| *** Piet has quit IRC | 00:02 | |
| *** doug-fish has joined #openstack-nova | 00:02 | |
| *** baoli has quit IRC | 00:03 | |
| *** doug-fish has quit IRC | 00:06 | |
| *** jamielennox is now known as jamielennox|away | 00:08 | |
| *** sacharya has joined #openstack-nova | 00:11 | |
| *** su_zhang has joined #openstack-nova | 00:11 | |
| *** tonytan4ever has quit IRC | 00:14 | |
| *** achanda has joined #openstack-nova | 00:15 | |
| *** sacharya has quit IRC | 00:15 | |
| *** su_zhang has quit IRC | 00:16 | |
| *** doug-fish has joined #openstack-nova | 00:17 | |
| *** xyang1 has quit IRC | 00:20 | |
| *** doug-fish has quit IRC | 00:21 | |
| *** jamielennox|away is now known as jamielennox | 00:21 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: tests_fields: Remove most enum-type field tests https://review.openstack.org/244370 | 00:23 |
|---|---|---|
| openstackgerrit | Stephen Finucane proposed openstack/nova: test_fields: Remove 'test_stringify' tests https://review.openstack.org/244371 | 00:23 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: test_fields: Remove all 'Enum' subclass tests https://review.openstack.org/244372 | 00:23 |
| *** EinstCrazy has quit IRC | 00:24 | |
| *** jwcroppe has quit IRC | 00:26 | |
| *** takedakn has joined #openstack-nova | 00:27 | |
| *** IanGovett has quit IRC | 00:27 | |
| *** kfarr has quit IRC | 00:27 | |
| *** yamahata has joined #openstack-nova | 00:28 | |
| *** Piet has joined #openstack-nova | 00:29 | |
| *** sfinucan has quit IRC | 00:32 | |
| *** angdraug has quit IRC | 00:34 | |
| *** oomichi has joined #openstack-nova | 00:34 | |
| *** achanda has quit IRC | 00:35 | |
| *** takedakn has quit IRC | 00:40 | |
| *** suro-patz has quit IRC | 00:40 | |
| *** mnestratov has quit IRC | 00:41 | |
| *** liusheng has quit IRC | 00:46 | |
| *** liusheng has joined #openstack-nova | 00:47 | |
| *** diogogmt has quit IRC | 00:48 | |
| *** vilobhmm has quit IRC | 00:48 | |
| *** yamahata has quit IRC | 00:49 | |
| *** jyuso1 has joined #openstack-nova | 00:51 | |
| *** stackdump has quit IRC | 00:54 | |
| *** doug-fish has joined #openstack-nova | 00:56 | |
| *** doug-fish has quit IRC | 01:00 | |
| *** zhangjn has joined #openstack-nova | 01:03 | |
| *** electrocucaracha has quit IRC | 01:03 | |
| *** EinstCrazy has joined #openstack-nova | 01:03 | |
| *** yamahata has joined #openstack-nova | 01:06 | |
| *** ssurana has quit IRC | 01:09 | |
| *** kmartin_ has joined #openstack-nova | 01:09 | |
| *** apoorvad has quit IRC | 01:10 | |
| *** kmartin has quit IRC | 01:13 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 01:15 | |
| *** sneti has joined #openstack-nova | 01:15 | |
| *** zhenguo has joined #openstack-nova | 01:18 | |
| *** otter768 has joined #openstack-nova | 01:18 | |
| *** yamahata has quit IRC | 01:21 | |
| *** smcginnis has quit IRC | 01:21 | |
| *** terryyao has joined #openstack-nova | 01:22 | |
| *** doug-fish has joined #openstack-nova | 01:22 | |
| openstackgerrit | Wen Zhi Yu proposed openstack/nova: Clean up network resources when reschedule fails https://review.openstack.org/243477 | 01:26 |
| *** doug-fish has quit IRC | 01:27 | |
| *** jwcroppe has joined #openstack-nova | 01:27 | |
| *** rfolco has joined #openstack-nova | 01:29 | |
| *** otter768 has quit IRC | 01:29 | |
| *** mriedem has joined #openstack-nova | 01:31 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 01:37 | |
| *** haomaiwa_ has joined #openstack-nova | 01:37 | |
| *** smcginnis has joined #openstack-nova | 01:40 | |
| openstackgerrit | octopuszhang proposed openstack/nova: update connect_volume test https://review.openstack.org/244002 | 01:43 |
| *** jwcroppe has quit IRC | 01:44 | |
| *** mtanino has quit IRC | 01:45 | |
| openstackgerrit | octopuszhang proposed openstack/nova: update connect_volume test https://review.openstack.org/244002 | 01:45 |
| *** Yanyanhu has joined #openstack-nova | 01:47 | |
| *** su_zhang has joined #openstack-nova | 01:48 | |
| *** janiceonchn has quit IRC | 01:54 | |
| *** sacharya has joined #openstack-nova | 01:54 | |
| *** nic1 has quit IRC | 01:56 | |
| *** su_zhang has quit IRC | 01:56 | |
| *** claudiub has quit IRC | 01:58 | |
| *** sacharya has quit IRC | 01:59 | |
| *** haomaiwa_ has quit IRC | 02:01 | |
| *** haomaiwa_ has joined #openstack-nova | 02:01 | |
| *** doug-fish has joined #openstack-nova | 02:02 | |
| *** mc_nair has quit IRC | 02:03 | |
| *** terryyao has quit IRC | 02:04 | |
| *** terryyao has joined #openstack-nova | 02:05 | |
| *** doug-fish has quit IRC | 02:06 | |
| *** stackdump has joined #openstack-nova | 02:06 | |
| *** smatzek has joined #openstack-nova | 02:11 | |
| *** diogogmt has joined #openstack-nova | 02:13 | |
| *** hemna has quit IRC | 02:13 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 02:13 | |
| *** ijw has quit IRC | 02:14 | |
| *** baoli has joined #openstack-nova | 02:15 | |
| *** thedodd has quit IRC | 02:16 | |
| openstackgerrit | octopuszhang proposed openstack/nova: Use JSON format instead of json format https://review.openstack.org/220875 | 02:16 |
| *** baoli has quit IRC | 02:17 | |
| *** baoli has joined #openstack-nova | 02:17 | |
| openstackgerrit | octopuszhang proposed openstack/nova: Use JSON format instead of json format https://review.openstack.org/220875 | 02:18 |
| *** baoli has quit IRC | 02:18 | |
| *** baoli has joined #openstack-nova | 02:19 | |
| *** seanw1 has quit IRC | 02:20 | |
| *** seanw1 has joined #openstack-nova | 02:21 | |
| *** sneti has quit IRC | 02:24 | |
| *** janiceonchn has joined #openstack-nova | 02:24 | |
| *** mdrabe has joined #openstack-nova | 02:26 | |
| *** doug-fish has joined #openstack-nova | 02:28 | |
| *** spandhe has quit IRC | 02:29 | |
| *** EinstCrazy has quit IRC | 02:29 | |
| *** venkat_p has joined #openstack-nova | 02:30 | |
| *** venkat_p has left #openstack-nova | 02:31 | |
| *** mdrabe has quit IRC | 02:32 | |
| *** stackdump has quit IRC | 02:32 | |
| *** doug-fish has quit IRC | 02:33 | |
| *** gcb has joined #openstack-nova | 02:36 | |
| *** lnxnut has joined #openstack-nova | 02:37 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Revert "Exclude all BDM checks for cells" https://review.openstack.org/242698 | 02:39 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: WIP: cells: create block device mappings in API cell https://review.openstack.org/242649 | 02:39 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make BlockDeviceMapping object support uuid https://review.openstack.org/242603 | 02:39 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add uuid column to BlockDeviceMapping https://review.openstack.org/242602 | 02:39 |
| *** janiceonchn has quit IRC | 02:39 | |
| *** klkumar has joined #openstack-nova | 02:48 | |
| *** rfolco has quit IRC | 02:49 | |
| *** achanda has joined #openstack-nova | 02:54 | |
| *** dims has quit IRC | 02:57 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** haomaiwa_ has joined #openstack-nova | 03:01 | |
| *** kmartin__ has joined #openstack-nova | 03:03 | |
| *** kmartin_ has quit IRC | 03:06 | |
| *** dims has joined #openstack-nova | 03:07 | |
| *** doug-fish has joined #openstack-nova | 03:07 | |
| openstackgerrit | huan proposed openstack/nova: XenAPI:Resolve Nova/Neutron race condition https://review.openstack.org/241127 | 03:07 |
| *** doug-fish has quit IRC | 03:11 | |
| *** su_zhang has joined #openstack-nova | 03:13 | |
| *** signed8bit has quit IRC | 03:14 | |
| *** dims has quit IRC | 03:14 | |
| *** dims has joined #openstack-nova | 03:16 | |
| *** spandhe has joined #openstack-nova | 03:18 | |
| *** mriedem has quit IRC | 03:18 | |
| *** terryyao_ has joined #openstack-nova | 03:21 | |
| *** terryyao has quit IRC | 03:24 | |
| *** dims has quit IRC | 03:25 | |
| *** otter768 has joined #openstack-nova | 03:30 | |
| *** doug-fish has joined #openstack-nova | 03:34 | |
| *** otter768 has quit IRC | 03:35 | |
| *** spandhe has quit IRC | 03:37 | |
| *** doug-fish has quit IRC | 03:38 | |
| *** jwcroppe has joined #openstack-nova | 03:40 | |
| *** doug-fish has joined #openstack-nova | 03:45 | |
| *** doug-fish has quit IRC | 03:49 | |
| *** stevemar_ has quit IRC | 03:49 | |
| *** stevemar_ has joined #openstack-nova | 03:50 | |
| *** jwcroppe has quit IRC | 03:50 | |
| *** Sree has joined #openstack-nova | 03:51 | |
| *** klkumar has quit IRC | 03:51 | |
| *** Sree has quit IRC | 03:51 | |
| *** Sree has joined #openstack-nova | 03:52 | |
| *** su_zhang has quit IRC | 03:52 | |
| *** stevemar_ has quit IRC | 03:53 | |
| *** klkumar has joined #openstack-nova | 03:53 | |
| *** smatzek has quit IRC | 03:57 | |
| *** haomaiwa_ has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-nova | 04:01 | |
| *** suro-patz has joined #openstack-nova | 04:10 | |
| *** suro-patz has quit IRC | 04:11 | |
| *** pixelbeat has quit IRC | 04:12 | |
| *** Marga_ has quit IRC | 04:14 | |
| *** spandhe has joined #openstack-nova | 04:15 | |
| *** klkumar has quit IRC | 04:15 | |
| openstackgerrit | javeme proposed openstack/nova: update task state for the confirm_resize https://review.openstack.org/244433 | 04:21 |
| *** suro-patz has joined #openstack-nova | 04:23 | |
| *** spandhe has quit IRC | 04:23 | |
| *** klkumar has joined #openstack-nova | 04:23 | |
| *** suro-patz has quit IRC | 04:23 | |
| *** suro-patz has joined #openstack-nova | 04:24 | |
| *** garyk1 has joined #openstack-nova | 04:28 | |
| *** terryyao has joined #openstack-nova | 04:32 | |
| *** terryyao_ has quit IRC | 04:33 | |
| *** haomaiwang has quit IRC | 04:37 | |
| *** haomaiwang has joined #openstack-nova | 04:42 | |
| *** jwcroppe has joined #openstack-nova | 04:44 | |
| *** haomaiwang has quit IRC | 04:56 | |
| *** edleafe has quit IRC | 04:57 | |
| *** achanda has quit IRC | 05:04 | |
| *** links has joined #openstack-nova | 05:07 | |
| *** chinmaya has joined #openstack-nova | 05:12 | |
| *** haomaiwang has joined #openstack-nova | 05:13 | |
| *** fawadkhaliq has joined #openstack-nova | 05:14 | |
| chinmaya | :garyk hi Gary. have a query on resize/cold migrate in vmware driver | 05:14 |
| garyk1 | sure, all ears | 05:17 |
| *** sacharya has joined #openstack-nova | 05:19 | |
| *** baoli has quit IRC | 05:20 | |
| chinmaya | :garyk1 suppose i have 2 cluster compute nodes, and i trigger resize, generally i see the scheduler picks up the other cluster as compute node. but the VM resides on the same cluster, | 05:23 |
| chinmaya | and nova show, will show the hypervisor as the other cluster | 05:23 |
| chinmaya | : garyk1 are you planning on considering the dest - param | 05:24 |
| garyk1 | chinmaya: the resize is done in place | 05:24 |
| garyk1 | that is we do not need to move it to another cluster. | 05:25 |
| garyk1 | chinmaya: we are planning to do that for live migrations | 05:25 |
| chinmaya | :garyk1 yeah, but cold migration also goes in the same flow | 05:25 |
| garyk1 | yup. agree. that too will be addressed | 05:26 |
| *** suro-patz has quit IRC | 05:27 | |
| chinmaya | garyk1: "we do not need to move it to another cluster" im not quite sure about this.. | 05:27 |
| *** haomaiwang has quit IRC | 05:27 | |
| chinmaya | if resize_one_same_host is false | 05:27 |
| garyk1 | chinmaya: i do not understand. if one does a resize then the properties of the instance are changed. | 05:27 |
| garyk1 | why do you need to move this to a new host? | 05:28 |
| chinmaya | *resize_on_same_host | 05:28 |
| garyk1 | in the vmware case a host == cluster | 05:28 |
| chinmaya | suppose there 2 clusters, and i have seen a behavior, scheduler was picking up another cluster | 05:29 |
| garyk1 | https://github.com/openstack/nova/blob/master/nova/virt/vmwareapi/driver.py#L131 | 05:29 |
| chinmaya | yeah right, 2 clusters = 2 hosts. | 05:29 |
| garyk1 | chinmaya: it does not have anything to do with the scheduler | 05:29 |
| garyk1 | chinmaya: can you please explain the problem that you saw | 05:30 |
| chinmaya | i'll . give me a sec :) | 05:31 |
| *** otter768 has joined #openstack-nova | 05:31 | |
| *** otter768 has quit IRC | 05:36 | |
| chinmaya | garyk1: actually i was trying these things on stable/kilo. https://github.com/openstack/nova/blob/stable/kilo/nova/compute/manager.py#L3907 | 05:36 |
| chinmaya | if allow_resize_to_same_host is set to false, | 05:36 |
| chinmaya | and i was observing after the resize, nova show used to show the other cluster | 05:37 |
| chinmaya | as the hypervisor_host_name | 05:37 |
| *** haomaiwang has joined #openstack-nova | 05:38 | |
| chinmaya | https://github.com/openstack/nova/blob/stable/kilo/nova/compute/manager.py#L4072 , "migration.dest_compute " was the name of the other compute cluster | 05:39 |
| garyk1 | chinmaya: let me check that and get back to you | 05:40 |
| garyk1 | the resize it self is done in place. i see no reason why it should be moved to another cluster | 05:40 |
| chinmaya | what if the new flavor needs a new compute | 05:41 |
| chinmaya | i mean, existing compute doesnt have that much space | 05:41 |
| garyk1 | then the compute will be at capacity | 05:42 |
| garyk1 | can you please provide a use case? | 05:44 |
| *** haomaiwang has quit IRC | 05:50 | |
| chinmaya | sure. im running two nova compute services for managing two clusters, while performing resize (and set allow_resize_to_same_host = false) i was getting this > https://github.com/openstack/nova/blob/stable/kilo/nova/compute/resource_tracker.py#L205 dest_compute as other cluster | 05:51 |
| *** jinxing has joined #openstack-nova | 05:54 | |
| garyk1 | at the moment the code supports allow_resize_to_same_host | 05:55 |
| garyk1 | if we want to set this as false then we need to add in the cold and live migration support | 05:55 |
| *** jinxing has quit IRC | 05:55 | |
| *** TakashiN has joined #openstack-nova | 05:56 | |
| *** TakashiN is now known as takashin | 05:56 | |
| *** takashin has left #openstack-nova | 05:57 | |
| *** takashin has joined #openstack-nova | 05:58 | |
| *** jinxing has joined #openstack-nova | 05:58 | |
| *** abhishekk has joined #openstack-nova | 06:00 | |
| chinmaya | garyk1: got it. and in cold migration support, we need to clone the vm anyways right, by treating the "dest" param ? | 06:01 |
| chinmaya | but generally, these dest param is the host ip, which wont work in vmware case. it needs manager changes, no ? | 06:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/242727 | 06:02 |
| garyk1 | yes, we need to do the clone | 06:02 |
| *** achanda has joined #openstack-nova | 06:05 | |
| *** Marga_ has joined #openstack-nova | 06:06 | |
| *** irenab has quit IRC | 06:07 | |
| *** emagana has joined #openstack-nova | 06:07 | |
| *** haomaiwa_ has joined #openstack-nova | 06:09 | |
| chinmaya | :garyk1 one more question. Are you planning on supporting cold migration on different hosts ? | 06:10 |
| chinmaya | is there any bug/bp ? | 06:10 |
| *** doug-fish has joined #openstack-nova | 06:10 | |
| garyk1 | chinmaya: it is something that we plan to do. i guess i need to post a spec… but feel free to do that if you want | 06:10 |
| *** achanda has quit IRC | 06:10 | |
| abhishekk | johnthetubaguy: hi you around? | 06:12 |
| chinmaya | garyk1: sure.. however i feel live migration is a better value add. but can only be supported from vc 6 and above ? | 06:13 |
| *** jerrygb has quit IRC | 06:14 | |
| *** doug-fish has quit IRC | 06:15 | |
| *** achanda has joined #openstack-nova | 06:15 | |
| garyk1 | chinmaya: yes, that is correct | 06:17 |
| garyk1 | we need to leverage v6 migrations across clusters | 06:17 |
| *** cfriesen__ has quit IRC | 06:17 | |
| garyk1 | that was the plan. | 06:17 |
| chinmaya | garyk1: i see. Thanks for your time Gary. | 06:18 |
| garyk1 | thank you. have a good one | 06:18 |
| *** abhishekk has quit IRC | 06:18 | |
| openstackgerrit | javeme proposed openstack/nova: update task state for the confirm_resize https://review.openstack.org/244433 | 06:21 |
| *** haomaiwa_ has quit IRC | 06:21 | |
| *** markmc has quit IRC | 06:22 | |
| *** gongysh has joined #openstack-nova | 06:22 | |
| *** browne has quit IRC | 06:23 | |
| *** markmc has joined #openstack-nova | 06:24 | |
| *** haomaiwang has joined #openstack-nova | 06:25 | |
| *** jnclyz12483 has joined #openstack-nova | 06:26 | |
| *** jnclyz12483 has quit IRC | 06:26 | |
| *** jnclyz12483 has joined #openstack-nova | 06:27 | |
| *** jnclyz12483 has quit IRC | 06:27 | |
| *** Sree_ has joined #openstack-nova | 06:28 | |
| *** jnclyz12483 has joined #openstack-nova | 06:28 | |
| *** Sree_ is now known as Guest70549 | 06:28 | |
| *** jnclyz12483 has quit IRC | 06:28 | |
| *** jnclyz12483 has joined #openstack-nova | 06:29 | |
| *** jnclyz12483 has quit IRC | 06:29 | |
| *** jnclyz12483 has joined #openstack-nova | 06:30 | |
| *** Sree has quit IRC | 06:30 | |
| *** jnclyz12483 has quit IRC | 06:31 | |
| *** jnclyz12483 has joined #openstack-nova | 06:31 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Fix wrong usage of config option use_neutron_default_nets https://review.openstack.org/243061 | 06:32 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test:clean up wrong value in method CONF.set_override https://review.openstack.org/242416 | 06:32 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Enhance value check for option notify_on_state_change https://review.openstack.org/242717 | 06:32 |
| *** jnclyz12483 has quit IRC | 06:33 | |
| *** mnestratov has joined #openstack-nova | 06:36 | |
| *** mjura has joined #openstack-nova | 06:39 | |
| *** irenab has joined #openstack-nova | 06:41 | |
| *** sc68cal has quit IRC | 06:45 | |
| *** links has quit IRC | 06:45 | |
| *** emagana has quit IRC | 06:46 | |
| *** emagana has joined #openstack-nova | 06:52 | |
| *** suro-patz has joined #openstack-nova | 06:54 | |
| *** jhesketh has quit IRC | 06:54 | |
| *** masayukig has quit IRC | 06:54 | |
| *** sacharya has quit IRC | 06:54 | |
| *** gszasz has joined #openstack-nova | 06:54 | |
| *** baoli has joined #openstack-nova | 06:55 | |
| *** jhesketh has joined #openstack-nova | 06:59 | |
| *** masayukig has joined #openstack-nova | 07:00 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-nova | 07:01 | |
| *** baoli has quit IRC | 07:01 | |
| *** josecastroleon has joined #openstack-nova | 07:03 | |
| *** haomaiwang has quit IRC | 07:04 | |
| *** links has joined #openstack-nova | 07:05 | |
| *** scheuran has joined #openstack-nova | 07:07 | |
| *** moshele has joined #openstack-nova | 07:08 | |
| *** emagana has quit IRC | 07:09 | |
| *** jwcroppe has quit IRC | 07:10 | |
| *** zhangjn has quit IRC | 07:12 | |
| *** sirushti has quit IRC | 07:14 | |
| *** gridinv has quit IRC | 07:16 | |
| *** gibi has quit IRC | 07:16 | |
| eliqiao | hi nova, I get No sql_connection parameter is established when I update nova to latest. anyone seen this? | 07:16 |
| eliqiao | is there workaround now/ | 07:17 |
| *** gridinv has joined #openstack-nova | 07:17 | |
| *** sirushti has joined #openstack-nova | 07:17 | |
| *** gibi has joined #openstack-nova | 07:17 | |
| *** nkrinner has quit IRC | 07:17 | |
| *** nkrinner has joined #openstack-nova | 07:17 | |
| eliqiao | seems related to blueprint new-oslodb-enginefacade | 07:18 |
| *** adun153 has joined #openstack-nova | 07:18 | |
| adun153 | What's the purpose of nova-cert and nova-conductor? | 07:19 |
| *** suro-patz has quit IRC | 07:19 | |
| *** achanda has quit IRC | 07:20 | |
| *** lpetrut has joined #openstack-nova | 07:20 | |
| eliqiao | adun153: http://docs.openstack.org/kilo/config-reference/content/section_conductor.html | 07:21 |
| *** stevemar_ has joined #openstack-nova | 07:21 | |
| *** salv-orlando has joined #openstack-nova | 07:23 | |
| adun153 | Could you please give me an example of data that nova-compute need to write to, or read from the database? | 07:23 |
| *** haomaiwang has joined #openstack-nova | 07:24 | |
| *** stevemar_ has quit IRC | 07:24 | |
| *** doron has joined #openstack-nova | 07:26 | |
| openstackgerrit | lyanchih proposed openstack/python-novaclient: Client repr hosts list return value will raise AttributeError: host https://review.openstack.org/244476 | 07:29 |
| *** adun153 has quit IRC | 07:29 | |
| *** zhangjn has joined #openstack-nova | 07:29 | |
| *** adun153 has joined #openstack-nova | 07:31 | |
| *** otter768 has joined #openstack-nova | 07:32 | |
| *** zhangjn has quit IRC | 07:32 | |
| *** zhangjn has joined #openstack-nova | 07:33 | |
| *** armax has quit IRC | 07:34 | |
| *** zhangjn has quit IRC | 07:34 | |
| *** otter768 has quit IRC | 07:36 | |
| eliqiao | adun153: such as nova-compute wants to get an instance by instance name/ip it will send request to nova-conductor | 07:37 |
| *** haomaiwang has quit IRC | 07:38 | |
| *** haomaiwang has joined #openstack-nova | 07:40 | |
| *** alex_klimov has joined #openstack-nova | 07:43 | |
| *** rdopiera has joined #openstack-nova | 07:43 | |
| *** alexschm has joined #openstack-nova | 07:44 | |
| *** haomaiwang has quit IRC | 07:51 | |
| *** doron is now known as doron_afk | 07:51 | |
| *** paul-carlton has joined #openstack-nova | 07:52 | |
| *** sacharya has joined #openstack-nova | 07:55 | |
| *** aix has joined #openstack-nova | 07:55 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: split check_can_live_migrate_destination https://review.openstack.org/244484 | 07:56 |
| *** haomaiwa_ has joined #openstack-nova | 07:56 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-migrate: Change the status Migration is created with https://review.openstack.org/244485 | 07:56 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Don't track migrations in 'accepted' state https://review.openstack.org/244486 | 07:56 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Move the Migration set-status-if-exists pattern to a method https://review.openstack.org/244487 | 07:56 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Mark migration as failed on fail to schedule https://review.openstack.org/244488 | 07:56 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 07:56 |
| *** sahid has joined #openstack-nova | 07:58 | |
| *** josecastroleon has quit IRC | 07:59 | |
| *** sacharya has quit IRC | 07:59 | |
| *** sahid has quit IRC | 08:00 | |
| *** haomaiwa_ has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-nova | 08:01 | |
| *** sahid has joined #openstack-nova | 08:05 | |
| *** lpetrut has quit IRC | 08:06 | |
| *** josecastroleon has joined #openstack-nova | 08:08 | |
| *** takashin has quit IRC | 08:10 | |
| *** paul-carlton has quit IRC | 08:12 | |
| *** sudipto has joined #openstack-nova | 08:12 | |
| *** liverpooler has joined #openstack-nova | 08:13 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 08:13 |
| *** markmc has quit IRC | 08:14 | |
| *** haomaiwang has quit IRC | 08:14 | |
| *** haomaiwang has joined #openstack-nova | 08:16 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 08:17 |
| *** liverpooler has quit IRC | 08:17 | |
| *** liverpooler has joined #openstack-nova | 08:17 | |
| *** jichen has joined #openstack-nova | 08:19 | |
| tangchen__ | Nikola: Hi nikola, please take a look at this series when you are free. https://review.openstack.org/#/c/241476/ | 08:25 |
| *** paul-carlton has joined #openstack-nova | 08:26 | |
| ndipanov | tangchen__, yeah have it marked already no worries | 08:27 |
| tangchen__ | ndipanov: Thx. :) | 08:27 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova-specs: Pause VM during live migration https://review.openstack.org/229040 | 08:29 |
| openstackgerrit | garyk proposed openstack/nova: VMware: expand support for Opaque networks https://review.openstack.org/165750 | 08:30 |
| bauzas | good morning Nova | 08:31 |
| *** sc68cal has joined #openstack-nova | 08:31 | |
| *** haomaiwang has quit IRC | 08:32 | |
| *** haomaiwang has joined #openstack-nova | 08:33 | |
| *** hparekh has quit IRC | 08:34 | |
| *** rotbeard has joined #openstack-nova | 08:34 | |
| *** andymaier has joined #openstack-nova | 08:35 | |
| *** andymaier_ has joined #openstack-nova | 08:35 | |
| *** andymaier_ has quit IRC | 08:38 | |
| *** Pablo|off| has joined #openstack-nova | 08:40 | |
| BobBall | yes bauzas, I know, and I'm working on it :) | 08:40 |
| BobBall | :D | 08:40 |
| BobBall | + good morning | 08:40 |
| bauzas | eh eh | 08:40 |
| *** ociuhandu has quit IRC | 08:41 | |
| bauzas | that's a traditional French time when going to work, y'know ;) | 08:41 |
| bauzas | that, and taking 30 mins for chatting and taking coffee before actually | 08:41 |
| bauzas | working | 08:41 |
| BobBall | I was expecting you to point out that the XenServer CI was broken last night by a Nova merge :) | 08:41 |
| *** hparekh has joined #openstack-nova | 08:43 | |
| bauzas | BobBall: oh nooooo, I'm just caffeinating now after a long pause because $day-off yesterday | 08:43 |
| *** mpavone has joined #openstack-nova | 08:43 | |
| BobBall | ahhh great :) | 08:43 |
| bauzas | that - and reading emails | 08:43 |
| bauzas | but glad to hear you're on the problem :) | 08:44 |
| BobBall | :D | 08:45 |
| *** salv-orlando has quit IRC | 08:46 | |
| *** achanda has joined #openstack-nova | 08:47 | |
| *** Pablo|off| is now known as pcaruana | 08:48 | |
| *** romainh has joined #openstack-nova | 08:49 | |
| *** rcernin has joined #openstack-nova | 08:50 | |
| *** danpb has joined #openstack-nova | 08:57 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes *Utils modules and unit tests https://review.openstack.org/215917 | 08:57 |
| *** lucas-dinner is now known as lucasagomes | 09:00 | |
| *** haomaiwang has quit IRC | 09:00 | |
| *** matrohon has joined #openstack-nova | 09:00 | |
| *** paul-carlton has quit IRC | 09:01 | |
| *** ociuhandu has joined #openstack-nova | 09:01 | |
| *** jistr has joined #openstack-nova | 09:03 | |
| *** gcb has quit IRC | 09:04 | |
| *** gcb has joined #openstack-nova | 09:05 | |
| *** yassine__ has joined #openstack-nova | 09:07 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 09:08 |
| *** haomaiwang has joined #openstack-nova | 09:08 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 09:08 |
| *** markmc has joined #openstack-nova | 09:10 | |
| openstackgerrit | Jianghua Wang proposed openstack/nova-specs: XenAPI: Add spec to support Virtual GPU (VGPU) https://review.openstack.org/229351 | 09:10 |
| *** aix has quit IRC | 09:19 | |
| *** claudiub has joined #openstack-nova | 09:19 | |
| *** doron_afk has left #openstack-nova | 09:20 | |
| *** aix has joined #openstack-nova | 09:21 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Libvirt: Fix SMBFS volume attach race condition https://review.openstack.org/187619 | 09:22 |
| *** eliqiao has quit IRC | 09:23 | |
| *** eliqiao has joined #openstack-nova | 09:24 | |
| *** rk4n has joined #openstack-nova | 09:28 | |
| openstackgerrit | Jianghua Wang proposed openstack/nova-specs: XenAPI: Add spec to support Virtual GPU (VGPU) https://review.openstack.org/229351 | 09:28 |
| *** pixelbeat has joined #openstack-nova | 09:28 | |
| *** terryyao has quit IRC | 09:29 | |
| *** otter768 has joined #openstack-nova | 09:32 | |
| tangchen__ | hi, just a little question. If I use nova lock xxx to lock an instance, is there any interface that I can check if the server is locked ? | 09:33 |
| tangchen__ | I tried nova show, nova list --long | 09:33 |
| tangchen__ | there is no such info | 09:33 |
| *** achanda has quit IRC | 09:34 | |
| *** haomaiwang has quit IRC | 09:34 | |
| *** ihrachys has joined #openstack-nova | 09:35 | |
| *** jinxing has quit IRC | 09:36 | |
| *** jinxing has joined #openstack-nova | 09:36 | |
| *** gcb has quit IRC | 09:37 | |
| *** otter768 has quit IRC | 09:37 | |
| andrearosa | tangchen__: nova show reports the locked field | 09:38 |
| *** haomaiwa_ has joined #openstack-nova | 09:38 | |
| tangchen__ | andrearosa: No, I didn't find that field in nova show output. | 09:38 |
| andrearosa | tangchen__: I did, tested just now on a devstack installation, maybe it is enabled by some specific extension, let me check | 09:39 |
| *** garyk has quit IRC | 09:42 | |
| tangchen__ | andrearosa: Please refer to: http://paste.openstack.org/show/478638/ | 09:43 |
| tangchen__ | andrearosa: This is my output. | 09:43 |
| andrearosa | do you have /me looking | 09:44 |
| andrearosa | sorry I am looking at your output | 09:44 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 09:44 |
| tangchen__ | andrearosa: Thx for helping. ;) | 09:44 |
| *** garyk1 has quit IRC | 09:45 | |
| alexschm | danpb: Hi, do you know any second core I could ask to review the fix for live migration with serial console? | 09:45 |
| alexschm | dapnb: https://review.openstack.org/#/c/215102/ | 09:45 |
| alexschm | danpb: you gave your +2 two months ago and nothing happened since that point in time | 09:46 |
| *** openstackgerrit has quit IRC | 09:46 | |
| *** openstackgerrit has joined #openstack-nova | 09:47 | |
| openstackgerrit | javeme proposed openstack/nova: update task state for the confirm_resize https://review.openstack.org/244433 | 09:49 |
| andrearosa | tangchen__: this is my output for an instance owned by demo user in devstack: http://paste.openstack.org/show/478640/ | 09:49 |
| bauzas | tangchen__: which version are you running ? | 09:50 |
| bauzas | tangchen__: IIRC, there was a bp to show it | 09:50 |
| bauzas | sec | 09:50 |
| bauzas | got it | 09:50 |
| bauzas | tangchen__: andrearosa: https://github.com/openstack/nova/blame/master/nova/api/openstack/compute/views/servers.py#L311 | 09:50 |
| *** Sree has joined #openstack-nova | 09:50 | |
| bauzas | tangchen__: which means that your client has to query for an API version >= 2.9 | 09:51 |
| *** flyingtt has joined #openstack-nova | 09:51 | |
| andrearosa | bauzas: thanks | 09:51 |
| bauzas | tangchen__: https://github.com/openstack/nova/commit/49a572a043f27623a15af5f1b8e54c3a560b805c | 09:51 |
| tangchen__ | bauzas, andrearosa: OK, I see. Thank you guys very much. | 09:52 |
| *** Guest70549 has quit IRC | 09:54 | |
| *** haomaiwa_ has quit IRC | 09:58 | |
| *** andymaier has quit IRC | 09:58 | |
| *** haomaiwa_ has joined #openstack-nova | 09:59 | |
| *** rk4n has quit IRC | 09:59 | |
| openstackgerrit | javeme proposed openstack/nova: update task state for the confirm_resize https://review.openstack.org/244433 | 10:00 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: add hacking check for help text length https://review.openstack.org/242078 | 10:00 |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** haomaiwa_ has joined #openstack-nova | 10:01 | |
| *** ZZelle has joined #openstack-nova | 10:02 | |
| *** gszasz has quit IRC | 10:02 | |
| *** Yanyanhu has quit IRC | 10:04 | |
| johnthetubaguy | gibi: just approving the notification blueprints, did I get the arrows the correct way around in the dependency tree? https://blueprints.launchpad.net/nova/+spec/service-status-notification | 10:04 |
| gibi | johnthetubaguy: cheking.., | 10:07 |
| *** gszasz has joined #openstack-nova | 10:07 | |
| *** stackdump has joined #openstack-nova | 10:07 | |
| gibi | johnthetubaguy: yes, this is my view as well to start with the new notification | 10:07 |
| *** adun153 has quit IRC | 10:08 | |
| *** derekh has joined #openstack-nova | 10:09 | |
| *** haomaiwa_ has quit IRC | 10:11 | |
| *** stackdump has quit IRC | 10:12 | |
| *** haomaiwang has joined #openstack-nova | 10:13 | |
| bauzas | andrearosa: catching you because PaulMurray isn't there, I'm missing a bit of information about Bristol :) | 10:14 |
| bauzas | in particular the train station and how far it is from the office | 10:14 |
| bauzas | andrearosa: I got most of the information by http://www.hpl.hp.com/bristol/ and http://www.hpl.hp.com/downloads/HP_Labs_Bristol.pdf but I want to make sure it's correct | 10:15 |
| bauzas | before me booking flights etc. | 10:15 |
| *** openstackgerrit has quit IRC | 10:16 | |
| *** openstackgerrit has joined #openstack-nova | 10:17 | |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V: Adds host maintenance implementation https://review.openstack.org/204157 | 10:19 |
| johnthetubaguy | gibi: cool, thanks for thecking | 10:20 |
| *** sacharya has joined #openstack-nova | 10:21 | |
| andrearosa | bauzas: sorry I was in a meeting with PaulMurray. Filton Abbey wood station or/and Bristol Parkway are the two stations close to the office, they are 10-15 mins walk | 10:21 |
| bauzas | okay, that's what I understood then | 10:22 |
| gibi | johnthetubaguy: thanks for approving :) | 10:22 |
| bauzas | andrearosa: Google tells me it's rather 20-30 mins walk for both of them :p | 10:22 |
| andrearosa | and if you need taxi Parkway is your choice as it is bigger than Filton. | 10:22 |
| andrearosa | 30 mins it too much, I think Filton is a bit shorted than Parkway | 10:23 |
| andrearosa | s/shorted/shorter | 10:23 |
| bauzas | andrearosa: https://www.google.fr/maps/dir/Filton+Abbey+Wood/HP+Bristol,+Long+Down+Ave,+Stoke+Gifford,+Bristol,+South+Gloucestershire+BS34+8QZ/@51.5034813,-2.5587824,17z/data=!3m1!4b1!4m14!4m13!1m5!1m1!1s0x487191c88fb332f7:0xc0ce3149aee89f36!2m2!1d-2.5635251!2d51.5035814!1m5!1m1!1s0x487191cc572dbc15:0x8348734b13d56a2c!2m2!1d-2.5541369!2d51.500937!3e2 | 10:24 |
| *** kaisers has quit IRC | 10:25 | |
| *** sacharya has quit IRC | 10:25 | |
| andrearosa | bauzas: that route suggests to take left when you leave the station but if you go right is easier, so let'say 20 mins | 10:28 |
| bauzas | ack | 10:28 |
| *** haomaiwang has quit IRC | 10:31 | |
| *** janiceonchn has joined #openstack-nova | 10:32 | |
| *** kaisers has joined #openstack-nova | 10:33 | |
| *** janiceonchn has quit IRC | 10:33 | |
| *** haomaiwang has joined #openstack-nova | 10:36 | |
| *** deepakcs has joined #openstack-nova | 10:36 | |
| *** paul-carlton has joined #openstack-nova | 10:36 | |
| paul-carlton1 | bauzas I walk from Filton Abbey Wood to the office everyday, it is a 10 to 15 minutes walk | 10:38 |
| *** garyk has joined #openstack-nova | 10:38 | |
| bauzas | okay, I don't want to run a contest, good to know :) | 10:38 |
| andrearosa | lol | 10:38 |
| bauzas | :D | 10:38 |
| paul-carlton1 | When you leave the station head for MOD office and turn right up the hill , path takes you round MOD office, less roads to cross and I think it is slightly quicker | 10:40 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Remove onSharedStorage from evacuate API https://review.openstack.org/243562 | 10:40 |
| *** jinxing has quit IRC | 10:45 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Rearranges to create new Compute API Guide https://review.openstack.org/230186 | 10:46 |
| openstackgerrit | John Garbutt proposed openstack/nova: Rearranges to create new Compute API Guide https://review.openstack.org/230186 | 10:47 |
| *** jistr_ has joined #openstack-nova | 10:48 | |
| *** oomichi has quit IRC | 10:48 | |
| *** haomaiwang has quit IRC | 10:48 | |
| *** ihrachys_ has joined #openstack-nova | 10:48 | |
| *** aix_ has joined #openstack-nova | 10:48 | |
| *** rcernin_ has joined #openstack-nova | 10:48 | |
| *** ihrachys has quit IRC | 10:48 | |
| *** jistr has quit IRC | 10:49 | |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V: Adds host maintenance implementation https://review.openstack.org/204157 | 10:49 |
| *** deepakcs has quit IRC | 10:49 | |
| *** rcernin has quit IRC | 10:49 | |
| *** rcernin has joined #openstack-nova | 10:50 | |
| *** ihrachys has joined #openstack-nova | 10:50 | |
| *** 7F1AAZM53 has joined #openstack-nova | 10:51 | |
| *** aix has quit IRC | 10:51 | |
| *** ihrachys_ has quit IRC | 10:52 | |
| *** jgjhgsonchn88523 has joined #openstack-nova | 10:53 | |
| *** aix_ has quit IRC | 10:53 | |
| *** rcernin_ has quit IRC | 10:53 | |
| *** eglynn has joined #openstack-nova | 10:53 | |
| *** jistr_ has quit IRC | 10:54 | |
| *** Sree has quit IRC | 10:56 | |
| *** Sree has joined #openstack-nova | 10:57 | |
| *** baoli has joined #openstack-nova | 10:58 | |
| *** 7F1AAZM53 has quit IRC | 11:01 | |
| *** Sree has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-nova | 11:01 | |
| *** haomaiwang has quit IRC | 11:02 | |
| *** baoli has quit IRC | 11:03 | |
| *** garyk has quit IRC | 11:04 | |
| *** garyk has joined #openstack-nova | 11:04 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add SerialConsoleOps class https://review.openstack.org/184040 | 11:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Removes support for Hyper-V Server 2008 R2 https://review.openstack.org/237643 | 11:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Nova serial console access support https://review.openstack.org/145004 | 11:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 11:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console handler class https://review.openstack.org/184039 | 11:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 11:05 |
| *** aix_ has joined #openstack-nova | 11:05 | |
| *** jistr_ has joined #openstack-nova | 11:05 | |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds Hyper-V vNic hot-plug implementation https://review.openstack.org/138255 | 11:07 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Moves OVS related code to a common location https://review.openstack.org/179727 | 11:07 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds Hyper-V OVS ViF driver https://review.openstack.org/140045 | 11:07 |
| *** zhenguo has quit IRC | 11:07 | |
| *** e0ne has joined #openstack-nova | 11:09 | |
| *** rk4n has joined #openstack-nova | 11:12 | |
| *** jamielennox is now known as jamielennox|away | 11:12 | |
| *** haomaiwang has joined #openstack-nova | 11:12 | |
| openstackgerrit | Jianghua Wang proposed openstack/nova: xenapi: OVS agent updates the wrong port when using XenServer + Neutron https://review.openstack.org/242846 | 11:12 |
| *** alex_klimov has quit IRC | 11:13 | |
| *** dims_ has joined #openstack-nova | 11:14 | |
| *** baoli has joined #openstack-nova | 11:14 | |
| *** irenab has quit IRC | 11:14 | |
| *** jerrygb has joined #openstack-nova | 11:15 | |
| *** lpetrut has joined #openstack-nova | 11:16 | |
| *** romainh has left #openstack-nova | 11:16 | |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: Blueprint to use libvirt storage pools https://review.openstack.org/232045 | 11:17 |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: Blueprint to use libvirt storage pools https://review.openstack.org/232045 | 11:17 |
| *** gszasz has quit IRC | 11:18 | |
| *** baoli has quit IRC | 11:19 | |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: Blueprint to use libvirt storage pools https://review.openstack.org/232045 | 11:19 |
| *** mgoddard_ has quit IRC | 11:19 | |
| *** mgoddard_ has joined #openstack-nova | 11:19 | |
| *** jerrygb has quit IRC | 11:19 | |
| dims_ | danpb : johnthetubaguy : couple of quick policy.json fixes if you get a chance https://review.openstack.org/#/c/242902/ and https://review.openstack.org/#/c/243090/ | 11:22 |
| *** haomaiwang has quit IRC | 11:24 | |
| johnthetubaguy | dims_: I do wonder if we can have something in oslo.polcy to help us find these missing policy rules in the gate, like a raise exceptions mode, that would be awesome stuff | 11:27 |
| dims_ | johnthetubaguy y, something that tracks these things and finally dump it at the end | 11:27 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 11:28 |
| johnthetubaguy | dims_: possibly, I would be fine with it raising an exception and triggering a 500 error, just for testing, like add a policy rule of "raise exception on missing rule" or something? | 11:28 |
| dims_ | right | 11:28 |
| *** haomaiwang has joined #openstack-nova | 11:30 | |
| *** terryyao has joined #openstack-nova | 11:30 | |
| *** josecastroleon has quit IRC | 11:30 | |
| dims_ | thanks danpb and johnthetubaguy - will whip up something soon | 11:31 |
| johnthetubaguy | dims_: awesome :) | 11:32 |
| dims_ | though johnthetubaguy - that would be in oslo.policy, which we are yet to adopt | 11:32 |
| johnthetubaguy | dims_: yeah, gives us a reason to get some good stuff | 11:32 |
| *** otter768 has joined #openstack-nova | 11:33 | |
| *** mwagner has quit IRC | 11:34 | |
| *** fawadkhaliq has quit IRC | 11:35 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova: Hyper-V: removes *Utils modules and unit tests https://review.openstack.org/215917 | 11:37 |
| *** otter768 has quit IRC | 11:38 | |
| *** smatzek has joined #openstack-nova | 11:39 | |
| *** lnxnut has quit IRC | 11:39 | |
| *** baoli has joined #openstack-nova | 11:41 | |
| *** haomaiwang has quit IRC | 11:41 | |
| *** alex_klimov has joined #openstack-nova | 11:42 | |
| *** baoli has quit IRC | 11:46 | |
| *** gcb has joined #openstack-nova | 11:47 | |
| johnthetubaguy | andrearosa: hey, I was just looking at this spec: https://review.openstack.org/#/c/226199/10/specs/mitaka/approved/enhance-instance-operations-during-live-migration.rst,cm | 11:47 |
| *** haomaiwang has joined #openstack-nova | 11:47 | |
| johnthetubaguy | andrearosa: I am not sure I understand how much we gain, vs the effort to ensure it works | 11:47 |
| andrearosa | johnthetubaguy: that is one of the main concerns raised against that proposal | 11:48 |
| andrearosa | johnthetubaguy: the idea comes from our experience where you can have a very long time for live migration | 11:48 |
| andrearosa | and we have ppl asking for still managing their instances | 11:49 |
| andrearosa | johnthetubaguy: for example if the live migration takes | 11:49 |
| andrearosa | 20 hours why the customer couldn't get access to the console or have a look at the console.log file for 20 hours? | 11:49 |
| *** ndipanov has quit IRC | 11:50 | |
| andrearosa | johnthetubaguy: I agree that is not one of the must to have feature for live-migration | 11:50 |
| johnthetubaguy | andrearosa: it feels like the cancel live migrate is more useful | 11:52 |
| johnthetubaguy | andrearosa: honestly, if we get live-migrate tested and working this release, I think adding that console access next release makes sense, just feels like its a cycle to early to worry about this | 11:53 |
| andrearosa | johnthetubaguy: it is, but why do you see one excluding the other? | 11:53 |
| johnthetubaguy | andrearosa: its more the testing | 11:53 |
| johnthetubaguy | andrearosa: I see cancel as more widely useful (if you need to snapshot, or resize, etc) | 11:53 |
| andrearosa | johnthetubaguy: your point is about priority? I mean you don't want to spend time on these things as there are more important feature to implement? | 11:55 |
| *** gszasz has joined #openstack-nova | 11:55 | |
| johnthetubaguy | andrearosa: its a combination of things really, priority is probably the main one | 11:55 |
| johnthetubaguy | andrearosa: we don't have live-migrate working correctly yet, it doesn't seem appropriate to start allowing concurrent operations during live-migrate until we have the basic things solid | 11:56 |
| johnthetubaguy | (and well tested) | 11:56 |
| johnthetubaguy | danpb: looking at the split network plane for live-migrate stuff | 11:56 |
| andrearosa | johnthetubaguy: I see your point and I agree, I thought to get it approved just in case we can make everything working in time, probably unlikely | 11:57 |
| johnthetubaguy | danpb: confused why we have both CONF.libvirt.live_migration_uri and CONF.my_live_migration_ip in this case, doesn't live_migration_uri already help, or is that the management traffic side? | 11:57 |
| *** terryyao has quit IRC | 11:57 | |
| *** LiuNanke has joined #openstack-nova | 11:57 | |
| *** terryyao has joined #openstack-nova | 11:58 | |
| johnthetubaguy | andrearosa: yeah, it feels too soon right now, lets not get distracted from all the live-migrate bug fixes, and keeping the gating tests stable, etc | 11:58 |
| danpb | johnthetubaguy: live migration URI is used by the source host to initiate connections | 11:58 |
| andrearosa | johnthetubaguy: ok I am fine to not get it approved for this cycle | 11:58 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test:clean up wrong value in method CONF.set_override https://review.openstack.org/242416 | 11:58 |
| danpb | johnthetubaguy: it needs to be expanded to contain the IP address of the *target* host, which CONF.my_live_migration_ip provides | 11:59 |
| danpb | johnthetubaguy: the target host IP is passed back to the source host in the pre_live_migration phase | 11:59 |
| *** gcb_ has joined #openstack-nova | 11:59 | |
| danpb | johnthetubaguy: you can't hardcode the IP address in CONF.libvirt.live_migration_uri, because on any one host you obviously need to be able to migrate to arbitrary target host | 11:59 |
| johnthetubaguy | danpb: ah, so thats the bit I was missing, its always got the substitution variables in there | 12:00 |
| johnthetubaguy | danpb: thanks, makes sense now | 12:00 |
| danpb | live_migration_ip is basically replacing the use of compute_nodes.hostname field | 12:00 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Remove onSharedStorage from evacuate API https://review.openstack.org/243562 | 12:00 |
| *** haomaiwang has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-nova | 12:01 | |
| *** gcb has quit IRC | 12:03 | |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Query and cancel ongoing live migrations https://review.openstack.org/228828 | 12:04 |
| openstackgerrit | jichenjc proposed openstack/nova: libvirt:Fix a typo of test cases https://review.openstack.org/243117 | 12:04 |
| *** rfolco has joined #openstack-nova | 12:04 | |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Query and cancel ongoing live migrations https://review.openstack.org/228828 | 12:05 |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Query and cancel ongoing live migrations https://review.openstack.org/228828 | 12:07 |
| *** marzif has joined #openstack-nova | 12:08 | |
| *** jgjhgsonchn88523 has quit IRC | 12:08 | |
| *** romainh has joined #openstack-nova | 12:10 | |
| *** haomaiwang has quit IRC | 12:11 | |
| *** haomaiwang has joined #openstack-nova | 12:11 | |
| alexschm | johnthetubaguy: can you add this review to the agenda of the meeting today? https://review.openstack.org/#/c/215102/ | 12:12 |
| alexschm | johnthetubaguy: I'm looking for another core to review this, it got a +2 already from danpb two months ago | 12:12 |
| johnthetubaguy | alexschm: its not really a "stuck" review as such, agreed we want someone else to take a look | 12:12 |
| johnthetubaguy | I have reviewed it, but don't know that code enough for a +2 | 12:13 |
| alexschm | johnthetubaguy: right, so should I raise it in the open discussion? | 12:13 |
| johnthetubaguy | alexschm: yeah, that works | 12:15 |
| alexschm | johnthetubaguy: ok thx | 12:15 |
| *** IanGovett has joined #openstack-nova | 12:18 | |
| *** chinmaya has quit IRC | 12:18 | |
| *** josecastroleon has joined #openstack-nova | 12:19 | |
| *** jistr_ is now known as jistr | 12:20 | |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V: Adds Hyper-V UEFI Secure Boot https://review.openstack.org/209581 | 12:20 |
| *** haomaiwang has quit IRC | 12:22 | |
| *** haomaiwang has joined #openstack-nova | 12:23 | |
| openstackgerrit | jichenjc proposed openstack/nova: Add DiskFormat as Enum in fields https://review.openstack.org/229854 | 12:25 |
| openstackgerrit | jichenjc proposed openstack/nova: Make ImageMeta:disk_format use DiskFormatField https://review.openstack.org/231460 | 12:25 |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Migrate libvirt volumes https://review.openstack.org/232053 | 12:25 |
| *** samueldmq has joined #openstack-nova | 12:29 | |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Cancel ongoing live migrations https://review.openstack.org/228828 | 12:30 |
| *** Sree__ has joined #openstack-nova | 12:31 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova-specs: Pause VM during live migration https://review.openstack.org/229040 | 12:37 |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: objects: added 'os_secure_boot' property to ImageMetaProps object https://review.openstack.org/237593 | 12:38 |
| *** baoli has joined #openstack-nova | 12:39 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova-specs: Pause VM during live migration https://review.openstack.org/229040 | 12:40 |
| johnthetubaguy | dansmith: sdague: I think this might break upgrade, if so, we might want to check if we have let any of these slip in already: https://review.openstack.org/#/c/243496/1 | 12:41 |
| * johnthetubaguy goes for lunch | 12:41 | |
| *** haomaiwang has quit IRC | 12:42 | |
| sdague | johnthetubaguy: oh, you are probably right | 12:42 |
| openstackgerrit | jichenjc proposed openstack/nova: Change Invalid exception to a specified exception https://review.openstack.org/232506 | 12:43 |
| *** baoli has quit IRC | 12:44 | |
| *** sacharya has joined #openstack-nova | 12:45 | |
| *** lbelivea has joined #openstack-nova | 12:45 | |
| *** sacharya has quit IRC | 12:49 | |
| *** zhangguoqing has joined #openstack-nova | 12:51 | |
| *** zhangguoqing has quit IRC | 12:52 | |
| *** zhangguoqing has joined #openstack-nova | 12:53 | |
| *** sudipto has quit IRC | 12:53 | |
| *** baoli has joined #openstack-nova | 12:53 | |
| *** arif-ali has quit IRC | 13:01 | |
| *** thorst has joined #openstack-nova | 13:02 | |
| *** LiuNanke has quit IRC | 13:02 | |
| *** jinxing has joined #openstack-nova | 13:04 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Removes support for Hyper-V Server 2008 R2 https://review.openstack.org/237643 | 13:08 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 13:08 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 13:08 |
| *** haomaiwang has joined #openstack-nova | 13:08 | |
| *** mwagner has joined #openstack-nova | 13:09 | |
| *** IanGovett has quit IRC | 13:11 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 13:11 |
| *** annegentle has joined #openstack-nova | 13:12 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 13:15 |
| *** jerrygb has joined #openstack-nova | 13:15 | |
| *** gcb_ has quit IRC | 13:17 | |
| *** annegentle has quit IRC | 13:17 | |
| *** jerrygb_ has joined #openstack-nova | 13:18 | |
| *** aysyd has joined #openstack-nova | 13:19 | |
| *** jerrygb has quit IRC | 13:22 | |
| *** IanGovett has joined #openstack-nova | 13:23 | |
| *** arif-ali has joined #openstack-nova | 13:24 | |
| *** jlanoux has joined #openstack-nova | 13:24 | |
| *** dims_ has quit IRC | 13:24 | |
| *** doug-fish has joined #openstack-nova | 13:26 | |
| *** dims has joined #openstack-nova | 13:26 | |
| *** burgerk has joined #openstack-nova | 13:26 | |
| *** MVenesio has joined #openstack-nova | 13:29 | |
| *** gcb_ has joined #openstack-nova | 13:30 | |
| *** terryyao has quit IRC | 13:30 | |
| alex_xu | johnthetubaguy: hi, should we mention virtual doc sprint in today meeting? just advert it and see if any feedback about the date or anything else. but looks likes fine, no concern from the maillist | 13:31 |
| johnthetubaguy | alex_xu: oh, yeah, we can add that to the dates list | 13:31 |
| johnthetubaguy | alex_xu: good idea | 13:32 |
| johnthetubaguy | alex_xu: do you want to quickly add that in there for me now? | 13:32 |
| alex_xu | johnthetubaguy: yea, let me add it | 13:32 |
| johnthetubaguy | alex_xu: thank you! | 13:32 |
| alex_xu | johnthetubaguy: np | 13:32 |
| *** edmondsw has joined #openstack-nova | 13:32 | |
| *** otter768 has joined #openstack-nova | 13:34 | |
| openstackgerrit | jichenjc proposed openstack/nova: Make lock policy default to admin or owner https://review.openstack.org/242213 | 13:37 |
| *** diogogmt has quit IRC | 13:37 | |
| *** stevemar_ has joined #openstack-nova | 13:38 | |
| *** otter768 has quit IRC | 13:39 | |
| *** jerrygb_ has quit IRC | 13:39 | |
| *** diogogmt has joined #openstack-nova | 13:40 | |
| *** edleafe has joined #openstack-nova | 13:41 | |
| *** Piet has quit IRC | 13:45 | |
| *** alex_klimov has quit IRC | 13:47 | |
| *** alex_klimov has joined #openstack-nova | 13:47 | |
| *** garyk has quit IRC | 13:47 | |
| *** boris-42 has quit IRC | 13:48 | |
| *** su_zhang has joined #openstack-nova | 13:49 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add enable-pci-api-function.rst https://review.openstack.org/244666 | 13:49 |
| *** vladikr has quit IRC | 13:51 | |
| *** e0ne has quit IRC | 13:51 | |
| *** vladikr has joined #openstack-nova | 13:51 | |
| *** e0ne has joined #openstack-nova | 13:52 | |
| *** ptm_away is now known as PaulMurray | 13:54 | |
| *** BobBall has quit IRC | 13:54 | |
| johnthetubaguy | sdague: there is an API doc thing, linked to a project change form annegentle to get the API concept guide going, I think its ready to merge now: https://review.openstack.org/#/c/230186/ | 13:58 |
| sdague | johnthetubaguy: great, I'll take a look soon | 13:58 |
| johnthetubaguy | sdague: thanks | 13:58 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'migration' https://review.openstack.org/243496 | 13:58 |
| *** mdrabe has joined #openstack-nova | 13:59 | |
| johnthetubaguy | mikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal mnestratov kashyap aloga rgeragnov bauzas xyang tpatil med_ nic scottda nagyz belliott dguitarbite sdague jroll diana_clarke _diana_ raildo jichen | 13:59 |
| johnthetubaguy | gjayavelu mdorman klindgren sorrison belmoreira mgagne mrda mmmpork PaulMurray gcb meeting time | 14:00 |
| pkholkin | johnthetubaguy, sdague: Hello guys! Please take a look at updated patch in migration section https://review.openstack.org/#/c/243496/2. We decided to ignore use_slave parameter and not to change object version | 14:00 |
| sdague | pkholkin: there was another patch that had a similar issue, right? | 14:01 |
| *** haomaiwang has quit IRC | 14:01 | |
| dguitarbite | hello | 14:01 |
| *** lykinsbd has joined #openstack-nova | 14:01 | |
| *** haomaiwang has joined #openstack-nova | 14:01 | |
| *** diogogmt has quit IRC | 14:01 | |
| pkholkin | johnthetubaguy, sdague: yes, we marked it as wip now and will change soon | 14:01 |
| *** andreykurilin has left #openstack-nova | 14:01 | |
| *** andreykurilin has joined #openstack-nova | 14:02 | |
| pkholkin | johnthetubaguy, sdague: also, we added 'async' decorator there, it is needed for enginefacade to use slave_connection if it is | 14:02 |
| *** EinstCrazy has joined #openstack-nova | 14:03 | |
| *** e0ne has quit IRC | 14:04 | |
| *** mriedem has joined #openstack-nova | 14:04 | |
| openstackgerrit | Niall Bunting proposed openstack/nova: Update GlanceClient exceptions to new versions https://review.openstack.org/244673 | 14:04 |
| *** rlrossit has joined #openstack-nova | 14:06 | |
| *** Sree__ has quit IRC | 14:07 | |
| *** korzen has joined #openstack-nova | 14:08 | |
| edmondsw | tonyb, we spoke about https://bugs.launchpad.net/nova/+bug/1415087 in Tokyo. Did you push up a WIP patch and can you point me to it? | 14:10 |
| openstack | Launchpad bug 1415087 in OpenStack Compute (nova) "[OSSA 2015-011] Format-guessing and file disclosure in image convert (CVE-2015-1850, CVE-2015-1851)" [High,Incomplete] - Assigned to Tony Breeds (o-tony) | 14:10 |
| openstackgerrit | Matthew Booth proposed openstack/nova: Fix typo in test_post_select_populate https://review.openstack.org/244178 | 14:11 |
| *** salv-orlando has joined #openstack-nova | 14:12 | |
| *** ctrath has joined #openstack-nova | 14:12 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add SerialConsoleOps class https://review.openstack.org/184040 | 14:13 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Removes support for Hyper-V Server 2008 R2 https://review.openstack.org/237643 | 14:13 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 14:13 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console handler class https://review.openstack.org/184039 | 14:13 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 14:13 |
| *** e0ne has joined #openstack-nova | 14:13 | |
| *** w_verdugo has quit IRC | 14:15 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes *Utils modules and unit tests https://review.openstack.org/215917 | 14:15 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: config options: centralize section "serial_console" https://review.openstack.org/244177 | 14:17 |
| mriedem | Vek: we're planning a novaclient minor release so let's hold up on approving anything right now | 14:19 |
| *** EinstCrazy has quit IRC | 14:19 | |
| *** nelsnelson has joined #openstack-nova | 14:20 | |
| *** jichen has quit IRC | 14:20 | |
| *** marzif has quit IRC | 14:20 | |
| *** mc_nair has joined #openstack-nova | 14:21 | |
| *** rotbeard has quit IRC | 14:21 | |
| *** ijuwang has joined #openstack-nova | 14:22 | |
| *** BobBall has joined #openstack-nova | 14:22 | |
| *** jerrygb has joined #openstack-nova | 14:22 | |
| *** annegentle has joined #openstack-nova | 14:22 | |
| *** nelsnels_ has quit IRC | 14:23 | |
| *** jyuso1 has quit IRC | 14:23 | |
| *** cfriesen__ has joined #openstack-nova | 14:23 | |
| *** pm90_ has joined #openstack-nova | 14:24 | |
| openstackgerrit | Bob Ball proposed openstack/nova: XenAPI: Correct hypervisor type in Horizon's admin view https://review.openstack.org/237374 | 14:24 |
| mriedem | sdague: so novaclient release today or tuesday? | 14:25 |
| sdague | mriedem: I'm fine with today | 14:26 |
| *** MVenesio has quit IRC | 14:26 | |
| *** w_verdugo has joined #openstack-nova | 14:26 | |
| *** liverpooler has quit IRC | 14:28 | |
| *** zhangguoqing has quit IRC | 14:29 | |
| *** fawadkhaliq has joined #openstack-nova | 14:31 | |
| *** MVenesio has joined #openstack-nova | 14:31 | |
| *** xyang1 has joined #openstack-nova | 14:34 | |
| *** smatzek has quit IRC | 14:34 | |
| johnthetubaguy | mriedem: sdague: today is cool | 14:34 |
| mriedem | ok i posted a heads up to the -dev list | 14:34 |
| openstackgerrit | Gábor Antal proposed openstack/nova: Replaced deprecated timeutils methods https://review.openstack.org/241179 | 14:35 |
| openstackgerrit | Marian Horban proposed openstack/nova: Double detach volume causes server fault https://review.openstack.org/244683 | 14:35 |
| openstackgerrit | Marian Horban proposed openstack/nova: Double detach volume causes server fault https://review.openstack.org/244683 | 14:35 |
| *** lpetrut has quit IRC | 14:39 | |
| openstackgerrit | Andrey Kurilin proposed openstack/nova: [Py34] Enable api.openstack.test_wsgi unit test https://review.openstack.org/231076 | 14:39 |
| *** lpetrut has joined #openstack-nova | 14:39 | |
| *** su_zhang has quit IRC | 14:41 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova-specs: Pause VM during live migration https://review.openstack.org/229040 | 14:41 |
| BobBall | johnthetubaguy: Regarding https://review.openstack.org/#/c/231103/4/nova/virt/xenapi/volume_utils.py (sorry - I forgot you'd made comments there) - which bit specifically are you meaning to use six for? | 14:41 |
| mriedem | what's the story with the dr? | 14:43 |
| bauzas | mriedem: I guess it's http://www.doctorwho.tv/events/doctor-who-experience/ | 14:45 |
| mriedem | oh "doctor who" | 14:45 |
| mriedem | now it makes sense | 14:45 |
| *** EinstCrazy has joined #openstack-nova | 14:46 | |
| *** eharney has joined #openstack-nova | 14:51 | |
| *** jgjhgsonchn88523 has joined #openstack-nova | 14:51 | |
| *** ssurana has joined #openstack-nova | 14:51 | |
| openstackgerrit | Andrey Kurilin proposed openstack/nova: [Py34] api.openstack.test_faults.TestFaultWrapper https://review.openstack.org/231473 | 14:54 |
| openstackgerrit | Andrey Kurilin proposed openstack/nova: [Py34] api.openstack.test_faults.TestFaultWrapper https://review.openstack.org/231473 | 14:55 |
| *** _yonglihe has joined #openstack-nova | 14:57 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: pci utils: fix the check if pci device is physical_function https://review.openstack.org/227160 | 14:57 |
| *** whenry_ has quit IRC | 14:58 | |
| *** haomaiwang has quit IRC | 14:58 | |
| *** pm90_ is now known as pratikmallya | 14:58 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Replace N block_device_mapping queries with 1 https://review.openstack.org/229964 | 14:59 |
| johnthetubaguy | BobBall: its str(v) | 15:00 |
| *** dims has quit IRC | 15:01 | |
| *** EinstCrazy has quit IRC | 15:01 | |
| *** mjura has quit IRC | 15:01 | |
| *** smatzek has joined #openstack-nova | 15:02 | |
| mriedem | sdague: we might want to slip this into novaclient 2.35.0 https://review.openstack.org/#/c/232605/ | 15:02 |
| mriedem | sdague: left off my +W to see if you wanted to skim it | 15:02 |
| *** stackdump has joined #openstack-nova | 15:03 | |
| sdague | yeh, it's tests, I'd say land it | 15:03 |
| mriedem | yeah, my thinking too, +W | 15:04 |
| *** cdent has joined #openstack-nova | 15:04 | |
| *** korzen has quit IRC | 15:04 | |
| BobBall | johnthetubaguy: I think I must be missing something... str(int) works in both 2.7 and 3; why do we need to use six? | 15:04 |
| *** dims has joined #openstack-nova | 15:07 | |
| *** diogogmt has joined #openstack-nova | 15:07 | |
| bauzas | BobBall: because str != unicode in py2.7 | 15:07 |
| *** diogogmt has quit IRC | 15:07 | |
| mdbooth | PaulMurray: Does your cabal of live migrators have a weekly IRC meeting, yet? Looking at https://etherpad.openstack.org/p/mitaka-live-migration | 15:08 |
| *** stevemar_ has quit IRC | 15:08 | |
| *** diogogmt has joined #openstack-nova | 15:10 | |
| kashyap | mdbooth: Seems like it's on Tuesdays at 1400 UTC | 15:10 |
| mdbooth | kashyap: From which magicall well did you divine this? | 15:10 |
| mriedem | bauzas: py3 | 15:10 |
| kashyap | mdbooth: Heh, it was asked here yesterday | 15:10 |
| mdbooth | Ah, ok | 15:11 |
| *** sfinucan has joined #openstack-nova | 15:11 | |
| mriedem | mdbooth: ML thread on it too | 15:11 |
| bauzas | mriedem: aaaaand ? :) | 15:11 |
| kashyap | mdbooth: I just looked up my logs for that; also it was announced on the list, but well-buried. | 15:11 |
| *** njirap has joined #openstack-nova | 15:11 | |
| mriedem | nvm | 15:11 |
| mdbooth | mriedem: D'oh so there is | 15:11 |
| *** sneti has joined #openstack-nova | 15:12 | |
| *** lnxnut has joined #openstack-nova | 15:16 | |
| *** ssurana has quit IRC | 15:16 | |
| *** mrkz has joined #openstack-nova | 15:16 | |
| *** breitz has quit IRC | 15:18 | |
| *** signed8bit has joined #openstack-nova | 15:18 | |
| *** breitz has joined #openstack-nova | 15:18 | |
| *** EinstCrazy has joined #openstack-nova | 15:19 | |
| *** edtubill has joined #openstack-nova | 15:20 | |
| *** marzif has joined #openstack-nova | 15:21 | |
| *** ssurana has joined #openstack-nova | 15:21 | |
| *** thangp has joined #openstack-nova | 15:24 | |
| *** mpavone has quit IRC | 15:26 | |
| *** _yonglihe has quit IRC | 15:27 | |
| *** gcb_ has quit IRC | 15:27 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'service' https://review.openstack.org/241178 | 15:29 |
| *** sneti has quit IRC | 15:29 | |
| *** stevemar_ has joined #openstack-nova | 15:29 | |
| *** sneti has joined #openstack-nova | 15:29 | |
| *** eglynn has quit IRC | 15:29 | |
| *** mtanino has joined #openstack-nova | 15:31 | |
| mhorban_ | @sdague: Hello, Could you please look at https://review.openstack.org/#/c/238983/ ? It is fix of "services no longer reliably stop in stable/liberty / master" bug. | 15:31 |
| BobBall | bauzas: I'm still confused... What should I use instead of str(int) from six? I see the difference that py3 will return unicode strings, but I don't understand why that causes a problem in this case https://review.openstack.org/#/c/231103/4/nova/virt/xenapi/volume_utils.py and what I need to do to fix it? | 15:31 |
| PaulMurray | mdbooth, https://wiki.openstack.org/wiki/Meetings/NovaLiveMigration | 15:31 |
| mdbooth | PaulMurray: Found it, thanks | 15:31 |
| PaulMurray | mdbooth, anything for the agenda, let me know | 15:32 |
| mdbooth | PaulMurray: The other way round. I was hoping I might take something from the agenda and work on it. | 15:32 |
| mdbooth | We'll see. | 15:32 |
| bauzas | BobBall: so, I'm not johnthetubaguy so my point is that it's a style thing - not really a problem | 15:32 |
| *** nkrinner has quit IRC | 15:32 | |
| PaulMurray | mdbooth, cool - anything at all or do you have a topic in mind? | 15:33 |
| bauzas | BobBall: I mean, you can mix both and that will work - but like johnthetubaguy it's better to be explicit | 15:33 |
| *** mriedem is now known as mriedem_away | 15:33 | |
| mdbooth | PaulMurray: Nothing in particular. I've poked a fair amount of the code over time so I mostly know what's going on, and I just thought I might pitch in wherever. | 15:34 |
| BobBall | ok, but I still don't see which methods from six are actually helpful here... | 15:35 |
| *** otter768 has joined #openstack-nova | 15:35 | |
| BobBall | It's probably very obvious to you but I've not written much explicitely 2/3 compatible code :D | 15:35 |
| PaulMurray | mdbooth, there are specs to review at the moment and of course bugs. For the features we can ask if there is anything specific in the meeting. If not I'm sure I can find something interesting to get your teeth into | 15:36 |
| mdbooth | PaulMurray: Ok, will see. Bugs are fine, too. | 15:36 |
| PaulMurray | mdbooth, and thanks for the offer | 15:37 |
| bauzas | BobBall: six.text_type() ;) | 15:38 |
| BobBall | thanks. *goes looking* | 15:39 |
| *** mpavone has joined #openstack-nova | 15:39 | |
| johnthetubaguy | bauzas: I thought it broke for some reason, I could be wrong | 15:39 |
| *** otter768 has quit IRC | 15:39 | |
| bauzas | johnthetubaguy: no, it works | 15:40 |
| bauzas | johnthetubaguy: that's not stylistic, but that works | 15:40 |
| johnthetubaguy | BobBall: hmm, odd, OK then | 15:40 |
| BobBall | ah - I see - so the suggestion is six.text_type(v) rather than str(v) | 15:40 |
| johnthetubaguy | yeah | 15:40 |
| *** fawadkhaliq has quit IRC | 15:40 | |
| *** scheuran has quit IRC | 15:40 | |
| BobBall | Very happy to change that! thanks for the explanation | 15:40 |
| johnthetubaguy | BobBall: if it still works with str() on python34, then maybe we don't need it | 15:41 |
| johnthetubaguy | BobBall: the best way to fix this stuff is to get all the XenAPI tests passing on python34 | 15:41 |
| johnthetubaguy | BobBall: I know hyper-v have their whole compute driver running on python34 now, which is cool | 15:41 |
| BobBall | Indeed | 15:42 |
| *** klkumar has quit IRC | 15:42 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'compute_node' https://review.openstack.org/241189 | 15:42 |
| krotscheck | sdague, johnthetubaguy: Thanks for the +2's on the CORS patch :) | 15:42 |
| BobBall | well I can't currently even run the py27 unit tests on my local machine now - let alone py34 :D | 15:43 |
| *** baoli has quit IRC | 15:44 | |
| bauzas | BobBall: johnthetubaguy: to make it perfectly clear, 'foo'.join(collection(str(v)) works | 15:44 |
| bauzas | in py34 and py27 | 15:44 |
| BobBall | awesome bauzas - that's nice | 15:44 |
| johnthetubaguy | bauzas: yeah, I should read up on that more, I thought str() was always bad for some reason | 15:44 |
| *** burgerk has quit IRC | 15:45 | |
| johnthetubaguy | I guess str() is unicode in py34 | 15:45 |
| *** baoli has joined #openstack-nova | 15:46 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: enginefacade: 'compute_node' https://review.openstack.org/241189 | 15:46 |
| *** salv-orlando has quit IRC | 15:46 | |
| *** baoli has quit IRC | 15:48 | |
| bauzas | johnthetubaguy: yup | 15:48 |
| *** tonytan4ever has joined #openstack-nova | 15:49 | |
| bauzas | johnthetubaguy: well, it's a string to be clear, but it's unicoded | 15:49 |
| johnthetubaguy | bauzas: right | 15:50 |
| *** sfinucan has quit IRC | 15:50 | |
| *** sfinucan has joined #openstack-nova | 15:50 | |
| *** Piet has joined #openstack-nova | 15:51 | |
| *** achanda has joined #openstack-nova | 15:51 | |
| dansmith | johnthetubaguy: where is the spec to remove the use_slave stuff? | 15:52 |
| *** baoli has joined #openstack-nova | 15:52 | |
| *** aix_ has quit IRC | 15:52 | |
| *** claudiub has quit IRC | 15:53 | |
| johnthetubaguy | dansmith: so thats a good question, it seems to be happening as part of this: https://blueprints.launchpad.net/nova/+spec/new-oslodb-enginefacade | 15:54 |
| johnthetubaguy | http://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/oslo_db-enginefacade.html#code-changes | 15:54 |
| dansmith | okay | 15:54 |
| *** ccarmack has joined #openstack-nova | 15:55 | |
| johnthetubaguy | dansmith: I think it now moves to using the read decorator to imply use_slave by default | 15:55 |
| dansmith | so we just added that to everything for a couple years for seemingly no reason? :) | 15:55 |
| johnthetubaguy | although I think they must be assuming a syncronous read slave, at that point | 15:55 |
| dansmith | I guess I'm not sure how we can always know if we can read from the slave | 15:55 |
| mdbooth | dansmith: use_slave is different | 15:55 |
| *** pradk_ has joined #openstack-nova | 15:55 | |
| mdbooth | It means it's acceptable for it to be async | 15:55 |
| dansmith | if we're reading something that depends on synchronization with a write.. | 15:55 |
| mdbooth | i.e. a bit out of date | 15:55 |
| johnthetubaguy | well, I think that was the folks in utar, when you have a lagging slave | 15:55 |
| *** angdraug has joined #openstack-nova | 15:55 | |
| *** pradk_ is now known as pradk | 15:56 | |
| johnthetubaguy | mdbooth: yeah, thats what I thought | 15:56 |
| mdbooth | So it's only used by periodic tasks, I think | 15:56 |
| johnthetubaguy | mdbooth: yeah | 15:56 |
| mdbooth | IIRC there's an async_reader decorator which is separate | 15:56 |
| dansmith | mdbooth: well, it was supposed to be used by lots of stuff, we just don't ever seem to go through the thought process of whether reading out of data data is okay for each case | 15:56 |
| *** aix has joined #openstack-nova | 15:56 | |
| johnthetubaguy | dansmith: it did seem a little random | 15:56 |
| mdbooth | dansmith: The new decorator should make it easier. The design point is that it's a decision the developer should make, not a user. | 15:57 |
| dansmith | I guess my primary concern is that we're not going to start reading from slaves more than we are now, in places where it might break things | 15:57 |
| mdbooth | Nope | 15:57 |
| mdbooth | The reader decorator doesn't use async | 15:57 |
| dansmith | mdbooth: define developer and user in that case | 15:57 |
| mdbooth | Has to be done explicitly | 15:57 |
| dansmith | right, which is what use_slave was supposed to get us | 15:57 |
| johnthetubaguy | right, use_slave was explicit | 15:58 |
| dansmith | like, anything that reads flavors mostly anywhere other than api should be able to read from a slave | 15:58 |
| *** baoli has quit IRC | 15:58 | |
| mdbooth | developer == person writing the python code for a periodic task, and is writing it in the explicit knowledge that the data might be a bit out of date | 15:58 |
| dansmith | we sucked at ever thinking about it and making those places actually declare that it's okay | 15:58 |
| johnthetubaguy | I guess the idea is to use the decorator and context instead? | 15:58 |
| dansmith | see, I don't actually think that periodics can always use the slave, | 15:58 |
| johnthetubaguy | dansmith: you mean we didn't use it in enough places to be useful? | 15:59 |
| dansmith | because we need to know about task_state changes immediately to prevent periodics from racing with things | 15:59 |
| mdbooth | Anyway, it makes it easier to use or not use. | 15:59 |
| dansmith | johnthetubaguy: we didn't consider "can this run against the slave" ever.. we added it everywhere so we could start making that decision, but we never actually started marking things as "async okay" | 15:59 |
| johnthetubaguy | dansmith: oh, I thought we did add a few of those | 15:59 |
| mtanino | Hi, I'd like to know the policy of feature proposal at Nova. I'd like to add an option to existing API, in this case, do I need both SPEC and BP? Are there good reference to know Nova's poroposal policy? | 16:00 |
| dansmith | and we can't really make that call at the db layer only | 16:00 |
| mdbooth | dansmith: It's not for the db layer. It's for the layer above. | 16:00 |
| johnthetubaguy | so I thought it was the object method caller that needs to decide right? | 16:00 |
| *** baoli has joined #openstack-nova | 16:00 | |
| dansmith | mdbooth: we can't make it at the object layer | 16:00 |
| *** signed8bit is now known as signed8bit_ZZZzz | 16:00 | |
| johnthetubaguy | right, has to be outside the object later | 16:00 |
| dansmith | johnthetubaguy: right, it's the caller of the object thing | 16:00 |
| dansmith | i.e. the code doing a lookup of flavors, for example | 16:01 |
| dansmith | we can't make the call inside list_flavors() | 16:01 |
| dansmith | s/call/decision/ for disambiguation | 16:01 |
| mdbooth | It can live above that, even. If you decorate a chunk of a periodic task with async_reader, anything in there would use a slave if it was available | 16:01 |
| johnthetubaguy | so I think we might want to revise this spec... | 16:01 |
| dansmith | mdbooth: but the periodic tasks never talk to the database | 16:01 |
| johnthetubaguy | need to replace calls that use use_slave to use the new thing, at a minimum | 16:01 |
| dansmith | mdbooth: so you can't decorate them with anything db-related and have it honored | 16:01 |
| dansmith | johnthetubaguy: I'm not sure how they can | 16:01 |
| *** achanda has quit IRC | 16:02 | |
| mdbooth | dansmith: The decorator stores state in the request context | 16:02 |
| dansmith | unless there is some other communication channel I'm missing | 16:02 |
| mdbooth | So it percolates | 16:02 |
| dansmith | ah, okay | 16:02 |
| dansmith | so just to be clear, | 16:02 |
| dansmith | the object stores the context, | 16:02 |
| dansmith | so you have to look up the object with the async-ified context, | 16:02 |
| mdbooth | Yup | 16:02 |
| dansmith | and then everything else done to that object will have that async flag | 16:02 |
| mdbooth | And you can set that anywhere you have a request context | 16:03 |
| dansmith | unless you specifically alter the in-object context | 16:03 |
| *** achanda has joined #openstack-nova | 16:03 | |
| dansmith | and of course, | 16:03 |
| mdbooth | Actually... I wonder if this is a wart | 16:03 |
| dansmith | if you pass that object to another service over rpc, it will lose the async flag, which is probably desirable | 16:03 |
| dansmith | but needs to be obvious I think | 16:03 |
| mdbooth | So the db context is stored as a thread locally type thing on the request context | 16:04 |
| *** achanda has quit IRC | 16:04 | |
| mdbooth | And it's modified based on lexical scope | 16:04 |
| mdbooth | If an object is storing a request context | 16:04 |
| mdbooth | ... | 16:04 |
| *** signed8bit_ZZZzz is now known as signed8bit | 16:04 | |
| mdbooth | Its db context is going to be different depending on the lexical scope of the code calling it | 16:04 |
| dansmith | storing the context is inbuilt into the object machinery, in case it's not clear | 16:05 |
| dansmith | it's not like only some objects do that | 16:05 |
| mdbooth | Right | 16:05 |
| *** kfarr has joined #openstack-nova | 16:05 | |
| *** pratikmallya has quit IRC | 16:06 | |
| mdbooth | So, any object operation which fetches from the db will fetch using the context of the calling code. | 16:06 |
| dansmith | you mean object code that calls db.* methods directly, right? | 16:07 |
| mdbooth | Directly or indirectly, doesn't matter | 16:07 |
| mdbooth | If a db operation results using its request context... | 16:07 |
| openstackgerrit | Sven Anderson proposed openstack/nova: Fix is_volume_backed_instance() for unset image_ref https://review.openstack.org/243616 | 16:07 |
| johnthetubaguy | I think I am starting to remember why we didn't do this using context in the first place... | 16:08 |
| *** electrocucaracha has joined #openstack-nova | 16:08 | |
| dansmith | I'm confused about how this does or does not propagate over RPC I guess | 16:08 |
| mdbooth | It doesn't propagate over rpc | 16:08 |
| mdbooth | at all | 16:08 |
| dansmith | that seems to be my concern then | 16:08 |
| mdbooth | So it needs to? | 16:09 |
| dansmith | because everything that should be deciding whether something is async-tolerant is on the calling side of RPC | 16:09 |
| *** tongli has joined #openstack-nova | 16:09 | |
| dansmith | everything is an exaggeration, but .. most things | 16:09 |
| mdbooth | This is when your fetching an object remotely over rpc... | 16:10 |
| mdbooth | That's a good point | 16:10 |
| mdbooth | Which I don't think I considered | 16:10 |
| mdbooth | you're | 16:10 |
| *** baoli has quit IRC | 16:10 | |
| mdbooth | Ok, so it sounds like some use_slave will continue to exist at the rpc layer | 16:11 |
| dansmith | yes, which is all object/db operations for a whole set of services, | 16:11 |
| bauzas | mriedem_away: meh, when you're back, I'd like to discuss about your good point on https://review.openstack.org/#/c/242008/5/releasenotes/source/index.rst,cm and explain why master | 16:11 |
| dansmith | to be more services during an upgrade if we throw a switch | 16:11 |
| mdbooth | Right, I realise now | 16:11 |
| dansmith | mdbooth: I don't particularly like having to have use_slave as a parameter for everything, mind you, but I think whatever we do has to have rpc in mind | 16:12 |
| mdbooth | Yes, I think it will have to remain for rpc calls which need it. However, it doesn't need to go further than that, because those rpc calls can then set the appropriate context immediately. | 16:12 |
| mdbooth | And it will propagate without being passed explicitly. | 16:12 |
| *** ihrachys has quit IRC | 16:13 | |
| dansmith | mdbooth: that means every object method needs to take that param, | 16:13 |
| dansmith | but that it tells the db layer what to do based on the context, right? | 16:13 |
| * mdbooth isn't familiar with this part of rpc. Are all object methods also rpcs? | 16:14 | |
| * mdbooth is thinking the top-level entry into a service | 16:14 | |
| mdbooth | But yes, it could still be a lot | 16:14 |
| mdbooth | This doesn't add uses of use_slave, though. It just doesn't get rid of as many as I thought. | 16:15 |
| mdbooth | Hmm, although you'll then need to explicitly pull out the async context when making an rpc call | 16:16 |
| *** thedodd has joined #openstack-nova | 16:16 | |
| mdbooth | Because the db facade definitely doesn't do that for you... | 16:16 |
| *** thedodd has quit IRC | 16:16 | |
| *** stevemar_ has quit IRC | 16:17 | |
| *** davideagnello has quit IRC | 16:17 | |
| *** mriedem_away is now known as mriedem | 16:17 | |
| mriedem | bauzas: i'm here now | 16:17 |
| mdbooth | dansmith: Yes, I think use_slave requires a rethink. | 16:17 |
| *** davideagnello has joined #openstack-nova | 16:17 | |
| *** links has quit IRC | 16:18 | |
| *** stevemar_ has joined #openstack-nova | 16:18 | |
| mdbooth | Probably not a big one, though. | 16:18 |
| dansmith | mdbooth: any object method that is remotable is an rpc yes | 16:18 |
| mdbooth | Right, we're not going to break that. | 16:19 |
| *** eglynn has joined #openstack-nova | 16:19 | |
| mdbooth | dansmith: I think a conservative first implementation, which is what this is, should leave use_slave in place. | 16:19 |
| bauzas | mriedem: so reno looks at the branches | 16:19 |
| mriedem | ah | 16:19 |
| bauzas | mriedem: not at the files | 16:19 |
| mdbooth | We'll then add a little decorator in sqla/api.py which pulls it out of kwargs and set @reader or @async_reader according to passed args | 16:20 |
| bauzas | mriedem: http://docs.openstack.org/developer/reno/sphinxext.html#sphinx-extension | 16:20 |
| mdbooth | That's functionally equivalent to what we have now | 16:20 |
| mdbooth | And leaves scope for improvement | 16:20 |
| bauzas | mriedem: like, when we'll release and cut the new mitaka branch, we'll have to add a new rst file asking to look at the mitaka branch | 16:20 |
| bauzas | mriedem: like I did for the liberty branch | 16:20 |
| mdbooth | Does that make sense? Where do you want me to write that? | 16:21 |
| *** mdrabe has quit IRC | 16:21 | |
| bauzas | mriedem: in the meantime, we can ask guys to write notes that 'll be seen in that specific page | 16:21 |
| *** kbyrne has joined #openstack-nova | 16:21 | |
| mriedem | bauzas: yeah, i guess we could also do :version: 13.0.0 | 16:21 |
| mriedem | but maybe that's getting ahead of ourselves | 16:22 |
| bauzas | mriedem: yeah that's another option | 16:22 |
| mriedem | when we start cutting stable point releases, i think we'll need to use :version: | 16:22 |
| bauzas | mriedem: right | 16:22 |
| mriedem | since stable/liberty could be 12.0.0 or 12.1.0 | 16:22 |
| bauzas | mriedem: yup | 16:22 |
| bauzas | mriedem: so, glance just discovered they actually needed that kind of 'show me the current' patch for the CI | 16:22 |
| *** burgerk has joined #openstack-nova | 16:23 | |
| bauzas | mriedem: hence me providing it | 16:23 |
| mriedem | bauzas: not following, you mean this https://review.openstack.org/#/c/242008/5/releasenotes/notes/start-using-reno-cd0063e9cdb87daa.yaml? | 16:23 |
| bauzas | mriedem: sec | 16:24 |
| snikitin | johnthetubaguy: Hi John. I want to ask you about very old patch 'instance_tags'. https://review.openstack.org/#/c/128940/ In Liberty it was blocked because Jay didn't want to use extensions. | 16:25 |
| bauzas | mriedem: http://lists.openstack.org/pipermail/openstack-dev/2015-November/078940.html | 16:25 |
| snikitin | johnthetubaguy:I changed my code to use plugin mechanism. In this way tags formally was an extension (it was just a child class of Extension), but it was a core extension, always turned on. | 16:25 |
| *** emagana has joined #openstack-nova | 16:25 | |
| snikitin | johnthetubaguy: So we planed that in Mitaka we will use some new way to create new API controllers without any extensions. Do we have this new way? if not, how I can create new API controller? | 16:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Replace N block_device_mapping queries with 1 https://review.openstack.org/229964 | 16:26 |
| *** mpavone has quit IRC | 16:27 | |
| bauzas | mriedem: oh I even see I should backport my 'master' change to liberty... | 16:27 |
| johnthetubaguy | snikitin: I wasn't sure what jaypipes was thinking there, it should be clearer in the code base now, it just needs to live here: https://github.com/openstack/nova/tree/master/nova/api/openstack/compute | 16:27 |
| mriedem | bauzas: yeah dhellmann called it 'unreleased' https://review.openstack.org/#/c/243302/1/releasenotes/source/unreleased.rst | 16:27 |
| dansmith | johnthetubaguy: so that means we put use_slave back into the patch you asked for it to be removed, yes? :) | 16:28 |
| bauzas | mriedem: yup, I'm going to follow that pattern | 16:28 |
| johnthetubaguy | dansmith: yeah, sorry about that | 16:28 |
| bauzas | mriedem: that's not very sexy, that's basically copy/paste from other projects :/ | 16:28 |
| dansmith | diana_clarke: ^ | 16:28 |
| johnthetubaguy | dansmith: do you have the link for that? | 16:28 |
| diana_clarke | dansmith: thanks | 16:28 |
| johnthetubaguy | I can go find it | 16:28 |
| mriedem | bauzas: i'm not sure it needs to be sexy | 16:29 |
| diana_clarke | dansmith: not that it was actrually using 'use_slave' before, but I guess someone could | 16:29 |
| *** MVenesio_ has joined #openstack-nova | 16:29 | |
| *** MVenesio has quit IRC | 16:30 | |
| dansmith | diana_clarke: it's an rpc interface, so getting it right first is beneficial | 16:30 |
| *** emagana has quit IRC | 16:30 | |
| *** emagana_ has joined #openstack-nova | 16:30 | |
| *** jdurgin1 has joined #openstack-nova | 16:31 | |
| johnthetubaguy | diana_clarke: ping me when ready for review again, I was basically +2 on that apart from the use_slave thing | 16:32 |
| *** garyk has joined #openstack-nova | 16:32 | |
| *** EinstCrazy has quit IRC | 16:35 | |
| pkholkin | johnthetubaguy: could you please take another look at the patch related to use_slave too) https://review.openstack.org/#/c/243496/ | 16:35 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add "unreleased" release notes page https://review.openstack.org/242008 | 16:35 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add reno for release notes management https://review.openstack.org/242007 | 16:35 |
| pkholkin | we changed two similar patches too | 16:35 |
| *** ssurana has quit IRC | 16:36 | |
| *** hemnafk is now known as hemna | 16:36 | |
| bauzas | mriedem: ^ | 16:36 |
| bauzas | mriedem: I made the changes to look like other changes | 16:36 |
| bauzas | mriedem: I'm not backporting them to the liberty branhc | 16:36 |
| bauzas | now* | 16:36 |
| *** kmartin__ has quit IRC | 16:36 | |
| mriedem | bauzas: +2 | 16:37 |
| bauzas | mriedem: I would hold until Monday to +W them | 16:38 |
| mriedem | why? | 16:38 |
| *** sacharya has joined #openstack-nova | 16:39 | |
| *** rdopiera has quit IRC | 16:39 | |
| *** fawadkhaliq has joined #openstack-nova | 16:40 | |
| *** mwagner has quit IRC | 16:41 | |
| bauzas | mriedem: well, nvm the CI job is not yet merged... | 16:41 |
| *** pratikmallya has joined #openstack-nova | 16:41 | |
| *** paul-carlton has quit IRC | 16:41 | |
| *** mdrabe has joined #openstack-nova | 16:42 | |
| *** romainh has left #openstack-nova | 16:42 | |
| *** r-daneel has joined #openstack-nova | 16:42 | |
| *** kmartin has joined #openstack-nova | 16:42 | |
| openstackgerrit | Bob Ball proposed openstack/nova: XenAPI: Workaround for 6.5 iSCSI bug https://review.openstack.org/231103 | 16:43 |
| *** moshele has quit IRC | 16:45 | |
| *** mgoddard__ has joined #openstack-nova | 16:45 | |
| *** pratikmallya has quit IRC | 16:45 | |
| *** jdurgin1 has quit IRC | 16:45 | |
| *** pratikmallya has joined #openstack-nova | 16:46 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Replace N block_device_mapping queries with 1 https://review.openstack.org/229964 | 16:46 |
| dansmith | johnthetubaguy: ^ | 16:47 |
| openstackgerrit | Niall Bunting proposed openstack/nova: Update GlanceClient exceptions to new versions https://review.openstack.org/244673 | 16:47 |
| *** mgoddard_ has quit IRC | 16:48 | |
| *** alexschm has quit IRC | 16:48 | |
| *** aix has quit IRC | 16:49 | |
| *** annegentle has quit IRC | 16:49 | |
| *** annegentle has joined #openstack-nova | 16:49 | |
| *** rcernin has quit IRC | 16:50 | |
| *** ctrath has quit IRC | 16:52 | |
| *** jinxing has quit IRC | 16:53 | |
| bauzas | mriedem: so, everything is cleaned up now, I'm going to write the CI patch you proposed by only running tempest and tests when no relnodes are touched | 16:53 |
| *** rlrossit has quit IRC | 16:53 | |
| mriedem | sweet | 16:54 |
| *** ctrath has joined #openstack-nova | 16:55 | |
| *** rlrossit has joined #openstack-nova | 16:55 | |
| *** alex_klimov has quit IRC | 16:57 | |
| diana_clarke | dansmith: I'm nervous about the hash change in test_objects.py. When I rebased that patch today, I need to update it & I don't fully understand why. | 16:57 |
| diana_clarke | dansmith: https://review.openstack.org/#/c/229964/14..15/nova/tests/unit/objects/test_objects.py | 16:57 |
| *** mgoddard has joined #openstack-nova | 16:58 | |
| *** armax has joined #openstack-nova | 16:58 | |
| *** mgoddard__ has quit IRC | 16:58 | |
| *** marzif has quit IRC | 16:59 | |
| dansmith | diana_clarke: it's because you changed the args of the remotable, right? | 16:59 |
| *** josecastroleon has quit IRC | 16:59 | |
| *** marzif has joined #openstack-nova | 17:00 | |
| diana_clarke | dansmith: Oh!!! I think it's finally coming together for me. Thanks! Also, I need to revert that then. | 17:00 |
| *** ihrachys has joined #openstack-nova | 17:00 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Replace N block_device_mapping queries with 1 https://review.openstack.org/229964 | 17:02 |
| *** stackdump has quit IRC | 17:02 | |
| *** e0ne has quit IRC | 17:02 | |
| mriedem | johnthetubaguy: fyi, soon we'll need a performance team CPL :) https://wiki.openstack.org/wiki/Meetings/Performance#Agenda_for_next_meeting | 17:04 |
| mriedem | i added conductor to that since it's come up twice now in the ops list as a performance issue | 17:05 |
| *** marzif has quit IRC | 17:05 | |
| *** marzif has joined #openstack-nova | 17:06 | |
| *** pcaruana has quit IRC | 17:06 | |
| *** lykinsbd has quit IRC | 17:06 | |
| johnthetubaguy | mriedem: yeah, unsure how much the pymsql stuff has helped us there | 17:06 |
| mriedem | did you see zombie mysqldb? https://review.openstack.org/#/c/243822/ | 17:06 |
| johnthetubaguy | mriedem: I know belliott was going to take a look at that | 17:06 |
| *** jlanoux has quit IRC | 17:07 | |
| johnthetubaguy | alaski: did we try pymysql yet, instead of the mysqldb stuff? | 17:07 |
| alaski | I don't think we have | 17:07 |
| *** baoli has joined #openstack-nova | 17:07 | |
| alaski | I haven't heard of it at least | 17:08 |
| johnthetubaguy | alaski: ah, so we only test pymysql in the gate now right, and its evetlet safe, but slower, but in theory it should mean the nova-cells workers and conductor get more concurrency per process than the old C lib stuff, in theory | 17:09 |
| *** njirap has quit IRC | 17:09 | |
| dansmith | johnthetubaguy: why slower? | 17:09 |
| dansmith | I think it's slower at some things, but ... | 17:09 |
| *** baoli_ has joined #openstack-nova | 17:09 | |
| dansmith | johnthetubaguy: you know we were able to halve a bunch of worker values because of that move right? | 17:09 |
| johnthetubaguy | dansmith: yeah, I mean slower at some things, but its no longer blocking, as I understood it | 17:09 |
| dansmith | we couldn't quite halve conductor, but we did reduce it | 17:09 |
| johnthetubaguy | dansmith: right | 17:10 |
| *** baoli has quit IRC | 17:12 | |
| *** exploreshaifali has joined #openstack-nova | 17:13 | |
| mriedem | johnthetubaguy: it's not only that, mysql-python doesn't support py3 | 17:14 |
| mriedem | pymysql does | 17:14 |
| *** gszasz has quit IRC | 17:15 | |
| *** rmart04 has joined #openstack-nova | 17:16 | |
| *** rk4n has quit IRC | 17:17 | |
| alaski | johnthetubaguy: our biggest issue is not really having a good testbed for comparing these things. It would be amazing if we could get something setup and get some actual numbers | 17:18 |
| *** mwagner has joined #openstack-nova | 17:20 | |
| mriedem | would testing it against the large ops job results help? | 17:20 |
| *** ssurana has joined #openstack-nova | 17:20 | |
| *** marzif has quit IRC | 17:20 | |
| *** rmart04 has quit IRC | 17:21 | |
| *** rmart04 has joined #openstack-nova | 17:21 | |
| johnthetubaguy | mriedem: agreed about py34, I think that was the deciding vote for pymysql | 17:22 |
| alaski | mriedem: as long as not too much is faked out. I don't remember what actually gets missed in the fake driver in that test. | 17:22 |
| johnthetubaguy | mriedem: I think large ops is worth a try, I thought the issue was more about the unpredictable performance of the VM it lands on | 17:22 |
| mriedem | alaski: just the hypervisor stuff really | 17:23 |
| johnthetubaguy | I do wonder about getting like a single on metal box, and running some test in a loop to check for regressions, i.e. trying to remove the VM variability out of the equation. Or maybe just a 30GB standard VM would do the trick | 17:23 |
| mriedem | alaski: it hits all the same nova db stuff (except probably volumes) | 17:24 |
| *** rcernin has joined #openstack-nova | 17:24 | |
| alaski | mriedem: that seems like a good start then | 17:25 |
| alaski | we recently switched off the mysqldb driver and ran into issues with instance_update_and_get_original backing up in the nova cells service | 17:25 |
| *** apoorvad has joined #openstack-nova | 17:25 | |
| dansmith | mriedem: what do you mean? we're running largeops with pymysql now aren't we? | 17:26 |
| alaski | and it wasn't much slower than the mysqldb version, but just enough to cause a wait when a lot came in | 17:26 |
| *** rmart04_ has joined #openstack-nova | 17:26 | |
| *** afaranha has left #openstack-nova | 17:26 | |
| sdague | danpb: do you have thoughts about why we're getting this error back from libvirt so much in the gate - "libvirtError: Error while building firewall: Some rules could not be created for interface vnet5: Failure to execute command '$EBT -t nat -A libvirt-P-vnet5 -p 0x0800 -j P-vnet5-ipv4' : 'Illegal target name 'P-vnet5-ipv4'.'." | 17:26 |
| mriedem | dansmith: i meant a DNM patch to test out https://review.openstack.org/#/c/243822/ | 17:27 |
| mriedem | like in devstack | 17:27 |
| dansmith | okay | 17:27 |
| *** rmart04 has quit IRC | 17:28 | |
| *** rmart04_ is now known as rmart04 | 17:28 | |
| *** nic has joined #openstack-nova | 17:30 | |
| *** suro-patz has joined #openstack-nova | 17:31 | |
| johnthetubaguy | alaski: I suspect if we moved to pymysql it might be a different story? | 17:32 |
| *** josecastroleon has joined #openstack-nova | 17:32 | |
| alaski | johnthetubaguy: it seems likely that we would get different performance characteristics. Whether or not it's enough of an improvement to work for us is hard to say without some testing | 17:34 |
| *** stackdump has joined #openstack-nova | 17:35 | |
| *** EinstCrazy has joined #openstack-nova | 17:35 | |
| sdague | is it just that the nwfilterDefineXML is async, and sometimes we get to the guest definition before that's taken | 17:36 |
| johnthetubaguy | alaski: agreed with the need to test it to find out | 17:36 |
| *** otter768 has joined #openstack-nova | 17:36 | |
| *** paul-carlton has joined #openstack-nova | 17:36 | |
| danpb | sdague: IIRC, "Illegal target name" means the target does not exist | 17:37 |
| danpb | sdague: which could be a sign that something else is racing with libvirt for firewall access | 17:37 |
| dansmith | johnthetubaguy: we need an ack from you on this before we can land it: https://review.openstack.org/#/c/190399/ | 17:38 |
| dansmith | and a release note about having to do it | 17:38 |
| sdague | danpb: right, we've got a bunch of nwfilterDefineXML calls (looks like 5 or so) to set up the nwfilter before guest start | 17:38 |
| danpb | sdague: seems like the version of libvirt you run there isn;t using ebtables --concurrent flag for locking | 17:38 |
| johnthetubaguy | dansmith: taking a peak | 17:39 |
| danpb | so its possible newer libvirt would not have the problem | 17:39 |
| dansmith | johnthetubaguy: could you also peek? | 17:39 |
| sdague | danpb: that is also true, but that's not what's failing in this case | 17:39 |
| sdague | http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/screen-n-cpu.txt.gz?level=TRACE#_2015-11-12_16_10_36_859 | 17:40 |
| *** MVenesio has joined #openstack-nova | 17:40 | |
| *** spandhe has joined #openstack-nova | 17:40 | |
| sdague | I am wondering if it's possible that the nwfilterDefineXML operations are all async, and thus backing up, then we request the guest gets created with a reference to the nwfilter, and devices we expect to be there are not | 17:41 |
| *** MVenesio_ has quit IRC | 17:41 | |
| johnthetubaguy | dansmith: heh | 17:42 |
| sdague | this seems to be a distinct failure different from the ebtables --concurrent issue | 17:42 |
| *** otter768 has quit IRC | 17:42 | |
| *** ccesario has quit IRC | 17:42 | |
| *** EinstCrazy has quit IRC | 17:42 | |
| sdague | maybe just understanding who / what creates the P-vnet5-ipv4 table entry | 17:42 |
| *** aix has joined #openstack-nova | 17:43 | |
| danpb | sdague: if you look in libvirtd.log.gz you'll see | 17:43 |
| danpb | This log message shows the bit where we create that table entry | 17:44 |
| danpb | 2015-11-12 16:10:36.690+0000: 15521: debug : virCommandRunAsync:2282 : About to run /bin/sh -c 'EBT="/sbin/ebtables" | 17:44 |
| *** tonytan4ever has quit IRC | 17:44 | |
| danpb | and the very next comment libvirt runs in that thread shows the failure | 17:44 |
| danpb | 2015-11-12 16:10:36.742+0000: 15521: error : virCommandWait:2399 : internal error: Child process (/bin/sh -c 'EBT="/sbin/ebtables" | 17:44 |
| danpb | this seems to rather suggest something else changing ebtables has raced with libvirt and killed the chain | 17:45 |
| dansmith | nova messes with ebtables itself at times, right? | 17:45 |
| danpb | yeah, that would be the prime suspect | 17:46 |
| sdague | will any touch of ebtables scramble that? | 17:46 |
| dansmith | danpb: are the calls we make to libvirt that touch that synchronous, or does the daemon async some of it? | 17:46 |
| *** david-lyle has quit IRC | 17:46 | |
| danpb | dansmith: iirc they're asynchronous | 17:47 |
| dansmith | then that's got to be it | 17:47 |
| danpb | libvirt will serialize everything it does, but not serialize with external things | 17:47 |
| danpb | this is why I fear lack of the --concurrent flag might be causing us the problem here | 17:47 |
| *** ccesario has joined #openstack-nova | 17:48 | |
| sdague | danpb: can it race with itself? | 17:48 |
| danpb | no, libvirt has a mutex that it serializes all its own calls on | 17:48 |
| danpb | and the logs don't show any other libvirt threads doing iptables stuff in this time period anyway | 17:49 |
| *** markmc has quit IRC | 17:49 | |
| *** e0ne has joined #openstack-nova | 17:50 | |
| openstackgerrit | Bob Ball proposed openstack/nova: XenAPI: Cope with more Cinder backends https://review.openstack.org/244789 | 17:50 |
| sdague | http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_08_30_956 | 17:50 |
| sdague | that's the assync command | 17:50 |
| sdague | which includes an rm_chain on libvirt-J-vnet5 | 17:51 |
| sdague | then 39ms later there is - http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_08_30_985 | 17:51 |
| sdague | which is where the fail is | 17:51 |
| danpb | sdague: libvirt-J-vnet5 is not the problem one - it was P-vnet5-ipv4 that has the fatal error | 17:53 |
| *** cdent has quit IRC | 17:53 | |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Cancel ongoing live migrations https://review.openstack.org/228828 | 17:53 |
| danpb | sdague: the error you point to is expected - libvirt is just cleaning up any historical cruft at that point & ignores that particular error | 17:54 |
| openstackgerrit | Bob Ball proposed openstack/nova: XenAPI: Cope with more Cinder backends https://review.openstack.org/244789 | 17:54 |
| *** derekh has quit IRC | 17:54 | |
| *** IanGovett has quit IRC | 17:55 | |
| sdague | ok, I guess we're running this big ebtables block here - http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_08_30_985 | 17:56 |
| sdague | and then about 150ms later get the failure - http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_10_36_742 | 17:57 |
| sdague | is that failure virCommandWait an async return, or another command? | 17:57 |
| *** su_zhang has joined #openstack-nova | 17:58 | |
| sdague | the only ebtables bits that nova does are the dhcp blocks in -t filter, so actual racing with libvirt which seems to only use -t nat seems weird | 18:01 |
| sdague | and I actually think in the gate, we skip all the nova ebtables code paths entirely | 18:01 |
| sdague | oh, nope there is one nat rule | 18:02 |
| *** signed8bit is now known as signed8bit_ZZZzz | 18:03 | |
| *** su_zhang has quit IRC | 18:03 | |
| danpb | sdague: presumably you meant to point to this in your first link http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_10_36_690 | 18:03 |
| sdague | danpb: yep, copy/paste link fail | 18:04 |
| *** su_zhang has joined #openstack-nova | 18:04 | |
| * danpb hates the eb/iptables code in this particular version of libvirt - debugging the auto-generated shell scripts is horrible | 18:05 | |
| sdague | so, yeh, we don't hit the code path for -t nat in the gate on ebtables, and I'm not even clear if we hit the ebtables bits of nova in our config (I've never seen anything in ebtables outside of the nat table) | 18:05 |
| danpb | do we do *anything* ebtables related in the gate | 18:06 |
| danpb | or pehraps nova-network/neutron do something | 18:06 |
| sdague | I'm looking at the nova net code | 18:06 |
| *** PaulMurray is now known as ptm_away | 18:07 | |
| sdague | we definitely don't ever do this - https://github.com/openstack/nova/blob/02de17c72ac0f0824df2b3d460739a58cb6b5285/nova/network/linux_net.py#L738 | 18:07 |
| sdague | because that depends on a forced config we don't have | 18:07 |
| *** claudiub has joined #openstack-nova | 18:08 | |
| sdague | oh, the floating ips code puts stuff in nat | 18:09 |
| *** lykinsbd has joined #openstack-nova | 18:09 | |
| *** jichen has joined #openstack-nova | 18:09 | |
| sdague | though not going near any of the interfaces listed | 18:10 |
| sdague | https://github.com/openstack/nova/blob/02de17c72ac0f0824df2b3d460739a58cb6b5285/nova/network/linux_net.py#L832-L836 | 18:10 |
| *** lucasagomes is now known as lucas-afk | 18:10 | |
| *** paul-carlton has quit IRC | 18:11 | |
| *** sahid has quit IRC | 18:12 | |
| *** ccarmack has quit IRC | 18:13 | |
| *** sfinucan has quit IRC | 18:15 | |
| *** sfinucan has joined #openstack-nova | 18:15 | |
| *** ccarmack has joined #openstack-nova | 18:16 | |
| sdague | well, I need to find lunch for reals. Anyway, trying to figure out if this is a preventable issue currently, because this one is pretty prevalent | 18:16 |
| sdague | 8 fails in 24 hrs / 177 fails in 10 days | 18:16 |
| *** josecastroleon has quit IRC | 18:16 | |
| *** ctrath has quit IRC | 18:16 | |
| sdague | https://bugs.launchpad.net/nova/+bug/1501366 | 18:17 |
| openstack | Launchpad bug 1501366 in OpenStack Compute (nova) "libvirtError: Error while building firewall: Some rules could not be created for interface" [High,Confirmed] | 18:17 |
| *** electrocucaracha has quit IRC | 18:17 | |
| *** vilobhmm has joined #openstack-nova | 18:19 | |
| *** ctrath has joined #openstack-nova | 18:19 | |
| *** bnemec has quit IRC | 18:21 | |
| *** ctrath has quit IRC | 18:21 | |
| *** rook- has joined #openstack-nova | 18:22 | |
| *** rlrossit has quit IRC | 18:22 | |
| *** ccarmack has quit IRC | 18:22 | |
| *** jistr has quit IRC | 18:22 | |
| *** sfinucan has quit IRC | 18:22 | |
| *** lykinsbd_ has joined #openstack-nova | 18:23 | |
| *** rk4n has joined #openstack-nova | 18:25 | |
| *** lykinsbd has quit IRC | 18:26 | |
| *** eme has joined #openstack-nova | 18:26 | |
| eme | Hi I am facing issue connecting to nova vm. | 18:26 |
| *** achanda has joined #openstack-nova | 18:26 | |
| eme | I have added icmp, tcp rule but still can't ping or ssh to it. | 18:26 |
| eme | Can anyone please help? | 18:26 |
| *** ccarmack has joined #openstack-nova | 18:27 | |
| *** ctrath has joined #openstack-nova | 18:27 | |
| *** rlrossit has joined #openstack-nova | 18:27 | |
| *** rk4n has quit IRC | 18:27 | |
| *** tonytan4ever has joined #openstack-nova | 18:27 | |
| *** paul-carlton has joined #openstack-nova | 18:28 | |
| *** browne has joined #openstack-nova | 18:29 | |
| *** lykinsbd has joined #openstack-nova | 18:29 | |
| *** lykinsbd has quit IRC | 18:30 | |
| *** nic has quit IRC | 18:30 | |
| *** lykinsbd_ has quit IRC | 18:31 | |
| *** nic has joined #openstack-nova | 18:31 | |
| *** lykinsb__ has joined #openstack-nova | 18:31 | |
| *** yassine__ has quit IRC | 18:31 | |
| *** tdurakov has joined #openstack-nova | 18:32 | |
| tdurakov | sdague, hi, are you around? | 18:32 |
| *** su_zhang has quit IRC | 18:32 | |
| *** ev1lkry has joined #openstack-nova | 18:32 | |
| dansmith | tdurakov: reportedly lunching at the moment :) | 18:32 |
| ev1lkry | everything I have read states that when editing the policy.json for nova that no services need to be restarted but when doing "compute_extension:flavormanage": [ ], it still says that the polixy doesnt hallow compute_extension:flavormanage to be perfomed on my other users | 18:32 |
| ev1lkry | i've edited the /nova/policy.json file | 18:33 |
| dansmith | ev1lkry: nova-api will need to be restarted at least | 18:33 |
| tdurakov | dansmith, oh, thank you:) | 18:33 |
| *** smatzek has quit IRC | 18:33 | |
| ev1lkry | dansmith that was it thank you | 18:34 |
| *** ijuwang_ has joined #openstack-nova | 18:36 | |
| *** ijuwang has quit IRC | 18:36 | |
| *** ijuwang_ is now known as ijuwang | 18:37 | |
| *** smatzek has joined #openstack-nova | 18:37 | |
| *** yamahata has joined #openstack-nova | 18:38 | |
| mriedem | ev1lkry: are there docs that are wrong somewhere? | 18:39 |
| *** fawadkhaliq has quit IRC | 18:39 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Removes support for Hyper-V Server 2008 R2 https://review.openstack.org/237643 | 18:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: adds os-win library https://review.openstack.org/212102 | 18:40 |
| *** signed8bit_ZZZzz is now known as signed8bit | 18:41 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Revert "Exclude all BDM checks for cells" https://review.openstack.org/242698 | 18:41 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: WIP: cells: create block device mappings in API cell https://review.openstack.org/242649 | 18:41 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Make BlockDeviceMapping object support uuid https://review.openstack.org/242603 | 18:41 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add uuid column to BlockDeviceMapping https://review.openstack.org/242602 | 18:41 |
| *** zul has quit IRC | 18:41 | |
| ev1lkry | dansmith : I have read in some non openstack.org sites that you do not need to restart services when editing the policy.json file however the docs i've see doesnt say either for example: http://docs.openstack.org/kilo/config-reference/content/policy-json-file.html | 18:42 |
| *** zul has joined #openstack-nova | 18:42 | |
| alaski | ev1lkry: for some projects that might be true, like Glance I think, but not for Nova | 18:42 |
| *** EmilienM has quit IRC | 18:42 | |
| *** ociuhandu has quit IRC | 18:42 | |
| *** sfinucan has joined #openstack-nova | 18:43 | |
| *** electrocucaracha has joined #openstack-nova | 18:44 | |
| *** su_zhang has joined #openstack-nova | 18:45 | |
| mriedem | ev1lkry: docs bug reported https://bugs.launchpad.net/openstack-manuals/+bug/1515729 | 18:45 |
| openstack | Launchpad bug 1515729 in openstack-manuals "Appendix A. The policy.json file in OpenStack Configuration Reference - liberty" [Undecided,New] | 18:45 |
| *** djchou has joined #openstack-nova | 18:45 | |
| *** EmilienM has joined #openstack-nova | 18:45 | |
| ev1lkry | awesome, thanks mridem | 18:46 |
| mriedem | sdague: dims: when we do novaclient 3.0 we should also drop py26 support | 18:46 |
| mriedem | since oslo wants to do that anyway | 18:46 |
| *** e0ne has quit IRC | 18:47 | |
| *** su_zhang has quit IRC | 18:47 | |
| *** rmart04 has quit IRC | 18:47 | |
| *** danpb has quit IRC | 18:48 | |
| sfinucan | mriedem: could you have a peek at this modification to the CPU thread pinning spec? It's not "trivial", but johnthetubaguy has signed off on it and the changes are minimal https://review.openstack.org/#/c/244198/3/specs/mitaka/approved/virt-driver-cpu-pinning.rst | 18:48 |
| sfinucan | figure I can't get any code merged on that until the specs line up | 18:49 |
| *** lpetrut has quit IRC | 18:49 | |
| sfinucan | also, johnthetubaguy: should I/can I rework the original spec to remove the references to CPU thread pinning? What's there now is out-of-date for a number of reasons (policies have changed names, wrong release etc.) | 18:50 |
| mriedem | sfinucan: who originally approved it? | 18:50 |
| mriedem | i'd ask them | 18:50 |
| *** Marga_ has quit IRC | 18:50 | |
| mriedem | i don't have context | 18:50 |
| sfinucan | I thought you did? | 18:50 |
| mriedem | i think i approved a re-approval from liberty | 18:50 |
| *** yamahata has quit IRC | 18:50 | |
| sfinucan | Ah, that's my mistake | 18:50 |
| sfinucan | OK, I'll do that. Sorry for the interruption | 18:51 |
| mriedem | np | 18:51 |
| *** su_zhang has joined #openstack-nova | 18:51 | |
| *** tdurakov has quit IRC | 18:52 | |
| *** lykinsbd has joined #openstack-nova | 18:54 | |
| *** paul-carlton has quit IRC | 18:54 | |
| *** tdurakov has joined #openstack-nova | 18:55 | |
| *** mnestratov has quit IRC | 18:55 | |
| *** electrocucaracha has quit IRC | 18:56 | |
| *** tonytan_brb has joined #openstack-nova | 18:56 | |
| *** lykinsb__ has quit IRC | 18:56 | |
| *** lykinsbd has quit IRC | 18:56 | |
| *** Piet has quit IRC | 18:57 | |
| *** lykinsbd has joined #openstack-nova | 18:57 | |
| sfinucan | alaski: Would you be able to look at this spec? https://review.openstack.org/#/c/244198/3/ | 18:58 |
| *** mriedem has quit IRC | 18:58 | |
| *** angdraug has quit IRC | 18:58 | |
| sfinucan | It's an extension of the original CPU pinning spec, which you reviewed way back when. Not much has changed, but I do need another +2 from someone familar with the idea https://review.openstack.org/#/c/93652/ | 18:59 |
| *** tonytan4ever has quit IRC | 18:59 | |
| alaski | sfinucan: I can look at it, but I'd really like to see input from danpb since he authored the original | 18:59 |
| *** tonytan4ever has joined #openstack-nova | 19:01 | |
| *** tonytan_brb has quit IRC | 19:01 | |
| sfinucan | alaski: fortunately, he's already +2d it :) | 19:01 |
| alaski | hmm, I'm not seeing that | 19:03 |
| *** Piet has joined #openstack-nova | 19:04 | |
| *** mriedem has joined #openstack-nova | 19:05 | |
| *** kmartin has quit IRC | 19:05 | |
| *** kmartin has joined #openstack-nova | 19:06 | |
| sfinucan | alaski: Ah, my mistake. Wrong name | 19:07 |
| sfinucan | drat | 19:07 |
| sfinucan | I'll reach out to danpb | 19:07 |
| *** ctrath has quit IRC | 19:08 | |
| *** Piet has quit IRC | 19:09 | |
| *** ctrath has joined #openstack-nova | 19:11 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Last sync from oslo-incubator https://review.openstack.org/244818 | 19:11 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/python-novaclient: Last sync from oslo-incubator https://review.openstack.org/244827 | 19:12 |
| *** ArxCruz has joined #openstack-nova | 19:12 | |
| *** aix has quit IRC | 19:15 | |
| *** ArxCruz has quit IRC | 19:17 | |
| *** lykinsbd has quit IRC | 19:17 | |
| *** sfinucan has quit IRC | 19:18 | |
| *** sfinucan has joined #openstack-nova | 19:18 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Replaced deprecated timeutils methods https://review.openstack.org/241179 | 19:19 |
| *** lykinsbd has joined #openstack-nova | 19:19 | |
| *** achanda has quit IRC | 19:21 | |
| *** achanda has joined #openstack-nova | 19:21 | |
| *** su_zhang has quit IRC | 19:28 | |
| openstackgerrit | Hans Lindgren proposed openstack/nova: Use TimeFixture from oslo_utils to override time in tests https://review.openstack.org/237616 | 19:32 |
| *** korzen has joined #openstack-nova | 19:32 | |
| *** ccarmack has quit IRC | 19:33 | |
| *** bnemec has joined #openstack-nova | 19:33 | |
| *** pratikmallya has quit IRC | 19:35 | |
| *** otter768 has joined #openstack-nova | 19:37 | |
| *** lykinsbd has quit IRC | 19:38 | |
| *** ijw has joined #openstack-nova | 19:39 | |
| *** med_ is now known as openstackgerrit_ | 19:39 | |
| *** angdraug has joined #openstack-nova | 19:40 | |
| *** openstackgerrit_ is now known as med_ | 19:40 | |
| *** pm90_ has joined #openstack-nova | 19:40 | |
| *** carolynvs has joined #openstack-nova | 19:40 | |
| annegentle | hey all, I'm looking in the nova codebase for an example of an already-microversioned API call. | 19:41 |
| annegentle | so far I think I'm seeing just RPC API calls that are versioned | 19:41 |
| mriedem | annegentle: sec | 19:41 |
| annegentle | mriedem: cool thanks | 19:41 |
| mriedem | annegentle: https://review.openstack.org/#/c/168966/ | 19:41 |
| mriedem | that one was pretty simple | 19:42 |
| mriedem | https://review.openstack.org/#/c/168966/8/nova/api/openstack/compute/plugins/v3/fixed_ips.py | 19:42 |
| sdague | dansmith: in looking at ebtables rules that are on a gate rule not by libvirt, it's one PREROUTING rule for floating ips | 19:42 |
| *** alex_klimov has joined #openstack-nova | 19:42 | |
| sdague | http://logs.openstack.org/94/244194/6/check/gate-grenade-dsvm-multinode-nv/fd8b32f/logs/worlddump-2015-11-12-141039-cinder_resources_created.txt.gz - -p IPv4 --logical-in br100 --ip-src 10.1.0.2 --ip-dst ! 10.1.0.0/20 -j redirect | 19:42 |
| *** otter768 has quit IRC | 19:42 | |
| annegentle | mriedem: dere it is. Ok carolynvs so looking at https://review.openstack.org/#/c/168966/8/doc/api_samples/versions/versions-get-resp.json that's the JSON | 19:42 |
| sdague | which looks like we delete and reinsert every time a floating ip is added, which seems a little odd | 19:42 |
| annegentle | carolynvs: for the response. Hm... | 19:42 |
| carolynvs | Thanks annegentle: I'm still getting used to gerrit :) | 19:42 |
| annegentle | carolynvs: heh yes. | 19:43 |
| *** ijw has quit IRC | 19:43 | |
| annegentle | mriedem: so to request a particular version, where do you put the header? (Or maybe I misunderstand) | 19:43 |
| carolynvs | I'm guessing that I need to include "X-OpenStack-Nova-API-Version: 2.4" in the header of my request | 19:44 |
| dansmith | sdague: I'm not familiar enough with ebtables to know what that means | 19:44 |
| dansmith | sdague: in iptables there would be a redirect --to $something | 19:44 |
| mriedem | carolynvs: right | 19:44 |
| sdague | right, I'm not hugely familiar as well, however it seems odd that it seems to be called on every floating ip add | 19:45 |
| sdague | but it's always the same rule | 19:45 |
| carolynvs | so I should go off of the sample json version response when figuring out what microversion should be used for new changes discussed in the release notes | 19:45 |
| dansmith | sdague: never a different IP? | 19:45 |
| sdague | which means it's flapping ebtables for no particular reason | 19:45 |
| dansmith | sdague: yeah, that's weird indeed | 19:45 |
| cburgess | sdague Is it always the same floating IP? | 19:45 |
| dansmith | sdague: I think cburgess knew something about ebtables at one point | 19:45 |
| sdague | well, we have 2 floating ips there | 19:45 |
| dansmith | speak of the devil | 19:45 |
| sdague | and one rule | 19:45 |
| cburgess | I should have kept my mouth shut.... :P | 19:45 |
| sdague | --ip-dst ! | 19:45 |
| sdague | cburgess: no, you totally shouldn't, because the dark magics should be explained | 19:46 |
| annegentle | mriedem: then if I'm reading rest_api_version_history.rst right, do you get the added keypair type for 2.2 also? Or does your request for keypair types need to have X-OpenStack-Nova-API-Version: 2.2" in it? | 19:46 |
| cburgess | LOL | 19:46 |
| cburgess | sdague OK let me look at this log you posted... | 19:46 |
| sdague | cburgess: well, it's a dump | 19:46 |
| mriedem | annegentle: it's cumulative | 19:46 |
| mriedem | so 2.4 includes 2.2 and 2.3 | 19:46 |
| annegentle | mriedem: ok so 2.4 would give me the rest | 19:47 |
| annegentle | mriedem: ok thanks! | 19:47 |
| mriedem | yup | 19:47 |
| cburgess | sdague The output of ebtables -t nat -L that I'm looking at, where does this come from? Compute node, or nova-network node? | 19:47 |
| *** greghaynes_ has quit IRC | 19:47 | |
| sdague | cburgess: this is an all in one for the gate | 19:47 |
| cburgess | OK | 19:47 |
| cburgess | and 10.1.0.2 is the fixed IP of the instance? | 19:47 |
| sdague | one sec | 19:48 |
| *** mc_nair has quit IRC | 19:48 | |
| sdague | it is the fixed ip for one instance | 19:48 |
| sdague | there is a 10.1.0.3 instance as well | 19:48 |
| sdague | yeh, that's the super confusing thing | 19:49 |
| cburgess | OK so... the specific test we are running here is what exactly? Launch instance, add floating, remove floating add floating? And the question is why the etables stuff? | 19:49 |
| sdague | at this point in the test we've launched 2 instances, add floating ips, and pinged them | 19:49 |
| *** greghaynes_ has joined #openstack-nova | 19:49 | |
| sdague | mostly just trying to understand | 19:50 |
| sdague | is it odd that there aren't rules for both of them? | 19:50 |
| *** rmart04 has joined #openstack-nova | 19:51 | |
| cburgess | Seems that way. Let me look at what this rule is actually doing. | 19:51 |
| *** carolynvs has left #openstack-nova | 19:51 | |
| cburgess | sdague dansmith yeah this is weird... redirect in an ebtables rule triggers a MAC address re-write. | 19:52 |
| dansmith | oh | 19:52 |
| dansmith | is that to make floatings work at L2? | 19:52 |
| suro-patz | rlrossit: please check https://review.openstack.org/#/c/243825/, need some input from you | 19:52 |
| cburgess | If I recall it basically changes the destination MAC to the interface the packets was recieved on, which would be the bridge in this case I think. | 19:52 |
| cburgess | dansmith sdague Let me go see where we configure this and why... | 19:53 |
| tdurakov | sdague, ping | 19:54 |
| sdague | tdurakov: what's up? | 19:54 |
| *** stackdump has quit IRC | 19:54 | |
| tdurakov | sdague, hi, need your help with https://review.openstack.org/#/c/243013/ | 19:54 |
| cburgess | sdague This is on master or a stable branch we are seeing this? | 19:55 |
| tdurakov | if everything is ok, it would be great to merge this, so i could debug config switching | 19:55 |
| sdague | cburgess: this is liberty, for unrelated reasons the only place we're regularly dumping this state in the middle of a run is on grenade testing | 19:56 |
| *** electrocucaracha has joined #openstack-nova | 19:56 | |
| sdague | so I have a ton of these dumps around in the gate to look at what "normal" looks like | 19:56 |
| sdague | this run passes | 19:56 |
| sdague | tdurakov: +2 | 19:57 |
| cburgess | sdague OK so in the test we have 2 instances in the same network each with a floating IP but only one of these ebtables rules? | 19:57 |
| sdague | yes | 19:58 |
| cburgess | sdague So yeah it looks like we should have 2 rules so thats odd. | 19:58 |
| tdurakov | sdague, thank you! | 19:58 |
| sdague | which I only just noticed in staring at this | 19:58 |
| cburgess | So the original bug from vish is https://bugs.launchpad.net/nova/+bug/1096985 | 19:58 |
| openstack | Launchpad bug 1096985 in OpenStack Compute (nova) "Floating ips on separate interface + external gateway does not work" [Medium,Fix released] - Assigned to Vish Ishaya (vishvananda) | 19:58 |
| sdague | it actually doesn't look like n-net did anything for the second guest | 19:58 |
| *** whenry__ has joined #openstack-nova | 19:58 | |
| sdague | which is even weirder | 19:58 |
| cburgess | I have a general idea of what vish is doing here and why. But yeah... it really seems like we should have two of those. Basically you should have 1 rule for each fixed IP in this configuration. | 19:59 |
| sdague | yeh, the code looks like it does that | 19:59 |
| sdague | however, something else went wrong here | 20:00 |
| sdague | oh... wait | 20:00 |
| sdague | nevermind, I picked a terrible job to look at | 20:00 |
| cburgess | OK | 20:00 |
| sdague | this is actually the multinode job | 20:00 |
| sdague | the other guest is on the other node | 20:00 |
| sdague | gdi | 20:00 |
| cburgess | sdague Also it seems like we aren't getting the full rule output there. | 20:01 |
| *** doug-fish has quit IRC | 20:01 | |
| cburgess | As in the full rule we are adding is... | 20:01 |
| sdague | cburgess: no? | 20:01 |
| cburgess | --logical-in %s -p ipv4 --ip-src %s ! --ip-dst %s -j redirect --redirect-target ACCEPT | 20:01 |
| sdague | if you know how I should invoke ebtables to get something better, let me know, I'll update the dump tool | 20:01 |
| cburgess | The output seems to stop after -j redirect. | 20:01 |
| mriedem | garyk: you have a couple of small issues to cleanup in https://review.openstack.org/#/c/165750/ and then i'm +2 | 20:01 |
| cburgess | sdague: nevermind ACCEPT is the default target so I suspect the tool just isn't showing you the default. | 20:02 |
| sdague | cburgess: hmmm... I'm running: sudo ebtables -t nat -L | 20:02 |
| cburgess | sdague Yeah that should be fine. | 20:02 |
| cburgess | Thats what I use. | 20:02 |
| cburgess | sdague So yeah if the other VM is on a different node then that makes sense why we only have a single rule. | 20:03 |
| *** ihrachys has quit IRC | 20:03 | |
| sdague | so, actually in the all in one case I'm seeing no rules | 20:04 |
| *** artom has joined #openstack-nova | 20:04 | |
| *** sfinucan has quit IRC | 20:04 | |
| *** sfinucan_ has joined #openstack-nova | 20:04 | |
| *** ihrachys has joined #openstack-nova | 20:04 | |
| *** sfinucan_ has quit IRC | 20:04 | |
| *** yassine__ has joined #openstack-nova | 20:04 | |
| *** lykinsbd has joined #openstack-nova | 20:04 | |
| *** iurygregory has quit IRC | 20:04 | |
| *** sfinucan has joined #openstack-nova | 20:04 | |
| sdague | ok, anyway, this makes a bit more sense now | 20:05 |
| cburgess | sdague dansmith So yeah in general what that rule is doing is saying.. if we get a packet from a fixed IP with a floating IP and VM is talking to a destination outside the local network and the floating IP is on a different device rewrite the L2 frame so that its coming from the device with the floating IP. | 20:05 |
| *** markmc has joined #openstack-nova | 20:05 | |
| cburgess | Its kind of a weird overly specific thing. | 20:05 |
| cburgess | But whatever. | 20:05 |
| *** ericksonsantos has quit IRC | 20:05 | |
| *** doug-fish has joined #openstack-nova | 20:05 | |
| cburgess | sdague dansmith The idea is that in a multi-host setup you could have a dedicate VLAN/NIC for floating IPs thats separate from where all your tenant traffic is. | 20:06 |
| *** rmart04 has quit IRC | 20:06 | |
| *** korzen has quit IRC | 20:06 | |
| dansmith | cburgess: surely this is part of some thing to avoid people being confused about floating ips on the same L2 where the NAT rules would otherwise kinda break us | 20:06 |
| *** artom_ has quit IRC | 20:06 | |
| cburgess | dansmith Its actually to guard against asymetric routing. | 20:07 |
| dansmith | where you take one path out and another path in, right? | 20:07 |
| cburgess | Yes | 20:07 |
| dansmith | yeah | 20:07 |
| *** IanGovett has joined #openstack-nova | 20:07 | |
| cburgess | So basically.. internet -> floating ip -> VM -> tenant gateway | 20:07 |
| *** electrocucaracha has quit IRC | 20:07 | |
| cburgess | Thats asymetric | 20:07 |
| dansmith | yeah | 20:08 |
| cburgess | With this rule you get a return back out the interface that hosts the floating IP. | 20:08 |
| *** lpetrut has joined #openstack-nova | 20:08 | |
| sdague | ok, so I think the code makes sense now. I'm slightly surprised that when we are running all in one their appear to be no rules created | 20:08 |
| cburgess | sdague Its only created in a specifc topo | 20:09 |
| cburgess | sdague brb | 20:09 |
| sdague | oh, right | 20:09 |
| mriedem | dims: was this the thing zzzeek was fixing in oslo.db? http://logs.openstack.org/04/240804/3/gate/gate-osc-dsvm-functional/9a34c38/logs/screen-n-api.txt.gz?level=TRACE#_2015-11-11_22_52_34_533 | 20:09 |
| sdague | because we only run multihost on the actual multinode job | 20:09 |
| sdague | so, right, I think that means on all non multinode jobs nova never runs ebtables in the gate | 20:10 |
| *** mc_nair has joined #openstack-nova | 20:11 | |
| dansmith | sdague: it's nice that we test reality now :) | 20:11 |
| sdague | dansmith: does that seem like a reasonable hypothesis | 20:11 |
| dansmith | dansmith: yeah | 20:12 |
| sdague | dansmith: sure, but it also means that all the ebtables explodes that were blamed on nova | 20:12 |
| sdague | nova never runs the command | 20:12 |
| cburgess | sdague bac | 20:17 |
| cburgess | er.. back | 20:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'hw_cpu_threads_policy' to ImageMetaProps https://review.openstack.org/202647 | 20:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'cpu_policy' and 'cpu_threads_policy' fields https://review.openstack.org/202648 | 20:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'hw:cpu_threads_policy=avoid' filtering https://review.openstack.org/202649 | 20:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: trivial: Add some logs to 'numa_topology_filter' https://review.openstack.org/202650 | 20:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add 'hw:cpu_threads_policy=require' scheduling https://review.openstack.org/202651 | 20:17 |
| *** raildo is now known as raildo-afk | 20:17 | |
| cburgess | Well nova runs ebtables for a number of different reasons. | 20:18 |
| cburgess | sdague So I don't think we can say it "never" runs ebtables. | 20:18 |
| sdague | cburgess: in an all in one without multihost set | 20:19 |
| cburgess | and if you mean those failures from last year that jogo looked it I'm pretty sure it was ebtables in nova since i wrote the fix for it.. :P | 20:19 |
| cburgess | sdague Fair. Some jobs might not run it. | 20:19 |
| sdague | cburgess: no, I mean a current set of fails | 20:20 |
| *** nikhil has joined #openstack-nova | 20:20 | |
| sdague | Bug 1501366 - libvirtError: Error while building firewall: Some rules could not be created for interface | 20:20 |
| openstack | bug 1501366 in OpenStack Compute (nova) "libvirtError: Error while building firewall: Some rules could not be created for interface" [High,Confirmed] https://launchpad.net/bugs/1501366 | 20:20 |
| *** tonytan4ever has quit IRC | 20:22 | |
| *** kfarr has quit IRC | 20:24 | |
| tdurakov | sdague, could you plus 2 once again? https://review.openstack.org/#/c/243013/7 | 20:24 |
| openstackgerrit | jichenjc proposed openstack/nova: Change Invalid exception to a specified exception https://review.openstack.org/232506 | 20:24 |
| sdague | actually, we do a log.debug whenever we run ebtables because of oslo.process | 20:25 |
| cburgess | sdague I'm not convinced that my retry fix is related. Also people are conflating my fix and what libvirt 1.2.11 buys you. | 20:25 |
| sdague | cburgess: right, I'm not actually saying anything about your fix | 20:26 |
| sdague | I actually just want to figure out what's up with some of these fails | 20:26 |
| cburgess | sdague Ok actually.... I'm wrong.. the log posted in https://bugs.launchpad.net/nova/+bug/1501366 is in fact the exact error my fix addresses. | 20:27 |
| openstack | Launchpad bug 1501366 in OpenStack Compute (nova) "libvirtError: Error while building firewall: Some rules could not be created for interface" [High,Confirmed] | 20:27 |
| cburgess | If my fix wasn't back ported to juno then that would be the cause. | 20:27 |
| bauzas | mriedem: dansmith: I can still see some BDM changes | 20:27 |
| cburgess | sdague Wait sorry I meant https://bugs.launchpad.net/nova/+bug/1501558 | 20:28 |
| openstack | Launchpad bug 1501558 in OpenStack Compute (nova) "nova-net: libvirtError: Error while building firewall: Some rules could not be created for interface: Unable to update the kernel" [High,Confirmed] | 20:28 |
| bauzas | mriedem: dansmith: since I was on PTO yesterday, is there still a discussion for that - given also the convo we had during today's meeting? | 20:28 |
| sdague | cburgess: right, that's another one | 20:28 |
| mriedem | bauzas: i was still debugging | 20:28 |
| bauzas | mriedem: tbh, I'm reviewing the changes for ramping up on BDMs | 20:29 |
| cburgess | sdague Yeah sorry to many similar bugs here. So yeah my fix is needed for 1501558 in Juno (mriedem you might care about that). But 1501366 look to be something else. | 20:29 |
| *** electrocucaracha has joined #openstack-nova | 20:29 | |
| mriedem | we're killing juno next week | 20:30 |
| dansmith | in a field | 20:30 |
| dansmith | to gangsta rap | 20:30 |
| mriedem | die mofo die mofo die | 20:30 |
| sdague | by looking at the n-net logs it looks like we don't tend to run any ebtables commands during a grenade run | 20:30 |
| dansmith | bauzas: there wasn't much that wasn't on the review | 20:31 |
| sdague | and only a few tempest tests trigger it happening during a longer run | 20:31 |
| dansmith | bauzas: apparently what I did was not what the others wanted, but I'm so distracted I haven't tried to figure out why | 20:31 |
| bauzas | meh | 20:31 |
| mriedem | now i'm listening to https://www.youtube.com/watch?v=qNa6xImZeo4 | 20:32 |
| mriedem | so i'm distracted | 20:32 |
| sdague | mriedem: heh | 20:32 |
| *** ijw has joined #openstack-nova | 20:33 | |
| *** kmartin has quit IRC | 20:35 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Move thread policy features between specs https://review.openstack.org/244857 | 20:36 |
| openstackgerrit | Anne Gentle proposed openstack/nova: docs: update the API faults concept guide https://review.openstack.org/243066 | 20:36 |
| *** tonytan4ever has joined #openstack-nova | 20:36 | |
| *** ijw_ has joined #openstack-nova | 20:36 | |
| *** ijw has quit IRC | 20:37 | |
| *** ctrath has quit IRC | 20:40 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 20:41 | |
| mriedem | rlrossit: damn your keen eye https://review.openstack.org/#/c/243852/ | 20:41 |
| mriedem | i don't want to write unit tests for that | 20:41 |
| *** signed8bit_ZZZzz is now known as signed8bit | 20:43 | |
| *** ociuhandu has joined #openstack-nova | 20:44 | |
| *** ctrath has joined #openstack-nova | 20:44 | |
| *** pixelbeat has quit IRC | 20:49 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: test_fields: Remove all 'Enum' subclass tests https://review.openstack.org/244372 | 20:52 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: tests_fields: Remove most enum-type field tests https://review.openstack.org/244370 | 20:53 |
| *** rlrossit has quit IRC | 20:53 | |
| *** eme has quit IRC | 20:53 | |
| *** tdurakov has quit IRC | 20:56 | |
| *** rlrossit has joined #openstack-nova | 20:58 | |
| *** gyee has joined #openstack-nova | 20:58 | |
| *** lpetrut has quit IRC | 20:58 | |
| mriedem | ugh, who wants to pad stats? https://review.openstack.org/#/c/232827/ | 20:58 |
| *** su_zhang has joined #openstack-nova | 20:59 | |
| *** lpetrut has joined #openstack-nova | 20:59 | |
| cburgess | mriedem sdague I posted a comment to https://bugs.launchpad.net/nova/+bug/1501558 to try and clarify the retry loop logic vs --concurrent and libvrit 1.2.11. | 21:01 |
| openstack | Launchpad bug 1501558 in OpenStack Compute (nova) "nova-net: libvirtError: Error while building firewall: Some rules could not be created for interface: Unable to update the kernel" [High,Confirmed] | 21:01 |
| *** yassine__ has quit IRC | 21:02 | |
| *** su_zhang has quit IRC | 21:03 | |
| sdague | mriedem: some of those changes annoy me a little, like filesystem => file system | 21:04 |
| sdague | as the no space version is fine | 21:04 |
| mriedem | sdague: i know, | 21:05 |
| *** djchou has quit IRC | 21:05 | |
| mriedem | sdague: i was about to say in my +2 comment that i felt bad acting like i was supporting this type of change | 21:05 |
| *** kmartin has joined #openstack-nova | 21:07 | |
| sdague | well, I commented | 21:07 |
| *** jichen has quit IRC | 21:07 | |
| *** pm90__ has joined #openstack-nova | 21:08 | |
| *** pm90__ has quit IRC | 21:08 | |
| *** kmartin has quit IRC | 21:08 | |
| *** kmartin has joined #openstack-nova | 21:08 | |
| *** suro-patz has quit IRC | 21:10 | |
| mriedem | sdague: also https://review.openstack.org/#/c/218730/ | 21:11 |
| mriedem | i noted the commit that removed the usage in my comment | 21:11 |
| mriedem | which should have been in the commit message | 21:11 |
| *** pm90_ has quit IRC | 21:11 | |
| *** Marga_ has joined #openstack-nova | 21:12 | |
| sdague | that one I can get behind | 21:12 |
| *** smatzek has quit IRC | 21:12 | |
| mriedem | also easy https://review.openstack.org/#/c/241976/ | 21:16 |
| *** tdurakov has joined #openstack-nova | 21:17 | |
| dansmith | got it | 21:18 |
| mriedem | BobBall: so xenproject ci was passing force_config_drive=always here? https://review.openstack.org/#/c/237762/ | 21:20 |
| mriedem | or where is the failure? | 21:20 |
| cburgess | sdague mriedem So for https://bugs.launchpad.net/nova/+bug/1501366 it is the ebtables race that we originally fixed in https://bugs.launchpad.net/nova/+bug/1316621 but from the other side (aka libvirt is hitting the race, not nova). I posted a comment on how to fix it. Short version libvirt 1.2.11 isn't good enough. | 21:21 |
| openstack | Launchpad bug 1501366 in OpenStack Compute (nova) "libvirtError: Error while building firewall: Some rules could not be created for interface" [High,Confirmed] | 21:21 |
| openstack | Launchpad bug 1316621 in OpenStack Compute (nova) "ebtables calls can race with libvirt" [Medium,Fix released] - Assigned to Chet Burgess (cfb-n) | 21:21 |
| sdague | cburgess: thanks much for the input there! | 21:22 |
| *** Piet has joined #openstack-nova | 21:22 | |
| mriedem | BobBall: nvm, i see it here http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/62/237762/9/15586/logs/nova.conf | 21:22 |
| mriedem | BobBall: ok, but that option was deprecated for a release, and the change wasn't xen specific, so we missed it, sorry | 21:22 |
| bauzas | mriedem: AFAIK, BobBall is on UK TZ so I guess he would reply by tomorrow :) | 21:23 |
| *** achanda has quit IRC | 21:23 | |
| mriedem | that's fine | 21:25 |
| mriedem | if yo'ure on irc, you're fair game for pings | 21:25 |
| mriedem | also commented in the review | 21:25 |
| *** stevemar_ has quit IRC | 21:25 | |
| *** achanda has joined #openstack-nova | 21:26 | |
| *** achanda has quit IRC | 21:26 | |
| *** su_zhang has joined #openstack-nova | 21:26 | |
| mriedem | sfinucan: weren't you supposed to add danpb to this? https://review.openstack.org/#/c/244370/ | 21:26 |
| *** cn28h has joined #openstack-nova | 21:26 | |
| *** achanda has joined #openstack-nova | 21:27 | |
| *** diogogmt has quit IRC | 21:27 | |
| sfinucan | I added him to the bug | 21:27 |
| mriedem | i doubt anyone follows bug email | 21:27 |
| mriedem | i added him to the review | 21:27 |
| sfinucan | Ah, OK. I didn't know that | 21:27 |
| sfinucan | There's two approaches. I'll stick him on the other | 21:27 |
| *** mdrabe has quit IRC | 21:29 | |
| *** otter768 has joined #openstack-nova | 21:38 | |
| *** pm90_ has joined #openstack-nova | 21:39 | |
| *** shardy has quit IRC | 21:39 | |
| *** claudiub has quit IRC | 21:40 | |
| *** pm90__ has joined #openstack-nova | 21:42 | |
| *** otter768 has quit IRC | 21:42 | |
| *** pm90_ has quit IRC | 21:44 | |
| *** thangp has quit IRC | 21:44 | |
| alaski | dansmith: no rush, but when you have some time your thoughts on the comment on line 321 at https://review.openstack.org/#/c/211753/27 would be appreciated | 21:48 |
| alaski | basically, what to do about serialized objects in the db and bumping object versions | 21:48 |
| dansmith | alaski: okay | 21:49 |
| sdague | cburgess: so I'm going to say a thing, and you are going to hate me, and I'll regret it. But I'll say it anyway. | 21:50 |
| *** jhesketh has quit IRC | 21:50 | |
| cburgess | sdague OK.... | 21:51 |
| mriedem | sdague thinks justin bieber is over rated | 21:51 |
| mriedem | there, he said it | 21:51 |
| sdague | it almost feels like we should have a thing replace ebtables with a version that is a shell script wrapper to the real one that includes a flock | 21:51 |
| *** ZZelle has quit IRC | 21:51 | |
| cburgess | mriedem No argument here | 21:51 |
| sdague | then we stop having to play this giant dance of who's got what version that supports what thing | 21:52 |
| * sdague waits for the hate | 21:52 | |
| cburgess | sdague OK so... I don't hate you, but I hate that. I don't think its reasonable for openstack to say "here replace your ebtables command and introduce all these potential issue around upgradability" | 21:52 |
| sdague | yeh, fair | 21:52 |
| sdague | we could document it as a way around the problem for people that can't upgrade the world | 21:53 |
| cburgess | sdague So if we want to provide a script some place in tree and say "if you can't run version X of this and Y of this you could also use this script" I think thats great. | 21:53 |
| *** ZZelle has joined #openstack-nova | 21:53 | |
| cburgess | But forcing/requiring that script... I don't see it. | 21:53 |
| *** jhesketh has joined #openstack-nova | 21:53 | |
| sdague | yeh, that seems like a plan | 21:54 |
| *** achanda has quit IRC | 21:54 | |
| cburgess | sdague We might even be able to be slick and look at how its called in libvirt and might be able to use just path tricks to use the script. | 21:54 |
| *** achanda has joined #openstack-nova | 21:54 | |
| *** e0ne has joined #openstack-nova | 21:54 | |
| sdague | hmmmm | 21:54 |
| sdague | now that's a thought | 21:54 |
| *** aysyd has quit IRC | 21:54 | |
| cburgess | sdague Still wouldn't be automatic. Would still be a documented thing for deployer to go change libvirt path to find script is they don't have the right versions. | 21:55 |
| cburgess | But thats way more palatable then replacing a system binary. | 21:55 |
| cburgess | sdague So are you saying we want to fix nova to do the right thing with the right versions and support a fallback to retry+script with docs around using the work around? | 21:56 |
| sdague | yeh | 21:56 |
| cburgess | I can get behind that. | 21:56 |
| cburgess | As a goal for M? Or back port as well? | 21:56 |
| sdague | well, master first | 21:56 |
| sdague | decisions can be made later after that seems to work | 21:57 |
| cburgess | sdague bug or spec? | 21:57 |
| sdague | the fallback script will get reasonably tested because we can't get to a new enough libvirt this cycle, so we'd use that in the gate to stop randomly exploding | 21:57 |
| *** spandhe has quit IRC | 21:58 | |
| sdague | cburgess: this is fine to be the solution to the bug at hand I think | 21:58 |
| sdague | dansmith / mriedem opinions? | 21:58 |
| dansmith | sdague: I think you're going to hell for this | 21:58 |
| *** Marga_ has quit IRC | 21:58 | |
| mriedem | i'm leaving for parent / teacher conferences | 21:58 |
| dansmith | now, let me read the scrollback | 21:58 |
| cburgess | LOL | 21:58 |
| mriedem | i'll ask them what they think | 21:58 |
| sdague | dansmith: I'm not convinced your answer will be different once you read it | 21:58 |
| dansmith | so, you can't use path tricks from inside nova to cause libvirt to use the script, | 21:59 |
| cburgess | dansmith He is going to hell, but it is the least ugly choice. | 21:59 |
| dansmith | you'll need it to be set for the daemon | 21:59 |
| cburgess | dansmith Right it would be documented. | 21:59 |
| dansmith | cburgess: I'm willing to sacrifice sdague's eternal soul, how about you? | 21:59 |
| cburgess | There would have to be a doc that says "If you can't run with these versions you should do the following...." | 21:59 |
| sdague | also, it looks like the commands out of the daemon are not using path | 21:59 |
| cburgess | dansmith I'm agree with sdague so I think my soul is on the line too. | 21:59 |
| *** mriedem has quit IRC | 22:00 | |
| sdague | they are /sbin/ebtables directly | 22:00 |
| cburgess | sdague As in its just 'ebtable' or its calling with full path? | 22:00 |
| dansmith | cburgess: if your soul isn't on the review, you're safe :) | 22:00 |
| cburgess | Damn. | 22:00 |
| cburgess | OK. | 22:00 |
| sdague | http://logs.openstack.org/86/230186/20/gate/gate-tempest-dsvm-postgres-full/f791773//logs/libvirt/libvirtd.txt.gz#_2015-11-12_16_00_43_480 | 22:00 |
| cburgess | sdague Well.. thats not a guarantee.. I would check the code first to be sure. | 22:00 |
| sdague | cburgess: sure, but I'm pretty sure this is generating that shell script as such | 22:01 |
| dansmith | they could be generating the shell with the path they've resolved, | 22:01 |
| dansmith | but it wouldn't make a lot of sense | 22:01 |
| cburgess | Right exactly | 22:01 |
| cburgess | Oh find... let me load the code and check... | 22:01 |
| cburgess | Give me a minute. | 22:01 |
| cburgess | dansmith sdague Actually path trick or not doesn't matter. Thats an implementation details of the proposed solution right? | 22:01 |
| sdague | yeh | 22:02 |
| dansmith | well, | 22:02 |
| cburgess | sdague Also instead of flock shouldn't we use oslo locking for this? | 22:02 |
| sdague | it looks like it's resolved at compile time | 22:02 |
| dansmith | it means that people will have to hack their environment | 22:02 |
| sdague | cburgess: no, because this would be a stupid shell script | 22:02 |
| cburgess | dansmith Yes | 22:02 |
| cburgess | sdague Ermm... | 22:02 |
| dansmith | cburgess: I expect there's zero chance we'll be okay with that | 22:02 |
| dansmith | since a yum update would break you | 22:02 |
| cburgess | dansmith I agree | 22:02 |
| cburgess | But... there is no solution to this problem without the proper versions. | 22:03 |
| dansmith | well, | 22:03 |
| sdague | dansmith: right, that's why it would be a script in tree with instructions for manual install | 22:03 |
| dansmith | there is of course | 22:03 |
| sdague | not something we'd do | 22:03 |
| dansmith | we could either: | 22:03 |
| cburgess | So... we propose fixing nova to do the right thing with the proper versions and also providing a work around. | 22:03 |
| dansmith | 1. Ask libvirt to take it as a runtime argument in config | 22:03 |
| sdague | and you could build a package with triggers that does the replace | 22:03 |
| dansmith | 2. Ask libvirt to flock the ebtables binary it's using, and we do the same | 22:03 |
| dansmith | well, maybe we can't flock the binary | 22:04 |
| sdague | dansmith: new libvirt already does that | 22:04 |
| dansmith | but some location | 22:04 |
| dansmith | oh | 22:04 |
| sdague | the --concurrency flag | 22:04 |
| dansmith | sorry I missed that | 22:04 |
| sdague | but you need 1.2.11 | 22:04 |
| cburgess | dansmith OK but given libvirt has already fixed the problem by using --concurrent and it would still require everyone to get a new version of libvirt.. why not just use --concurrent? | 22:04 |
| dansmith | so this only matters for older libvirt? | 22:04 |
| dansmith | sorry I glazed over that | 22:04 |
| sdague | and an ebtables that's new enough to support it | 22:04 |
| cburgess | dansmith Yeah | 22:04 |
| *** Marga_ has joined #openstack-nova | 22:04 | |
| sdague | yeh, this is a "well, you can't upgrade, you can do this thing instead even though it's hacky" | 22:04 |
| dansmith | well, then if it's a hack for older systems, then sure, a sample script and a box of chocolates | 22:05 |
| dansmith | yep | 22:05 |
| *** jamielennox|away is now known as jamielennox | 22:05 | |
| *** vilobhmm has quit IRC | 22:05 | |
| *** exploreshaifali has quit IRC | 22:05 | |
| cburgess | Where old = gate | 22:05 |
| *** vilobhmm has joined #openstack-nova | 22:05 | |
| sdague | cburgess: also, ebtables path is set in configure.ac it looks like | 22:05 |
| sdague | so compile time | 22:05 |
| sdague | but variable per system | 22:05 |
| *** vilobhmm has quit IRC | 22:06 | |
| cburgess | OK so they have to replace the binary then. Oh well. | 22:06 |
| cburgess | sdague dansmith OK so we agree, its ugly, and hacky, and not great, but best we can do? | 22:08 |
| *** MVenesio has quit IRC | 22:08 | |
| dansmith | cburgess: actually I think we agree that people that haven't upgraded will take whatever god-awful solution we poop out, which I'm okay with :) | 22:08 |
| cburgess | dansmith OK | 22:09 |
| cburgess | sdague How fast did you want this? | 22:09 |
| sdague | cburgess: if you are signing up for a thing, and I don't have to write this, as soon as you can :) | 22:09 |
| cburgess | sdague Well the how long determines if I can do it... but... I did loose ATC and need to land something again... :P | 22:10 |
| sdague | fwiw, --concurrent is supported in ebtables as old as trusty, so I don't actually understand why we don't use it | 22:10 |
| cburgess | Are you sure? | 22:10 |
| cburgess | I thought I looked and it wasn't in trusty. | 22:10 |
| sdague | os1:~> ebtables --version | 22:10 |
| sdague | ebtables v2.0.10-4 (December 2011) | 22:10 |
| sdague | os1:~> ebtables --help | grep concur | 22:11 |
| sdague | --concurrent : use a file lock to support concurrent scripts | 22:11 |
| cburgess | root@mcp1.trial6.mc:/home/cfb$ dpkg -l | grep ebtables | 22:11 |
| cburgess | ii ebtables 2.0.9.2-2ubuntu2 Ethernet bridge frame table administration | 22:11 |
| cburgess | root@mcp1.trial6.mc:/home/cfb$ ebtables --concurrent | 22:11 |
| cburgess | Unknown argument: '--concurrent'. | 22:11 |
| cburgess | I don't know where you got that ebtables from. | 22:11 |
| cburgess | I don't think thats the default trust one. | 22:11 |
| *** Marga_ has quit IRC | 22:11 | |
| cburgess | trusty | 22:11 |
| sdague | ii ebtables 2.0.10.4-3ubuntu1 amd64 Ethernet bridge frame table administration | 22:11 |
| cburgess | WTF | 22:11 |
| sdague | you didn't take updates :) | 22:11 |
| cburgess | Can you show me... 'apt-cache showpkg ebtables'? | 22:11 |
| dansmith | alaski: comments commentified | 22:12 |
| sdague | Package: ebtables | 22:12 |
| sdague | Versions: | 22:12 |
| sdague | 2.0.10.4-3ubuntu1 (/var/lib/apt/lists/gallifrey_ubuntu_dists_trusty_main_binary-amd64_Packages) (/var/lib/dpkg/status) | 22:12 |
| sdague | I have a local mirror | 22:12 |
| sdague | but it's just trusty | 22:12 |
| dansmith | trusty? | 22:12 |
| dansmith | more like rusty! | 22:12 |
| dansmith | hey-o | 22:12 |
| cburgess | sdague I don't get it... I have the same... wtf... | 22:13 |
| * dansmith channels mriedem while he's at PTA or whatever | 22:13 | |
| cburgess | sdague http://packages.ubuntu.com/precise/net/ebtables | 22:13 |
| cburgess | So... I'm really not surer where you are getting that from. | 22:14 |
| sdague | http://packages.ubuntu.com/trusty/ebtables | 22:14 |
| dansmith | that's precise homeboy | 22:14 |
| sdague | yeh, that's precise :) | 22:14 |
| cburgess | dansmith LOL sorry yes... | 22:14 |
| cburgess | sdague dansmith When I wrote the original retry loop gate was still precise, thats why I didn't do it. I forgot gate was upgraded since then. | 22:14 |
| sdague | yeh | 22:14 |
| cburgess | So yeah.. we can fix it now that we have trusty. | 22:14 |
| cburgess | but.. | 22:15 |
| alaski | dansmith: thanks | 22:15 |
| cburgess | We also need libvirt 1.2.11. We can't just call --concurrent and assume it will work. We need both sides to use --concurrent or we have to retry. | 22:15 |
| cburgess | We should probably just have it use --concurrent and do retry to make it easier. | 22:15 |
| sdague | cburgess: sure, though there is no harm if we call --concurrent | 22:15 |
| sdague | yeh, that's what I was thinking | 22:15 |
| *** david-lyle has joined #openstack-nova | 22:16 | |
| cburgess | sdague OK I'll work on it. Do we want to try and detect versions or just require the newer version and explode if we don't have --concurrent? | 22:16 |
| sdague | then if you have 1.2.11, you're fine | 22:16 |
| sdague | for mitaka, I think we can probably just assume you need ebtables > has this | 22:17 |
| sdague | document in the release notes when we add that flag that it's a requirement | 22:17 |
| sdague | I assume 2.0.10 was the version that mattered? | 22:18 |
| *** eharney has quit IRC | 22:18 | |
| cburgess | Yeah I think so. | 22:18 |
| cburgess | sdague Wait add flag? You want a flag? | 22:18 |
| *** sacharya has quit IRC | 22:19 | |
| sdague | cburgess: by flag I meant --concurrent always being passed to ebtables | 22:19 |
| sdague | no, this is not configurable | 22:19 |
| *** boris-42 has joined #openstack-nova | 22:19 | |
| cburgess | sdague Got it sorry ok... | 22:19 |
| sdague | cburgess: then you can be an atc :) | 22:21 |
| sdague | honestly, that patch is probably 30 minutes | 22:21 |
| cburgess | sdague OK assigned https://bugs.launchpad.net/nova/+bug/1501366 to myself | 22:21 |
| openstack | Launchpad bug 1501366 in OpenStack Compute (nova) "libvirtError: Error while building firewall: Some rules could not be created for interface" [High,Confirmed] - Assigned to Chet Burgess (cfb-n) | 22:21 |
| *** thorst has quit IRC | 22:22 | |
| sdague | cburgess: thanks! | 22:22 |
| cburgess | sdague Yeah its the script thats a bit more involved... | 22:22 |
| sdague | yeh, the script will be | 22:22 |
| openstackgerrit | Sumant Murke proposed openstack/nova-specs: Add new attribute to nova show API response https://review.openstack.org/244888 | 22:22 |
| cburgess | sdague but first step is get nova doing the right thing. | 22:22 |
| *** thorst has joined #openstack-nova | 22:22 | |
| *** ArxCruz has joined #openstack-nova | 22:22 | |
| *** dims_ has joined #openstack-nova | 22:23 | |
| *** tdurakov has quit IRC | 22:24 | |
| sdague | cburgess: honestly, I think you can just do this - https://github.com/openstack-infra/puppet-elastic_recheck/blob/master/files/er_safe_run.sh | 22:25 |
| *** dims has quit IRC | 22:25 | |
| sdague | flock /well/know/lock/location /sbin/ebtables.real $@ | 22:25 |
| cburgess | sdague So I think the issue is going to be that the "well known location" is different between different OSes | 22:26 |
| cburgess | On ubunut it should be /var/run/lock but I think on RH its different. | 22:26 |
| cburgess | So we might need a quick lsb_release check and change path, but I get where you are doing. | 22:26 |
| *** thorst has quit IRC | 22:27 | |
| sdague | cburgess: /var/lock is FHS | 22:27 |
| sdague | ubuntu links it to /var/run/lock | 22:27 |
| sdague | but /var/lock is probably the right answer | 22:27 |
| sdague | http://www.pathname.com/fhs/pub/fhs-2.3.html#VARLOCKLOCKFILES | 22:27 |
| cburgess | sdague You assume everyone uses FHS but ok... we can run with that. | 22:31 |
| sdague | yes, I assume that if you are running linux that doesn't comply with FHS then I hate you | 22:31 |
| sdague | because that's one of the few things the distros all generally agreed on | 22:32 |
| cburgess | sdague I'm going to do it as 2 patches though. One to fix it the right way, then one to add the work around. | 22:32 |
| *** Marga_ has joined #openstack-nova | 22:32 | |
| sdague | cburgess: that is perfectly kosher | 22:32 |
| *** lnxnut has quit IRC | 22:32 | |
| dansmith | and appreciated | 22:33 |
| *** annegentle has quit IRC | 22:33 | |
| cburgess | nod | 22:33 |
| sdague | cburgess: yes, definitely thanks in advance for jumping in here | 22:36 |
| *** emagana_ has quit IRC | 22:36 | |
| *** alex_klimov has quit IRC | 22:37 | |
| *** emagana has joined #openstack-nova | 22:37 | |
| *** lpetrut has quit IRC | 22:39 | |
| *** tongli has quit IRC | 22:41 | |
| *** pm90_ has joined #openstack-nova | 22:44 | |
| openstackgerrit | Sumant Murke proposed openstack/nova-specs: Add new attribute to nova show API response https://review.openstack.org/244888 | 22:44 |
| *** kaisers has quit IRC | 22:45 | |
| *** burgerk has quit IRC | 22:45 | |
| *** thorst has joined #openstack-nova | 22:46 | |
| *** ctrath has quit IRC | 22:46 | |
| *** pm90__ has quit IRC | 22:47 | |
| *** thorst_ has joined #openstack-nova | 22:48 | |
| *** rlrossit has quit IRC | 22:49 | |
| *** annegentle has joined #openstack-nova | 22:49 | |
| *** ev1lkry has quit IRC | 22:50 | |
| *** annegentle has quit IRC | 22:50 | |
| *** thorst has quit IRC | 22:51 | |
| *** thorst_ has quit IRC | 22:53 | |
| *** tonytan4ever has quit IRC | 22:54 | |
| *** ctrath1 has joined #openstack-nova | 22:55 | |
| *** edtubill has quit IRC | 22:55 | |
| *** kaisers has joined #openstack-nova | 22:56 | |
| *** stevemar_ has joined #openstack-nova | 22:56 | |
| *** pumaranikar has joined #openstack-nova | 22:57 | |
| *** IanGovett has quit IRC | 22:57 | |
| *** stevemar_ has quit IRC | 22:59 | |
| pumaranikar | Hi I am working on bug https://bugs.launchpad.net/nova/+bug/1250066. It says I have to update "updated_at" field when attach/detach volume to an instance | 23:03 |
| openstack | Launchpad bug 1250066 in OpenStack Compute (nova) "Need change instance's property updated_at when attach/detach volume" [Low,Confirmed] - Assigned to Pushkar Umaranikar (pushkar-umaranikar) | 23:03 |
| pumaranikar | But instance.save methond won't work here as updates set will be empty | 23:03 |
| pumaranikar | Any suggestions about How Can I approach to reslove ? | 23:04 |
| *** ZZelle has quit IRC | 23:08 | |
| *** rk4n has joined #openstack-nova | 23:12 | |
| openstackgerrit | Sumant Murke proposed openstack/nova-specs: Add new attribute to nova show API response https://review.openstack.org/244888 | 23:12 |
| *** jamielennox is now known as jamielennox|away | 23:13 | |
| *** obondarev has quit IRC | 23:14 | |
| *** rk4n has quit IRC | 23:15 | |
| *** obondarev has joined #openstack-nova | 23:15 | |
| *** rk4n has joined #openstack-nova | 23:16 | |
| *** rk4n has quit IRC | 23:16 | |
| *** pixelbeat has joined #openstack-nova | 23:17 | |
| *** aix has joined #openstack-nova | 23:18 | |
| *** darrenc is now known as darrenc_afk | 23:18 | |
| *** gyee has quit IRC | 23:19 | |
| *** mc_nair has quit IRC | 23:20 | |
| *** gyee has joined #openstack-nova | 23:21 | |
| *** ihrachys has quit IRC | 23:21 | |
| *** jamielennox|away is now known as jamielennox | 23:22 | |
| *** ctrath1 has quit IRC | 23:23 | |
| *** kmartin has quit IRC | 23:23 | |
| *** sneti has quit IRC | 23:24 | |
| *** rk4n has joined #openstack-nova | 23:24 | |
| *** rk4n has quit IRC | 23:29 | |
| *** ihrachys has joined #openstack-nova | 23:30 | |
| *** ijuwang has quit IRC | 23:31 | |
| *** pradk has quit IRC | 23:33 | |
| *** pradk has joined #openstack-nova | 23:35 | |
| *** nikhil_k_ has joined #openstack-nova | 23:37 | |
| *** rk4n has joined #openstack-nova | 23:38 | |
| *** otter768 has joined #openstack-nova | 23:39 | |
| *** darrenc_afk is now known as darrenc | 23:39 | |
| *** nikhil_k has quit IRC | 23:40 | |
| *** otter768 has quit IRC | 23:43 | |
| *** rk4n has quit IRC | 23:47 | |
| *** ArxCruz has quit IRC | 23:47 | |
| *** ArxCruz has joined #openstack-nova | 23:48 | |
| *** takashin has joined #openstack-nova | 23:49 | |
| *** ArxCruz has quit IRC | 23:49 | |
| *** emagana has quit IRC | 23:49 | |
| *** hparekh has quit IRC | 23:50 | |
| *** pumaranikar has quit IRC | 23:50 | |
| *** rk4n has joined #openstack-nova | 23:50 | |
| *** hwoarang has quit IRC | 23:51 | |
| *** markmc has quit IRC | 23:52 | |
| *** hwoarang has joined #openstack-nova | 23:52 | |
| *** jgjhgsonchn88523 has quit IRC | 23:54 | |
| *** achanda has quit IRC | 23:55 | |
| *** markmc has joined #openstack-nova | 23:55 | |
| *** e0ne has quit IRC | 23:56 | |
| *** ihrachys has quit IRC | 23:57 | |
| *** hparekh has joined #openstack-nova | 23:57 | |
| *** pm90_ has quit IRC | 23:57 | |
| *** xyang1 has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!