
| mriedem | jroll: time to shave your beard into mutton chops if you're going to listen to radar love | 22:39 |
| * jroll picks up a razor and a guitar | 22:40 | |
| *** mtreinish has quit IRC | 22:41 | |
| *** _ducttape_ has quit IRC | 22:42 | |
| *** kevinz has joined #openstack-nova | 22:42 | |
| *** esberglu has joined #openstack-nova | 22:43 | |
| *** sdague has quit IRC | 22:43 | |
| openstackgerrit | STEW TY proposed openstack/nova: Transform instance.rescue notifications https://review.openstack.org/385644 | 22:44 |
| *** esberglu has quit IRC | 22:45 | |
| *** mlavalle has quit IRC | 22:46 | |
| openstackgerrit | STEW TY proposed openstack/nova: Transform instance.rescue notifications (WIP) https://review.openstack.org/385644 | 22:47 |
| *** mtreinish has joined #openstack-nova | 22:47 | |
| *** smatzek has quit IRC | 22:49 | |
| *** anush_ has joined #openstack-nova | 22:49 | |
| *** artom has joined #openstack-nova | 22:49 | |
| *** thingee has quit IRC | 22:52 | |
| *** takashin has joined #openstack-nova | 22:56 | |
| *** thingee has joined #openstack-nova | 22:58 | |
| *** anush_ has quit IRC | 23:00 | |
| *** artom has quit IRC | 23:00 | |
| openstackgerrit | Scott DAngelo proposed openstack/nova: WIP DO NOT MERGE https://review.openstack.org/385682 | 23:00 |
| *** tjones- has joined #openstack-nova | 23:01 | |
| *** gus_ is now known as gus | 23:06 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Remove deprecated use_usb_tablet config option https://review.openstack.org/385683 | 23:06 |
| *** ducttape_ has joined #openstack-nova | 23:07 | |
| *** edmondsw has quit IRC | 23:08 | |
| *** suresh12 has quit IRC | 23:12 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove deprecated commands https://review.openstack.org/385637 | 23:12 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Update docs for instructions on deprecating commands https://review.openstack.org/385639 | 23:12 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Remove unused helper volume methods in v2 shell https://review.openstack.org/385638 | 23:12 |
| *** mriedem is now known as mriedem_afk | 23:13 | |
| *** anush_ has joined #openstack-nova | 23:15 | |
| *** artom has joined #openstack-nova | 23:15 | |
| *** suresh12 has joined #openstack-nova | 23:21 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Corrects the type of a base64 encoded string https://review.openstack.org/385686 | 23:24 |
| edleafe | melwitt: ^^ This takes care of the py3 issue you noted earlier today | 23:25 |
| edleafe | melwitt: there are still py3 issues, but at least the tests run (and fail) now | 23:25 |
| openstackgerrit | Nicolas Simonds proposed openstack/python-novaclient: Raise the correct exception if policy is unspecified https://review.openstack.org/385690 | 23:35 |
| *** rama_y has quit IRC | 23:36 | |
| *** esberglu has joined #openstack-nova | 23:36 | |
| *** zz_dimtruck is now known as dimtruck | 23:38 | |
| *** markvoelker has quit IRC | 23:39 | |
| *** rfolco has joined #openstack-nova | 23:45 | |
| *** dimtruck is now known as zz_dimtruck | 23:47 | |
| *** gjayavelu has quit IRC | 23:55 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Correct bug in microversion headers in placement https://review.openstack.org/384806 | 23:55 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 23:55 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add get_all_by_root_provider_uuid() https://review.openstack.org/377215 | 23:55 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 23:55 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: update client to set parent provider https://review.openstack.org/385693 | 23:55 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Use the ResourceProvider id instead of compute node. https://review.openstack.org/382000 | 23:58 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: handle generic PCI devices in inventory https://review.openstack.org/374681 | 23:58 |
| openstackgerrit | Jay Pipes proposed openstack/nova: placement: allocations for generic PCI devices https://review.openstack.org/375168 | 23:58 |
| *** esberglu has quit IRC | 23:59 | |
| *** ducttape_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
