| *** browne has quit IRC | 00:01 | |
| *** browne has joined #openstack-nova | 00:02 | |
| *** jamielennox|away is now known as jamielennox | 00:02 | |
| *** eglynn-afk has joined #openstack-nova | 00:03 | |
| openstackgerrit | melanie witt proposed openstack/nova: Add a TODO to make ComputeNode.cpu_info non-nullable https://review.openstack.org/271105 | 00:03 |
|---|---|---|
| *** ivase has joined #openstack-nova | 00:04 | |
| *** ijw has joined #openstack-nova | 00:07 | |
| *** ivase has quit IRC | 00:08 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Add some validations for BDM https://review.openstack.org/258788 | 00:09 |
| *** Sukhdev has quit IRC | 00:09 | |
| *** Sukhdev has joined #openstack-nova | 00:10 | |
| *** ljxiash has joined #openstack-nova | 00:11 | |
| *** zz_dimtruck is now known as dimtruck | 00:11 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Pass BDM in swap_volume method of LibvirtDriver https://review.openstack.org/257135 | 00:11 |
| *** Daisy_ has joined #openstack-nova | 00:13 | |
| *** Guest80711 is now known as med_ | 00:15 | |
| *** med_ has quit IRC | 00:15 | |
| *** med_ has joined #openstack-nova | 00:15 | |
| *** Daisy_ has quit IRC | 00:16 | |
| *** eglynn-afk has quit IRC | 00:18 | |
| *** thumpba has quit IRC | 00:20 | |
| *** edtubill has joined #openstack-nova | 00:21 | |
| *** thorst has joined #openstack-nova | 00:24 | |
| *** abitha has joined #openstack-nova | 00:25 | |
| *** diana_clarke has joined #openstack-nova | 00:25 | |
| *** zenoway has joined #openstack-nova | 00:27 | |
| *** penick has quit IRC | 00:31 | |
| *** zenoway has quit IRC | 00:32 | |
| *** thorst has quit IRC | 00:32 | |
| *** jed56 has quit IRC | 00:33 | |
| *** ivase has joined #openstack-nova | 00:33 | |
| *** ivase_ has joined #openstack-nova | 00:36 | |
| *** zengine has quit IRC | 00:37 | |
| *** nic has quit IRC | 00:38 | |
| *** jamielennox is now known as jamielennox|away | 00:38 | |
| *** ivase has quit IRC | 00:38 | |
| *** jamielennox|away is now known as jamielennox | 00:39 | |
| *** thorst has joined #openstack-nova | 00:40 | |
| *** ivase_ has quit IRC | 00:40 | |
| *** thorst has quit IRC | 00:42 | |
| *** ducttape_ has quit IRC | 00:42 | |
| *** zhangjn has joined #openstack-nova | 00:42 | |
| *** lucasagomes has quit IRC | 00:42 | |
| openstackgerrit | Dina Belova proposed openstack/nova: Integrate OSprofiler and Nova https://review.openstack.org/254703 | 00:43 |
| *** lucasagomes has joined #openstack-nova | 00:49 | |
| *** EinstCrazy has joined #openstack-nova | 00:50 | |
| *** edtubill has quit IRC | 00:54 | |
| *** qeelee has joined #openstack-nova | 00:55 | |
| *** EinstCrazy has quit IRC | 00:55 | |
| *** mriedem_machell is now known as mriedem_afk | 00:58 | |
| *** eil397 has quit IRC | 00:58 | |
| *** zenoway has joined #openstack-nova | 01:03 | |
| *** hoangcx has joined #openstack-nova | 01:04 | |
| *** ducttape_ has joined #openstack-nova | 01:06 | |
| *** zenoway has quit IRC | 01:06 | |
| *** haomaiwang has joined #openstack-nova | 01:12 | |
| *** Swami_ has quit IRC | 01:12 | |
| gmann | alex_xu: sdague johnthetubaguy : etherpad for all extensions testing - https://etherpad.openstack.org/p/api-sample-tests-with-all-extensions | 01:14 |
| gmann | linked the same in our newton api etherpad | 01:14 |
| alex_xu | gmann: cool, thanks | 01:14 |
| gmann | alex_xu: found the old one i created, i added yesterday point there | 01:15 |
| alex_xu | gmann: :) | 01:15 |
| *** gongysh has joined #openstack-nova | 01:22 | |
| *** pradk has quit IRC | 01:22 | |
| openstackgerrit | Felix Ma proposed openstack/nova: Catch exceptions in _destroy_evacuated_instances https://review.openstack.org/265681 | 01:23 |
| *** haomaiwang has quit IRC | 01:27 | |
| eliqiao | hi tdurakov_ there? | 01:28 |
| openstackgerrit | Jinxing Fang proposed openstack/nova: Update the home-page https://review.openstack.org/275467 | 01:29 |
| *** lixiaoy1 has joined #openstack-nova | 01:30 | |
| *** abitha has quit IRC | 01:30 | |
| *** jamielennox is now known as jamielennox|away | 01:30 | |
| *** ctrath has quit IRC | 01:30 | |
| *** achanda has joined #openstack-nova | 01:32 | |
| *** shuquan has joined #openstack-nova | 01:32 | |
| *** shuquan has quit IRC | 01:32 | |
| eliqiao | dansmith: hi dan, still there, I have a question about my patch https://review.openstack.org/#/c/254101/16/nova/virt/libvirt/driver.py, tdurakov_ suggest that I need to add a method inside of migrate_data obj to calculate block_migration value but I think that would require to bump migrat_data object version again, can you help to take a look? | 01:32 |
| *** shuquan has joined #openstack-nova | 01:33 | |
| dansmith | eliqiao: only if it's a remotable method, which it wouldn't be, right? | 01:34 |
| *** ljxiash has quit IRC | 01:35 | |
| *** thorst has joined #openstack-nova | 01:36 | |
| eliqiao | dansmith: no, it's not a remotable method. | 01:37 |
| dansmith | eliqiao: then no, no version bump required | 01:37 |
| *** Daisy has joined #openstack-nova | 01:38 | |
| eliqiao | dansmith: Okay, gotcha, thanks :) | 01:39 |
| *** Sukhdev has quit IRC | 01:39 | |
| dansmith | eliqiao: no problem | 01:40 |
| *** dimtruck is now known as zz_dimtruck | 01:40 | |
| *** baoli has joined #openstack-nova | 01:42 | |
| *** baoli has quit IRC | 01:43 | |
| *** thorst has quit IRC | 01:44 | |
| *** thorst has joined #openstack-nova | 01:45 | |
| *** mtanino has quit IRC | 01:45 | |
| notmorgan | dansmith: do you have a moment? i need to understand something so i can solve the comment and we can make novaclient all OCC/KSA enabled. | 01:45 |
| notmorgan | dansmith: i can bug you later this week if that is better | 01:45 |
| notmorgan | dansmith: i am asking you cause we're in the same timezone ;) | 01:46 |
| dansmith | notmorgan: tell me what you want to understand and then I'll decide if now is good :) | 01:46 |
| dansmith | notmorgan: we are not in the same timezone | 01:46 |
| notmorgan | oh travelling? | 01:46 |
| notmorgan | LCAing or such? | 01:46 |
| dansmith | notmorgan: no, I'm being sarcastic, nevermind :) | 01:46 |
| notmorgan | sorry brain is only 1/2 ... functioning | 01:46 |
| notmorgan | no coffee today | 01:46 |
| *** zz_dimtruck is now known as dimtruck | 01:47 | |
| notmorgan | dansmith: so looking at https://review.openstack.org/#/c/245200/2/novaclient/shell.py and the comment re the microversioning (line 587) | 01:47 |
| notmorgan | dansmith: with OCC and clouds.yaml in mind, how should i attack this. | 01:48 |
| *** su_zhang has quit IRC | 01:48 | |
| notmorgan | dansmith: i get what the design of novaclient is meant to be, but i'm struggling with the specifics on how to handle the case of set vs unset vs <whatever> on the microversion | 01:49 |
| notmorgan | dansmith: and like i said, feel free to punt ;) i'll bug you again next week when back in PDX if now is bad. | 01:49 |
| dansmith | I don't know what "OCC and clouds.yaml" means | 01:49 |
| notmorgan | OCC is os-cloud-config | 01:49 |
| notmorgan | clouds.yaml is a way to communicate basic capabilities for a given cloud | 01:50 |
| notmorgan | devstack, public, private cloud, etc | 01:50 |
| notmorgan | with various profiles | 01:50 |
| dansmith | I don't recognize those as things | 01:50 |
| notmorgan | example is neutron uses NAT vs provider network in a given deployment profile | 01:50 |
| dansmith | that aside, I'm trying to parse what is changing in this, in terms of the version impact, give me a sec | 01:51 |
| notmorgan | dansmith: sounds good. | 01:51 |
| notmorgan | i just figure i can get your input and then try and resolve the problem in a way that works right for nova. | 01:51 |
| notmorgan | :) | 01:51 |
| notmorgan | i also apologize for the lack of rebasing - but i wanted to rebase/resolve comment at the same time | 01:53 |
| *** thorst has quit IRC | 01:53 | |
| dansmith | notmorgan: well, I'm really not the right person to ask about the details of this api versioning stuff, especially as it relates to the client, BUT: | 01:54 |
| dansmith | notmorgan: I think the reviewer is complaining that you're defaulting to the latest version, maybe? | 01:54 |
| dansmith | which isn't how clients are supposed to behave, with the exception (I believe) of the shel | 01:54 |
| dansmith | er, shell | 01:54 |
| notmorgan | dansmith: ok ill take another hard look at the comment and try and grok it | 01:54 |
| notmorgan | dansmith: and then poke sdague or someone else too | 01:54 |
| dansmith | the client is supposed to use specific versions it knows about, but I think the shell is supposed to always use the latest | 01:55 |
| notmorgan | dansmith: if i can't figure it out. Thnx. :) | 01:55 |
| notmorgan | right. | 01:55 |
| dansmith | notmorgan: in this timezone, ping oomichi when he's up | 01:55 |
| notmorgan | good idea | 01:55 |
| notmorgan | will do so after my trip to seattle | 01:55 |
| dansmith | or alex_xu | 01:55 |
| notmorgan | :) | 01:55 |
| notmorgan | thnx | 01:55 |
| notmorgan | have a great evening | 01:55 |
| notmorgan | :) | 01:55 |
| dansmith | notmorgan: next time ask me something I know the answer to so I can feel useful, okay? | 01:55 |
| notmorgan | oh oh.. i know i know.. sooooo is nova "cool"? | 01:56 |
| notmorgan | (that one should be easy, right?) | 01:56 |
| * dansmith clears his throat | 01:56 | |
| dansmith | Yes, yes it is. | 01:56 |
| notmorgan | fantastic | 01:56 |
| *** Marga_ has quit IRC | 01:56 | |
| * dansmith can rest easy now | 01:56 | |
| *** ljxiash has joined #openstack-nova | 01:56 | |
| notmorgan | dansmith: also we shall need to food again sometime not-too-far-off | 01:57 |
| * notmorgan goes back to things. | 01:57 | |
| dansmith | notmorgan: I require food multiple times _EACH_DAY_ so that should be doable | 01:57 |
| * dansmith goes back to his B.B. King documentary | 01:57 | |
| *** shuquan has quit IRC | 01:58 | |
| *** mdrabe has joined #openstack-nova | 01:59 | |
| openstackgerrit | Ruby Loo proposed openstack/nova: No need to have ironicclient parameter in methods https://review.openstack.org/273755 | 02:03 |
| openstackgerrit | huan proposed openstack/nova: XenAPI: Do not rely on XenAPI to get device names https://review.openstack.org/213112 | 02:06 |
| *** salv-orl_ has joined #openstack-nova | 02:09 | |
| *** salv-orlando has quit IRC | 02:12 | |
| *** shuquan has joined #openstack-nova | 02:15 | |
| *** shuquan has quit IRC | 02:15 | |
| openstackgerrit | Miguel Lavalle proposed openstack/nova: Use instance hostname for Neutron DNS https://review.openstack.org/271578 | 02:15 |
| *** shuquan has joined #openstack-nova | 02:15 | |
| *** EinstCrazy has joined #openstack-nova | 02:17 | |
| *** Marga_ has joined #openstack-nova | 02:17 | |
| *** browne has quit IRC | 02:18 | |
| *** Marga_ has quit IRC | 02:22 | |
| *** Jeffrey4l has joined #openstack-nova | 02:22 | |
| *** haomaiwang has joined #openstack-nova | 02:23 | |
| *** shuquan has quit IRC | 02:23 | |
| *** amotoki has joined #openstack-nova | 02:24 | |
| *** shuquan has joined #openstack-nova | 02:26 | |
| *** jamielennox|away is now known as jamielennox | 02:31 | |
| *** angdraug has quit IRC | 02:32 | |
| *** Marga_ has joined #openstack-nova | 02:34 | |
| *** mdrabe has quit IRC | 02:38 | |
| *** Marga_ has quit IRC | 02:38 | |
| *** dims_ has quit IRC | 02:40 | |
| *** Apoorva has quit IRC | 02:41 | |
| *** dimtruck is now known as zz_dimtruck | 02:41 | |
| *** Jeffrey4l has quit IRC | 02:42 | |
| *** zz_dimtruck is now known as dimtruck | 02:42 | |
| *** shuquan has quit IRC | 02:45 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Create image for suspended instance booted from volume https://review.openstack.org/223382 | 02:47 |
| *** vilobhmm11 has quit IRC | 02:48 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Only return latest fault of each instance in db api https://review.openstack.org/223392 | 02:48 |
| *** Marga_ has joined #openstack-nova | 02:48 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Calculate block_migration in virt drivers if it's None https://review.openstack.org/254101 | 02:50 |
| openstackgerrit | Eli Qiao proposed openstack/nova: Return host and block_migration in task_api._execute() https://review.openstack.org/254102 | 02:50 |
| *** Marga_ has quit IRC | 02:50 | |
| *** Marga_ has joined #openstack-nova | 02:51 | |
| *** thorst has joined #openstack-nova | 02:52 | |
| openstackgerrit | Wen Zhi Yu proposed openstack/nova: Make sure destination will rollback if live migration fails https://review.openstack.org/270093 | 02:56 |
| *** thorst has quit IRC | 02:58 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-nova | 03:01 | |
| *** rook-laptop has quit IRC | 03:02 | |
| openstackgerrit | Dina Belova proposed openstack/nova: Integrate OSprofiler and Nova https://review.openstack.org/254703 | 03:03 |
| *** shuquan has joined #openstack-nova | 03:04 | |
| *** links has joined #openstack-nova | 03:06 | |
| *** shuquan has quit IRC | 03:08 | |
| *** moshele has joined #openstack-nova | 03:09 | |
| *** browne has joined #openstack-nova | 03:10 | |
| *** Daisy has quit IRC | 03:11 | |
| *** ducttape_ has quit IRC | 03:17 | |
| *** mdrabe has joined #openstack-nova | 03:17 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limmits when booting an instance. https://review.openstack.org/275379 | 03:18 |
| *** ducttape_ has joined #openstack-nova | 03:21 | |
| *** shuquan has joined #openstack-nova | 03:22 | |
| *** ijw has quit IRC | 03:23 | |
| *** ijw has joined #openstack-nova | 03:23 | |
| *** ijw has quit IRC | 03:23 | |
| openstackgerrit | likun proposed openstack/nova: Add correct form 'pci_requests' objects in build instance request on child cell. https://review.openstack.org/275484 | 03:24 |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 03:28 |
| *** rook-laptop has joined #openstack-nova | 03:28 | |
| *** gyee has quit IRC | 03:28 | |
| *** ducttape_ has quit IRC | 03:28 | |
| *** tiendc has joined #openstack-nova | 03:34 | |
| *** doug-fish has quit IRC | 03:34 | |
| *** doug-fish has joined #openstack-nova | 03:35 | |
| *** doug-fish has quit IRC | 03:36 | |
| *** rook-laptop has quit IRC | 03:36 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: libvirt: Add methods to parse of domain metadata from xml https://review.openstack.org/275388 | 03:38 |
| *** gyee has joined #openstack-nova | 03:38 | |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for migration. https://review.openstack.org/241479 | 03:41 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for evacuation. https://review.openstack.org/241478 | 03:41 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for resize. https://review.openstack.org/241477 | 03:41 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for live-migration. https://review.openstack.org/241476 | 03:41 |
| *** shuquan has quit IRC | 03:41 | |
| *** ducttape_ has joined #openstack-nova | 03:41 | |
| *** ducttape_ has quit IRC | 03:44 | |
| *** unicell1 has quit IRC | 03:46 | |
| *** dedery has joined #openstack-nova | 03:47 | |
| *** deva_ has joined #openstack-nova | 03:48 | |
| *** dedery_ has joined #openstack-nova | 03:51 | |
| *** deva_ has quit IRC | 03:53 | |
| *** qeelee has quit IRC | 03:53 | |
| *** dedery has quit IRC | 03:54 | |
| *** thorst has joined #openstack-nova | 03:56 | |
| *** gongysh has quit IRC | 04:00 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-nova | 04:01 | |
| *** ducttape_ has joined #openstack-nova | 04:02 | |
| *** thorst has quit IRC | 04:03 | |
| *** ducttape_ has joined #openstack-nova | 04:03 | |
| *** gyee has quit IRC | 04:07 | |
| *** irenab_ has joined #openstack-nova | 04:09 | |
| *** jamielennox is now known as jamielennox|away | 04:09 | |
| *** irenab has quit IRC | 04:11 | |
| *** irenab_ is now known as irenab | 04:11 | |
| *** su_zhang has joined #openstack-nova | 04:13 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Improve os-migrateLive input and output values https://review.openstack.org/259319 | 04:14 |
| *** su_zhang has quit IRC | 04:15 | |
| *** su_zhang has joined #openstack-nova | 04:16 | |
| *** moshele has quit IRC | 04:17 | |
| *** deepakcs has joined #openstack-nova | 04:20 | |
| *** moshele has joined #openstack-nova | 04:21 | |
| *** mriedem_afk has quit IRC | 04:22 | |
| *** ljxiash has quit IRC | 04:23 | |
| *** ljxiash has joined #openstack-nova | 04:24 | |
| *** su_zhang_ has joined #openstack-nova | 04:26 | |
| *** tiendc has quit IRC | 04:26 | |
| *** su_zhang has quit IRC | 04:27 | |
| *** achanda has quit IRC | 04:28 | |
| *** ljxiash has quit IRC | 04:28 | |
| *** dimtruck is now known as zz_dimtruck | 04:29 | |
| *** shyama has joined #openstack-nova | 04:30 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 04:32 |
| *** qeelee has joined #openstack-nova | 04:33 | |
| openstackgerrit | Ankit Agrawal proposed openstack/nova: Pass bdm info to _get_instance_disk_info method https://review.openstack.org/221162 | 04:36 |
| *** vishwanathj has joined #openstack-nova | 04:37 | |
| *** vishwanathj has quit IRC | 04:37 | |
| *** Daisy has joined #openstack-nova | 04:38 | |
| *** shuquan has joined #openstack-nova | 04:38 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 04:39 |
| *** achanda has joined #openstack-nova | 04:43 | |
| *** Daisy has quit IRC | 04:43 | |
| *** ljxiash has joined #openstack-nova | 04:44 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: libvirt: Add methods to parse of domain metadata from xml https://review.openstack.org/275388 | 04:46 |
| *** ljxiash has quit IRC | 04:49 | |
| *** SridharG has joined #openstack-nova | 04:50 | |
| *** qeelee has quit IRC | 04:50 | |
| *** SridharG has quit IRC | 04:50 | |
| *** ivase has joined #openstack-nova | 04:52 | |
| *** EinstCrazy has quit IRC | 04:53 | |
| *** ljxiash has joined #openstack-nova | 04:53 | |
| *** mdrabe has quit IRC | 04:53 | |
| *** abhishekk has quit IRC | 04:54 | |
| *** ljxiash has quit IRC | 04:54 | |
| *** ljxiash has joined #openstack-nova | 04:55 | |
| *** amotoki has quit IRC | 04:57 | |
| *** ivase has quit IRC | 04:57 | |
| *** ljxiash has quit IRC | 04:59 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** thorst has joined #openstack-nova | 05:01 | |
| *** haomaiwang has joined #openstack-nova | 05:01 | |
| *** vilobhmm11 has joined #openstack-nova | 05:02 | |
| *** ducttape_ has quit IRC | 05:03 | |
| *** wanghao has quit IRC | 05:07 | |
| *** wanghao has joined #openstack-nova | 05:07 | |
| *** thorst has quit IRC | 05:08 | |
| *** abhishekk has joined #openstack-nova | 05:09 | |
| openstackgerrit | Ren Qiaowei proposed openstack/nova: add checking for new image metadata property 'hw_cpu_realtime_mask' https://review.openstack.org/266233 | 05:13 |
| *** abitha has joined #openstack-nova | 05:16 | |
| *** moshele has quit IRC | 05:17 | |
| *** amotoki has joined #openstack-nova | 05:18 | |
| *** wyang has joined #openstack-nova | 05:21 | |
| *** Sukhdev has joined #openstack-nova | 05:23 | |
| *** teju has joined #openstack-nova | 05:26 | |
| *** sudipto_ has joined #openstack-nova | 05:27 | |
| *** Akanksha08 has joined #openstack-nova | 05:29 | |
| *** amotoki_ has joined #openstack-nova | 05:30 | |
| *** amotoki has quit IRC | 05:33 | |
| *** ljxiash has joined #openstack-nova | 05:39 | |
| *** ljxiash has quit IRC | 05:41 | |
| *** ljxiash has joined #openstack-nova | 05:41 | |
| *** takashin has joined #openstack-nova | 05:45 | |
| *** abitha has quit IRC | 05:47 | |
| *** ssurana has quit IRC | 05:47 | |
| *** qeelee has joined #openstack-nova | 05:49 | |
| *** nkrinner has joined #openstack-nova | 05:50 | |
| *** amotoki_ has quit IRC | 05:52 | |
| *** Marga_ has quit IRC | 05:58 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:01 | |
| *** Jeffrey4l has joined #openstack-nova | 06:02 | |
| *** qeelee has quit IRC | 06:04 | |
| *** shuquan has quit IRC | 06:05 | |
| *** amotoki has joined #openstack-nova | 06:05 | |
| *** thorst has joined #openstack-nova | 06:07 | |
| *** cfriesen has quit IRC | 06:11 | |
| *** armax_ has joined #openstack-nova | 06:11 | |
| *** unicell has joined #openstack-nova | 06:11 | |
| *** armax has quit IRC | 06:12 | |
| *** armax_ is now known as armax | 06:12 | |
| *** thorst has quit IRC | 06:13 | |
| *** penick has joined #openstack-nova | 06:15 | |
| *** amotoki has quit IRC | 06:18 | |
| *** mubirru has joined #openstack-nova | 06:18 | |
| *** SridharG has joined #openstack-nova | 06:23 | |
| *** amotoki has joined #openstack-nova | 06:26 | |
| *** jaypipes has quit IRC | 06:26 | |
| *** garyk has joined #openstack-nova | 06:26 | |
| *** Marga_ has joined #openstack-nova | 06:27 | |
| openstackgerrit | Abhilash Goyal proposed openstack/nova: initial comit Closes-bug: 1414529 which says eval was being used in session.py, which is not considered safe, hence instead of eval I used literal_eval function of ast library in python which is considered safe as it runs only pyhton's valid syntax and ra https://review.openstack.org/275513 | 06:28 |
| openstack | bug 1414529 in OpenStack Compute (nova) "eval being used in session.py" [Low,Confirmed] https://launchpad.net/bugs/1414529 - Assigned to Abhilash Goyal (abhilash-goyal) | 06:28 |
| *** eliqiao has quit IRC | 06:31 | |
| *** eliqiao has joined #openstack-nova | 06:32 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: fixed log warning in sqlalchemy/api.py https://review.openstack.org/246901 | 06:33 |
| *** zul has joined #openstack-nova | 06:33 | |
| *** qeelee has joined #openstack-nova | 06:33 | |
| *** ssurana has joined #openstack-nova | 06:34 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring network model https://review.openstack.org/149627 | 06:36 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring the disk bus https://review.openstack.org/149628 | 06:36 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: adding fake libosinfo module https://review.openstack.org/275519 | 06:36 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: virt: adjusting the osinfo tests to use fakelibosinfo https://review.openstack.org/275520 | 06:36 |
| openstackgerrit | venkatamahesh proposed openstack/nova: Fix some word spellings in messages https://review.openstack.org/275521 | 06:39 |
| *** dedery_ has quit IRC | 06:45 | |
| *** amotoki has quit IRC | 06:46 | |
| *** oomichi has joined #openstack-nova | 06:47 | |
| *** dave-mccowan has quit IRC | 06:48 | |
| *** zul has quit IRC | 06:51 | |
| *** chhavi has joined #openstack-nova | 06:51 | |
| *** armax has quit IRC | 06:52 | |
| *** dedery has joined #openstack-nova | 06:54 | |
| *** esracelik has joined #openstack-nova | 06:59 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-nova | 07:01 | |
| *** hoangcx has quit IRC | 07:03 | |
| *** browne has quit IRC | 07:06 | |
| *** amotoki has joined #openstack-nova | 07:06 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/275530 | 07:06 |
| *** zenoway has joined #openstack-nova | 07:08 | |
| *** paul-carlton1 has joined #openstack-nova | 07:11 | |
| *** mjura has joined #openstack-nova | 07:11 | |
| *** omnipresent has joined #openstack-nova | 07:12 | |
| *** thorst has joined #openstack-nova | 07:13 | |
| *** zenoway has quit IRC | 07:14 | |
| *** itamarl has joined #openstack-nova | 07:14 | |
| *** moshele has joined #openstack-nova | 07:15 | |
| *** lbragstad has quit IRC | 07:16 | |
| *** lpetrut has joined #openstack-nova | 07:16 | |
| *** thorst has quit IRC | 07:18 | |
| *** yamahata has quit IRC | 07:19 | |
| *** diana_clarke has quit IRC | 07:19 | |
| *** moshele has quit IRC | 07:21 | |
| *** lbragstad has joined #openstack-nova | 07:21 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: remove the unnecessary parem of set_vm_state_and_notify https://review.openstack.org/275537 | 07:24 |
| *** ekuris has joined #openstack-nova | 07:31 | |
| *** tdurakov has joined #openstack-nova | 07:31 | |
| tdurakov | pkoniszewski, hi, are you around? | 07:32 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Split-network-plane-for-live-migration https://review.openstack.org/245005 | 07:39 |
| *** coreycb` is now known as coreycb | 07:40 | |
| eliqiao | hi tdurakov, there? | 07:42 |
| tdurakov | eliqiao, yep | 07:42 |
| tdurakov | hi | 07:42 |
| *** Akanksha08 has quit IRC | 07:43 | |
| eliqiao | tdurakov: it's about https://review.openstack.org/#/c/254101/17 , I see you just -1 on it, would like to discuss it now, free ? | 07:43 |
| tdurakov | eliqiao, sure | 07:45 |
| eliqiao | tdurakov: if you check https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L5532 , you will find out that volume backed instance will be covered. | 07:45 |
| *** dedery has quit IRC | 07:46 | |
| tdurakov | eliqiao, and what about xen? | 07:47 |
| *** belmoreira has joined #openstack-nova | 07:47 | |
| *** Marga_ has quit IRC | 07:47 | |
| *** eglynn-afk has joined #openstack-nova | 07:48 | |
| *** penick has quit IRC | 07:48 | |
| *** dedery has joined #openstack-nova | 07:49 | |
| *** Akanksha08 has joined #openstack-nova | 07:50 | |
| *** avishay has joined #openstack-nova | 07:50 | |
| avishay | Hi all, from what I can see in the code Cinder isn't called to detach volume in shelve_offload_instance - is this correct? | 07:51 |
| eliqiao | tdurakov: does xen support boot from volume? I have no idea about it. | 07:53 |
| *** jwcroppe has quit IRC | 07:55 | |
| tdurakov | eliqiao, https://review.openstack.org/#/c/8156/ | 07:56 |
| tdurakov | eliqiao, should work | 07:56 |
| *** mvk_ has quit IRC | 07:56 | |
| tdurakov | johnthetubaguy, hi, are you around? got question about xen. Is it possible to boot from volume in nova using xenapi? | 07:57 |
| *** Sukhdev has quit IRC | 07:57 | |
| *** hoangcx has joined #openstack-nova | 07:57 | |
| eliqiao | tdurakov: john maybe still not get up now. | 07:57 |
| tdurakov | right | 07:58 |
| *** mpavone has joined #openstack-nova | 07:58 | |
| *** jwcroppe has joined #openstack-nova | 07:58 | |
| *** vilobhmm11 has quit IRC | 07:58 | |
| *** takashin has left #openstack-nova | 08:00 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwa_ has joined #openstack-nova | 08:01 | |
| pkoniszewski | tdurakov: hi | 08:01 |
| *** jwcroppe has quit IRC | 08:02 | |
| tdurakov | pkoniszewski, have you start ml thread about live-migration progress? | 08:03 |
| pkoniszewski | tdurakov: not yet, i will start it today in 1-2 hours | 08:03 |
| *** jwcroppe has joined #openstack-nova | 08:04 | |
| tdurakov | pkoniszewski, cool, could you please ping me in irc as thread started | 08:05 |
| pkoniszewski | tdurakov: sure, will do | 08:05 |
| tdurakov | pkoniszewski, thanks! | 08:06 |
| *** liverpooler has joined #openstack-nova | 08:06 | |
| *** zhangjn has quit IRC | 08:07 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Split-network-plane-for-live-migration https://review.openstack.org/245005 | 08:07 |
| openstackgerrit | esracelik proposed openstack/nova: Config options: centralize "configdrive" options https://review.openstack.org/254092 | 08:08 |
| *** zhangjn has joined #openstack-nova | 08:09 | |
| *** zhangjn has quit IRC | 08:10 | |
| *** salv-orlando has joined #openstack-nova | 08:10 | |
| openstackgerrit | esracelik proposed openstack/nova: Config options: centralize section "hyperv" https://review.openstack.org/255124 | 08:10 |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: libvirt: clear instance.root_device_name while ignoring user device name https://review.openstack.org/214314 | 08:10 |
| *** zhangjn has joined #openstack-nova | 08:10 | |
| *** zhangjn has quit IRC | 08:11 | |
| *** zhangjn has joined #openstack-nova | 08:11 | |
| *** zhangjn has quit IRC | 08:11 | |
| *** zhangjn has joined #openstack-nova | 08:12 | |
| *** zhangjn has quit IRC | 08:12 | |
| *** salv-orl_ has quit IRC | 08:12 | |
| openstackgerrit | esracelik proposed openstack/nova: Improve the help text for hyperv options https://review.openstack.org/268175 | 08:13 |
| *** lpetrut has quit IRC | 08:13 | |
| *** jistr has joined #openstack-nova | 08:14 | |
| *** salv-orlando has quit IRC | 08:14 | |
| *** qeelee has quit IRC | 08:14 | |
| openstackgerrit | esracelik proposed openstack/nova: Improve the help text for configdrive options https://review.openstack.org/266875 | 08:14 |
| *** salv-orlando has joined #openstack-nova | 08:14 | |
| *** zhangjn has joined #openstack-nova | 08:15 | |
| *** ihrachys has joined #openstack-nova | 08:15 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: add ploop support into qemu-img info https://review.openstack.org/260636 | 08:15 |
| *** thorst has joined #openstack-nova | 08:16 | |
| *** zhangjn has quit IRC | 08:16 | |
| *** moshele has joined #openstack-nova | 08:16 | |
| *** zhangjn has joined #openstack-nova | 08:17 | |
| *** ekuris has quit IRC | 08:17 | |
| *** boris-42 has joined #openstack-nova | 08:18 | |
| *** ekuris has joined #openstack-nova | 08:20 | |
| *** markus_z has joined #openstack-nova | 08:21 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Expose quiesce/unquiesce API https://review.openstack.org/248989 | 08:22 |
| *** jichen has joined #openstack-nova | 08:22 | |
| *** thorst has quit IRC | 08:22 | |
| *** jistr is now known as jistr|mtg | 08:22 | |
| *** amotoki has quit IRC | 08:25 | |
| *** sinese has joined #openstack-nova | 08:25 | |
| *** zhangjn has quit IRC | 08:25 | |
| *** nikhil has joined #openstack-nova | 08:25 | |
| bauzas | good morning Nova | 08:26 |
| tdurakov | eliqiao, seen your comment, that's my concern. Will it work correctly for volume-backed instances in xen, or we miss something | 08:27 |
| *** mvk_ has joined #openstack-nova | 08:27 | |
| *** nikhil_k has quit IRC | 08:28 | |
| eliqiao | tdurakov: for now, if an xen instance is boot from volume, then when doing livemigration, we need to specify block_migration = False, correct ? | 08:29 |
| eliqiao | tdurakov: I have no idear who is more familar with xen server. seems johnthetubaguy worked for xen before? | 08:29 |
| *** ihrachys has quit IRC | 08:30 | |
| tdurakov | eliqiao, yep, and here is the point that new calculation for block-migration could broke smth. I'd rather wait johnthetubaguy comments | 08:30 |
| eliqiao | tdurakov: pardon ? what is smth ? | 08:31 |
| tdurakov | eliqiao, something | 08:31 |
| eliqiao | tdurakov: oh, good to know :) | 08:32 |
| *** mubirru has quit IRC | 08:32 | |
| *** su_zhang_ has quit IRC | 08:33 | |
| *** Jeffrey4l has quit IRC | 08:33 | |
| *** SridharG has quit IRC | 08:33 | |
| *** garyk has quit IRC | 08:34 | |
| *** garyk has joined #openstack-nova | 08:35 | |
| *** sudipto_ has quit IRC | 08:35 | |
| *** ihrachys has joined #openstack-nova | 08:36 | |
| *** zakora has joined #openstack-nova | 08:37 | |
| *** ihrachys has quit IRC | 08:38 | |
| *** oomichi is now known as oomichi_away | 08:38 | |
| *** bnemec has joined #openstack-nova | 08:39 | |
| *** mubirru has joined #openstack-nova | 08:39 | |
| *** jaosorior has joined #openstack-nova | 08:39 | |
| *** zul has joined #openstack-nova | 08:40 | |
| *** efoley_ has joined #openstack-nova | 08:41 | |
| *** lpetrut has joined #openstack-nova | 08:42 | |
| *** matrohon has joined #openstack-nova | 08:42 | |
| bauzas | oomichi_away: evening | 08:43 |
| bauzas | alex_xu: morning | 08:43 |
| bauzas | oops s/morning/eveing | 08:43 |
| *** jlanoux has joined #openstack-nova | 08:43 | |
| *** SridharG has joined #openstack-nova | 08:44 | |
| *** ihrachys has joined #openstack-nova | 08:44 | |
| bauzas | oomichi_away: alex_xu: we have a FFE for https://review.openstack.org/#/c/253087/7 I'd appreciate if you could take a look on it ? | 08:44 |
| *** mfedosin has joined #openstack-nova | 08:44 | |
| *** rk4n has joined #openstack-nova | 08:46 | |
| *** ihrachys has quit IRC | 08:46 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Improve os-migrateLive input and output values https://review.openstack.org/259319 | 08:48 |
| *** rdopiera has joined #openstack-nova | 08:48 | |
| *** ihrachys has joined #openstack-nova | 08:49 | |
| *** sudipto_ has joined #openstack-nova | 08:52 | |
| *** salv-orlando has quit IRC | 08:52 | |
| *** ihrachys has quit IRC | 08:56 | |
| *** haomaiwa_ has quit IRC | 09:01 | |
| *** EinstCrazy has joined #openstack-nova | 09:01 | |
| *** haomaiwa_ has joined #openstack-nova | 09:01 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Calculate block_migration in virt drivers if it's None https://review.openstack.org/254101 | 09:04 |
| openstackgerrit | Eli Qiao proposed openstack/nova: Return host and block_migration in task_api._execute() https://review.openstack.org/254102 | 09:04 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Improve os-migrateLive input and output values https://review.openstack.org/259319 | 09:04 |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Do not use "cmp" builtin https://review.openstack.org/275187 | 09:06 |
| eliqiao | tdurakov: I just update the patch to address Balazs' comment, can you please help to check with johnthetubaguy about live-migration for volume-backed xen instance P.S I am in evening time now. | 09:06 |
| tdurakov | eliqiao, sure | 09:06 |
| eliqiao | tdurakov: thank you :) | 09:06 |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Do not use 'unicode' but six.text_type https://review.openstack.org/275186 | 09:06 |
| *** ihrachys has joined #openstack-nova | 09:10 | |
| *** ihrachys has quit IRC | 09:10 | |
| *** MCoLo has joined #openstack-nova | 09:11 | |
| *** rdopiera has quit IRC | 09:11 | |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Do not use print statements, but print functions https://review.openstack.org/275184 | 09:12 |
| alex_xu | bauzas: will reach that this evening | 09:13 |
| *** sudipto_ has quit IRC | 09:13 | |
| *** achanda has quit IRC | 09:14 | |
| bauzas | alex_xu: cool, ty | 09:15 |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Do not use "file" builtin, but "open" instead https://review.openstack.org/275183 | 09:15 |
| *** kodokuu has joined #openstack-nova | 09:16 | |
| *** chhavi has quit IRC | 09:16 | |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: xrange() is renamed to range() in Python 3 https://review.openstack.org/275182 | 09:17 |
| kodokuu | garyk Hi | 09:18 |
| *** achanda has joined #openstack-nova | 09:18 | |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: xrange() is renamed to range() in Python 3 https://review.openstack.org/275182 | 09:18 |
| kodokuu | garyk I see you are owner from https://review.openstack.org/#/q/topic:bp/vmware-better-display-names Do you think is possible to add project_name on folder for VC ? | 09:19 |
| kodokuu | Even if I can't find project_name in instance.py :/ | 09:19 |
| *** thorst has joined #openstack-nova | 09:21 | |
| *** ankit_ag has joined #openstack-nova | 09:21 | |
| *** yassine__ has joined #openstack-nova | 09:21 | |
| *** [BNC]markusz has joined #openstack-nova | 09:22 | |
| *** beagles has quit IRC | 09:22 | |
| *** markus_z has quit IRC | 09:22 | |
| *** zhangjn has joined #openstack-nova | 09:22 | |
| *** [BNC]markusz is now known as markus_z | 09:22 | |
| *** rdopiera has joined #openstack-nova | 09:23 | |
| *** bharath_ has joined #openstack-nova | 09:23 | |
| alex_xu | bauzas: np | 09:23 |
| *** zhangjn has quit IRC | 09:24 | |
| bharath_ | Facing " Cannot access storage file '/opt/stack/data/nova/instances/86ef8e22-622c-4315-9cdc-882db1e4ce7f/disk' (as uid:106, gid:114): Permission denied" issue, how to fix this? | 09:24 |
| *** achanda has quit IRC | 09:24 | |
| *** mfedosin has quit IRC | 09:24 | |
| johnthetubaguy | tdurakov: eliqiao: sorry, missing context, how can I help about live-migration? | 09:25 |
| *** dguitarbite has quit IRC | 09:25 | |
| eliqiao | hi johnthetubaguy , it's about the block_migration flag calculatation for xen api. | 09:25 |
| tdurakov | johnthetubaguy, hi, could you look at this patch: https://review.openstack.org/#/c/254101/ what will happen if we try block-migrate volume backed instance on xen? | 09:26 |
| *** dguitarbite has joined #openstack-nova | 09:27 | |
| *** danpb has joined #openstack-nova | 09:27 | |
| johnthetubaguy | eliqiao: tdurakov: so at rackspace, we only every run with block_migration=True as we don't use shared storage for the ephemeral disks | 09:28 |
| *** thorst has quit IRC | 09:28 | |
| *** efoley__ has joined #openstack-nova | 09:29 | |
| tdurakov | johnthetubaguy, no volume-backed instances? | 09:29 |
| johnthetubaguy | tdurakov: if you have a volume backed instance, block_migration will still work just fine, those get disconnected/reconnected and not copied | 09:29 |
| *** EinstCrazy has quit IRC | 09:29 | |
| *** sudipto_ has joined #openstack-nova | 09:29 | |
| bharath_ | Facing " Cannot access storage file '/opt/stack/data/nova/instances/86ef8e22-622c-4315-9cdc-882db1e4ce7f/disk' (as uid:106, gid:114): Permission denied" issue, how to fix this? | 09:29 |
| *** EinstCrazy has joined #openstack-nova | 09:29 | |
| *** gszasz has joined #openstack-nova | 09:30 | |
| *** efoley_ has quit IRC | 09:30 | |
| johnthetubaguy | eliqiao: tdurakov: your check in vmops is correct, the aggregate check should do the trick | 09:31 |
| tdurakov | johnthetubaguy, thank you | 09:31 |
| eliqiao | johnthetubaguy: okay, goacha, thanks, it's really help, good to have you :) | 09:32 |
| *** BobBall has joined #openstack-nova | 09:32 | |
| tdurakov | eliqiao, seems correct to me, will review this ps again and leave comments | 09:32 |
| johnthetubaguy | cool, the xenapi meeting is on now | 09:32 |
| eliqiao | tdurakov: okay, cool, thanks. | 09:32 |
| johnthetubaguy | eliqiao: tdurakov: slight thing, that feels like two separate patches | 09:33 |
| johnthetubaguy | I will pass this link into them | 09:33 |
| *** b3nt_pin has joined #openstack-nova | 09:33 | |
| eliqiao | johnthetubaguy: do you meant I need to split it as libvirt and xen patches? | 09:33 |
| andrearosa | danpb: are you around? I noticed that the roolback live migration alway sets the miration state to 'error' https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L5509 I am not sure that is always correct, for example in libvirt we know whne a job has been cancelled https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L6317 and I think in that case we should set the | 09:33 |
| andrearosa | migration state to "cancelled", do you agree with me? | 09:33 |
| *** b3nt_pin is now known as Guest29540 | 09:33 | |
| *** garyk has quit IRC | 09:33 | |
| johnthetubaguy | eliqiao: thats more normal at least | 09:33 |
| johnthetubaguy | eliqiao: its a different set of folks that would be able to review each half of the patch really well | 09:34 |
| *** garyk has joined #openstack-nova | 09:34 | |
| eliqiao | johnthetubaguy: okay, I feel like to do it my tommorrow time. thanks | 09:35 |
| danpb | andrearosa: yes in the cancelled case it ought to be safe to leave it in normal state | 09:35 |
| andrearosa | danpb: cool, I am going to put a simple patch to give us more flexibility in managing the migration state, I'll ping you when it's ready | 09:36 |
| *** dguitarbite has quit IRC | 09:37 | |
| *** garyk has quit IRC | 09:37 | |
| *** nkrinner has quit IRC | 09:38 | |
| openstackgerrit | jinquanni proposed openstack/nova: Modify the exception handling in order to solve the following questions: nova service-list failed when one or some cell-service exception of multiple child cells close-bug:#1539056 https://review.openstack.org/275583 | 09:39 |
| *** mfedosin has joined #openstack-nova | 09:40 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Xen: Calculate block_migration if it's None https://review.openstack.org/254101 | 09:42 |
| openstackgerrit | Eli Qiao proposed openstack/nova: Return host and block_migration in task_api._execute() https://review.openstack.org/254102 | 09:42 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Improve os-migrateLive input and output values https://review.openstack.org/259319 | 09:42 |
| openstackgerrit | Eli Qiao proposed openstack/nova: Libvirt: Calculate block_migration if it's None https://review.openstack.org/275585 | 09:42 |
| eliqiao | johnthetubaguy: I did a quick shift, :) | 09:43 |
| johnthetubaguy | eliqiao: cool, thank you! | 09:43 |
| *** e0ne has joined #openstack-nova | 09:45 | |
| *** ljxiash has quit IRC | 09:47 | |
| *** dguitarbite has joined #openstack-nova | 09:47 | |
| *** ljxiash has joined #openstack-nova | 09:47 | |
| openstackgerrit | Sean Dague proposed openstack/nova: Enable volume operations for shelved instances https://review.openstack.org/265898 | 09:47 |
| *** mfedosin has quit IRC | 09:48 | |
| *** ljxiash has quit IRC | 09:52 | |
| *** jaosorior has quit IRC | 09:53 | |
| *** lpetrut has quit IRC | 09:54 | |
| *** jaosorior has joined #openstack-nova | 09:54 | |
| *** tdurakov has quit IRC | 09:55 | |
| *** yasemin has joined #openstack-nova | 09:55 | |
| *** zhangjn_ has joined #openstack-nova | 09:57 | |
| *** zhangjn_ has quit IRC | 09:58 | |
| *** PaulMurray has joined #openstack-nova | 09:58 | |
| *** gjayavelu has joined #openstack-nova | 09:59 | |
| *** zhangjn_ has joined #openstack-nova | 10:00 | |
| *** zhangjn_ has quit IRC | 10:00 | |
| *** vivekd has joined #openstack-nova | 10:00 | |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** haomaiwang has joined #openstack-nova | 10:01 | |
| *** efoley__ has quit IRC | 10:01 | |
| *** zhangjn_ has joined #openstack-nova | 10:02 | |
| *** gjayavelu has quit IRC | 10:02 | |
| *** zhangjn_ has quit IRC | 10:02 | |
| *** zhangjn_ has joined #openstack-nova | 10:02 | |
| *** zul has quit IRC | 10:03 | |
| *** mfedosin has joined #openstack-nova | 10:03 | |
| *** e0ne has quit IRC | 10:05 | |
| *** nkrinner has joined #openstack-nova | 10:05 | |
| *** zul has joined #openstack-nova | 10:06 | |
| *** cdent has joined #openstack-nova | 10:07 | |
| *** jaosorior has quit IRC | 10:09 | |
| *** obondarev_ has joined #openstack-nova | 10:14 | |
| *** obondarev has quit IRC | 10:14 | |
| *** zhangjn_ has quit IRC | 10:15 | |
| *** zhangjn_ has joined #openstack-nova | 10:15 | |
| *** jaosorior has joined #openstack-nova | 10:16 | |
| *** salv-orlando has joined #openstack-nova | 10:16 | |
| *** ociuhandu has joined #openstack-nova | 10:18 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/nova: Ironic: Use ironicclient native retries for connection errors https://review.openstack.org/211097 | 10:18 |
| *** salv-orlando has quit IRC | 10:21 | |
| *** wyang has quit IRC | 10:22 | |
| *** BobBall has quit IRC | 10:23 | |
| *** ljxiash has joined #openstack-nova | 10:25 | |
| *** ljxiash has quit IRC | 10:25 | |
| *** hoangcx has quit IRC | 10:26 | |
| *** thorst has joined #openstack-nova | 10:26 | |
| *** rook-laptop has joined #openstack-nova | 10:27 | |
| *** ildikov has quit IRC | 10:27 | |
| *** omnipresent has quit IRC | 10:28 | |
| johnthetubaguy | gibi: I think its these notifications that need converting next: https://review.openstack.org/#/c/248989/93/nova/compute/manager.py | 10:29 |
| johnthetubaguy | gibi: we keep wanting to add more | 10:30 |
| sdague | PaulMurray / andrearosa are you folks fixing the microversion issue on the top patch? | 10:30 |
| sdague | I kind of missed that because the spec was a little vague, and dan had approved it, so I assumed some other agreement was in place | 10:31 |
| PaulMurray | sdague, yes - up soon | 10:32 |
| *** lpetrut has joined #openstack-nova | 10:32 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: add hacking check for help text length https://review.openstack.org/242078 | 10:32 |
| andrearosa | sdague: I will address it, Dan actually -1 the patch | 10:32 |
| sdague | andrearosa: yeh, I noticed this morning | 10:32 |
| *** avishay has quit IRC | 10:32 | |
| sdague | I had to rebase your patch so the depends on tests would work right, but after that I have a tempest test that tests the detach for shelved | 10:33 |
| openstackgerrit | Andrey Pavlov proposed openstack/nova: remove EC2 legacy code https://review.openstack.org/266425 | 10:33 |
| PaulMurray | sdague, also have tempest tests up https://review.openstack.org/#/c/258391 | 10:33 |
| *** thorst has quit IRC | 10:33 | |
| sdague | PaulMurray: ah, great | 10:33 |
| PaulMurray | sdague, we are sorting out depends on | 10:33 |
| PaulMurray | sdague, does it look right to you? | 10:34 |
| sdague | yeh, it's remarkably close to what I have :) | 10:34 |
| sdague | though in the attach case, I think you actually want to boot after it | 10:34 |
| sdague | because that was the use case, right? | 10:35 |
| PaulMurray | Yes, I'll get jlanoux to sort it | 10:35 |
| andrearosa | jlanoux: ^^ | 10:36 |
| andrearosa | ops too late | 10:36 |
| sdague | if you need help on this one, let me know. I'm happy to help sling some code on these tests. | 10:36 |
| pkoniszewski | tdurakov_: email sent, its called Migration progress | 10:37 |
| PaulMurray | sdague, BTW https://review.openstack.org/#/c/259528 wasn't +A - I think dansmith intended johnthetubaguy to do that after removing -2 | 10:37 |
| pkoniszewski | PaulMurray: ^^ | 10:37 |
| PaulMurray | pkoniszewski, thanks | 10:37 |
| sdague | PaulMurray: I think he wanted test results as well | 10:38 |
| *** efoley__ has joined #openstack-nova | 10:38 | |
| sdague | I think it's fine to get the results, and just approve the 2 patches in one go later today. Because the infrastructure patch doesn't make sense to land without the API landing | 10:39 |
| PaulMurray | sdague, ok - that sounds fine to me - just worried about rebasing | 10:39 |
| PaulMurray | but ok | 10:39 |
| *** rfolco has joined #openstack-nova | 10:40 | |
| *** aysyd has joined #openstack-nova | 10:40 | |
| sdague | right, understood, there are 14 patches in flight that currently conflict with this, but none of them seem super problematic | 10:40 |
| sdague | we can always quick rebase and approve if that's the issue after a patch is already approved | 10:41 |
| *** Daisy has joined #openstack-nova | 10:42 | |
| *** zenoway has joined #openstack-nova | 10:42 | |
| PaulMurray | sdague, re tempest tests - ideally we should ssh into the vms and check they can see the volumes as appropriate | 10:43 |
| PaulMurray | sdague, do you want us to do that? | 10:43 |
| openstackgerrit | jichenjc proposed openstack/nova: move disassociate to network_associate.py https://review.openstack.org/275615 | 10:43 |
| sdague | for the attach case, yes | 10:44 |
| sdague | I feel like for the shelve detach case it's fine, as it's mostly resource cleanup | 10:44 |
| sdague | but for attaching to a shelved instance than unshelving, seeing the attachment there would be good | 10:45 |
| sdague | so it's not just the API doing hand waving | 10:45 |
| *** dedery has quit IRC | 10:46 | |
| *** zenoway has quit IRC | 10:46 | |
| sdague | I'll leave review comments | 10:46 |
| sdague | there is also a thing this test will need to do once the microversions are fixed otherwise it won't run | 10:47 |
| gibi | johnthetubaguy: thanks for the info, I started collecting the list of patches that are proposing new notifications here https://etherpad.openstack.org/p/nova-versioned-notifications | 10:47 |
| *** aysyd has quit IRC | 10:47 | |
| *** aysyd has joined #openstack-nova | 10:48 | |
| *** eglynn has joined #openstack-nova | 10:50 | |
| *** dedery has joined #openstack-nova | 10:51 | |
| *** e0ne has joined #openstack-nova | 10:51 | |
| *** ljxiash has joined #openstack-nova | 10:52 | |
| *** obondarev_ has quit IRC | 10:52 | |
| *** obondarev has joined #openstack-nova | 10:53 | |
| *** eglynn-afk has quit IRC | 10:54 | |
| openstackgerrit | jichenjc proposed openstack/nova: Add TODO for 202 to 204 change on networks delete https://review.openstack.org/275620 | 10:55 |
| jlanoux | ok - thanks sdague | 10:56 |
| sdague | jlanoux: otherwise, looks pretty good | 10:57 |
| *** ljxiash has quit IRC | 10:57 | |
| *** jaosorior has quit IRC | 10:58 | |
| *** zenoway has joined #openstack-nova | 10:58 | |
| *** jaosorior has joined #openstack-nova | 10:59 | |
| *** deepakcs has quit IRC | 10:59 | |
| *** zul has quit IRC | 11:00 | |
| *** ndipanov has joined #openstack-nova | 11:00 | |
| *** ndipanov has quit IRC | 11:00 | |
| *** haomaiwang has quit IRC | 11:01 | |
| *** zul has joined #openstack-nova | 11:01 | |
| *** ljxiash has joined #openstack-nova | 11:01 | |
| *** ndipanov has joined #openstack-nova | 11:01 | |
| *** haomaiwang has joined #openstack-nova | 11:01 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: [WIP] POC OVS offload https://review.openstack.org/275624 | 11:02 |
| *** efoley__ has quit IRC | 11:04 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: ironic driver: Adding InfiniBand support https://review.openstack.org/266540 | 11:04 |
| *** yassine__ has quit IRC | 11:05 | |
| *** chhavi has joined #openstack-nova | 11:07 | |
| *** salv-orlando has joined #openstack-nova | 11:07 | |
| *** efoley__ has joined #openstack-nova | 11:12 | |
| *** gszasz has quit IRC | 11:12 | |
| *** zhangjn_ has quit IRC | 11:13 | |
| *** deepakcs has joined #openstack-nova | 11:13 | |
| *** zhangjn has joined #openstack-nova | 11:13 | |
| *** sfinucan has joined #openstack-nova | 11:13 | |
| *** efoley__ is now known as efoley | 11:16 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Tidy up scheduler_evolution.rst https://review.openstack.org/262561 | 11:20 |
| cdent | bauzas: would be nice to get that through ^ | 11:20 |
| *** jaosorior has quit IRC | 11:21 | |
| *** ptm_away has joined #openstack-nova | 11:22 | |
| *** mleroy has left #openstack-nova | 11:24 | |
| *** PaulMurray has quit IRC | 11:26 | |
| *** ndipanov has quit IRC | 11:27 | |
| *** jistr|mtg has quit IRC | 11:27 | |
| *** thorst has joined #openstack-nova | 11:32 | |
| openstackgerrit | Wen Zhi Yu proposed openstack/nova: Make sure destination will rollback if live migration fails https://review.openstack.org/270093 | 11:32 |
| *** rotbeard has joined #openstack-nova | 11:37 | |
| *** thorst has quit IRC | 11:38 | |
| *** deepakcs has quit IRC | 11:39 | |
| *** Akanksha08 has quit IRC | 11:40 | |
| *** sahid has joined #openstack-nova | 11:41 | |
| *** tdurakov__ has joined #openstack-nova | 11:43 | |
| *** boris-42 has quit IRC | 11:43 | |
| *** vivekd_ has joined #openstack-nova | 11:44 | |
| *** vivekd has quit IRC | 11:45 | |
| *** vivekd_ is now known as vivekd | 11:45 | |
| *** bkumar has joined #openstack-nova | 11:45 | |
| *** dedery has quit IRC | 11:46 | |
| openstackgerrit | jichenjc proposed openstack/nova: Add rebuild action descriptions in support-matrix https://review.openstack.org/275641 | 11:48 |
| *** dedery has joined #openstack-nova | 11:50 | |
| *** wyang has joined #openstack-nova | 11:51 | |
| *** efoley has quit IRC | 11:53 | |
| *** gszasz has joined #openstack-nova | 11:54 | |
| *** deepakcs has joined #openstack-nova | 11:55 | |
| *** deepakcs has quit IRC | 11:55 | |
| Anticimex | heh, /me just realized that nova can't use cinders multi-backend thing, allowing an instance to use storage from two or more Ceph pools | 11:56 |
| *** pkoniszewski has quit IRC | 11:56 | |
| Anticimex | saw this blueprint, abandoned in juno cycle: https://review.openstack.org/#/c/91957/ | 11:56 |
| *** pkoniszewski has joined #openstack-nova | 11:56 | |
| Anticimex | has that been implemented already and I missed it, or is it possible to resurrect and improve on that blueprint perhaps | 11:57 |
| Anticimex | +? if anyone knows. | 11:57 |
| *** yasemin has left #openstack-nova | 11:59 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: reset task_state after select_destinations failed. https://review.openstack.org/275650 | 11:59 |
| *** vivekd_ has joined #openstack-nova | 12:00 | |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwa_ has joined #openstack-nova | 12:01 | |
| *** rook-laptop has quit IRC | 12:03 | |
| tdurakov__ | eliqiao, are you still online? | 12:03 |
| *** smatzek has joined #openstack-nova | 12:04 | |
| *** tdurakov__ is now known as tdurakov | 12:04 | |
| *** vivekd has quit IRC | 12:04 | |
| *** vivekd_ is now known as vivekd | 12:04 | |
| *** sudipto_ has quit IRC | 12:06 | |
| *** ildikov has joined #openstack-nova | 12:07 | |
| *** rook-laptop has joined #openstack-nova | 12:08 | |
| *** Akanksha08 has joined #openstack-nova | 12:09 | |
| *** avishay has joined #openstack-nova | 12:10 | |
| *** ihrachys has joined #openstack-nova | 12:10 | |
| *** rodrigods has quit IRC | 12:11 | |
| *** rodrigods has joined #openstack-nova | 12:11 | |
| *** links has quit IRC | 12:11 | |
| *** ihrachys has quit IRC | 12:13 | |
| *** Akanksha08 has quit IRC | 12:15 | |
| *** zz_dimtruck has quit IRC | 12:15 | |
| *** raildo-afk is now known as raildo | 12:17 | |
| *** zul has quit IRC | 12:24 | |
| *** lucasagomes is now known as lucas-hungry | 12:25 | |
| tdurakov | markus_z, hi are you around? | 12:25 |
| markus_z | tdurakov: hi, yeah, I'm here. Finally recovered. | 12:26 |
| tdurakov | markus_z, good to hear that, so about job for latest packages? | 12:27 |
| tdurakov | markus_z, do you need anything for this? any help? | 12:28 |
| *** aagrawal has joined #openstack-nova | 12:28 | |
| markus_z | tdurakov: we could have a short chart about that projects.yaml file where they are definded if you have some minutes? | 12:29 |
| tdurakov | markus_z, sure | 12:29 |
| markus_z | It's this huge think I guess https://github.com/openstack/governance/blob/master/reference/projects.yaml | 12:29 |
| *** ociuhandu_ has joined #openstack-nova | 12:30 | |
| markus_z | *thing | 12:30 |
| tdurakov | markus_z, you need to define something like: https://review.openstack.org/#/c/243013/7/jenkins/jobs/projects.yaml | 12:30 |
| markus_z | tdurakov: ah, ok | 12:31 |
| *** dims has joined #openstack-nova | 12:32 | |
| markus_z | tdurakov: can I add a filter or something? that it only starts when a change in the /virt/libvirt/ folder happens? | 12:32 |
| *** haomaiw__ has joined #openstack-nova | 12:32 | |
| *** ptm__ has joined #openstack-nova | 12:33 | |
| tdurakov | markus_z, so, you want to this job to be triggered only if changes are happened in /virt/libvirt package only? | 12:33 |
| markus_z | tdurakov: yes | 12:33 |
| *** qeelee has joined #openstack-nova | 12:33 | |
| *** ankit_ag has quit IRC | 12:34 | |
| *** haomaiwa_ has quit IRC | 12:34 | |
| *** lixiaoy1 has quit IRC | 12:34 | |
| *** ptm_away has quit IRC | 12:34 | |
| *** shyama has quit IRC | 12:34 | |
| *** ociuhandu has quit IRC | 12:34 | |
| *** ociuhandu_ is now known as ociuhandu | 12:34 | |
| *** shyama has joined #openstack-nova | 12:34 | |
| tdurakov | markus_z these criteria as set here: https://review.openstack.org/#/c/267669/1/zuul/layout.yaml | 12:34 |
| *** abhishekk has quit IRC | 12:36 | |
| markus_z | tdurakov: I have to read that as "skip if NOT project openstack/nova", right? | 12:37 |
| tdurakov | markus_z, it should be read as if project is openstack/nova and files changed not in directories listed bellow then execute this job | 12:39 |
| *** zhangjn has quit IRC | 12:39 | |
| sdague | markus_z: right, if project == nova and all files changed are in the following matches, then skip | 12:39 |
| sdague | so basically the part of the tree that's irrelevant | 12:39 |
| openstackgerrit | Abhilash Goyal proposed openstack/nova: Close-bug: 1414529 Change-Id: Ib6145408360fa57cccc8d77c590a203e5088b193 https://review.openstack.org/275513 | 12:40 |
| openstack | bug 1414529 in OpenStack Compute (nova) "eval being used in session.py" [Low,In progress] https://launchpad.net/bugs/1414529 - Assigned to Abhilash Goyal (abhilash-goyal) | 12:40 |
| -openstackstatus- NOTICE: Infra running with lower capacity now, due to a temporary problem affecting one of our nodepool providers. Please expect some delays in your jobs. Apologies for any inconvenience caused. | 12:40 | |
| *** baoli has joined #openstack-nova | 12:40 | |
| markus_z | tdurakov: sdague: ok, thanks, I think I got it. | 12:41 |
| tdurakov | markus_z, cool:) | 12:41 |
| *** zhangjn has joined #openstack-nova | 12:42 | |
| *** ekuris has quit IRC | 12:42 | |
| *** rook-laptop has quit IRC | 12:42 | |
| *** baoli_ has joined #openstack-nova | 12:43 | |
| *** yassine has joined #openstack-nova | 12:43 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 12:45 |
| *** baoli has quit IRC | 12:45 | |
| *** dedery has quit IRC | 12:46 | |
| *** thorst_ has joined #openstack-nova | 12:46 | |
| *** zhangjn has quit IRC | 12:47 | |
| *** bharath_ has quit IRC | 12:48 | |
| *** zhangjn has joined #openstack-nova | 12:48 | |
| *** zhangjn has quit IRC | 12:48 | |
| *** zhangjn has joined #openstack-nova | 12:49 | |
| *** daemontool has joined #openstack-nova | 12:52 | |
| *** atuvenie has joined #openstack-nova | 12:54 | |
| *** Jeffrey4l has joined #openstack-nova | 12:55 | |
| *** zul has joined #openstack-nova | 12:55 | |
| *** dedery has joined #openstack-nova | 12:55 | |
| *** sinese has quit IRC | 12:56 | |
| *** zenoway has quit IRC | 12:56 | |
| *** atuvenie has quit IRC | 12:57 | |
| *** jaosorior has joined #openstack-nova | 12:57 | |
| *** dguitarbite has quit IRC | 12:57 | |
| *** sinese has joined #openstack-nova | 12:57 | |
| *** atuvenie has joined #openstack-nova | 12:58 | |
| *** zenoway has joined #openstack-nova | 13:00 | |
| *** haomaiw__ has quit IRC | 13:01 | |
| *** haomaiwa_ has joined #openstack-nova | 13:01 | |
| *** sinese_ has joined #openstack-nova | 13:01 | |
| *** openstackgerrit has quit IRC | 13:02 | |
| *** openstackgerrit has joined #openstack-nova | 13:03 | |
| *** sinese has quit IRC | 13:04 | |
| *** toure has quit IRC | 13:04 | |
| *** toure has joined #openstack-nova | 13:04 | |
| *** rook-laptop has joined #openstack-nova | 13:08 | |
| *** ihrachys has joined #openstack-nova | 13:08 | |
| openstackgerrit | Sean Dague proposed openstack/nova: libvirt: deprecate live/block_migration_flag opts https://review.openstack.org/263436 | 13:10 |
| *** efoley has joined #openstack-nova | 13:12 | |
| *** jistr|mtg has joined #openstack-nova | 13:12 | |
| *** claudiub has joined #openstack-nova | 13:13 | |
| *** haomaiwa_ has quit IRC | 13:16 | |
| *** ducttape_ has joined #openstack-nova | 13:17 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Cleanup NovaObjectDictCompat from BandwidthUsage https://review.openstack.org/263638 | 13:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Cleanup NovaObjectDictCompat from NumaTopology https://review.openstack.org/260016 | 13:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Move wsgi options into a group https://review.openstack.org/263812 | 13:18 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: centralize section "wsgi" https://review.openstack.org/252687 | 13:18 |
| * dims provides a heads up - watch out of oslo.messaging 4.1.0 breaks | 13:18 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: test_fields: Remove all 'Enum' subclass tests https://review.openstack.org/244372 | 13:19 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Add new API to force live migration to complete https://review.openstack.org/245921 | 13:21 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Add new DB API method to retrieve migration for instance https://review.openstack.org/257270 | 13:21 |
| bauzas | sdague: do you know if we have some problems with the apt caches ? http://logs.openstack.org/62/258962/2/check/gate-tempest-dsvm-full/45b878c/console.html#_2016-02-03_11_54_58_169 | 13:22 |
| *** xyang1 has joined #openstack-nova | 13:23 | |
| sdague | bauzas: yes, rax iad was failing to talk ipv6 | 13:23 |
| sdague | it's been pulled out of nodepool | 13:23 |
| sdague | -openstackstatus/#openstack-nova- NOTICE: Infra running with lower capacity now, due to a temporary problem affecting one of our nodepool providers. Please expect some delays in your jobs. Apologies for any inconvenience caused. | 13:23 |
| bauzas | okay, was just looking at logstash | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.firewall' options https://review.openstack.org/257995 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.netutils' options https://review.openstack.org/258009 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Improve 'virt.images' conf options documentation https://review.openstack.org/259553 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.imagecache' options https://review.openstack.org/258984 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.images' options https://review.openstack.org/258004 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.disk' options https://review.openstack.org/258912 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'virt.driver' options https://review.openstack.org/257956 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Improve 'virt.driver' conf options documentation https://review.openstack.org/266364 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Improve 'virt.firewall' conf options documentation https://review.openstack.org/266365 | 13:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Improve 'pci' conf options documentation https://review.openstack.org/243335 | 13:24 |
| sdague | I pushed a signature | 13:25 |
| sdague | it just hasn't hit yet | 13:25 |
| bauzas | sdague: okay | 13:27 |
| bauzas | sdague: do you have the bug # ? | 13:27 |
| *** ducttape_ has quit IRC | 13:28 | |
| *** lucas-hungry is now known as lucasagomes | 13:29 | |
| *** erlarese has joined #openstack-nova | 13:30 | |
| *** ihrachys has quit IRC | 13:30 | |
| dedery | hi, i'm trying to back-port bug fix #1522112 from master to liberty. Jenkins fails on things that seem unrelated ([ERROR] /opt/stack/new/devstack/functions-common:1004 Failed to update apt repos, we're dead now). what can be done? patch-set: https://review.openstack.org/#/c/270686 | 13:30 |
| bauzas | dedery: see my question just above | 13:30 |
| bauzas | it's an infra issue | 13:30 |
| bauzas | so please wait before rechecking | 13:31 |
| sdague | bauzas: you can recheck now | 13:31 |
| sdague | we just disabled that region | 13:31 |
| bauzas | ack, bug number for the recheck ? | 13:31 |
| bauzas | sdague: good to know | 13:31 |
| dedery | bauzas: thanks! | 13:31 |
| sdague | 1541364 | 13:32 |
| *** ihrachys has joined #openstack-nova | 13:33 | |
| *** aagrawal has quit IRC | 13:33 | |
| *** sinese_ has quit IRC | 13:36 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Improve 'vnc' conf options documentation https://review.openstack.org/266273 | 13:36 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Make noVNC proxy into vnc group https://review.openstack.org/269635 | 13:36 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: config options: Centralise 'vnc' options https://review.openstack.org/263763 | 13:36 |
| *** haomaiwang has joined #openstack-nova | 13:36 | |
| bauzas | sdague: ty | 13:37 |
| *** mubirru has quit IRC | 13:38 | |
| *** edmondsw has joined #openstack-nova | 13:38 | |
| johnthetubaguy | markus_z: glad you are feeling better now, welcome back! | 13:39 |
| *** daemontool has quit IRC | 13:39 | |
| johnthetubaguy | markus_z: +1 your concerns around the doc bugs | 13:39 |
| markus_z | johnthetubaguy: thanks! | 13:39 |
| *** cdelatte has quit IRC | 13:39 | |
| *** sinese_ has joined #openstack-nova | 13:39 | |
| markus_z | johnthetubaguy: yeah. But I guess pushing the problem on the docs team was wrong from the beginning. | 13:41 |
| *** itamarl has quit IRC | 13:41 | |
| *** itamarl has joined #openstack-nova | 13:41 | |
| *** qeelee has quit IRC | 13:42 | |
| johnthetubaguy | markus_z: yeah, I figured this was going to happen about a year ago, its about time we cleaned up our own mess! | 13:42 |
| johnthetubaguy | I mean we have started doing that with API docs, effectively | 13:42 |
| johnthetubaguy | right now, I feel thats the most important mess, and your work on the config options docs is attacking the other side of things quite nicely | 13:42 |
| markus_z | agreed | 13:44 |
| markus_z | johnthetubaguy: speaking of them, can we keep up the work there or is it better to stop it? | 13:44 |
| *** rodolfo-alonso-h has joined #openstack-nova | 13:44 | |
| bauzas | that's refactoring, so only impacted by M-3 FF IMHO | 13:45 |
| johnthetubaguy | markus_z: yeah, I promised you I would decide and I haven't. if we can stop all the other similar tasks, its temping to keep that going. | 13:45 |
| johnthetubaguy | bauzas: yeah, its just the review distraction that worries me | 13:45 |
| *** rodolfo-alonso-h is now known as rodolfo-alonso | 13:45 | |
| *** dedery has quit IRC | 13:45 | |
| johnthetubaguy | but the docs stuff is important | 13:46 |
| markus_z | they are not easy to review, I have to admit that | 13:46 |
| dims | johnthetubaguy : dashboards don't help? | 13:46 |
| bauzas | we can just maybe an optimistic approach and not fence proactively but wait for M3 ? | 13:46 |
| *** atuvenie has quit IRC | 13:46 | |
| johnthetubaguy | dims: only so far, we just don't have enough people doing reviews | 13:46 |
| dims | johnthetubaguy : right | 13:46 |
| *** vivekd has quit IRC | 13:46 | |
| johnthetubaguy | dims: hence the non-priority feature freeze, so we get the super important stuff getting review bandwidth | 13:47 |
| dims | johnthetubaguy : yep, understood. | 13:47 |
| sdague | markus_z: here's the thing, DocImpact triggering a bug in our tracker means that reviewers will actually maybe care about autogenerating bugs, as they are on our team | 13:48 |
| dims | johnthetubaguy : so one person saw two -2's in a row in different reviews in different oslo projects and complained on list :) | 13:48 |
| sdague | right now it's a way to auto generate bugs on another team that no one understands | 13:48 |
| *** Jeffrey4l has quit IRC | 13:48 | |
| sdague | at least nova people should have the context about what the Doc bug means | 13:49 |
| sdague | and maybe stop randomly using DocImpact | 13:49 |
| johnthetubaguy | sdague: yeah, feeling the pain we are creating, often helps fix these things | 13:49 |
| sdague | when it's not needed | 13:49 |
| *** itamarl has quit IRC | 13:49 | |
| *** Jeffrey4l has joined #openstack-nova | 13:49 | |
| dims | sdague : +1 | 13:49 |
| bauzas | sdague: since we have reno now in place, I hope it could reduce that problem | 13:49 |
| sdague | bauzas: me too | 13:49 |
| bauzas | by asking people to provide a relnote instead of a tag | 13:50 |
| sdague | others weren't so sure, but at least we'll be able to control that | 13:50 |
| markus_z | sdague: right, the generous use of docimpact will hit us now ourselves. | 13:50 |
| sdague | markus_z: the only new bugs generated here is when 2 core reviewers believe DocImpact is correct | 13:50 |
| sdague | which we'll just have to remember to care about now | 13:50 |
| sdague | markus_z: and you should feel free to yell at folks about it :) | 13:51 |
| sdague | when they get lazy | 13:51 |
| bauzas | yeah, that's a good point | 13:51 |
| bauzas | I maybe also did that by not thinking about asking to remove the tag if the reno file was there | 13:51 |
| bauzas | very good catch | 13:51 |
| johnthetubaguy | yeah, its a good conversation to be having | 13:52 |
| bauzas | markus_z: remember that conf opts are explicitely mentioned in the reno usage thing http://docs.openstack.org/developer/nova/code-review.html#when-a-release-note-is-needed | 13:52 |
| johnthetubaguy | do folks know if the config options get auto regenerated in the admin docs? | 13:52 |
| johnthetubaguy | or is that a manual thing? | 13:52 |
| markus_z | sdague: I don't have the resources to skim all bug reports. But let's wait what happens and when we have data. | 13:52 |
| *** teju has quit IRC | 13:53 | |
| bauzas | johnthetubaguy: you mean the opt list ? | 13:53 |
| johnthetubaguy | yeah | 13:53 |
| bauzas | like http://docs.openstack.org/liberty/config-reference/content/list-of-compute-config-options.html | 13:53 |
| markus_z | johnthetubaguy: I guess the do now. At least some reviews left that impression at me. | 13:53 |
| johnthetubaguy | yeah, that | 13:53 |
| bauzas | johnthetubaguy: I remember someone told me once it was manual | 13:53 |
| bauzas | but that would require a doublecheck | 13:54 |
| *** ljxiash has quit IRC | 13:54 | |
| johnthetubaguy | looking at that, its got none of our recent longer text for the config options | 13:54 |
| markus_z | bauzas: The manual creation gets triggered manually, but the content gets slurped directly from the code. | 13:54 |
| bauzas | yeah that I was sure | 13:54 |
| johnthetubaguy | so I think we generated lots of DocImpact to just trigger a refresh of those docs | 13:54 |
| johnthetubaguy | thats all I am thinking | 13:54 |
| bauzas | markus_z: but I mean, someone has to call the content, right? | 13:55 |
| *** cdelatte has joined #openstack-nova | 13:55 | |
| *** claudiub has quit IRC | 13:55 | |
| bauzas | markus_z: it's not like if tox -egenconfig was running periodically or on every commit | 13:55 |
| *** sean-k-mooney has joined #openstack-nova | 13:55 | |
| markus_z | bauzas: right, I don't remember who's doing that on which event. | 13:56 |
| *** Akanksha08 has joined #openstack-nova | 13:56 | |
| bauzas | markus_z: lemme check the docs repo | 13:56 |
| johnthetubaguy | if someone fancies creating a bot for that... it could save us a heap of DocImpact bugs | 13:56 |
| johnthetubaguy | well basically most of them | 13:56 |
| bauzas | johnthetubaguy: honestly, we no longer need to have DocImpact for that | 13:56 |
| *** amotoki has joined #openstack-nova | 13:57 | |
| *** erlon has joined #openstack-nova | 13:57 | |
| bauzas | johnthetubaguy: if doc people look at our relnotes, they'll get the message | 13:57 |
| bauzas | TBH, I see our relnotes as the communication interface with others (CDers, docs team, infra etc.) | 13:58 |
| johnthetubaguy | I think we need to take more responsibility for making sure that happens, thats all I was thinking. | 13:58 |
| johnthetubaguy | yeah, it sure helps | 13:58 |
| johnthetubaguy | its also reminding us what the hell we did yesterday | 13:58 |
| *** gjayavelu has joined #openstack-nova | 13:58 | |
| *** sean-k-mooney has left #openstack-nova | 14:00 | |
| *** haomaiwang has quit IRC | 14:01 | |
| johnthetubaguy | danpb: I would love it if you could have a look at this one, and see if we are likely to get this merged by Friday? https://blueprints.launchpad.net/nova/+spec/libvirt-hardware-policy-from-libosinfo | 14:01 |
| *** haomaiwa_ has joined #openstack-nova | 14:01 | |
| *** ihrachys has quit IRC | 14:01 | |
| markus_z | just to double-check that I get it right, is the combination of "docimpact" AND a reno file a valid combination? Right now I think yes, but only if a doc change in a manual (in a non-nova repo) is necessary. | 14:01 |
| *** Kevin_Zheng has quit IRC | 14:02 | |
| *** efoley_ has joined #openstack-nova | 14:03 | |
| *** efoley has quit IRC | 14:03 | |
| danpb | johnthetubaguy: yeah should be doable | 14:03 |
| danpb | it was just waiting for some test case changes sdague wanted | 14:04 |
| bauzas | johnthetubaguy: so see https://raw.githubusercontent.com/openstack/openstack-manuals/26510e84ddabe9ead475bd67b5671557dd1c5893/doc/config-reference/source/compute/config-options.rst | 14:04 |
| *** ZZelle-web has joined #openstack-nova | 14:04 | |
| ZZelle-web | johnthetubaguy: hi | 14:05 |
| bauzas | johnthetubaguy: which points to eg. https://raw.githubusercontent.com/openstack/openstack-manuals/26510e84ddabe9ead475bd67b5671557dd1c5893/doc/config-reference/source/tables/nova-api.rst | 14:05 |
| ZZelle-web | johnthetubaguy: it's unclear to me what i should add in https://etherpad.openstack.org/p/mitaka-nova-non-priority-ff-tracking to request a FFE for https://review.openstack.org/130526 | 14:06 |
| *** Kevin_Zheng has joined #openstack-nova | 14:07 | |
| *** rotbeard has quit IRC | 14:09 | |
| *** sean-k-mooney has joined #openstack-nova | 14:10 | |
| *** daemontool has joined #openstack-nova | 14:10 | |
| *** ivase has joined #openstack-nova | 14:11 | |
| markus_z | bauzas: right, and that tables creation gets triggered by a human from the docs team AFAIK. | 14:11 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: adding fake libosinfo module https://review.openstack.org/275519 | 14:11 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring network model https://review.openstack.org/149627 | 14:12 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: virt: adjusting the osinfo tests to use fakelibosinfo https://review.openstack.org/275520 | 14:12 |
| bauzas | markus_z: I know the last committer so I can ask him | 14:12 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring the disk bus https://review.openstack.org/149628 | 14:12 |
| markus_z | yes, that would be good. | 14:12 |
| *** ivase_ has joined #openstack-nova | 14:12 | |
| *** mdrabe has joined #openstack-nova | 14:13 | |
| *** carlito has joined #openstack-nova | 14:13 | |
| *** speller has joined #openstack-nova | 14:14 | |
| bauzas | markus_z: https://review.openstack.org/#/c/259889/ is the related change | 14:14 |
| *** speller_ has joined #openstack-nova | 14:15 | |
| *** speller_ has quit IRC | 14:15 | |
| *** ivase has quit IRC | 14:15 | |
| carlito | My instances that are created with some users in some tenants do not get an IP address. I found that the project_id field of the network table contained one of our tenant id's, and vm's were indeed working there. Just set the project_id field to NULL, but new instances still don't get an IP. Any ideas? | 14:15 |
| carlito | I did `debug=True` and now I see: "Networks retrieved for instance: |[]|" | 14:16 |
| *** thumpba has joined #openstack-nova | 14:16 | |
| carlito | So it just cannot find any network at all it seems. The vlan field is cleared. This is on Juno btw. | 14:16 |
| *** rlrossit has joined #openstack-nova | 14:16 | |
| *** thumpba has quit IRC | 14:17 | |
| *** ljxiash has joined #openstack-nova | 14:17 | |
| *** amotoki has quit IRC | 14:17 | |
| *** thumpba has joined #openstack-nova | 14:17 | |
| *** ZZelle-web has quit IRC | 14:18 | |
| *** dave-mccowan has joined #openstack-nova | 14:20 | |
| *** jistr|mtg is now known as jistr | 14:20 | |
| *** amotoki has joined #openstack-nova | 14:21 | |
| *** cdelatte has quit IRC | 14:22 | |
| *** su_zhang has joined #openstack-nova | 14:23 | |
| johnthetubaguy | ZZelle: just need to put a link in that FFE request section, with a reason why its more important that the existing FFEs | 14:25 |
| *** jichen has quit IRC | 14:25 | |
| bauzas | sean-k-mooney: around ? wanted to chat on https://review.openstack.org/#/c/275128/ | 14:25 |
| *** jwcroppe has quit IRC | 14:26 | |
| *** mrkz has joined #openstack-nova | 14:26 | |
| *** daemontool has quit IRC | 14:27 | |
| bauzas | rodolfo-alonso: around too ? | 14:29 |
| *** bkumar has quit IRC | 14:30 | |
| sean-k-mooney | bauzas: intermitently sorry i am not sure how long my battery will last. | 14:30 |
| sean-k-mooney | bauzas: but yes i can talk | 14:30 |
| bauzas | sean-k-mooney: fair enough, just wanted to explain my views and discuss with you synchronously | 14:31 |
| bauzas | sean-k-mooney: so, I'm totally fine adding a new filter, but I want to deprecate the old one | 14:31 |
| *** ihrachys has joined #openstack-nova | 14:32 | |
| bauzas | sean-k-mooney: because adding 2 filters for the same usage (even if the inclusive/exclusive way is different) sounds confusing for our operators | 14:32 |
| sean-k-mooney | buazas: ok do you have an opion on the mechanisum i suggested for enableing the inclusive/exclusive symantics | 14:33 |
| bauzas | sean-k-mooney: also, since it's really easy to use an out-of-tree filter by modifying the conf opt, I really want to make sure that we keep the in-tree filters very clear | 14:33 |
| rodolfo-alonso | bauzas: hi | 14:33 |
| bauzas | sean-k-mooney: it looks maybe fine, but lemme explain what I want | 14:33 |
| sean-k-mooney | sure | 14:33 |
| *** burgerk has quit IRC | 14:33 | |
| bauzas | as an op, I'm already using AggregateInstanceExtraSpecsFilter with some aggregate metadata already there + some flavor extra specs | 14:34 |
| bauzas | so, I'll see that with Mitaka, there will be a new filter providing the same behaviour but also adding new semantics | 14:35 |
| bauzas | so, I'll only change my conf opt for no longer using AggregateInstanceExtraSpecsFilter but rather using the new filter and it will still work | 14:35 |
| bauzas | later, as an op, I could modify my aggregates to refine my logic - or create new aggregates if I wish with different metadata | 14:35 |
| *** Swami has joined #openstack-nova | 14:36 | |
| *** edtubill has joined #openstack-nova | 14:36 | |
| bauzas | but I'll be sure that changing my filter will not need either my aggs metadata or my flavors to be modified | 14:36 |
| bauzas | sean-k-mooney: see the UX ? | 14:36 |
| rodolfo-alonso | bauzas: but the new filter doen't gives the same behaviour | 14:37 |
| bauzas | rodolfo-alonso: I know, hence my points | 14:37 |
| *** ihrachys has quit IRC | 14:37 | |
| rodolfo-alonso | bauzas: maybe we need to re write the doc | 14:37 |
| sean-k-mooney | yes i think that would be a good UX. the gap as the new fiter is currently specifed dose not support the old behavior as rodolf said | 14:37 |
| sean-k-mooney | s/rodolf/redolfo | 14:38 |
| *** Guest29540 is now known as beagles | 14:38 | |
| bauzas | sean-k-mooney: the main problem I see is that operators would need to change their agg metadata if they would want to use your filter | 14:38 |
| *** gjayavelu has quit IRC | 14:38 | |
| rodolfo-alonso | bauzas: yes, and doesn't collides with the old one | 14:38 |
| bauzas | plus the fact that you don't support neither the scoping of the flavor specs, nor the possibility to use some semantics like <= | 14:39 |
| rodolfo-alonso | bauzas: sure, and we can add it | 14:39 |
| bauzas | rodolfo-alonso: well, it still checks aggs vs. flavors right? | 14:39 |
| sean-k-mooney | i agree to enable the filter they would have to modify there aggregate. we could write a db migrate to do the conversion. | 14:39 |
| bauzas | sean-k-mooney: please no | 14:39 |
| bauzas | operators don't want to change their metadata | 14:40 |
| sean-k-mooney | bauzas: yes it would be rather complicated so i would prefer not to also | 14:40 |
| *** cdelatte has joined #openstack-nova | 14:40 | |
| bauzas | it would even be a terrible experience given that changing the metadata is an API resource | 14:40 |
| bauzas | and we can have some API admins that are not ops on the machine | 14:41 |
| rodolfo-alonso | bauzas: I know, and if you use both filters, you don't need to change the old config, just add your new metadata (if you need it) | 14:41 |
| *** efoley_ has quit IRC | 14:41 | |
| sean-k-mooney | bauzas: so it will still be comparing aggregated to flavors yes but the relation ship is different like the difference between cinder and swift both are storage solution with differnt semantics | 14:41 |
| bauzas | rodolfo-alonso: the idea is that we would deprecate the old filter by the same time | 14:42 |
| *** delattec has joined #openstack-nova | 14:42 | |
| bauzas | rodolfo-alonso: so technically, operators could still use both at the same time, but that's not really useful since both will return the same hosts | 14:42 |
| rodolfo-alonso | bauzas: to deprecate the old filter without changing metadata, we need to rewrite it again | 14:43 |
| bauzas | rodolfo-alonso: I know hence my point to ask "please provide backwards compat'" | 14:43 |
| sean-k-mooney | bauzas: so the first too change you suggested regarding the scope and match op i think can be done easily. | 14:43 |
| rodolfo-alonso | bauzas: please, read my answer (https://review.openstack.org/#/c/275128/) | 14:43 |
| bauzas | rodolfo-alonso: if the semantics were identical, I would just have asked to provide a deprecation notice for the old filter | 14:44 |
| sean-k-mooney | bauzas: actully if you set the same requirement in both filter they would return different hosts in most cases | 14:44 |
| rodolfo-alonso | bauzas: sean it's right | 14:44 |
| bauzas | sean-k-mooney: that would be weird | 14:44 |
| *** _RuiChen has quit IRC | 14:44 | |
| *** tdurakov has quit IRC | 14:44 | |
| *** Marga_ has joined #openstack-nova | 14:44 | |
| sean-k-mooney | they are desined for different usecases | 14:45 |
| *** burt has joined #openstack-nova | 14:45 | |
| *** _RuiChen has joined #openstack-nova | 14:45 | |
| rodolfo-alonso | bauzas: both filter data in the opposite way | 14:45 |
| rodolfo-alonso | bauzas: that's why we made the new one | 14:45 |
| sean-k-mooney | bauzas: the new filter is intened to preserve host with sepcifics consumable resources form beeing used by systems that dont request them | 14:45 |
| *** cdelatte has quit IRC | 14:45 | |
| bauzas | rodolfo-alonso: that's a technical reason why not doing that - not an user story | 14:45 |
| bauzas | rodolfo-alonso: if you want to match key/values, the old filter was just ensuring that *ALL* key/value pairs of flavors have to be present in the aggregate | 14:47 |
| *** tongli has joined #openstack-nova | 14:47 | |
| bauzas | that's what I want to see in your filter | 14:47 |
| bauzas | you can iterate over the agg keys, I'm fine | 14:47 |
| sean-k-mooney | bauzas: yes but ours makes sure all key/values in the aggreate is in the flavor | 14:47 |
| bauzas | but you need to ensure the same assertion, ie. that all flavor key/value pairs are present | 14:47 |
| bauzas | sean-k-mooney: that's a semantic nit | 14:48 |
| bauzas | lemme explain | 14:48 |
| sean-k-mooney | bauzas: i have to grab a chager but i should so ill be back in a sec but keep going | 14:48 |
| bauzas | ok | 14:48 |
| bauzas | so, the point is that you write some flavor specs | 14:49 |
| bauzas | in order to compose | 14:49 |
| *** anteaya has quit IRC | 14:49 | |
| *** cfriesen has joined #openstack-nova | 14:50 | |
| bauzas | if you consider that all aggregate keys have to be present on the flavor extra spec, then it's just exactly the same as if all the flavor specs are present on the agg, right ? | 14:50 |
| sean-k-mooney | no | 14:51 |
| rodolfo-alonso | bauzas: no | 14:51 |
| bauzas | if you want to use an aggregate metadata key for something else, then create another aggregate | 14:51 |
| sean-k-mooney | if the aggregate is a subset of the flavor it is different | 14:51 |
| *** ctrath has joined #openstack-nova | 14:52 | |
| bauzas | okay, let's consider the old usecase | 14:52 |
| bauzas | three cases | 14:52 |
| bauzas | 1. both are perfect matches | 14:52 |
| bauzas | in that case, inverting the logic doesn't really cate | 14:53 |
| bauzas | care | 14:53 |
| sean-k-mooney | agreed | 14:53 |
| *** efoley_ has joined #openstack-nova | 14:53 | |
| bauzas | 2. we have more flavor keys than agg keys | 14:53 |
| *** Daisy has quit IRC | 14:53 | |
| bauzas | in that case, the old filter was denying | 14:53 |
| sean-k-mooney | and the new flavor would allow | 14:53 |
| *** ducttape_ has joined #openstack-nova | 14:54 | |
| bauzas | that's what I want to change :) | 14:54 |
| *** esracelik has quit IRC | 14:54 | |
| *** anteaya has joined #openstack-nova | 14:54 | |
| *** ducttape_ has quit IRC | 14:54 | |
| sean-k-mooney | that is the main thing we want to support | 14:54 |
| rodolfo-alonso | to change in the new filter? | 14:54 |
| *** ducttape_ has joined #openstack-nova | 14:54 | |
| *** Daisy has joined #openstack-nova | 14:54 | |
| rodolfo-alonso | that's one of the reasons of the new filter | 14:54 |
| *** pushkaru has joined #openstack-nova | 14:54 | |
| *** mriedem has joined #openstack-nova | 14:54 | |
| bauzas | sean-k-mooney: because you could add some sentinels | 14:55 |
| bauzas | ie. | 14:55 |
| sean-k-mooney | the rational is i would like to specify that an aggregate reuires all instnace to use say cpu_policy=dedicated | 14:55 |
| bauzas | say that you want to support case #2 | 14:55 |
| *** Daisy has quit IRC | 14:55 | |
| bauzas | flavor says a=1, b=2 | 14:55 |
| bauzas | your agg is saying a=1 | 14:55 |
| bauzas | then using the new filter would fail, but if I would invert the logic | 14:56 |
| rodolfo-alonso | no, will pass | 14:56 |
| rodolfo-alonso | with the new filter will pass | 14:56 |
| bauzas | I could write in my agg meta a=1,* | 14:56 |
| bauzas | what I want with your filter is | 14:56 |
| *** ljxiash has quit IRC | 14:57 | |
| bauzas | for a=1, b=1 in flavor extra spec | 14:57 |
| bauzas | a=1 in metadata fails | 14:57 |
| bauzas | but | 14:57 |
| rodolfo-alonso | yes | 14:57 |
| bauzas | a=1,* (or something equivalent) succeeds | 14:57 |
| rodolfo-alonso | nonono | 14:57 |
| bauzas | why ? | 14:57 |
| rodolfo-alonso | a=1,b=1 in agg | 14:57 |
| bauzas | :) | 14:57 |
| rodolfo-alonso | and a=1 in flavor | 14:57 |
| rodolfo-alonso | --> pass | 14:57 |
| *** bklei has joined #openstack-nova | 14:58 | |
| rodolfo-alonso | because it's and inclusive filter | 14:58 |
| sean-k-mooney | so how a about a simple usecase | 14:58 |
| *** Akanksha08 has left #openstack-nova | 14:58 | |
| bauzas | sorry, I explained why I want :) | 14:58 |
| sean-k-mooney | for cpu pinning to work all instnace on a host must use the same policy shared or dedicated | 14:59 |
| bauzas | because what I explained provides the exaxt same behaviour as the old filter | 14:59 |
| bauzas | like I said, if I want to use wildcards, I'm okay | 14:59 |
| *** pradk has joined #openstack-nova | 14:59 | |
| sean-k-mooney | with the new filter i can exress just that requirement by setting hw:cpu_policy=value | 14:59 |
| *** zhangjn has quit IRC | 15:00 | |
| sean-k-mooney | but i want the instance to be able to request other values | 15:00 |
| bauzas | sean-k-mooney: and that's what I want to change :) | 15:00 |
| sean-k-mooney | the issue it with the shared in this case | 15:00 |
| bauzas | hw:cpu_policy=value,whatever is fine | 15:00 |
| bauzas | whatever being * or something you want as a sentinel | 15:00 |
| *** haomaiwa_ has quit IRC | 15:01 | |
| *** ljxiash has joined #openstack-nova | 15:01 | |
| sean-k-mooney | i need to be able to say hw:cpu_policy=shared or hw:cpu_policy is not set | 15:01 |
| *** haomaiwang has joined #openstack-nova | 15:01 | |
| *** delatte has joined #openstack-nova | 15:03 | |
| *** delattec has quit IRC | 15:03 | |
| sean-k-mooney | bauzas: ok so in our new filter if we had the following would it be ok with you | 15:03 |
| sean-k-mooney | agg( a=1) flavor (a=1 b=2) fail | 15:03 |
| sean-k-mooney | agg(a=1,*) flavor(a=1 b=2) pass | 15:04 |
| *** delatte has quit IRC | 15:04 | |
| *** zhangjn has joined #openstack-nova | 15:04 | |
| sean-k-mooney | agg(a=1,a=~,*) flavor(b=2) pass | 15:04 |
| johnthetubaguy | isn't the issue here really that we have two very similar looking filters, and its hard to work out which one is best, and hard to work out why we should maintain both? | 15:05 |
| *** Daisy has joined #openstack-nova | 15:05 | |
| *** tdurakov has joined #openstack-nova | 15:05 | |
| bauzas | sorry was afk | 15:05 |
| bauzas | sean-k-mooney: your semantics looks fine to me | 15:06 |
| johnthetubaguy | if the filter docs say exactly why you want to use each of them, does that get us closer? | 15:06 |
| bauzas | johnthetubaguy: I think we're getting closer | 15:06 |
| johnthetubaguy | bauzas: this new filter sounds more powerful than the old, so can we just drop the old one? or are they properly different? | 15:06 |
| bauzas | johnthetubaguy: yup, that's what I asked :) | 15:07 |
| johnthetubaguy | yeah, just wondering if we have an answer to that yet | 15:07 |
| bauzas | johnthetubaguy: but for that, we need to care of the compatibility with the existing filter | 15:07 |
| bauzas | hence us disserting on the semantics | 15:07 |
| bauzas | I was explaining that the old filter cares about having all the flavor extra specs mentioned in the agg metadata | 15:08 |
| johnthetubaguy | OK, gotcha | 15:08 |
| sean-k-mooney | ok i think we can represent our semantics with addtional sentenil if by defalt we use the old semantics | 15:08 |
| *** amotoki has quit IRC | 15:08 | |
| bauzas | sean-k-mooney: that was my point :) | 15:08 |
| bauzas | I'm all up for adding new semantics for your usecases | 15:08 |
| johnthetubaguy | that sounds better, expanding the use cases of the old filter, slightly | 15:09 |
| bauzas | I just want to keep the default semantics being compatible with AggregateInstanceExtraSpecsFilter | 15:09 |
| sean-k-mooney | with the old filter though am i correct that i cant have namespaced keys | 15:09 |
| sean-k-mooney | e.g. hw:* | 15:09 |
| johnthetubaguy | but we can add that, if we want it? | 15:09 |
| bauzas | sean-k-mooney: which is good for you, right ? | 15:09 |
| bauzas | sean-k-mooney: I don't want to constraint your new logic, just making sure that we keep the existing logic | 15:10 |
| *** Daisy has quit IRC | 15:10 | |
| bauzas | if you want to use wildcards, I'm fine | 15:10 |
| johnthetubaguy | oh, I see, that confuses the logic | 15:10 |
| sean-k-mooney | bauzas: no we need namespaced keys | 15:10 |
| *** eharney has joined #openstack-nova | 15:10 | |
| sean-k-mooney | or at least | 15:10 |
| sean-k-mooney | i need to be able to handel both a:b and c:b seperatly | 15:11 |
| sean-k-mooney | so if i have a:b in the aggreate and c:b in the flavor they should not match | 15:11 |
| *** cdelatte has joined #openstack-nova | 15:12 | |
| sean-k-mooney | if i have just b in the aggreagte then i would be ok with it matching a:b in the flavor i gues | 15:12 |
| sean-k-mooney | *guess | 15:12 |
| bauzas | so the old filter was only caring about the keys | 15:12 |
| bauzas | and then the values | 15:12 |
| sean-k-mooney | the old fileter was only caring about un namespaced key or keys in it own namespace | 15:13 |
| *** jwcroppe has joined #openstack-nova | 15:13 | |
| bauzas | sec, checking | 15:13 |
| *** jed56 has joined #openstack-nova | 15:13 | |
| bauzas | I see your point on namespaces | 15:13 |
| bauzas | not on values | 15:13 |
| sean-k-mooney | basically i would like the namespace to be considerd part of the key | 15:14 |
| *** ajayaa has joined #openstack-nova | 15:14 | |
| sean-k-mooney | does that make sense? | 15:14 |
| bauzas | sean-k-mooney: so, see https://github.com/openstack/nova/blob/master/nova/scheduler/filters/aggregate_instance_extra_specs.py#L53-L60 | 15:14 |
| *** ajayaa has quit IRC | 15:14 | |
| bauzas | sean-k-mooney: it was not really namespaced | 15:15 |
| *** ajayaa has joined #openstack-nova | 15:15 | |
| rodolfo-alonso | you need "scope" to scape the namespace | 15:15 |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Mox -> mock for _save_helper() in test_instance_info_cache.py https://review.openstack.org/271711 | 15:15 |
| bauzas | sean-k-mooney: because we were getting the keyname in the agg by getting the first item of the namespace (but the scope) | 15:15 |
| bauzas | AFAICT, the scope is purely a flavor thing | 15:15 |
| bauzas | not an aggregate metadata thing | 15:16 |
| *** boris-42 has joined #openstack-nova | 15:16 | |
| bauzas | so, scope:x:y=foo in the flavor was looking for x=foo in the metadata | 15:16 |
| bauzas | man | 15:17 |
| bauzas | sorry | 15:17 |
| *** kurtmartin has joined #openstack-nova | 15:17 | |
| bauzas | scope:x:y=foo in the flavor was looking for x:y=foo in the metadata | 15:17 |
| sean-k-mooney | yes | 15:17 |
| sean-k-mooney | bauzas: so you would have to prefix hw:cpu_policy with scope: | 15:18 |
| bauzas | it was optional but you would have to support it, yes | 15:18 |
| bauzas | but say that I was writing that : | 15:18 |
| bauzas | in the flavor => scope:hw:*=foo | 15:18 |
| sean-k-mooney | well https://github.com/openstack/nova/blob/master/nova/scheduler/filters/aggregate_instance_extra_specs.py#L55 would skip just hw:cpu_policy | 15:18 |
| bauzas | in the agg hw:*=foo, it worked | 15:19 |
| pkholkin | sdague: hello Sean! could you please look once again oslo.cache patch https://review.openstack.org/#/c/274468/ | 15:19 |
| pkholkin | you are our main reviewer) | 15:19 |
| *** mtanino has joined #openstack-nova | 15:19 | |
| johnthetubaguy | sahid: I was looking at the last two patches for the libvirt-real-time, can we get those rebased please? https://review.openstack.org/#/q/topic:bp/libvirt-real-time,n,z | 15:19 |
| *** delattec has joined #openstack-nova | 15:19 | |
| sdague | pkholkin: looking | 15:19 |
| johnthetubaguy | sahid: or are those patches not really needed anymore? | 15:20 |
| bauzas | sean-k-mooney: mmm, correct | 15:20 |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Simple mox removals from objects/test_instance_info_cache.py https://review.openstack.org/275753 | 15:20 |
| bauzas | sean-k-mooney: because you would be using the ':' operator | 15:20 |
| sdague | pkholkin: +2 | 15:20 |
| pkholkin | sdague: great news) thanks! | 15:21 |
| sean-k-mooney | bauzas: yes ideally i would like for the operator to not have to change there existing flavor to use the new filter | 15:21 |
| bauzas | sean-k-mooney: me neither :) | 15:22 |
| *** cdelatte has quit IRC | 15:22 | |
| *** clayton has joined #openstack-nova | 15:23 | |
| bauzas | but operators weren't using x:y in the flavors right? unless they were writing scope:x:y | 15:23 |
| *** ljxiash has quit IRC | 15:23 | |
| sean-k-mooney | bauzas: they were | 15:23 |
| *** ihrachys has joined #openstack-nova | 15:23 | |
| sean-k-mooney | bauzas: for cpu policy, hugepages | 15:23 |
| sean-k-mooney | bauzas: trusted compute and more | 15:24 |
| bauzas | sean-k-mooney: I guess not with that filter :) | 15:24 |
| sean-k-mooney | yep | 15:24 |
| bauzas | I know that | 15:24 |
| sean-k-mooney | this filter would not work with it | 15:24 |
| bauzas | I'll have to drop | 15:25 |
| clayton | does anyone know why nova creates both a console log and a console pty serial devices via libvirt? | 15:25 |
| sean-k-mooney | bauzas: ok thanks for your time | 15:25 |
| bauzas | alaski: cells meeting at 0500UTC right? | 15:25 |
| bauzas | oops | 15:25 |
| rodolfo-alonso | bauzas: thanks! | 15:25 |
| *** ljxiash has joined #openstack-nova | 15:25 | |
| bauzas | 1700UTC I mean | 15:25 |
| alaski | bauzas: 1700 yeah | 15:25 |
| bauzas | alaski: okay, I'll try to be there :/ | 15:25 |
| sean-k-mooney | bauzas: ill follow up with rodolfo offline and see if i can addapt the new filter to be backwards compatible. | 15:26 |
| *** delatte has joined #openstack-nova | 15:27 | |
| sean-k-mooney | bauzas: we can continue the discution on the review and we can pop back into the nova channel someother time if we need to discuss more | 15:27 |
| bauzas | coolness | 15:27 |
| *** delattec has quit IRC | 15:28 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: neutron: pci_request logic considers 'direct-physical' vnic type https://review.openstack.org/262584 | 15:28 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: pci: Add utility method for getting the MAC addr https://review.openstack.org/262585 | 15:28 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: neutron: Update the port with a MAC address for PFs https://review.openstack.org/262586 | 15:28 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: Add a vnic type for PF passthrough and a new libvirt vif driver https://review.openstack.org/262583 | 15:28 |
| *** delatte has quit IRC | 15:29 | |
| *** mgoddard_ has joined #openstack-nova | 15:30 | |
| dims | sdague : thanks. pkholkin, snikitin : great job! | 15:31 |
| *** mdrabe has quit IRC | 15:31 | |
| *** liverpooler has quit IRC | 15:31 | |
| pkholkin | dims: thank you for support!) | 15:31 |
| *** Swami_ has joined #openstack-nova | 15:32 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Move config options from nova/objects/network.py https://review.openstack.org/269240 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Improve the help text for the linuxnet options (4) https://review.openstack.org/275423 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Improve the help text for the linuxnet options (2) https://review.openstack.org/271489 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Improve the help text for the linuxnet options (3) https://review.openstack.org/272231 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Move config options from nova/network/linux_net.py https://review.openstack.org/270441 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Improve the help text for the linuxnet options (1) https://review.openstack.org/270959 | 15:32 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Improve help text for the network object options https://review.openstack.org/269313 | 15:32 |
| sahid | johnthetubaguy: yes i will do | 15:33 |
| *** knikolla has joined #openstack-nova | 15:33 | |
| *** mgoddard has quit IRC | 15:33 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 15:33 |
| johnthetubaguy | sahid: cool, thank you | 15:34 |
| *** efoley_ has quit IRC | 15:36 | |
| *** Swami has quit IRC | 15:36 | |
| *** Marga_ has quit IRC | 15:37 | |
| *** cdelatte has joined #openstack-nova | 15:37 | |
| *** Marga_ has joined #openstack-nova | 15:37 | |
| *** efoley_ has joined #openstack-nova | 15:39 | |
| *** doug-fish has joined #openstack-nova | 15:39 | |
| *** delattec has joined #openstack-nova | 15:40 | |
| *** Sukhdev has joined #openstack-nova | 15:41 | |
| *** dimtruck has joined #openstack-nova | 15:42 | |
| *** cdelatte has quit IRC | 15:43 | |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Remove mox from tests/unit/objects/test_quotas.py https://review.openstack.org/271712 | 15:45 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: reset task_state after select_destinations failed. https://review.openstack.org/275650 | 15:46 |
| *** zul has quit IRC | 15:46 | |
| *** tdurakov has quit IRC | 15:46 | |
| *** rotbeard has joined #openstack-nova | 15:47 | |
| *** bharath_ has joined #openstack-nova | 15:48 | |
| *** mdrabe has joined #openstack-nova | 15:48 | |
| *** mlavalle has joined #openstack-nova | 15:48 | |
| mlavalle | alaski: ping | 15:48 |
| alaski | mlavalle: hi | 15:48 |
| *** tdurakov has joined #openstack-nova | 15:49 | |
| mlavalle | alaski: hi.... could you help me a little bit with a patchset I submitted to Nova? https://review.openstack.org/#/c/271578/ | 15:49 |
| mlavalle | alaski: I am getting +1 from Jenkins, but it is failing 1 test in dsvm-tempest-xen | 15:50 |
| *** zhangjn has quit IRC | 15:50 | |
| mlavalle | alaski: can I trigger a re-execution of just dsvm-tempest-xen? | 15:50 |
| alaski | mlavalle: xen: recheck | 15:51 |
| mlavalle | alaski: thanks! | 15:51 |
| *** zhangjn has joined #openstack-nova | 15:51 | |
| alaski | mlavalle: if you toggle on the CI comments each system should include instructions on triggering a recheck | 15:51 |
| *** zhangjn has quit IRC | 15:51 | |
| mlavalle | alaski: ok, i'll take a look | 15:51 |
| alaski | mlavalle: that xen failure does look unrelated to your change | 15:51 |
| *** efoley_ has quit IRC | 15:52 | |
| mlavalle | alaski: yeah, I already analyzed the console log and I am sure it is not related to my patch. The test failing uses Nova network. My patch is with Neutron | 15:52 |
| markus_z | dansmith: Do you have a minute? I'm enhancing the "migrate_data" object you introduced and need a hand. | 15:53 |
| dansmith | markus_z: sure | 15:53 |
| dansmith | markus_z: by "enhancing" do you mean "fixing some bug" I introduced? :D | 15:53 |
| markus_z | dansmith: No, fixing the serial console live migration bug by using your flawless code ;) | 15:54 |
| dansmith | \o/ | 15:54 |
| markus_z | dansmith: I haven't touched the RPC stuff yet, that's why I'm not sure what I'm doing | 15:54 |
| *** salv-orl_ has joined #openstack-nova | 15:54 | |
| *** thumpba has quit IRC | 15:55 | |
| *** efoley_ has joined #openstack-nova | 15:55 | |
| markus_z | dansmith: I added a new field to the "fields" dict of LibvirtLiveMigrateData and added a "obj_make_compatible" method. | 15:55 |
| markus_z | with a version bump to 1.1 | 15:55 |
| dansmith | cool | 15:55 |
| markus_z | the from/to_legacy_dict confuses me a bit | 15:56 |
| markus_z | do I consider this new field there too? | 15:56 |
| *** salv-orlando has quit IRC | 15:57 | |
| *** nkrinner has quit IRC | 15:57 | |
| dansmith | markus_z: well, not in to_legacy_dict(), because this field didn't exist in the code that just passed dicts around (i.e. before my change). In from_legacy_dict, you might need to synthesize the value of the new field you added, but that depends on what you're doing | 15:57 |
| dansmith | markus_z: how about you put it up as a WIP and let me comment there? | 15:57 |
| markus_z | dansmith: I can do that. The unit tests aren't changed then, because they make use of "to_legacy_dict". | 15:59 |
| *** burgerk has joined #openstack-nova | 15:59 | |
| markus_z | dansmith: I'll need still some time, I'll ping you then, thanks in advance. | 15:59 |
| dansmith | markus_z: yeah, that was partly my way of making sure that we weren't changing the expectations of the current code, but obviously going forward some of those will need to change to more objecty comparisons once we're adding more things | 16:00 |
| dansmith | sure | 16:00 |
| *** jgriffith_away is now known as jgriffith | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-nova | 16:01 | |
| *** jdurgin1 has joined #openstack-nova | 16:02 | |
| *** lennyb has quit IRC | 16:03 | |
| tjones_zzzz | sdague: hi do you have time to chat about https://review.openstack.org/#/c/270303/? | 16:03 |
| *** rodolfo-alonso has quit IRC | 16:06 | |
| *** ptm__ is now known as PaulMurray | 16:08 | |
| *** bharath_ has quit IRC | 16:08 | |
| openstackgerrit | Dina Belova proposed openstack/nova: Integrate OSprofiler and Nova https://review.openstack.org/254703 | 16:08 |
| *** SridharG has quit IRC | 16:09 | |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Remove mox from simple test_instance.py tests https://review.openstack.org/264292 | 16:10 |
| *** mgoddard_ has quit IRC | 16:12 | |
| *** mgoddard has joined #openstack-nova | 16:12 | |
| *** zenoway has quit IRC | 16:13 | |
| *** mjura has quit IRC | 16:13 | |
| *** shyama has quit IRC | 16:15 | |
| *** shyama has joined #openstack-nova | 16:15 | |
| *** Marga_ has quit IRC | 16:16 | |
| *** lennyb has joined #openstack-nova | 16:16 | |
| *** Marga_ has joined #openstack-nova | 16:18 | |
| *** ihrachys has quit IRC | 16:20 | |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Mox -> mock for test_refresh test in test_instance.py https://review.openstack.org/275794 | 16:21 |
| johnthetubaguy | mriedem: pumaranikar is looking at this bug https://bugs.launchpad.net/nova/+bug/1338735 and it seems very related to this: https://review.openstack.org/#/c/202770/ and all our disconnect_problems, due you have any ideas? also just checking as we backport this one a few places | 16:22 |
| openstack | Launchpad bug 1338735 in OpenStack Compute (nova) "Live-migration with volumes creates orphan access records" [Low,Confirmed] - Assigned to Pushkar Umaranikar (pushkar-umaranikar) | 16:22 |
| *** matrohon has quit IRC | 16:24 | |
| *** rook-laptop has quit IRC | 16:24 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Acquire TCP ports for console during live migration https://review.openstack.org/215102 | 16:25 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: prevent serial console port exhaustion https://review.openstack.org/275800 | 16:25 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: WIP: fill serial ports in migrate data https://review.openstack.org/275801 | 16:25 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: remove the unnecessary parem of set_vm_state_and_notify https://review.openstack.org/275537 | 16:26 |
| mriedem | johnthetubaguy: hmmm, i wasn't on that review, but remember seeing it before. might want ndipanov or hemna to look at that bug | 16:26 |
| mriedem | johnthetubaguy: the bug was opened before that change was made wasn't it? | 16:26 |
| mriedem | i'd probably ask what version of nova they are on | 16:26 |
| *** ljxiash has quit IRC | 16:26 | |
| *** sc68cal has joined #openstack-nova | 16:26 | |
| mriedem | johnthetubaguy: it was reported with a dell storage driver, maybe smcginnis can get some test people to see if that's recreatable | 16:27 |
| smcginnis | mriedem: Sorry, cinder meeting right now. I'll follow up after. | 16:27 |
| markus_z | dansmith: That's the change with "migrate_data": https://review.openstack.org/#/c/215102/13 | 16:27 |
| *** belmoreira has quit IRC | 16:28 | |
| *** speller has quit IRC | 16:28 | |
| markus_z | dansmith: When I execute the unit tests in https://review.openstack.org/#/c/275801/1 I get the error "NotImplementedError: Cannot load 'serial_listen_ports' in the base class" which I don't understand. | 16:28 |
| tdurakov | alaski, ping | 16:29 |
| alaski | tdurakov: hi | 16:29 |
| dansmith | markus_z: that means you have an object that doesn't have your new property set, and then you tried to touch it | 16:29 |
| dansmith | markus_z: basically, it's trying to lazy-load that field, but you didn't provide a lazy-load helper to actually load it | 16:29 |
| tdurakov | alaski, hi, about https://review.openstack.org/#/c/249110/4/nova/tests/unit/network/test_rpcapi.py - it's better to pass context_obj instead, right? | 16:29 |
| johnthetubaguy | mriedem: ah, good points, thanks, smcginnis sounds like a good person to ask, pumaranikar that sound OK? | 16:31 |
| markus_z | dansmith: something like "obj_load_attr" in instance.py? | 16:31 |
| *** rook-laptop has joined #openstack-nova | 16:31 | |
| dansmith | markus_z: right, but, there is no lazy-loading possible here since this is not DB-backed, so you don't need to add it | 16:31 |
| *** ljxiash has joined #openstack-nova | 16:32 | |
| dansmith | markus_z: let me pull it down and I'll see where you're hitting it and then explain | 16:32 |
| danpb | dansmith: is there guidance for projects using ovo wrt namespaces ? | 16:32 |
| markus_z | dansmith: ok, thanks | 16:32 |
| alaski | tdurakov: yes, you should pass in a context object. my comment is just that it shouldn't be an admin context in this case | 16:32 |
| danpb | dansmith: ie should os-vif be setting OBJ_SERIAL_NAMESPACE / OBJ_PROJECT_NAMESPACE ? | 16:32 |
| danpb | or is it permissible for libraries to just use the default namespace | 16:32 |
| dansmith | danpb: one of those should be set yeah.. one is only there for nova compat.. I think there are comments in ovo/base.py about which one to use right? | 16:33 |
| tdurakov | alaski, acked, thanks | 16:33 |
| alaski | tdurakov: np | 16:33 |
| danpb | dansmith: yea comment says SERIAL_NAMESPACE is not needed except for back compat | 16:33 |
| *** carlito has quit IRC | 16:33 | |
| danpb | but it wasn't clear whether changing these is mandatory or not | 16:33 |
| dansmith | danpb: yeah, so ignore that one and set the other one to something yeah | 16:33 |
| danpb | ok | 16:33 |
| dansmith | danpb: okay, I'll cook up a comment change to make it more clear | 16:34 |
| *** ihrachys has joined #openstack-nova | 16:34 | |
| *** diana_clarke has joined #openstack-nova | 16:35 | |
| dansmith | markus_z: hmm, which test fails? all of test_migrate_data passes for me | 16:35 |
| *** zenoway has joined #openstack-nova | 16:35 | |
| *** thumpba has joined #openstack-nova | 16:35 | |
| markus_z | dansmith: test_pre_live_migration_works_correctly_mocked | 16:36 |
| markus_z | in the libvirt/test_driver.py | 16:36 |
| * dansmith runs | 16:36 | |
| dansmith | markus_z: also works for me.. should I be running on the "fill serial ports .." patch? | 16:37 |
| markus_z | dansmith: yes, please. That one shows then the "cannot load" error | 16:38 |
| dansmith | okay | 16:38 |
| dansmith | markus_z: ah, so you haven't set that property yet, but you .append() to it | 16:38 |
| *** diana_clarke has quit IRC | 16:39 | |
| dansmith | markus_z: so since it's not set, it's trying to lazy-load it, before running .append() on whatever value it finds | 16:39 |
| dansmith | so you just need to initialize it to [] first | 16:39 |
| dansmith | migrate_data.serial_listen_ports = [] | 16:39 |
| *** mgarza_ has joined #openstack-nova | 16:39 | |
| markus_z | dansmith: ok, my assumption was that this is already done when I declare it as a specific field type. | 16:39 |
| *** freealvin has quit IRC | 16:40 | |
| *** zenoway has quit IRC | 16:40 | |
| dansmith | markus_z: nope :) | 16:40 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Use instance in setup_networks_on_host https://review.openstack.org/249110 | 16:40 |
| dansmith | markus_z: the field type specifies the data type, but not the default value.. for loooots of reasons | 16:40 |
| *** mriedem is now known as mriedem_afk | 16:40 | |
| *** cdent has quit IRC | 16:41 | |
| markus_z | dansmith: ok, I just believe it :) | 16:41 |
| *** ljxiash has quit IRC | 16:41 | |
| markus_z | dansmith: So I just use __init__ in the LibvirtLiveMigrateData class? | 16:41 |
| dansmith | markus_z: no, just set it on the object right before the for loop | 16:41 |
| *** tdurakov has quit IRC | 16:41 | |
| markus_z | dansmith: oh, ok, I had that once and it felt somehow odd. Nevertheless, thanks for pointing me to the issue. I'll push tomorrow new patch sets for that. | 16:43 |
| *** efoley_ has quit IRC | 16:43 | |
| *** Swami__ has joined #openstack-nova | 16:43 | |
| dansmith | markus_z: it's because the object has to be able to represent a part of what is in the database so we're not passing around tons of data all the time, and without overwriting things in the DB when we save a partial object | 16:44 |
| dansmith | markus_z: it makes less sense here, but it's the way the infrastructure works | 16:44 |
| *** tonyb has quit IRC | 16:45 | |
| *** ljxiash has joined #openstack-nova | 16:45 | |
| *** armax has joined #openstack-nova | 16:46 | |
| *** tonyb has joined #openstack-nova | 16:46 | |
| *** ijw has joined #openstack-nova | 16:46 | |
| *** kevinbenton has quit IRC | 16:46 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Replace flavors.get_all_flavors_sorted_list() with object call https://review.openstack.org/268801 | 16:46 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Refactor and objectify flavor fakes used in api tests https://review.openstack.org/268800 | 16:46 |
| markus_z | dansmith: I skimmed through your blog entries about this topic before I pinged you, but I think I have to read that more attentively again to get it. | 16:47 |
| *** Swami_ has quit IRC | 16:47 | |
| *** ijw has quit IRC | 16:47 | |
| *** ijw has joined #openstack-nova | 16:48 | |
| *** ijw has quit IRC | 16:48 | |
| *** ndipanov has joined #openstack-nova | 16:48 | |
| *** aarefiev has quit IRC | 16:48 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove unused db parameter from scheduler.utils.set_vm_state_and_notify() https://review.openstack.org/267682 | 16:48 |
| *** aarefiev has joined #openstack-nova | 16:48 | |
| * markus_z is leaving for today | 16:49 | |
| *** diana_clarke has joined #openstack-nova | 16:49 | |
| dansmith | markus_z: might not be in there.. it's a detail :) | 16:49 |
| openstackgerrit | Rushi Agrawal proposed openstack/nova: Remove one instance of mox from unit/objects/test_instance.py https://review.openstack.org/270921 | 16:50 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Gracefully handle a deleting instance during rebuild https://review.openstack.org/243005 | 16:50 |
| *** kevinbenton has joined #openstack-nova | 16:51 | |
| *** delattec has quit IRC | 16:51 | |
| *** su_zhang has quit IRC | 16:53 | |
| *** yassine has quit IRC | 16:54 | |
| *** lpetrut has quit IRC | 16:54 | |
| smcginnis | pumaranikar: I've pinged our test group and the owner of the EQLX driver to see if they can weigh in on that bug. | 16:55 |
| *** whenry has quit IRC | 16:55 | |
| pumaranikar | smcginnis, thanks | 16:56 |
| *** EinstCrazy has quit IRC | 16:56 | |
| *** zakora has quit IRC | 16:56 | |
| *** belmoreira has joined #openstack-nova | 16:57 | |
| *** jistr has quit IRC | 16:57 | |
| *** ijw has joined #openstack-nova | 16:58 | |
| *** mgoddard_ has joined #openstack-nova | 16:58 | |
| *** mgoddard has quit IRC | 16:58 | |
| *** e0ne has quit IRC | 16:59 | |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: Add test coverage to functional api tests _compare_result method https://review.openstack.org/263440 | 17:00 |
| *** haomaiwang has quit IRC | 17:01 | |
| *** Apoorva has joined #openstack-nova | 17:01 | |
| *** moshele has quit IRC | 17:01 | |
| *** 7F1AA87I3 has joined #openstack-nova | 17:01 | |
| *** yamahata has joined #openstack-nova | 17:02 | |
| *** dims has quit IRC | 17:04 | |
| *** shyama has quit IRC | 17:05 | |
| *** shyama has joined #openstack-nova | 17:05 | |
| *** dims has joined #openstack-nova | 17:06 | |
| *** zenoway has joined #openstack-nova | 17:08 | |
| *** jlanoux has quit IRC | 17:08 | |
| *** ihrachys has quit IRC | 17:10 | |
| *** dims has quit IRC | 17:12 | |
| *** prometheanfire has joined #openstack-nova | 17:12 | |
| prometheanfire | there docs about adding the nova-api db? | 17:12 |
| *** zenoway has quit IRC | 17:12 | |
| *** sinese_ has quit IRC | 17:13 | |
| *** Swami__ has quit IRC | 17:13 | |
| auggy | sdague: sorry i didn't realize you had +2'ed my change, i fixed some whitespace issues and added some comments per lxsli_web 's feedback | 17:15 |
| auggy | sdague: https://review.openstack.org/263440 | 17:15 |
| *** mvk_ has quit IRC | 17:16 | |
| *** rdopiera has quit IRC | 17:17 | |
| *** dims has joined #openstack-nova | 17:18 | |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Make _poll_for_status more user-friendly https://review.openstack.org/272459 | 17:20 |
| *** eglynn has quit IRC | 17:21 | |
| *** danpb has quit IRC | 17:25 | |
| *** jgriffith is now known as jgriffith_away | 17:25 | |
| *** avishay has quit IRC | 17:26 | |
| *** achanda has joined #openstack-nova | 17:27 | |
| *** toure has quit IRC | 17:29 | |
| *** chhavi has quit IRC | 17:30 | |
| *** toure has joined #openstack-nova | 17:30 | |
| *** zenoway has joined #openstack-nova | 17:32 | |
| *** toure_ has joined #openstack-nova | 17:32 | |
| *** unicell has quit IRC | 17:36 | |
| *** zenoway has quit IRC | 17:36 | |
| *** toure_ has quit IRC | 17:38 | |
| *** jgriffith_away is now known as jgriffith | 17:39 | |
| *** Swami has joined #openstack-nova | 17:39 | |
| openstackgerrit | Devananda van der Veen proposed openstack/nova: Update driver support matrix for Ironic https://review.openstack.org/275860 | 17:40 |
| openstackgerrit | Devananda van der Veen proposed openstack/nova: Update driver matrix for Ironic network integration https://review.openstack.org/275861 | 17:40 |
| *** zenoway has joined #openstack-nova | 17:41 | |
| *** sfinucan has quit IRC | 17:41 | |
| *** achanda has quit IRC | 17:42 | |
| *** speller has joined #openstack-nova | 17:42 | |
| *** ljxiash has quit IRC | 17:43 | |
| *** vilobhmm11 has joined #openstack-nova | 17:43 | |
| openstackgerrit | Devananda van der Veen proposed openstack/nova: [WIP] Update driver matrix for Ironic network integration https://review.openstack.org/275861 | 17:43 |
| *** wyang has quit IRC | 17:44 | |
| *** vilobhmm11 has quit IRC | 17:44 | |
| *** belmoreira has quit IRC | 17:45 | |
| *** shyama has quit IRC | 17:45 | |
| bauzas | prometheanfire: yup in the release notes | 17:45 |
| *** jaosorior has quit IRC | 17:46 | |
| bauzas | but those were probably trampled by a reno bug when we cut M-2 | 17:46 |
| *** vilobhmm11 has joined #openstack-nova | 17:46 | |
| alaski | it's also in the patch that requires the api db | 17:46 |
| *** yamahata has quit IRC | 17:46 | |
| bauzas | prometheanfire: https://github.com/openstack/nova/blob/master/releasenotes/notes/api-database-now-required-6245f39d36885d1c.yaml and https://github.com/openstack/nova/blob/master/releasenotes/notes/request-spec-api-db-b9cc6e0624d563c5.yaml | 17:47 |
| bauzas | we can only see the latter until https://bugs.launchpad.net/reno/+bug/1537451 is fixed and released | 17:47 |
| openstack | Launchpad bug 1537451 in reno "notes missing/overridden in latest version" [High,Fix released] - Assigned to Doug Hellmann (doug-hellmann) | 17:47 |
| bauzas | (speaking of http://docs.openstack.org/releasenotes/nova/unreleased.html ) | 17:48 |
| dhellmann | bauzas : the fix for that is merged, and I'm working on the release today | 17:49 |
| johnthetubaguy | bauzas: grr, how did the The disk_weight_multiplier option is now available. bit get in there again | 17:51 |
| bauzas | dhellmann: oh, I missed that when copy/pasting again | 17:51 |
| bauzas | dhellmann: excellent news, thanks | 17:51 |
| bauzas | johnthetubaguy: wat ? | 17:51 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: neutron: Update the port with a MAC address for PFs https://review.openstack.org/262586 | 17:52 |
| * bauzas looking | 17:52 | |
| johnthetubaguy | bauzas: was looking in the unreleased notes, we have a stray | 17:52 |
| bauzas | sec | 17:52 |
| openstackgerrit | Andrea Rosa proposed openstack/nova: Enable volume operations for shelved instances https://review.openstack.org/265898 | 17:52 |
| bauzas | with the increase of reno files, using git grep is my friend | 17:53 |
| dhellmann | johnthetubaguy : should that not be there at all, or is it in the wrong version? | 17:53 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Add new configuration option to turn auto converge on/off https://review.openstack.org/266001 | 17:53 |
| bauzas | dhellmann: I guess johnthetubaguy is sad because it's a prelude section | 17:54 |
| bauzas | http://docs.openstack.org/releasenotes/nova/unreleased.html | 17:54 |
| bauzas | johnthetubaguy: https://github.com/openstack/nova/commit/13c1301f41edee6559220f2dfa92211714925f8d | 17:54 |
| bauzas | here is the thing | 17:54 |
| bauzas | yeah https://review.openstack.org/#/c/261108/6/releasenotes/notes/disk-weight-scheduler-98647f9c6317d21d.yaml | 17:55 |
| *** ssurana has quit IRC | 17:55 | |
| johnthetubaguy | dhellmann: yeah, sorry, that bit is totally not your fault, its just stuff we shouldn't have approved | 17:55 |
| johnthetubaguy | bauzas: yeah, I figured its something I approved | 17:56 |
| *** Daisy has joined #openstack-nova | 17:56 | |
| bauzas | johnthetubaguy: lemme fix that | 17:56 |
| *** wyang has joined #openstack-nova | 17:56 | |
| dhellmann | johnthetubaguy : ah, ok. let me know if you spot other errors with content that you think are related to bugs in reno. Some of these git log commands are subtle. | 17:56 |
| johnthetubaguy | bauzas: OK , I was just going to do that, but thank you | 17:56 |
| johnthetubaguy | dhellmann: no worries, will do, I think bauzas is going through the big ones with you now | 17:57 |
| dhellmann | johnthetubaguy, bauzas: yep | 17:57 |
| bauzas | dhellmann: nothing really catched yet apart that one | 17:57 |
| bauzas | I mean the milestone cut | 17:57 |
| *** lucasagomes has quit IRC | 17:58 | |
| bauzas | what's unfortunate is that this change is yet another one I was having in scope but with no time to review it :( | 17:58 |
| *** bkumar has joined #openstack-nova | 17:58 | |
| *** nic has joined #openstack-nova | 17:59 | |
| *** Sukhdev has quit IRC | 17:59 | |
| *** su_zhang has joined #openstack-nova | 17:59 | |
| *** Daisy has quit IRC | 18:00 | |
| PaulMurray | dansmith, the volume ops patch has been updated chages re your and sdague comments: https://review.openstack.org/#/c/265898 | 18:00 |
| PaulMurray | dansmith, also the tempest patch is here: https://review.openstack.org/#/c/258391 | 18:00 |
| *** 7F1AA87I3 has quit IRC | 18:01 | |
| *** ndipanov has quit IRC | 18:01 | |
| *** haomaiwang has joined #openstack-nova | 18:01 | |
| *** su_zhang has quit IRC | 18:01 | |
| sdague | PaulMurray: I don't see the microversion logic in it | 18:01 |
| dansmith | PaulMurray: that patch tells me not to review it | 18:02 |
| PaulMurray | sdague, looking.... | 18:02 |
| sdague | oh, it's a layer down | 18:02 |
| *** su_zhang has joined #openstack-nova | 18:02 | |
| sdague | no, wait, it's not right | 18:03 |
| sdague | oh, sorry... I get it now. I didn't quite realize the entry point | 18:04 |
| *** sinese has joined #openstack-nova | 18:04 | |
| *** lucasagomes has joined #openstack-nova | 18:04 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Remove prelude from disk-weight-sch reno https://review.openstack.org/275871 | 18:04 |
| bauzas | johnthetubaguy: ^ | 18:05 |
| * bauzas heads to dinner | 18:05 | |
| *** mpavone has quit IRC | 18:05 | |
| johnthetubaguy | bauzas: thanks | 18:05 |
| johnthetubaguy | I should go get dinner too, good point | 18:05 |
| *** gszasz has quit IRC | 18:05 | |
| *** bnemec has quit IRC | 18:05 | |
| bauzas | I'm already 30 mins late :p | 18:05 |
| *** speller has quit IRC | 18:06 | |
| *** su_zhang has quit IRC | 18:07 | |
| *** su_zhang_ has joined #openstack-nova | 18:07 | |
| *** vilobhmm11 has quit IRC | 18:08 | |
| *** bkumar has quit IRC | 18:09 | |
| *** liverpooler has joined #openstack-nova | 18:14 | |
| *** liverpooler has quit IRC | 18:15 | |
| *** liverpooler has joined #openstack-nova | 18:15 | |
| *** rook-laptop has quit IRC | 18:16 | |
| *** Apoorva is now known as apoorvad | 18:18 | |
| *** vilobhmm11 has joined #openstack-nova | 18:19 | |
| PaulMurray | dansmith, I put a comment about the title - please do check it out. I will get jlanaux to change it. | 18:20 |
| PaulMurray | sdague, BTW - jlanoux put the to detach/attach tests into one test to avoid the extra unnecessary boot and shelve. I think it may need tweaking. | 18:22 |
| *** rk4n has quit IRC | 18:23 | |
| *** _ducttape_ has joined #openstack-nova | 18:24 | |
| *** e0ne has joined #openstack-nova | 18:24 | |
| prometheanfire | bauzas: thanks | 18:25 |
| *** Jeffrey4l has quit IRC | 18:25 | |
| prometheanfire | bauzas: db_sync will create the DB as well? | 18:26 |
| *** ducttape_ has quit IRC | 18:27 | |
| prometheanfire | ah, you do have to define the connection string, see it now | 18:29 |
| *** achanda has joined #openstack-nova | 18:29 | |
| *** rotbeard has quit IRC | 18:31 | |
| *** unicell has joined #openstack-nova | 18:33 | |
| *** jang has quit IRC | 18:34 | |
| *** bkumar has joined #openstack-nova | 18:36 | |
| *** unicell1 has joined #openstack-nova | 18:36 | |
| *** rook-laptop has joined #openstack-nova | 18:37 | |
| *** unicell1 has quit IRC | 18:37 | |
| *** unicell has quit IRC | 18:37 | |
| *** unicell1 has joined #openstack-nova | 18:37 | |
| *** ctrath has quit IRC | 18:38 | |
| *** unicell1 has quit IRC | 18:39 | |
| *** unicell has joined #openstack-nova | 18:39 | |
| *** PaulMurray has quit IRC | 18:39 | |
| *** eil397 has joined #openstack-nova | 18:39 | |
| *** tdurakov has joined #openstack-nova | 18:39 | |
| *** su_zhang_ has quit IRC | 18:39 | |
| *** su_zhang has joined #openstack-nova | 18:40 | |
| *** mvk_ has joined #openstack-nova | 18:41 | |
| tdurakov | dansmith, alaski, hi, if you got some time could you please review this patch again https://review.openstack.org/#/c/249110/ | 18:42 |
| *** bkumar has quit IRC | 18:42 | |
| *** ctrath has joined #openstack-nova | 18:43 | |
| *** yamahata has joined #openstack-nova | 18:43 | |
| *** doug-fish has quit IRC | 18:45 | |
| *** ssurana has joined #openstack-nova | 18:46 | |
| alaski | tdurakov: sure, will get to it in a bit | 18:47 |
| *** liverpooler has quit IRC | 18:47 | |
| *** tdurakov has quit IRC | 18:47 | |
| prometheanfire | bauzas: the charset option given only works for mysqltypes | 18:48 |
| *** vilobhmm111 has joined #openstack-nova | 18:49 | |
| *** burt has quit IRC | 18:51 | |
| *** vilobhmm11 has quit IRC | 18:51 | |
| *** tcammann has joined #openstack-nova | 18:51 | |
| *** deva_ has joined #openstack-nova | 18:51 | |
| *** browne has joined #openstack-nova | 18:52 | |
| *** jgriffith is now known as jgriffith_away | 18:53 | |
| *** mc_nair_ has joined #openstack-nova | 18:55 | |
| *** mc_nair_ is now known as mc_nair | 18:56 | |
| *** vilobhmm111 has quit IRC | 18:57 | |
| *** jgriffith_away is now known as jgriffith | 18:57 | |
| *** vilobhmm11 has joined #openstack-nova | 18:57 | |
| *** tcammann has quit IRC | 18:57 | |
| *** liverpooler has joined #openstack-nova | 18:58 | |
| *** rlrossit has quit IRC | 18:59 | |
| *** gjayavelu has joined #openstack-nova | 19:00 | |
| *** deva_ has quit IRC | 19:00 | |
| *** haomaiwang has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-nova | 19:01 | |
| openstackgerrit | Pushkar Umaranikar proposed openstack/nova: Config options: Centralize debugger options https://review.openstack.org/274298 | 19:03 |
| *** browne has quit IRC | 19:03 | |
| *** browne has joined #openstack-nova | 19:05 | |
| *** _ducttape_ has quit IRC | 19:05 | |
| *** mylu has joined #openstack-nova | 19:06 | |
| *** ssurana1 has joined #openstack-nova | 19:06 | |
| *** ducttape_ has joined #openstack-nova | 19:06 | |
| *** penick has joined #openstack-nova | 19:07 | |
| *** ssurana has quit IRC | 19:08 | |
| *** rook-laptop has quit IRC | 19:12 | |
| *** Sukhdev has joined #openstack-nova | 19:14 | |
| *** deva_ has joined #openstack-nova | 19:16 | |
| *** ccarmack has joined #openstack-nova | 19:17 | |
| *** achanda has quit IRC | 19:17 | |
| *** jgriffith is now known as jgriffith_away | 19:18 | |
| ccarmack | Hi, Can someone take a quick look at https://bugs.launchpad.net/nova/+bug/1540939 and maybe explain why nova was written that way? | 19:18 |
| openstack | Launchpad bug 1540939 in OpenStack Compute (nova) "Instance delete causing port leak" [Undecided,New] | 19:18 |
| *** bklei has quit IRC | 19:20 | |
| *** bklei has joined #openstack-nova | 19:21 | |
| *** gjayavelu has quit IRC | 19:21 | |
| alaski | ccarmack: the comment was added by garyk so you might try to catch him when he's on sometime | 19:22 |
| ccarmack | alaski: ok, thanks | 19:22 |
| alaski | ccarmack: just looking at it though, if the admin creds are needed it may have just been missed on the delete side | 19:23 |
| *** deva_ has quit IRC | 19:23 | |
| *** bklei has quit IRC | 19:24 | |
| ccarmack | alaski: I was thinking to either remove the admin cred on the create side, or add the admin on the delete side | 19:24 |
| ccarmack | …but I don't know the history | 19:24 |
| alaski | me neither... | 19:24 |
| ccarmack | We have a set up where the nova policy allows "operator" to create/delete an instance, but requires admin to create/delete ports | 19:25 |
| *** bklei has joined #openstack-nova | 19:28 | |
| *** jgriffith_away is now known as jgriffith | 19:29 | |
| *** tdurakov has joined #openstack-nova | 19:29 | |
| alaski | I've struggled with setting up asymmetrical policies like that between nova and glance before. I've wondered if keystone trusts would help, but haven't learned enough to know if they address that case | 19:31 |
| *** su_zhang has quit IRC | 19:32 | |
| *** su_zhang has joined #openstack-nova | 19:33 | |
| *** su_zhang has quit IRC | 19:33 | |
| *** su_zhang has joined #openstack-nova | 19:34 | |
| doffm | ccarmack: Its been in since the dawn of neutron. Might no longer be needed. I'm just going to put up a patch removing it to see what happens. :/ | 19:34 |
| *** e0ne has quit IRC | 19:35 | |
| *** achanda has joined #openstack-nova | 19:35 | |
| ccarmack | thanks doffm | 19:35 |
| *** jichen has joined #openstack-nova | 19:37 | |
| melwitt | yeah, I think it makes sense that nova should be able to create/delete ports but that a normal user shouldn't be able to. and so far that means nova has to present admin credentials. and I don't understand why it wouldn't be admin on the delete side | 19:37 |
| *** ihrachys has joined #openstack-nova | 19:37 | |
| *** liverpooler has quit IRC | 19:37 | |
| *** rook-laptop has joined #openstack-nova | 19:37 | |
| *** liverpooler has joined #openstack-nova | 19:38 | |
| *** daemontool has joined #openstack-nova | 19:38 | |
| *** bklei has quit IRC | 19:38 | |
| ccarmack | thanks melwitt, I'll upload a patch adding the admin check to the delete | 19:39 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Use instance in setup_networks_on_host https://review.openstack.org/249110 | 19:39 |
| *** su_zhang has quit IRC | 19:40 | |
| melwitt | ccarmack: does it only leak ports if the port binding extension is in use? I didn't understand that part in the lp bug | 19:41 |
| *** ducttape_ has quit IRC | 19:41 | |
| * melwitt is looking at the code links in the bug now | 19:41 | |
| ccarmack | melwitt: yes, because nova calls neutron to check if the port binding is in use, and if so, passes admin creds to port create | 19:41 |
| melwitt | ccarmack: I should have looked at your links before asking. yeah | 19:42 |
| ccarmack | That code dates back to 2013, apparently when someone switched from calling quasar (?) to calling neutron | 19:42 |
| *** achanda has quit IRC | 19:43 | |
| ccarmack | melwitt, sorry quantum, not quasar | 19:43 |
| melwitt | quasar, I like that. it was quantum :) | 19:43 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Use instance in setup_networks_on_host https://review.openstack.org/249110 | 19:43 |
| doffm | ccarmack: Yes, was added in the 7de916f5 commit, didn't exist in the quantum code. | 19:43 |
| *** boris-42 has quit IRC | 19:43 | |
| *** Sukhdev has quit IRC | 19:44 | |
| *** bklei has joined #openstack-nova | 19:44 | |
| melwitt | ccarmack: from what I see there, I agree with you the same logic should be on the delete side where it checks for the port binding extension and use admin or not accordingly | 19:44 |
| *** bklei has quit IRC | 19:44 | |
| ccarmack | melwitt: I'm trying to track down gongysh, who committed that | 19:44 |
| melwitt | I don't know all the history either but I don't see why it shouldn't be matching | 19:44 |
| *** su_zhang has joined #openstack-nova | 19:46 | |
| melwitt | you might also ask our neutron liason sc68cal, he could probably shed some light on this | 19:47 |
| *** ociuhandu has quit IRC | 19:47 | |
| ccarmack | ah, thanks | 19:47 |
| sc68cal | hello | 19:47 |
| melwitt | hi :) | 19:48 |
| sc68cal | is this the port binding thing? we are discussing this in -neutron | 19:48 |
| melwitt | it's about port binding and admin vs not creds | 19:48 |
| melwitt | for nova calling neutron | 19:48 |
| ccarmack | https://bugs.launchpad.net/nova/+bug/1540939 | 19:49 |
| openstack | Launchpad bug 1540939 in OpenStack Compute (nova) "Instance delete causing port leak" [Undecided,New] | 19:49 |
| melwitt | so yeah, same thing. doffm is over tehre | 19:49 |
| sc68cal | ok, shall we continue this in -neutron? | 19:49 |
| ccarmack | yea | 19:49 |
| *** mgarza__ has joined #openstack-nova | 19:50 | |
| openstackgerrit | Mark Doffman proposed openstack/nova: Make os-instance-action read deleted instances. https://review.openstack.org/253087 | 19:51 |
| *** daemontool has quit IRC | 19:51 | |
| *** paul-carlton1 has quit IRC | 19:51 | |
| *** liverpooler has quit IRC | 19:51 | |
| *** daemontool has joined #openstack-nova | 19:53 | |
| *** mgarza_ has quit IRC | 19:53 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Make use of new Console and ConsolePool objects https://review.openstack.org/267540 | 19:53 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Add Console and ConsolePool objects https://review.openstack.org/267539 | 19:53 |
| *** ducttape_ has joined #openstack-nova | 19:54 | |
| *** mfedosin has quit IRC | 19:55 | |
| *** vilobhmm11 has quit IRC | 19:55 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove and deprecate conductor provider_fw_rule_get_all() https://review.openstack.org/216241 | 19:57 |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Remove provider_fw_rules tables and associated db api methods https://review.openstack.org/216242 | 19:57 |
| *** diana_clarke has quit IRC | 19:57 | |
| *** achanda has joined #openstack-nova | 19:59 | |
| *** baoli_ has quit IRC | 19:59 | |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Do not catch exceptions using Python2 sintax https://review.openstack.org/275185 | 20:01 |
| *** haomaiwang has quit IRC | 20:01 | |
| *** 7JTAAES9V has joined #openstack-nova | 20:01 | |
| *** rlrossit has joined #openstack-nova | 20:01 | |
| *** diana_clarke has joined #openstack-nova | 20:02 | |
| *** ssurana1 has quit IRC | 20:03 | |
| *** gtt116 has joined #openstack-nova | 20:06 | |
| *** gtt116_ has quit IRC | 20:09 | |
| *** ihrachys has quit IRC | 20:10 | |
| *** rlrossit has quit IRC | 20:12 | |
| *** ccarmack has left #openstack-nova | 20:14 | |
| melwitt | dansmith: I'm reviewing https://review.openstack.org/#/c/274870/ and I was wondering if this is one where we could see a difference in lazy-loads in n-cpu logs? | 20:14 |
| dansmith | melwitt: you mean you want to see some lazy-load that doesn't happen in these logs? | 20:17 |
| *** mgarza__ has quit IRC | 20:19 | |
| melwitt | dansmith: with the heal info cache patch when I looked in the logs I saw a drop in the number of 'lazy-loading ...' messages and I was wondering if this patch is one where we could see that too. I wasn't sure since it has to do with migrations | 20:19 |
| dansmith | melwitt: yeah, you should | 20:19 |
| dansmith | it's easier to see them with the DNM patch below it instead of above, but hang on | 20:19 |
| *** rlrossit_ has joined #openstack-nova | 20:20 | |
| melwitt | dansmith: on the surface the count of 'lazy-loading ...' messages is higher than the heal info cache patch, which I don't understand since this patch depends on that one | 20:21 |
| dansmith | melwitt: it is? my browser counts 10 migration_context loads in the one below and 8 in this one | 20:22 |
| *** mylu has quit IRC | 20:23 | |
| *** vilobhmm11 has joined #openstack-nova | 20:23 | |
| dansmith | melwitt: so if you look, there are still 3 migration_context lazy loads by the end of the set: http://logs.openstack.org/14/274714/7/check/gate-tempest-dsvm-full/f19c7a6/logs/screen-n-cpu.txt.gz | 20:24 |
| dansmith | but they're in paths I don't address set | 20:25 |
| dansmith | *yet | 20:25 |
| edleafe | is there an API command for migrating a VM to a specific host, rather than letting the scheduler choose it? | 20:25 |
| dansmith | so between the bottom and the top, we go from 10 to 3 by my count | 20:25 |
| *** mylu has joined #openstack-nova | 20:26 | |
| melwitt | dansmith: I was looking at the count of 'lazy-loading' in general. the migration_context specific ones were close, for whatever reason my browser shows 10 on the patch and 8 on the one below | 20:26 |
| dansmith | melwitt: you sure you're not reversing those? | 20:27 |
| dansmith | another way to look at it is to see the ones that we were hitting in the first run of that dnm patch, compared to the current one | 20:27 |
| *** Sukhdev has joined #openstack-nova | 20:28 | |
| dansmith | melwitt: this pattern: http://logs.openstack.org/14/274714/2/check/gate-tempest-dsvm-full/603f85c/logs/screen-n-cpu.txt.gz#_2016-02-01_17_46_34_135 | 20:28 |
| melwitt | dansmith: what about the total lazy load count? basically to check if some unrelated lazy-loads might have come out of the change | 20:29 |
| dansmith | melwitt: which I think you won't find in this latest one: http://logs.openstack.org/14/274714/7/check/gate-tempest-dsvm-full/f19c7a6/logs/screen-n-cpu.txt.gz | 20:29 |
| *** sahid has quit IRC | 20:29 | |
| dansmith | so the total amount can vary depending on when our periodics run | 20:30 |
| dansmith | so if a periodic happens to run when we have a few instances in active state, then we'll trigger a bunch of flavor things | 20:30 |
| dansmith | if we don't then we'll trigger fewer | 20:30 |
| *** vilobhmm11 has quit IRC | 20:30 | |
| melwitt | okay, that's the main thing I was wondering about | 20:30 |
| *** tdurakov has quit IRC | 20:31 | |
| dansmith | melwitt: % grep 'Lazy-loading' screen-n-cpu.txt-before screen-n-cpu.txt.gz -c | 20:31 |
| dansmith | screen-n-cpu.txt-before:624 | 20:31 |
| dansmith | screen-n-cpu.txt.gz:395 | 20:31 |
| melwitt | I thought maybe on average they were consistent but I guess they aren't | 20:31 |
| dansmith | we definitely don't vary by ~200 per run :) | 20:31 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Test that new models don't use SoftDeleteMixin https://review.openstack.org/275912 | 20:32 |
| openstackgerrit | Dina Belova proposed openstack/nova: Integrate OSprofiler and Nova https://review.openstack.org/254703 | 20:32 |
| melwitt | dansmith: I think I got my tabs mixed up, sorry about that | 20:34 |
| dansmith | melwitt: it's cool | 20:35 |
| dansmith | it's too confusing, so I've been pulling down the logs and grepping | 20:35 |
| tjones_zzzz | edleafe: did you get an answer on migrating to a specific host? | 20:36 |
| melwitt | dansmith: yeah, good call :P | 20:36 |
| edleafe | tjones_zzzz: no | 20:37 |
| *** prometheanfire has left #openstack-nova | 20:37 | |
| tjones_zzzz | edleafe: oh cause i was wondering too. | 20:37 |
| edleafe | tjones_zzzz: I assume admin-only, of course | 20:37 |
| tjones_zzzz | edleafe: yeah i would assume so but im not sure there is an api to specify the host at all | 20:38 |
| *** su_zhang has quit IRC | 20:40 | |
| bauzas | tjones_zzzz: what is the question about that ? | 20:41 |
| tjones_zzzz | edleafe: was asking is there an API command for migrating a VM to a specific host, rather than letting the scheduler choose it? and i wondered too | 20:42 |
| bauzas | live-migrating ? | 20:43 |
| bauzas | evacuating ? | 20:43 |
| edleafe | live | 20:43 |
| bauzas | tbc, yes, providing a destination host for live-migrating just bypasses the scheduler | 20:43 |
| bauzas | well, until my lovely BP is implemented and merged | 20:44 |
| tjones_zzzz | very nice - i have some customers asking for that. do you have a link handy? | 20:44 |
| edleafe | bauzas: but even then, check destination will only forbid a host, not select one, right? | 20:44 |
| bauzas | tjones_zzzz: not really, it's secret sauce | 20:44 |
| edleafe | bauzas: and is that an API-level call? I don't see it anywhere in the docs | 20:45 |
| bauzas | sec | 20:45 |
| dansmith | edleafe: really? you didn't even run nova help live-migration ? :) | 20:45 |
| edleafe | dansmith: not working as a CLI | 20:45 |
| edleafe | dansmith: looking at http://developer.openstack.org/api-ref-compute-v2.1.html | 20:45 |
| bauzas | sooo | 20:46 |
| bauzas | the CLI just says that you can optionally provide an host | 20:46 |
| bauzas | tjones_zzzz: http://developer.openstack.org/api-ref-compute-v2.1.html#migrateLive | 20:46 |
| sdague | as far as I remember live only | 20:47 |
| bauzas | sdague: you're right | 20:47 |
| bauzas | only for live-migrate and evacuate | 20:47 |
| tjones_zzzz | bauzas: thanks | 20:47 |
| bauzas | not for a resize | 20:47 |
| *** vilobhmm11 has joined #openstack-nova | 20:47 | |
| edleafe | bauzas: thanks - that's the API call I was looking for | 20:48 |
| bauzas | but the doc doesn't say that if you provide an host, then you bypass the sched | 20:48 |
| sdague | right, because normal users are not expected to know the topology | 20:48 |
| bauzas | exactly | 20:48 |
| bauzas | what sdague said | 20:48 |
| *** ssurana has joined #openstack-nova | 20:48 | |
| *** su_zhang has joined #openstack-nova | 20:48 | |
| *** mylu has quit IRC | 20:49 | |
| *** mylu has joined #openstack-nova | 20:49 | |
| bauzas | tjones_zzzz: edleafe: so, sec, I'll give you the recipe for the secret sauce | 20:49 |
| bauzas | tjones_zzzz: edleafe: https://github.com/openstack/nova/blob/master/nova/conductor/tasks/live_migrate.py#L57-L62 | 20:50 |
| edleafe | bauzas: long term, when request specs are persisted and check destination is implemented, is it reasonable to add a call to validate potential target hosts as valid targets? | 20:50 |
| bauzas | ^ here is the secret sauce that I want to modify | 20:50 |
| bauzas | edleafe: http://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/check-destination-on-migrations.html is what I'd love to see | 20:50 |
| edleafe | bauzas: me too | 20:51 |
| bauzas | I have a WIP that I need to work on https://review.openstack.org/#/c/273104/ | 20:51 |
| bauzas | I hope to propose it by the next week, so it would mean that for Mitaka, evacuate and live-migrate would at least check the existing Spec if there | 20:51 |
| tjones_zzzz | bauzas: yes that is great! | 20:52 |
| edleafe | bauzas: but in addition, could you see an API call that accepts an instance and a list of potential hosts, and returns info if that instance can be migrated to that host? | 20:52 |
| bauzas | edleafe: I'd not love that | 20:53 |
| *** apoorvad has quit IRC | 20:53 | |
| bauzas | edleafe: because it means that you could get some computes for one call | 20:53 |
| bauzas | and then, 2 secs later, those computes could no longer be accepting new instances | 20:53 |
| edleafe | bauzas: sure, but it would be no racier than anything else | 20:54 |
| edleafe | bauzas: it would be optimistic, of course | 20:54 |
| bauzas | instead, I'd prefer to propose an admin API resource for getting the resources, aka. cburgess_'s spec "tell me my resources, scheduler" | 20:54 |
| sdague | bauzas: ++ | 20:54 |
| edleafe | bauzas: I was more concerned with affinity, pci, etc., that would be constraints in the original reqspec | 20:55 |
| cburgess_ | bauzas ++++++ | 20:55 |
| sdague | I think expanding the external specification of targets until the "resource planner" is in seems weird | 20:55 |
| *** burt has joined #openstack-nova | 20:55 | |
| bauzas | edleafe: since we're cloud, you can spin off an instance and kill it, it would be like a dryrun | 20:55 |
| *** dims has quit IRC | 20:56 | |
| *** acabot_ has joined #openstack-nova | 20:56 | |
| edleafe | bauzas: providing a list of resources is one thing, but checking for anti/affinity stuff is another matter | 20:56 |
| edleafe | bauzas: a dry run won't really be a good model for a migration | 20:57 |
| bauzas | ah, for migrations | 20:57 |
| edleafe | bauzas: here is the use case: | 20:58 |
| edleafe | bauzas: an external system wants to optimize usage in the cloud | 20:58 |
| jed56 | :) | 20:58 |
| bauzas | oh oh, Watcher midcycle, eh ? | 20:58 |
| edleafe | bauzas: it wants to move some vms to other hosts | 20:58 |
| jed56 | yes | 20:58 |
| edleafe | bauzas: exactly | 20:58 |
| jed56 | hello bauzas | 20:59 |
| bauzas | I have a magic bowl for knowing what you guys do | 20:59 |
| bauzas | so | 20:59 |
| edleafe | bauzas: so it selects the best hosts from an optimization POV, but needs to make sure that the hosts can accept those instances | 20:59 |
| edleafe | bauzas: trying to avoid the try/fail cycle | 21:00 |
| bauzas | edleafe: it's more an alarm/trigger situation right? | 21:00 |
| edleafe | bauzas: yes, that's one path for watcher | 21:00 |
| bauzas | edleafe: where that external system would detect a policy violation and then ask to move the instance | 21:00 |
| *** 7JTAAES9V has quit IRC | 21:01 | |
| edleafe | bauzas: yes, or any of a bunch of other causes for the move | 21:01 |
| bauzas | edleafe: in that case, it's a reactive approach where moving the instance would be checked by the scheduler | 21:01 |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance https://review.openstack.org/275922 | 21:01 |
| *** haomaiwa_ has joined #openstack-nova | 21:01 | |
| edleafe | bauzas: that's the try/fail approach we're trying to avoid | 21:01 |
| bauzas | but when you'd be moving, you'be nearly sure that either you have capacity for that, or it would fail | 21:01 |
| bauzas | edleafe: mmm, what if the external system was knowing the resources then ? | 21:02 |
| edleafe | bauzas: capacity is one thing. My concern would be the extra_specs stuff that would be persisted in the req spec | 21:02 |
| bauzas | I see | 21:02 |
| bauzas | but that would only be solved for each instance, right? | 21:03 |
| bauzas | like say that I have a stupid rule saying that I want to have an affinity with a specific existing uuid | 21:03 |
| bauzas | instance uuid | 21:03 |
| *** smatzek has quit IRC | 21:04 | |
| bauzas | in that case, moving that instance having the --same_host flag would always fail unless you migrate the dependent instance, right? | 21:04 |
| edleafe | bauzas: yup | 21:04 |
| edleafe | bauzas: that's a great example | 21:05 |
| edleafe | bauzas: using the try/fail approach, we could potentially cycle throough every host | 21:05 |
| *** edtubill has quit IRC | 21:05 | |
| *** su_zhang has quit IRC | 21:05 | |
| bauzas | mmm | 21:05 |
| *** edtubill has joined #openstack-nova | 21:05 | |
| *** dims has joined #openstack-nova | 21:05 | |
| *** kurtmartin has quit IRC | 21:06 | |
| bauzas | (considering the problem) | 21:06 |
| bauzas | the problem I see is that it is per-instance | 21:07 |
| *** mylu has quit IRC | 21:07 | |
| *** mylu has joined #openstack-nova | 21:07 | |
| *** baoli has joined #openstack-nova | 21:07 | |
| edleafe | bauzas: why is that a problem? | 21:08 |
| bauzas | if you want to proactively check if it's possible to live-migrate all the host, you need to verify that for every instance of the host | 21:09 |
| bauzas | which is bummers | 21:09 |
| edleafe | bauzas: ah, I see | 21:09 |
| *** ccarmack has joined #openstack-nova | 21:09 | |
| edleafe | bauzas: what we're looking at is more of a single instance at a time thing | 21:09 |
| edleafe | bauzas: so here's what I think would work: | 21:10 |
| edleafe | bauzas: watcher comes up with a list on N acceptable target hosts | 21:10 |
| edleafe | bauzas: it sends that list and the instance uuid to a Nova API | 21:11 |
| *** raildo is now known as raildo-afk | 21:11 | |
| edleafe | bauzas: it gets back a {host1:True, host2:True, host3:False...hostN:False} dict | 21:11 |
| edleafe | bauzas: with T/F being if that host would be acceptable given the persisted req spec | 21:12 |
| edleafe | bauzas: does that seem reasonable? | 21:12 |
| bauzas | like I said, host1 being True is okay for the next 1 secs | 21:12 |
| edleafe | bauzas: that limitation is understood | 21:13 |
| *** apoorvad has joined #openstack-nova | 21:13 | |
| bauzas | edleafe: for Watcher | 21:13 |
| bauzas | edleafe: not for all operators :) | 21:13 |
| doffm | bauzas: edleafe: They way the scheduler is set up its an expensive operation to find if 'host1' is valid. | 21:13 |
| doffm | Since we have to pass it through all the filters? | 21:13 |
| doffm | And same for 'host2'...'host3'. | 21:14 |
| doffm | Correct? :/ | 21:14 |
| doffm | Oh sorry. | 21:14 |
| edleafe | doffm: correct | 21:14 |
| *** clayton has quit IRC | 21:14 | |
| bauzas | doffm: yup but that's not where the scheduler is late | 21:14 |
| *** ajayaa has quit IRC | 21:14 | |
| bauzas | that is quite quick | 21:14 |
| doffm | OK. | 21:14 |
| bauzas | that's rather getting the list of hosts which is expensive | 21:15 |
| *** ajayaa has joined #openstack-nova | 21:15 | |
| *** boris-42 has joined #openstack-nova | 21:15 | |
| bauzas | looping over memory and filtering is quick | 21:15 |
| *** ajayaa has quit IRC | 21:15 | |
| edleafe | doffm: we'd supply a small number of hosts to consider, rather than all hosts. | 21:15 |
| bauzas | still considering the possibility to get all the resources for the hosts | 21:16 |
| edleafe | doffm: the difference is instead of returning a single (host,node) tuple, it would return all passing hsots | 21:16 |
| edleafe | hosts, even | 21:16 |
| bauzas | like GET /scheduler/capacity {'host': 'myhost'} | 21:16 |
| *** ajayaa has joined #openstack-nova | 21:17 | |
| dansmith | omg I might merge a patch today | 21:17 |
| * dansmith is so excited | 21:17 | |
| *** Swami has quit IRC | 21:17 | |
| *** ajayaa has quit IRC | 21:17 | |
| dansmith | from the look of the gate, I thought that was not possible | 21:17 |
| bauzas | dansmith: waaaaat ? | 21:17 |
| doffm | Hooray! <confetti emoji> | 21:17 |
| *** su_zhang has joined #openstack-nova | 21:17 | |
| * edleafe hands dansmith a valium | 21:17 | |
| *** auggy has left #openstack-nova | 21:18 | |
| melwitt | lol | 21:19 |
| bauzas | edleafe: trying to loop over and over on the problem, I don't really see a cool solution | 21:20 |
| bauzas | maybe logging a backlog spec could help ? | 21:21 |
| alaski | a slightly less horrible solution would be to pass in the list of hosts and leave it to the scheduler | 21:21 |
| edleafe | alaski: you mean call live migrate with a list of hosts, instead of just one? | 21:22 |
| alaski | yes | 21:22 |
| bauzas | alaski: you mean accepting a list of values for the host field when live-migrating N? | 21:22 |
| bauzas | snap | 21:22 |
| bauzas | well | 21:22 |
| alaski | I don't love it, but it's better than having a list passed out of the scheduler | 21:22 |
| bauzas | that's just forcing the destination :) | 21:22 |
| *** Swami has joined #openstack-nova | 21:23 | |
| *** auggy has joined #openstack-nova | 21:23 | |
| edleafe | not really | 21:23 |
| edleafe | it's just saying "try A, and if that fails, try B..." | 21:23 |
| alaski | it's saying let the scheduler make the True/False list you want it to expose in the API and then work from that | 21:24 |
| bauzas | not sure I got that | 21:24 |
| *** daemontool_ has joined #openstack-nova | 21:25 | |
| alaski | edleafe asked for a list like {host1:True, host2:True, host3:False...hostN:False}, well a dict. I'm saying rather than return it, just have the scheduler pick one and use it, falling back to others if that fails | 21:25 |
| bauzas | http://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/check-destination-on-migrations.html is accepting a destination as part of the ReqSpec given to the scheduler | 21:25 |
| bauzas | alaski: ah that | 21:25 |
| edleafe | alaski: that could work | 21:26 |
| bauzas | alaski: so yeah, what you propose goes to the same direction than the above BP | 21:26 |
| bauzas | provided that the field we pass to the scheduler is not a string, but a list | 21:26 |
| bauzas | that's even a good point | 21:27 |
| dansmith | *even( | 21:27 |
| dansmith | that's _even_ a good point! :) | 21:27 |
| *** daemontool has quit IRC | 21:27 | |
| alaski | \o/ | 21:27 |
| bauzas | alaski++ | 21:27 |
| bauzas | meh, no karma points | 21:27 |
| dansmith | no, it's just that alaski has infinity points already | 21:28 |
| dansmith | he has NaN points | 21:28 |
| bauzas | oh man | 21:28 |
| bauzas | I'm memory-bounded | 21:28 |
| alaski | I don't know about that. if so it was probably an underflow error | 21:28 |
| *** zul has joined #openstack-nova | 21:29 | |
| bauzas | alaski: so what you propose is just a slight modification of the already approved spec | 21:29 |
| bauzas | edleafe: jed56: that could be a new spec that you would fill in, depending on the one I have | 21:29 |
| openstackgerrit | Mark Sturdevant proposed openstack/nova: Add instance host to evacuate terminate connector https://review.openstack.org/274318 | 21:29 |
| *** tdurakov has joined #openstack-nova | 21:29 | |
| dansmith | 851640 | 21:30 |
| *** ctrath has quit IRC | 21:30 | |
| bauzas | dansmith: 42 | 21:30 |
| dansmith | doh | 21:30 |
| edleafe | bauzas: I think the valium is kicking in... | 21:30 |
| * bauzas is just thinking of a new filter which would always return 42 | 21:30 | |
| edleafe | bauzas: sure, I could propose a spec that builds on that. | 21:31 |
| bauzas | or amending the one I made | 21:31 |
| bauzas | I honestly dunno what would the right format, but I just want to make sure that we all agree on the direction | 21:31 |
| edleafe | bauzas: didn't realize I could amend an accepted spec | 21:31 |
| *** tdurakov has quit IRC | 21:31 | |
| edleafe | bauzas: especially after spec freeze | 21:32 |
| doffm | alaski just did his best superman impression, flew in with a good idea and then he was gone. | 21:33 |
| edleafe | bauzas: thinking about it, it's actually proposing to change the live migration API, not check destinations | 21:33 |
| *** ctrath has joined #openstack-nova | 21:33 | |
| edleafe | bauzas: so yeah, it's a new spec for Newton | 21:33 |
| *** abitha has joined #openstack-nova | 21:33 | |
| bauzas | edleafe: sure, it would be anyway a newton update | 21:34 |
| alaski | doffm: I have a strict limit on number of words I allow myself per day :) also it's harder to find a telephone booth to change in nowadays | 21:34 |
| bauzas | I also need to propose again my spec, since I won't have time to have it implemented by mitaka :( | 21:35 |
| bauzas | edleafe: so, some of you guys could add a dependent spec on the new one I'll propose | 21:35 |
| bauzas | but that's nits | 21:35 |
| edleafe | bauzas: sounds good | 21:35 |
| bauzas | edleafe: FWIW, I'm also planning to modify the live-migrate API resource by adding a force flag | 21:35 |
| *** achanda has quit IRC | 21:35 | |
| edleafe | bauzas: didn't you already propose that? | 21:36 |
| bauzas | edleafe: it's part of the spec, I meant | 21:36 |
| edleafe | bauzas: that's what I remembered from the summit discussion | 21:36 |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: Add test coverage to functional api tests _compare_result method https://review.openstack.org/263440 | 21:39 |
| auggy | sdague: hey sorry to bug you, i need a +2 again on account of angering the pep8 gods | 21:39 |
| *** aysyd has quit IRC | 21:39 | |
| *** daemontool_ has quit IRC | 21:39 | |
| *** edtubill has quit IRC | 21:39 | |
| openstackgerrit | Sean Dague proposed openstack/nova: virt: adjusting the osinfo tests to use fakelibosinfo https://review.openstack.org/275520 | 21:40 |
| *** daemontool_ has joined #openstack-nova | 21:40 | |
| *** zul has quit IRC | 21:40 | |
| *** su_zhang has quit IRC | 21:41 | |
| *** clayton has joined #openstack-nova | 21:43 | |
| *** eglynn has joined #openstack-nova | 21:43 | |
| *** achanda has joined #openstack-nova | 21:43 | |
| *** ijw has quit IRC | 21:46 | |
| *** jistr has joined #openstack-nova | 21:46 | |
| *** jistr has quit IRC | 21:47 | |
| *** edtubill has joined #openstack-nova | 21:50 | |
| *** edtubill has quit IRC | 21:50 | |
| bauzas | melwitt: in case you missed the convo, we would not want to use the prelude section for reno notes, so I fixed that and I need a +W https://review.openstack.org/#/c/275871/ | 21:51 |
| *** mylu has quit IRC | 21:52 | |
| *** mylu has joined #openstack-nova | 21:53 | |
| *** salv-orlando has joined #openstack-nova | 21:54 | |
| sdague | bauzas: +W | 21:54 |
| sdague | I was wondering about that the other day | 21:54 |
| bauzas | cool, ty | 21:54 |
| sdague | I expect we're going to need to spend a day crisping up all the release notes at the end of the cycle | 21:54 |
| sdague | as they are definitely in very different voices | 21:54 |
| sdague | but that's fine, that's at least just an editing job | 21:55 |
| bauzas | sdague: I usually do that at the end of each milestone | 21:55 |
| sdague | bauzas: ok, cool | 21:55 |
| bauzas | sdague: I also have an etherpad | 21:55 |
| bauzas | for tracking what needs a reno file | 21:55 |
| bauzas | https://etherpad.openstack.org/p/nova-milestone-reno | 21:55 |
| *** thumpba has quit IRC | 21:55 | |
| *** mylu has quit IRC | 21:56 | |
| *** mylu has joined #openstack-nova | 21:56 | |
| *** mylu has quit IRC | 21:56 | |
| *** salv-orl_ has quit IRC | 21:56 | |
| *** edtubill has joined #openstack-nova | 21:57 | |
| *** haomaiwa_ has quit IRC | 22:01 | |
| *** haomaiwang has joined #openstack-nova | 22:01 | |
| *** penick has quit IRC | 22:02 | |
| *** su_zhang has joined #openstack-nova | 22:04 | |
| auggy | hey folks i have a bug triaging question regarding https://bugs.launchpad.net/nova/+bug/1475786 | 22:05 |
| openstack | Launchpad bug 1475786 in OpenStack Compute (nova) "Cannot ping to a same subnet VM via floating IP" [Undecided,New] | 22:05 |
| *** thorst_ has quit IRC | 22:05 | |
| auggy | before i get too deep in attempting to repro it, how much are we actually supporting nova-network related things? | 22:05 |
| *** PsionTheory has joined #openstack-nova | 22:05 | |
| auggy | like should i push the reporter to try using neutron instead and then come back if they still have a problem? | 22:05 |
| *** thorst_ has joined #openstack-nova | 22:06 | |
| *** erlarese has quit IRC | 22:07 | |
| *** thorst__ has joined #openstack-nova | 22:08 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Database not needed for test class: PciGetInstanceDevs https://review.openstack.org/275945 | 22:08 |
| auggy | i'm also still figuring out the neutron/nova-network boundary so pardon if my question is a little naive ;) | 22:09 |
| *** penick has joined #openstack-nova | 22:09 | |
| *** thorst_ has quit IRC | 22:09 | |
| melwitt | bauzas: thanks, I didn't know that about prelude | 22:11 |
| *** ducttape_ has quit IRC | 22:11 | |
| bauzas | no worries | 22:11 |
| *** thorst__ has quit IRC | 22:12 | |
| *** achanda has quit IRC | 22:16 | |
| *** dims_ has joined #openstack-nova | 22:17 | |
| *** su_zhang has quit IRC | 22:17 | |
| *** mylu has joined #openstack-nova | 22:18 | |
| *** sinese has quit IRC | 22:18 | |
| *** dims has quit IRC | 22:21 | |
| cfriesen | can someone point me at where the limits get set up when creating an instance? I'm trying to figure out why a disk limit isn't specified. | 22:23 |
| *** eglynn has quit IRC | 22:23 | |
| *** ducttape_ has joined #openstack-nova | 22:24 | |
| cfriesen | do I just need to enable the DiskFilter sched filter? | 22:24 |
| *** pushkaru has quit IRC | 22:26 | |
| *** dims_ has quit IRC | 22:27 | |
| *** farid_ has joined #openstack-nova | 22:28 | |
| *** rook-laptop has quit IRC | 22:30 | |
| *** thorst has joined #openstack-nova | 22:31 | |
| *** hblixt has joined #openstack-nova | 22:31 | |
| *** farid_ has quit IRC | 22:34 | |
| *** thorst has quit IRC | 22:35 | |
| openstackgerrit | Anand Shanmugam proposed openstack/nova: Handle Instance Exceptions in add_fixed_ip and remove_fixed_ip https://review.openstack.org/266675 | 22:41 |
| *** thorst has joined #openstack-nova | 22:42 | |
| bauzas | cfriesen: the limits are indeed set by the respective filters | 22:45 |
| bauzas | and passed back to the conductor when returning the destination | 22:45 |
| bauzas | (in the returned tuple) | 22:45 |
| *** ccarmack has quit IRC | 22:49 | |
| *** su_zhang has joined #openstack-nova | 22:52 | |
| *** mylu has quit IRC | 22:53 | |
| *** ctrath has quit IRC | 22:53 | |
| *** mylu has joined #openstack-nova | 22:54 | |
| *** rook-laptop has joined #openstack-nova | 22:55 | |
| *** rlrossit_ has quit IRC | 22:58 | |
| *** jwcroppe has quit IRC | 22:59 | |
| *** burgerk has quit IRC | 22:59 | |
| *** ctrath has joined #openstack-nova | 22:59 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** haomaiwa_ has joined #openstack-nova | 23:01 | |
| *** jichen has quit IRC | 23:05 | |
| *** erlon has quit IRC | 23:06 | |
| *** mdrabe has quit IRC | 23:06 | |
| *** thorst has quit IRC | 23:07 | |
| *** thorst has joined #openstack-nova | 23:07 | |
| *** dave-mccowan has quit IRC | 23:08 | |
| *** xyang1 has quit IRC | 23:08 | |
| *** ccarmack has joined #openstack-nova | 23:09 | |
| *** achanda has joined #openstack-nova | 23:09 | |
| *** eharney has quit IRC | 23:09 | |
| *** ccarmack has quit IRC | 23:09 | |
| *** Sukhdev has quit IRC | 23:10 | |
| *** thorst has quit IRC | 23:12 | |
| *** achanda has quit IRC | 23:12 | |
| *** Sukhdev has joined #openstack-nova | 23:18 | |
| *** mylu has quit IRC | 23:21 | |
| *** Marga_ has quit IRC | 23:23 | |
| *** rfolco has quit IRC | 23:23 | |
| *** e0ne has joined #openstack-nova | 23:24 | |
| *** ducttape_ has quit IRC | 23:24 | |
| *** ctrath has quit IRC | 23:25 | |
| *** mylu has joined #openstack-nova | 23:25 | |
| *** mylu has quit IRC | 23:26 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: libvirt: live-resize vcpus https://review.openstack.org/275412 | 23:28 |
| openstackgerrit | Anand Shanmugam proposed openstack/nova: Handle Instance Exceptions in add_fixed_ip and remove_fixed_ip https://review.openstack.org/266675 | 23:29 |
| *** zenoway has quit IRC | 23:29 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: virt/libvirt: Consider live-resize limits when booting an instance. https://review.openstack.org/275379 | 23:32 |
| *** farid has joined #openstack-nova | 23:32 | |
| *** baoli has quit IRC | 23:33 | |
| *** e0ne has quit IRC | 23:33 | |
| *** edtubill has quit IRC | 23:33 | |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: libvirt: Add methods to parse of domain metadata from xml https://review.openstack.org/275388 | 23:34 |
| openstackgerrit | Ahmed ElKhouly proposed openstack/nova: Compute: Add check on live-resize constraints https://review.openstack.org/275383 | 23:35 |
| *** baoli has joined #openstack-nova | 23:36 | |
| *** mlavalle has quit IRC | 23:38 | |
| *** angdraug has joined #openstack-nova | 23:40 | |
| *** dave-mccowan has joined #openstack-nova | 23:40 | |
| *** baoli has quit IRC | 23:41 | |
| *** zenoway has joined #openstack-nova | 23:41 | |
| *** dimtruck is now known as zz_dimtruck | 23:42 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Use instance in setup_networks_on_host https://review.openstack.org/249110 | 23:43 |
| *** mylu has joined #openstack-nova | 23:44 | |
| *** zenoway has quit IRC | 23:44 | |
| *** zenoway has joined #openstack-nova | 23:45 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Database not needed for RPC serializer test classes https://review.openstack.org/275981 | 23:46 |
| *** jamielennox|away is now known as jamielennox | 23:47 | |
| *** mylu has quit IRC | 23:48 | |
| *** mylu has joined #openstack-nova | 23:48 | |
| *** edmondsw has quit IRC | 23:49 | |
| *** pradk has quit IRC | 23:51 | |
| openstackgerrit | Corey Wright proposed openstack/nova: Avoid lazy-loading flavor during usage audit https://review.openstack.org/274984 | 23:55 |
| *** ivase_ has quit IRC | 23:55 | |
| *** crinkle has quit IRC | 23:55 | |
| *** crinkle has joined #openstack-nova | 23:57 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/nova: (WIP)Implement setup_networks_on_host for Neutron networks https://review.openstack.org/275073 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!