Friday, 2018-07-06

*** heyongli has quit IRC00:03
*** heyongli has joined #openstack-nova00:03
*** shaohe_feng has quit IRC00:05
*** shaohe_feng has joined #openstack-nova00:06
*** edmondsw has joined #openstack-nova00:07
*** edmondsw has quit IRC00:12
melwittalex_xu, gmann: I wanted to get your thoughts on this problem http://lists.openstack.org/pipermail/openstack-dev/2018-July/132021.html having to do with nova-api redirect from /v2.1 => /v2.1/ being absolute URL00:12
*** moshele has joined #openstack-nova00:13
*** hongbin has joined #openstack-nova00:13
*** heyongli has quit IRC00:13
*** heyongli has joined #openstack-nova00:14
*** shaohe_feng has quit IRC00:16
*** shaohe_feng has joined #openstack-nova00:17
gmannmelwitt: ack, will check today00:17
melwittthanks gmann. what's possible with routes redirect capabilities are beyond my knowledge :)00:18
*** hemna_ has joined #openstack-nova00:23
*** heyongli has quit IRC00:23
*** heyongli has joined #openstack-nova00:24
*** shaohe_feng has quit IRC00:26
*** shaohe_feng has joined #openstack-nova00:27
*** medberry has joined #openstack-nova00:34
*** heyongli has quit IRC00:34
*** hemna_ has quit IRC00:34
*** medberry is now known as Guest6213100:34
*** heyongli has joined #openstack-nova00:34
*** shaohe_feng has quit IRC00:36
*** shaohe_feng has joined #openstack-nova00:37
*** claudiub has quit IRC00:40
*** heyongli has quit IRC00:44
*** heyongli has joined #openstack-nova00:44
*** shaohe_feng has quit IRC00:46
*** shaohe_feng has joined #openstack-nova00:47
*** oanson has quit IRC00:54
*** heyongli has quit IRC00:54
openstackgerritMerged openstack/nova master: Use valid UUID in the placement gabbits  https://review.openstack.org/58035700:55
*** heyongli has joined #openstack-nova00:55
*** oanson has joined #openstack-nova00:55
openstackgerritMerged openstack/nova master: cover migration cases with functional tests  https://review.openstack.org/49386500:55
*** ttsiouts has quit IRC00:55
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (5)  https://review.openstack.org/57411000:56
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (6)  https://review.openstack.org/57411300:57
*** shaohe_feng has quit IRC00:57
*** Guest62131 has quit IRC00:57
*** shaohe_feng has joined #openstack-nova00:57
*** moshele has quit IRC00:58
*** heyongli has quit IRC01:04
*** heyongli has joined #openstack-nova01:05
*** yedongcan has joined #openstack-nova01:06
*** shaohe_feng has quit IRC01:07
*** shaohe_feng has joined #openstack-nova01:09
*** mrsoul has joined #openstack-nova01:11
*** mrsoul_ has joined #openstack-nova01:11
*** tiendc has joined #openstack-nova01:12
*** Guest62131 has joined #openstack-nova01:13
*** mrsoul` has quit IRC01:14
*** mschuppert has quit IRC01:15
*** heyongli has quit IRC01:15
*** heyongli has joined #openstack-nova01:15
*** shaohe_feng has quit IRC01:17
openstackgerritMerged openstack/nova master: VMware: save VC reads for information that is static  https://review.openstack.org/57455201:17
*** shaohe_feng has joined #openstack-nova01:20
*** heyongli has quit IRC01:25
*** heyongli has joined #openstack-nova01:25
*** shaohe_feng has quit IRC01:27
*** Guest62131 has quit IRC01:27
*** shaohe_feng has joined #openstack-nova01:28
*** heyongli has quit IRC01:35
*** heyongli has joined #openstack-nova01:36
*** shaohe_feng has quit IRC01:38
*** yamahata has quit IRC01:39
*** shaohe_feng has joined #openstack-nova01:39
*** gyee has quit IRC01:45
*** s10 has quit IRC01:45
*** heyongli has quit IRC01:45
*** heyongli has joined #openstack-nova01:46
*** shaohe_feng has quit IRC01:48
*** shaohe_feng has joined #openstack-nova01:49
*** edmondsw has joined #openstack-nova01:55
*** heyongli has quit IRC01:56
*** heyongli has joined #openstack-nova01:56
*** shaohe_feng has quit IRC01:58
*** shaohe_feng has joined #openstack-nova01:59
*** edmondsw has quit IRC02:00
*** heyongli has quit IRC02:06
*** heyongli has joined #openstack-nova02:06
*** Guest62131 has joined #openstack-nova02:08
*** shaohe_feng has quit IRC02:08
*** shaohe_feng has joined #openstack-nova02:09
gmannmelwitt: alex_xu not sure why we did redirect relative path to obsolete but it seems legacy code not just while moving to route - https://github.com/openstack/nova/blob/stable/ocata/nova/api/openstack/compute/versionsV21.py#L5902:14
gmanni had quick glance and i think we should be able to support relative patch also02:14
gmann we shouold be able to map /v2.1 to 'GET': [version_controller, 'show']02:15
gmannlet me try that out02:15
*** lei-zh has joined #openstack-nova02:15
*** heyongli has quit IRC02:16
*** heyongli has joined #openstack-nova02:17
alex_xugmann: melwitt I'm also checking the RFC, it said the Location value is absolute URL https://tools.ietf.org/html/rfc2616.html#section-14.3002:17
*** shaohe_feng has quit IRC02:19
*** shaohe_feng has joined #openstack-nova02:19
*** annp has joined #openstack-nova02:19
gmannalex_xu: yeah, that is correct for our case also. location value is absolete url something like https://example.com/compute/v2.1/02:22
*** Guest90874 has quit IRC02:23
gmannmelwitt: alex_xu issue what i understood is redirecting from relative to absolete by API route caused https->http. so if we support relative path also it fix the issue. correct me if wrong .02:25
*** heyongli has quit IRC02:26
*** heyongli has joined #openstack-nova02:27
alex_xugmann: melwitt sorry, found rfc7231 update the rfc2616 https://tools.ietf.org/html/rfc7231#page-68, the relative url is ok02:27
alex_xugmann: so we are ok to fix with relative url I think02:28
*** shaohe_feng has quit IRC02:29
gmannalex_xu: yeah02:29
*** shaohe_feng has joined #openstack-nova02:31
alex_xugmann: we have a lot of places use absolute url https://github.com/openstack/nova/blob/master/doc/api_samples/servers/server-get-resp.json#L2002:31
*** yamamoto has quit IRC02:33
gmannohk02:34
*** jmlowe has quit IRC02:34
openstackgerritGhanshyam Mann proposed openstack/nova master: WIP Support Relative URL for version API  https://review.openstack.org/58054402:35
gmannmelwitt: alex_xu something like this ^^ and if that seems ok then we can discuss to introduce as microversion or directly ?02:35
openstackgerritGhanshyam Mann proposed openstack/nova master: WIP Support Relative URL for version API  https://review.openstack.org/58054402:36
*** mdnadeem has joined #openstack-nova02:36
*** heyongli has quit IRC02:37
*** heyongli has joined #openstack-nova02:37
*** Guest62131 has quit IRC02:38
*** lei-zh has quit IRC02:39
*** shaohe_feng has quit IRC02:39
*** lei-zh has joined #openstack-nova02:39
*** psachin` has joined #openstack-nova02:40
*** shaohe_feng has joined #openstack-nova02:41
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (7)  https://review.openstack.org/57497402:42
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (8)  https://review.openstack.org/57531102:43
*** heyongli has quit IRC02:47
openstackgerritGhanshyam Mann proposed openstack/nova master: WIP: Stop redirect of '' to '/' for version API  https://review.openstack.org/58054402:47
*** heyongli has joined #openstack-nova02:47
*** yamamoto has joined #openstack-nova02:47
*** shaohe_feng has quit IRC02:49
*** shaohe_feng has joined #openstack-nova02:50
*** yamamoto has quit IRC02:53
*** udesale has joined #openstack-nova02:55
*** heyongli has quit IRC02:57
*** heyongli has joined #openstack-nova02:58
*** shaohe_feng has quit IRC03:00
*** shaohe_feng has joined #openstack-nova03:01
*** yamamoto has joined #openstack-nova03:03
*** yamamoto has quit IRC03:07
*** heyongli has quit IRC03:07
*** heyongli has joined #openstack-nova03:08
*** shaohe_feng has quit IRC03:10
*** shaohe_feng has joined #openstack-nova03:11
*** heyongli has quit IRC03:18
*** yamamoto has joined #openstack-nova03:18
*** heyongli has joined #openstack-nova03:18
*** shaohe_feng has quit IRC03:20
*** shaohe_feng has joined #openstack-nova03:21
*** yamamoto has quit IRC03:22
*** mdnadeem has quit IRC03:24
*** yamahata has joined #openstack-nova03:26
*** heyongli has quit IRC03:28
*** blkart has quit IRC03:28
*** heyongli has joined #openstack-nova03:28
*** shaohe_feng has quit IRC03:30
*** shaohe_feng has joined #openstack-nova03:32
*** yamamoto has joined #openstack-nova03:33
*** yamamoto has quit IRC03:37
*** heyongli has quit IRC03:38
*** heyongli has joined #openstack-nova03:39
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (9)  https://review.openstack.org/57558103:40
*** mdnadeem has joined #openstack-nova03:40
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (10)  https://review.openstack.org/57601703:40
*** shaohe_feng has quit IRC03:41
*** blkart has joined #openstack-nova03:41
*** lei-zh has quit IRC03:41
*** shaohe_feng has joined #openstack-nova03:41
*** edmondsw has joined #openstack-nova03:43
*** edmondsw has quit IRC03:48
*** heyongli has quit IRC03:48
*** heyongli has joined #openstack-nova03:49
*** lpetrut_ has joined #openstack-nova03:49
*** shaohe_feng has quit IRC03:51
*** shaohe_feng has joined #openstack-nova03:52
*** hongbin has quit IRC03:52
*** Bhujay has joined #openstack-nova03:57
*** Bhujay has quit IRC03:58
*** Bhujay has joined #openstack-nova03:58
*** Bhujay has quit IRC03:58
*** Bhujay has joined #openstack-nova03:59
*** heyongli has quit IRC03:59
*** heyongli has joined #openstack-nova03:59
*** Bhujay has quit IRC03:59
*** Bhujay has joined #openstack-nova04:00
*** shaohe_feng has quit IRC04:01
*** shaohe_feng has joined #openstack-nova04:03
*** heyongli has quit IRC04:09
*** heyongli has joined #openstack-nova04:09
*** shaohe_feng has quit IRC04:11
*** vladikr has quit IRC04:13
*** shaohe_feng has joined #openstack-nova04:13
*** vladikr has joined #openstack-nova04:14
*** heyongli has quit IRC04:19
*** heyongli has joined #openstack-nova04:20
openstackgerritTetsuro Nakamura proposed openstack/nova master: Update root providers in same tree  https://review.openstack.org/58024604:20
*** shaohe_feng has quit IRC04:22
*** shaohe_feng has joined #openstack-nova04:22
*** heyongli has quit IRC04:29
*** heyongli has joined #openstack-nova04:30
*** shaohe_feng has quit IRC04:32
*** shaohe_feng has joined #openstack-nova04:33
*** yamamoto has joined #openstack-nova04:34
*** sahid has joined #openstack-nova04:34
*** lpetrut_ has quit IRC04:36
*** yamamoto has quit IRC04:38
*** yamamoto has joined #openstack-nova04:38
*** andreas_s has joined #openstack-nova04:39
*** heyongli has quit IRC04:40
*** heyongli has joined #openstack-nova04:40
*** lpetrut_ has joined #openstack-nova04:41
*** shaohe_feng has quit IRC04:42
*** shaohe_feng has joined #openstack-nova04:43
*** andreas_s has quit IRC04:44
*** lei-zh has joined #openstack-nova04:44
*** tianhui has joined #openstack-nova04:47
*** tianhui_ has quit IRC04:49
*** heyongli has quit IRC04:50
*** heyongli has joined #openstack-nova04:50
*** lpetrut_ has quit IRC04:51
openstackgerritShilpa Devharakar proposed openstack/nova master: WIP: Consistency_Groups assignment"  https://review.openstack.org/58055704:51
*** shaohe_feng has quit IRC04:52
*** shaohe_feng has joined #openstack-nova04:55
*** abhishekk has joined #openstack-nova05:00
*** heyongli has quit IRC05:00
*** heyongli has joined #openstack-nova05:01
*** shaohe_feng has quit IRC05:03
*** shaohe_feng has joined #openstack-nova05:03
*** heyongli has quit IRC05:10
*** heyongli has joined #openstack-nova05:11
*** blkart_ has joined #openstack-nova05:13
*** shaohe_feng has quit IRC05:13
*** shaohe_feng has joined #openstack-nova05:14
*** blkart has quit IRC05:14
*** janki has joined #openstack-nova05:20
*** heyongli has quit IRC05:21
*** heyongli has joined #openstack-nova05:21
*** ratailor has joined #openstack-nova05:22
*** shaohe_feng has quit IRC05:23
*** shaohe_feng has joined #openstack-nova05:24
*** lpetrut_ has joined #openstack-nova05:28
*** heyongli has quit IRC05:31
*** heyongli has joined #openstack-nova05:31
*** edmondsw has joined #openstack-nova05:31
*** shaohe_feng has quit IRC05:33
*** lpetrut_ has quit IRC05:34
*** armaan has joined #openstack-nova05:35
*** quiquell|off is now known as quiquell05:35
*** edmondsw has quit IRC05:36
*** andreas_s has joined #openstack-nova05:37
*** shaohe_feng has joined #openstack-nova05:38
*** masber has joined #openstack-nova05:39
*** yamahata has quit IRC05:39
*** yamahata has joined #openstack-nova05:40
*** armaan has quit IRC05:40
*** armaan has joined #openstack-nova05:41
*** heyongli has quit IRC05:41
*** heyongli has joined #openstack-nova05:42
*** shaohe_feng has quit IRC05:44
*** shaohe_feng has joined #openstack-nova05:46
*** armaan has quit IRC05:47
*** armaan has joined #openstack-nova05:47
*** heyongli has quit IRC05:51
*** heyongli has joined #openstack-nova05:52
*** Luzi has joined #openstack-nova05:52
*** shaohe_feng has quit IRC05:54
*** shaohe_feng has joined #openstack-nova05:55
*** adreznec has quit IRC05:57
*** adreznec has joined #openstack-nova05:58
*** heyongli has quit IRC06:02
*** heyongli has joined #openstack-nova06:02
*** yamahata has quit IRC06:04
*** shaohe_feng has quit IRC06:04
*** blkart_ has quit IRC06:05
*** blkart_ has joined #openstack-nova06:05
*** shaohe_feng has joined #openstack-nova06:05
*** nicolasbock has joined #openstack-nova06:08
openstackgerritVu Cong Tuan proposed openstack/os-vif master: Remove unnecessary pyNN testenv sections  https://review.openstack.org/57768606:08
*** heyongli has quit IRC06:12
*** heyongli has joined #openstack-nova06:12
*** mrsoul_ is now known as mschuppert06:13
*** shaohe_feng has quit IRC06:14
*** shaohe_feng has joined #openstack-nova06:15
*** dpawlik has joined #openstack-nova06:19
*** abhishekk has quit IRC06:20
*** abhishekk has joined #openstack-nova06:22
*** heyongli has quit IRC06:22
*** heyongli has joined #openstack-nova06:23
*** shaohe_feng has quit IRC06:25
*** shaohe_feng has joined #openstack-nova06:26
*** tiendc_ has joined #openstack-nova06:29
*** lei-zh has quit IRC06:30
*** tiendc has quit IRC06:31
*** lpetrut has joined #openstack-nova06:32
*** heyongli has quit IRC06:32
*** lei-zh has joined #openstack-nova06:33
*** heyongli has joined #openstack-nova06:33
*** shaohe_feng has quit IRC06:35
*** shaohe_feng has joined #openstack-nova06:36
*** tiendc__ has joined #openstack-nova06:39
*** tiendc_ has quit IRC06:42
*** mdnadeem has quit IRC06:43
*** heyongli has quit IRC06:43
*** heyongli has joined #openstack-nova06:43
openstackgerritOpenStack Proposal Bot proposed openstack/nova master: Imported Translations from Zanata  https://review.openstack.org/57801906:44
*** shaohe_feng has quit IRC06:45
*** shaohe_feng has joined #openstack-nova06:45
*** tssurya has joined #openstack-nova06:47
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (11)  https://review.openstack.org/57601806:49
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (12)  https://review.openstack.org/57601906:50
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (13)  https://review.openstack.org/57602006:50
*** s10 has joined #openstack-nova06:50
*** heyongli has quit IRC06:53
*** heyongli has joined #openstack-nova06:53
*** shaohe_feng has quit IRC06:55
*** shaohe_feng has joined #openstack-nova06:56
*** quiquell is now known as quiquell|bbl06:58
*** mdnadeem has joined #openstack-nova06:58
*** heyongli has quit IRC07:03
*** heyongli has joined #openstack-nova07:04
*** tiendc has joined #openstack-nova07:05
openstackgerritTakashi NATSUME proposed openstack/nova master: Transform metrics.update notification  https://review.openstack.org/58056707:05
*** Shadowphax has joined #openstack-nova07:06
*** shaohe_feng has quit IRC07:06
*** shaohe_feng has joined #openstack-nova07:06
*** tiendc__ has quit IRC07:07
*** ccamacho has quit IRC07:08
*** ccamacho has joined #openstack-nova07:08
openstackgerritTakashi NATSUME proposed openstack/nova master: Transform metrics.update notification  https://review.openstack.org/58056707:08
*** heyongli has quit IRC07:13
*** heyongli has joined #openstack-nova07:14
*** shaohe_feng has quit IRC07:16
*** shaohe_feng has joined #openstack-nova07:17
*** edmondsw has joined #openstack-nova07:20
*** gibi is now known as gibilet07:22
* gibilet forgot to post notes to the agenda of the yesterdays meeting :/07:22
*** heyongli has quit IRC07:24
*** heyongli has joined #openstack-nova07:24
*** edmondsw has quit IRC07:24
*** vladikr has quit IRC07:25
*** vladikr has joined #openstack-nova07:25
*** shaohe_feng has quit IRC07:26
*** shaohe_feng has joined #openstack-nova07:27
*** rpittau has quit IRC07:27
*** rcernin has quit IRC07:29
*** alexchadin has joined #openstack-nova07:33
*** gibilet is now known as giblet07:33
*** heyongli has quit IRC07:34
*** heyongli has joined #openstack-nova07:34
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (14)  https://review.openstack.org/57602707:35
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (15)  https://review.openstack.org/57603107:35
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (16)  https://review.openstack.org/57629907:35
*** shaohe_feng has quit IRC07:36
*** shaohe_feng has joined #openstack-nova07:39
*** quiquell|bbl is now known as quiquell07:39
*** rpittau has joined #openstack-nova07:39
*** vladikr has quit IRC07:43
*** heyongli has quit IRC07:44
*** trungnv has quit IRC07:45
*** heyongli has joined #openstack-nova07:45
*** amoralej|off is now known as amoralej07:46
*** vladikr has joined #openstack-nova07:47
*** shaohe_feng has quit IRC07:47
*** shaohe_feng has joined #openstack-nova07:47
*** heyongli has quit IRC07:47
*** jpena|off is now known as jpena07:47
*** vladikr has quit IRC07:53
*** ttsiouts has joined #openstack-nova07:53
*** lpetrut_ has joined #openstack-nova07:54
*** ttsiouts has quit IRC07:54
*** lpetrut has quit IRC07:54
*** ttsiouts has joined #openstack-nova07:54
*** vladikr has joined #openstack-nova07:55
*** shaohe_feng has quit IRC07:57
*** shaohe_feng has joined #openstack-nova07:59
*** moshele has joined #openstack-nova08:01
*** moshele has quit IRC08:03
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (17)  https://review.openstack.org/57634408:03
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (18)  https://review.openstack.org/57667308:04
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (19)  https://review.openstack.org/57667608:04
*** takashin has left #openstack-nova08:04
*** s10 has quit IRC08:05
*** mgoddard has joined #openstack-nova08:07
*** shaohe_feng has quit IRC08:07
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Add InstanceGroupPolicy object  https://review.openstack.org/57362808:10
*** shaohe_feng has joined #openstack-nova08:11
*** nicolasbock has quit IRC08:15
*** nicolasbock has joined #openstack-nova08:15
*** lei-zh has quit IRC08:16
*** shaohe_feng has quit IRC08:17
*** shaohe_feng has joined #openstack-nova08:19
*** tidwellr has joined #openstack-nova08:19
*** nicolasbock_ has joined #openstack-nova08:23
*** tidwellr has quit IRC08:24
*** nicolasbock has quit IRC08:25
*** shaohe_feng has quit IRC08:28
*** ratailor has quit IRC08:28
*** ratailor has joined #openstack-nova08:29
*** shaohe_feng has joined #openstack-nova08:30
*** vladikr has quit IRC08:31
*** vladikr has joined #openstack-nova08:32
*** nicolasbock_ has quit IRC08:36
*** shaohe_feng has quit IRC08:38
*** lei-zh has joined #openstack-nova08:39
*** derekh has joined #openstack-nova08:39
*** vladikr has quit IRC08:39
*** shaohe_feng has joined #openstack-nova08:41
*** vladikr has joined #openstack-nova08:42
*** nicolasbock_ has joined #openstack-nova08:43
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Add InstanceGroupPolicy object  https://review.openstack.org/57362808:48
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Refactor policies to policy in InstanceGroup DB model  https://review.openstack.org/57911308:48
*** shaohe_feng has quit IRC08:48
*** shaohe_feng has joined #openstack-nova08:49
*** armaan has quit IRC08:49
*** armaan has joined #openstack-nova08:50
*** dtantsur|afk is now known as dtantsur08:57
*** shaohe_feng has quit IRC08:58
*** armaan has quit IRC08:58
*** armaan has joined #openstack-nova08:59
*** ispp has joined #openstack-nova09:00
*** shaohe_feng has joined #openstack-nova09:01
*** armaan has quit IRC09:04
*** lei-zh has quit IRC09:05
*** lei-zh has joined #openstack-nova09:05
*** edmondsw has joined #openstack-nova09:08
*** shaohe_feng has quit IRC09:09
*** shaohe_feng has joined #openstack-nova09:09
*** sambetts|afk is now known as sambetts09:10
*** edmondsw has quit IRC09:12
*** lei-zh has quit IRC09:19
*** shaohe_feng has quit IRC09:19
*** shaohe_feng has joined #openstack-nova09:21
*** Shadowphax has quit IRC09:26
*** Shadowphax has joined #openstack-nova09:26
openstackgerritChris Dent proposed openstack/nova master: Use nova.db.api directly  https://review.openstack.org/54326209:28
*** cdent has joined #openstack-nova09:28
*** Zames has joined #openstack-nova09:29
*** shaohe_feng has quit IRC09:29
cdentjohnthetubaguy: i've fixed the rebase issues on https://review.openstack.org/#/c/543262/09:29
*** shaohe_feng has joined #openstack-nova09:30
*** Shadowphax has quit IRC09:31
*** Zames_ has joined #openstack-nova09:31
*** Zames has quit IRC09:31
*** Zames_ has quit IRC09:32
*** Shadowphax has joined #openstack-nova09:32
*** shaohe_feng has quit IRC09:39
*** shaohe_feng has joined #openstack-nova09:42
*** blkart_ has quit IRC09:49
*** shaohe_feng has quit IRC09:50
*** dtantsur is now known as dtantsur|brb09:55
*** yamamoto has quit IRC09:55
*** shaohe_feng has joined #openstack-nova09:56
*** shaohe_feng has quit IRC10:00
*** shaohe_feng has joined #openstack-nova10:01
*** blkart_ has joined #openstack-nova10:02
openstackgerritDeepak Mourya proposed openstack/nova master: Resource_provider API handler does not return specific error codes  https://review.openstack.org/57942310:09
*** shaohe_feng has quit IRC10:10
*** vladikr has quit IRC10:10
*** vladikr has joined #openstack-nova10:11
*** yamamoto has joined #openstack-nova10:15
*** shaohe_feng has joined #openstack-nova10:17
*** vladikr has quit IRC10:17
*** vladikr has joined #openstack-nova10:17
*** annp has quit IRC10:18
*** mdnadeem has quit IRC10:18
*** stakeda has quit IRC10:19
*** shaohe_feng has quit IRC10:20
*** shaohe_feng has joined #openstack-nova10:22
*** alexchadin has quit IRC10:24
*** tommylikehu has quit IRC10:26
*** tommylikehu has joined #openstack-nova10:26
*** ispp has quit IRC10:29
*** shaohe_feng has quit IRC10:31
*** abhishekk has quit IRC10:33
*** shaohe_feng has joined #openstack-nova10:33
*** zhuli__ has quit IRC10:34
*** zhuli__ has joined #openstack-nova10:34
*** shaohe_feng has quit IRC10:41
*** yedongcan has left #openstack-nova10:41
*** shaohe_feng has joined #openstack-nova10:43
*** yamamoto has quit IRC10:49
*** shaohe_feng has quit IRC10:51
*** shaohe_feng has joined #openstack-nova10:52
*** edmondsw has joined #openstack-nova10:56
*** andreas_s has quit IRC10:58
*** andreas_s has joined #openstack-nova10:58
*** alexchadin has joined #openstack-nova10:59
*** edmondsw has quit IRC11:00
*** shaohe_feng has quit IRC11:01
*** andreas_s has quit IRC11:03
*** shaohe_feng has joined #openstack-nova11:03
*** alexchadin has quit IRC11:04
*** shaohe_feng has quit IRC11:12
*** shaohe_feng has joined #openstack-nova11:14
*** armaan has joined #openstack-nova11:19
*** shaohe_feng has quit IRC11:22
*** shaohe_feng has joined #openstack-nova11:23
*** ratailor has quit IRC11:25
*** ispp has joined #openstack-nova11:28
*** Guest62131 has joined #openstack-nova11:28
*** vladikr has quit IRC11:29
*** vladikr has joined #openstack-nova11:29
*** vivsoni_ has quit IRC11:32
*** shaohe_feng has quit IRC11:32
*** cdent has quit IRC11:34
*** shaohe_feng has joined #openstack-nova11:34
*** Guest62131 has quit IRC11:36
*** cdent has joined #openstack-nova11:38
*** shaohe_feng has quit IRC11:42
*** quiquell is now known as quiquell|lunch11:43
*** shaohe_feng has joined #openstack-nova11:44
*** edmondsw has joined #openstack-nova11:44
*** armaan has quit IRC11:47
*** jpena is now known as jpena|lunch11:47
*** armaan has joined #openstack-nova11:47
*** cdent has quit IRC11:48
*** yamamoto has joined #openstack-nova11:50
*** Guest62131 has joined #openstack-nova11:50
*** janki has quit IRC11:51
*** armaan has quit IRC11:51
*** cdent has joined #openstack-nova11:52
*** shaohe_feng has quit IRC11:53
*** claudiub has joined #openstack-nova11:53
*** shaohe_feng has joined #openstack-nova11:53
cdentjaypipes, johnthetubaguy : you guys were +2 on https://review.openstack.org/#/c/543262/ before, but when the gate rebased it, it has some failures. those are fixed now.11:55
*** yamamoto has quit IRC11:56
jaypipescdent: re-added my +211:59
*** Guest62131 has quit IRC11:59
*** alexchadin has joined #openstack-nova12:00
cdentthanks12:01
*** Shadowphax has quit IRC12:03
*** shaohe_feng has quit IRC12:03
*** Shadowphax has joined #openstack-nova12:03
*** shaohe_feng has joined #openstack-nova12:05
*** armaan has joined #openstack-nova12:07
*** quiquell|lunch is now known as quiquell12:08
*** Shadowphax has quit IRC12:08
*** Shadowphax has joined #openstack-nova12:09
*** yamamoto has joined #openstack-nova12:10
*** claudiub has quit IRC12:10
*** vladikr has quit IRC12:13
*** shaohe_feng has quit IRC12:13
*** shaohe_feng has joined #openstack-nova12:14
*** vladikr has joined #openstack-nova12:15
*** jdillaman has joined #openstack-nova12:16
*** yamamoto has quit IRC12:19
*** shaohe_feng has quit IRC12:23
*** sridharg has joined #openstack-nova12:24
*** shaohe_feng has joined #openstack-nova12:26
*** pchavva has joined #openstack-nova12:27
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Add InstanceGroupPolicy object  https://review.openstack.org/57362812:28
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Refactor policies to policy in InstanceGroup DB model  https://review.openstack.org/57911312:28
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Add policy to InstanceGroup object  https://review.openstack.org/56337512:28
*** shaohe_feng has quit IRC12:34
*** tiendc has quit IRC12:34
openstackgerritYikun Jiang (Kero) proposed openstack/nova master: Add policy field to ServerGroup notification object  https://review.openstack.org/56340112:34
*** shaohe_feng has joined #openstack-nova12:37
*** udesale has quit IRC12:37
*** udesale has joined #openstack-nova12:38
*** lyan has joined #openstack-nova12:44
*** lyan is now known as Guest5066612:44
*** shaohe_feng has quit IRC12:44
*** dillaman has joined #openstack-nova12:45
*** shaohe_feng has joined #openstack-nova12:46
*** s10 has joined #openstack-nova12:46
*** jmlowe has joined #openstack-nova12:49
s10Hello. Is there any way to pass log_config_append option to Nova wsgi app (https://github.com/openstack/nova/blob/master/nova/api/openstack/wsgi_app.py) other than passing different nova.conf to nova-api-wsgi and nova-metadata-wsgi?12:51
*** eharney has joined #openstack-nova12:54
*** shaohe_feng has quit IRC12:54
*** alexchad_ has joined #openstack-nova12:54
*** mdnadeem has joined #openstack-nova12:54
*** medberry has joined #openstack-nova12:55
*** alexchadin has quit IRC12:56
*** yamamoto has joined #openstack-nova12:58
*** jangutter has quit IRC12:58
*** janki has joined #openstack-nova12:58
*** jpena|lunch is now known as jpena12:58
*** shaohe_feng has joined #openstack-nova12:59
*** edleafe is now known as figleaf12:59
*** amoralej is now known as amoralej|lunch13:02
*** shaohe_feng has quit IRC13:04
*** yamamoto_ has joined #openstack-nova13:05
*** shaohe_feng has joined #openstack-nova13:06
*** yamamoto_ has quit IRC13:07
*** yamamoto has quit IRC13:08
*** felipemonteiro_ has joined #openstack-nova13:09
*** stephenfin is now known as finucannot13:10
openstackgerritAndreas Jaeger proposed openstack/os-vif master: move legacy-tempest-dsvm-nova-os-vif to repo  https://review.openstack.org/55940113:10
*** felipemonteiro__ has joined #openstack-nova13:10
*** hemna_ has joined #openstack-nova13:12
*** alexchad_ has quit IRC13:13
*** felipemonteiro_ has quit IRC13:14
*** shaohe_feng has quit IRC13:15
*** shaohe_feng has joined #openstack-nova13:15
*** mriedem has joined #openstack-nova13:18
*** armaan has quit IRC13:18
*** armaan has joined #openstack-nova13:19
*** mlavalle has joined #openstack-nova13:21
openstackgerritsahid proposed openstack/nova master: hardware: fix memory check usage for small/large pages  https://review.openstack.org/53216813:21
openstackgerritsahid proposed openstack/nova master: hardware: fix hugepages memory usage per intances  https://review.openstack.org/58065713:21
*** armaan has quit IRC13:23
*** shaohe_feng has quit IRC13:25
*** armaan has joined #openstack-nova13:26
tssuryagood morning mriedem: so I was trying to resurrect my patch on trying to clean up RPs/CNs/Service and aggregate_hosts records on deleting a host by adding a new flag --recursively-force delete the host (https://review.openstack.org/#/c/546660/), but then thought after the recent changes we already did with nova service-delete/ what is the whole point of having nova-manage delete_host command ? I mean deleting the service and deleting the13:26
tssurya(if its still too early for you now, feel free to answer later whenever you get time)13:27
*** shaohe_feng has joined #openstack-nova13:27
*** yamamoto has joined #openstack-nova13:29
*** dtantsur|brb is now known as dtantsur13:29
mriedemtssurya: you mean this https://review.openstack.org/#/c/563698/13:30
mriedemidk, it sounds pretty redundant13:30
tssuryayes13:30
tssuryajust to have a command delete_host to delete the host_mappings also sounds lame,13:31
tssuryabut incorporating all that again into this is redundant, specially considering you also solved the bug of deleting a service hosting instances13:31
*** shaohe_feng has quit IRC13:35
*** armaan has quit IRC13:36
*** armaan has joined #openstack-nova13:37
*** shaohe_feng has joined #openstack-nova13:37
*** medberry has quit IRC13:39
*** armaan has quit IRC13:40
*** armaan has joined #openstack-nova13:40
*** psachin` has quit IRC13:42
*** alexchadin has joined #openstack-nova13:45
*** armaan has quit IRC13:45
*** shaohe_feng has quit IRC13:45
*** shaohe_feng has joined #openstack-nova13:46
*** BrinZhang has joined #openstack-nova13:47
*** awaugama has joined #openstack-nova13:52
*** sapd_ has joined #openstack-nova13:53
*** sapd has quit IRC13:53
*** shaohe_feng has quit IRC13:56
*** medberry has joined #openstack-nova13:56
*** medberry has quit IRC13:56
*** medberry has joined #openstack-nova13:56
*** burt has joined #openstack-nova13:57
*** shaohe_feng has joined #openstack-nova13:57
*** med_ has quit IRC13:57
*** ispp has quit IRC13:59
*** josecastroleon has quit IRC14:00
*** josecastroleon has joined #openstack-nova14:01
*** ispp has joined #openstack-nova14:01
mriedemtssurya: agree. i guess my question is, what happens if i run nova-manage cell_v2 delete_host *before* nova service-delete? are those compatible or is there a required order now?14:02
*** medberry has quit IRC14:03
mriedemif we run delete_host, i'm not sure if service-delete will work14:03
tssuryamriedem: well if you run delete_host before service-delete; in the part of service-delete trying to delete the host_mapping14:03
tssuryait might can HostMappingNotFound14:03
tssuryasay*14:03
mriedembut we handle and ignore that don't we?14:04
*** ispp has quit IRC14:04
*** amoralej|lunch is now known as amoralej14:04
mriedemok so we don't need the host mapping to lookup the service record14:04
tssuryahttps://github.com/openstack/nova/blob/master/nova/objects/host_mapping.py#L13914:05
*** med_ has joined #openstack-nova14:05
*** med_ has quit IRC14:05
*** med_ has joined #openstack-nova14:05
mriedemi'm talking about this https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/services.py#L24414:05
mriedemso if i delete the host mapping via delete_host, nova service-delete will fail14:05
mriedemwith a 500 actually14:06
*** shaohe_feng has quit IRC14:06
mriedemi really thought we had a try/except HostMappingNotFound: pass around this https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/services.py#L24314:06
mriedemsince if the host mapping is already gone, we shouldn't care since we want it gone anyway14:06
tssuryayea14:06
tssuryaI guess its fine then14:07
mriedemso, maybe that's a new bug now and rather than https://review.openstack.org/#/c/546660/ we should just fix that14:07
tssuryabecause it wouldn't find the host_mapping14:07
mriedemso nova service-delete and nova-manage cell_v2 delete_host are idempotent14:07
mriedemtssurya: well, a 500 response from service-delete isn't fine14:07
*** shaohe_feng has joined #openstack-nova14:07
mriedemthe bug is:14:07
mriedem1. run nova-manage cell_v2 delete_host14:07
mriedem2. run nova service-delete - 500 error14:08
mriedemand the service doesn't get deleted, but the aggregate relationship and resource provider are14:08
tssuryawe won't get a 500 right ? because14:08
tssuryaah yea14:08
tssuryashoot got it got it14:08
*** hongbin has joined #openstack-nova14:08
*** yamahata has joined #openstack-nova14:09
tssuryaIMHO, delete_host should be removed14:09
tssuryaright ?14:09
mriedemhttps://review.openstack.org/#/c/554920/4/nova/api/openstack/compute/services.py@24114:09
mriedemidk, is there a use case for deleting the host mapping but not the compute service?14:10
*** mdnadeem has quit IRC14:10
mriedemif you're trying to delete the cell14:11
*** yamahata has quit IRC14:11
*** alexchadin has quit IRC14:11
mriedemlet's say i need to move my services to another cell db or something, i'd have to remove the host mappings to remove the cell14:11
tssuryawell there is a delete_cell command14:11
*** yamahata has joined #openstack-nova14:11
mriedembut that won't pass unless the host mappings are gone first14:11
mriedemunless you use --force14:11
mriedemhow about for now we just fix this https://review.openstack.org/#/c/554920/4/nova/api/openstack/compute/services.py@24114:12
mriedemsince that's an obvious bug14:12
tssuryamriedem: cool,14:12
tssuryawill open a bug14:13
*** Luzi has quit IRC14:14
*** shaohe_feng has quit IRC14:16
*** shaohe_feng has joined #openstack-nova14:20
*** dpawlik has quit IRC14:20
*** bnemec is now known as beekneemech14:22
*** antosh has joined #openstack-nova14:23
*** edmondsw has quit IRC14:23
*** edmondsw has joined #openstack-nova14:24
*** shaohe_feng has quit IRC14:26
*** edmondsw has quit IRC14:27
*** shaohe_feng has joined #openstack-nova14:28
mriedemthese json compare tests are still failing and it's not clear to me why14:31
mriedemhttp://logs.openstack.org/13/579113/10/check/openstack-tox-lower-constraints/163ef96/job-output.txt.gz#_2018-07-06_12_41_18_61174714:31
mriedemb"testtools.matchers._impl.MismatchError: '/dev/disk/path/lun-X' != '/dev/disk/path/lun-Z': path:14:31
*** edmondsw has joined #openstack-nova14:32
mriedemas far as i can tell the expected and actual for root.nova_object.data.bdms[0].nova_object.data.connection_info_json.data.device_path are the same14:32
mriedemonce deserialized14:32
*** Bhujay has quit IRC14:33
BrinZhangbauzas: Hi, can you review this commit? https://review.openstack.org/#/c/579520/14:34
BrinZhangthank you.14:34
mriedemBrinZhang: we've already rejected that same blueprint at least half a dozen times14:36
BrinZhangWhy?14:36
*** shaohe_feng has quit IRC14:37
BrinZhangWe learned from customers that this is very meaningful.14:37
mriedemsee my comments in the review14:38
mriedemyou can already create a volume with the desired type and provide that to nova14:38
*** shaohe_feng has joined #openstack-nova14:38
*** felipemonteiro__ has quit IRC14:38
*** lpetrut_ has quit IRC14:40
lyarwoodmriedem: ah crap, I'll take a look now, happy to revert the original change as it only introduced this test btw14:40
BrinZhangThis way you said can achieve the goal, but if there are multiple backends, it is not more convenient to create an instance directly by specifying the backend with an image.14:42
*** alexchadin has joined #openstack-nova14:44
mriedemBrinZhang: i'm not sure what your point is, if there are multiple volume types, the user still has to pick one if they don't want the default14:44
mriedemso if they have to do that anyway, they might as well just create the volume in cinder directly and then provider the volume id to nova during server create14:45
mriedem*provide14:45
mriedemif these are the types of users that can't use a command line to tie 2 things together, surely your cloud has a UI that your users are already using for most of their interaction with the cloud14:46
mriedemand the UI could orchestrate this14:46
*** quiquell is now known as quiquell|off14:47
*** shaohe_feng has quit IRC14:47
s10It might be not to user friendly, because volume can be in creating state for several minutes, and because usually users network connection to Cloud API is less reliable then connection from nova to cinder, it's better to orchestrate this inside cloud (nova or heat), not in UI14:48
*** shaohe_feng has joined #openstack-nova14:48
BrinZhangmriedem: After all, the UI is a combined operation, which is efficient and convenient through the underlying implementation.14:49
s10With nova boot user have to make only one POST connection, with Cinder+Nova — what he should do, poll cinder-api every 10-30 seconds?14:49
*** purplerbot has quit IRC14:49
*** hongbin has quit IRC14:49
*** purplerbot has joined #openstack-nova14:49
*** hongbin has joined #openstack-nova14:49
*** rpittau has quit IRC14:50
*** tidwellr has joined #openstack-nova14:52
mriedemcburgess was, at one point, investigating how nova could boot from a volume that's not yet 'available' but i don't think that every went anywhere14:53
mriedem*ever14:53
s10In our openstack deployment we use our own patch for specifying volume type on nova boot since 2015 year, It's just disappointing that this feature will never be upstream because of this policy...14:54
mriedemit's also disappointing that everyone wants to cram all of their orchestration into nova, which overly complicates all of the code, which leads to more bugs and brittle architecture14:55
*** medberry has joined #openstack-nova14:55
*** medberry is now known as Guest1861014:55
mriedemwe don't even support a lot of the server action APIs for volume-backed instances, and bfv was added several several years ago14:56
mriedemso that goes to show what kinds of gaps we already have in functionality with existing capabilities14:56
mriedemso i'm personally just not that interested in piling onto that existing technical debt14:56
mriedemi'm one person though, so other maintainers of the project can speak up if they think this is something we should add, i didn't -2 the spec14:57
*** shaohe_feng has quit IRC14:57
*** shaohe_feng has joined #openstack-nova14:58
*** ispp has joined #openstack-nova14:59
melwittmriedem: on your unit test thing, I think what's happening is, while what it shows in the output matches, but it's not showing what it internally compared https://github.com/openstack/nova/blob/master/nova/test.py#L502-L503 and failed. so the list order must be changing through try_deserialize?15:00
*** Guest18610 is now known as med__15:01
*** med__ is now known as medberry215:02
*** ispp has quit IRC15:02
mriedemmelwitt: i'm not sure i'm following, it says it compared these:15:02
mriedemb"testtools.matchers._impl.MismatchError: '/dev/disk/path/lun-X' != '/dev/disk/path/lun-Z': path:15:02
*** deepak_mourya has quit IRC15:03
*** medberry2 is now known as medberry15:03
*** medberry has quit IRC15:03
*** medberry has joined #openstack-nova15:03
melwittmriedem: right, that's from the mismatch exception raised internally from the inner() method. but it try_deserialize both expected and observed before it compared them15:03
BrinZhangmriedem: thank you . We did not change the original openstack architecture, just specify the volume type when creating the volume. This option is optional and I don't think this will burden the current architecture.15:04
*** eharney has quit IRC15:04
melwittmriedem: when assertJsonEqual shows you the items it compared https://github.com/openstack/nova/blob/master/nova/test.py#L553-L554 it's not showing what it actually compared and failed (the try_deserialize'd expected and observed)15:04
mriedemmelwitt: so you think because now we deserialize the bdms entry, that unsorted list could be out of order15:05
melwittso it misleads. it should be outputting what the results from try_deserialize were in the error message15:05
melwittI think when assertJsonEqual does, yes. here https://github.com/openstack/nova/blob/master/nova/test.py#L49615:05
BrinZhangmelwitt: Users can choose not to pass this option to configure the default volume type, at present, openstack tends to be stable, and individuals feel that it is an improvement point in ease of use..15:06
openstackgerritChen proposed openstack/nova master: Fix server_group_members quota check  https://review.openstack.org/58068415:07
*** shaohe_feng has quit IRC15:07
openstackgerritClaudiu Belu proposed openstack/nova master: WIP: replace spec with autospec  https://review.openstack.org/55729915:10
mriedemmelwitt: ok i can't really grok the sort_key in that method either, so i guess i'll revert my fix and go back to the brute force deserialize directly in the test which is what i started with15:10
*** Shadowphax has quit IRC15:10
openstackgerritMerged openstack/nova master: Fix whitespace damage  https://review.openstack.org/58037015:10
*** shaohe_feng has joined #openstack-nova15:11
melwittmriedem: yeah, I was about to say, I see there is some code for isinstance(..., list) and sorting but I don't grok it either. but it's not hitting that path, based on the error output15:11
melwittthat one would say "Different list items"15:11
melwittalthough maybe you mean L54015:12
melwittso maybe it's using a sort key that isn't making it sort in this case15:13
melwittlike something that is equal between the two items15:13
BrinZhangmriedem: Thank you, we will conduct further research and consider this issue.15:13
mriedemidk, i'm too dumb to figure out that code, i'm just going to revert and do the original fix i had for this bug15:13
mriedemBrinZhang: maybe it will come up again at the ptg in september and we'll finally just cave and agree to do it in stein15:15
melwittyeah, I'm finding it very hard to understand and don't completely understand it. it would help a lot just to change the error output to output what the results of the try_deserialize were since that's what it's actually comparing15:15
BrinZhangmriedem: Ok, this makes me look forward to :)15:17
melwittor rather, output something from the exception object for expected/observed if they are stored in there somewhere15:17
melwittin the MismatchError object15:17
*** shaohe_feng has quit IRC15:18
*** shaohe_feng has joined #openstack-nova15:18
*** eharney has joined #openstack-nova15:19
*** openstackgerrit has quit IRC15:19
*** ttsiouts has quit IRC15:19
*** ispp has joined #openstack-nova15:20
*** josecastroleon has quit IRC15:21
*** tssurya has quit IRC15:22
mgagnemriedem: support for volume type in BDM is something we would be interested in for the reasons mentioned above.15:22
melwittbecause it's easier than using Heat and we create a volume anyway?15:23
mgagnemelwitt: none of our users use heat, it's just the reality. they don't want to use Heat, they don't need it for such simple task.15:24
melwittjust wanted to make sure I understood the reasoning15:24
mgagnemelwitt: about not using Heat or adding the feature to Nova?15:24
melwittand am not missing other pieces15:24
*** openstackgerrit has joined #openstack-nova15:25
openstackgerritStephen Finucane proposed openstack/nova master: objects: Add NUMATopologyLimits.networks  https://review.openstack.org/57548615:25
openstackgerritStephen Finucane proposed openstack/nova master: hardware: Start accounting for networks in NUMA fitting  https://review.openstack.org/56444815:25
openstackgerritStephen Finucane proposed openstack/nova master: objects: Add RequestSpec.network_metadata  https://review.openstack.org/56444215:25
openstackgerritStephen Finucane proposed openstack/nova master: scheduler: Start utilizing RequestSpec.network_metadata  https://review.openstack.org/56445215:25
openstackgerritStephen Finucane proposed openstack/nova master: conf: Add '[neutron] physnets' and related options  https://review.openstack.org/56444015:25
openstackgerritStephen Finucane proposed openstack/nova master: libvirt: Start populating NUMACell.network_metadata field  https://review.openstack.org/56444115:25
dansmithdon't forget, if we do this for volume, there are a hundred similar things for networks we'll have to take, plus all that cyborg goodness to come :)15:25
*** Shilpa has quit IRC15:25
melwittadding the feature. I was asking if it's because 1) it's easier than using Heat and 2) we create a volume anyway and adding this would be basically passing one extra request param and we wouldn't have to change anything else15:25
mgagnemelwitt: 1) as said, our users don't care about Heat 2) yep15:26
melwittack15:27
*** shaohe_feng has quit IRC15:28
*** shaohe_feng has joined #openstack-nova15:29
*** chyka has joined #openstack-nova15:29
*** alexchadin has quit IRC15:31
mgagnethat the fact that it's a feature available on alternative platforms might play into it too.15:31
*** ispp has quit IRC15:31
melwittit's something I would consider but only because of the level of simplicity it is. and I'd be more concerned about what precedent it sets. we _don't_ want to add more orchestration in general15:33
*** medberry has quit IRC15:34
mriedemmgagne: what alternative platforms?15:35
mriedemi don't know how volume types work in aws15:35
mgagnemriedem: you can pass the volume type when creating an instance15:35
mgagnemelwitt: it's a recurrent theme driven by user needs which might also come from features they have found elsewhere and want to find in openstack. at this point, it might be the time to reconsider. redirecting users to a full blown orchestration service is overkill, especially if you want to boot a single instance and other platforms offer it.15:36
mriedemwhy can't osc or horizon handle this?15:37
mgagnemriedem: workflow is completely different here. you need to wait on client side for that.15:37
mriedemopenstack server create --block-device-mapping type=rbd,size=1 --flavor foo --image bar my-rbd-backed-server15:37
mgagneit's not fire and forget15:38
mriedemopenstack server create has a --wait option15:38
mriedemif you specify the volume type, it implies --wait15:38
*** shaohe_feng has quit IRC15:38
mgagneI don't follow you. why would it implies --wait ?15:38
mriedemand then osc polls cinder until the volume is available15:38
mriedemthen it's not 'fire and forget'15:38
mriedemif i added this to osc, i'm saying i'd do it that way15:39
mriedemif i want to rely on osc to orchestrate this for me, i'm opting into the polling15:39
*** shaohe_feng has joined #openstack-nova15:39
mgagneI think it's misjudging the need of the user15:39
mriedemat some point the user is waiting for the server to be active right?15:39
mriedemso they either wait client side, or wait for nova to say the server is active, and nova is doing the polling under the covers15:40
mriedemeither way the client is waiting15:40
mgagnenot 100% true, you can boot and make it auto bootstrap later with user-data.15:40
mgagneand it's ready when it's ready without you having to wait on CLI15:40
melwittmgagne: I've never been super against the idea being that we already create a volume, we already wait for it to be ready, it's already part of the boot flow and this is adding one request param and nothing else changes15:41
*** yamamoto_ has joined #openstack-nova15:41
mgagneI just feel that people are misunderstanding the need and advantages of doing it server-side.15:41
mgagneand just throwing ideas/hacks to make it go away from Nova.15:42
*** hemna_ has quit IRC15:42
*** tianhui_ has joined #openstack-nova15:44
*** yamamoto has quit IRC15:44
*** lyarwood has quit IRC15:45
*** mriedem is now known as mriedem_away15:45
*** tianhui has quit IRC15:46
openstackgerritMatt Riedemann proposed openstack/nova master: Handle compare in test_pre_live_migration_volume_backed* directly  https://review.openstack.org/58070215:46
*** medberry has joined #openstack-nova15:46
*** medberry is now known as Guest5433815:47
*** moshele has joined #openstack-nova15:48
*** shaohe_feng has quit IRC15:48
*** shaohe_feng has joined #openstack-nova15:49
*** hemna_ has joined #openstack-nova15:51
*** moshele has quit IRC15:57
*** shaohe_feng has quit IRC15:59
*** shaohe_feng has joined #openstack-nova16:00
*** lucasagomes is now known as lucas-afk16:01
*** hemna_ has quit IRC16:01
*** Swami has joined #openstack-nova16:02
*** BrinZhang has quit IRC16:04
openstackgerritBalazs Gibizer proposed openstack/nova master: Functional test for booting with nested resources  https://review.openstack.org/52772816:04
*** sahid has quit IRC16:06
*** shaohe_feng has quit IRC16:09
*** shaohe_feng has joined #openstack-nova16:10
*** yamahata has quit IRC16:12
mgagneinteresting... when 'services' (dns) where added per networks in metadata, no new metadata version was created: https://github.com/openstack/nova/commit/4b333b989dfc778a8b61db4a1b8552e988a1047116:12
openstackgerritBalazs Gibizer proposed openstack/nova master: Functional test for booting with nested resources  https://review.openstack.org/52772816:13
*** sambetts is now known as sambetts|afk16:14
*** wwriverrat has joined #openstack-nova16:15
*** jmlowe has quit IRC16:17
*** mdbooth_ has quit IRC16:19
*** shaohe_feng has quit IRC16:19
*** shaohe_feng has joined #openstack-nova16:20
*** lyarwood has joined #openstack-nova16:26
*** shaohe_feng has quit IRC16:29
*** shaohe_feng has joined #openstack-nova16:30
*** jchhatbar has joined #openstack-nova16:35
*** dtantsur is now known as dtantsur|afk16:37
*** janki has quit IRC16:38
*** shaohe_feng has quit IRC16:40
*** shaohe_feng has joined #openstack-nova16:41
*** yamamoto has joined #openstack-nova16:43
*** sridharg has quit IRC16:44
*** ttsiouts has joined #openstack-nova16:45
*** yamamoto_ has quit IRC16:47
*** ttsiouts has quit IRC16:47
*** shaohe_feng has quit IRC16:50
*** Swami has quit IRC16:51
*** tianhui has joined #openstack-nova16:52
*** shaohe_feng has joined #openstack-nova16:53
*** tianhui_ has quit IRC16:53
*** yamahata has joined #openstack-nova16:54
openstackgerritDan Smith proposed openstack/nova master: Avoid requesting DISK_GB allocation for root_gb on BFV instances  https://review.openstack.org/58072016:54
*** shaohe_feng has quit IRC17:00
*** shaohe_feng has joined #openstack-nova17:01
*** derekh has quit IRC17:02
*** mgoddard has quit IRC17:03
*** amoralej is now known as amoralej|off17:04
*** Swami has joined #openstack-nova17:07
*** smcginnis is now known as smcginnis_afk17:08
*** udesale has quit IRC17:08
*** jchhatbar has quit IRC17:10
*** mriedem_away is now known as mriedem17:10
*** shaohe_feng has quit IRC17:10
*** shaohe_feng has joined #openstack-nova17:11
*** Shadowphax has joined #openstack-nova17:13
*** burt has quit IRC17:18
*** shaohe_feng has quit IRC17:21
*** shaohe_feng has joined #openstack-nova17:22
*** yamamoto has quit IRC17:24
*** med_ has quit IRC17:25
*** yamamoto has joined #openstack-nova17:26
*** jpena is now known as jpena|off17:26
*** itlinux has joined #openstack-nova17:28
*** yamamoto has quit IRC17:30
*** shaohe_feng has quit IRC17:31
*** shaohe_feng has joined #openstack-nova17:32
melwittmriedem: the idea of counting build requests in addition to instance records for quota rears its head again (in my comments) for this quota bug https://review.openstack.org/58068417:34
*** alexpilotti has quit IRC17:35
mriedemlikely need a functional test to recreate that issue to determine the appropriate fix17:37
*** yamamoto has joined #openstack-nova17:37
*** Guest54338 has quit IRC17:37
*** yamamoto has quit IRC17:37
mriedemisn't it just, set server group member quota to 2, create 1 instance in the group, then request 2 more instances from the same request in the same group and it will pass the check but ultimately end up with 3 members in the group?17:38
melwittyeah, I could give that a shot later, maybe next week. I have to pick up penick from the (far) airport today17:38
melwittI reproduced the problem in devstack and know why it's happening. just set quota to 2 and boot 3 instances into the group with --max-count 317:39
mriedemah ok17:39
melwittwe count instance records for the user for the member count and we don't create instance records till conductor17:39
mriedemok i could hack on a func regression test for this once i'm done reviewing finucannot's numa series17:40
mriedemdo we have any way to link build requests to groups?17:40
melwittthe first part (initial quota check) of the proposed fix makes sense, check the total multi-create delta against the count. but for the recheck, nothing really makes sense unless we were to count build requests too17:40
melwittyeah, group members contains a instance_uuid column17:41
mriedemso are they 2 tests then? the one i described which goes over quota in a 2nd request? and yours goes over quota in a single request17:41
*** shaohe_feng has quit IRC17:41
melwittI'm only considering the single request, the multi-create scenario17:42
*** shaohe_feng has joined #openstack-nova17:42
mriedembut doesn't the build request idea happen with the recheck in a subsequent request?17:42
melwittI think in the separate single instance request scenario, the initial check is fine but the recheck won't be right17:43
*** sridharg has joined #openstack-nova17:43
mriedemright, so 2 tests?17:43
melwittyeah, sounds like it. I only had the multi-create context in my head17:43
*** sridharg has quit IRC17:43
melwittbut in either case recheck is wrong. in the multi-create case the initial check is also wrong17:44
mriedemok, well i'll wrap up what i'm doing and then start on some tests so everyone can be on the same page17:44
melwittbecause it uses a delta of 1, assuming instance records are being created in between, which they're not. if we counted build requests, then probably delta of 1 will also work17:44
melwittcool, thank you. I'm also reminded of the bug Kevin_Zheng was looking to solve around quotas, same thing, idea to count build requests and de-dupe instance uuids. I think that one was to avoid leaving a potentially large number of instances in ERROR state because of a quota fail during multi-create17:48
mriedemyes it was17:48
*** med_ has joined #openstack-nova17:48
*** med_ has quit IRC17:48
*** med_ has joined #openstack-nova17:48
mriedemso _server_group_count_members_by_user would just need to get a list of build requests by the list of group.members uuids, then remove duplicates from any group members found in the cells17:49
melwittso I think we just need to go that route if we can, otherwise a lot of this stuff doesn't make sense. I spent awhile trying to think how to deal with it _without_ build requests and it's just trying to predict things before resources are actually created17:49
*** Swami has quit IRC17:49
mriedemwe have BuildRequestList.get_by_filters but that wouldn't be very efficient, but it would be backportable17:50
mriedemcould add a BuildRequestList.get_by_instance_uuids later on master only17:50
melwittthat's what I'm thinking. it's kind of indirect. I'm not sure if we can look those up by user_id. and there's also the danger of, do build requests get orphaned17:51
mriedemno they don't get orphaned17:51
melwittbecause that would throw off quota count stuff17:51
mriedem_provision_instances creates the BuildRequest with the instance_uuid set17:51
*** cdent has quit IRC17:51
mriedembuild requests don't have a user_id field set...17:51
*** shaohe_feng has quit IRC17:52
mriedemonly project_id17:52
melwittI guess maybe even if they did it doesn't matter because we'd de-dupe with instance uuids so even if it got stuck somehow it wouldn't hurt. just trying to think of things17:52
mriedemwell i think BuildRequestList.get_by_filters could be used for now and backportable17:52
*** shaohe_feng has joined #openstack-nova17:52
mriedemand then add BuildRequest.user_id in master, and add a BuildRequestList.get_by_instance_uuid_and_user_id or something in master17:52
mriedemanyway, tests first17:53
* melwitt nods17:53
melwittwhat you're saying sounds legit17:54
*** alexpilotti has joined #openstack-nova17:54
*** itlinux has quit IRC17:56
*** itlinux has joined #openstack-nova17:57
*** itlinux has quit IRC17:57
*** Sundar has joined #openstack-nova17:58
Sundarefried: Please LMK when you are available.17:58
*** alexpilotti has quit IRC17:59
*** chyka has quit IRC17:59
*** chyka has joined #openstack-nova18:00
*** shaohe_feng has quit IRC18:02
*** jmlowe has joined #openstack-nova18:02
*** shaohe_feng has joined #openstack-nova18:04
*** shaohe_feng has quit IRC18:12
*** moshele has joined #openstack-nova18:14
*** shaohe_feng has joined #openstack-nova18:14
*** moshele has quit IRC18:16
*** ccamacho has quit IRC18:17
*** tidwellr has quit IRC18:20
*** tssurya has joined #openstack-nova18:20
*** jmlowe_ has joined #openstack-nova18:21
*** shaohe_feng has quit IRC18:22
*** jmlowe has quit IRC18:23
*** shaohe_feng has joined #openstack-nova18:23
*** AlexeyAbashkin has joined #openstack-nova18:24
*** Sundar has quit IRC18:25
*** Sundar has joined #openstack-nova18:30
*** shaohe_feng has quit IRC18:33
*** shaohe_feng has joined #openstack-nova18:35
*** yamamoto has joined #openstack-nova18:38
*** alexpilotti has joined #openstack-nova18:39
*** cargonza_ has joined #openstack-nova18:41
*** imacdonn has quit IRC18:41
*** imacdonn has joined #openstack-nova18:41
*** alexpilo_ has joined #openstack-nova18:42
*** shaohe_feng has quit IRC18:43
*** AlexeyAbashkin has quit IRC18:43
*** mgoddard has joined #openstack-nova18:43
*** shaohe_feng has joined #openstack-nova18:44
*** alexpilotti has quit IRC18:45
*** alexpilo_ has quit IRC18:47
*** AlexeyAbashkin has joined #openstack-nova18:47
*** yamamoto has quit IRC18:48
*** mgoddard has quit IRC18:48
*** shaohe_feng has quit IRC18:53
*** shaohe_feng has joined #openstack-nova18:55
*** jmlowe_ has quit IRC18:56
*** AlexeyAbashkin has quit IRC18:59
*** s10 has quit IRC19:02
*** Guest54338 has joined #openstack-nova19:02
*** shaohe_feng has quit IRC19:03
*** shaohe_feng has joined #openstack-nova19:04
*** mschuppert has quit IRC19:04
*** shaohe_feng has quit IRC19:14
*** shaohe_feng has joined #openstack-nova19:16
openstackgerritMathieu Gagné proposed openstack/nova master: Add support for multiple fixed-ips in metadata  https://review.openstack.org/58074219:20
*** shaohe_feng has quit IRC19:24
*** shaohe_feng has joined #openstack-nova19:25
*** gyee has joined #openstack-nova19:32
*** shaohe_feng has quit IRC19:34
*** eharney has quit IRC19:34
*** shaohe_feng has joined #openstack-nova19:36
*** cargonza_ has quit IRC19:39
*** cargonza_ has joined #openstack-nova19:39
openstackgerritDan Smith proposed openstack/nova master: Avoid requesting DISK_GB allocation for root_gb on BFV instances  https://review.openstack.org/58072019:43
*** shaohe_feng has quit IRC19:44
*** shaohe_feng has joined #openstack-nova19:45
*** shaohe_feng has quit IRC19:55
figleafSundar: Eric is on PTO today. He'll be back Monday19:56
*** AlexeyAbashkin has joined #openstack-nova19:58
*** shaohe_feng has joined #openstack-nova19:59
*** shaohe_feng has quit IRC20:05
*** AlexeyAbashkin has quit IRC20:05
*** shaohe_feng has joined #openstack-nova20:06
*** Guest54338 has quit IRC20:07
*** alexpilotti has joined #openstack-nova20:09
*** edmondsw_ has joined #openstack-nova20:09
*** alexpilo_ has joined #openstack-nova20:11
*** edmondsw has quit IRC20:11
*** alexpilotti has quit IRC20:14
*** alexpilo_ has quit IRC20:15
*** shaohe_feng has quit IRC20:15
Sundarfigleaf: Thanks for letting me know.20:16
*** shaohe_feng has joined #openstack-nova20:16
*** Guest54338 has joined #openstack-nova20:18
*** awaugama has quit IRC20:21
*** shaohe_feng has quit IRC20:25
*** Guest54338 is now known as medberry20:26
*** medberry has joined #openstack-nova20:26
*** shaohe_feng has joined #openstack-nova20:27
*** alexpilotti has joined #openstack-nova20:35
*** shaohe_feng has quit IRC20:36
*** Sundar has quit IRC20:36
openstackgerritMatt Riedemann proposed openstack/nova master: WIP: Fix server_group_members quota check  https://review.openstack.org/58068420:36
openstackgerritMatt Riedemann proposed openstack/nova master: Add functional regressions tests for server_group_members OverQuota  https://review.openstack.org/58075520:36
mriedemmelwitt: ^20:36
*** shaohe_feng has joined #openstack-nova20:37
*** alexpilotti has quit IRC20:39
*** shaohe_feng has quit IRC20:46
*** shaohe_feng has joined #openstack-nova20:47
*** antosh has quit IRC20:53
*** shaohe_feng has quit IRC20:56
*** Shadowphax has quit IRC20:57
*** shaohe_feng has joined #openstack-nova20:57
openstackgerritMatt Riedemann proposed openstack/nova master: Add functional regressions tests for server_group_members OverQuota  https://review.openstack.org/58075520:57
openstackgerritMatt Riedemann proposed openstack/nova master: Fix server_group_members quota check  https://review.openstack.org/58068420:57
mriedemthis time with the fix for both scenarios, basically re-wrote the persons patch20:57
*** dpawlik has joined #openstack-nova20:58
*** Shadowphax has joined #openstack-nova20:59
*** chyka has quit IRC21:01
*** jmlowe has joined #openstack-nova21:01
*** dpawlik has quit IRC21:02
*** Shadowphax has quit IRC21:05
openstackgerritMatt Riedemann proposed openstack/nova master: Avoid joins in _server_group_count_members_by_user  https://review.openstack.org/58076421:06
*** shaohe_feng has quit IRC21:06
*** shaohe_feng has joined #openstack-nova21:07
*** jmlowe has quit IRC21:11
*** nicolasbock_ has quit IRC21:15
*** antosh has joined #openstack-nova21:15
*** eharney has joined #openstack-nova21:16
*** shaohe_feng has quit IRC21:17
*** shaohe_feng has joined #openstack-nova21:19
*** shaohe_feng has quit IRC21:27
*** shaohe_feng has joined #openstack-nova21:28
*** antosh_ has joined #openstack-nova21:30
*** oomichi has joined #openstack-nova21:32
*** antosh has quit IRC21:33
*** alexpilotti has joined #openstack-nova21:37
*** shaohe_feng has quit IRC21:37
*** shaohe_feng has joined #openstack-nova21:38
mriedemdansmith: you should probably approve this spec amendment since you and i were the ones to originally approve the spec and then the author (or gmann) realized it missed something which lead to the amendment https://review.openstack.org/#/c/558125/21:40
mriedemand for whatever reason the person is waiting on this before writing code21:40
*** oomichi has quit IRC21:46
*** Guest50666 has quit IRC21:46
*** shaohe_feng has quit IRC21:47
*** shaohe_feng has joined #openstack-nova21:48
*** jmlowe has joined #openstack-nova21:49
openstackgerritMatt Riedemann proposed openstack/nova stable/pike: cleanup mapping/reqspec after archive instance  https://review.openstack.org/58077521:52
*** edmondsw_ has quit IRC21:53
openstackgerritEric Fried proposed openstack/nova master: WIP: Tighten up ReportClient use of generation  https://review.openstack.org/55666921:54
*** shaohe_feng has quit IRC21:58
*** shaohe_feng has joined #openstack-nova21:58
mriedemvdrok: our runways queue is getting pretty thin, if https://blueprints.launchpad.net/nova/+spec/allow-reserved-equal-total-inventory is ready to go you might want to throw it in the queue22:02
mriedemhttps://etherpad.openstack.org/p/nova-runways-rocky22:02
*** shaohe_feng has quit IRC22:08
*** shaohe_feng has joined #openstack-nova22:08
*** mriedem has quit IRC22:12
*** moshele has joined #openstack-nova22:13
*** shaohe_feng has quit IRC22:18
*** pchavva has quit IRC22:19
*** moshele has quit IRC22:19
*** shaohe_feng has joined #openstack-nova22:20
*** hongbin has quit IRC22:25
*** tssurya has quit IRC22:27
*** shaohe_feng has quit IRC22:28
*** shaohe_feng has joined #openstack-nova22:29
*** shaohe_feng has quit IRC22:39
*** shaohe_feng has joined #openstack-nova22:40
openstackgerritRick Bartra proposed openstack/nova master: Add granularity to os-services API  https://review.openstack.org/57855322:40
*** figleaf is now known as edleafe22:46
*** shaohe_feng has quit IRC22:49
*** edmondsw has joined #openstack-nova22:49
*** shaohe_feng has joined #openstack-nova22:51
*** edmondsw has quit IRC22:54
*** wwriverrat has quit IRC22:59
*** shaohe_feng has quit IRC22:59
*** shaohe_feng has joined #openstack-nova23:02
*** eharney has quit IRC23:04
*** shaohe_feng has quit IRC23:09
*** shaohe_feng has joined #openstack-nova23:09
*** vladikr has quit IRC23:14
openstackgerritKen'ichi Ohmichi proposed openstack/nova master: Remove unnecessary TODO on ubuntu installation  https://review.openstack.org/58077923:14
*** shaohe_feng has quit IRC23:20
*** shaohe_feng has joined #openstack-nova23:20
*** vladikr has joined #openstack-nova23:28
*** shaohe_feng has quit IRC23:30
*** shaohe_feng has joined #openstack-nova23:31
*** antosh_ has quit IRC23:34
*** rcernin has joined #openstack-nova23:36
*** masuberu has joined #openstack-nova23:38
*** shaohe_feng has quit IRC23:40
*** masber has quit IRC23:41
*** shaohe_feng has joined #openstack-nova23:42
*** shaohe_feng has quit IRC23:50
*** shaohe_feng has joined #openstack-nova23:54
*** _pewp_ has quit IRC23:56
*** _pewp_ has joined #openstack-nova23:56

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!