Tuesday, 2015-12-08

*** jerrygb has quit IRC00:00
*** jerrygb has joined #openstack-nova00:02
*** tonytan4ever has joined #openstack-nova00:02
*** salv-orl_ has quit IRC00:03
*** jamielennox|away is now known as jamielennox00:05
*** takedakn has quit IRC00:07
*** takedakn has joined #openstack-nova00:07
*** s-moriya has joined #openstack-nova00:08
*** salv-orlando has joined #openstack-nova00:09
*** aginwala has quit IRC00:11
zhenq1when running command nova --os-compute-api-version 2.0 list, it still shows calling v2.1 api, can anyone tell me why this happened?00:13
*** mirrorbox has quit IRC00:16
*** Jeffrey4l has quit IRC00:17
*** aginwala has joined #openstack-nova00:17
*** mirrorbox has joined #openstack-nova00:17
*** jerrygb has quit IRC00:18
*** jerrygb has joined #openstack-nova00:18
*** ianbrown has quit IRC00:20
*** ianbrown has joined #openstack-nova00:20
*** smatzek has quit IRC00:20
*** ljxiash has joined #openstack-nova00:23
*** jerrygb has quit IRC00:23
*** Jeffrey4l has joined #openstack-nova00:23
*** tonytan4ever has quit IRC00:23
*** ljxiash_ has joined #openstack-nova00:24
*** ianbrown has quit IRC00:25
*** ianbrown_ has joined #openstack-nova00:25
*** mirrorbox has quit IRC00:27
*** ljxiash has quit IRC00:27
*** ljxiash_ has quit IRC00:28
*** zenoway has quit IRC00:29
*** betherly has quit IRC00:32
*** patrickeast has quit IRC00:32
*** mtanino has quit IRC00:33
*** dougwig has quit IRC00:35
*** chenli has joined #openstack-nova00:35
*** dougwig has joined #openstack-nova00:35
alex_xutoday is Nova API doc sprint!00:38
alex_xugmann: oomichi tangchen ^00:38
*** aginwala has quit IRC00:39
*** mc_nair_ has joined #openstack-nova00:40
*** mc_nair has quit IRC00:40
*** ducttape_ has quit IRC00:40
*** ducttape_ has joined #openstack-nova00:40
*** otter768 has joined #openstack-nova00:41
*** yonglihe has quit IRC00:41
*** shaohe_feng has joined #openstack-nova00:42
*** takedakn has quit IRC00:43
*** ijw has joined #openstack-nova00:44
*** alejandrito has quit IRC00:46
gmannalex_xu: yea, Thanks00:47
gmannalex_xu: so we start working on API etherpad right00:48
alex_xugmann: yea00:48
gmannalex_xu: cool, i will start :)00:48
alex_xugmann: cool, thanks :)00:48
*** ijw has quit IRC00:51
*** ijw has joined #openstack-nova00:53
*** aginwala has joined #openstack-nova00:53
*** mirrorbox has joined #openstack-nova00:54
*** aginwala has quit IRC00:55
*** EinstCrazy has joined #openstack-nova00:56
*** RichardRaseley has quit IRC00:56
*** Sukhdev has joined #openstack-nova00:56
*** jinxing has joined #openstack-nova00:56
*** Sukhdev has quit IRC00:57
*** ijw has quit IRC00:57
*** thorst has joined #openstack-nova00:57
*** otter768 has quit IRC00:59
*** takashin has joined #openstack-nova00:59
*** changbl has quit IRC01:00
*** wyang has joined #openstack-nova01:02
*** hemna is now known as hemnafk01:04
*** thorst has quit IRC01:05
*** andrewbogott has joined #openstack-nova01:05
*** ljxiash has joined #openstack-nova01:08
*** andrewbogott has quit IRC01:08
*** andrewbogott has joined #openstack-nova01:08
oomichialex_xu: yeah, nice day to do that!01:09
alex_xuoomichi: yeah!01:09
oomichialex_xu: gmann: how about https://review.openstack.org/#/c/253417/ ?01:09
alex_xuoomichi: I'm trying the microversion one. I'm not sure I picked a right one01:09
oomichialex_xu: api-wg guideline?01:10
alex_xuoomichi: our concept doc01:10
alex_xuoomichi: we have section for it http://docs-draft.openstack.org/13/253513/1/check/gate-compute-api-guide/c0fa091//api-guide/build/html/microversions.html01:10
oomichialex_xu: ah, I see. microversion concept is really necessary for us01:11
alex_xuoomichi: review guide, that is cool01:11
alex_xuoomichi: thanks for doing that!01:11
gmannoomichi: that nice.01:12
gmannalex_xu: yea, that will be nice. So you will update about more on how to use microversion from user perspective right01:12
oomichialex_xu: thanks, nice to share a part of api-wg guideline01:12
*** ljxiash_ has joined #openstack-nova01:13
oomichialex_xu: that comes from johnthetubaguy01:13
*** changbl has joined #openstack-nova01:13
alex_xugmann: yea, let me try a build structure, then I will ask you guys help me review, if that is right, then let's put more detail01:13
alex_xugmann: I guess we probably copy some use-case from sdague's blog https://dague.net/tag/microversions/01:14
gmannalex_xu: yea, Thanks01:14
*** changbl has quit IRC01:14
*** ZZelle_ has quit IRC01:14
*** ljxiash__ has joined #openstack-nova01:14
*** ljxias___ has joined #openstack-nova01:15
*** ljxiash has quit IRC01:15
*** suro-patz has quit IRC01:16
*** ljxias___ has quit IRC01:16
*** armax has quit IRC01:17
tangchenHi alex_xu, oomichi01:17
alex_xutangchen: hi01:17
*** ljxiash has joined #openstack-nova01:17
tangchenalex_xu, oomichi: I just  got to my office01:17
oomichitangchen: morning01:17
alex_xutangchen: cool, morning :)01:17
tangchenalex_xu: morning01:17
alex_xutangchen: which city you are based?01:18
*** penick has quit IRC01:18
*** ljxiash_ has quit IRC01:18
tangchenalex_xu: Nanjing, china01:18
alex_xutangchen: ah, cool, I'm at Beijing01:18
*** ljxiash__ has quit IRC01:18
tangchenalex_xu: Oh, I thought you are abroad.01:19
tangchenalex_xu: Seeing your conversation just now. Was that about the API doc ?01:20
*** betherly has joined #openstack-nova01:20
alex_xutangchen: yeh, I'm not, still enjoy the Beijing's air pollution01:20
alex_xutangchen: yes, today is doc sprint01:20
*** ianbrown_ has quit IRC01:20
*** nithyag_ has joined #openstack-nova01:21
*** ijw has joined #openstack-nova01:21
*** jyuso1 has joined #openstack-nova01:21
*** NithyaG has quit IRC01:23
tangchenalex_xu: Sorry I didn't pay much time on it. Will start to investigate the one I signed up for.01:23
openstackgerritAndrew Bogott proposed openstack/nova: Prevent scheduling of instances on full hosts.  https://review.openstack.org/24225101:23
*** coolsvap has quit IRC01:23
*** jyuso1 has quit IRC01:23
*** jyuso1 has joined #openstack-nova01:23
*** nikhil_k has joined #openstack-nova01:23
*** zhenguo has joined #openstack-nova01:23
alex_xutangchen: no worry, free to take task from https://etherpad.openstack.org/p/nova-v2.1-api-doc01:23
*** tonytan4ever has joined #openstack-nova01:24
tangchenalex_xu: Sure. Thanks.:)01:24
tangchenoomichi: Good, morning. Thanks again for reviewing and pushing the NMI BP. :)01:24
alex_xutangchen: np :)01:25
*** nikhil has quit IRC01:25
*** salv-orlando has quit IRC01:25
*** thorst has joined #openstack-nova01:25
*** thorst has quit IRC01:25
*** coolsvap has joined #openstack-nova01:25
oomichitangchen: welcome, nmi is one of the APIs I prefer ;)01:25
*** ducttape_ has quit IRC01:25
*** yangyapeng has joined #openstack-nova01:26
*** links has joined #openstack-nova01:27
*** Sukhdev has joined #openstack-nova01:28
*** tonytan4ever has quit IRC01:28
oomichialex_xu: gmann: now nova repo contains multiple microversion docs like "how to bump", "how to implement" and basic design01:29
oomichialex_xu: gmann: what will we put microversion thing as concept doc?01:30
alex_xuoomichi: that is for developer01:30
alex_xuoomichi: johnthetubaguy gives suggestion put the use-case01:30
gmannoomichi: we need more on how to use microversion with some example01:30
alex_xuoomichi: like sdague's blog mentioned01:30
alex_xugmann: +101:30
alex_xugmann: I guess that will be in the api-ref?01:31
openstackgerritRyan McNair proposed openstack/nova: Add retry logic for detaching device using LibVirt  https://review.openstack.org/22785101:31
gmannalex_xu: example ?01:31
oomichialex_xu: gmann: OK, concept doc should be use-case, right?01:31
oomichialex_xu: gmann: basically01:31
alex_xuoomichi: yes01:31
oomichialex_xu: for all docs? or only for microversions?01:32
alex_xugmann: probably a top section at http://developer.openstack.org/api-ref-compute-v2.1.html01:32
alex_xugmann: if we said concept doc only for use-case, I only found that is for how-to01:32
*** yangyapeng has quit IRC01:33
alex_xuoomichi: concept doc? yes, concept doc is more about use-case01:33
*** ducttape_ has joined #openstack-nova01:33
oomichialex_xu: I see, nice to contain the purpose of concept doc also in nova repo01:34
*** changbl has joined #openstack-nova01:34
alex_xuoomichi: good idea01:34
gmannalex_xu: yea, so with 'how to' + use case (like new feature can be get with microversion etc) right01:34
*** rook has joined #openstack-nova01:34
gmannoomichi: alex_xu yea, so at high level somewhere we can explain what doc is for what (concept guide, api ref)01:35
oomichialex_xu: gmann: basically, I don't want to put the similar docs on different places because they easily get old and difficult to maintain01:35
gmannoomichi: True01:35
alex_xuoomichi: yes01:35
alex_xuoomichi: gmann there is 'Intended audience' in http://docs-draft.openstack.org/13/253513/1/check/gate-compute-api-guide/c0fa091//api-guide/build/html/01:36
*** baoli has joined #openstack-nova01:36
*** yasu has joined #openstack-nova01:39
*** patrickeast has joined #openstack-nova01:39
*** aginwala has joined #openstack-nova01:39
oomichialex_xu: nice pointing01:39
oomichialex_xu: I am reading...01:39
alex_xuoomichi: probably we can put something at there to clear the scope01:40
* oomichi much time is necessary for reading English doc :(01:40
oomichialex_xu: nice idea !01:40
* alex_xu have same English problem :(01:40
gmannalex_xu: Thanks, yea that helps people01:41
*** ssurana has quit IRC01:41
alex_xugmann: np01:41
*** ssurana has joined #openstack-nova01:41
openstackgerritOpenStack Proposal Bot proposed openstack/python-novaclient: Updated from global requirements  https://review.openstack.org/25391101:42
*** ducttape_ has quit IRC01:42
openstackgerritVilobh Meshram proposed openstack/nova: Introduce Flavor Classes  https://review.openstack.org/25330601:43
*** smoriya has joined #openstack-nova01:43
*** r-daneel has quit IRC01:43
*** unicell1 has joined #openstack-nova01:43
*** unicell has quit IRC01:45
*** unicell1 has quit IRC01:45
*** unicell has joined #openstack-nova01:45
*** ssurana has quit IRC01:45
*** s-moriya has quit IRC01:47
*** changbl has quit IRC01:47
oomichialex_xu: in the doc, words are inconsistent. eg: operator vs. administrator01:47
*** ianbrown_ has joined #openstack-nova01:48
oomichiI worked for news-paper company project before, and I tend to find this kind of conner cases..01:49
alex_xuoomichi: I thought we should use the user which described at https://review.openstack.org/#/c/249812/4/api-guide/source/users.rst01:49
*** Yingxin_ has joined #openstack-nova01:49
alex_xuoomichi: wow, you have many different experience :)01:50
*** akshai has quit IRC01:51
oomichialex_xu: thanks for the link. the patch doesn't mention the difference between operator vs. administrator01:51
*** apoorvad has quit IRC01:51
oomichialex_xu: -1 even if johnthetubaguy's patch ;)01:52
*** Yingxin has quit IRC01:53
*** Yingxin_ is now known as Yingxin01:53
*** jyuso1 has quit IRC01:53
alex_xuoomichi: :)01:53
openstackgerritTony Breeds proposed openstack/nova: Mask the token used to allow access to consoles  https://review.openstack.org/22062201:54
*** wyang has quit IRC01:55
*** rfolco_ has quit IRC01:56
openstackgerritAugustina Ragwitz proposed openstack/nova: Remove required project_id from functional test templates  https://review.openstack.org/25446501:57
*** 7F1AA32U1 has joined #openstack-nova01:58
*** changbl has joined #openstack-nova02:00
*** 7F1AA32U1 has quit IRC02:01
*** haomaiwang has joined #openstack-nova02:01
*** XuZhixing has joined #openstack-nova02:01
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Make "ReSTful service" consistent  https://review.openstack.org/25446602:02
openstackgerritVilobh Meshram proposed openstack/nova-specs: Introduce Disk and Memory Monitor Plugin for Compute  https://review.openstack.org/25446702:03
*** tjones has quit IRC02:04
*** yalie has joined #openstack-nova02:05
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Fix the endpoint of /v2 on concept doc  https://review.openstack.org/25447302:07
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Fix the endpoint of /v2 on concept doc  https://review.openstack.org/25447302:09
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/25390102:09
*** Sukhdev has quit IRC02:11
*** wyang has joined #openstack-nova02:11
openstackgerritAugustina Ragwitz proposed openstack/nova: Generate api sample data without project id's to support v2.13  https://review.openstack.org/25448402:13
openstackgerritAugustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id  https://review.openstack.org/25448502:13
*** Jeffrey4l has quit IRC02:13
*** terryyao_ has joined #openstack-nova02:13
*** Jeffrey4l has joined #openstack-nova02:14
*** jerrygb has joined #openstack-nova02:14
*** yonglihe has joined #openstack-nova02:15
*** terryyao_ is now known as terry_yao02:16
openstackgerritAugustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id  https://review.openstack.org/25448502:16
*** jerrygb has quit IRC02:16
*** terry_yao is now known as yterry02:16
*** yterry is now known as terry_yao02:17
*** zhenq1 has quit IRC02:17
openstackgerritEli Qiao proposed openstack/nova: Calculate block_migration in virt drivers if it's None  https://review.openstack.org/25410102:21
openstackgerritEli Qiao proposed openstack/nova: Return host and block_migration in task_api._execute()  https://review.openstack.org/25410202:21
openstackgerritAlex Xu proposed openstack/nova: docs: add microversions description in the concept doc  https://review.openstack.org/25449002:21
*** yangyape_ has joined #openstack-nova02:22
openstackgerritAugustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id  https://review.openstack.org/25448502:22
*** thorst has joined #openstack-nova02:24
*** thorst has quit IRC02:25
*** thorst has joined #openstack-nova02:25
*** suro-patz has joined #openstack-nova02:27
alex_xugmann: oomichi let me know if this is ok for microversion https://review.openstack.org/#/c/254490/1 , and forgive me if it looks like terible...02:28
* alex_xu spends an hour just write such few lines...02:28
oomichialex_xu: the content seems nice. nice coverage02:29
alex_xuoomichi: thanks :)02:29
oomichialex_xu: but -1 for small points, sorry02:29
alex_xuoomichi: no, thanks for any -1, that will help me02:30
oomichialex_xu: I finished to  review patches on https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/complete-todo-in-api-concept-doc,n,z02:30
alex_xuoomichi: thanks, I saw that, will check with my patch02:31
oomichialex_xu: yeah, please. I will re-check them soon after your post new patches02:31
alex_xuoomichi: ok, got it02:31
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/25390102:32
oomichialex_xu: the cache memory of  my brain is not big, so nice to re-spin them quickly before forgetting the points ;)02:32
alex_xuoomichi: :)02:32
*** otter768 has joined #openstack-nova02:32
*** thorst has quit IRC02:34
*** ijuwang has quit IRC02:38
*** browne has quit IRC02:39
*** jinxing has quit IRC02:40
*** vilobhmm11 has quit IRC02:41
*** vilobhmm11 has joined #openstack-nova02:41
*** ducttape_ has joined #openstack-nova02:41
*** houming has joined #openstack-nova02:42
oomichialex_xu: can you take a look at https://review.openstack.org/#/c/254466/ and https://review.openstack.org/#/c/254473/ ?02:42
oomichialex_xu: both are easy for you02:43
alex_xuoomichi: ok, got it02:43
oomichialex_xu: both changes a single line on each, thanks02:44
chenlihi, anyone can help review https://review.openstack.org/#/c/252237/02:46
*** shihanzhang has joined #openstack-nova02:46
*** shz has quit IRC02:49
*** yamahata has quit IRC02:51
*** baoli has quit IRC02:55
*** jgjhgsonchn88523 has joined #openstack-nova02:56
*** ianbrown_ has quit IRC02:57
*** Marga_ has quit IRC02:57
gmannalex_xu: nice start, just few comments.03:00
*** ducttape_ has quit IRC03:00
alex_xugmann: thanks03:00
*** suro-patz has quit IRC03:00
*** haomaiwang has quit IRC03:01
*** haomaiwang has joined #openstack-nova03:01
*** aginwala has quit IRC03:03
*** akshai has joined #openstack-nova03:04
openstackgerritAlex Xu proposed openstack/nova: Add separated section for extra specs and image properties  https://review.openstack.org/25351203:04
openstackgerritAlex Xu proposed openstack/nova: Add more concepts for servers  https://review.openstack.org/25329203:04
openstackgerritAlex Xu proposed openstack/nova: Add separated section for configure guest os  https://review.openstack.org/25351303:04
alex_xuoomichi: ^ just updated03:04
oomichialex_xu: gmann: one small question: How about updating "updated" on "list versions" API when bumping a microversion?03:05
oomichiI am checking http://developer.openstack.org/api-ref-compute-v2.1.html#listVersionsv2.103:05
oomichibut the updated seems strange for me because of "2013-07-23T11:33:21Z"03:06
oomichialex_xu: ok, will recheck soon03:06
*** suro-patz has joined #openstack-nova03:07
*** aginwala has joined #openstack-nova03:07
*** edmondsw has quit IRC03:07
*** aginwala has quit IRC03:07
gmannoomichi: i see, yea, we can update but that can be done by regenerating the sample file right03:08
gmannoomichi: alex_xu currently functional tests create new sample file (under /doc/api_sample) only if they donot exist03:09
*** shaohe_feng has quit IRC03:09
oomichigmann: nice point. yeah, and new valid sample files are blocked to be created by the existing malformed sample files sometimes due to the behavior.03:10
*** aginwala has joined #openstack-nova03:11
gmannoomichi: yea,03:11
*** ljxiash has quit IRC03:11
gmannoomichi: i remember sdague added new env in tox.ini for sample file creation, let me search the patch03:12
*** ljxiash has joined #openstack-nova03:12
*** unicell has quit IRC03:13
oomichigmann: that is interesting one :)03:13
gmannoomichi: https://review.openstack.org/#/c/253239/03:13
gmannoomichi: yea, very simple way to generate03:13
*** ljxiash_ has joined #openstack-nova03:13
gmannoomichi: but your point is that we should have always updated one on api ref right?03:14
*** ljxiash has quit IRC03:14
*** haomaiwang has quit IRC03:15
oomichialex_xu: done on your patches again, please see https://review.openstack.org/#/c/253513/03:15
*** haomaiwang has joined #openstack-nova03:15
*** haomaiwang has quit IRC03:16
*** haomaiwang has joined #openstack-nova03:16
*** haomaiwang has quit IRC03:17
oomichigmann: now we are genarating sample files by putting some values on the corresponding tpl files.03:17
oomichigmann: and sometimes tpl files are malformed against JSON03:17
oomichigmann: then malformed JSON sample files were generated03:18
*** haomaiwang has joined #openstack-nova03:18
oomichigmann: ideally, it is nice to genarate valid JSON sample files with using python json library.03:18
*** haomaiwang has quit IRC03:19
*** 17WAA5E9R has joined #openstack-nova03:19
*** 17WAA5E9R has quit IRC03:20
*** ijw has quit IRC03:20
*** haomaiwang has joined #openstack-nova03:20
*** haomaiwang has quit IRC03:21
openstackgerritAugustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id  https://review.openstack.org/25448503:21
*** shuquan has joined #openstack-nova03:23
alex_xuoomichi: emm..I should use deployed in line 600 also?03:24
*** browne has joined #openstack-nova03:24
*** haomaiwang has joined #openstack-nova03:24
*** haomaiwang has quit IRC03:25
*** haomaiwang has joined #openstack-nova03:25
*** vladikr has joined #openstack-nova03:25
*** haomaiwang has quit IRC03:26
*** haomaiwang has joined #openstack-nova03:26
*** haomaiwang has quit IRC03:27
*** haomaiwang has joined #openstack-nova03:27
*** haomaiwang has quit IRC03:28
*** shuquan has quit IRC03:28
*** ducttape_ has joined #openstack-nova03:28
*** haomaiwang has joined #openstack-nova03:28
openstackgerritTakashi NATSUME proposed openstack/nova: Ensure consistency of ENV_REQUEST_ID  https://review.openstack.org/25452903:28
*** haomaiwang has quit IRC03:29
*** salv-orlando has joined #openstack-nova03:29
*** haomaiwang has joined #openstack-nova03:29
*** haomaiwang has quit IRC03:30
openstackgerritVilobh Meshram proposed openstack/nova: Introduce Flavor Classes  https://review.openstack.org/25330603:30
*** 7F1AA34GJ has joined #openstack-nova03:30
*** 7F1AA34GJ has quit IRC03:31
*** shuquan has joined #openstack-nova03:31
*** ccard__ has joined #openstack-nova03:31
*** haomaiwang has joined #openstack-nova03:31
*** thorst has joined #openstack-nova03:32
*** haomaiwang has quit IRC03:32
alex_xuoomichi: re: https://review.openstack.org/#/c/253513/2/api-guide/source/server_concepts.rst after reread those words, deploying server and deployed server are right, they describe different time03:33
*** haomaiwa_ has joined #openstack-nova03:33
*** haomaiwa_ has quit IRC03:34
*** salv-orlando has quit IRC03:34
*** ccard_ has quit IRC03:34
*** haomaiwang has joined #openstack-nova03:34
*** haomaiwang has quit IRC03:35
*** jerrygb has joined #openstack-nova03:35
*** haomaiwa_ has joined #openstack-nova03:38
*** thorst has quit IRC03:39
*** vilobhmm11 has quit IRC03:42
*** terry_yao has quit IRC03:42
*** ccard_ has joined #openstack-nova03:44
*** unicell has joined #openstack-nova03:45
*** ccard__ has quit IRC03:46
openstackgerritjaveme proposed openstack/nova: update task state for the confirm_resize  https://review.openstack.org/24443303:48
*** jgjhgsonchn88523 has quit IRC03:49
gmannoomichi: yea nice point03:50
*** ducttape_ has quit IRC03:54
*** ssurana has joined #openstack-nova03:54
*** ducttape_ has joined #openstack-nova03:54
*** ianbrown has joined #openstack-nova03:55
*** jgjhgsonchn88523 has joined #openstack-nova03:57
openstackgerritZhenyu Zheng proposed openstack/nova: Add project-id and user-id when list server-groups  https://review.openstack.org/20901803:57
*** haomaiwa_ has quit IRC04:01
*** baoli has joined #openstack-nova04:01
*** haomaiwang has joined #openstack-nova04:01
*** jgjhgsonchn88523 has quit IRC04:04
*** tonytan4ever has joined #openstack-nova04:05
*** darrenc is now known as darrenc_afk04:06
*** eliqiao has joined #openstack-nova04:06
*** eliqiao_ has quit IRC04:06
openstackgerritJinxing Fang proposed openstack/nova: remove the qpid message driver  https://review.openstack.org/25453504:07
*** mc_nair_ has quit IRC04:09
*** tonytan4ever has quit IRC04:09
*** aginwala has quit IRC04:13
*** Jeffrey4l has quit IRC04:14
*** aginwala has joined #openstack-nova04:16
*** ducttape_ has quit IRC04:17
*** chenli has quit IRC04:19
openstackgerritGhanshyam Mann proposed openstack/nova: docs: add concept guide for certificate  https://review.openstack.org/25453704:20
*** aginwala has quit IRC04:20
gmannalex_xu: oomichi ^^ for certificate, please have look04:20
*** shuquan has quit IRC04:25
*** shuquan has joined #openstack-nova04:26
*** otter768 has quit IRC04:27
*** tpatil has quit IRC04:34
*** mdrabe has joined #openstack-nova04:34
*** cfarquhar has quit IRC04:34
*** ianbrown has quit IRC04:35
*** baoli has quit IRC04:37
*** thorst has joined #openstack-nova04:38
*** darrenc_afk is now known as darrenc04:42
*** boris-42_ has quit IRC04:43
*** pratikmallya has joined #openstack-nova04:43
*** thorst has quit IRC04:44
openstackgerritjaveme proposed openstack/python-novaclient: remove the default arguments "[]"  https://review.openstack.org/25417104:45
*** masayukig has quit IRC04:47
*** terryyao_ has joined #openstack-nova04:47
*** shuquan has quit IRC04:50
*** chenli has joined #openstack-nova04:52
openstackgerritNanZhang proposed openstack/nova: live-migration: Update instance availability_zone  https://review.openstack.org/25404104:52
*** yamahata has joined #openstack-nova04:53
*** cfarquhar has joined #openstack-nova04:54
*** cfarquhar has quit IRC04:54
*** cfarquhar has joined #openstack-nova04:54
*** masayukig has joined #openstack-nova04:54
*** dave-mccowan has quit IRC04:55
*** dave-mccowan has joined #openstack-nova04:57
*** abhishekk has joined #openstack-nova05:00
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-nova05:01
*** ianbrown has joined #openstack-nova05:02
*** suro-patz1 has joined #openstack-nova05:03
*** suro-patz has quit IRC05:05
*** fawadkhaliq has joined #openstack-nova05:12
openstackgerritAlex Xu proposed openstack/nova: Add separated section for configure guest os  https://review.openstack.org/25351305:14
*** dave-mccowan has quit IRC05:14
*** tbh has joined #openstack-nova05:14
*** sudipto has joined #openstack-nova05:15
*** dave-mccowan has joined #openstack-nova05:17
oomichigmann: I send a mail related to http203: http://lists.openstack.org/pipermail/openstack-dev/2015-December/081690.html05:19
gmannoomichi: nice Thanks.05:20
*** amotoki has joined #openstack-nova05:20
*** Marga_ has joined #openstack-nova05:20
alex_xuoomichi: just reploy some comments for https://review.openstack.org/#/c/254490/1/api-guide/source/microversions.rst05:20
*** akshai has quit IRC05:22
oomichialex_xu: replied again ;)05:24
alex_xuoomichi: thanks, learned :)05:24
*** Marga_ has quit IRC05:25
*** Sukhdev has joined #openstack-nova05:26
*** ankit_ag has quit IRC05:28
*** cfriesen has quit IRC05:28
*** shuquan has joined #openstack-nova05:30
*** Marga_ has joined #openstack-nova05:32
*** Marga_ has quit IRC05:32
*** Marga_ has joined #openstack-nova05:33
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Make admin consistent  https://review.openstack.org/25454905:37
tbhHi, I am new to nova, I am following https://wiki.openstack.org/wiki/Nova/Mentoring, I am interested in https://bugs.launchpad.net/nova/+bug/135775105:40
openstackLaunchpad bug 1357751 in OpenStack Compute (nova) "nova.tests.network.test_manager.AllocateTestCase should use mock" [Low,Confirmed] - Assigned to Aniruddha Singh Gautam (aniruddha-gautam)05:40
tbhas there is no activity for that bug, can I assign that bug to myself05:40
*** aginwala has joined #openstack-nova05:41
oomichialex_xu: question about https://etherpad.openstack.org/p/nova-v2.1-api-doc05:41
*** thorst has joined #openstack-nova05:42
*** claudiub has quit IRC05:42
oomichialex_xu: the etherpad seems to concentrate on api-site without concept guideline, right?05:42
*** ljxiash_ has quit IRC05:42
*** ljxiash has joined #openstack-nova05:43
oomichialex_xu: when we found a problem on concept guideline(under api-guide/source/), where should we put patches?05:43
*** dave-mccowan has quit IRC05:45
*** shuquan has quit IRC05:46
*** ljxiash has quit IRC05:47
*** shuquan has joined #openstack-nova05:47
*** amotoki has quit IRC05:48
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Remove XML description from extension concept  https://review.openstack.org/25455105:48
openstackgerritSudipta Biswas proposed openstack/nova: Remove version hardcoding for 'show' method  https://review.openstack.org/22857805:49
*** thorst has quit IRC05:49
*** jerrygb has quit IRC05:50
*** mdrabe has quit IRC05:51
*** jaosorior has joined #openstack-nova05:51
openstackgerritNanZhang proposed openstack/nova: live-migration: Update instance availability_zone  https://review.openstack.org/25404105:53
*** ianbrown has quit IRC05:57
*** jaypipes has joined #openstack-nova05:57
*** ianbrown has joined #openstack-nova05:57
*** ianbrown_ has joined #openstack-nova05:57
*** ianbrown_ has quit IRC05:57
*** dims has quit IRC05:57
*** yamahata has quit IRC05:58
*** yamahata has joined #openstack-nova05:58
*** ljxiash has joined #openstack-nova05:59
*** ianbrown_ has joined #openstack-nova06:00
*** ianbrown has quit IRC06:00
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-nova06:01
*** pradipta has joined #openstack-nova06:02
*** wolsen has quit IRC06:04
*** yamahata has quit IRC06:04
alex_xuoomichi: yea, we didn't have place for concept guideline, probably if it is a todo, just add a patch for it.06:04
*** wolsen has joined #openstack-nova06:05
*** chinmaya has joined #openstack-nova06:06
alex_xuoomichi: but if you put in the etherpad, sounds ok also06:06
oomichialex_xu: maybe no problem without the otherpad. https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/complete-todo-in-api-concept-doc,n,z works fine06:07
*** moshele_1 has joined #openstack-nova06:07
openstackgerritAlex Xu proposed openstack/nova: docs: add microversions description in the concept doc  https://review.openstack.org/25449006:08
*** moshele_1 has quit IRC06:08
*** donghao has joined #openstack-nova06:09
*** yamahata has joined #openstack-nova06:09
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata  https://review.openstack.org/25219706:10
*** shuquan has quit IRC06:15
*** ildikov has quit IRC06:16
*** ianbrown_ has quit IRC06:17
*** pratikmallya has quit IRC06:18
*** Sukhdev has quit IRC06:20
*** dims has joined #openstack-nova06:20
*** Sukhdev has joined #openstack-nova06:20
alex_xuoomichi: ok, thanks06:20
*** ianbrown has joined #openstack-nova06:23
*** Sukhdev has quit IRC06:25
*** moshele has joined #openstack-nova06:25
*** esracelik has joined #openstack-nova06:26
*** suro-patz1 has quit IRC06:28
*** suro-patz has joined #openstack-nova06:28
*** otter768 has joined #openstack-nova06:28
*** yangyape_ is now known as Yangyapeng06:28
*** takashin has left #openstack-nova06:29
*** XuZhixing has quit IRC06:31
*** haomaiwang has quit IRC06:32
*** otter768 has quit IRC06:33
*** 18WABB396 has joined #openstack-nova06:33
*** rook has quit IRC06:34
*** rook has joined #openstack-nova06:34
*** ljxiash has quit IRC06:34
*** ljxiash has joined #openstack-nova06:35
*** donghao has quit IRC06:37
*** garyk has joined #openstack-nova06:37
*** shuquan has joined #openstack-nova06:38
*** e0ne has joined #openstack-nova06:38
*** Jeffrey4l has joined #openstack-nova06:39
*** ljxiash has quit IRC06:40
*** ianbrown has quit IRC06:40
*** sahid has joined #openstack-nova06:40
*** ianbrown has joined #openstack-nova06:40
*** ianbrown has quit IRC06:41
*** ianbrown_ has joined #openstack-nova06:41
*** suro-patz has quit IRC06:43
*** ljxiash has joined #openstack-nova06:43
*** aginwala has quit IRC06:43
*** ianbrown__ has joined #openstack-nova06:43
*** ianbrown_ has quit IRC06:43
*** e0ne has quit IRC06:45
*** aginwala has joined #openstack-nova06:46
*** eliqiao has quit IRC06:48
*** thorst has joined #openstack-nova06:48
*** salv-orlando has joined #openstack-nova06:49
*** smoriya has quit IRC06:49
*** e0ne has joined #openstack-nova06:49
*** vilobhmm11 has joined #openstack-nova06:50
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/25390106:50
*** e0ne has quit IRC06:52
*** aginwala has quit IRC06:52
*** salv-orlando has quit IRC06:54
*** thorst has quit IRC06:54
*** unicell has quit IRC06:57
*** unicell has joined #openstack-nova06:58
*** ankit_ag has joined #openstack-nova07:00
*** 18WABB396 has quit IRC07:01
*** haomaiwang has joined #openstack-nova07:01
openstackgerritYingxin Cheng proposed openstack/nova: Use stevedore for scheduler host manager  https://review.openstack.org/24647607:01
*** e0ne has joined #openstack-nova07:02
*** mjura has joined #openstack-nova07:03
*** ljxiash has quit IRC07:04
*** ljxiash has joined #openstack-nova07:05
*** eliqiao has joined #openstack-nova07:07
*** houming has quit IRC07:08
*** haomaiwang has quit IRC07:09
*** haomaiwang has joined #openstack-nova07:09
*** dims has quit IRC07:11
*** houming has joined #openstack-nova07:11
*** scheuran has joined #openstack-nova07:14
*** houming has quit IRC07:16
*** vilobhmm11 has quit IRC07:18
*** e0ne has quit IRC07:20
*** e0ne has joined #openstack-nova07:21
*** terryyao_ has quit IRC07:21
*** shuquan has quit IRC07:22
*** nkrinner has joined #openstack-nova07:22
*** aginwala has joined #openstack-nova07:25
*** terryyao_ has joined #openstack-nova07:25
*** shuquan has joined #openstack-nova07:25
*** haomaiwang has quit IRC07:28
*** haomaiwa_ has joined #openstack-nova07:28
*** dims has joined #openstack-nova07:31
*** ljxiash has quit IRC07:31
*** terryyao_ has quit IRC07:31
*** ljxiash has joined #openstack-nova07:33
*** doron_afk has joined #openstack-nova07:36
*** terryyao_ has joined #openstack-nova07:39
*** fawadkhaliq has quit IRC07:41
*** fawadkhaliq has joined #openstack-nova07:42
*** houming has joined #openstack-nova07:42
*** jaypipes has quit IRC07:44
*** lpetrut has joined #openstack-nova07:45
tangchenHi alex_xu, oomichi, I think we can start to push the NMI (crash dump) work. Would you please review this patch when you are free ?07:48
tangchenalex_xu, oomichi: https://review.openstack.org/#/c/208794/07:48
tangchenalex_xu, oomichi: It is an improvement for the merged libvirt inject_nmi() API.07:49
tangchenalex_xu, oomichi: Thanks.07:49
*** jichen has joined #openstack-nova07:51
*** dims has quit IRC07:52
*** shuquan has quit IRC07:53
*** thorst has joined #openstack-nova07:53
*** shuquan has joined #openstack-nova07:54
*** wyang has quit IRC07:55
*** e0ne has quit IRC07:56
*** thorst has quit IRC07:58
openstackgerritMichal Pryc proposed openstack/nova: Implements proper UUID format for the ComputeAPITestCase  https://review.openstack.org/25439808:00
*** zenoway has joined #openstack-nova08:00
openstackgerritCindia-blue proposed openstack/nova: VMware: Enable Fault tolerance on VM  https://review.openstack.org/25180708:02
*** ildikov has joined #openstack-nova08:03
*** jerrygb has joined #openstack-nova08:06
*** matrohon has joined #openstack-nova08:07
*** rk4n has joined #openstack-nova08:07
BobBall_AWOLmordred edleafe: Seems like a temporary failure to download some packages - has resolved itself.08:09
*** BobBall_AWOL is now known as BobBall08:09
*** jerrygb has quit IRC08:12
*** rdopiera has joined #openstack-nova08:13
openstackgerritNanZhang proposed openstack/nova: live-migration: Update instance availability_zone  https://review.openstack.org/25404108:13
*** eliqiao_ has joined #openstack-nova08:15
openstackgerritClaudiu Belu proposed openstack/nova: Compute: Adds driver disk_gb instance overhead estimation  https://review.openstack.org/23428308:15
openstackgerritCindia-blue proposed openstack/nova: VMware: Enable Fault tolerance on VM  https://review.openstack.org/25180708:17
*** rook has quit IRC08:18
*** eliqiao has quit IRC08:18
*** rook has joined #openstack-nova08:18
*** klkumar has joined #openstack-nova08:23
*** liverpooler has joined #openstack-nova08:26
*** liverpooler has quit IRC08:26
*** gcb has joined #openstack-nova08:27
*** otter768 has joined #openstack-nova08:29
*** liverpooler has joined #openstack-nova08:30
*** mpavone has joined #openstack-nova08:30
*** moshele has quit IRC08:33
*** otter768 has quit IRC08:34
*** markus_z has joined #openstack-nova08:37
*** klkumar has quit IRC08:38
*** yonglihe has quit IRC08:41
*** yamahata has quit IRC08:43
*** jlanoux has joined #openstack-nova08:44
*** lyanchih has joined #openstack-nova08:48
*** aginwala has quit IRC08:49
*** klkumar has joined #openstack-nova08:52
*** aginwala has joined #openstack-nova08:53
*** eliqiao_ is now known as eliqiao08:54
*** rk4n has quit IRC08:54
*** links has quit IRC08:55
*** browne has quit IRC08:56
*** thorst has joined #openstack-nova08:56
*** paul-carlton2 has joined #openstack-nova08:58
bauzasgood morning Nova09:00
*** aginwala has quit IRC09:01
*** aginwala has joined #openstack-nova09:01
markus_zgood morning bauzas09:02
oomichibauzas: morning09:02
bauzasmarkus_z: are you better ? :)09:02
markus_zbauzas: Yeah, it will do I guess, thanks for asking09:03
oomichijohnthetubaguy: bauzas: can you take a look at https://review.openstack.org/#/c/253417 ?09:03
markus_zbauzas: We already have hit spec-freeze, right?09:03
*** mpavone has quit IRC09:03
*** e0ne has joined #openstack-nova09:03
bauzasmarkus_z: yeah09:03
oomichijohnthetubaguy: bauzas: it is nice to know reviewing points related to api thing09:04
*** thorst has quit IRC09:04
bauzasoomichi: reviewing it now09:04
markus_zbauzas: Because my blueprint is still open: https://review.openstack.org/#/c/234291/09:04
bauzasoomichi: alex_xu: correct me if I'm wrong but it's doc day?09:04
oomichibauzas: yeah, right. doc day09:05
bauzaswell, it's end of the day for folks living in an earlier TZ :p09:05
bauzasoomichi: okay, any pointers I should look to ? I saw some etherpad09:05
oomichibauzas: patches are ready to be reviewed: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/complete-todo-in-api-concept-doc,n,z09:05
*** mpavone has joined #openstack-nova09:05
oomichibauzas: the etherpad is https://etherpad.openstack.org/p/nova-v2.1-api-doc09:05
bauzasoomichi: okay, I'll pick those too09:06
oomichibauzas: cool! :)09:06
markus_zbauzas: I guess I did the status wrong again: https://blueprints.launchpad.net/nova/+spec/libvirt-virtlogd09:06
esracelikmarkus_z: hi markus, i hope you're better09:06
markus_zesracelik: Hi esra, yes I am, thanks for asking.09:07
esracelikmarkus_z: markus could you check https://review.openstack.org/#/c/254092/09:07
bauzasmarkus_z: I guess johnthetubaguy had no time to look thru all the open specs yet09:07
markus_zesracelik: yepp, checking it right now09:07
esracelikmarkus_z: ok, i will have some questions then09:07
*** links has joined #openstack-nova09:09
*** aginwala_ has joined #openstack-nova09:10
openstackgerritVladik Romanovsky proposed openstack/nova: pci: adding support to specify a device_type in pci requests  https://review.openstack.org/24901709:11
openstackgerritVladik Romanovsky proposed openstack/nova: pci: do not filter out any SRIOV Physical Functions  https://review.openstack.org/24901609:11
openstackgerritVladik Romanovsky proposed openstack/nova: pci: changing the claiming and allocation logic for PF/VF assignment  https://review.openstack.org/24901809:11
openstackgerritVladik Romanovsky proposed openstack/nova: db: adding columns to PciDevice table  https://review.openstack.org/24901309:11
openstackgerritVladik Romanovsky proposed openstack/nova: objects: adding a parent_addr field to the PciDevice object  https://review.openstack.org/24901509:11
openstackgerritVladik Romanovsky proposed openstack/nova: db: querry to retrieve all pci device by parent address  https://review.openstack.org/24901409:11
*** rgerganov has joined #openstack-nova09:11
markus_zesracelik: looks good! I have to double-check the help text. It would also be good to separate the work into two patches. One does the movement and the second patch enhances the help text. It easier to grasp what happened when doing a review.09:12
*** ndipanov has quit IRC09:12
markus_zesracelik: Which open questions do you have?09:12
esracelikmarkus_z: ok should also I seperate this one?09:13
*** ndipanov has joined #openstack-nova09:13
markus_zesracelik: Yes, that would be helpful.09:13
*** shakamunyi has joined #openstack-nova09:13
rgerganovjohnthetubaguy: could you please take a look at https://review.openstack.org/#/c/228778/ ? it is trivial to implement and very important for the VM console story; it also has a lot of +1s09:13
bauzasoomichi: I'm reviewing https://review.openstack.org/#/c/253417/3/doc/source/code-review.rst,cm but I need more understanding09:14
*** aginwala has quit IRC09:14
oomichibauzas: ok, please go ahead09:14
bauzasoomichi: is that you saw some API changes not enforcing https://review.openstack.org/#/c/253417/3/doc/source/code-review.rst,cm ?09:14
oomichibauzas: yeah, many times09:15
bauzasoomichi: I guess it's also for new changes, right? not a clear enforcement for the existing code?09:15
*** aginwala_ has quit IRC09:15
esracelikmarkus_z: ok, i will.. my first question is: there are lots of config files in different directories, but we are merging them in one directory. I mean wouldn't it better to create subdirectories in /nova/conf09:15
oomichibauzas: only for new API or new API changes09:15
esracelikmarkus_z: like /nova/conf/virt/configdrive.py09:15
bauzasoomichi: so I guess adding an hacking check wouldn't help because we would fail because of the existing bits?09:15
oomichibauzas: yeah, right.09:16
*** ildikov has quit IRC09:16
*** eglynn has joined #openstack-nova09:16
markus_zesracelik: What would you move to the folder "virt"?09:16
bauzasoomichi: are you planning to have some consistency eventually ? I know mitaka prios so I wonder if it could be a Nutella stuff09:16
esracelikmarkus_z: nova/conf/virt/configdrive.py,netutils.py,agent.py so on09:17
bauzasoomichi: see, you said "reviewers should read all of the guidelines, but they are very long", so my concern is maybe to see how we could automatically prevent that09:18
*** ildikov has joined #openstack-nova09:18
bauzasoomichi: because indeed, I have to admit that I'm very not familiar with the API-WG rules09:18
bauzas(that's also why I should refrain to +W an API change :) )09:18
esracelikmarkus_z: or just one virt.py in nova/conf that includes all options in /nova/virt directory09:19
oomichibauzas: and it is difficult to apply this guideline to the existing APIs09:19
oomichibauzas: then, as you said, difficult to have some pep8 rules for blocking invalid api changes.09:20
*** shuquan has quit IRC09:20
markus_zesracelik: Let me rephrase it, what would be the grouping criteria of "nova/conf/virt"?09:20
oomichibauzas: so that is a reason why I required you also to see it ;)09:20
markus_zesracelik: The first usage of an option?09:20
openstackgerritVladik Romanovsky proposed openstack/nova: pci: changing the claiming and allocation logic for PF/VF assignment  https://review.openstack.org/24901809:20
oomichibauzas: I don't ask you +A for that ;)09:20
esracelikmarkus_z: just the current location of the option. for example all options in nova/virt directory09:21
*** yassine__ has joined #openstack-nova09:21
bauzasoomichi: true thing, hence me not +W'ing that change :)09:21
oomichibauzas: that is very nice way as reviewer!09:22
*** tonytan4ever has joined #openstack-nova09:22
oomichibauzas: thanks much!09:22
markus_zesracelik: Are you worried about the number of modules in "nova/conf" or do you want to have a "location indicator" or another reasoning?09:22
*** shakamunyi has quit IRC09:23
*** shuquan has joined #openstack-nova09:23
*** sfinucan has joined #openstack-nova09:23
esracelikmarkus_z yes actually the number of modules in nova/conf.. maybe not that important09:23
*** danpb has joined #openstack-nova09:24
markus_zesracelik: I wouldn't worry about that right now, to be honest. If we want to introduce subfolders in the future, then it can be easily done within the "nova/conf/" folder and its "__init__.py" without anyone noticing it outside.09:26
*** pradipta has quit IRC09:26
markus_zbecause they "outside" world imports "nova.conf"09:26
*** tonytan4ever has quit IRC09:26
markus_zthat's the abstraction layer and interface here09:27
esracelikmarkus_z: yes right.. second question: for my following patches, will I add my previous works as a dependency09:28
openstackgerritStephen Finucane proposed openstack/nova: Change logging level for 'oslo_db'  https://review.openstack.org/24027109:28
esracelikmarkus_z: like this one  https://review.openstack.org/#/c/252687/09:29
markus_zesracelik: Yes, that would be the best. In Gerrit you will see them as "related changes".09:29
markus_ztracy builds unrelated dependencies there, I wouldn't recommend it. I will talk to her.09:30
openstackgerritesracelik proposed openstack/nova: Config options: centralize "configdrive" options  https://review.openstack.org/25409209:30
esracelikmarkus_z: i removed help texts for now09:31
markus_zThey way she is doing it will prevent that the volume changes will merge before the guestfs changes are merged. But they are unrelated.09:31
*** rotbeard has joined #openstack-nova09:32
markus_zesracelik: cool, I have to admit you're pretty fast09:32
esracelikmarkus_z: oh, I meant adding unrelated dependencies as she did.. i will not then.09:32
*** tbh has quit IRC09:33
esracelikmarkus_z: :) thanks, i hope it allways goes that way09:33
openstackgerritStephen Finucane proposed openstack/nova: config options: Centralise PCI options  https://review.openstack.org/25373209:33
openstackgerritStephen Finucane proposed openstack/nova: Improve 'pci_passthrough_whitelist' documentation  https://review.openstack.org/24333509:33
*** pratikmallya has joined #openstack-nova09:34
*** chenli has quit IRC09:34
openstackgerritSergey Nikitin proposed openstack/nova: enginefacade: 'migration'  https://review.openstack.org/24349609:34
markus_zesracelik: The workflow doesn't change that much after this task, that's the cool thing about it.09:35
esracelikmarkus_z: this work is pretty fun for me. and instructive too09:37
alex_xubauzas: yea, today is doc sprint :)09:37
*** macsz has quit IRC09:39
ankit_agndipanov: hi09:40
ndipanovhi ankit_ag09:40
ankit_agndipanov: Could you please check my reply on https://review.openstack.org/#/c/221162/09:40
ndipanovankit_ag, good question09:41
sfinucansdague: Morning. Could you cast an eye over this change once more. Interested in your take on it, given my feedback https://review.openstack.org/#/c/240271/09:41
ankit_agndipanov: :D09:41
*** terryyao_ has quit IRC09:41
ankit_agndipanov: The only issue with previous PS I see is an additional DB call09:42
ndipanovankit_ag, yeah09:42
ndipanovthat and the fact that it is going to be an inaccurate list of instances that it uses09:42
openstackgerritPaul Carlton proposed openstack/nova: Mask the token used to allow access to consoles  https://review.openstack.org/22062209:43
alex_xubauzas: free to review those patch https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/complete-todo-in-api-concept-doc,n,z and free to pick any one you are expert on that.09:43
*** ildikov has quit IRC09:45
openstackgerritgaryk proposed openstack/nova: Scheduler: honor the glance metadata for hypervisor details  https://review.openstack.org/25420609:45
*** macsz has joined #openstack-nova09:46
ankit_agndipanov: do you have any quick solution in mind09:47
*** ljxiash has quit IRC09:48
*** tbh has joined #openstack-nova09:48
alex_xuFor the people they just wakeup, today is nova api doc sprint!09:48
*** ljxiash has joined #openstack-nova09:49
*** ljxiash_ has joined #openstack-nova09:50
ndipanovankit_ag, will comment in a sec09:50
ankit_agndipanov: sure, thank you09:51
*** moshele has joined #openstack-nova09:51
*** ljxiash_ has quit IRC09:52
*** ljxiash has quit IRC09:54
*** pratikmallya has quit IRC09:54
bauzasalex_xu: sure, that's what I'm doing09:54
*** StevenK has quit IRC09:56
bauzasoomichi: still around ?09:56
bauzasoomichi: https://review.openstack.org/#/c/254551/1/api-guide/source/extensions.rst,cm removes all the section, not only about XML09:56
bauzasoomichi: including the headers doc, etc.09:57
*** moshele has quit IRC09:57
*** StevenK has joined #openstack-nova09:57
oomichibauzas: yep, untill the end of nova-api meeting ;-)   I got +A from my family for overtime work today09:57
alex_xubauzas: thanks :)09:57
*** moshele has joined #openstack-nova09:57
*** markmc has joined #openstack-nova09:57
alex_xuoomichi: heh :)09:57
oomichibauzas: yeah, that seems confused. actually I removed the first part on my env. but actually, the section mentioned about XML thing09:58
bauzasoomichi: hah09:58
*** ildikov has joined #openstack-nova09:59
eliqiaohi what the problem with xen ci?09:59
*** houming has quit IRC09:59
oomichibauzas: In JSON, there is not any namespace.09:59
eliqiaowhy can not access ci logs http://logs.openstack.xenproject.org/02/254102/2/check/dsvm-tempest-xen/71cf9fb10:00
*** macsz has quit IRC10:00
*** aix has joined #openstack-nova10:01
*** macsz has joined #openstack-nova10:01
*** wyang has joined #openstack-nova10:01
oomichieliqiao: oops, I am not sure the reason10:02
oomichieliqiao: do you face the same problem even after rechecks?10:02
openstackgerritjichenjc proposed openstack/nova: Add hypervisor, aggregates, migration description  https://review.openstack.org/25073710:02
eliqiaooomichi: I have 2 patches facing this issue, just did recheck, will see it later.10:03
*** zhenguo has quit IRC10:03
ankit_agndipanov: Thanks for the time and your valuable comments10:04
oomichieliqiao: the first xen job succeeded and we can see the log10:04
*** thorst has joined #openstack-nova10:04
oomichieliqiao: on your patch10:05
bauzasoomichi: right, but it was saying "In JSON, the alias must be used" so I felt that "The volumes element is defined in the ``RS-CBS`` namespace" was referring to the alias, not the namespace with JSON10:05
ndipanovankit_ag, thanks for the kind words :)10:05
bauzasoomichi: that said, I totally get we can't just cut one phrase :)10:05
eliqiaooomichi: okay, I will double check it again.10:05
oomichieliqiao: nice to do just "xen: recheck"10:05
eliqiaooomichi: sure, thanks.10:06
oomichieliqiao: welcome :)10:06
oomichibauzas: ok, I need to dig it more ;)10:07
bauzasoomichi: that said, that whole sentence is just japanese to me, so maybe I'm wrong10:07
*** tbh has quit IRC10:08
openstackgerritjichenjc proposed openstack/nova: Change some wording on server_concepts.rst  https://review.openstack.org/24806310:09
*** thorst has quit IRC10:09
oomichibauzas: heh.. yeah, that seems French for me instead :)10:09
oomichibauzas: anyways, I need to add the reason why we can remove the sentence on the commit message10:10
markus_zalex_xu: Where will I see the result of the api doc sprint? At http://developer.openstack.org/api-ref-compute-v2.1.html ?10:13
alex_xumarkus_z: there is etherpad to check the work https://etherpad.openstack.org/p/nova-v2.1-api-doc10:13
alex_xumarkus_z: and there are two api doc need improve, one is concept doc, another one is api-ref http://developer.openstack.org/api-ref-compute-v2.1.html10:14
*** jistr has joined #openstack-nova10:14
markus_zalex_xu: Ah, cool. And it's not yet swagger, right?10:15
alex_xumarkus_z: this is the link for concept doc http://developer.openstack.org/api-guide/compute/10:15
alex_xumarkus_z: yes, it's not yet swagger. whatever we have swagger or not, we still need those content for docs10:16
alex_xuso current the api team focus on the doc content improvement10:16
*** ljxiash has joined #openstack-nova10:16
openstackgerritjichenjc proposed openstack/nova: Change some wording on server_concepts.rst  https://review.openstack.org/24806310:16
markus_zalex_xu: still awesome! It was always a bit hard for me to understand the rest api in detail.10:17
alex_xumarkus_z: yeh, the current api ref is really need more detail10:18
*** usdsd has joined #openstack-nova10:21
*** fawadkhaliq has quit IRC10:21
*** ljxiash has quit IRC10:21
*** aix has quit IRC10:21
*** fawadkhaliq has joined #openstack-nova10:21
johnthetubaguyrgerganov: I am not sure I understood the big picture with how those tokens get used, seemed to be assuming code out of the Nova tree was using the tokens, which seems confusing. But we are frozen for new specs and blueprints now, so I have added it to the exception request list10:24
*** salv-orlando has joined #openstack-nova10:25
*** ssurana has quit IRC10:26
johnthetubaguyrgerganov: actually I get it now, you just get the connection info from the token10:27
*** usdsd has left #openstack-nova10:28
*** gcb has quit IRC10:29
*** salv-orlando has quit IRC10:30
*** otter768 has joined #openstack-nova10:30
*** gcb has joined #openstack-nova10:31
*** rk4n has joined #openstack-nova10:32
*** aix has joined #openstack-nova10:33
*** otter768 has quit IRC10:35
garykdanpb: can you please look at https://review.openstack.org/25420610:36
openstackgerrithuan proposed openstack/nova: XenAPI: Do not rely on XenAPI to get device names  https://review.openstack.org/21311210:37
*** cdent has joined #openstack-nova10:37
*** links has quit IRC10:37
tangchenjohnthetubaguy: Hi John, since the NMI (crash dump) BP has been approved, and the spec was also merged, would you please help to remove the -2 of these patches ?10:38
tangchenjohnthetubaguy: I'll restart this work from this week.10:39
tangchenjohnthetubaguy: Thanks.10:39
johnthetubaguytangchen: must have missed those, will go check that out10:40
tangchenjohnthetubaguy: Not in a hurry. Please take your time.  Thanks a lot. :)10:41
openstackgerritAdelina Tuvenie proposed openstack/nova: Moves OVS related code to a common location  https://review.openstack.org/17972710:41
johnthetubaguytangchen: no worries, doing it now before I forget, should have all gone now10:42
tangchenjohnthetubaguy: Yes. Thank you:)10:43
ndipanovjohnthetubaguy, what am I missing here https://review.openstack.org/#/c/199205/10:43
*** ljxiash has joined #openstack-nova10:44
ndipanovjohnthetubaguy, iiuc how major versions work we should currently support 4. and 5. with the proxy class, and then drop 4 on both client and server in Nandos right?10:45
johnthetubaguyndipanov: I have basically got bauzas not to bump the major version, as there is no need to that, and its easier to do that at the end of the release10:45
ndipanovjohnthetubaguy, my thoughts exactly10:45
johnthetubaguyyeah, its overly complicated as we add extra stuff to do it now10:45
bauzasndipanov: johnthetubaguy: yeah I need to respon10:46
ndipanovbauzas, OK let me just comment so that there's no confusion10:46
johnthetubaguysorry, I may not have updated my comments after changing10:46
johnthetubaguyndipanov: cool, thanks10:46
*** mpavone has quit IRC10:47
ndipanovdid we decide on the N name?10:47
bauzasNutella !10:47
ndipanovis it Nandos?10:47
bauzasNando is fair10:47
johnthetubaguyndipanov: not sure we did the vote yet, but I may have just missed it10:47
johnthetubaguyeither of those, an I am going to be hungry for a whole 6 months10:48
bauzashttps://wiki.openstack.org/wiki/Release_Naming nothing yet10:48
bauzasNachos !10:48
*** zhenq has joined #openstack-nova10:48
*** ljxiash has quit IRC10:48
ndipanovbauzas, not sure why the bottom patch doesn't want to recheck10:49
ndipanovyou may actually have to re-push the series10:49
bauzasndipanov: lemme see10:49
bauzasndipanov: I saw you rechecking this morning10:49
*** rook has quit IRC10:49
*** rook has joined #openstack-nova10:50
*** shuquan has quit IRC10:51
ndipanovbauzas, I don't think it did anything10:51
*** rfolco_ has joined #openstack-nova10:54
*** shihanzhang has quit IRC10:55
*** shz has joined #openstack-nova10:56
johnthetubaguybauzas: ndipanov: I think that might really need a reverify rather than recheck, but its usually easier to just rebase at this point10:57
openstackgerritAdelina Tuvenie proposed openstack/nova: Adds Hyper-V OVS ViF driver  https://review.openstack.org/14004510:57
bauzasjohnthetubaguy: reverify = recheck now10:57
*** links has joined #openstack-nova10:58
bauzasbut lemme doublecheck the zuul layout10:58
openstackgerritStephen Finucane proposed openstack/nova: config options: Centralise PCI options  https://review.openstack.org/25373210:58
openstackgerritStephen Finucane proposed openstack/nova: Improve 'pci_passthrough_whitelist' documentation  https://review.openstack.org/24333510:58
johnthetubaguybauzas: hmm, yeah, I remember something about that now10:58
johnthetubaguybauzas: you have pep8 failures on the follow up patch now, somehow?10:58
*** Yangyapeng has quit IRC10:58
ndipanovyeah I thought that too10:58
johnthetubaguybauzas: it says: ./nova/scheduler/filters/numa_topology_filter.py:48:41: F821 undefined name 'instance'10:58
bauzasjohnthetubaguy: yeah it's due to an automatic rebase when testing10:58
bauzasjohnthetubaguy: so gerrit doesn't show the mere10:59
johnthetubaguybauzas: if you refresh it, I can fast approve it for you10:59
bauzasjohnthetubaguy: but I was wanting to land the metrics stuff10:59
* bauzas interesting...10:59
johnthetubaguyyeah, I would just refresh the lot10:59
*** EinstCrazy has quit IRC10:59
openstackgerritAlex Xu proposed openstack/nova: docs: update old stuff in version section  https://review.openstack.org/25468710:59
bauzasunless I'm blind, I no longer see a recheck way for the check pipeline10:59
*** lyanchih has quit IRC11:01
bauzasnvm, I *am* blind https://github.com/openstack-infra/project-config/blob/master/zuul/layout.yaml#L2011:01
*** openstackgerrit has quit IRC11:02
bauzasndipanov: so Zuul was probably ill this morning11:02
*** openstackgerrit has joined #openstack-nova11:03
bauzasndipanov: I just did a recheck and it works11:03
ndipanovI don't yaml bauzas11:03
*** lucas-dinner is now known as lucasagomes11:04
bauzasit just says that you can trigger again a new jenkins call if a comment is left using the above regex :)11:04
bauzasfor the check pipeline11:04
bauzasjohnthetubaguy: and if you see the yaml file, you'll see that reverify triggers the check pipeline, and no longer the gate pipeline11:05
*** salv-orlando has joined #openstack-nova11:06
*** john5223 is now known as zz_john522311:06
*** haomaiwa_ has quit IRC11:06
*** mgoddard has joined #openstack-nova11:07
*** thorst has joined #openstack-nova11:08
*** aysyd has joined #openstack-nova11:09
*** ljxiash has joined #openstack-nova11:11
*** paul-carlton2 has quit IRC11:11
*** tbh has joined #openstack-nova11:12
*** sahid has quit IRC11:13
*** thorst has quit IRC11:13
*** dguitarbite has quit IRC11:13
*** Kennan has quit IRC11:14
*** ljianbj has quit IRC11:14
*** ljxiash has quit IRC11:14
*** salv-orlando has quit IRC11:14
*** ljianbj has joined #openstack-nova11:14
*** Jeffrey4l has quit IRC11:15
*** Kennan has joined #openstack-nova11:17
*** sahid has joined #openstack-nova11:21
openstackgerritDina Belova proposed openstack/python-novaclient: Add profiling support to novaclient  https://review.openstack.org/25469911:23
tbhHi, I am new to nova, I am following https://wiki.openstack.org/wiki/Nova/Mentoring, I am interested in https://bugs.launchpad.net/nova/+bug/135775111:23
openstackLaunchpad bug 1357751 in OpenStack Compute (nova) "nova.tests.network.test_manager.AllocateTestCase should use mock" [Low,Confirmed] - Assigned to Aniruddha Singh Gautam (aniruddha-gautam)11:23
tbhas there is no activity for that bug, can I assign that bug to myself11:23
tbhor do you have any suggestions for the beginners?11:24
*** EinstCrazy has joined #openstack-nova11:24
*** paul-carlton2 has joined #openstack-nova11:25
*** dguitarbite has joined #openstack-nova11:26
*** sahid has quit IRC11:26
*** gszasz has joined #openstack-nova11:27
*** jlanoux has quit IRC11:27
*** jlanoux has joined #openstack-nova11:27
*** smatzek has joined #openstack-nova11:28
openstackgerritDina Belova proposed openstack/python-novaclient: Add profiling support to novaclient  https://review.openstack.org/25469911:28
*** Kennan has quit IRC11:29
*** ljianbj has quit IRC11:29
*** ptm_away is now known as PaulMurray11:29
*** ljianbj has joined #openstack-nova11:29
sfinucantbh: I don't know what the "official" stance is, but I normally count a three month period to mean you can take it11:30
sfinucanTwo months is still long enough however as the bug doesn't look like a significant one. I'd say you can start working on it (Ask forgiveness, not permission etc.)11:31
*** ociuhandu has quit IRC11:31
rgerganovalaski, mikal: could you please review https://review.openstack.org/#/c/228778/ ? it already has +2 and I believe it's straightforward11:31
rgerganovjohnthetubaguy, thanks a lot for the review11:32
*** Kennan has joined #openstack-nova11:32
tbhsfinucan, I have already sent the mail to the assignee, but no reply yet, so asking in IRC for guidelines11:33
openstackgerritAlex Xu proposed openstack/nova: docs: update old stuff in version section  https://review.openstack.org/25468711:34
sfinucantbh: Yes, so you have. IMO it's been long enough: work away and just apologise if necessary (which is v. unlikely)11:34
tbhsfinucan, sure, I will mention in the bug comments, thanks for the suggestion.11:35
sfinucantbh: No problem. Best of luck11:36
*** nagyz has joined #openstack-nova11:36
nagyzdanpb, are you around?11:36
danpbnagyz: hi11:37
*** ljxiash has joined #openstack-nova11:39
*** Kennan has quit IRC11:39
*** ljianbj has quit IRC11:39
*** ljxiash has quit IRC11:39
*** ljianbj has joined #openstack-nova11:40
openstackgerritDina Belova proposed openstack/nova: Integrate OSprofiler and Nova  https://review.openstack.org/25470311:40
*** jlanoux has quit IRC11:40
*** tbh has quit IRC11:40
*** jinxing has joined #openstack-nova11:40
*** jlanoux has joined #openstack-nova11:41
sfinucanalex_xu: Question on the doc sprint?11:41
alex_xusfinucan: yup11:41
sfinucanalex_xu: The etherpad refers to 'https://github.com/openstack/nova/tree/master/doc/source/v2' to find TODOs11:42
*** sahid has joined #openstack-nova11:42
sfinucanI was expecting a TODO.rst file or similar, but the link is dead11:42
sfinucanStupid question, but do I just grep the docs folder for 'TODO'? :)11:42
*** Kennan has joined #openstack-nova11:42
openstackgerritjichenjc proposed openstack/python-novaclient: Allow command line for virtual-interface-list  https://review.openstack.org/25470711:43
*** ljianbj has quit IRC11:43
*** Kennan has quit IRC11:43
*** moorryan has quit IRC11:44
*** abhishekk has quit IRC11:44
alex_xusfinucan: oops, the api concept doc was moved before, let me correct the link, thanks for point it out11:44
alex_xusfinucan: there is the right link https://github.com/openstack/nova/tree/master/api-guide/source11:44
openstackgerritDina Belova proposed openstack/python-novaclient: Add profiling support to novaclient  https://review.openstack.org/25469911:44
openstackgerritDina Belova proposed openstack/nova: Integrate OSprofiler and Nova  https://review.openstack.org/25470311:44
garykjohnthetubaguy: bauzas: any chance that you guys can please look at https://review.openstack.org/254206. this breaks stable/liberty11:45
sfinucanalex_xu: OK, thanks. Do I just grep for TODOs and make sure they're not already in progress?11:45
*** mikal has quit IRC11:46
sdagueBobBall: gah, in python 2.4 urlparse is not a named tuple it's just a tuple11:46
sfinucanAgain, sorry if it's a stupid question heh11:46
*** ljianbj has joined #openstack-nova11:46
alex_xusfinucan: yes, you can, but you should check https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/complete-todo-in-api-concept-doc,n,z also, maybe there is patch for some TODO already11:46
*** Kennan has joined #openstack-nova11:46
*** yassine__ has quit IRC11:46
*** mikal has joined #openstack-nova11:47
alex_xusfinucan: there also have patch add more todo, like this https://review.openstack.org/25329211:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce freeze filesystems  https://review.openstack.org/24914011:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce thaw filesystems  https://review.openstack.org/24914111:47
openstackgerritsahid proposed openstack/nova: libvirt: replace direct libvirt's call AbortJobBlock  https://review.openstack.org/24913811:47
openstackgerritsahid proposed openstack/nova: libvirt: move find_disk to Guest's object  https://review.openstack.org/24913911:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce "snapshot" to Guest's object  https://review.openstack.org/24935311:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce block_stats to return device stats  https://review.openstack.org/25128811:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce "pause" to Guest's object  https://review.openstack.org/24983411:47
openstackgerritsahid proposed openstack/nova: libvirt: wrapper list_guests to Host's object  https://review.openstack.org/24983511:47
*** ljianbj has quit IRC11:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce "shutdown" to Guest's object  https://review.openstack.org/24983311:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce get_job_info to Guest's object  https://review.openstack.org/25074011:47
openstackgerritsahid proposed openstack/nova: libvirt: configure migrate max downtime  https://review.openstack.org/25073911:47
*** Kennan has quit IRC11:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce "abort_job" to Guest's object  https://review.openstack.org/25073811:47
openstackgerritsahid proposed openstack/nova: libvirt: introduce "migrate" to Guest's object  https://review.openstack.org/25036511:47
openstackgerritsahid proposed openstack/nova: libvirt: check graphic/serial consoles at pre_live_migration step  https://review.openstack.org/25470911:47
*** yasu has quit IRC11:47
sfinucanalex_xu: OK, I'll see if I can find something I can do. Thanks for the help :)11:47
*** ljianbj has joined #openstack-nova11:47
alex_xusfinucan: and free to update any part of the doc if you found the info is old, or something missed11:49
sfinucanalex_xu: Yup, I have a few ideas11:50
*** Kennan has joined #openstack-nova11:50
bauzasgaryk: sorry I missed your bugfix11:51
bauzasI'll have to take some PTO this afternoon but I'll be back by the EU evening11:52
alex_xusfinucan: thanks for idea and help :)11:52
openstackgerritPaul Murray proposed openstack/nova: Drop JSON decoding for supported_instances  https://review.openstack.org/25423111:56
*** tbh has joined #openstack-nova11:56
openstackgerritSean Dague proposed openstack/nova: retool xen glance plugin to work with urls  https://review.openstack.org/25433111:56
garykbauzas: np11:57
*** pratikmallya has joined #openstack-nova11:57
* alex_xu reminders api meeting in 2mins11:58
*** houming has joined #openstack-nova11:59
ankit_agHi Cores: can someone please help reviewing https://review.openstack.org/#/c/185549/12:02
*** e0ne has quit IRC12:03
BobBall'fraid so sdague :)12:04
*** dedery_ has joined #openstack-nova12:04
openstackgerritSudipta Biswas proposed openstack/nova: libvirt: Add ppc to supported arch for NUMA  https://review.openstack.org/24576512:05
sdagueBobBall: ok, well that new version hopefully works12:05
sdagueBobBall: it would be really nice if there was a way to test that stuff without having to wait for xenci full runs12:06
*** jerrygb has joined #openstack-nova12:09
BobBallI know.12:10
BobBalljohnthetubaguy has some WIP patches for unit testing the plugins12:10
openstackgerritStephen Finucane proposed openstack/nova: config options: Centralise PCI options  https://review.openstack.org/25373212:10
openstackgerritStephen Finucane proposed openstack/nova: Improve 'pci_passthrough_whitelist' documentation  https://review.openstack.org/24333512:10
*** e0ne has joined #openstack-nova12:11
*** thorst has joined #openstack-nova12:12
*** Kevin_Zheng has joined #openstack-nova12:13
openstackgerritClaudiu Belu proposed openstack/nova: Adds HostCapabilities object model and updates ComputeNode object  https://review.openstack.org/22828012:13
openstackgerritClaudiu Belu proposed openstack/nova: Adds host_capabilities conf options group  https://review.openstack.org/25016612:13
openstackgerritClaudiu Belu proposed openstack/nova: driver: Adds get_hypervisor_capabilities method  https://review.openstack.org/25412412:13
openstackgerritClaudiu Belu proposed openstack/nova: WIP: Hyper-V: Adds get_hypervisor_capabilities implementation  https://review.openstack.org/25412512:13
*** jerrygb has quit IRC12:13
nagyzdanpb, regarding the qemu patches you're planning - is there anything we can help with? is the code available somewhere in git or accessible form?12:14
*** fawadkhaliq has quit IRC12:14
openstackgerritSudipta Biswas proposed openstack/nova: libvirt: Add ppc to supported arch for NUMA  https://review.openstack.org/24576512:17
kashyapnagyz: This is probably what is you're looking for - https://lists.gnu.org/archive/html/qemu-devel/2015-11/msg04748.html12:17
nagyzkashyap, ah thanks12:17
kashyapnagyz: I have setup locally a tree to test it, been meaning to diligently test it, but got distracted.12:19
*** thorst has quit IRC12:19
*** Kennan has quit IRC12:20
*** tbh has quit IRC12:21
*** tbh has joined #openstack-nova12:23
paul-carlton2is Microsoft Hyper-V CI check voting for nova?12:24
*** Kennan has joined #openstack-nova12:27
markus_zpaul-carlton2: According to https://wiki.openstack.org/wiki/ThirdPartySystems/Hyper-V_CI it is not. But that could be out-dated.12:28
markus_zpaul-carlton2: but there is a yaml file which configures that, let me look for a sec12:28
*** otter768 has joined #openstack-nova12:31
*** ankit_ag has quit IRC12:33
markus_zpaul-carlton2: I thought it's hidden in here https://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml but I don't find the specific bit, sorry.12:33
paul-carlton2thanks, no worries12:34
*** EinstCra_ has joined #openstack-nova12:35
*** sudipto has quit IRC12:35
*** links has quit IRC12:36
*** otter768 has quit IRC12:36
*** shuquan has joined #openstack-nova12:36
*** fawadkhaliq has joined #openstack-nova12:37
*** snikitin has quit IRC12:37
*** EinstCrazy has quit IRC12:37
ndipanovmoshele, lbelivea hey I might be a bit late for the meeting today12:38
ndipanovfeel free to start without me12:39
ndipanovor we can start at 20 past12:39
lbeliveandipanov: np12:39
lbeliveandipanov: better start with all of us, will wait12:40
*** fawadkhaliq has quit IRC12:40
*** jaosorior has quit IRC12:40
*** fawadkhaliq has joined #openstack-nova12:40
*** shuquan has quit IRC12:41
*** jaosorior has joined #openstack-nova12:42
*** thorst has joined #openstack-nova12:44
*** Kennan has quit IRC12:48
*** yassine__ has joined #openstack-nova12:49
*** mgoddard has quit IRC12:52
openstackgerritKen'ichi Ohmichi proposed openstack/nova: Fix the endpoint of /v2 on concept doc  https://review.openstack.org/25447312:54
*** Kennan has joined #openstack-nova12:54
*** hparekh2 has joined #openstack-nova12:56
moshelendipanov: ok12:57
esracelikmarkus_z: markus, i want to start working on hyperv options, is it ok?12:57
markus_zesracelik: Yeah, sure, go ahead!12:58
*** ducttape_ has joined #openstack-nova12:58
markus_zesracelik: Any further open questions to this point?12:59
*** yangyapeng has joined #openstack-nova12:59
esracelikmarkus_z: so i need a base repository, do you checkout nova in another directory or open a branch in the current?12:59
markus_zesracelik: I open a new branch in the current repository. That's the best way imo.12:59
markus_zesracelik: Git makes it super easy to switch among those branches.13:00
*** gmann_ has joined #openstack-nova13:00
esracelikmarkus_z: but with git checkout -b newbranch I get all previous changes in the new branch13:00
sdaguealex_xu / gmann ok api_samples discussion time13:00
gmann_sdague: yup13:00
alex_xusdague: yup13:00
markus_zesracelik: You have to specify that the base is remote/master branch13:01
alex_xusdague: so that will add 2.xx for all the apis?13:01
sdagueok, so first up lets talk about doc samples location13:01
*** ducttape_ has quit IRC13:02
sdaguewe're like to have a set of doc samples which don't have project_id in them so that we can test with the existing api_samples mechanism13:02
gmann_sdague: so we will maintain both with and without project-id sample?13:03
sdagueand not change all the existing docs, because it should still work with project id13:03
danpbnagyz: most recent posting of patches was https://lists.gnu.org/archive/html/qemu-devel/2015-11/msg04748.html13:03
sdaguegmann_: right13:03
sdagueso my thought is we're adding a microversion for this13:03
sdaguewe can build a whole new doc set at that microversion13:03
markus_zesracelik: something like this:  git checkout -b bp/centralize-config-options remotes/origin/master13:03
gmann_sdague: humm, i thought we will remove that from latest samples so that people can start using without project-id13:03
danpbnagyz: i keep my repo here https://github.com/berrange/qemu/tree/qcrypto-secrets-luks-113:04
sdaguegmann_: well, they are going to need a cloud that's at least got v2.13 for that to work13:04
gmann_sdague: oh, with microversion. then yes13:04
danpbnagyz: what is most beneficial would be getting review comments on qemu patches13:04
sdagueso from a documentation perspective, it doesn't feel like we can just delete it13:04
gmann_sdague: yup13:04
sdagueany cloud with v2.13+ will support it on any API revision13:05
sdaguebecause of where it is on the processing13:05
danpband after i post a newer version, probably need some kind of functional testing to ensure that it is fully interoperable with dm-crypt13:05
markus_zesracelik: although, maybe it makes sense to name the local branch like "bp/centralize-config-options_hyperv" or similar13:05
sdaguehowever, that's the signaling to know it's ok13:05
gmann_sdague: alex_xu so current structure for version sample is resource/vx_y/13:05
*** mwagner_lap has quit IRC13:05
*** ducttape_ has joined #openstack-nova13:06
alex_xusdague: do you mean you want to keep two sets of api samples, one for with project id, one for without project id?13:06
sdaguealex_xu: yes13:06
esracelikmarkus_z: i tried git checkout -b hypervopt remotes/origin/master but that branch but i can still see my previous configdrive.py in nova/conf13:06
sdaguealex_xu: otherwise you can't test both13:06
sdaguethe samples are the verification13:06
gmann_sdague: i feel instead of having api_sample/resource/v2_13 we have all under api_sample/v2_13 ?13:06
sdaguegmann_: I would be ok if we flipped the version like that13:06
sdagueit would make some things easier13:07
gmann_sdague: alex_xu yea for all before 2.13 we need with project id also13:07
alex_xusdague: we support regax one some field validation I remember, we can test both case I guess13:07
sdaguealex_xu: we can with 1 set of templates13:07
sdaguebut we still need 2 sets of doc samples13:07
markus_zesracelik: Hm, that's odd, let me think for a second13:08
gmann_sdague: yea flipped looks nice idea13:08
sdaguealso, from a documentation to the user perspective, having the second set of examples makes a lot of sense13:08
gmann_sdague: initially that was one of the option for microversion sample structure13:08
*** mwagner_lap has joined #openstack-nova13:08
sdaguegmann_: yeh, I think until this point, either was fine, but getting a whole additional set of samples makes it clear it would be better with the version first in our directory13:09
alex_xuok, I didn't get other idea except have two set docs13:09
sdaguegmann_: is that a patch you could work up13:09
gmann_sdague: sure13:09
sdagueok, second question13:09
esracelikmarkus_z: oww sorry sorry that was nova/conf/configdrive.pyc, everything seems good13:10
sdaguehttps://github.com/openstack/nova/blob/7c02bde804434dfd769fdaf7fb6d1ccbcb946d66/nova/tests/functional/api_sample_tests/api_sample_base.py#L59 - when can _test not be defined?13:10
*** tonytan4ever has joined #openstack-nova13:10
sdaguebecause I didn't think that was really a thing13:10
gmann_sdague: alex_xu and that gives more clear idea for what all changed in that version from top folder itself13:10
sdaguegmann_: yep13:10
markus_zesracelik: Ah, ok :) In such cases you could also do a "git reset --hard" <= resets any changes13:10
esracelikmarkus_z: ok, thanks a lot13:11
gmann_sdague: that is for microversion tests class13:11
sdaguegmann_: can you explain more?13:11
gmann_sdague: https://github.com/openstack/nova/blob/master/nova/tests/functional/api_sample_tests/test_keypairs.py#L11813:11
* gmann_ checking detail13:13
PaulMurrayndipanov, do you have anything to say about https://review.openstack.org/#/c/215483/3 - looks like it will come up in the live migration meeting later13:13
alex_xuit is for distinguish different behaviour for different scenarios in baseclass13:13
ndipanovPaulMurray, ah well13:14
*** tonytan4ever has quit IRC13:14
sdagueah, ok, so the scenario list is overwritten in some subclasses13:14
sdaguegmann_: how many more hours are you online today?13:15
gmann_sdague: yea for all microversion tests13:15
alex_xubut sounds like for v2.1 we have two value13:15
*** amotoki has joined #openstack-nova13:15
sdaguebecause I'd like to refactor this a bit for clarity13:15
gmann_sdague: :) may be 20 more min13:15
sdagueok, so I'll have a patch for you tomorrow to look at13:15
gmann_sdague: cool13:16
sdaguewhen you wake up, because I think we should do the testscenarios a bit differently13:16
ndipanovPaulMurray, the thing is - I don't think the fix that was proposed is a good way to go13:16
*** ducttape_ has quit IRC13:16
sdagueinstead of setting some constant like _test13:16
ndipanovfor the related security fix13:16
openstackgerritClaudiu Belu proposed openstack/nova: Adds HostCapabilities object model and updates ComputeNode object  https://review.openstack.org/22828013:16
openstackgerritClaudiu Belu proposed openstack/nova: Adds host_capabilities conf options group  https://review.openstack.org/25016613:16
openstackgerritClaudiu Belu proposed openstack/nova: driver: Adds get_hypervisor_capabilities method  https://review.openstack.org/25412413:16
openstackgerritClaudiu Belu proposed openstack/nova: Adds host_capabilities to HostState  https://review.openstack.org/25474713:16
*** dims has joined #openstack-nova13:16
sdaguewe should be setting the things the test cares about13:16
ndipanovso I hate to see even more code that "legitimizes" it13:16
sdaguelike _api_version13:16
* alex_xu probably learned something from sdague again13:17
sdaguebasically call out all the attributes we are changing per test13:17
openstackgerritJohn Garbutt proposed openstack/nova: docs: add test strategy and feature classification  https://review.openstack.org/21566413:17
gmann_sdague: yea that better idea.13:17
sdagueinstead of going through a global constant and having our own if block13:17
sdaguegmann_: ok, cool13:17
sdagueI'll put up a patch during my day, and will be interested to see what you think13:17
PaulMurrayndipanov, at the moment you and I hae -1 on it, but you have a better understanding than I do - I just wasn't sure about the use of error vs fail.13:18
sdaguegmann_: thanks much for the info13:18
gmann_sdague: sure, so you will refactor for _test right13:18
sdaguegmann_: yes13:18
PaulMurrayso I will probably just direct him to you13:18
gmann_sdague: and for me refactoring the sample structure with version at top folder13:18
ndipanovI will look into that one more closely too13:18
sdaguegmann_: yes13:18
gmann_sdague: cool13:18
sdaguegmann_: great, thanks much13:19
gmann_sdague: alex_xu Thanks.13:19
gmann_sdague: alex_xu hanging up if nothing else?13:19
alex_xugmann_: yeh, see you tomorrow13:20
gmann_alex_xu: see you. i will put that up in morning so that you can look early13:21
alex_xugmann_: yeh, I will help for review, thanks for the work13:21
*** rpodolyaka has quit IRC13:21
*** baoli has joined #openstack-nova13:22
*** tlbr has quit IRC13:22
*** ianbrown__ has quit IRC13:22
*** ianbrown_ has joined #openstack-nova13:22
gmann_alex_xu: thanks,13:23
*** gmann_ has quit IRC13:23
openstackgerritJohn Garbutt proposed openstack/nova: api-guide: add a doc on users  https://review.openstack.org/24981213:24
*** baoli_ has joined #openstack-nova13:25
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Fixes disk overhead claim issue  https://review.openstack.org/24630713:26
* johnthetubaguy goes to get some food13:26
yalieping alex_xu13:26
*** baoli has quit IRC13:28
yalieping johnthetubaguy13:28
*** ljxiash has joined #openstack-nova13:28
sdagueBobBall: so, out of curiosity, is there any plan for xenserver to update to not ancient base os any time soon?13:29
BobBallYes - tech preview currently out is on CentOS 7.113:30
*** baoli_ has quit IRC13:31
*** edmondsw has joined #openstack-nova13:31
*** ljxiash_ has joined #openstack-nova13:32
*** baoli has joined #openstack-nova13:32
*** ljxiash has quit IRC13:34
openstackgerritAdelina Tuvenie proposed openstack/nova: Test patch for nova CI  https://review.openstack.org/25475613:35
*** jaypipes has joined #openstack-nova13:35
openstackgerritIldiko Vancsa proposed openstack/nova: virt: Allow multiple volume attachments  https://review.openstack.org/19313513:36
openstackgerritIldiko Vancsa proposed openstack/nova: comp: Allow multiple volume attachments  https://review.openstack.org/19313413:36
openstackgerritIldiko Vancsa proposed openstack/nova: WIP: REST API changes to support multi-attach  https://review.openstack.org/15303813:36
*** chinmaya has quit IRC13:36
*** baoli has quit IRC13:37
*** baoli has joined #openstack-nova13:37
sdagueBobBall: ok, cool13:38
*** mgoddard has joined #openstack-nova13:38
BobBallIt's been a long time coming, but it's progress.  At this rate, I expect XenServer to use Python 3 by 2027.13:38
openstackgerritAndrew Bogott proposed openstack/nova: Prevent scheduling of instances on full hosts.  https://review.openstack.org/24225113:39
*** baoli has quit IRC13:42
lxsliPlease could I get reviews of https://review.openstack.org/#/c/215613/ ?13:42
*** terryyao has joined #openstack-nova13:42
openstackgerritStephen Finucane proposed openstack/nova: Improve 'pci' conf options documentation  https://review.openstack.org/24333513:42
sdagueBobBall: is xenci running the last patch?13:42
BobBall 1 sec - will double check13:42
BobBall| 18535 | openstack/nova         | refs/changes/31/254331/3  | Collecting |   |        | 0.05        | 0.91     |13:43
BobBallExpect the vote in a few minutes ;)13:43
*** tbh has quit IRC13:44
*** jerrygb has joined #openstack-nova13:44
*** eliqiao has quit IRC13:45
sdagueBobBall: ok, follow on question, let me push a broken patch and get your opinion on it13:45
BobBalldon't push it to that changeset yet13:45
openstackgerritSean Dague proposed openstack/nova: WIP: convert to download/upload vhd2 which uses url  https://review.openstack.org/25476013:46
*** eliqiao has joined #openstack-nova13:46
sdagueno, it's above it13:46
BobBallit'll abort the current run just as it's finishing13:46
*** ociuhandu has joined #openstack-nova13:46
BobBallah right - cool13:46
sdagueoh, I guess I mixed the last patch a little with the vhd2 functions13:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce freeze filesystems  https://review.openstack.org/24914013:46
openstackgerritsahid proposed openstack/nova: libvirt: check graphic/serial consoles at pre_live_migration step  https://review.openstack.org/25470913:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce thaw filesystems  https://review.openstack.org/24914113:46
*** eliqiao has quit IRC13:46
openstackgerritsahid proposed openstack/nova: libvirt: retrive root block device  https://review.openstack.org/24913913:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce "snapshot" to Guest's object  https://review.openstack.org/24935313:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce block_stats to return device stats  https://review.openstack.org/25128813:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce "pause" to Guest's object  https://review.openstack.org/24983413:46
openstackgerritsahid proposed openstack/nova: libvirt: wrapper list_guests to Host's object  https://review.openstack.org/24983513:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce "shutdown" to Guest's object  https://review.openstack.org/24983313:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce get_job_info to Guest's object  https://review.openstack.org/25074013:46
openstackgerritsahid proposed openstack/nova: libvirt: configure migrate max downtime  https://review.openstack.org/25073913:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce "abort_job" to Guest's object  https://review.openstack.org/25073813:46
openstackgerritsahid proposed openstack/nova: libvirt: introduce "migrate" to Guest's object  https://review.openstack.org/25036513:46
BobBallDid you forget to add the plugin?13:46
*** eliqiao has joined #openstack-nova13:46
*** ljxiash_ has quit IRC13:47
BobBalloh my bad13:47
BobBallthat was in the original change - sorry13:47
sdagueright, so this is the actually cut over to the new function, how do we want to handled that?13:48
BobBallJust like that's fine.13:48
BobBallBut we need to update the plugin version numbers in the two changes13:48
openstackgerritdstepanenko proposed openstack/nova: WIP: This is 3rd part of changes according to pci-generate-stats  https://review.openstack.org/24769213:48
sdaguewhere is the plugin version number indicated?13:48
sdagueI guess I haven't stumbled over that yet13:48
*** pratikmallya has quit IRC13:48
BobBallSo the first change needs to update https://git.openstack.org/cgit/openstack/nova/tree/plugins/xenserver/xenapi/etc/xapi.d/plugins/nova_plugin_version#n3113:48
sdaguecan you -1 with that info, and I'll fix it13:49
BobBallAnd the second change needs to update https://git.openstack.org/cgit/openstack/nova/tree/nova/virt/xenapi/client/session.py#n8413:49
BobBallWill do13:49
*** ljxiash has joined #openstack-nova13:49
BobBallXS CI gave it a +113:49
sdagueok, I'm going to work through these unit tests to get the vhd2 bits passing13:50
sdagueso, I guess the only question is with the versioning is it ok to change change the signature on vhd functions?13:50
sdagueor is this vhd2 path the prefered one13:50
openstackgerritdstepanenko proposed openstack/nova: WIP: This is 4th patch of changes according to pci-generate-stats blueprint.  https://review.openstack.org/25075713:51
BobBallCan't change the signature without a major version update (cuz it'd be incompatible).  Really _really_ want to have the function deprecated for at least 1 cycle otherwise ppl deploying it must update compute + plugins in sync and have a fixed restart time of the compute13:51
sdagueyep, that's fine13:52
sdagueso, I'm going to just flip the unit tests to using vhd213:53
* BobBall stands back13:53
sdagueif you want to still test vhd functions, I'll leave to to you to fill back in13:53
BobBallUnit tests don't call through to plugins currently - so you're just testing nova code, which will only be using vhd213:54
*** dave-mccowan has joined #openstack-nova13:56
openstackgerritEd Leafe proposed openstack/nova: Edit the text to be more native-English sounding  https://review.openstack.org/25476613:57
sdagueyeh, so I guess that's sensible13:57
yaliejohnthetubaguy: ping13:57
*** diogogmt has quit IRC13:58
*** diogogmt has joined #openstack-nova13:58
johnthetubaguyyalie: sorry, was having lunch, how can I help?13:59
sdagueBobBall: also, I had mox :)14:00
*** sudipto has joined #openstack-nova14:01
*** mgoddard_ has joined #openstack-nova14:02
yaliejohnthetubaguy: thanks john, I have a bp about removing the nova's verification on port with IP and network with subnet when booting a VM14:02
*** lykinsbd has joined #openstack-nova14:02
*** mgoddard has quit IRC14:02
andrearosanova-live-migration subteam meeting is now14:02
*** wyang has quit IRC14:02
esracelikmarkus_z: markus i have a question. sorry for disturbing frequently14:02
*** derekh has joined #openstack-nova14:02
openstackgerritSudipta Biswas proposed openstack/nova: Remove version hardcoding for 'show' method  https://review.openstack.org/22857814:02
markus_zesracelik: nah, no worries14:02
yaliejohnthetubaguy: https://review.openstack.org/#/c/239276/   I am not sure if it could get a exception of nova bp14:03
*** jinxing has quit IRC14:03
esracelikmarkus_z: can I merge all options in directory nova/conf/hyperv in one file14:04
*** akshai has joined #openstack-nova14:04
yaliejohnthetubaguy: I disscussed with kevin benton in neutron, this feature has been supported in neutron, it would not depend some feature in neutron.14:04
esracelikmarkus_z: i am asking this because if I create nova/conf/vif.py it is not clear if it is nova/virt/hyperv/vif.py or nova/virt/vmwareapi/vif.py14:05
*** ankit has joined #openstack-nova14:05
*** rook-desktop has joined #openstack-nova14:05
johnthetubaguyyalie: we can add that to the list in the etherpad, and see how it goes14:05
*** nikhil_k is now known as nikhil14:05
openstackgerritYingxin Cheng proposed openstack/nova: Use stevedore for scheduler driver  https://review.openstack.org/25476814:05
markus_zesracelik: let me check the code, one second14:06
yaliejohnthetubaguy: thanks johnthetubaguy14:06
*** mgoddard has joined #openstack-nova14:06
openstackgerritAndrew Laski proposed openstack/nova: Clean up filter_properties/availability_zone in boot  https://review.openstack.org/25443414:06
*** claudiub has joined #openstack-nova14:06
*** mgoddard_ has quit IRC14:06
*** atuvenie has joined #openstack-nova14:07
*** rlrossit has joined #openstack-nova14:07
markus_zesracelik: I think it's sane to have a "nova/conf/hyperv.py" which contains config opt "vswitch_name" and "nova/conf/vmware" which contains "vlan_interface" and "intergration_bridge".14:08
*** hparekh2 has quit IRC14:08
*** ducttape_ has joined #openstack-nova14:08
markus_zesracelik: Is that what you meant with merging them into one file?14:08
*** ctrath has joined #openstack-nova14:08
*** pratikmallya has joined #openstack-nova14:10
*** changbl has quit IRC14:10
esracelikmarkus_z: so maybe current position of options are not relevant, right?14:10
esracelikmarkus_z: when creating nova/conf/new.py file I should collect all related options together?14:11
markus_zesracelik: The semantic of them has priority over the location.14:11
markus_zesracelik: yes, they should build a group of related options.14:11
esracelikmarkus_z: allright, got it, thanks14:11
markus_zesracelik: finding the group is especially funny in the "DEFAULT" section.14:12
markus_zesracelik: If you think that a current grouping doesn't make sense and should be re-evaluated, add it at the end of etherpad https://etherpad.openstack.org/p/config-options14:13
*** tlbr has joined #openstack-nova14:14
*** ducttape_ has quit IRC14:15
*** rpodolyaka has joined #openstack-nova14:15
*** Daisy has joined #openstack-nova14:16
esracelikmarkus_z: ok, it seem like it won't be that easy :) for example there are two vlan_interface options, one in DEFAULT section one in vmware section.14:16
*** cfriesen has joined #openstack-nova14:18
*** breitz has quit IRC14:18
*** breitz has joined #openstack-nova14:19
markus_zesracelik: yeah, that's one of those things. We have the code, cloud admins don't have it. How should they decide here?14:19
sdagueBobBall: ok, I've got what I think is the stack supporting this about ready to go, just doing one more check on unit tests before pushing14:20
esracelikmarkus_z: yes, I am getting familiar.. thanks14:21
markus_zesracelik: Not sure if you already know it, but we still have the service "nova-network" in use. So some of the options will configure this service.14:21
esracelikmarkus_z: yes I used nova-network. when will it be removed completely?14:22
markus_zesracelik: I was discussed during the last summit, but I don't remember the conclusion, sorry.14:23
esracelikmarkus_z: ok, removal approach would bu fun as well :)14:23
*** ijuwang has joined #openstack-nova14:24
markus_zesracelik: I guess people much smarter than me have to do that :)14:24
*** diogogmt has quit IRC14:25
esracelikmarkus_z: oh, dont think like that. we have a sentence in turkish for this case, but i don't know how to say it in english , anyway:)14:26
*** mdrabe has joined #openstack-nova14:27
openstackgerritdstepanenko proposed openstack/nova: WIP: This is 4th patch of changes according to pci-generate-stats blueprint.  https://review.openstack.org/25075714:27
*** mriedem has joined #openstack-nova14:28
*** dane-fichter has joined #openstack-nova14:28
*** yalie has quit IRC14:28
openstackgerritStephen Finucane proposed openstack/nova: trivial: Add additional logs for NUMA scheduling  https://review.openstack.org/25478314:30
*** zhangjn has joined #openstack-nova14:32
*** zhangjn has quit IRC14:32
*** otter768 has joined #openstack-nova14:32
*** busterswt has joined #openstack-nova14:32
*** zhangjn has joined #openstack-nova14:32
openstackgerritSean Dague proposed openstack/nova: Convert xenapi driver to using up/down vhd2  https://review.openstack.org/25476014:33
openstackgerritSean Dague proposed openstack/nova: retool xen glance plugin to work with urls  https://review.openstack.org/25433114:33
openstackgerritSean Dague proposed openstack/nova: Implement GlanceEndpoint object  https://review.openstack.org/25421414:33
openstackgerritSean Dague proposed openstack/nova: always create clients with GlanceEndpoint  https://review.openstack.org/25430614:33
openstackgerritSean Dague proposed openstack/nova: Add upload/download vhd2 interfaces  https://review.openstack.org/25478514:33
sdagueBobBall: ok, the last 3 patches in that stack should be reviewed by you14:34
sdagueto make sure I got that right14:34
*** otter768 has quit IRC14:37
*** smatzek has quit IRC14:38
*** eglynn has quit IRC14:38
*** Daisy has quit IRC14:38
*** Daisy has joined #openstack-nova14:39
*** tonytan4ever has joined #openstack-nova14:41
*** lucasagomes is now known as lucas-hungry14:42
*** coreycb has joined #openstack-nova14:45
*** tonytan4ever has quit IRC14:45
*** fawadkhaliq has quit IRC14:45
*** fawadkhaliq has joined #openstack-nova14:46
coreycbwill there be a liberty point release in the near future?14:46
*** ankit has quit IRC14:47
*** edmondsw has quit IRC14:48
*** armax has joined #openstack-nova14:50
*** dims_ has joined #openstack-nova14:52
nagyzdanpb, we'll engage our qemu team and check with them14:52
*** derekh is now known as derekh_afk14:53
nagyzdanpb, would setting up a CI system that tests dm-crypt -> in-qemu migration be helpful?14:53
*** Daisy has quit IRC14:55
*** yalie has joined #openstack-nova14:55
*** dims has quit IRC14:55
*** eharney has joined #openstack-nova14:56
*** mrkz has joined #openstack-nova14:56
sdagueflaper87 / johnthetubaguy how important is it to *not* send keystone headers for glanceclient?14:57
*** Daisy has joined #openstack-nova14:57
*** baoli has joined #openstack-nova14:58
openstackgerritEd Leafe proposed openstack/nova: Edit the text to be more native-English sounding  https://review.openstack.org/25476614:58
*** esracelik has quit IRC14:58
flaper87sdague: mmh, not sure I understand your question (or what it relates to) but I'd assume it refers to the patches you just pushed for review14:58
*** ducttape_ has joined #openstack-nova14:59
sdagueflaper87: https://review.openstack.org/#/c/254306/2/nova/image/glance.py,cm14:59
dansmithThis set on live migration data objectification could use eyes: https://review.openstack.org/#/c/247719/15:00
*** yonglihe has joined #openstack-nova15:00
danpbnagyz: not sure we need a full on CI system, but certainly will want some kind of functional test suite integrated into QEMU to test it15:00
*** mgoddard has quit IRC15:00
danpbQEMU's I/O layer has a fairly extensive test harness to build on15:00
openstackgerritDave McCowan proposed openstack/nova: Check context before returning cached value  https://review.openstack.org/25435815:00
*** zhangjn has quit IRC15:01
nagyzdanpb, I more meant it in an OpenStack setting, for example to live migrate between a host running the current encryptor path and the new one15:01
*** dustins|gone is now known as dustins15:01
nagyzdanpb, as I guess that's something that would need to be supported to replace the current IO path?15:01
*** yangyapeng has quit IRC15:01
danpbnagyz: oh i see what you mean15:02
danpbyeah, that could well be something that's useful now we can do migration testing in CI15:02
*** mgoddard has joined #openstack-nova15:03
*** ducttape_ has quit IRC15:03
*** ducttape_ has joined #openstack-nova15:03
*** Jeffrey4l has joined #openstack-nova15:03
nagyzdanpb, ok, I'll need to talk to our internal folks but sounds like a plan from my side15:04
*** inc0 has quit IRC15:05
*** pratikmallya has quit IRC15:05
*** Jeffrey4l_ has joined #openstack-nova15:07
*** baoli has quit IRC15:08
*** inc0 has joined #openstack-nova15:08
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Fix help message in case of microversions  https://review.openstack.org/25433515:09
*** jerrygb has quit IRC15:09
*** jerrygb has joined #openstack-nova15:10
*** artom has joined #openstack-nova15:10
*** yarkot has joined #openstack-nova15:13
*** Daisy has quit IRC15:13
*** atuvenie has quit IRC15:13
*** zenoway has quit IRC15:13
*** rdopiera has quit IRC15:13
*** edmondsw has joined #openstack-nova15:13
*** paul-carlton_ has joined #openstack-nova15:13
*** smatzek has joined #openstack-nova15:14
mriedemcoreycb: i think so15:14
mriedemthere have been some high priority regression fixes going into stable/liberty in the last week or so15:14
*** Daisy has joined #openstack-nova15:14
mriedemcoreycb: we're trying to round up release notes for a liberty release15:14
mriedemgaryk: had another one that has to go in that too15:14
*** jerrygb has quit IRC15:15
*** jinxing has joined #openstack-nova15:15
*** zenoway has joined #openstack-nova15:15
*** sneti has joined #openstack-nova15:15
garykmriedem: sorry do not follow15:16
coreycbmriedem, ok good to hear, I'll keep an eye out for it15:16
mriedemthat will go back to liberty right?15:16
*** READ10 has joined #openstack-nova15:16
garykmriedem: yes, planning to backport that when the gate recovers15:16
*** fawadkhaliq has quit IRC15:16
mriedemwas there a gate outage?15:17
*** lucas-hungry is now known as lucasagomes15:17
lxsliclaudiub: I removed https://review.openstack.org/#/c/234283/ from the trivial patch list because 1) it's your patch, not supposed to propose your own; 2) it's not that trivial. Hope that's OK15:17
andreykurilinjohnthetubaguy, sdague, mriedem, alex_xu, melwitt: hi! Can you look at update for novaclient's spec(https://review.openstack.org/#/c/211206/ )? Also, it can be moved to implemented...15:18
*** Daisy has quit IRC15:19
*** zenoway has quit IRC15:19
mriedemdansmith: hit me up on https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/objectify-live-migrate-data,n,z when things settle down a bit, i want to get https://review.openstack.org/#/c/254428/ done first15:19
*** Jeffrey4l has quit IRC15:20
andreykurilinmriedem: I know you want to release novaclient this week. When you will do? I want to get merge several patches in this release:)15:20
*** atuvenie has joined #openstack-nova15:20
dansmithmriedem: ... are you asking me to remind you to look at that set, or that you will look after you look at that other patch?15:20
mriedemdansmith: just to get on me about reviewing it15:21
dansmithmriedem: okay15:21
mriedemi was planning on it anyway15:21
mriedembut need the whip15:21
*** _ducttape_ has joined #openstack-nova15:21
mriedemandreykurilin: we're waiting on https://review.openstack.org/#/c/245200/15:21
dansmithmriedem, you yella-belly, no-good, trough-drinkin lowlife... review that set15:22
dansmithmriedem: good? ^^15:22
mriedemthems are right proper fightin words, but not what i had in mind15:22
*** moshele has quit IRC15:22
*** jerrygb has joined #openstack-nova15:24
andreykurilinmriedem: got it. thanks15:24
*** ducttape_ has quit IRC15:24
openstackgerritdstepanenko proposed openstack/nova: Duplicated the logic of nova.pci.stats.PciDevStats.supports_request() method into the nova.objects.PciDevicePoolList object.  https://review.openstack.org/24208715:25
openstackgerritdstepanenko proposed openstack/nova: Duplicated the logic of nova.pci.stats.PciDevStats.supports_request() method into the nova.objects.PciDevicePoolList object.  https://review.openstack.org/24208715:26
*** atuvenie has quit IRC15:26
*** gszasz has quit IRC15:27
flaper87sdague: As you mentioned it in your comment on that patch, I don't think sending them hurts at all. I'd just do what you did there.15:28
* flaper87 <- brb15:28
sdagueflaper87: ok, cool, there is one test later that tests for not sending those headers, but I think it's mostly silly15:30
*** gszasz has joined #openstack-nova15:30
openstackgerritBalazs Gibizer proposed openstack/nova: Add service status notification  https://review.openstack.org/24567815:30
openstackgerritBalazs Gibizer proposed openstack/nova: Add infra for versioned notifications  https://review.openstack.org/24702415:30
openstackgerritBalazs Gibizer proposed openstack/nova: Adds json sample for the versioned notifications  https://review.openstack.org/24816715:30
openstackgerritBalazs Gibizer proposed openstack/nova: Make emitting versioned notifications configurable  https://review.openstack.org/24756415:30
*** claudiub has quit IRC15:32
mriedemthis is confusing,15:32
mriedemstable/liberty reno job, liberty series release notes are blank http://docs-draft.openstack.org/24/253224/4/check/gate-nova-releasenotes/46de2bb//releasenotes/build/html/liberty.html15:32
mriedemthe stable/liberty reno change's notes are in the 'current series' section: http://docs-draft.openstack.org/24/253224/4/check/gate-nova-releasenotes/46de2bb//releasenotes/build/html/unreleased.html15:32
mriedemdhellmann: ^ is that right?15:32
*** mtanino has joined #openstack-nova15:33
*** xyang1 has joined #openstack-nova15:33
*** sneti has quit IRC15:35
*** sneti has joined #openstack-nova15:35
dansmithmarkus_z: can you re-review this? https://review.openstack.org/#/c/243335/815:36
*** diogogmt has joined #openstack-nova15:36
markus_zdansmith: yepp, looking15:36
*** erhudy has joined #openstack-nova15:36
*** thangp has joined #openstack-nova15:37
*** thumpba has joined #openstack-nova15:39
*** terryyao has quit IRC15:40
*** mjura has quit IRC15:41
openstackgerritSean Dague proposed openstack/nova: drop _create_glance_client function  https://review.openstack.org/25482715:41
openstackgerritSean Dague proposed openstack/nova: convert GlanceClientWrapper to endpoint  https://review.openstack.org/25482815:41
*** doron_afk has left #openstack-nova15:41
*** rpodolyaka has quit IRC15:41
*** tlbr has quit IRC15:41
sdaguemriedem: are there fixes up for liberty volumes regressions somewhere?15:43
*** tonytan4ever has joined #openstack-nova15:43
mriedemliberty volumes regressions?15:43
*** jistr is now known as jistr|mtg15:43
*** pratikmallya has joined #openstack-nova15:44
sdaguemriedem: it seems that's the typically grenade multinode fail now15:46
sdaguebugs 2 - 6 are all volumes15:46
*** andrewbogott has quit IRC15:46
mriedemsdague: so something must have changed in tempest if we're getting regressions on all branches all of a sudden15:46
mriedemwhich is what i was getting at least week15:47
sdaguemriedem: well, maybe, maybe not15:47
sdaguewe've now got ovh and bluebox in our node rotation15:47
sdagueand they are both slower than hp or rax15:47
mriedemso thinking slower nodes causing issues?15:48
sdagueyeh, could be15:48
mriedemi also don't know how much i trust at 12/1 date in logstash15:48
sdagueanyway, mostly wondering if there are any cinder people that are looking into any of these?15:48
mriedemnot that i'm aware of15:48
mriedemsdague: this was the thing i spent all of yesterday on https://review.openstack.org/#/c/254428/15:49
mriedemstill need to clean it up and add tests15:49
sdagueah, gotcha15:50
mriedemanyway, that's a clear race and an easy change for backporting, and that ebs test is hitting all branches15:51
mriedemas is whatever the ec2 volume tests are doing15:51
mriedembasically doing boot from volume + quick server delete exposes this race15:51
*** derekh_afk has quit IRC15:51
*** mc_nair has joined #openstack-nova15:52
*** nkrinner has quit IRC15:53
mriedemdansmith: you might want to weigh in here, but i think this is too complicated for a latent perf bug to backport to stable/liberty https://review.openstack.org/#/c/245656/15:53
mriedemplus it has an object version bump in it15:54
dansmithmriedem: oh yeah, definitely not for backport.. I'll comment15:54
*** burt has quit IRC15:57
*** gokrokve has joined #openstack-nova15:59
sdaguehmmm... so the grenade-dsvm-multinode test is > 50% fail right now, largely for volumes fails on liberty16:01
sdagueno every nova change is currently a coin flip16:02
sdaguewhich is why the gate is backed up as bad as it is16:02
dansmithsdague: so I just saw one that was ssh timeout16:02
*** Vek has quit IRC16:02
dansmithsdague: on the old side.. is that how it manifests?16:02
sdaguedansmith: all my spot checking is that it's on the old side16:02
openstackgerritSudipta Biswas proposed openstack/nova: libvirt: Add ppc to supported arch for NUMA  https://review.openstack.org/24576516:03
sdaguehonestly, I haven't looked much yet, because I had this other context stack in my head16:03
mriedemthere was a big ssh change series in tempest16:03
mriedemlast week16:03
*** salv-orlando has joined #openstack-nova16:03
*** salv-orlando has quit IRC16:03
sdaguemriedem: did it land?16:03
*** lpetrut has quit IRC16:03
*** salv-orlando has joined #openstack-nova16:03
sdagueok, I've got another couple of things to look at before I can shift gears here and dig deeper16:04
mriedemi don't see any stable/liberty regressions in nova16:04
*** smurke has joined #openstack-nova16:05
BobBallWill do sdague - ta16:05
*** smurke has joined #openstack-nova16:05
sdagueyeh, I think the difference is that when we're running smoke these 2 volume tests often happen at the same time16:05
*** tlbr has joined #openstack-nova16:05
*** hemnafk is now known as hemna16:06
markus_zsfinucan: I'm reviewing your patch https://review.openstack.org/#/c/243335/8/nova/conf/pci.py,cm right now and would like to ask you a question. Have a minute?16:06
mriedemguess it wouldn't just be ceph since it's also multinode16:06
sdaguemriedem: yeh16:07
*** smurke has quit IRC16:07
*** tjones has joined #openstack-nova16:07
sdagueanyway, if we are lacking people to dig into this, we should consider flipping back to non-voting16:07
*** smurke has joined #openstack-nova16:07
edleafemarkus_z: got a question for you about the config opt changes16:07
sdaguebecause right now it's just getting in the way of anything else non related landing16:07
*** mariusv has joined #openstack-nova16:07
*** mariusv has quit IRC16:07
*** mariusv has joined #openstack-nova16:07
dansmithsdague: if we do that we lose all upgrade gating, right?16:08
*** aix has quit IRC16:08
edleafemarkus_z: looking at https://review.openstack.org/#/c/248963/8/nova/virt/opts.py,cm as an example of the issue16:08
markus_zedleafe: looking16:08
sdaguewe do, but right now people are going to just recheck grind on these fails anyway.16:08
sdaguewell, we lose the partial upgrade -116:09
markus_zedleafe: What's the issue with that?16:09
mriedemsdague: do we have any idea when this started? or only the 12/1?16:09
*** smatzek has quit IRC16:09
dansmithsdague: I kinda don't want to turn that off...16:09
*** gszasz has quit IRC16:09
*** rpodolyaka has joined #openstack-nova16:09
edleafemarkus_z: won't list_opts() return something like (('DEFAULT', [bunch of opts]), ('DEFAULT', [more opts]), ('hyper-v', [hyper opts]), ...)16:09
sdaguethe fail rate is drifting up pretty continuously - http://tinyurl.com/qhdwtuz16:09
edleafemarkus_z: IOW, multiple tuples with 'DEFAULT' as the first element16:10
sdaguedansmith: I hear you, but 50% failure rate in the gate is pretty much untenable16:10
mriedemare these all ssh failures?16:10
sdagueunless you are ready to -2 all nova patches until this is solved16:10
mriedemor other timeouts?16:10
edleafemarkus_z: oops, that wasn't the one16:10
dansmithsdague: sure, but we wouldn't turn off the main devstack test..16:10
edleafemarkus_z: since 'guestfs' is its own section16:10
*** tlbr has quit IRC16:10
sdaguemriedem: like I said, this is all the info I've got. And I can't look into it until later today16:10
*** stackdump has joined #openstack-nova16:11
dansmithsdague: mriedem: maybe we have a new deadlock we need to retry?16:11
mriedemsdague: ok, i can poke on it after cleaning up my fix for one of those volume things (the rbd one)16:11
dansmithhaven't we been merging some oslo db stuff that might be causing that?16:11
edleafemarkus_z: but others have list_opts() return a tuple with 'DEFAULT' - I can dig one up16:11
mriedemdansmith: that was in mitaka16:11
mriedembut oslo.db releases....16:11
*** ccarmack has joined #openstack-nova16:11
markus_zedleafe: I guess I know what you mean, let me look it up16:11
*** vilobhmm11 has joined #openstack-nova16:12
edleafemarkus_z: foound it: https://review.openstack.org/#/c/252687/6/nova/opts.py,cm16:12
edleafethe 'wsgi' opts are all 'DEFAULT'16:12
mriedemdansmith: looks like that old side devstack installed oslo.db 3.1.0 http://logs.openstack.org/32/253732/4/check/gate-grenade-dsvm-multinode/9f726ed/logs/old/devstacklog.txt.gz#_2015-12-08_14_41_32_14916:12
*** fawadkhaliq has joined #openstack-nova16:12
mriedemwhich was released on 10/1916:12
mriedemlet's check logstash for that db trace though16:12
dansmithit's also possible that we're not retrying something that we're suddenly hitting or something16:13
dansmiththis is a pymysql exception too16:13
markus_zedleafe: yeah, that's the one I wasn't sure but didn't yet double-check. Let me check it16:13
*** tlbr has joined #openstack-nova16:13
dansmithmriedem: like, if we're not catching the pymysql version of this exception16:13
mriedemdansmith: hmm, yeah, maybe due to slower nodes16:13
mriedemcausing the deadlock hit16:14
mriedemoslo.db is converting that for us16:14
dansmithI don't see version tags in the oslo.db repo16:14
markus_zedleafe: The result in the docs job makes me believe the oslo's sample config generator can handle this: http://docs-draft.openstack.org/87/252687/6/check/gate-nova-docs/13acb07//doc/build/html/sample_config.html16:15
*** changbl has joined #openstack-nova16:15
markus_zedleafe: checking oslo's sample config generator though16:15
dansmithoh, nevermind I'm dumb16:15
dansmithmriedem: so we're getting an actual pymysql exception not an oslo.db one, which looks like it wouldn't be retried by that wrapper16:16
dansmithdims_: who do we ask about oslo.db stuff?16:16
*** browne has joined #openstack-nova16:16
mriedemrpodolyaka: ^16:17
*** ianbrown_ has quit IRC16:17
mriedemthat db deadlock actually shows up a lot http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22DBDeadlock:%20(pymysql.err.InternalError)%20(1213,%20u'Deadlock%20found%20when%20trying%20to%20get%20lock;%20try%20restarting%20transaction')%20%5BSQL:%20u'SELECT%20reservations.created_at%5C%22%20AND%20tags:%5C%22screen-n-api.txt%5C%22%20AND%20voting:%5C%221%5C%2216:18
edleafemarkus_z: ah, so even if there are duplicate names, the generator chains them together?16:18
rpodolyakamriedem: hmm, interesting16:18
mriedem154 hits in 7 days, but they are mostly in successful runs16:18
*** stackdump has quit IRC16:18
edleafemarkus_z: still, it seems shaky16:18
mriedemso we must be retrying on deadlock16:18
markus_zedleafe: looks so, yes. But I have to read the generator code to be sure.16:18
*** pradk has joined #openstack-nova16:18
markus_zedleafe: What's the shaky part here?16:19
dansmithmriedem: but we don't raise/log unless we give up or get something unexpected right?16:19
*** pratikmallya has quit IRC16:19
openstackgerritRyan Moore proposed openstack/nova: Add better help to rdp options  https://review.openstack.org/25355616:19
dansmither, no we do16:19
mriedemdansmith: no, oslo.db logs every one it hits i think16:19
mriedemthe neutron server logs were always full of these16:19
mriedembut in mostly successful runs b/c of the retry16:19
mriedemit'd be nice if oslo.db would only trace on the last failure before giving up16:20
edleafemarkus_z: the value returned by nova/opts.py's list_opts()16:20
edleafemarkus_z: with the double 'DEFAULT' tuples.16:20
edleafemarkus_z: it fells shaky to generate a bad structure knowing that something else will clean it up16:20
*** gszasz has joined #openstack-nova16:21
edleafemarkus_z: I'd rather it be correct to start with16:21
mriedemi looked into this once, i think it's actually something lower in the stack that is logging it16:21
edleafemarkus_z: in case the generator ever changes16:21
mriedemin oslo.db16:21
mriedemit gets logged here though https://github.com/openstack/oslo.db/blob/3.1.0/oslo_db/api.py#L16916:21
dansmithso I think the instance is started16:21
dansmithand tempest gets back "instance not found"16:21
dansmithstarted: http://logs.openstack.org/32/253732/4/check/gate-grenade-dsvm-multinode/9f726ed/logs/grenade.sh.txt.gz#_2015-12-08_15_11_33_63016:22
edleafemarkus_z: since the nova/conf/wsgi.py's list_opts is only called from nova/opts.py, why not have it just return the list, and put that in the itertools.chain?16:22
*** pumaranikar has joined #openstack-nova16:22
edleafemarkus_z: or even skip the list_opts(), and just add nova.conf.wsgi.ALL_OPTS to the itertools.chain16:22
sfinucanmarkus_z: Sure. What's up?16:23
*** baoli has joined #openstack-nova16:23
*** mriedem is now known as rodermang16:23
*** rodermang is now known as mriedem16:23
dansmithI guess the not found is bogus fail after the ssh fails..16:24
markus_zedleafe: Yepp, I've seen how you did it for the scheduler opts. My initial intention was to keep the knowledge of the group name in one place. The place where the config options got declared.16:24
mriedemthe not found is on cleanup16:24
*** gszasz has quit IRC16:24
dansmithmriedem: yeah, that's confusing16:24
mriedemdansmith: i don't think this is one of the volume failures16:24
mriedemthe build error is:16:25
mriedemDetails: {u'message': u"[Errno 2] No such file or directory: '/opt/stack/data/nova/instances/bdbcf072-c793-4f85-931f-14b169d54b13/disk'", u'code': 500, u'created': u'2015-12-08T15:05:22Z'}16:25
edleafemarkus_z: I did that with the scheduler opts because, well, that's the way it already was :)16:25
dansmithmriedem: really? we get the console dump, which means the instance started16:25
rpodolyakamriedem: dansmith: agreed, we should not really log every exception before we give up :( but it's properly wrapped in DBDeadlock and it's *not* a raw pymysql exception right?16:25
mriedemdansmith: which is this http://status.openstack.org/elastic-recheck/index.html#152182316:25
*** changbl has quit IRC16:25
mriedemdansmith: yeah, this is a reboot test16:26
edleafemarkus_z: but I think that nova/opts should be the place where grouping is declared, not the individual files16:26
*** changbl has joined #openstack-nova16:26
mriedemso it starts, and then i think we try to reboot and kaboom16:26
markus_zsfinucan: I updated your review with the questions I had. I just needed to think a bit more about it.16:26
*** salv-orl_ has joined #openstack-nova16:26
sfinucanndipanov: Question on one of your review comments?16:26
mriedemdansmith: 2015-12-08 15:05:22,667 29753 INFO     [tempest.common.waiters] State transition "HARD_REBOOT/rebooting_hard" ==> "ERROR/reboot_started_hard" after 1 second wait16:26
dansmithmriedem: hmm16:26
sfinucanmarkus_z: cool - I'll check it out16:27
markus_zedleafe: Oh, why is that so? Those "opts.py" files are "only" for the generation of the sample conf file.16:27
markus_zedleafe: The "real" grouping which impacts the usage in the code is done in "register_opts" within the individual files.16:27
markus_zedleafe: But I might misunderstand what you say.16:27
mriedemdansmith: anyway, not volume related so i don't think that's one of the ones sean was talking about16:28
dansmithmriedem: ah, I forgot this is multinode16:28
mriedemstill a high percent failure in the multinode job though16:28
edleafemarkus_z: hmmm, then why specify the sections in nova/opts.py list_opts?16:28
sfinucanactually, anyone can answer this: is this not allowed --> 'from nova.objects.field import X'? If so, how come?16:28
dansmithmriedem: well, it's a volume test that fails to ssh, which is I think the only data he had, no?16:28
dansmithsfinucan: because you don't import things inside a module you import a module16:28
mriedemrpodolyaka: i'm not even sure how this is logged https://github.com/openstack/oslo.db/blob/3.1.0/oslo_db/api.py#L16916:29
*** ianbrown_ has joined #openstack-nova16:29
dansmithsfinucan: should be all over the developer docs16:29
*** markmc has quit IRC16:29
*** salv-orlando has quit IRC16:29
mriedemrpodolyaka: oh wait, yeah i do16:29
mriedemb/c it's the pymysql error16:29
*** mariusv_ has joined #openstack-nova16:29
sfinucandansmith: Damn. None of the tests picked it up :( OK, I'll fix16:29
dansmithmriedem: http://logs.openstack.org/32/253732/4/check/gate-grenade-dsvm-multinode/9f726ed/logs/subnode-2/old/screen-n-cpu.txt.gz?level=TRACE16:29
mriedemrpodolyaka: well, it's the deadlock error i gues16:29
dansmithsfinucan: hacking should fail16:29
mriedemdansmith: yup16:30
*** pratikmallya has joined #openstack-nova16:30
sfinucandansmith: Nope. Something to do with objects, Nikola thinks https://review.openstack.org/#/c/202648/23/nova/virt/hardware.py16:30
mriedemdansmith: that's what we're tracking in http://status.openstack.org/elastic-recheck/index.html#152182316:30
*** changbl has quit IRC16:30
dansmithmriedem: gotcha.. kinda seems like the old two-computes-one-hostname bug16:30
rpodolyakamriedem: hmm, I was sure DBDeadlock is 'our' exception. or there is one with the same name in pymysql too?16:30
markus_zedleafe: The oslo generator cannot determine this from the registering of the config opts. I don't know why.16:31
*** jdurgin1 has joined #openstack-nova16:31
mriedemrpodolyaka: no it's the DBDeadlock from oslo.db16:31
mriedemdansmith: hmm, i never heard how that was sorted out16:31
ndipanovsfinucan, shoot16:31
mriedemthought it was some infra thing?16:31
rpodolyakamriedem: ah,ok, that's better, at least we wrap it :P16:31
markus_zedleafe: I don't see a use case where I want this to be different.16:31
dansmithmriedem: it was, but we have new clouds16:31
mriedemrpodolyaka: yeah, so i'm going to push a change to remove that logging16:31
rpodolyakamriedem: cool! send me a link when you do16:31
mriedemrpodolyaka: since if we are done retrying, we log here https://github.com/openstack/oslo.db/blob/3.1.0/oslo_db/api.py#L14316:31
*** mariusv has quit IRC16:31
*** dedery_ has quit IRC16:31
edleafemarkus_z: that's what I was assuming, and why I think the grouping has to be explicit in nova/opts.py16:32
sfinucanndipanov: It's OK - dansmith explained the error of my ways ^^^16:32
markus_zedleafe: A "popular" issue is fixed by this: https://review.openstack.org/#/c/235396/16:32
mriedemand if its unexpected, we'd reraise here https://github.com/openstack/oslo.db/blob/3.1.0/oslo_db/api.py#L14116:32
*** smatzek has joined #openstack-nova16:32
ndipanovsfinucan, he's been know to do that16:32
dansmithmriedem: yeah, hostname is "ubuntu" on this one I think16:33
edleafemarkus_z: that reinforces my thought that the individual files shouldn't be determining the grouping16:33
*** otter768 has joined #openstack-nova16:33
edleafemarkus_z: just nova/opts.py16:33
dansmithmriedem: lemme go to -infra16:33
markus_zedleafe: Oh, I interpret it differently. The change which moved the vnc options to their own group did not change the "opts.py" because the grouping was in two different locations.16:34
edleafemarkus_z: the reason it had to change in nova/opts.py is because that's what determines what gets generated16:36
edleafemarkus_z: without that change there was no 'vnc' section, even with the register_opts specifying the vnc group16:37
openstackgerritBalazs Gibizer proposed openstack/nova: Add service status notification  https://review.openstack.org/24567816:37
openstackgerritBalazs Gibizer proposed openstack/nova: Add infra for versioned notifications  https://review.openstack.org/24702416:37
openstackgerritBalazs Gibizer proposed openstack/nova: Adds json sample for the versioned notifications  https://review.openstack.org/24816716:37
openstackgerritBalazs Gibizer proposed openstack/nova: Make emitting versioned notifications configurable  https://review.openstack.org/24756416:37
openstackgerritjichenjc proposed openstack/nova: Remove flavor.extra_spec  on hw_watchdog_action  https://review.openstack.org/25485216:37
*** otter768 has quit IRC16:38
markus_zedleafe: That's the change which moved the vnc options to their own group. https://review.openstack.org/#/c/193923/2/nova/vnc/__init__.py,cm I believe if there would have been a "list_opts" method the issue I linked earlier wouldn't have happended.16:38
edleafemarkus_z: so ideally there would be no groupings in nova/opts.py, and simply a series of list_opts() calls to each section of opts?16:40
markus_zedleafe: yes, exactly16:40
markus_zedleafe: I want to refactor those "opts.py" in a last step when all options are moved.16:40
markus_zedleafe: They are spread in the tree and you have to *know* that they are there and have to be changed.16:41
markus_zedleafe: I didn't make that clear I guess, sorry.16:41
openstackgerritMatt Riedemann proposed openstack/nova: Add hv testing for ImageMetaProps._legacy_property_map  https://review.openstack.org/25485616:41
mriedemgaryk: ^ added testing for your change16:41
edleafemarkus_z: that was why I preferred one place to name them and group them16:41
edleafemarkus_z: but spreading them out among all the files will work, too, once they are in the same directory16:42
openstackgerritMatt Riedemann proposed openstack/nova: Add hv testing for ImageMetaProps._legacy_property_map  https://review.openstack.org/25485616:42
markus_zedleafe: I'm missing something, sorry. What did you mean with "one place to name them"?16:43
*** changbl has joined #openstack-nova16:43
edleafemarkus_z: nova/opts.py, as opposed to doing that in many nova/conf/*py files16:44
*** boris-42_ has joined #openstack-nova16:44
*** armax has quit IRC16:44
markus_zedleafe: Ah. And that "nova/opts.py" has all config options we have?16:44
*** unicell has quit IRC16:45
markus_zedleafe: If "yes", there were concerns about merge conflicts and that file would be overwhelming.16:45
edleafemarkus_z: it certainly appears that it does, although I haven't gone through each one :)16:45
*** unicell has joined #openstack-nova16:45
*** yarkot has quit IRC16:46
edleafemarkus_z: since each individual option isn't spelled out16:46
edleafemarkus_z: just the list from each file16:46
*** jichen has quit IRC16:46
edleafemarkus_z: e.g., nova.db.api.db_opts16:47
edleafemarkus_z: not each individual db opt16:47
*** openstackgerrit has quit IRC16:47
*** openstackgerrit has joined #openstack-nova16:47
edleafemarkus_z: so that if another db opt is added, nova/opts.py would pick it up16:47
openstackgerritIldiko Vancsa proposed openstack/nova: virt: Allow multiple volume attachments  https://review.openstack.org/19313516:48
openstackgerritIldiko Vancsa proposed openstack/nova: WIP: REST API changes to support multi-attach  https://review.openstack.org/15303816:48
*** Daisy has joined #openstack-nova16:48
*** yarkot has joined #openstack-nova16:49
markus_zedleafe: Isn't that the same when adding a new option to "nova/conf/<module>.py" and adding it to the "ALL_OPTS" list?16:50
*** armax has joined #openstack-nova16:51
*** zhenq has quit IRC16:51
edleafemarkus_z: well, if the ALL_OPTS is then returned by list_opts()16:51
markus_zedleafe: yes, true16:51
edleafemarkus_z: the point was trying to understand why you thought there would be merge conflicts16:52
*** Daisy has quit IRC16:52
markus_zedleafe: It seems that we made bad experiences with the exceptions.py module which contains all of our exception classes.16:53
edleafemarkus_z: ah16:54
edleafemarkus_z: I think that a series of .py files for each logical opt grouping, and a central place for gathering them makes a lot of sense16:55
edleafeas long as the central place isn't listing them one-by-one16:55
edleafebut getting them from each .py file16:55
*** ianbrown__ has joined #openstack-nova16:55
*** alejandrito has joined #openstack-nova16:55
*** unicell1 has joined #openstack-nova16:56
*** apoorvad has joined #openstack-nova16:56
*** apoorvad has quit IRC16:56
*** unicell has quit IRC16:57
markus_zedleafe: yepp. I didn't yet try it but I think there will be a "nova/conf/opts.py" file which calls all of the "list_opts" methods. The other "opts.py" in the tree would then be obsolete.16:57
*** ianbrown_ has quit IRC16:57
edleafemarkus_z: that sounds sane.16:58
markus_zedleafe: I forget to put some important words in my sentences... I'm not yet 100% fit.16:58
*** yamahata has joined #openstack-nova16:58
markus_zedleafe: It does? Or did I filibuster you?16:58
*** mgoddard has quit IRC16:59
edleafemarkus_z: I'm perfectly healthy and I forget to make my thoughts clear :)16:59
*** danpb has quit IRC16:59
edleafemarkus_z: no, it's along the lines of how I see it. Whether the central place is nova/opts.py or nova/conf/opts.py isn't as important17:00
dansmithsdague: so can you link me to one of the logs where we're failing a volume test with the ssh failure that isn't the same as the one where I noticed the hostname thing?17:00
markus_zedleafe: True. As long as simple bugs like this get prevented: https://bugs.launchpad.net/nova/+bug/150635617:00
openstackLaunchpad bug 1506356 in OpenStack Compute (nova) "There is no "[vnc]" option group in nova.conf.sample" [Low,In progress] - Assigned to shunya kitada (skitada)17:00
dansmithsdague: I would assume that if this is happening, any number of tests would fail for various reasons as calls get directed to the wrong compute node(s)17:01
*** vishwanathj has joined #openstack-nova17:01
markus_zedleafe: OK, cool, we have a conclusion, right?17:02
*** ildikov has quit IRC17:02
sdaguedansmith: sure, let me look17:03
edleafemarkus_z: not sure - do you now agree that a central place for defining the groups is the way to go?17:03
edleafemarkus_z: i.e., nova/conf/opts.py?17:03
*** vilobhmm11 has quit IRC17:03
*** busterswt has quit IRC17:04
*** jlanoux has quit IRC17:04
*** alex_xu has quit IRC17:04
markus_zedleafe: let me write some (pseudo) code, one sec17:04
*** yalie has quit IRC17:04
sdaguedansmith: yeh - http://logs.openstack.org/92/253792/7/check/gate-grenade-dsvm-multinode/d52d36a/logs/old/screen-n-cond.txt.gz#_2015-12-07_20_42_59_53117:04
*** Yingxin has quit IRC17:04
*** gzhai3 has quit IRC17:04
sdaguehost = ubuntu17:04
dansmithsdague: yeah17:04
sdaguegood catch17:04
sdagueit will be good to see what the fail rate settles into once this fix is out there17:05
markus_zedleafe: I'm putting it at the end of this etherpad: https://etherpad.openstack.org/p/config-options17:05
*** apoorvad has joined #openstack-nova17:06
*** zhenq has joined #openstack-nova17:06
*** Kennan has quit IRC17:06
*** houming has quit IRC17:07
edleafemarkus_z: cool17:07
dansmithmarkus_z: you should -1 that pci doc patch.. hard to tell there are pending comments otherwise17:07
*** Kennan has joined #openstack-nova17:07
markus_zdansmith: ah, ok, I'll do that17:07
dansmithmarkus_z: thanks17:08
cfriesenndipanov:  you've been looking at sfinucan's threading policy patches, but the spec change is still in progress...any chance we can get https://review.openstack.org/#/c/244198/ approved?17:08
*** sudipto has quit IRC17:09
*** gzhai3 has joined #openstack-nova17:09
sdagueBobBall: I can fix the registration, but https://review.openstack.org/#/c/254760/2 failure is pretty opaque to me17:09
*** busterswt has joined #openstack-nova17:09
ndipanovcfriesen, let's land patches first?17:10
*** alex_xu has joined #openstack-nova17:10
*** Yingxin has joined #openstack-nova17:11
markus_zedleafe: I added the code as I understand it. Would you change it to show what you mean?17:12
ndipanovcfriesen, I think actual patches are more important and the BP is approved17:12
rlrossitdansmith: prepare for me bugging you some more... if possible, could you take a look at https://review.openstack.org/#/c/252594/ to let me know if I'm on the right track before the nova notifications meeting this afternoon?17:12
edleafemarkus_z: sure17:12
rlrossitI will be eternally grateful :)17:12
sfinucancfriesen: I don't think we need the spec reworked to get code in, per earlier discussion with ndipanov. Suits me as there's enough work in the code alone :)17:13
dansmithrlrossit: does tuple support += ?17:13
sfinucanHowever, I'm off home so I'll let you discuss it among yourselves, if necessary17:13
rlrossitdansmith: it did for me when I was testing at least17:13
dansmithrlrossit: doesn't for me just now17:13
dansmithrlrossit: er wait, it does17:14
dansmithhmm, that's weird17:14
*** amotoki has quit IRC17:14
rlrossitdid you try to append something like (stuff) instead of (stuff,) ?17:14
dansmithyeah, but tuples are supposed to be immutable17:14
dansmithso whatever17:14
*** ccarmack has quit IRC17:15
rlrossitdansmith: depends on what you mean by "immutable"17:15
rlrossitthey still are her17:15
dansmithrlrossit: well, clearly17:15
*** yalie has joined #openstack-nova17:16
*** salv-orl_ has quit IRC17:16
*** RichardRaseley has joined #openstack-nova17:17
*** unicell1 has quit IRC17:18
rlrossitdansmith: :)17:18
markus_zrlrossit: rofl :D17:20
rpodolyakamriedem: btw, I completely forgot about this one - https://review.openstack.org/#/c/222783/417:21
rpodolyakamriedem: maybe you could simply update it17:21
*** jistr|mtg is now known as jistr17:21
*** paul-carlton_ has quit IRC17:22
openstackgerritLudovic Beliveau proposed openstack/nova: Allow saving empty pci_device_pools in ComputeNode object  https://review.openstack.org/21604917:23
edleafemarkus_z: ok, changed it to centralize the groups in nova/conf/opts.py17:23
markus_zedleafe: Ah, ok, I did understand you correctly. My issue with that approach is that the group name "vnc" is spread over two files and one has to know that these two must be the same.17:24
*** matrohon has quit IRC17:25
edleafemarkus_z: yeah, I see that17:25
*** rfolco_ has quit IRC17:25
edleafemarkus_z: let me change it once again...17:25
*** Yingxin has quit IRC17:26
mriedemrpodolyaka: i can clean that up, i don't like the kwarg though,17:26
mriedemrpodolyaka: b/c that menas nova has to pass LOG.DEBUG wherever the retry decorator is used17:26
*** mrkz has quit IRC17:26
mriedemrpodolyaka: oslo.db should just be sane and not trace exceptions for things it expects to hit in a retry decorator17:27
*** yassine__ has quit IRC17:27
edleafemarkus_z: how about now?17:27
mriedemrpodolyaka: i have a nova change to cleanup first and then i'll update kevinbenton's change17:27
*** rotbeard has quit IRC17:27
markus_zedleafe: yepp, that's very good imo.17:28
edleafemarkus_z: IOW, make the grouping explicit, and that can be a pattern for all conf/*.py files17:28
kevinbentonmriedem: my oslo db patch?17:28
mriedemyeah https://review.openstack.org/#/c/222783/17:28
*** rfolco_ has joined #openstack-nova17:28
mriedemi'd rather not have to pass the log level kwarg in17:28
kevinbentonmriedem: yeah, feel free to take over that. i'm swamped17:28
mriedemoslo.db should just be smarter17:28
rpodolyakamriedem: I'm ok with .debug()17:28
markus_zedleafe: yes. that's better than my first version. Let me change it one more time :)17:28
mriedem#success: a small handful of people agree on sanity!17:29
openstackstatusmriedem: Added success to Success page17:29
mriedemkevinbenton: btw, i watched your mitaka summit session on perf profiling in neutron db stuff, very nice17:29
kevinbentonthanks! glad someone else saw it. the room had like 15 people in it :)17:30
mriedemi watched on youtube17:30
mriedembut yeah17:30
markus_zedleafe: what about now?17:30
*** Yingxin has joined #openstack-nova17:31
mriedemdansmith: i forget, does x in y where y is a nova object trigger lazy load?17:31
*** tonytan4ever has quit IRC17:31
edleafemarkus_z: how will that work with DEFAULT?17:31
*** mrkz has joined #openstack-nova17:31
dansmithmriedem: shouldn't if it's using obj_attr_is_set, which is what you did right?17:31
mriedemthis is in the bdm object17:31
johnthetubaguymriedem: talking of which, we have this os-profiler spec and patches submitted: https://blueprints.launchpad.net/nova/+spec/osprofiler-support-in-nova17:31
markus_zedleafe: hmm... I guess it doesn't. But with all the other groups.17:32
mriedemjohnthetubaguy: i don't think kevinbenton used any of the osprofiler stuff17:32
edleafemarkus_z: that's why I thought a constant with the group name would be cleaner17:32
*** sahid has quit IRC17:32
johnthetubaguymriedem: interesting, I was curious about that17:32
edleafemarkus_z: if you're using DEFAULT17:32
kevinbentonosprofiler would have been nice, this was just based on measuring response times of calls over the AMQP bus17:33
markus_zedleafe: Yes, If you're using DEFAULT than your example was the better approach, yes.17:33
mriedemjohnthetubaguy: ^17:33
mriedemthat was the session17:33
edleafemarkus_z: and at least for now, most are in DEFAULT17:33
mriedemthe funny thing was the parts where 'we added x to improve performance and it degraded performance somehow'17:34
markus_zedleafe: A lot, true17:34
kevinbentonyeah, like whackamole17:34
rpodolyakaheh :)17:34
kevinbentonimproved one call and punished another17:34
ndipanovdansmith, move this to +0 at least? https://review.openstack.org/#/c/148277/17:35
johnthetubaguymriedem: kevinbenton: totally went through that with the scheduler, turning off my original enhancement idea improved the performance17:35
rpodolyakakevinbenton:  yeah, such 'improvements' without proper benchmarking are profiling are often painful17:35
*** scheuran has quit IRC17:36
*** aginwala has joined #openstack-nova17:36
*** pratikmallya has quit IRC17:36
kevinbentonthe challenge is that the different agents need things grouped in different ways. so turning on a joined load would help the one but then slow down one that didn't need the extra info17:37
dansmithndipanov: you asked me to -1 this in the first place and now you want me to remove it?17:37
dansmithndipanov: as I said before, I think we should get bauzas' read on it17:37
*** gmann has quit IRC17:38
dansmithgiven that bauzas' set is, IMHO, more important, I surely don't care about jeopardizing this notifications set while making his work harder17:38
*** smatzek has quit IRC17:39
dane-fichterCan anyone give me some pointers on getting the py34 test environment to work correctly locally? It does not run my tests even though they are included in tests-py3.txt17:39
mriedemdane-fichter: there is a blacklist17:39
*** klkumar has quit IRC17:39
markus_zedleafe: how about the conclusion at the end in the etherpad?17:40
*** gmann has joined #openstack-nova17:40
dansmithndipanov: oh, he did rebase on bauzas' set, right? you're just asking me to clear the -1 since jenkins re-added it?17:40
mriedemdane-fichter: https://github.com/openstack/nova/blob/master/tests-py3.txt17:40
mriedemdane-fichter: https://github.com/openstack/nova/blob/master/tox.ini#L5017:40
*** lykinsbd has quit IRC17:40
mriedemso you have to remove the tests from the blacklist if you want to run them with the py34 target17:40
dane-fichtermriedem: ah ok. why does the gate on Gerrit run them then?17:41
ndipanovdansmith, yes17:41
dansmithI'm not sure why it re-added that after a rebase17:41
ndipanovwell the patch was not affected17:42
*** aginwala has quit IRC17:42
ndipanovit was a clean rebase17:42
dansmithwell, I know, but usually it does17:42
dansmiththere must be some detail I'm missing about when it does and doesn't17:42
ndipanovfor a clean rebase it won't clear reviews17:42
dansmithwell, it does clear mine when I rebase cleanly on master and resubmit17:43
dansmithall the time17:43
ndipanovit clears jenkins but not humans17:43
*** garyk has quit IRC17:43
dansmithmaybe there are multiple levels of clean-ness17:43
ndipanovhumans > jenkinses17:43
edleafemarkus_z: updated the etherpad17:44
ndipanovalleged humans I should say17:44
markus_zedleafe: absolutely sane that way17:44
dane-fichtermriedem: does the gate-nova-python34 gate run all tests regardless of the tests-py3.txt blacklist?17:44
edleafemarkus_z: ok, I'll update the scheduler option bits17:44
*** aginwala has joined #openstack-nova17:45
tjonesmarkus_z: i'll break the dependency on my other patches from the volume one.  was just trying to avoid the rebase on __init__ but it's trivial.  :-)17:45
*** ccarmack has joined #openstack-nova17:45
markus_zedleafe: cool, thanks17:45
*** Redcavalier has joined #openstack-nova17:45
markus_ztjones: ok, great, I wanted to talk to you about that. Thanks for that! Please also upload follow up patches which describe the options more in detail.17:46
openstackgerritLudovic Beliveau proposed openstack/nova: Support configuration of multiple pci_alias within an array  https://review.openstack.org/24255517:47
tjonesyep will do17:47
openstackgerritEd Leafe proposed openstack/nova: Replace os.path.join() for URLs  https://review.openstack.org/21830917:47
markus_ztjones: awesome, thanks!17:47
mriedemdane-fichter: no17:48
mriedemit's the same as local17:48
RedcavalierHi guys, got a question about the nova network code and how it interacts with openvswitch. We notice that nova creates two veth and a linux bridge. We're trying yo find where exactly the second veth is linked to the OVS port. I'm pretty sure it's in the nova code, but please correct me if I'm wrong.17:48
mriedemdane-fichter: the gate jobs wouldn't / shouldn't be running them17:48
mriedemdane-fichter: i'd need the actual test id though17:49
dane-fichtermriedem: https://review.openstack.org/#/c/189843/17:49
mriedemdane-fichter: so nova.tests.unit.image.test_glance.TestDownloadNoDirectUri.test_download_with_signature_verification_fail17:50
dane-fichterso my tests in test_glance are failing, even though they are part of the TestDownloadNoDirectURI17:50
openstackgerritAlexis Lee proposed openstack/nova: Only consider requested_destination  https://review.openstack.org/24883317:50
openstackgerritAlexis Lee proposed openstack/nova: Factor _rebuild_pick_host out of rebuild_instance  https://review.openstack.org/25488717:50
*** e0ne has quit IRC17:50
mriedemdane-fichter: so that test class isn't blacklisted17:51
mriedemso you should be able to run it locally fine17:51
edleafemarkus_z: question on the scheduler opts. How does the generator find that?17:51
mriedemtox -r -e py34 -- nova.tests.unit.image.test_glance.TestDownloadNoDirectUri.test_download_with_signature_verification_fail17:51
cfriesensfinucan: ndipanov: sorry for the lag, wasn't paying attention.  I guess I'm a bit confused because some of the proposed patches are only valid if we approve the modifications to the spec.17:51
*** ianbrown__ has quit IRC17:52
markus_zedleafe: There is the file "etc/nova/nova-config-generator.conf".17:52
markus_zedleafe: the namespace is a folder which must contain a "opts.py" module with a "list_opts" function17:52
edleafemarkus_z: ah, so just remove the reference from there17:52
cfriesensfinucan: ndipanov: otherwise the spec won't reflect the implementation, leading to confusion later on when people read the spec and not the code17:52
dane-fichtermriedem: it looks blacklisted to me17:53
edleafemarkus_z: so we can move them into nova/conf17:53
mriedemdane-fichter: nova.tests.unit.image.test_glance is not blacklisted17:54
mriedemit doesn't show up in https://github.com/openstack/nova/blob/master/tests-py3.txt17:54
dane-fichtermriedem: I guess I locally have an old version of tests-py3 (?) I just cloned this repo from github though17:54
markus_zedleafe: You want to delete the "nova/scheduler/opts.py" file? If "yes", than you have to delete the reference in "nova-config-generator" I guess. As I said, I didn't yet try it.17:55
mriedemdane-fichter: could be an old os-testr?17:55
edleafemarkus_z: ok17:55
dane-fichtermriedem: perhaps. thanks for your help17:55
mriedemdane-fichter: it uses os-testr to trim the tests https://github.com/openstack/nova/blob/master/tox.ini#L5017:55
cdentmordred++ that letter to the tc list is excellent17:56
*** ianbrown__ has joined #openstack-nova17:56
markus_zedleafe: I have to leave for today. See you tomorrow.17:56
edleafemarkus_z:  ok - enjoy your evening17:57
*** busterswt has quit IRC17:57
markus_zedleafe: thanks, have a nice day17:57
*** markus_z has quit IRC17:57
*** yarkot has quit IRC17:58
mordredcdent: thanks!17:58
*** suro-patz has joined #openstack-nova17:58
*** shz has quit IRC18:00
*** jaosorior has quit IRC18:00
*** vishwanathj has quit IRC18:00
*** Vek has joined #openstack-nova18:01
*** shz has joined #openstack-nova18:01
*** smatzek has joined #openstack-nova18:02
*** yalie1 has joined #openstack-nova18:05
*** jistr has quit IRC18:05
*** PaulMurray is now known as ptm_away18:06
*** pratikmallya has joined #openstack-nova18:07
*** yalie has quit IRC18:09
*** gzhai has joined #openstack-nova18:09
*** unicell has joined #openstack-nova18:10
*** gzhai3 has quit IRC18:12
*** derekh has joined #openstack-nova18:14
mriedemndipanov: when converting a BlockDeviceDict to a nova.virt.block_device.DriverBlockDevice, that wraps a bdm object (creates one if the bdm passed in isn't a NovaObject), where do we ever set the context on the internal bdm nova object that we create?18:15
mriedemsince bdm.save() requires a context18:15
*** tonytan4ever has joined #openstack-nova18:16
*** derekh has quit IRC18:16
mriedemthe nova.tests.virt.test_block_device tests just mock out bdm.save() like this:18:16
mriedemself.mox.StubOutWithMock(driver_bdm._bdm_obj, 'save')18:16
mriedemso i'm not sure if we're just working in devstack b/c the DriverBlockDevice is always constructed with a BlockDeviceMapping object (that has a context in it), or if i'm missing something18:17
*** salv-orlando has joined #openstack-nova18:17
ndipanovmriedem, not sure might be a bug18:18
openstackgerritLudovic Beliveau proposed openstack/nova: Support configuration of multiple pci_alias within an array  https://review.openstack.org/24255518:18
ndipanovprobably it's always constructed from an object meaning we should remove the other option <- mriedem18:18
mriedemndipanov: yeah....i'm thinking so18:18
sfinucancfriesen: That would make sense, but assuming the enum gets +2d, the only possible changes could be with wording18:19
sfinucanso not really that important18:19
sfinucanat least that's my take :)18:19
*** alejandrito has quit IRC18:19
mriedemndipanov: bdm.save() does take a context, so we could pass that in, if we wanted to preserve the old transformation code, but it's probably not worht it18:19
*** alejandrito has joined #openstack-nova18:19
sfinucannow I'm *really* going home :)18:19
*** sfinucan has quit IRC18:19
mriedemdansmith: am i misremembering the stance on passing context to objects CRUD methods, but was save(context) still ok?18:19
mriedemno, BlockDeviceMapping.save() doesn't take context anymore, even if the parent class does18:20
dansmithmriedem: if it's remotable (not classmethod) then it should not take a context18:20
mriedem    @base.remotable18:21
mriedem    def save(self):18:21
*** tonytan4ever has quit IRC18:21
*** ccarmack has quit IRC18:21
*** ssurana has joined #openstack-nova18:21
*** salv-orlando has quit IRC18:21
mriedemndipanov: so i'm going to remove this https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L128-L13118:21
mriedemsince it's busted18:21
mriedemever since we did https://review.openstack.org/#/c/164268/18:22
mriedemwe just papered over the fact in the tests18:22
mriedemby mocking out save()18:22
*** ctrath has quit IRC18:23
cfriesensfinucan: fair enough.  I just thought it was sort of backwards to approve code changes before approving the spec driving the changes. :)18:24
*** gokrokve has quit IRC18:24
ndipanovmriedem, sounds good18:24
ndipanovping me for a +2 if in next 45 minutes as I have a call so will be around18:25
*** vilobhmm11 has joined #openstack-nova18:25
*** thangp has quit IRC18:25
*** shakamunyi has joined #openstack-nova18:25
*** gokrokve has joined #openstack-nova18:25
mriedemndipanov: this will be tomorrow18:25
mriedemi have to hack around it for now to get  some tests passing18:26
mriedemi opened a bug for tracking though https://bugs.launchpad.net/nova/+bug/152403518:26
openstackLaunchpad bug 1524035 in OpenStack Compute (nova) "nova.virt.block_device.DriverBlockDevice cannot save to DB if bdm passed in was not already an object" [Undecided,New]18:26
*** mgoddard has joined #openstack-nova18:26
ndipanovmriedem, cool18:27
*** jaypipes has quit IRC18:27
*** ctrath has joined #openstack-nova18:28
*** pumaranikar has quit IRC18:29
*** ccarmack has joined #openstack-nova18:29
*** gokrokve has quit IRC18:30
*** dims_ has quit IRC18:30
*** ijw has joined #openstack-nova18:30
*** ircuser-1 has quit IRC18:30
*** xyang1 has quit IRC18:31
*** xyang1 has joined #openstack-nova18:31
*** gokrokve has joined #openstack-nova18:31
*** ijw_ has joined #openstack-nova18:32
*** ijw has quit IRC18:32
*** gokrokve has quit IRC18:33
*** otter768 has joined #openstack-nova18:34
*** gokrokve has joined #openstack-nova18:34
*** tonytan4ever has joined #openstack-nova18:35
*** paul-carlton2 has quit IRC18:37
*** tjones has quit IRC18:37
*** xyang1 has quit IRC18:38
*** smurke has quit IRC18:38
*** smurke has joined #openstack-nova18:38
*** otter768 has quit IRC18:39
*** jerrygb has quit IRC18:39
*** e0ne has joined #openstack-nova18:39
*** gzhai1 has joined #openstack-nova18:41
*** tjones has joined #openstack-nova18:42
*** pumaranikar has joined #openstack-nova18:43
openstackgerritVladik Romanovsky proposed openstack/nova: pci: adding support to specify a device_type in pci requests  https://review.openstack.org/24901718:43
openstackgerritVladik Romanovsky proposed openstack/nova: pci: changing the claiming and allocation logic for PF/VF assignment  https://review.openstack.org/24901818:43
*** Kennan has quit IRC18:43
*** Kennan has joined #openstack-nova18:44
*** gzhai has quit IRC18:44
*** cdent has quit IRC18:45
*** ljxiash has quit IRC18:45
*** jerrygb has joined #openstack-nova18:46
*** rlrossit has quit IRC18:46
*** aginwala has quit IRC18:47
*** david-ly_ has joined #openstack-nova18:50
*** rook has quit IRC18:51
*** david-ly_ is now known as david-lyle_18:52
*** rlrossit has joined #openstack-nova18:52
*** eliqiao has quit IRC18:52
openstackgerritAugustina Ragwitz proposed openstack/nova: Update functional tests to support optional project id  https://review.openstack.org/25448518:53
*** david-lyle has quit IRC18:54
*** eliqiao has joined #openstack-nova18:55
*** ildikov has joined #openstack-nova18:55
*** ctrath has quit IRC18:56
openstackgerritTracy Jones proposed openstack/nova: config options: centralize section "volume"  https://review.openstack.org/25374718:56
rlrossitgibi: what's the plan for developer documentation for the notification work?18:57
openstackgerritSean Dague proposed openstack/nova: tighten regex on objectify  https://review.openstack.org/25490918:57
sdaguedansmith: because your fingerprints remain in api_samples I expect some blood money / review of that ^^^18:58
sdaguedon't worry, it's small18:58
*** penick has joined #openstack-nova18:58
sdagueand almost entirely a comment explaining wtf is going on18:59
*** READ10 has quit IRC18:59
*** penick_ has joined #openstack-nova19:00
* dansmith runs19:00
openstackgerritVilobh Meshram proposed openstack/nova: Introduce Flavor Classes  https://review.openstack.org/25330619:00
*** openstackgerrit has quit IRC19:02
dansmithsdague: doesn't that change integers into strings?19:02
*** openstackgerrit has joined #openstack-nova19:03
*** salv-orlando has joined #openstack-nova19:04
*** penick has quit IRC19:04
*** penick_ is now known as penick19:04
openstackgerritSudipta Biswas proposed openstack/nova: Remove version hardcoding for 'show' method  https://review.openstack.org/22857819:04
*** ctrath has joined #openstack-nova19:05
dansmithsdague: I guess there is some magic in the int: bit but I don't know where that comes from19:06
sdaguedansmith: so that's always been there19:06
sdagueI'm just changing the anchor on the sub regex19:06
dansmithsdague: yeah, I just don't really know how this is really safe and/or why we're doing this19:07
sdaguefrom ! " to ': '19:07
sdagueso if you don't have it, you can't load the templates as json19:07
sdaguebecause it's a parse error19:07
dansmithoh, I see, I see because we've got these raw things19:07
dansmithguh, yuck19:07
*** gokrokve has quit IRC19:08
sdagueyeh, well, blame vishy :)19:08
sdaguethere is only so much unwind we can do here, but I'm trying to simplify an upper layer, and this just got in the way of it19:08
sdagueat least I payed it forward with more comments19:08
*** Sukhdev has joined #openstack-nova19:09
*** lpetrut has joined #openstack-nova19:10
*** salv-orlando has quit IRC19:11
*** burt has joined #openstack-nova19:13
*** david-lyle_ is now known as david-lyle19:14
*** aginwala has joined #openstack-nova19:15
*** gokrokve has joined #openstack-nova19:16
*** rook has joined #openstack-nova19:17
*** garyk has joined #openstack-nova19:26
openstackgerritJim Rollenhagen proposed openstack/nova: Allow virt driver to define binding:host_id  https://review.openstack.org/19441319:28
*** hblixt has joined #openstack-nova19:28
*** ZZelle_ has joined #openstack-nova19:29
*** aginwala has quit IRC19:29
*** aginwala has joined #openstack-nova19:30
*** gokrokve has quit IRC19:31
*** whenry_ has joined #openstack-nova19:31
*** pratikmallya has quit IRC19:33
*** pratikmallya has joined #openstack-nova19:36
*** harlowja has quit IRC19:38
*** harlowja has joined #openstack-nova19:38
gibirlrossit: notification samples as in the forth patch, devref about how to add a new versioned notification19:40
*** erhudy has quit IRC19:40
gibirlrossit: and also agreed to do some user facing documentation, like a list of notifications, maybe with swagger19:41
*** irenab has quit IRC19:42
rlrossitgibi: I guess my question was more directed at: if I want to make a new notification, what are the steps needed?19:42
rlrossitlike how do I make the schema work19:42
gibirlrossit: that is what we will cover with a devref19:43
gibirlrossit: at least in my plans :)19:44
rlrossitgibi: oh, I read your message before wrong. I thought you meant we'll be using the json samples as devref. My bad19:44
*** burt has quit IRC19:45
gibirlrossit: sorry I wasn't clear enough19:45
*** mc_nair_ has joined #openstack-nova19:51
*** mc_nair has quit IRC19:52
*** penick has quit IRC19:52
*** jichen has joined #openstack-nova19:55
migidansmith, Hi19:56
dansmithmigi: o/19:56
*** aysyd has quit IRC19:57
migidansmith, o/ = ?19:57
openstackgerritJim Rollenhagen proposed openstack/nova: Allow virt driver to define binding:host_id  https://review.openstack.org/19441319:57
jrollmigi: o/ = https://media.giphy.com/media/JGKwzgpKXVYaY/giphy.gif19:58
dansmithright, that ^19:58
dansmithexcept less dorky19:58
*** lbelivea has quit IRC20:00
migiso I am doing that quite easy change and wondering if it's right... in short does all id's needs to be uuids? Example number 1 was given to fixed_ip20:01
dansmithmigi: we have ids and uuids.. ids are integers and uuids are ... uuids, so you should only change the latter20:02
openstackgerritDina Belova proposed openstack/nova: Integrate OSprofiler and Nova  https://review.openstack.org/25470320:03
*** ctrath has quit IRC20:03
*** rfolco_ has quit IRC20:03
migidansmith, right, just the logic makes them same in few tests20:04
migiso will rewrite those parts20:05
*** aginwala has quit IRC20:05
dansmithmigi: have you pushed any patches up yet?20:05
*** lbelivea has joined #openstack-nova20:05
*** artom has quit IRC20:05
dansmithagain, I'd recommend not waiting until it's all done before pushing up early pieces :)20:05
migidansmith, just one:20:06
migitoday got more, but prefer to fix few things before push20:07
*** ctrath has joined #openstack-nova20:07
*** aginwala has joined #openstack-nova20:09
*** ctrath has quit IRC20:09
*** ociuhandu has quit IRC20:09
*** pece has joined #openstack-nova20:09
*** lbelivea has quit IRC20:10
*** ctrath has joined #openstack-nova20:12
*** lucasagomes is now known as lucas-dinner20:13
*** mariusv_ has quit IRC20:13
*** mariusv has joined #openstack-nova20:15
*** mariusv has quit IRC20:15
*** mariusv has joined #openstack-nova20:15
*** yarkot has joined #openstack-nova20:18
dansmithmigi: commented on one thing I think is inconsistent (but I may be wrong). otherwise that looks good to me20:18
dansmithmigi: also, that's about the right (i.e. max) amount of change for one patch, IMHO20:18
dansmithmigi: it's not a 36-char string field in the DB: https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/models.py#L247-L24720:21
migidansmith, running test to show exception, will reply to comment20:21
dansmithsdague: mriedem: do you know if Instance.image_ref can be anything other than the image uuid?20:21
dansmithmigi: okay cool20:21
migithat is why I've asked if all should be uuids as I was getting errors in places where I shouldn't (not real uuids imo)20:22
sdaguedansmith: so, in the glance code image_ref is allowed to be a full url20:22
dansmithsdague: orly20:22
mriedemdansmith: i wouldn't want to guess20:22
sdaguein the nova glance code that I've been unwinding20:22
sdaguewhen it can be that, is not super clear20:22
sdaguebut the code supports it20:22
dansmithsdague: okay thanks20:22
dansmithsdague: it's a varchar(255) in the database (like everything else), which is probably rather dangerous for a url20:23
sdagueit might not be from the db20:23
dansmithright, I mean we could truncate it20:23
dansmithin the db20:23
*** ociuhandu has joined #openstack-nova20:23
sdagueyeh, I can look more tomorrow, tc meeting now and then I have a hard stop20:23
dansmithsdague: migi is cleaning up our tests to remove all cases where we use "fake-uuid", etc for instance uuids, in advance of making the Instance object actually enforce that20:24
mriedemdansmith: also https://github.com/openstack/nova/blob/master/nova/cloudpipe/pipelib.py#L6920:24
openstackgerritSamuel Matzek proposed openstack/nova: glance: secure https when getting client version  https://review.openstack.org/25494420:24
mriedemused here https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L162120:24
dansmithmriedem: ah20:25
dansmithso probably either id or uuid20:25
dansmithbecause we hate ourselves and our users20:25
mriedemyeah man20:25
mriedemi suppress those feelings20:25
*** mariusv has quit IRC20:26
sdaguedansmith: yeh, so we should be a little cautious there on the image_ref side20:26
sdaguethough, honestly, some code simplifies if we don't let it ever be a url20:26
dansmithsdague: well, for the moment it sounds like not assuming it's a uuid is the safe bet.. not really related to his cleanups anyway,20:27
dansmithsince it's not a uuidfield on the instance20:27
RichardRaseleyCan anyone provide some detail around the mechanism nova-consoleauth uses to validate tokens which are passed to it?20:27
mriedemjgriffith: http://logs.openstack.org/07/251407/2/check/gate-tempest-dsvm-full/144f7fc/logs/libvirt/qemu/instance-00000066.txt.gz20:27
mriedemdoes this mean anything to you?20:27
mriedemqemu-system-x86_64: /build/qemu-5LgLIn/qemu-2.0.0+dfsg/block.c:3491: bdrv_error_action: Assertion `error >= 0' failed.20:27
*** lykinsbd has joined #openstack-nova20:27
mriedemkashyap: ^20:27
*** rhagarty has quit IRC20:30
openstackgerritDina Belova proposed openstack/nova: Integrate OSprofiler and Nova  https://review.openstack.org/25470320:30
*** rhagarty has joined #openstack-nova20:30
migidansmith, replied to comment20:33
dansmithmigi: I don't see it20:33
*** otter768 has joined #openstack-nova20:35
*** apoorvad has quit IRC20:35
migidansmith, my comment is (draft)20:36
migiwhere the heck is save btn20:36
dansmithmigi: you have to reply again at the main patch screen to commit it20:36
dansmithmigi: i.e. vote again20:37
migipretty user friendly20:37
dansmithmigi: ah, I see20:38
migisimilar type for number of networking stuff20:38
dansmithmigi: so danpb already made the assumption about the image uuid on ImageMeta20:38
migiand one field so far where uuid is None (which may be valid for some types)20:39
*** mariusv_ has joined #openstack-nova20:39
migiyes, but I am looking here for ideas if we want to fix it not only in tests20:39
*** otter768 has quit IRC20:40
migior keep tests working and then if necessary change image to non-uuid20:40
*** mariusv has joined #openstack-nova20:41
openstackgerritMatt Riedemann proposed openstack/nova: Make sure bdm.volume_id is set after auto-creating volumes  https://review.openstack.org/25442820:41
*** mariusv_ has quit IRC20:41
migidansmith, see http://paste.openstack.org/show/481231/20:43
*** ircuser-1 has joined #openstack-nova20:43
bauzasmriedem: have you seen that ? https://jenkins03.openstack.org/job/gate-nova-python34/3088/console20:44
*** markmc has joined #openstack-nova20:44
migiI am just unsure if this won't brake nova (not only tests), cause as in the previous examle db may allow int rather then full uuids20:44
bauzasmriedem: "mox3.mox.UnexpectedMethodCallError: Unexpected method call function.__call__(<nova.virt.xenapi.client.session.XenAPISession object at 0x7fa86a6a4c18>, 'test', False) -> None""20:44
dansmithmigi: we treat instance uuids and ids different by query at the DB layer, so those tests are just wrong I think20:45
dansmithmigi: so you're good to change I think20:45
mriedembauzas: not specifically, but there is some race with mox and neutronv2.api unit tests in nova on py3420:45
vilobhmm11alaski : ping20:45
migidansmith, ok20:45
bauzasmriedem: lemme see logstash20:45
alaskivilobhmm11: hi20:45
*** xyang1 has joined #openstack-nova20:45
alaskivilobhmm11: I'm commenting on the patch right now btw20:45
vilobhmm11https://review.openstack.org/#/c/253306/ I still feel it should be a listofString or listofObjects20:45
vilobhmm11alaski : ok :)20:46
openstackgerritChuck Carmack proposed openstack/nova: REST API changes for user settable server description  https://review.openstack.org/25495020:46
*** haypo_ has joined #openstack-nova20:46
*** yarkot has quit IRC20:49
mriedemi've re-opened this old qemu bug that's actually associated to fc2020:49
openstackbugzilla.redhat.com bug 1147282 in qemu "qemu vm guests hard shutdown unexpectedly" [Medium,New] - Assigned to virt-maint20:49
mriedembut it's a thing we hit upstream20:49
mriedemdansmith: kashyap: ^ not sure how that should be redirected20:50
openstackgerritjichenjc proposed openstack/nova: Change some wording on server_concepts.rst  https://review.openstack.org/24806320:50
mriedemthey had closed it b/c fc20 went EOL20:50
mriedemit showed up in a test_volume_boot_pattern ssh failure20:50
openstackgerritChuck Carmack proposed openstack/nova: REST API changes for user settable server description  https://review.openstack.org/25495020:52
*** suro-patz has quit IRC20:53
*** suro-patz has joined #openstack-nova20:54
*** haypo_ has quit IRC20:54
migidansmith, as for the second question - do you really think we should go across all the files and change the appropriate fields for consistency - if yes this is something I was proposing yesterday, but it was said the change will be too massive, so focus only of getting 100% running20:55
*** rook has quit IRC20:55
*** fawadkhaliq has quit IRC20:55
migiI am fine doing any way20:55
dansmithmigi: define "change the appropriate fields for consistency" more clearly20:56
*** suro-patz has quit IRC20:56
*** suro-patz has joined #openstack-nova20:57
*** suro-patz has quit IRC20:57
dansmithmigi: I think it would be ideal if you can make the test changes you need to survive the o.vo change, without changing any fields on the actual objects20:57
dansmithmigi: but I don't think you should change things you don't need to change20:57
miginot all of them, but just idea of how many (~50%) will require changes:20:58
dansmithmigi: meaning, I don't think you should include the image_ref='fake-uuid' change since it's not necessary (and you didn't)20:58
migi# grep image_ref unit/compute/test_compute.py|wc -l20:58
*** ljxiash has joined #openstack-nova20:58
dansmithright, but those are not needed right now, correct?20:58
*** suro-patz has joined #openstack-nova20:58
migino, but you gave -1 with comment those should be changed as well20:59
dansmithwell, no,20:59
dansmithI -1d because I thought that you were changing some and not others.. I was missing the difference between image_href passed to create and the image_ref property as set on the Instance object20:59
migiok see it's +1 now :)20:59
dansmithI was conflating the two into the latter20:59
migihaven't seen that20:59
migieverything is fine then21:00
dansmithit's +2 actually21:00
dansmithbut yeah :)21:00
dansmithmriedem: kashyap is probably a good person to chase that down21:00
*** thumpba has quit IRC21:02
*** ljxiash has quit IRC21:03
*** aginwala has quit IRC21:05
*** rk4n has quit IRC21:08
*** Sukhdev has quit IRC21:08
bauzasmeh, the py34 job is bad for logstash...21:09
*** aginwala has joined #openstack-nova21:09
sdaguedansmith: the gate fail rate for multinode is definitely falling, will know more in the morning what's still left on the fails21:09
*** zenoway has joined #openstack-nova21:09
sdaguethis python 34 unit test fail is problematic though - https://jenkins04.openstack.org/job/gate-nova-python34/3086/console21:11
*** liverpooler has quit IRC21:11
*** rfolco_ has joined #openstack-nova21:13
*** zenoway has quit IRC21:14
mriedembauzas: yeah, the py34 output is terrible for logstash21:15
mriedemsdague: the py34 failures on nova.tests.unit.network.test_neutronv2.TestNeutronv2.test_deallocate_for_instance_2* are known issues21:16
bauzasmriedem: just wonder how we could help that21:16
mriedemfix logging in py34?21:16
openstackLaunchpad bug 1521599 in OpenStack Compute (nova) "py34 unit tests fail randomly for network.test_neutronv2" [Medium,Confirmed]21:17
mriedemis the tracking bu21:17
*** zenoway has joined #openstack-nova21:18
sdaguemriedem: then where is the patch to put them in the blacklist?21:18
sdaguebecause leaving known racing unit tests enabled is just not ok21:18
mriedembecause it's not failing that often from what i can tell21:19
*** rfolco_ has quit IRC21:19
*** burt has joined #openstack-nova21:19
*** ianbrown__ has quit IRC21:19
mriedemthe blacklist is more for 'these just don't work with py34'21:19
mriedemat least that was my understanding21:19
sdagueracing unit tests aren't ok to keep enabled21:19
*** rook has joined #openstack-nova21:20
bauzasmriedem: IIUC, that's this builder which has a unicode problem https://github.com/openstack-infra/project-config/blob/552521848a4a856eb9bbdd2369fbf589e2a26bf4/jenkins/jobs/macros.yaml#L49021:20
*** claudiub has joined #openstack-nova21:20
bauzasmriedem: tbc, probably that one https://github.com/openstack/nova/blob/master/tox.ini#L5021:21
mriedemnot following - to blacklist those tests?21:22
mriedemi know where to do that21:22
mriedemwith a fail rate of 12 hits in 7 days, i didn't think it was high priority21:22
sdaguemriedem: http://tinyurl.com/pmx7cel - note, gate-nova-python27 has 0% failure rate in the gate, and that's where we really need to keep it21:22
*** ianbrown has joined #openstack-nova21:23
*** cdelatte has quit IRC21:23
bauzasmriedem: no, was speaking of the bad logging output sorry21:23
mriedembauzas: ok, so you think ostestr?21:23
bauzasmriedem: which creates a logstash problem21:23
bauzasmriedem: yeah21:23
*** smatzek has quit IRC21:23
mriedembauzas: we actually have a query for that one in e-r https://review.openstack.org/#/c/251868/21:23
mriedembut for some reason we don't have any hits21:23
sdaguemriedem: yeh, I think we should blacklist them, because they just cause people to go recheck grind21:23
mriedemhere http://status.openstack.org/elastic-recheck/index.html21:24
sdagueand any additional gate fails are really bad at our current load and fail level21:24
mriedembut i do have hits here http://logstash.openstack.org/#dashboard/file/logstash.json?query=message:%5C%22Client.delete_port('my_portid1')%5C%22%20AND%20build_name:%5C%22gate-nova-python34%5C%22%20AND%20tags:%5C%22console%5C%2221:24
bauzaswell, since the ouput is busted, I'm not really trusting logstash21:24
mriedemsdague: we could make that argument for just skipping test_volume_boot_pattern21:24
sdaguethis seems like a good one to just blacklist given that no one is fixing it, so it can be looked at without chewing up all the resources21:24
*** yarkot has joined #openstack-nova21:24
bauzasbecause the output is only one line21:24
sdagueunit tests should always pass21:24
bauzasfor the whole output21:24
bauzassdague: +121:24
sdaguewe control the whole system21:24
sdagueI think the bar has to be higher there21:25
mriedemblurg, that's fine, whatevs, i can push that up21:25
* mriedem kind of feels like he's being yelled at21:25
mriedembtw, this should fix a gate failure https://review.openstack.org/#/c/254428/21:25
mriedemit's actually 3-4 LOC for the fix21:26
mriedemalso i have a revert for gmr stuff here https://review.openstack.org/#/c/253821/21:26
mriedemthat we should probably get on21:26
*** ianbrown has quit IRC21:27
*** ianbrown has joined #openstack-nova21:28
*** dane-fichter has quit IRC21:28
*** doug-fis_ has joined #openstack-nova21:30
*** earlephilhower has quit IRC21:30
*** dims has joined #openstack-nova21:31
*** doug-fish has quit IRC21:32
*** vishwanathj has joined #openstack-nova21:32
dansmithsdague: it'd be kinda scary if sending 50% of the requests down the pipe to the wrong compute host _didn't_ make a big impact on the fails :)21:32
mriedemmtreinish: the ostestr blacklist takes test ids down to the test case, right?21:33
*** aginwala has quit IRC21:33
*** ianbrown_ has joined #openstack-nova21:33
*** ianbrown has quit IRC21:34
*** lbelivea has joined #openstack-nova21:34
*** apoorvad has joined #openstack-nova21:34
*** lbelivea has quit IRC21:35
*** lbeliveau has joined #openstack-nova21:35
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/25390121:35
*** xymind has joined #openstack-nova21:35
*** lbeliveau has quit IRC21:36
*** lbeliveau has joined #openstack-nova21:36
*** ianbrown has joined #openstack-nova21:41
*** Marga_ has quit IRC21:45
mtreinishmriedem: it's a regex match, it just constructs an exclude regex like what people were doing manually21:46
*** ljxiash has joined #openstack-nova21:47
mtreinishmriedem: https://github.com/openstack/os-testr/blob/master/os_testr/os_testr.py#L13621:47
*** cdelatte has joined #openstack-nova21:49
*** lpetrut has quit IRC21:50
*** salv-orlando has joined #openstack-nova21:50
*** ljxiash has quit IRC21:51
*** salv-orlando has quit IRC21:55
snetihi..can anyone suggest me how to run tests on python3.4. I get the below error:21:56
snetiInvocationError: '/opt/stack/nova/.tox/py34/bin/ostestr --blacklist_file tests-py3.txt'21:56
snetiERROR:   py34: commands failed21:56
mriedemsneti: tox -r -e py3421:56
mriedemas for why they fail, that would require more details21:57
*** rk4n has joined #openstack-nova21:58
snetimriedem: thank you. But I still get the same error with tox -r -e py3421:59
thomasemSo, trying unit tests on a 40 core box21:59
*** dims has quit IRC21:59
thomasemkeep seeing the API tests seeming to step on eachother since another test is running attached to
mtreinishsneti: can you paste the full traceback on paste.openstack.org21:59
*** suro-patz has quit IRC22:01
thomasemlets try this with concurrency=222:03
*** suro-patz has joined #openstack-nova22:04
snetimriedem: http://paste.openstack.org/show/481238/22:05
*** salv-orlando has joined #openstack-nova22:05
*** rk4n has quit IRC22:05
*** gokrokve has joined #openstack-nova22:05
*** rk4n has joined #openstack-nova22:06
*** rk4n has quit IRC22:06
*** jogo has quit IRC22:07
*** gokrokve has quit IRC22:07
*** gokrokve has joined #openstack-nova22:07
*** Marga_ has joined #openstack-nova22:08
*** vilobhmm11 has quit IRC22:10
*** david-ly_ has joined #openstack-nova22:10
*** vilobhmm11 has joined #openstack-nova22:10
mtreinishsneti: you ran python2 testr first: https://bugs.launchpad.net/testrepository/+bug/121290922:10
openstackLaunchpad bug 1212909 in Testrepository "times.dbm incompatible based on Python version it was created with" [Critical,Triaged]22:10
*** pratikma_ has joined #openstack-nova22:10
mtreinishsneti: you can wipe out your .testrepository directory and then re-run the tests22:10
mtreinishit'll make a new times.dbm file and it should work fine22:11
*** lykinsbd has quit IRC22:11
*** rk4n has joined #openstack-nova22:11
mtreinishsneti: if you run py34 tests first it creates a dbm file in a format that works on py2 and py322:11
mtreinishbut a py2 created one doesn't work on py3... :(22:12
*** jogo has joined #openstack-nova22:12
snetimtreinish, thank you! will try that22:12
*** david-lyle has quit IRC22:13
openstackgerritMatt Riedemann proposed openstack/nova: neutron: skip test_deallocate_for_instance_2* in py34 job  https://review.openstack.org/25497922:13
mriedemsdague: bauzas: ^22:13
*** david-ly_ is now known as david-lyle22:13
*** gonzalo2kx has joined #openstack-nova22:13
*** pratikmallya has quit IRC22:14
*** jichen has quit IRC22:14
*** pratikma_ has quit IRC22:15
*** belmoreira has joined #openstack-nova22:19
*** inc0 has quit IRC22:20
*** BadCub has left #openstack-nova22:20
*** ndipanov has quit IRC22:21
belmoreirabauzas: still here?22:22
belmoreiraI have some questions regarding the scheduler host_manager and a very weird behaviour22:22
*** ccarmack has quit IRC22:23
belmoreirawhen the scheduler starts it calls: _async_init_instance_info()22:23
belmoreirathis includes all deleted instances! Why they are needed?22:24
sdaguemriedem: thanks for that22:24
belmoreirain a cell with >200 nodes I have per node >6000 instances (deleted) registered in the DB22:26
*** e0ne has quit IRC22:26
*** dustins is now known as dustins|gone22:27
belmoreirathe scheduler is loading all this info when it starts (several GBs), breaking at some point22:27
*** gonzalo2kx has quit IRC22:28
*** mgoddard has quit IRC22:29
edleafebelmoreira: not sure why it's loading deleted instances22:30
*** changbl has quit IRC22:31
belmoreiraedleafe: I don't see why...22:32
edleafebelmoreira: it's just using the InstanceList.get_by_filters(), and I don't see anything about read_deleted=True22:32
*** thorst has quit IRC22:33
belmoreiraedleafe: it uses: instance_get_all_by_filters_sort22:33
*** rk4n has quit IRC22:33
openstackgerritSean Dague proposed openstack/nova: convert GlanceClientWrapper to endpoint  https://review.openstack.org/25482822:33
openstackgerritSean Dague proposed openstack/nova: retool xen glance plugin to work with urls  https://review.openstack.org/25433122:33
openstackgerritSean Dague proposed openstack/nova: drop _create_glance_client function  https://review.openstack.org/25482722:33
openstackgerritSean Dague proposed openstack/nova: Implement GlanceEndpoint object  https://review.openstack.org/25421422:33
openstackgerritSean Dague proposed openstack/nova: Clean up glance url handling  https://review.openstack.org/25421322:33
openstackgerritSean Dague proposed openstack/nova: always create clients with GlanceEndpoint  https://review.openstack.org/25430622:33
openstackgerritSean Dague proposed openstack/nova: Add upload/download vhd2 interfaces  https://review.openstack.org/25478522:33
belmoreiraand "Deleted instances will be returned by default, unless there's a filter that says otherwise"22:33
edleafebelmoreira: ugh22:34
belmoreiraedleafe: I just tested adding the filter: "deleted": False22:35
*** smurke has quit IRC22:35
belmoreirain _async_init_instance_info()22:35
*** pumaranikar has quit IRC22:35
*** otter768 has joined #openstack-nova22:35
belmoreiraand it didn't query deleted instances... However not sure about the consequences for the scheduler22:36
edleafebelmoreira: should deleted instances affect affinity decisions?22:36
*** thorst has joined #openstack-nova22:37
*** _ducttape_ has quit IRC22:38
*** rk4n has joined #openstack-nova22:38
openstackgerritMatt Riedemann proposed openstack/nova: Add a note about fixing "db type could not be determined" with py34  https://review.openstack.org/25498922:38
belmoreiraedleafe: previous scheduling decisions (on deleted instances) should not influence new ones. At least I'm not seeing any use case22:39
*** yarkot has quit IRC22:39
openstackgerritMatt Riedemann proposed openstack/nova: Add a note about fixing "db type could not be determined" with py34  https://review.openstack.org/25498922:39
mriedemmtreinish: ^ you're my inspiration22:39
*** gokrokve has quit IRC22:39
vilobhmm11alaski : https://blueprints.launchpad.net/nova/+spec/flavor-classes-unifying-flavors22:40
mtreinishmriedem: heh, ok22:40
edleafebelmoreira: the only one I can think of is if there was an anti-affinity decison that placed the new instance on a host with a deleted instance, and then that instance is undeleted22:40
*** otter768 has quit IRC22:40
*** thorst has quit IRC22:41
*** rlrossit has quit IRC22:42
*** aginwala has joined #openstack-nova22:42
edleafebelmoreira: can you enter a bug for this and assign it to me?22:42
edleafebelmoreira: I have to run to pick up the kid, but I'll discuss with bauzas tomorrow.22:42
*** pratikmallya has joined #openstack-nova22:43
belmoreirayes, I'm filling the bug22:43
*** zenoway has quit IRC22:43
belmoreiraedleafe: ping me as well22:44
openstackgerritCale Rath proposed openstack/nova: Purge soft-deleted instances cmd  https://review.openstack.org/20375122:45
*** gokrokve has joined #openstack-nova22:49
*** gokrokve has quit IRC22:50
*** sneti has quit IRC22:52
*** gokrokve has joined #openstack-nova22:53
*** ctrath has quit IRC22:53
*** zenoway has joined #openstack-nova22:53
*** salv-orlando has quit IRC22:53
*** salv-orlando has joined #openstack-nova22:54
*** ianbrown_ has quit IRC22:55
*** gokrokve_ has joined #openstack-nova22:55
*** gokrokve has quit IRC22:56
*** eharney has quit IRC22:56
*** cp16net has quit IRC22:56
*** samueldmq has quit IRC22:56
*** ianbrown_ has joined #openstack-nova22:57
*** devananda has quit IRC22:57
*** hblixt has quit IRC22:58
openstackgerritSean Dague proposed openstack/nova: convert GlanceClientWrapper to endpoint  https://review.openstack.org/25482822:58
openstackgerritSean Dague proposed openstack/nova: retool xen glance plugin to work with urls  https://review.openstack.org/25433122:58
openstackgerritSean Dague proposed openstack/nova: drop _create_glance_client function  https://review.openstack.org/25482722:58
openstackgerritSean Dague proposed openstack/nova: Implement GlanceEndpoint object  https://review.openstack.org/25421422:58
openstackgerritSean Dague proposed openstack/nova: Clean up glance url handling  https://review.openstack.org/25421322:58
openstackgerritSean Dague proposed openstack/nova: always create clients with GlanceEndpoint  https://review.openstack.org/25430622:58
openstackgerritSean Dague proposed openstack/nova: Add upload/download vhd2 interfaces  https://review.openstack.org/25478522:58
*** burgerk has quit IRC22:58
*** ctrath has joined #openstack-nova22:59
*** suro-patz has quit IRC23:00
*** apoorvad has quit IRC23:02
*** rook has quit IRC23:02
*** thorst has joined #openstack-nova23:02
*** ianbrown has quit IRC23:02
*** gokrokve_ has quit IRC23:03
mriedemman, sdague burning the early evening oil23:04
sdagueyeh, well, some of those patches need xenserver to run on them, so I wanted to post them before leaving the terminal for the day23:04
sdaguemriedem: an unrelated and hopefully not too yucky one I'd like in to unblock some clarifying patches in api_samples - https://review.openstack.org/#/c/254909/23:05
sdaguewhich I've got in a branch locally and will push tomorrow23:05
mriedem"This is terrible."23:06
mriedemi don't think i can grok that one tonight, have to leave soon23:07
*** thorst has quit IRC23:07
*** xyang1 has quit IRC23:08
*** mriedem is now known as mriedem_away23:09
openstackgerritEric Brown proposed openstack/nova: Fix config option types  https://review.openstack.org/16848123:13
*** mkoderer has quit IRC23:15
*** stackdump has joined #openstack-nova23:15
openstackgerritEric Brown proposed openstack/nova: Fix config option types  https://review.openstack.org/16848123:15
openstackgerritJim Rollenhagen proposed openstack/nova: Allow virt driver to define binding:host_id  https://review.openstack.org/19441323:16
*** mkoderer has joined #openstack-nova23:17
*** baoli has quit IRC23:18
*** edmondsw has quit IRC23:21
*** mdrabe has quit IRC23:23
*** ijuwang has quit IRC23:24
*** rook has joined #openstack-nova23:28
*** burgerk has joined #openstack-nova23:29
*** suro-patz has joined #openstack-nova23:31
*** baoli has joined #openstack-nova23:32
*** pradk has quit IRC23:35
*** baoli has quit IRC23:35
*** hblixt has joined #openstack-nova23:35
*** Sukhdev has joined #openstack-nova23:35
*** baoli has joined #openstack-nova23:36
*** burt has quit IRC23:36
*** alex_xu has quit IRC23:36
*** aginwala has quit IRC23:37
*** Daisy has joined #openstack-nova23:41
*** alex_xu has joined #openstack-nova23:41
*** Daisy has quit IRC23:41
*** Daisy has joined #openstack-nova23:42
*** pece has quit IRC23:44
notmorganbrought this up in -infra but for XenProject CI check  on change https://review.openstack.org/#/c/253792/ a "failed" run produces no logs23:44
notmorganBobBall: ^ cc23:44
notmorgansdague: mriedem_away ^ cc as well23:44
notmorgani am getting a 404 instead of something useful.23:45
notmorgansdague:, mriedem_away, mordred: for what it's worth https://review.openstack.org/#/c/253792/ is really just in transient (unrelated) error states now afaict.23:45
notmorganeverything is looking good.23:45
notmorganBobBall: sorry didn't mean to ping you, was mis-informed who ran the xen one.23:46
*** mrkz has quit IRC23:48
*** Yingxin_ has joined #openstack-nova23:53
*** baoli has quit IRC23:54
*** jerrygb has quit IRC23:55
*** aginwala has joined #openstack-nova23:55
*** jerrygb has joined #openstack-nova23:56
*** burgerk has quit IRC23:56
openstackgerritTracy Jones proposed openstack/nova: config options: centralize section "guestfs"  https://review.openstack.org/24896323:56
*** Daisy has quit IRC23:56
*** Yingxin has quit IRC23:56
*** Yingxin_ is now known as Yingxin23:56
*** takedakn has joined #openstack-nova23:56
*** Daisy has joined #openstack-nova23:57
*** andrewbogott has joined #openstack-nova23:57
*** smoriya has joined #openstack-nova23:57
*** andrewbogott has quit IRC23:58
*** andrewbogott has joined #openstack-nova23:58
*** zenoway has quit IRC23:58

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!