Tuesday, 2015-07-07

*** dannywilson has quit IRC00:00
*** dims has joined #openstack-nova00:01
*** hemna is now known as hemnafk00:03
*** ijw has joined #openstack-nova00:05
*** piyanai has quit IRC00:09
*** Sukhdev has quit IRC00:10
*** IanGovett has quit IRC00:11
*** shyu has joined #openstack-nova00:12
*** r-daneel has quit IRC00:13
*** baoli has quit IRC00:14
*** baoli has joined #openstack-nova00:15
*** rfolco has joined #openstack-nova00:18
*** asettle has joined #openstack-nova00:20
*** asettle has quit IRC00:20
*** asettle has joined #openstack-nova00:21
*** tsekiyama has quit IRC00:26
*** SourabhP has quit IRC00:27
*** SourabhP has joined #openstack-nova00:31
*** zzzeek has quit IRC00:35
*** promulo__ has joined #openstack-nova00:44
*** ssurana has quit IRC00:46
*** ssurana has joined #openstack-nova00:46
*** promulo has quit IRC00:47
*** dims has quit IRC00:50
*** ssurana has quit IRC00:50
*** fawadkhaliq has joined #openstack-nova00:53
*** wanghua has joined #openstack-nova00:53
*** jyuso1 has quit IRC00:54
*** jyuso1 has joined #openstack-nova00:54
*** fawadkhaliq has quit IRC00:56
*** dims has joined #openstack-nova00:58
*** nelsnels_ has joined #openstack-nova00:59
*** wanghua has quit IRC00:59
*** nelsnels_ has quit IRC00:59
*** wanghua has joined #openstack-nova00:59
*** humble_ has joined #openstack-nova01:00
*** nelsnelson has quit IRC01:02
*** zhenguo has joined #openstack-nova01:03
*** wanghua has quit IRC01:03
*** tongli has joined #openstack-nova01:09
*** erlon has quit IRC01:10
openstackgerritGhanshyam Mann proposed openstack/nova: Merge virtual interface tests between v2 and v2.1  https://review.openstack.org/19781101:10
openstackgerritZhenyu Zheng proposed openstack/nova: CONF.allow_resize_to_same_host should check only once in controller  https://review.openstack.org/18623601:10
openstackgerritZhenyu Zheng proposed openstack/nova: CONF.allow_resize_to_same_host should check only once in controller  https://review.openstack.org/18623601:11
*** ijw has quit IRC01:12
*** jasondotstar has quit IRC01:12
*** achanda has joined #openstack-nova01:13
*** tongli_ has joined #openstack-nova01:13
*** tongli has quit IRC01:14
openstackgerritGhanshyam Mann proposed openstack/nova-specs: Spec to add vif 'net-id' in virtual interfaces API  https://review.openstack.org/19862201:16
*** piyanai has joined #openstack-nova01:16
*** annashen has joined #openstack-nova01:17
*** andreykurilin__ has quit IRC01:17
openstackgerritZhenyu Zheng proposed openstack/nova: DB:support for migration objects multi-statuses  https://review.openstack.org/16960401:17
*** tongli_ has quit IRC01:18
*** Longgeek has joined #openstack-nova01:18
*** jwcroppe has quit IRC01:21
openstackgerritZhenyu Zheng proposed openstack/nova: add support for muti_statuses_query from migration  https://review.openstack.org/16963101:21
openstackgerritZhenyu Zheng proposed openstack/nova: DB:support for migration objects multi-statuses  https://review.openstack.org/16960401:21
*** annashen has quit IRC01:22
*** jwcroppe has joined #openstack-nova01:22
*** mtanino has quit IRC01:22
*** SourabhP has quit IRC01:24
*** jasondotstar has joined #openstack-nova01:25
*** jwcroppe has quit IRC01:26
*** slong has quit IRC01:31
*** slong has joined #openstack-nova01:31
*** andrewbogott is now known as andrewbogott_afk01:36
*** patrickeast has quit IRC01:38
*** lyanchih has joined #openstack-nova01:39
*** tongli has joined #openstack-nova01:43
openstackgerritGhanshyam Mann proposed openstack/nova: Removing extension "OS-EXT-VIF-NET" from v2.1 extension-list  https://review.openstack.org/19893401:44
*** asettle has quit IRC01:45
*** tongli has quit IRC01:47
*** Longgeek has quit IRC01:51
openstackgerritJim Rollenhagen proposed openstack/nova: Ironic: block storage support  https://review.openstack.org/18465201:51
*** tonytan4ever has joined #openstack-nova01:52
*** ajayaa has joined #openstack-nova01:58
*** kumar has joined #openstack-nova01:58
*** patrickeast has joined #openstack-nova02:08
*** lyanchih has quit IRC02:09
*** yamahata has quit IRC02:10
openstackgerritliaonanhai proposed openstack/nova-specs: Add user_id info: add 'user_id' to REST os-simple-tenant-usage output  https://review.openstack.org/19893702:11
*** mutoulbj has joined #openstack-nova02:12
openstackgerritRui Chen proposed openstack/nova-specs: CreateVM supports subnet specified  https://review.openstack.org/16963802:12
*** arnaud____ has joined #openstack-nova02:13
openstackgerritRui Chen proposed openstack/nova-specs: Split network plane for live migration  https://review.openstack.org/19499002:14
*** annashen has joined #openstack-nova02:18
*** arnaud____ has quit IRC02:21
*** jskinner has joined #openstack-nova02:22
*** annashen has quit IRC02:22
*** dboik_ has joined #openstack-nova02:24
*** setuid has quit IRC02:25
*** mutoulbj_ has joined #openstack-nova02:27
*** mutoulbj has quit IRC02:30
*** otter768 has quit IRC02:30
*** dboik_ has quit IRC02:31
*** pixelbeat has joined #openstack-nova02:33
*** pixelbeat_ has quit IRC02:34
*** setuid has joined #openstack-nova02:35
alex_xugmann: hi, can you reach oomichi, I have question for him02:36
gmannalex_xu: yea, ll ask him to contact you.02:37
*** oomichi has joined #openstack-nova02:37
alex_xugmann: thanks02:37
oomichialex_xu: morning02:37
alex_xuoomichi: hey, morning :)02:38
alex_xuoomichi: I'm think if this get agreement https://review.openstack.org/196918 , what we can do for nova and ironic02:38
*** pixelbeat has quit IRC02:38
alex_xuoomichi: the microversions already released, we can't just remove the old header02:38
oomichialex_xu: yeah, nice question.02:39
oomichialex_xu: I think we need to change the existing header as the deprecated and run with the double header in short term.02:39
alex_xuoomichi: emm...sounds like the only way, I guess at least one release cycle02:40
oomichialex_xu: or we will rename it without any depreacated header because microversion still is baby, and actually nobody(novaclient, tempest, etc.) is using microversion.02:41
oomichialex_xu: but that is not acceptable maybe,  because ext-nif extension should be also microversion as the conclusion of gmann patch.02:41
alex_xuoomichi: I guess this will be argument. as we discussion ext-vif-info, we think the v2.1 already released02:41
alex_xuoomichi: yea02:41
oomichialex_xu: yeah, that is the same as me :)02:42
oomichialex_xu: so, the only way is the deprecation ;)02:42
alex_xuoomichi: yea, at least there is something get agreement, thanks for push that api-wg guideline :)02:43
oomichialex_xu: now many projects(heat, magnum, neutron, etc.) start considering to use microversions. and nice guideline of microversions will help them.02:43
alex_xuoomichi: yea, ture02:43
oomichialex_xu: even if the guideline is against current implementation of nova and ironic.02:43
alex_xuoomichi: yea, agree02:44
alex_xuoomichi: btw, do you know when tempest support microversion?02:44
alex_xuany plan?02:44
oomichialex_xu: I have a qa-spec for microversions: https://review.openstack.org/#/c/169126/02:45
oomichialex_xu: but I am thinking the microversions tests of Tempest should be based on the guideline of API-WG02:45
*** tongli has joined #openstack-nova02:46
oomichialex_xu: because we need to test microversions with consistent manner between projects.02:46
alex_xuoomichi: but I still think the API-WG guideline need very long time02:46
oomichialex_xu: yeah, right.02:46
alex_xuoomichi: why we can't depend on current implementation?02:47
openstackgerritliaonanhai proposed openstack/nova: Add user_id info: add 'user_id' to REST os-simple-tenant-usage output  https://review.openstack.org/15152202:47
oomichialex_xu: for example, API header name. I'd like to implement the framework of microversions in Tempest and the framework can generate API header based on the same manner.02:48
oomichialex_xu: but now, we cannot get a consensus about that.02:48
oomichialex_xu: so if implementing the separated manner, we can do. but if doing that, we cannot detect the inconsistency between projects by Tempest.02:49
oomichialex_xu: I think Tempest is good tool for detecting this kind of inconsistency, and I hope we implement the same framework on Tempest.02:50
jrollthe primary goal is to test if the code is correct and the API is working, right?02:50
jrollso I think we should do that first02:50
jrolland then we can talk about testing headers and consistency between projects02:50
oomichijroll: yeah, right. we need to consider the balance between them.02:50
openstackgerritZhenyu Zheng proposed openstack/nova: resize-revert can break anti-affinity policy  https://review.openstack.org/16011002:50
jrollthe goal of tempest isn't to test if API guidelines are followed, it's to ensure an API works as expected.02:51
alex_xuoomichi: if we have consistent header, we still need test the old one?02:51
*** iamjarvo has joined #openstack-nova02:51
openstackgerritZhenyu Zheng proposed openstack/nova: add support for muti_statuses_query from migration  https://review.openstack.org/16963102:51
alex_xuas we said, there is deprecated time for old header02:51
openstackgerritZhenyu Zheng proposed openstack/nova: resize-revert can break anti-affinity policy  https://review.openstack.org/16011002:51
*** patrickeast_ has joined #openstack-nova02:51
*** patrickeast has quit IRC02:51
*** patrickeast_ is now known as patrickeast02:51
oomichijroll: so if the progress of API-WG is slow, I agree with that we move forward without api-wg consensus.02:51
jrolloomichi: well, microversions have been untested in master for months, is that slow enough?02:52
jroll(untested by tempest)02:52
oomichialex_xu: I don't feel we need to test old header if we can get a consensus of consistent header name.02:53
*** hakimo has quit IRC02:54
oomichijroll: umm, yeah, I can agree. I will update the qa-spec at current situation for moving forward.02:55
*** hakimo has joined #openstack-nova02:55
alex_xuoomichi: let me know if you need any help02:55
oomichialex_xu: thanks, your help is always helpful ;-)02:55
alex_xuoomichi: :)02:56
openstackgerritEric Brown proposed openstack/nova: Replace openssl calls with cryptography lib  https://review.openstack.org/19824602:57
*** arnaud____ has joined #openstack-nova02:57
*** tonytan4ever has quit IRC02:58
openstackgerritGhanshyam Mann proposed openstack/nova: Removing extension "OS-EXT-VIF-NET" from v2.1 extension-list  https://review.openstack.org/19893402:58
*** iamjarvo has quit IRC03:01
openstackgerritZhenyu Zheng proposed openstack/nova: Add instance action events for live migration  https://review.openstack.org/19537203:02
openstackgerritliaonanhai proposed openstack/nova-specs: Add user_id info: add 'user_id' to REST os-simple-tenant-usage output  https://review.openstack.org/19893703:03
openstackgerritRong Han proposed openstack/nova: Catch NovaException from post_live_migration_at_destination, set instance vm_state error.  https://review.openstack.org/18663403:12
*** Longgeek has joined #openstack-nova03:18
*** annashen has joined #openstack-nova03:19
*** jwcroppe has joined #openstack-nova03:21
*** annashen has quit IRC03:23
openstackgerritZhenyu Zheng proposed openstack/nova: Ironic:check the configuration item api_max_retries  https://review.openstack.org/12331903:24
*** asettle has joined #openstack-nova03:28
*** Longgeek has quit IRC03:29
*** lyanchih has joined #openstack-nova03:29
*** kiwik-chenrui has joined #openstack-nova03:36
*** RuiChen has quit IRC03:37
*** dims has quit IRC03:39
*** piyanai has quit IRC03:42
*** piyanai has joined #openstack-nova03:43
*** piyanai has quit IRC03:43
*** slong has quit IRC03:44
*** piyanai has joined #openstack-nova03:44
*** slong has joined #openstack-nova03:44
*** piyanai has quit IRC03:48
*** arnaud____ has quit IRC03:54
*** ssurana has joined #openstack-nova03:58
*** tongli has quit IRC04:08
*** deepthi has joined #openstack-nova04:10
*** lyanchih has quit IRC04:12
*** arnaud__ has joined #openstack-nova04:14
gmannalex_xu: hi04:14
gmannalex_xu: regarding describing removal of extension entry in https://review.openstack.org/#/c/198622/04:15
gmannalex_xu: as we need to remove extension entry on base v2.1 and backport to kilo, I think we should mention reentry of same in that microversion spec instead of describing extension removal04:16
*** annashen has joined #openstack-nova04:19
*** jskinner has quit IRC04:22
*** xyhuang has joined #openstack-nova04:22
*** annashen has quit IRC04:24
*** xyang1 has quit IRC04:24
*** kumar has quit IRC04:25
*** otter768 has joined #openstack-nova04:31
*** Marga_ has joined #openstack-nova04:32
*** Marga_ has quit IRC04:32
*** Marga_ has joined #openstack-nova04:33
*** otter768 has quit IRC04:36
*** Sukhdev has joined #openstack-nova04:38
*** liusheng has quit IRC04:38
*** irenab has quit IRC04:38
gmannalex_xu: oomichi : as I am thinking to remove VIF extension from v2.1 extension list as bug fix as in -https://review.openstack.org/#/c/198934/04:39
*** liusheng has joined #openstack-nova04:39
gmannalex_xu: oomichi : should we mention about extension removal in mivroversion spec - https://review.openstack.org/#/c/198622/04:39
openstackgerritGhanshyam Mann proposed openstack/nova: Removing extension "OS-EXT-VIF-NET" from v2.1 extension-list  https://review.openstack.org/19893404:40
*** dims has joined #openstack-nova04:40
oomichigmann: yeah, it will be nice to mention about extension removal in the mivroversion spec04:40
*** dims_ has joined #openstack-nova04:41
*** baoli has quit IRC04:43
gmannoomichi: ok, i will mention that in problem section as those will be fixed separately not in that microversion04:44
*** dims has quit IRC04:45
oomichigmann: thanks :)04:46
*** dims_ has quit IRC04:46
*** ajayaa has quit IRC04:47
*** shihanzhang has quit IRC04:48
*** shihanzhang has joined #openstack-nova04:49
*** dave-mccowan has quit IRC04:52
openstackgerritGhanshyam Mann proposed openstack/nova-specs: Spec to add vif 'net-id' in virtual interfaces API  https://review.openstack.org/19862204:57
*** nihilifer has joined #openstack-nova04:59
*** lpetrut has joined #openstack-nova05:02
*** slong has quit IRC05:03
*** slong has joined #openstack-nova05:03
*** SourabhP has joined #openstack-nova05:04
*** prashup has joined #openstack-nova05:04
*** dave-mccowan has joined #openstack-nova05:05
*** SourabhP_ has joined #openstack-nova05:09
*** SourabhP has quit IRC05:11
*** SourabhP_ is now known as SourabhP05:11
openstackgerritFeodor Tersin proposed openstack/nova: Fix collection of metadata for a snapshot of a volume-backed instance  https://review.openstack.org/17024305:11
openstackgerritFeodor Tersin proposed openstack/nova: Simplify interface for creating snapshot of volume-backed instance  https://review.openstack.org/19895805:11
openstackgerritFeodor Tersin proposed openstack/nova: Consolidate initialization of instance snapshot metadata  https://review.openstack.org/19895905:11
*** patrickeast has quit IRC05:13
openstackgerritEric Brown proposed openstack/nova: Replace openssl calls with cryptography lib  https://review.openstack.org/19824605:15
*** exploreshaifali has joined #openstack-nova05:17
*** sahid has joined #openstack-nova05:17
*** ssurana has quit IRC05:18
*** Marga_ has quit IRC05:18
*** Marga_ has joined #openstack-nova05:19
*** garyk has joined #openstack-nova05:19
*** annashen has joined #openstack-nova05:21
*** sudipto has joined #openstack-nova05:21
*** achanda has quit IRC05:23
*** dannywilson has joined #openstack-nova05:23
*** nkrinner has joined #openstack-nova05:24
*** ildikov has joined #openstack-nova05:24
*** dannywilson has quit IRC05:25
*** annashen has quit IRC05:25
*** dannywilson has joined #openstack-nova05:26
*** tojuvone has joined #openstack-nova05:26
*** Longgeek has joined #openstack-nova05:26
*** ajayaa has joined #openstack-nova05:28
*** irenab has joined #openstack-nova05:28
*** exploreshaifali has quit IRC05:28
*** _gryf|afk is now known as _gryf05:30
*** dave-mccowan has quit IRC05:31
*** gmann has quit IRC05:33
*** iamjarvo has joined #openstack-nova05:36
*** iamjarvo has quit IRC05:37
*** iamjarvo has joined #openstack-nova05:37
*** Murali has joined #openstack-nova05:39
*** Nikolay_St has quit IRC05:40
*** lyanchih has joined #openstack-nova05:40
*** unicell has joined #openstack-nova05:44
*** yamahata has joined #openstack-nova05:50
*** ig0r_ has joined #openstack-nova05:51
*** Longgeek has quit IRC05:52
*** wanghua has joined #openstack-nova05:52
*** humble_ has quit IRC05:53
*** humble_ has joined #openstack-nova05:55
*** yingjun has joined #openstack-nova05:55
*** ig0r__ has quit IRC05:55
*** Longgeek has joined #openstack-nova05:56
*** unicell has quit IRC05:56
*** gibi has joined #openstack-nova05:58
*** wanghua has quit IRC05:58
*** unicell has joined #openstack-nova05:58
openstackgerritAbhishek Kekane proposed openstack/nova: libvirt: Kill rsync/scp processes before deleting instance  https://review.openstack.org/19298605:59
*** kumar has joined #openstack-nova05:59
*** arnaud__ has quit IRC06:01
*** kiran-r has joined #openstack-nova06:02
*** lpetrut has quit IRC06:04
openstackgerritSudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors  https://review.openstack.org/19897206:06
*** eglynn-afk has joined #openstack-nova06:08
*** jichen has joined #openstack-nova06:09
openstackgerritRong Han proposed openstack/nova: Catch NovaException from post_live_migration_at_destination, set instance vm_state error.  https://review.openstack.org/18663406:15
*** gmann has joined #openstack-nova06:18
*** lpetrut has joined #openstack-nova06:19
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex  https://review.openstack.org/19721306:21
*** annashen has joined #openstack-nova06:21
openstackgerritwanghao proposed openstack/nova: Image/block device mapping combination is not valid  https://review.openstack.org/17198406:26
*** annashen has quit IRC06:26
*** radek__ has joined #openstack-nova06:28
*** moshele has joined #openstack-nova06:29
*** armax has quit IRC06:29
openstackgerritEric Brown proposed openstack/nova: Replace openssl calls with cryptography lib  https://review.openstack.org/19824606:29
*** otter768 has joined #openstack-nova06:32
*** Sukhdev has quit IRC06:34
*** cfriesen has quit IRC06:35
*** otter768 has quit IRC06:37
*** leseb has quit IRC06:38
*** eglynn-afk has quit IRC06:38
*** belmoreira has joined #openstack-nova06:43
*** fifieldt has joined #openstack-nova06:43
*** leseb has joined #openstack-nova06:44
*** asettle has quit IRC06:45
*** scheuran has joined #openstack-nova06:45
*** sahid has quit IRC06:49
*** sahid has joined #openstack-nova06:49
*** afazekas has joined #openstack-nova06:50
*** hshiina has joined #openstack-nova06:58
*** mjkang has joined #openstack-nova07:01
*** eglynn-afk has joined #openstack-nova07:03
*** Marga__ has joined #openstack-nova07:03
*** dannywilson has quit IRC07:04
*** Marga_ has quit IRC07:06
*** SourabhP has quit IRC07:07
*** pbandzi has joined #openstack-nova07:07
*** SourabhP has joined #openstack-nova07:09
*** mpavone has joined #openstack-nova07:10
*** iamjarvo has quit IRC07:12
*** slong has quit IRC07:17
*** oomichi has quit IRC07:22
*** annashen has joined #openstack-nova07:22
*** belmoreira has quit IRC07:26
*** annashen has quit IRC07:27
*** Nikolay_St has joined #openstack-nova07:32
*** browne has quit IRC07:33
*** deshipu has joined #openstack-nova07:33
*** moshele has quit IRC07:36
*** jistr has joined #openstack-nova07:37
*** kiwik-chenrui has quit IRC07:37
*** RuiChen has joined #openstack-nova07:38
*** lpetrut has quit IRC07:40
*** dims has joined #openstack-nova07:43
*** dims has quit IRC07:48
*** lpetrut has joined #openstack-nova07:48
*** hshiina has quit IRC07:53
*** SourabhP has quit IRC07:55
*** dguitarbite has joined #openstack-nova07:55
*** SourabhP has joined #openstack-nova07:58
*** lucasagomes has joined #openstack-nova08:02
*** exploreshaifali has joined #openstack-nova08:02
*** mjkang has quit IRC08:02
*** karimb has joined #openstack-nova08:05
*** SourabhP has quit IRC08:06
*** eliqiao has joined #openstack-nova08:06
eliqiaosahid ping08:07
*** gszasz has joined #openstack-nova08:09
sahidhellot eliqiao08:09
openstackgerritheijlong proposed openstack/nova: Fix the typo from _pre_upgrade_294 to _pre_upgrade_295 for tests/unit/\ db/test_migration  https://review.openstack.org/19900608:10
*** hshiina has joined #openstack-nova08:11
*** SourabhP has joined #openstack-nova08:12
*** haypo has joined #openstack-nova08:12
eliqiaosahid, for https://review.openstack.org/#/c/165838 , does it on your plan?08:12
openstackgerritheijlong proposed openstack/nova: Fix the typo from _pre_upgrade_294 to _pre_upgrade_295 for tests/unit/db/test_migration  https://review.openstack.org/19900608:13
*** ZZelle has joined #openstack-nova08:13
*** SourabhP_ has joined #openstack-nova08:15
sahideliqiao: yes i'm currently working o this i should push something soon08:16
*** SourabhP has quit IRC08:16
*** SourabhP_ is now known as SourabhP08:16
*** derekh has joined #openstack-nova08:17
*** ildikov has quit IRC08:17
eliqiaosahid: cool, I am interested on it. and have you look at https://launchpad.net/bugs/1243306, this may related08:19
openstackLaunchpad bug 1243306 in OpenStack Compute (nova) "consoleauth cannot be run in HA configuration without external memcache" [Wishlist,In progress] - Assigned to Eli Qiao (taget-9)08:19
sahideliqiao: yes interesting - but since the memcached support will probably be depretaced in future release i do not know if it make sense to fix this part08:21
sahidthe database effort will convert HA08:21
sahidcover08:21
*** jcoufal has joined #openstack-nova08:22
sahidabout what i know to memcached there is no way to do HA from the server itself, we should ask clients to store key in every memcached08:23
*** annashen has joined #openstack-nova08:23
sahidhum i think netflix push something like a proxy to do that work for memcached... let me see08:24
eliqiaosahid: hmm.. if all ha node use one memcache server , would that be okay?08:25
sahidyeah https://github.com/Netflix/EVCache08:25
sahideliqiao: well i do not think so :)08:25
eliqiaosahid: okay, then, we will abandon memcache after one release ?  thanks for addone 'console_tokens_backend'08:26
*** mpaolino has joined #openstack-nova08:28
*** annashen has quit IRC08:28
eliqiaosahid:  besides, I have some outdated patches on adding new console model, it 's start from https://review.openstack.org/#/c/177078/ , maybe useful for reference if you like.08:28
*** aloga_ has joined #openstack-nova08:28
sahideliqiao: yes we will probably abandon memcached as you pointed it's difficult to do HA with it. the console_tokens_backend has been introduced to perhaps see the support of other persistent storages like redis or cassandra...08:29
openstackgerritRoman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down  https://review.openstack.org/18408608:29
sahideliqiao: thank you08:30
eliqiaosahid: np. thanks for pushing that spec too  :)08:30
*** eliqiao has left #openstack-nova08:30
garykBobBall: around?08:30
*** eliqiao has joined #openstack-nova08:32
*** ildikov has joined #openstack-nova08:32
*** hshiina has quit IRC08:33
*** otter768 has joined #openstack-nova08:33
*** SourabhP has quit IRC08:36
*** eglynn-afk is now known as eglynn08:37
*** otter768 has quit IRC08:38
openstackgerritChangBo Guo(gcb) proposed openstack/nova: doc: update threading.rst  https://review.openstack.org/19901308:38
*** eliqiao has left #openstack-nova08:38
*** kumar has quit IRC08:39
*** yassine_ has joined #openstack-nova08:40
*** sambetts has quit IRC08:44
*** sambetts has joined #openstack-nova08:45
openstackgerritAndreas Scheuring proposed openstack/nova: Add macvtap as new virtual interface (vif) type to Nova's libvirt driver.  https://review.openstack.org/18228308:57
bauzasGood (UGT) Morning, Nova !08:57
*** moshele has joined #openstack-nova09:07
*** boris-42 has quit IRC09:22
*** annashen has joined #openstack-nova09:24
*** annashen has quit IRC09:29
openstackgerritSergey Vilgelm proposed openstack/nova: Switch to oslo.policy  https://review.openstack.org/19902009:30
*** yingjun has quit IRC09:30
BobBallgaryk: Just09:32
BobBall:)09:32
*** pbandzi has quit IRC09:33
*** pbandzi has joined #openstack-nova09:34
garykBobBall: other than wanting to know how you are, i'd also like to know why the following is failing: https://review.openstack.org/198702 with xen?09:36
BobBallI'm good thanks09:37
BobBallYourself?09:37
BobBallInteresting failure there...09:38
BobBallhttp://jenkins.openstack.xenproject.org/job/dsvm-tempest-xen/6047/console09:40
BobBall12:43:25 Cloning into 'devstack-gate'...09:40
BobBall12:45:26 fatal: read error: Connection reset by peer09:40
BobBallSeems like networking was OK at the start of the job but then immediately died09:40
BobBallOne issue is that because networking had died we couldn't upload the logs09:41
*** dims has joined #openstack-nova09:44
*** dims_ has joined #openstack-nova09:45
*** dims__ has joined #openstack-nova09:46
*** pixelbeat has joined #openstack-nova09:47
*** dims has quit IRC09:48
*** ociuhandu has joined #openstack-nova09:49
garykBobBall: all is well thanks09:49
*** dims_ has quit IRC09:50
*** dims__ has quit IRC09:50
BobBallglad to hear it!09:53
*** ZZelle has quit IRC09:55
*** piyanai has joined #openstack-nova10:07
sudiptobauzas, hello, one nova objects query - minor one...may i ask?10:07
bauzassudipto: yup, sure10:07
sudiptobauzas, i have changed one of the fields in the MonitorMetric object here: https://review.openstack.org/#/c/198972/ Now the field for 'value' will accept a MonitorMetricValue object...10:09
sudiptobauzas, So i am guessing i will have to define a obj_relationships dict for this? Can you basically given me an idea on what might be the object related changes i would have to make for a field change?10:09
bauzassudipto: the main problem is that you're changing a field10:10
openstackgerritRadoslav Gerganov proposed openstack/nova-specs: Update spec to match implementation  https://review.openstack.org/19903310:10
bauzassudipto: so, you need to provide a backwards compatible object10:10
sudiptobauzas, ok...10:10
openstackgerritChris Gacsal proposed openstack/nova: virt: Allow multiple volume attachments  https://review.openstack.org/19313510:11
openstackgerritChris Gacsal proposed openstack/nova: comp: Allow multiple volume attachments  https://review.openstack.org/19313410:11
openstackgerritChris Gacsal proposed openstack/nova: WIP: REST API changes to support multi-attach  https://review.openstack.org/15303810:11
bauzassudipto: if something is using object 1.0 and then providing an integer for the 1.1 object, it will create an AttributeError10:12
sudiptobauzas, yeah10:12
sudiptobauzas, so we need to some magic in the obj_make_compatible method?10:12
sudipto*do10:12
bauzassudipto: exactly10:13
*** claudiub has joined #openstack-nova10:13
sudiptobauzas, ok - also - what is the significance of the obj_relationships dict?10:13
sudiptobauzas, as in - how is it defined?10:13
bauzassudipto: that's for nested objects10:13
openstackgerritClaudiu Belu proposed openstack/nova: Adds Hyper-V Cluster implementation  https://review.openstack.org/19903710:14
* sudipto nova object novice10:14
*** IanGovett has joined #openstack-nova10:14
bauzassudipto: ie. you want to know the versions of the nested objects when you want to hydrate back the dict into an object10:14
sudiptobauzas, i have a 'value' field that accepts another object.10:14
bauzassudipto: oh yeah, that, I agree10:14
sudiptobauzas, ah got it...10:15
sudiptobauzas, so it's like saying - when we want to hydrate object 1.1 - for x field - take the version 1.X for that field object and hydrate - is that correct?10:16
bauzassudipto: instead of using child_versions, just provide a obj_relationship saying 'value': (1.1: <latestMonitorMetricValue object version>)10:16
bauzassudipto: yeah kind of10:16
bauzassudipto: that said, in general we dislike changing the type of a field10:16
sudiptobauzas, I have this in my working directory now:     obj_relationships = {10:17
sudipto        'value': [('1.1', '1.0')],10:17
sudipto    }10:17
bauzasright10:17
bauzasbut you also need to provide a backward check for 1.010:17
sudiptobauzas, along with this i have to implement the obj_make_compatible method.10:17
*** ociuhandu has quit IRC10:17
sudiptobauzas, to address the 1.0 concern right?10:18
bauzasyup10:18
sudiptobauzas, hmm ok - so in this case, it would be that if i get a primitive version - i should have a way to convert it into the newer version...10:18
bauzasdisclaimer, I'm not an object ninja so some of my thoughts could be wrong :)10:18
bauzassudipto: that's for example when you're passing an object over RPC10:19
bauzassudipto: the object is primitived and then rehydrated10:19
sudiptobauzas, yeah that actually makes sense to me...10:19
bauzassudipto: what's pretty cool with the NovaObjects is that the rehydratation is possible even if the object version is not the same10:20
bauzassudipto: just because it will actually look at the obj_make_compatible() method10:20
sudiptobauzas, yeah - you mean if an older object comes over, we can still hydrate it to the newer object via that method...10:20
sudipto?10:20
bauzassudipto: so, for example, you have a n-api having an 1.1 object10:21
bauzasor a conductor rather10:21
bauzasso, the conductor is primitiving it using obj_to_primitive into a dict and by marking its version10:21
bauzasit goes over the wire as a dict10:21
bauzasand then, the manager method is rehydrating it back10:22
sudiptobauzas, yup - got it...10:22
bauzasif the manager talks the 1.2 code (because it's newer), then it can still hydrate the object10:22
bauzasor 1.0 whatever10:22
sudiptobauzas, yeah10:23
*** mutoulbj_ has quit IRC10:23
bauzassudipto: that's also why you need to pin the versions for the nested objects10:23
sudiptobauzas, got it. Thanks for all the explanation :)10:23
sudiptobauzas, learning slowly :)10:24
bauzasnp10:24
*** annashen has joined #openstack-nova10:25
openstackgerritSergey Vilgelm proposed openstack/nova: Switch to oslo.policy  https://review.openstack.org/19902010:29
*** annashen has quit IRC10:30
*** exploreshaifali has quit IRC10:32
*** otter768 has joined #openstack-nova10:34
*** otter768 has quit IRC10:39
*** ericksonsantos has joined #openstack-nova10:40
*** eglynn has quit IRC10:46
*** dims has joined #openstack-nova10:46
*** eglynn has joined #openstack-nova10:46
*** iamjarvo has joined #openstack-nova10:47
*** deepthi has quit IRC10:49
*** dims has quit IRC10:51
*** iamjarvo has quit IRC10:52
*** ociuhandu has joined #openstack-nova11:00
*** Marga__ has quit IRC11:00
*** lucasagomes is now known as lucas-hungry11:01
*** Marga_ has joined #openstack-nova11:02
*** davideagnello has quit IRC11:04
*** lpetrut has quit IRC11:07
*** mutoulbj has joined #openstack-nova11:08
*** Longgeek has quit IRC11:08
*** exploreshaifali has joined #openstack-nova11:09
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler  https://review.openstack.org/19905111:11
*** lyanchih has quit IRC11:13
*** Marga_ has quit IRC11:15
*** arnaud__ has joined #openstack-nova11:15
*** rotbeard has joined #openstack-nova11:15
*** Marga_ has joined #openstack-nova11:17
rgerganovalex_xu: around?11:19
*** humble_ has quit IRC11:20
*** claudiub has quit IRC11:23
*** thorst has joined #openstack-nova11:24
*** annashen has joined #openstack-nova11:26
*** eliqiao has joined #openstack-nova11:27
*** andreykurilin has quit IRC11:29
*** claudiub has joined #openstack-nova11:30
*** annashen has quit IRC11:30
*** prashup has quit IRC11:30
*** andreykurilin has joined #openstack-nova11:32
openstackgerritAndreas Scheuring proposed openstack/nova: Add macvtap as new virtual interface (vif) type to Nova's libvirt driver.  https://review.openstack.org/18228311:45
*** tongli has joined #openstack-nova11:45
*** Longgeek has joined #openstack-nova11:45
alex_xurgerganov: here11:49
rgerganovalex_xu: I'd like to ask you an API question if you don't mind11:50
alex_xurgerganov: sure, please11:50
openstackgerritSergey Vilgelm proposed openstack/nova: Switch to oslo.policy  https://review.openstack.org/19902011:50
rgerganovalex_xu: I am going to add a new protocol for remote consoles11:50
*** shz has joined #openstack-nova11:50
rgerganovalex_xu: does that need a new API microversion?11:50
alex_xurgerganov: it need new version11:51
*** aloga_ has quit IRC11:51
*** shihanzhang has quit IRC11:51
*** aloga_ has joined #openstack-nova11:51
rgerganovalex_xu: ok, what about the validation schema11:51
rgerganovalex_xu: do I need to modify the existing one?11:51
rgerganovalex_xu: or add a new one?11:52
alex_xurgerganov: our schema support mircoversion, let me find example for you11:52
alex_xurgerganov: https://github.com/openstack/nova/blob/master/nova/api/validation/__init__.py#L2511:54
alex_xuI didn't found devref describe this, I should add some describe for this11:54
rgerganovalex_xu: hm, but this is the case when the validation schema is the same11:55
rgerganovin my case the validation schema is extended with one more value for protocol11:55
rgerganovi.e. from ['vnc', 'spice', 'rdp', 'serial'] it becomes ['vnc', 'spice', 'rdp', 'serial', 'mks']11:56
alex_xurgerganov: yes11:57
alex_xurgerganov: https://github.com/openstack/nova/blob/master/nova/tests/unit/api/openstack/compute/test_plugins/microversions.py#L7211:57
*** jasondotstar has quit IRC11:57
*** baoli has joined #openstack-nova11:57
rgerganovalaski: ah, so I need to add one more validation decorator11:58
alex_xurgerganov: yes :)11:58
rgerganovsorry, the last message was for alex_xu not alaski :)11:58
rgerganovalex_xu: thanks11:58
alex_xurgerganov: np11:58
*** markvoelker has quit IRC11:59
*** markvoelker has joined #openstack-nova11:59
*** baoli has quit IRC12:01
*** derekh has quit IRC12:01
*** baoli has joined #openstack-nova12:01
*** derekh has joined #openstack-nova12:02
*** dims has joined #openstack-nova12:02
*** jaypipes has joined #openstack-nova12:02
*** arnaud___ has joined #openstack-nova12:04
*** kiran-r has quit IRC12:06
*** arnaud__ has quit IRC12:07
*** lucas-hungry is now known as lucasagomes12:07
*** dims has quit IRC12:07
*** ferest has joined #openstack-nova12:08
*** tojuvone has quit IRC12:11
*** tojuvone has joined #openstack-nova12:11
*** dave-mccowan has joined #openstack-nova12:13
openstackgerritSudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors  https://review.openstack.org/19897212:15
*** ajayaa has quit IRC12:15
openstackgerritRoman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down  https://review.openstack.org/18408612:15
*** dprince has joined #openstack-nova12:15
*** ferest has quit IRC12:15
*** browne has joined #openstack-nova12:18
*** piyanai has quit IRC12:20
*** sudipto has quit IRC12:22
*** arnaud___ has quit IRC12:23
*** ankit_ag has joined #openstack-nova12:25
*** rdopiera has joined #openstack-nova12:25
*** tongli has quit IRC12:26
*** lpetrut has joined #openstack-nova12:26
*** tongli has joined #openstack-nova12:27
*** annashen has joined #openstack-nova12:27
*** aysyd has joined #openstack-nova12:28
*** Murali has quit IRC12:29
*** Murali has joined #openstack-nova12:30
*** ajayaa has joined #openstack-nova12:31
*** annashen has quit IRC12:31
*** markvoelker has quit IRC12:31
*** tongli has quit IRC12:32
*** jlanoux has joined #openstack-nova12:32
*** ivasev has joined #openstack-nova12:33
*** otter768 has joined #openstack-nova12:35
openstackgerritMichal Rostecki proposed openstack/nova: Dedicate aggregates for specific tenants  https://review.openstack.org/19578312:37
alex_xufffdsafd12:37
bauzasalex_xu: agreed12:38
bauzas:p12:38
*** edmondsw has joined #openstack-nova12:38
*** otter768 has quit IRC12:40
*** markvoelker has joined #openstack-nova12:40
*** eglynn has quit IRC12:44
*** eglynn_ has joined #openstack-nova12:44
jaypipesedleafe: you workin' on https://review.openstack.org/#/c/192184/ ? looks like there's a real test failure...12:45
jaypipesalex_xu: well said.12:45
*** moshele has quit IRC12:46
edleafejaypipes: yes, but still digging out from vacation. It's on my todo list12:46
alex_xujaypipes: thanks12:46
jaypipesalex_xu: :P12:46
edleafealex_xu: I totally disagree12:46
edleafe:)12:46
alex_xu....12:47
jaypipesalaski: around? I have concerns around injecting SQLAlchemy calls directly into the cellsv2 nova.objects. Is dansmith aware of this decision?12:47
*** rotbeard has quit IRC12:47
alaskijaypipes: I'm around12:48
jaypipesalaski: the reason I am concerned is that I was very much interested in using a non-relational DB for certain top-level operations that need to be less transactional and more "available"...12:48
*** jichen_ has joined #openstack-nova12:49
alaskitop-level being above an object?12:49
bauzasjaypipes: AFAIR, it was agreed to incrementally remove some DB API methods and directly call them from the objects12:49
bauzasjaypipes: because that Facade is not really helpful IMHO12:49
bauzascall SQLA, not "them"12:50
alex_xuedleafe: btw, I updated https://review.openstack.org/#/c/193725/ last week, to get rid of chaning api-paste.ini.12:50
*** jichen has quit IRC12:50
*** jichen_ is now known as jichen12:50
jaypipesbauzas: the problem with that is we limit ourselves to only relational databases.12:51
bauzasjaypipes: I understand you12:51
edleafealex_xu: yes, I saw that. There is a simple solution for the pep8 failures that I was hoping to post, but my devstack is broken12:51
jaypipesalaski: I mean things like the instance_uuid -> cell mappings12:51
*** moshele has joined #openstack-nova12:51
bauzasjaypipes: but for the moment, the DB API is only implemented by SQLA, right?12:51
*** eglynn_ has quit IRC12:51
*** Piet has joined #openstack-nova12:51
jaypipesalaski: it's not the ultimate source of truth, and it's something that should have higher levels of availability guarantee than an individual cell's database.12:51
alex_xuedleafe: yea, I can update that. And the functional test still need a lot of correct for 'import'12:51
alaskijaypipes: so the big thing that changes is that rather than a new db/api.py IMPL you would subclass the object and override the *_from_db *_to_db methods12:52
*** deshipu has left #openstack-nova12:53
jaypipesalaski: I see...12:54
jaypipesalaski: so the extensibility is kept, just a different implementation of that extensibility?12:54
*** lyanchih has joined #openstack-nova12:54
alaskijaypipes: right.  tbh we don't have a selection mechanism yet, but the idea was to maintain the ability to be extensible12:55
jaypipesalaski: I see. OK, cool with me. thanks for talking me off the cliff ;)12:55
alex_xusdague: gmann oomichi gilliard, a small clarify for remove v3 https://review.openstack.org/#/c/193589/ , hope get +1 from api sub team member, then get it merge.12:55
alaskijaypipes: and be more targeted with it12:55
alaskijaypipes: heh, cool12:55
bauzasgood to see alaski explaining my thoughts :)12:56
bauzasalaski: jaypipes: can I bummer you with a question ?12:57
jaypipesalaski: :) thx.12:57
jaypipesbauzas: sorry, Jay is away from the phone right now. Please leave a message.12:57
*** tojuvone has quit IRC12:57
bauzasjaypipes: heh, I can text you :p12:57
jaypipesbauzas: j/k :) what's up?12:57
bauzasjaypipes: so, I'm pretty done with all the ReqSpec usage internally12:58
*** ctrath has joined #openstack-nova12:58
jaypipesbauzas: cool.12:58
bauzasjaypipes: now, I'm about to update the RPC AI12:58
bauzasAPI12:58
bauzaseven12:58
bauzasjaypipes: I was pretty lazy on the spec by saying I was adding a new method called 'select_nodes'12:59
*** mdrabe has joined #openstack-nova12:59
bauzasjaypipes: but IMHO, it would be *faaaaar* better to instead bump a new version of the RPC API (even a major why not) and provide directly a new signature for select_destinations()12:59
jaypipesbauzas: that works for me.13:00
bauzasjaypipes: something like in the client : if client.can_send_version(new_vers): call.('select_destinations', ReqSpec)13:00
bauzasjaypipes: do you think a major version is good, or just a 4.3 ?13:00
*** vladikr has joined #openstack-nova13:00
alaskithat would be preferable for me as well.  you just might have to fight the rpc framework a bit13:00
bauzasalaski: that would be easy using a separate proxy for 5.013:01
jaypipesbauzas: 4.3 is fine with me.13:01
bauzas(or 4.3)13:01
bauzasalaski: so the proxy would be overriding the select_dest() method only13:01
bauzasalaski: jaypipes: thanks for your quick feedback, will see how I can achieve that13:02
alaskibauzas: I think you would end up needing to proxy everything, because I don't think you can switch between sending two different major versions easily13:02
alaskibut a minor version is fine, as long as it's additive and maintains backwards compat13:03
bauzasalaski: okay, lemme draft out some stuff so dansmith and you both can yell at me13:03
*** tojuvone has joined #openstack-nova13:04
openstackgerritNikola Dipanov proposed openstack/nova: rt: generalize claim code to be useful for other move actions  https://review.openstack.org/19713813:06
openstackgerritNikola Dipanov proposed openstack/nova: compute: rename ResizeClaim to MoveClaim  https://review.openstack.org/19713713:06
openstackgerritNikola Dipanov proposed openstack/nova: Define migration states and transitions and use when changing status  https://review.openstack.org/19766913:06
openstackgerritNikola Dipanov proposed openstack/nova: Introduce a simple state machine module  https://review.openstack.org/19766813:06
*** eharney has joined #openstack-nova13:07
*** rlrossit has joined #openstack-nova13:07
*** changbl has quit IRC13:08
*** jskinner has joined #openstack-nova13:08
*** haomaiw__ has quit IRC13:08
*** jasondotstar has joined #openstack-nova13:09
*** browne has quit IRC13:10
openstackgerritjichenjc proposed openstack/nova: Check flavor type before add tenant access  https://review.openstack.org/12433813:10
*** jasondot_ has joined #openstack-nova13:10
*** ankit_ag has quit IRC13:11
*** jasondot_ has quit IRC13:13
*** jasondotstar has quit IRC13:14
*** jasondotstar has joined #openstack-nova13:14
*** exploreshaifali has quit IRC13:15
*** kiran-r has joined #openstack-nova13:18
*** lyanchih has quit IRC13:19
*** dims has joined #openstack-nova13:19
*** dboik has joined #openstack-nova13:20
*** wanghua has joined #openstack-nova13:20
*** lyanchih has joined #openstack-nova13:20
*** sudipto has joined #openstack-nova13:21
openstackgerritAnthony PERARD proposed openstack/nova: libvirt-vif: Allow to configure a script on bridge interface  https://review.openstack.org/19909213:22
openstackgerritAnthony PERARD proposed openstack/nova: libvirt-xen: Add xen_vif_bridge_script_path config option.  https://review.openstack.org/19909313:22
*** haomaiwa_ has joined #openstack-nova13:22
*** nihilifer has quit IRC13:23
*** mriedem_away is now known as mriedem13:27
*** shz has quit IRC13:27
*** shz has joined #openstack-nova13:27
*** annashen has joined #openstack-nova13:27
*** piyanai has joined #openstack-nova13:29
*** Piet has quit IRC13:30
*** browne has joined #openstack-nova13:30
*** mtanino has joined #openstack-nova13:31
*** annashen has quit IRC13:32
*** boris-42 has joined #openstack-nova13:33
*** tongli has joined #openstack-nova13:34
*** mpavone has quit IRC13:36
openstackgerritSergey Vilgelm proposed openstack/nova: Switch to oslo.policy  https://review.openstack.org/19902013:37
*** jecarey has joined #openstack-nova13:38
*** tongli has quit IRC13:39
*** jasondotstar has quit IRC13:40
*** annegentle has joined #openstack-nova13:42
*** haomaiwa_ has quit IRC13:42
*** ajayaa has quit IRC13:45
*** jichen has quit IRC13:46
*** jichen has joined #openstack-nova13:47
*** haomaiwang has joined #openstack-nova13:48
*** piyanai_ has joined #openstack-nova13:49
*** piyanai has quit IRC13:49
*** piyanai_ is now known as piyanai13:49
*** mpavone has joined #openstack-nova13:50
mriedemhas anyone ever requested that we automatically rename (change instance.display_name) an instance after it's created to the hostname --fqdn of the instance? that seems like something that would have to be done outside of nova once the instance is created13:51
*** nihilifer has joined #openstack-nova13:51
mriedeminject an ssh key and then use that to get the hostname off the instance and rename it via nova-api13:52
*** claudiub has quit IRC13:52
johnthetubaguymriedem: I have had folks asking for that, although never done it13:54
mriedemi was looking at designate but there wasn't an obvious fit there - designate does have event listeners for nova instance create end13:55
johnthetubaguymriedem: the designate/neutron dns discussions brought that back up13:55
*** signed8bit_ZZZzz is now known as signed8__13:55
mriedemjohnthetubaguy: jinx :)13:55
johnthetubaguy:)13:55
johnthetubaguyso the instance certainly doesn't get its hostname updated13:55
johnthetubaguyat least not automatically13:55
*** r-daneel has joined #openstack-nova13:56
mriedemyeah, like could designate provide the hostname to the instance and then the event listener in designate use the fixed ip from the nova event to do the hostname lookup in designate?13:56
johnthetubaguymriedem: honestly I wish the hostname was separate and is stuck as the initial value on instance creation13:56
johnthetubaguyyeah, thats true13:56
*** annegentle has quit IRC13:57
johnthetubaguyI mean in some ways I wish we called that a server lablel13:57
johnthetubaguylabel13:57
johnthetubaguythen it wouldn't get confused with hostname13:57
johnthetubaguybut the ship has sorta sailed already13:57
alaskiI would prefer that hostname and display_name were completely decoupled13:57
mriedemalaski: yeah, that was the other confusing thing, the instance.hostname is the display_name in most cases13:57
johnthetubaguyalaski: +1 thats a better way of saying what I mean13:57
*** lpetrut has quit IRC13:58
mriedemit's probably a good thing that nova show doesn't list the hostname value13:58
johnthetubaguymriedem: unless you use neutron, and the dns hostname is based on the IP, although the instance uses the nova label13:58
tojuvonejohnthetubaguy: Could this be also approved as part of the mark-host-dow, it is just the novaclient side: https://blueprints.launchpad.net/python-novaclient/+spec/support-force-down-service13:58
*** nelsnelson has joined #openstack-nova13:59
*** lpetrut has joined #openstack-nova14:00
johnthetubaguytojuvone: we don't usually do a separate blueprint for python-novaclient, you can just add it to your nova blueprint, if thats easier?14:00
*** andrewbogott_afk is now known as andrewbogott14:00
tojuvoneYes, it was perhaps my misunderstanding, just how it is the easiest14:01
*** ijw has joined #openstack-nova14:01
openstackgerritTimofey Durakov proposed openstack/nova: Split instance resize operation into 2 phases  https://review.openstack.org/19508814:01
openstackgerritTimofey Durakov proposed openstack/nova: Create class hierarchy for tasks in conductor  https://review.openstack.org/18449514:01
tojuvoneso to move the review to the other BP side I guess?14:01
*** annegentle has joined #openstack-nova14:02
johnthetubaguytojuvone: yeah, I think that is the easiest way forward14:03
*** Kennan2 has joined #openstack-nova14:03
tojuvoneThanks, will do that :)14:03
*** zhenguo has quit IRC14:03
*** Kennan has quit IRC14:03
*** signed8__ is now known as signed8bit_ZZZzz14:04
sudiptodansmith, If i add a new subclass of the nova objects,  i believe i'd need to generate a hash for the object and add it to the test_objects.py - could you please tell me how i could generate the hash?14:04
*** xyang1 has joined #openstack-nova14:05
*** _gryf is now known as _gryf|afk14:05
dansmithsudipto: http://github.org/openstack/nova/tree/master/nova/tests/unit/objects/test_objects.py#L1244-124414:06
sudiptodansmith, ok thx...14:07
*** annegentle has quit IRC14:07
*** burt has joined #openstack-nova14:07
*** armax has joined #openstack-nova14:08
*** iamjarvo has joined #openstack-nova14:09
*** irenab has quit IRC14:09
*** iamjarvo has joined #openstack-nova14:10
*** yingjun has joined #openstack-nova14:11
openstackgerritChris Gacsal proposed openstack/nova: virt: Allow multiple volume attachments  https://review.openstack.org/19313514:13
openstackgerritChris Gacsal proposed openstack/nova: comp: Allow multiple volume attachments  https://review.openstack.org/19313414:13
openstackgerritChris Gacsal proposed openstack/nova: obj: Allow multiple volume attachments  https://review.openstack.org/19313314:13
openstackgerritChris Gacsal proposed openstack/nova: WIP: REST API changes to support multi-attach  https://review.openstack.org/15303814:13
*** mdrabe has quit IRC14:13
dansmithsudipto: er, wait14:15
dansmithsudipto: that's not the line I wanted you to see :D14:15
sudiptodansmith, but i think i saw it :D14:15
dansmithsudipto: GENERATE_HASHES right?14:15
*** tojuvone has quit IRC14:16
sudiptodansmith, basically i ran the Unit Test once - it generated the hashes, (and pointed the missmatch) - i copied it - will that not work?14:16
dansmithsorry about that, my local tree was out of sync with master so my auto-point-to-github routine failed :)14:16
dansmithyep, you can do that too14:16
sudiptodansmith, cool :)14:16
*** signed8bit has joined #openstack-nova14:16
*** signed8bit_ZZZzz has quit IRC14:17
*** mpaolino_ has joined #openstack-nova14:19
*** yingjun has quit IRC14:19
*** mpaolino has quit IRC14:20
*** cfriesen has joined #openstack-nova14:20
*** yingjun has joined #openstack-nova14:20
*** tongli has joined #openstack-nova14:20
jaypipesdansmith, sdague: quick one for ya. https://review.openstack.org/#/c/198244/14:21
*** zsais has joined #openstack-nova14:22
openstackgerritSudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors  https://review.openstack.org/19897214:23
dansmithdone14:24
*** yingjun has quit IRC14:24
*** lucas_ has joined #openstack-nova14:24
sudiptodansmith, if I may ask for a review of  https://review.openstack.org/19897214:25
dansmithsudipto: I'll look after jaypipes is happy with the monitor part14:26
sudiptodansmith, sure thing!14:26
*** gszasz has quit IRC14:28
*** annashen has joined #openstack-nova14:28
bauzassudipto: left some comments14:28
bauzasthe to_dict() method is not a remote method14:29
bauzasso, that's fine to leave the value.value14:29
*** mdrabe has joined #openstack-nova14:29
bauzasbut if you want to hydrate a 1.0 version of your object, then you need to provide an integer14:29
sudiptobauzas, thanks! The hydrate part yeah. I understood the opposite.14:30
bauzasas the 1.0 version of the object will have a to_dict() method directly checking the value, that's fine14:30
bauzas... unless I made a mistake, because I'm not running DanSmith().review()14:31
*** gszasz has joined #openstack-nova14:31
ndipanovso jaypipes I know you appreciate good names, but are you sure you want to stick with 'instance_type'14:31
ndipanovhttps://review.openstack.org/#/c/197138/4/nova/compute/resource_tracker.py14:31
ndipanovI thought we wanted to get rid of that mostly14:31
ndipanovshouldn't it be new_flavor at least14:32
jaypipesndipanov: I friggin hate the term "flavor".14:32
ndipanovoh boy14:32
*** annegentle has joined #openstack-nova14:32
ndipanovI feel that that ship has sailed :D14:32
bauzasor, flavour instead ? :D14:32
*** yingjun has joined #openstack-nova14:32
bauzasand schedular :)14:32
dansmitha consistent bad name is better than inconsistency, IMHO14:32
jaypipesndipanov: I don't really see a reason to change it from instance type.14:32
dansmith(although I don't have a problem with flavor)14:32
*** annashen has quit IRC14:33
ndipanovfair enough14:33
sudiptobauzas, ok - the way I thought about is exactly the opposite. So i guess, we don't need that routine/method at all.14:35
*** aloga_ has quit IRC14:35
bauzassudipto: well, imagine that you have to provide a path for accepting an old object14:36
*** otter768 has joined #openstack-nova14:36
openstackgerritRadoslav Gerganov proposed openstack/nova: Add MKS protocol for remote consoles  https://review.openstack.org/19912914:36
sudiptobauzas, i thought hydrating a old object using the new field semantics.14:36
*** BadCub-away is now known as BadCub14:36
*** ijuwang has joined #openstack-nova14:37
bauzasoh boy14:37
bauzassec14:37
*** aloga_ has joined #openstack-nova14:37
*** thangp has joined #openstack-nova14:38
openstackgerritSudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors  https://review.openstack.org/19897214:39
sudiptobauzas, new to objects, but thought primitive usually refers to the older version?14:39
bauzassudipto: yeah, obj_make_compatible is for providing a primitive dict compatible with the expected version14:39
bauzassudipto: so, here, you have a semantic change14:40
*** otter768 has quit IRC14:40
bauzassudipto: so, yeah I was wrong14:40
sudiptobauzas, so in this case, the expected version is the MonitorMetricValue, so i thought i should change the primitive int to this form...14:40
sudiptobauzas, ok darn, i was too quick to revert it. I will change it back.14:40
bauzassudipto: ok, here is the thing14:40
bauzassudipto: hold on14:41
bauzassudipto: so, you have a 1.1 object and you've been said that your target only accepts 1.014:41
bauzassudipto: then, it will run obj_make_compat() with the target set to 1.014:41
bauzassudipto: in that case, you need to change the value of the field to what the 1.0 object expects, ie. an integer14:42
sudiptobauzas, oh k, i thought i am saying - if i receive a 1.0 object, i know it's going to need a conversion, and hence convert into an object?14:43
*** annegentle has quit IRC14:43
bauzassudipto: here is the logic https://github.com/openstack/nova/blob/master/nova/objects/base.py#L407-L41214:43
*** otter768 has joined #openstack-nova14:44
bauzassudipto: you're first dehydrating the 1.1 object into a dict14:44
bauzassudipto: and then, you're tranforming that dict into another dict compatible with 1.014:44
sudiptobauzas, oh k. that way value.value will ensure that it's transformed properly...14:45
*** jichen has quit IRC14:46
bauzaswhat you'll have in the primitive['value'] is the value.to_primitive()14:46
*** jichen has joined #openstack-nova14:47
*** jichen has quit IRC14:47
bauzassudipto: which should be a dict of {'int_val': sth(), 'numa_nodes_value': sth_else()}14:47
*** jichen has joined #openstack-nova14:48
*** tonytan4ever has joined #openstack-nova14:49
*** Murali has quit IRC14:49
*** ajayaa has joined #openstack-nova14:49
sudiptobauzas, so target_version is 1.0 in that case right?14:49
*** aloga_ has quit IRC14:49
bauzassudipto: correct14:49
*** aloga_ has joined #openstack-nova14:49
bauzassudipto: and primitive is the result of https://github.com/openstack/nova/blob/master/nova/objects/base.py#L407-L41014:50
bauzas(or possibly a db dict)14:50
sudiptobauzas, in the 1.0 version this value field should be just and integer...so should be then saying in the make_comp method - something like:14:51
bauzassudipto: NO14:51
bauzassudipto: again, imagine that you have a 1.114:51
bauzassudipto: and then, you want to send it over RPC14:52
sudiptobauzas, yeah14:52
sudiptowhich is something like this:  {'int_val': sth(), 'numa_nodes_value': sth_else()}14:52
bauzassudipto: but your compute node is stupidly only accepting 1.0 objects14:52
bauzassudipto: right14:52
bauzassudipto: because objects are primitived into dicts with their fields as keys14:52
bauzassudipto: so, given that, you need to change primitive['value'] because if not, when rehydrating the 1.0 object, it will raise an exception14:54
*** emagana has joined #openstack-nova14:54
*** VW_ has joined #openstack-nova14:55
bauzassudipto: like dansmith said here https://github.com/openstack/nova/blob/master/nova/objects/base.py#L38114:55
*** bradjones|away is now known as bradjones14:56
*** achanda has joined #openstack-nova14:56
sudiptobauzas, got it - so what you are saying is - now the data would be in that dict format and the 1.0 accepts an integer right?14:57
dansmithare you guys talking about changing the type of a field?14:57
*** mgoddard has quit IRC14:59
*** garyk1 has joined #openstack-nova14:59
*** mgoddard has joined #openstack-nova15:00
sudiptodansmith, yeah - and how do we ensure backward compatibility.15:00
*** zzzeek has joined #openstack-nova15:02
*** garyk has quit IRC15:02
*** achanda has quit IRC15:02
dansmithsudipto: don't change field types please :)15:02
dansmithsudipto: add a new field and deprecate the old one if necessary15:02
*** ijuwang has quit IRC15:03
*** annegentle has joined #openstack-nova15:04
sudiptodansmith, oh k, actually this field was just added by jaypipes, so i changed it...15:05
*** _kiran_ has joined #openstack-nova15:05
*** thedodd has joined #openstack-nova15:05
*** _kiran_ has quit IRC15:06
*** ijw has quit IRC15:06
sudiptodansmith, just added = merged yesterday...so i thought of writing one version of the object...is it too quick for a version bump?15:07
mriedemdansmith: don't add new field types huh? https://review.openstack.org/#/c/197208/15:07
mriedems/add/change/15:07
mriedemgoing from string type to enum type seems ok15:08
ndipanovmriedem, yeah that's fine15:08
ndipanovif they end up a string and you just change the code that's fine15:09
mriedemndipanov: yeah, that's all that's in that series15:09
mriedemi plan on actually replacing the hard-coded strings in the code later with the enums15:09
sc68caljaypipes: got an easy one for you to review - https://review.openstack.org/#/c/198833/15:10
dansmithsudipto: no, you can't cheat :)15:10
dansmithmriedem: you're not changing the type there15:11
dansmithmriedem: : the only thing you could run into there is if your enum wasn't complete for some scenario15:12
*** annegentle has quit IRC15:13
jaypipessc68cal: reviewed.15:13
*** agireud has joined #openstack-nova15:14
sc68caljaypipes: thx15:14
jaypipessc68cal: np15:14
*** dannywilson has joined #openstack-nova15:16
*** dims has quit IRC15:16
*** sudipto has quit IRC15:16
*** mpaolino has joined #openstack-nova15:18
*** mpaolino_ has quit IRC15:18
*** emagana has quit IRC15:19
*** dannywilson has quit IRC15:20
*** tsekiyama has joined #openstack-nova15:20
mriedemsc68cal: jaypipes: oh hell yeah http://code.metager.de/source/xref/linux/utils/bridge-utils/brctl/brctl_cmd.c#3815:21
*** SourabhP has joined #openstack-nova15:21
*** annegentle has joined #openstack-nova15:21
mriedemEEXIST and default are both exiting with 1 :(15:21
*** tpeoples has joined #openstack-nova15:21
*** gszasz has quit IRC15:21
openstackgerritNikola Dipanov proposed openstack/nova: rt: generalize claim code to be useful for other move actions  https://review.openstack.org/19713815:23
openstackgerritNikola Dipanov proposed openstack/nova: Define migration states and transitions and use when changing status  https://review.openstack.org/19766915:23
openstackgerritNikola Dipanov proposed openstack/nova: Introduce a simple state machine module  https://review.openstack.org/19766815:23
*** SourabhP_ has joined #openstack-nova15:23
*** nkrinner is now known as nkrinner_afk15:23
*** lyanchih has quit IRC15:24
jaypipesmriedem: poo.15:25
mriedem+2ed it15:25
mriedemwith some nits15:25
*** SourabhP has quit IRC15:26
*** SourabhP_ is now known as SourabhP15:26
ndipanovdansmith, https://review.openstack.org/#/c/197669/ so tell me more abou the 2 status fields15:26
ndipanovI think I don't fully get how we would use them so I don't understand why they would be useful15:27
dansmithndipanov: I'm on a call, so I can't really concentrate, but we were just thinking of leaving status as more free-form15:27
*** ijuwang has joined #openstack-nova15:27
ndipanovok ping me later when you have more time ?15:27
*** dannywilson has joined #openstack-nova15:27
dansmithndipanov: well, we can constrict status to the available states or we can add another field that is tightly constrained, and let the status take on a bunch of non-constrained statuses, like I do in my second evacuate patch15:28
dansmithndipanov: sure15:28
openstackgerritSergey Nikitin proposed openstack/nova: Added method exists to the Tag object  https://review.openstack.org/17810315:28
*** annashen has joined #openstack-nova15:29
*** mpaolino has quit IRC15:29
*** ijw has joined #openstack-nova15:31
*** changbl has joined #openstack-nova15:31
johnthetubaguybauzas: I have attempted to pencil in some remote times in the midcycle agenda, I hope that helps: https://etherpad.openstack.org/p/liberty-nova-midcycle15:32
bauzasjohnthetubaguy: oh coolness !15:32
johnthetubaguymriedem: hey, so I filled out the midcycle etherpad a bit, does that look how you would expect? https://etherpad.openstack.org/p/liberty-nova-midcycle15:32
mriedemoh yeah, today is the last day for registratoin15:32
johnthetubaguyoh, good point15:32
johnthetubaguyI will drop a note to the ML if you like15:32
johnthetubaguywas going to point out the above etherpad15:32
dansmithmikal did a few days ago15:32
sc68calmriedem: nice find15:32
johnthetubaguydansmith: yeah, I can reply to that one15:33
*** kiran-r has quit IRC15:33
bauzasjohnthetubaguy: nice to see that, will raise the etherpad to the sched folks15:33
mriedemjohnthetubaguy: i assume you'll post the etherpad to the ML?15:33
*** annashen has quit IRC15:34
bauzasjohnthetubaguy: like I said, I should be enough happy if I could just lurk at the convos thanks to a mic15:34
johnthetubaguymriedem: yeah, just writing that now15:35
*** gszasz has joined #openstack-nova15:35
bauzasjohnthetubaguy: but I can hassle n0ano or someone else to open his mic15:35
*** Nikolay_St has quit IRC15:35
mriedembauzas: i think there are conf phones spread out throughout the room(s) but not sure how many or how well it will sound without repetition and yelling15:35
*** lyanchih_ has joined #openstack-nova15:35
bauzasmriedem: well, I'm fine with just listening15:36
*** garyk has joined #openstack-nova15:36
*** emagana has joined #openstack-nova15:36
bauzasmriedem: in case I need to say something, I can just ping you guys15:37
bauzasmriedem: if you don't get it in time, that's fair15:37
bauzasmriedem: I don't want to slow down the convo15:37
bauzas(and promised, I shouldn't talk about cheese)15:37
mriedemooo the tasty cheese15:37
*** sudipto has joined #openstack-nova15:37
bauzasso, yeah, I just need either a call number or a conference bridge, that's good enough :)15:38
*** garyk1 has quit IRC15:38
*** davidlenwell has quit IRC15:38
mriedemi have an international call in if needed15:38
bauzasmriedem: fair enough :)15:39
bauzasmriedem: I'll put myself on mute, no need to listen to me15:39
johnthetubaguymriedem: bauzas: sounds like we have options we can try15:40
johnthetubaguyand take questions via IRC I guess, :)15:40
bauzaswould be awesome15:40
bauzasjohnthetubaguy: yeah, zactly15:40
johnthetubaguyso we can try it, we might have to give up, but we can try15:40
bauzasjohnthetubaguy: we can just have a policy15:41
bauzasjohnthetubaguy: saying 'if you're remote, don't try to speak synchronously, but rather try async'15:41
bauzasso IRC sounds the best medium for thaty15:41
bauzasif the point is missed, no need to hassle it15:41
*** emagana has quit IRC15:42
*** dims has joined #openstack-nova15:43
*** xyhuang has quit IRC15:44
sudiptojaypipes, dansmith , this field that we are changing the value of - is not a DB column, (the value field for the monitors) - would you still press for deprecating it?15:46
dansmithsudipto: it's an RPC field, which means it needs extreme care15:46
dansmithsudipto: even more so than a db field15:46
ndipanovmriedem, https://review.openstack.org/#/c/189632/ <- good points there15:47
ndipanovsee my comment on why I think it's reasonable to not mock out everything there15:47
sudiptodansmith, alrite!15:47
*** nihilifer has quit IRC15:47
ndipanovwill update the patch - thanks for those reviews15:47
*** unicell has quit IRC15:49
*** tongli has quit IRC15:50
*** tongli has joined #openstack-nova15:50
mriedemlooking15:50
*** whenry_ has quit IRC15:51
jaypipescores: https://review.openstack.org/#/c/102649/ looks worth another review...15:51
mriedemndipanov: are you also going to add the UpgradeImpact to https://review.openstack.org/#/c/189632/ ?15:52
*** ijw has quit IRC15:53
*** davidlenwell has joined #openstack-nova15:53
openstackgerritVladik Romanovsky proposed openstack/nova: Adding user_id handling to keypair index, show and create api calls  https://review.openstack.org/7048515:54
*** tongli_ has joined #openstack-nova15:54
openstackgerritVladik Romanovsky proposed openstack/nova: tests: functional tests for keypairs create in API microversion 2.7  https://review.openstack.org/18364515:54
*** davidlenwell is now known as daviddavid15:54
*** tongli has quit IRC15:54
*** daviddavid is now known as davidlenwell15:54
*** annegentle has quit IRC15:56
*** moshele has quit IRC15:57
*** artom has joined #openstack-nova15:58
*** davidlenwell has quit IRC15:58
*** tongli_ has quit IRC15:58
*** davidlenwell has joined #openstack-nova15:58
*** salv-orlando has joined #openstack-nova15:58
ndipanovmriedem, sorry was away15:59
ndipanovwell I am but I am a little bit hazy on how that translates to release notes15:59
ndipanover I mean15:59
ndipanovhow it gets turned into release notes15:59
*** jistr has quit IRC16:00
*** tongli has joined #openstack-nova16:00
*** hemnafk is now known as hemna16:00
*** mgarza_ has joined #openstack-nova16:00
ndipanovmriedem, so I'll just add the tag and then describe inline that it's really something to be relnoted16:00
mriedemndipanov: i think we just point out in the release notes that device_name is ignored in bdm requests with the libvirt driver16:02
*** aswadr has joined #openstack-nova16:02
mriedemndipanov: DocImpact is used way too often and isn't really used for release notes all the time, i think at the end of the release we at least search for commits with UpgradeImpact and make sure those are doc'ed in the release notes16:02
mriedemwe could start using a new tag like ReleaseNotes or something, but meh16:02
ndipanovmriedem, yes16:03
*** annegentle has joined #openstack-nova16:03
kashyapRandomly chiming in, yeah ReleaseNotes tag sounds good. Could even be a `git notes`16:04
*** tongli_ has joined #openstack-nova16:04
*** tongli has quit IRC16:04
kashyapBut haven't seen someone use git-notes in context of OpenStack, though.16:05
* kashyap recalls an idea (from Rich Jones of libguestfs) to use 'git notes' to mark patches for cherry-picking for stable branches.16:07
*** scheuran has quit IRC16:07
mriedemi'll send something to the ML on a new tag16:07
*** jistr has joined #openstack-nova16:07
*** davidlenwell has quit IRC16:08
clarkbkashyap: every commit in openstack uses git notes to note review info16:08
clarkbfwiw16:08
*** jlanoux has quit IRC16:08
*** davidlenwell has joined #openstack-nova16:08
*** tongli_ has quit IRC16:09
*** ssurana has joined #openstack-nova16:09
*** annegentle has quit IRC16:10
*** jichen has quit IRC16:10
kashyapclarkb: Oh indeed, this tells me -- git fetch origin refs/notes/review:refs/notes/review16:11
kashyapclarkb: Thanks for the nice hint.16:11
mriedemthis libvirt change for bdm device_name is really probably an APIImpact tag...16:12
*** davidlenwell is now known as omnibus716:12
*** omnibus7 is now known as davidlenwell16:12
*** lucas_ has quit IRC16:12
*** davidlenwell is now known as david_enwell16:12
*** fawadkhaliq has joined #openstack-nova16:15
*** haomaiw__ has joined #openstack-nova16:16
*** haomaiwang has quit IRC16:18
*** david_enwell is now known as davidlenwell16:19
*** ajayaa has quit IRC16:20
*** mpavone has quit IRC16:20
*** davidlenwell has quit IRC16:20
*** davidlenwell has joined #openstack-nova16:20
*** hogepodge has quit IRC16:21
*** unicell has joined #openstack-nova16:23
*** davidlenwell has quit IRC16:23
*** davidlenwell has joined #openstack-nova16:24
*** tongli has joined #openstack-nova16:24
*** ildikov has quit IRC16:25
*** SourabhP has quit IRC16:26
*** davidlenwell has quit IRC16:27
*** davidlenwell has joined #openstack-nova16:28
*** unicell has quit IRC16:28
kashyapmriedem: Just learnt something, your (Mail) User-Agent says "Windows NT 6.1; WOW64; rv:31.0". Wonder if I should be unpleasantly surpried :P16:28
*** annashen has joined #openstack-nova16:28
*** unicell has joined #openstack-nova16:28
*** jcoufal has quit IRC16:28
*** armax has quit IRC16:28
mriedemmeh16:29
kashyapJ/k, please disregard me.16:29
mriedemi don't feel like trying to run all of ibm's mandatory software on red hat16:30
*** Piet has joined #openstack-nova16:31
*** davidlenwell has quit IRC16:31
kashyapDidn't meant that.  And, don't feed the troll (me).16:31
bauzasI don't exactly remember, can someone tell me how to nest 2 mocks in a contextmanager without using contextlib ?16:31
*** nic has joined #openstack-nova16:31
rdopierabauzas: with Foo, Bar as foo, bar: ...16:32
mriedembauzas: indentation16:32
bauzasI remember something about removing the usage of contextlib.nested16:32
bauzasmmm, doesn't work for me now, probably a miss then16:32
mriedembauzas: ex: https://review.openstack.org/#/c/181781/3/nova/tests/unit/virt/libvirt/test_driver.py16:33
mriedemline 868816:33
bauzasmriedem: okay, they're explicitely nested, but I was thinking more like rdopiera16:34
*** armax has joined #openstack-nova16:34
mriedemyeah, but foo and bar are usually so long that you have to put them on multiple lines16:34
*** haypo has left #openstack-nova16:34
mriedemwith mock.patch.object('foo......'), mock.patch.object('bar.......') as x,y16:34
mriedemdoesn't really work on a single line16:34
*** jistr has quit IRC16:35
*** exploreshaifali has joined #openstack-nova16:35
*** davidlenwell has joined #openstack-nova16:35
mriedemso if you have more than 2, explicit nesting isn't usually any better i don't think16:35
bauzasmriedem: http://paste.openstack.org/show/352439/16:35
bauzasmriedem: yeah, I added some parenthesis, but it sounds bad, see $16:35
*** davidlenwell is now known as davidlenwell_16:36
mriedemyeah, so i seem to remember that as long as you give a var, you can do them stacked like that16:36
*** agireud has quit IRC16:36
mriedemw/o parantehsis16:36
mriedembut you might need to use \16:36
mriedemidk, it's weird and not great options, dansmith looked it up once and i think we gave up16:36
dansmithon a with statement?16:37
bauzasmriedem: ergh16:37
mriedemhttps://docs.python.org/2/reference/compound_stmts.html#the-with-statement16:37
mriedemyeah16:37
bauzasdansmith: yup16:37
bauzasdansmith: see http://paste.openstack.org/show/352439/16:37
dansmithyeah, it's terrible16:37
mriedemso explicitly nest, or use contextlib.nested to stack them16:38
mriedemor use @mock.patch.object method decorators16:38
mriedemto stack16:38
*** davidlenwell_ is now known as davidlenwell16:38
bauzasmriedem: yeah, sounds like contextlib.nested is the best option16:38
mriedemwell, it's one of the lesser of many evils16:38
*** armax has quit IRC16:38
bauzasmriedem: well, the mocked object is created within the unittest so that's why I'm using a contextmanager16:39
bauzasmriedem: yeah, but at Ieast I'm not alone :)16:39
mriedemyup16:39
bauzasmriedem: we have plenty of that :(16:39
*** pbandzi_ has joined #openstack-nova16:39
bauzasmriedem: until someone steps up and find a good way to not use backslashes16:39
openstackgerritNikola Dipanov proposed openstack/nova: libvirt: Always default device names at boot  https://review.openstack.org/18963216:40
openstackgerritNikola Dipanov proposed openstack/nova: libvirt: implement get_device_name_for_instance  https://review.openstack.org/18063816:40
*** annegentle has joined #openstack-nova16:40
mriedembauzas: i'm waiting for python 416:40
mriedempython 4: the reckoning16:40
*** pbandzi has quit IRC16:41
bauzasmriedem: or golang even ?16:41
*** mdrabe has quit IRC16:42
mriedemno, i'm not cool enough for that16:42
dimsmriedem: jaypipes: when you get a chance, please take a look at https://review.openstack.org/#/c/187785/ (py3 additional tests)16:42
*** Nikolay_St has joined #openstack-nova16:42
* bauzas remembers a 2 hours meetup on Go, brrrr...16:43
*** rdopiera has quit IRC16:43
dimsmriedem: bauzas: in above, see nova/test.py, i have a equivalent for contextlib.nested that works with py316:43
*** tjones1 has joined #openstack-nova16:43
bauzasdims: ha, nice timing !16:44
tjones1hi - i think this is ready to go but could use some reviews https://review.openstack.org/#/c/127427/16:44
*** annegentle has quit IRC16:44
*** patrickeast has joined #openstack-nova16:45
*** agireud has joined #openstack-nova16:46
*** whenry has joined #openstack-nova16:46
*** lucas_ has joined #openstack-nova16:46
*** claudiub has joined #openstack-nova16:46
*** Longgeek has quit IRC16:47
*** kiran-r has joined #openstack-nova16:47
*** pbandzi_ has quit IRC16:48
*** veena has joined #openstack-nova16:48
*** tonytan4ever has quit IRC16:48
*** gszasz has quit IRC16:50
*** tongli has quit IRC16:50
*** lucas_ has quit IRC16:51
*** tongli has joined #openstack-nova16:51
*** lucasagomes is now known as lucas-dinner16:52
*** claudiub has quit IRC16:52
*** claudiub has joined #openstack-nova16:52
*** ZZelle has joined #openstack-nova16:53
sudiptobauzas, when we introduce a new field to the objects, and let's say the previous UTs were working without this field - do we just go about setting the new field to None to make the tests pass?16:53
*** lucas_ has joined #openstack-nova16:54
bauzassudipto: I maybe misunderstand your question, but no, you have to update the tests too16:54
*** karimb has quit IRC16:54
mriedemdid anyone realize that our sphinx docs conf file was building latex docs?16:54
mriedemmtreinish: ^ i know you're latex crazy16:54
*** yingjun has quit IRC16:54
*** lucas_ has quit IRC16:54
mriedemhttps://github.com/openstack/nova/blob/master/doc/source/conf.py#L24716:54
*** rlrossit has left #openstack-nova16:55
*** lucas_ has joined #openstack-nova16:55
mtreinishmriedem: yep, I knew16:55
sudiptobauzas, ok - basically based on dansmith 's comment, i decided to chuck the MonitorMetricValue Object and decided to add another field for the numa_nodes_values into the MonitorMetric object itself.16:55
mriedemmtreinish: is there any reason to build that out?16:55
mtreinishmriedem: but, I've never done anything with that output16:55
mriedemyeah it's been there since 201216:55
bauzassudipto: yeah, remember what we discussed this morning16:55
mtreinishmriedem: a local doc16:55
mtreinishwithout needing a web browser16:55
bauzasmriedem: we also build man pages IIRC16:56
mriedemrequires that you have latex locally to build it too, which sucks16:56
*** tongli has quit IRC16:56
mtreinishbauzas: yep we do16:56
sudiptobauzas, yeah i have taken care of the make_compat* method, but just wondering if for existing UTs we should just introduce this field and set it to None...16:56
sudiptobauzas, let me put out a patch, maybe better that way...16:56
bauzasmriedem: so I don't really see a problem with having latex support, provided it doesn't use more time than needed if we don't have latex16:56
*** davideagnello has joined #openstack-nova16:56
*** tongli has joined #openstack-nova16:56
mriedembauzas: the problem is just having to install the package from the distro manually16:57
mriedemwhich kind of sucks16:57
bauzasmriedem: that's not like if we were having a numpy dependency for a websockets stuff in Nova :p16:57
bauzassorry websockify16:57
*** diogogmt has joined #openstack-nova16:57
bauzasmriedem: that doesn't block the RST generation, does it ?16:57
mriedembauzas: it does if you set autodoc_index_modules and warnerrors=True in setup.cfg16:58
mriedemand run tox -e docs16:58
bauzasmriedem: I mean, I built some docs a couple of times and I don't remember anything about that16:58
bauzasmeh16:58
*** lucas__ has joined #openstack-nova16:58
mriedembauzas: that's b/c you're not building it all right now16:58
mriedemi'm trying to get that to work again16:58
bauzasmriedem: yup gotcha16:59
*** lucas_ has quit IRC16:59
bauzasmriedem: I see, no pip install, then ?16:59
bauzasnvm, I said something silly17:00
* bauzas is hiding 17:00
*** afazekas has quit IRC17:01
*** moshele has joined #openstack-nova17:01
*** lucas__ has quit IRC17:01
mriedemlunch time17:01
*** tongli has quit IRC17:01
*** lucas_ has joined #openstack-nova17:01
*** tongli has joined #openstack-nova17:04
*** ociuhandu has quit IRC17:04
*** lyanchih_ has quit IRC17:04
mtreinishmriedem: I just checked the latex output doesn't build. at least on my system17:06
mtreinishI'll dig into it later17:06
*** aloga_ has quit IRC17:06
*** aswadr has quit IRC17:07
*** aix has quit IRC17:08
*** piyanai has quit IRC17:08
*** salv-orl_ has joined #openstack-nova17:08
*** ildikov has joined #openstack-nova17:08
*** tongli has quit IRC17:09
*** angdraug has joined #openstack-nova17:09
*** garyk has quit IRC17:09
*** yamahata has quit IRC17:09
*** ajayaa has joined #openstack-nova17:10
*** salv-orlando has quit IRC17:11
*** gyee has joined #openstack-nova17:12
*** derekh has quit IRC17:12
openstackgerritNikola Dipanov proposed openstack/nova: Define migration states and transitions and use when changing status  https://review.openstack.org/19766917:14
*** piyanai has joined #openstack-nova17:15
*** ijw has joined #openstack-nova17:15
openstackgerritJesse J. Cook proposed openstack/nova: Cells: Decrement available slots  https://review.openstack.org/19742517:15
*** zzzeek has quit IRC17:16
*** lpetrut has quit IRC17:16
*** promulo__ has quit IRC17:16
*** promulo has joined #openstack-nova17:17
*** lpetrut has joined #openstack-nova17:17
*** tongli has joined #openstack-nova17:17
mtreinishmriedem: it's the unicode check marks and x's in the devref17:18
*** david-lyle has quit IRC17:19
*** _gryf|afk is now known as _gryf17:20
openstackgerritMarian Horban proposed openstack/nova: Switch to oslo.service library  https://review.openstack.org/19290017:20
*** tongli has quit IRC17:22
*** browne has quit IRC17:22
*** kiran-r has quit IRC17:23
mtreinishmriedem: err and the svg architecture diagram, latex didn't understand svg as a graphics format17:24
*** ijw has quit IRC17:24
*** lucas_ has quit IRC17:24
mtreinishremove those 2 bits and it worked17:24
*** achanda has joined #openstack-nova17:25
*** Mike_D_laptop has joined #openstack-nova17:25
*** tjones1 has quit IRC17:26
mtreinishmriedem: http://blog.kortar.org/wp-content/uploads/2015/07/Nova.pdf17:27
mtreinishI just removed the hypervisor support matrix and the architecture diagram for that17:28
mtreinishwe totally need to add an infra job that publishes pdfs for the docs17:29
*** lpetrut has quit IRC17:29
jrollandrearosa: hey, I'm looking at https://bugs.launchpad.net/nova/+bug/1449221 and wondering if you've looked into making exception handling better such that it just doesn't get stuck detaching?17:29
openstackLaunchpad bug 1449221 in OpenStack Compute (nova) "Nova volume-detach lacks '--force' command for cleanup" [Undecided,In progress] - Assigned to Andrea Rosa (andrea-rosa-m)17:29
jrollandrearosa: e.g. I'm seeing this if for example the cinder detach call fails17:29
jrolland I'd rather it just error or not get stuck17:29
*** annegentle has joined #openstack-nova17:30
*** Nikolay_St has quit IRC17:33
*** zzzeek has joined #openstack-nova17:35
*** Nikolay_St has joined #openstack-nova17:35
*** annegentle has quit IRC17:35
*** piyanai has quit IRC17:36
*** lucas_ has joined #openstack-nova17:38
mriedemmtreinish: ha17:39
*** piyanai has joined #openstack-nova17:39
*** piyanai has quit IRC17:39
mriedemmtreinish: well, i removed the sphinx.ext.pngmath extension but then sphinx complains that i don't have that but i'm trying to build math stuff with the html builder17:40
*** piyanai has joined #openstack-nova17:40
mriedemmtreinish: i can't really just remove the arch diagram or hypervisor support matrix17:40
*** rlrossit has joined #openstack-nova17:41
*** SourabhP has joined #openstack-nova17:42
*** Nikolay_St has quit IRC17:42
*** yamahata has joined #openstack-nova17:44
*** armax has joined #openstack-nova17:46
mriedemmrodden: i found the math culprit17:46
mriedemhttps://github.com/openstack/nova/blob/master/nova/virt/hyperv/vhdutilsv2.py#L12917:46
*** SourabhP has quit IRC17:48
*** SourabhP has joined #openstack-nova17:48
*** Nikolay_St has joined #openstack-nova17:49
*** moshele has quit IRC17:49
*** moshele has joined #openstack-nova17:50
*** Longgeek has joined #openstack-nova17:53
ndipanovhmmm dansmith I see that with this https://review.openstack.org/#/c/183354/12/nova/compute/api.py we don't get the same thing for the rebuild command17:55
ndipanovis that by design or?17:55
openstackgerritSylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests()  https://review.openstack.org/19920517:56
*** moshele has quit IRC17:57
*** browne has joined #openstack-nova17:59
*** agireud has quit IRC17:59
openstackgerritSudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors  https://review.openstack.org/19897218:00
*** mgarza_ has quit IRC18:00
dansmithndipanov: still onthis call18:00
ndipanovdansmith, kk - ping when done18:01
dansmith...if ever18:01
mriedemjust hang up18:02
*** agireud has joined #openstack-nova18:02
ndipanovdansmith, or start crying and sobbing without mute18:02
ndipanovand then just excuse yourself18:02
ndipanovno one would question your reasons at that point18:03
*** marzif_ has joined #openstack-nova18:03
*** annashen has quit IRC18:03
*** agireud has quit IRC18:04
*** tongli has joined #openstack-nova18:04
*** Longgeek has quit IRC18:04
*** ajayaa has quit IRC18:04
*** tonytan4ever has joined #openstack-nova18:04
*** agireud has joined #openstack-nova18:04
*** coolsvap|away is now known as coolsvap18:04
*** READ10 has joined #openstack-nova18:05
*** moshele has joined #openstack-nova18:05
*** changbl has quit IRC18:05
*** jwcroppe has quit IRC18:05
*** annashen has joined #openstack-nova18:05
*** jwcroppe has joined #openstack-nova18:06
*** mgarza_ has joined #openstack-nova18:06
*** moshele has quit IRC18:07
*** tongli has quit IRC18:09
openstackgerritEd Leafe proposed openstack/nova: Add logging when filtering returns nothing.  https://review.openstack.org/19218418:11
*** hogepodge has joined #openstack-nova18:11
*** tongli has joined #openstack-nova18:11
*** irenab has joined #openstack-nova18:13
*** Nikolay_St has quit IRC18:16
*** ZZelle has quit IRC18:16
*** lpetrut has joined #openstack-nova18:16
mtreinishmriedem: sure, that was just the quick solution18:17
mtreinishmriedem: I'm not sure how to fix the unicode thing, but the svg thing is simple enough18:18
mriedemmtreinish: texlive-latex-extra18:18
mriedemprovides utfx.def18:18
mtreinishjust convert it to an image format latex understands before building18:18
mriedemmtreinish: i can drop the pngmath extension if i just remove .. math:: from the hyper-v docstrings18:18
mtreinishI should have the texlive extra package installed, let me double check18:19
mriedemmtreinish: the thing is, i doubt the vm's that build the nova-python-docs jobs have these packages on them18:19
mtreinishmriedem: that's easy enough to fix though18:19
dansmithndipanov: not done yet, but finishing up18:20
dansmithndipanov: you're asking why rebuild doesn't create and update a migration object?18:20
*** hogepodge has quit IRC18:20
mriedemmtreinish: some project-config thing for the job?18:20
*** marzif_ has quit IRC18:21
mtreinishmriedem: heh, yeah more or less18:21
ndipanovdansmith, I was yes18:21
dansmithndipanov: it could, but a rebuild doesn't normally move the instance unless it's an evacuation, right?18:22
*** salv-orl_ has quit IRC18:22
ndipanovdansmith, hmmm18:22
* ndipanov checks18:22
*** lpetrut has quit IRC18:23
*** annashen has quit IRC18:23
ndipanovyes that is correct18:23
*** jwcroppe has quit IRC18:23
ndipanovfair enough18:24
*** annashen has joined #openstack-nova18:24
*** piyanai has quit IRC18:25
*** annashen has quit IRC18:28
*** whenry has quit IRC18:31
*** tjones1 has joined #openstack-nova18:31
openstackgerritRoman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down  https://review.openstack.org/18408618:31
*** piyanai has joined #openstack-nova18:33
*** sudipto has quit IRC18:35
*** lpetrut has joined #openstack-nova18:36
*** signed8bit is now known as signed8bit_ZZZzz18:37
*** mgarza_ has quit IRC18:38
*** veena has quit IRC18:38
*** mdrabe has joined #openstack-nova18:39
*** agireud has quit IRC18:39
*** lpetrut has quit IRC18:42
*** agireud has joined #openstack-nova18:43
*** mgarza_ has joined #openstack-nova18:45
*** Sukhdev has joined #openstack-nova18:45
*** iamjarvo has quit IRC18:48
*** salv-orlando has joined #openstack-nova18:49
*** yassine_ has quit IRC18:49
*** agireud has quit IRC18:50
*** achanda has quit IRC18:51
*** Nikolay_St has joined #openstack-nova18:52
*** baoli has quit IRC18:52
*** dims has quit IRC18:53
*** mriedem has quit IRC18:55
*** baoli has joined #openstack-nova18:56
*** annegentle has joined #openstack-nova18:56
*** zsais has quit IRC18:57
*** dims has joined #openstack-nova18:58
*** hogepodge has joined #openstack-nova19:00
*** pixelbeat has quit IRC19:00
openstackgerritNikola Dipanov proposed openstack/nova: libvirt: implement get_device_name_for_instance  https://review.openstack.org/18063819:01
*** ildikov has quit IRC19:02
openstackgerritRoman Dobosz proposed openstack/nova: Introducing new forced_down field for a Service object  https://review.openstack.org/18528019:02
openstackgerritRoman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down  https://review.openstack.org/18408619:02
openstackgerritRoman Dobosz proposed openstack/nova: Service group drivers forced_down flag utilization  https://review.openstack.org/19706319:02
*** harlowja_ has joined #openstack-nova19:02
*** whenry has joined #openstack-nova19:02
*** harlowja has quit IRC19:03
*** armax has quit IRC19:03
*** ildikov has joined #openstack-nova19:03
openstackgerritEric Brown proposed openstack/nova: Replace openssl calls with cryptography lib  https://review.openstack.org/19824619:04
*** ociuhandu has joined #openstack-nova19:05
*** mriedem has joined #openstack-nova19:06
*** annashen has joined #openstack-nova19:06
*** nihilifer has joined #openstack-nova19:09
*** armax has joined #openstack-nova19:10
*** agireud has joined #openstack-nova19:12
mriedemjaypipes: wow, ever seen this? https://github.com/openstack/nova/blob/master/nova/tests/unit/scheduler/test_caching_scheduler.py#L16919:13
mriedemi guess it's john's19:13
*** ayoung is now known as ayoung-afk19:13
*** obondarev has quit IRC19:14
jaypipesmriedem: no, never seen that...19:14
*** mgarza_ has quit IRC19:14
mriedemi was trying to see what was using graphviz, if anything19:14
*** signed8bit_ZZZzz is now known as signed8bit19:14
*** obondarev has joined #openstack-nova19:15
*** ndipanov has quit IRC19:20
*** dguitarbite has quit IRC19:21
*** changbl has joined #openstack-nova19:22
*** mgarza has joined #openstack-nova19:24
*** mriedem has quit IRC19:25
*** signed8bit is now known as signed8bit_ZZZzz19:28
*** signed8bit_ZZZzz is now known as signed8bit19:28
*** jichen has joined #openstack-nova19:29
*** mriedem has joined #openstack-nova19:29
*** agireud has quit IRC19:30
*** READ10 has quit IRC19:35
edleafeGetting a failure in the check-tempest-dsvm-cells for https://review.openstack.org/#/c/192184/19:35
edleafeI see this line in the logs: WARNING: DEVSTACK_GATE_SETTINGS file does not exist: '/opt/stack/new/nova/devstack/tempest-dsvm-cells-rc'19:36
edleafeAny ideas on how to debug further?19:36
*** tonytan4ever has quit IRC19:37
*** achanda has joined #openstack-nova19:37
*** Mike_D_laptop has quit IRC19:37
*** piyanai has quit IRC19:38
melwittedleafe: looking19:39
*** piyanai has joined #openstack-nova19:39
rlrossitdansmith: not sure if you want to play around with this guy but it's at least up there and I *think* this is the solution https://review.openstack.org/#/c/198904/19:40
dansmithrlrossit: oh yeah, let me pull it down and play with it19:40
*** tonytan4ever has joined #openstack-nova19:42
*** angdraug has quit IRC19:45
*** ajayaa has joined #openstack-nova19:45
*** agireud has joined #openstack-nova19:46
*** dims has quit IRC19:48
*** david-lyle has joined #openstack-nova19:50
mriedemmrodden: surprise surprise, excluding nova.tests.* makes the docs build faster :)19:50
dansmithrlrossit: left some comments, see if it makes sense19:51
*** ayoung-afk is now known as ayoung19:52
*** agireud has quit IRC19:52
*** jwcroppe has joined #openstack-nova19:53
rlrossitdansmith: I'll take a look. thanks!19:54
_gryfdansmith, could you take a look at https://review.openstack.org/#/c/185279/ and make it flow :)?19:54
dansmith_gryf: make it flow?19:54
mriedemlike flow rida19:55
mriedems/flow/flo/19:55
mriedemsorry19:55
_gryfdansmith, err, sorry there is a couple +1, but not +2…19:55
melwittedleafe: talking about it in #openstack-qa, looks like something unrelated to the cells job is wrong19:55
edleafemelwitt: ok, thanks.19:56
*** VW_ has quit IRC19:56
melwittedleafe: possibly a bad node, it failed to clone anything19:57
edleafemelwitt: yeah, from the limited info it looked like a bad configuration or a bad node19:58
melwittedleafe: yeah, I wasn't sure so I asked in -qa. I couldn't tell if it was a separate error or if the failure to find the devstack settings rc file caused it to stop there19:59
*** VW_ has joined #openstack-nova19:59
*** VW_ has quit IRC19:59
*** VW_ has joined #openstack-nova20:00
rlrossitdansmith: so if I'm reading this correctly, we should be building the tree once with versions in obj_relationships, and then building it again with the current subobject versions, and then comparing those trees instead of comparing against the static tree?20:00
*** agireud has joined #openstack-nova20:00
*** mgarza has quit IRC20:00
*** sahid has quit IRC20:01
*** ivasev has quit IRC20:01
*** Longgeek has joined #openstack-nova20:01
dansmithrlrossit: that's one way to do it, sure20:01
melwittedleafe: I'm making an e-r query for it right now, you can just recheck yours20:01
dansmithrlrossit: the nice thing about building both trees is that you can diff the whole thing and show the developer what all changed, instead of just failing on the first one, then failing again on the next one after they fix the first, etc20:02
edleafemelwitt: ok, but should I wait until the problem is fixed?20:02
*** zsais has joined #openstack-nova20:02
mriedemedleafe: just recheck20:03
mriedemit's a bad node probably20:03
*** mgarza_ has joined #openstack-nova20:03
rlrossitdansmith: so I'm trying to understand this from the standpoint of using the objects too... so if we up the version of an object, we also need to up the versions of everything that holds that object also to now use the most recent version?20:03
edleafemriedem: thanks. didn't want to pile on if there was a problem20:03
*** irenab has quit IRC20:04
*** irenab has joined #openstack-nova20:05
*** nic has quit IRC20:05
_gryfdansmith, thx :D20:05
*** jichen_ has joined #openstack-nova20:06
melwittedleafe: it turns out we can't do an e-r query either because it doesn't have fingerprintable logging that we know of https://bugs.launchpad.net/openstack-ci/+bug/135482920:06
openstackLaunchpad bug 1354829 in OpenStack Core Infrastructure "Booting from stale snapshots in rax-dfw leads to miscellaneous failures" [Critical,In progress] - Assigned to Jeremy Stanley (fungi)20:06
dansmithrlrossit: it's actually not "to use the most recent version"20:06
dansmithrlrossit: the rest of what you said is right, but the mapping is actually for backports20:06
*** _gryf is now known as _gryf|afk20:07
dansmithrlrossit: where this comes into play is: say we send an Instance object from a new nova-api worker to an older nova-compute worker20:07
*** marzif_ has joined #openstack-nova20:07
dansmithrlrossit: the newer API knows about v1.2 of the instance, but the older compute only knows about 1.120:07
dansmithrlrossit: so we have to be able to backport the object to 1.1 for the older node20:07
dansmithrlrossit: now, when we do that, we kindof "page fault" the object over to conductor to do the actual backporting20:07
*** armax has quit IRC20:08
dansmithrlrossit: if we did that because instance was 1.2, only to find out that instancefault was also 1.7 and we only know about 1.6, then we have to kick that to conductor,20:08
dansmithrlrossit: only to find out that info_cache is 1.8 and we only support 1.2, etc20:08
*** lucas-dinner has quit IRC20:08
dansmithrlrossit: so the goal of the mapping is to say "okay, he wants an instance 1.1, that means info_cache needs to be 1.2 and fault needs to be 1.6" so we can do it all at once20:09
*** jichen has quit IRC20:09
*** jichen_ is now known as jichen20:09
dansmithrlrossit: those are all made up numbers of course20:09
rlrossitdansmith: so possibly stupid question... how does this play into something like floatingip having a stale version of fixedip? why do we need to update the version of fixed ip held in floatingip, even if floatingip doesn't "need" it? is it just to totally prevent backporting?20:12
rlrossitI'm trying to fit this situation into what you said above20:12
dansmithrlrossit: to prevent two backports instead of one, yeah20:13
dansmithrlrossit: so if I pass a FloatingIP to an older compute, and that FloatingIP includes a FixedIP,20:13
dansmithrlrossit: it will ask conductor to backport FloatingIP to some version, and conductor will not backport the fixed inside it, because it's not in the mapping20:13
dansmithor, won't backport it to the right spot, depending on what the gap looks like20:14
rlrossitdansmith: so cascading backports == bad?20:14
dansmithwell, depends on what you mean.. I consider the correct situation to be cascading, because we're asked to do the top one, so we just do it all the way down the hierarchy in one step20:15
*** fawadkhaliq has quit IRC20:15
dansmithbut yes, if we send an object (with a subobject) to a compute and it has to ask to backport the top thing, and then again to backport the bottom thing, that's really expensive20:16
dansmitha backport is already very expensive20:16
*** marzif__ has joined #openstack-nova20:16
rlrossitalrighty I think I get it now. Thanks!20:16
dansmiththe other detail here is that conductor doesn't actually support doing the second backport, because it depends on this mapping being right20:16
*** armax has joined #openstack-nova20:17
dansmithso if we ask for the first backport and then encounter another in the same message, we'll ask conductor to backport the top again and conductor will look at us like o.O20:17
*** lpetrut has joined #openstack-nova20:17
*** ijw has joined #openstack-nova20:18
*** tongli has quit IRC20:18
*** marzif_ has quit IRC20:18
*** thangp has quit IRC20:19
*** salv-orlando has quit IRC20:20
*** lucasagomes has joined #openstack-nova20:21
*** penick has joined #openstack-nova20:22
*** piyanai has quit IRC20:23
*** Longgeek has quit IRC20:24
*** marzif__ has quit IRC20:26
*** lucasagomes has quit IRC20:27
*** htruta has quit IRC20:31
*** dprince has quit IRC20:34
*** htruta has joined #openstack-nova20:34
*** Piet has quit IRC20:34
*** nic has joined #openstack-nova20:35
*** coolsvap is now known as coolsvap|away20:39
*** jamielennox is now known as jamielennox|away20:41
*** piyanai has joined #openstack-nova20:43
*** jdandrea has joined #openstack-nova20:45
jdandreaQ: Is it possible to get nova to work with a remote OpenStack installation in a manner similar to Heat's Standalone Mode? (continued ...)20:45
*** Sukhdev has quit IRC20:45
jdandreaI ask because this comes in handy for development of Heat in conjunction with an existing cluster, and without adversely impacting users of that cluster (e.g., I could use it for testing and tracing scheduler filter plugins, for instance). Context: http://docs.openstack.org/developer/heat/getting_started/standalone.html20:45
*** erlon has joined #openstack-nova20:46
*** jasondotstar has joined #openstack-nova20:50
*** jamielennox|away is now known as jamielennox20:52
*** jwcroppe has quit IRC20:53
*** jskinner has quit IRC20:54
*** whenry has quit IRC20:54
*** igordcard_ has joined #openstack-nova20:56
mriedemjogo: i'm cooking up an epic docs thing that i think you'll appreciate20:57
mriedemsoon my precious, soon20:57
*** marzif__ has joined #openstack-nova20:58
*** agireud has quit IRC20:58
*** exploreshaifali has quit IRC21:00
*** jdurgin2 has joined #openstack-nova21:01
*** mutoulbj has quit IRC21:01
*** mutoulbj has joined #openstack-nova21:01
*** mutoulbj has quit IRC21:01
*** jdurgin has quit IRC21:02
*** irenab has quit IRC21:03
*** thorst has quit IRC21:03
*** jwcroppe has joined #openstack-nova21:04
*** rdo has quit IRC21:04
*** lucas_ has quit IRC21:06
*** rdo has joined #openstack-nova21:06
*** aysyd has quit IRC21:07
*** jwcroppe_ has joined #openstack-nova21:07
*** lucas_ has joined #openstack-nova21:09
*** jichen has quit IRC21:09
*** jwcroppe has quit IRC21:09
jogomriedem: what is it a recipe for?21:11
mriedemautodoc_index_modules=True21:11
mriedemand making that work w/o exploding tox -e docs21:11
mriedemor putting the entire module index in the index.rst21:11
jogomriedem: woot!21:12
jogoI'll even review that patch21:13
*** dims has joined #openstack-nova21:13
mriedemi know, you're the only one who would actually be interested :)21:13
mriedemand maybe mtreinish21:13
*** lucas_ has quit IRC21:13
*** ociuhandu has quit IRC21:13
mtreinishmriedem: I'm working on a script to fix the latex building21:13
openstackgerritMatt Riedemann proposed openstack/nova: Set autodoc_index_modules=True so tox -e docs builds module docs again  https://review.openstack.org/19929721:14
mriedemmtreinish: once i removed the math stuff latex wasn't a problem21:14
mriedemthere is the change ^21:14
*** piyanai has quit IRC21:14
mtreinishthere are obvious sphinx bugs we have to report in the latex output21:14
mtreinishmriedem: did you try to compile the latex?21:14
mriedemno21:14
mtreinishsphinx is generating invalid latex21:14
mtreinishthat's what I meant :)21:14
*** piyanai has joined #openstack-nova21:14
*** lucas_ has joined #openstack-nova21:15
* jogo wants more latex in his life21:15
jogoI used to write all my homework in it21:15
*** jwcroppe has joined #openstack-nova21:16
*** jasondotstar has quit IRC21:16
mtreinishjogo: you could use it for presentations like I do21:16
mtreinishjogo: http://blog.kortar.org/wp-content/uploads/2015/07/Nova.pdf21:17
*** armax has quit IRC21:17
*** piyanai has quit IRC21:17
*** jwcroppe_ has quit IRC21:17
*** annegentle has quit IRC21:18
*** Piet has joined #openstack-nova21:19
mtreinishI've got an updated version that has the support matrix and your architecture diagram included, but I've got to find a latex equiv for the unicode x mark that sphinx is using21:19
*** lucas_ has quit IRC21:19
*** ctrath has quit IRC21:20
*** lucas_ has joined #openstack-nova21:20
*** salv-orlando has joined #openstack-nova21:22
*** lucas__ has joined #openstack-nova21:22
*** agireud has joined #openstack-nova21:23
jogomtreinish: haha nice21:24
mtreinishjogo: oh and file sphinx bugs because it generates bogus latex21:24
*** lucas_ has quit IRC21:24
*** dboik has quit IRC21:25
mriedemwell i finally found out the correct form for :param and :returns21:25
mriedem:param thing:21:25
mriedem:returns:21:25
mriedemanything else doesn't render properly21:25
*** marzif_ has joined #openstack-nova21:26
*** marzif__ has quit IRC21:26
*** lucas__ has quit IRC21:27
mtreinishmriedem: yep, I think you can use the java one with the @ too. I remember seeing that somewhere21:27
mtreinishbut I could be making that up21:27
mriedemmtreinish: @ doesn't work i don't think21:27
mriedemi've tried before21:27
mriedemsince i was used to javadoc21:27
melwittI have seen the java style @param in docstring21:27
mriedemand rendered properly?21:27
*** VW_ has quit IRC21:28
melwittno, I mean I've seen it in the code. if it won't render that's good to know so I can look for that in reviews21:28
mtreinishmriedem, melwitt: yeah maybe I just saw it in code but autodoc probably wasn't turned on21:28
*** eharney has quit IRC21:29
*** lucas_ has joined #openstack-nova21:29
*** salv-orlando has quit IRC21:29
mtreinishmriedem: but yeah ':param type thing: desc', ':returns: desc', ':rtype: type', etc is what is correct21:29
*** salv-orlando has joined #openstack-nova21:30
*** dims has quit IRC21:30
mriedemmtreinish: god, i could just blast some stats by cleaning these all up huh :)21:30
mtreinishmriedem: go for it, I'll never be able to catch up :)21:31
mriedemi wonder if sphinx nitpick=True would complain about the wrong param stuff21:31
mriedemdoing it by hand would suck otherwise21:31
mtreinishmriedem: make sure you script it and do 1 patch per file21:32
mriedemmtreinish: good idea21:32
*** marzif_ has quit IRC21:32
melwitt:D21:32
*** VW_ has joined #openstack-nova21:33
*** lucas_ has quit IRC21:34
*** ssurana has quit IRC21:35
*** dims has joined #openstack-nova21:35
*** dims has quit IRC21:35
*** Guest7393 has joined #openstack-nova21:35
*** rlrossit has left #openstack-nova21:36
*** mriedem is now known as mriedem_away21:37
mtreinishjogo, mriedem_away: http://blog.kortar.org/wp-content/uploads/2015/07/Nova1.pdf now with the missing pieces21:39
mtreinishI just need to script fixing the latex and I'll push a patch with it21:40
*** karimb has joined #openstack-nova21:44
*** browne has quit IRC21:44
openstackgerritMichal Rostecki proposed openstack/nova: Dedicate aggregates for specific tenants  https://review.openstack.org/19578321:45
jogomtreinish: I forgot how fun that diagram was21:48
*** agireud has quit IRC21:48
*** Longgeek has joined #openstack-nova21:51
*** piyanai has joined #openstack-nova21:52
*** tonytan4ever has quit IRC21:52
*** lpetrut has quit IRC21:54
*** radek__ has quit IRC21:55
*** tonytan4ever has joined #openstack-nova21:56
*** mgarza_ has quit IRC21:57
*** annegentle has joined #openstack-nova21:57
*** Longgeek has quit IRC21:57
*** zsais has quit IRC21:58
*** changbl has quit IRC21:59
*** mgarza has joined #openstack-nova21:59
*** andrewbogott is now known as andrewbogott_afk22:00
*** jwcroppe_ has joined #openstack-nova22:04
*** burt has quit IRC22:04
*** nihilifer has quit IRC22:05
*** zzzeek has quit IRC22:05
*** jwcroppe has quit IRC22:06
*** zzzeek has joined #openstack-nova22:07
*** thorst has joined #openstack-nova22:07
*** mdrabe has quit IRC22:10
*** unicell has quit IRC22:10
*** unicell has joined #openstack-nova22:10
openstackgerritWalter A. Boring IV (hemna) proposed openstack/nova: Switch to using os-brick  https://review.openstack.org/17556922:12
*** jecarey has quit IRC22:13
*** ijw_ has joined #openstack-nova22:16
*** thorst has quit IRC22:18
*** ijw has quit IRC22:19
*** Raildo_ has joined #openstack-nova22:20
*** fawadkhaliq has joined #openstack-nova22:22
*** jwcroppe_ has quit IRC22:22
*** jwcroppe has joined #openstack-nova22:23
*** Nikolay_St has quit IRC22:26
*** jdurgin2 has quit IRC22:27
*** jdurgin has joined #openstack-nova22:27
*** wwriverrat has joined #openstack-nova22:28
*** jwcroppe has quit IRC22:28
*** armax has joined #openstack-nova22:28
*** annegentle has quit IRC22:29
wwriverratRespectfully requesting a core review of mindlessly simple patch: https://review.openstack.org/#/c/179978/22:29
*** piyanai has quit IRC22:32
*** edmondsw has quit IRC22:33
openstackgerritMatthew Treinish proposed openstack/nova: Add tool to build a doc latex pdf  https://review.openstack.org/19933222:35
*** browne has joined #openstack-nova22:37
*** GLaupre has joined #openstack-nova22:39
openstackgerritMatthew Treinish proposed openstack/nova: Add tool to build a doc latex pdf  https://review.openstack.org/19933222:39
mtreinishmriedem_away, jogo: ^^^22:40
mtreinishthe next step is to add the infra job to do something with that22:41
GLaupreHi all, can someone tell me where I can change the emulator use by libvirt? It seems that the XML config file use by default the PATH /usr/libexec. I have installed a new qemu and I want to add a 'emulator' parameter in the XML cfg file with my new PATH, Where the xml is built before the instance spawned? I don't find where I can add this?22:42
GLaupres/use/sued22:43
GLauprewell22:43
GLaupre:P22:43
*** Raildo_ has quit IRC22:43
*** jwcroppe has joined #openstack-nova22:44
*** Sukhdev has joined #openstack-nova22:45
mtreinishmriedem_away: http://blog.kortar.org/wp-content/uploads/2015/07/Nova2.pdf with your autodoc patch applied22:45
*** tonytan4ever has quit IRC22:48
mtreinishit doesn't really look any different22:48
*** pixelbeat has joined #openstack-nova22:49
*** annashen has quit IRC22:54
*** ijw_ has quit IRC22:56
*** jwcroppe has quit IRC22:56
*** ijw has joined #openstack-nova22:57
*** r-daneel has quit IRC23:00
*** tjones1 has left #openstack-nova23:01
*** fawadkhaliq has quit IRC23:02
openstackgerritJosh Gachnang proposed openstack/nova: Expose Neutron network data in metadata service  https://review.openstack.org/15309723:07
*** slong has joined #openstack-nova23:08
jaypipessdague, alaski: any chance of a review on https://review.openstack.org/#/c/196181/ ? :)23:11
jaypipesor any other cores! :)23:11
*** thedodd has quit IRC23:11
*** annegentle has joined #openstack-nova23:12
*** zhenguo has joined #openstack-nova23:14
*** hemna is now known as hemnafk23:16
*** mleroy has quit IRC23:18
*** nelsnelson has quit IRC23:20
*** ssurana has joined #openstack-nova23:20
*** salv-orlando has quit IRC23:26
*** krtaylor has quit IRC23:29
*** diogogmt has quit IRC23:30
*** ijw has quit IRC23:31
*** ijw has joined #openstack-nova23:33
*** annegentle has quit IRC23:34
*** diogogmt has joined #openstack-nova23:35
*** vladikr has quit IRC23:37
*** krtaylor has joined #openstack-nova23:41
*** chaochin has quit IRC23:42
*** chaochin has joined #openstack-nova23:42
*** Longgeek has joined #openstack-nova23:44
openstackgerritJosh Gachnang proposed openstack/nova: Build Neutron network data for metadata service  https://review.openstack.org/10264923:46
*** Longgeek has quit IRC23:49
*** erlon has quit IRC23:50
*** jasondotstar has joined #openstack-nova23:51
*** gyee has quit IRC23:51
*** baoli has quit IRC23:53
*** IanGovett has quit IRC23:54
*** karimb has quit IRC23:56
*** jwcroppe has joined #openstack-nova23:57

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