Wednesday, 2015-09-23

*** vilobhmm111 has quit IRC00:00
*** tsekiyam_ has joined #openstack-nova00:04
*** jerrygb_ has joined #openstack-nova00:05
*** mtanino has quit IRC00:07
*** tsekiyama has quit IRC00:08
*** jerrygb_ has quit IRC00:08
*** tsekiyam_ has quit IRC00:09
*** jerrygb has quit IRC00:09
*** rushil has quit IRC00:10
*** su_zhang has joined #openstack-nova00:13
*** Guest18166 is now known as dims_00:17
*** su_zhang has quit IRC00:18
*** thorst has joined #openstack-nova00:18
*** RA has joined #openstack-nova00:19
*** RA is now known as Guest3588600:19
*** su_zhang has joined #openstack-nova00:22
*** leeantho has quit IRC00:26
*** ijw has quit IRC00:28
*** lucas-dinner has quit IRC00:28
*** stevemar has joined #openstack-nova00:33
*** exploreshaifali has quit IRC00:33
*** eliqiao has joined #openstack-nova00:34
eliqiaohi sdague, around?00:34
*** stevemar has quit IRC00:36
*** takedakn has joined #openstack-nova00:36
*** zzzeek has quit IRC00:36
*** davideagnello has quit IRC00:37
*** davideagnello has joined #openstack-nova00:37
*** hitalia has quit IRC00:37
*** ssurana has quit IRC00:39
*** markvoelker has joined #openstack-nova00:40
*** thorst has quit IRC00:40
*** achanda has quit IRC00:41
*** xyang1 has quit IRC00:42
*** vilobhmm11 has joined #openstack-nova00:47
*** ijw has joined #openstack-nova00:48
*** gyee has quit IRC00:49
*** davideagnello has quit IRC00:49
*** davideagnello has joined #openstack-nova00:49
*** asettle has joined #openstack-nova00:50
*** whenry_ has joined #openstack-nova00:55
openstackgerritNicolas Simonds proposed openstack/nova: Fix deleted non-public flavor lookups for non-admin users  https://review.openstack.org/21180900:55
*** asettle has quit IRC00:56
*** Apoorva_ has joined #openstack-nova00:56
*** asettle has joined #openstack-nova00:56
*** EinstCrazy has joined #openstack-nova00:57
*** nic has quit IRC00:59
*** Apoorva has quit IRC00:59
*** Apoorva_ has quit IRC00:59
*** takedakn has quit IRC01:00
*** Apoorva has joined #openstack-nova01:00
*** Apoorva has quit IRC01:03
*** ijw has quit IRC01:03
*** Apoorva_ has joined #openstack-nova01:03
*** angdraug has quit IRC01:04
*** Apoorva_ has quit IRC01:05
*** Apoorva has joined #openstack-nova01:05
*** asettle has quit IRC01:06
*** asettle has joined #openstack-nova01:07
*** Apoorva has quit IRC01:09
*** Apoorva has joined #openstack-nova01:09
*** achanda has joined #openstack-nova01:11
*** stevemar has joined #openstack-nova01:11
*** tonytan4ever has quit IRC01:13
*** whenry_ has quit IRC01:19
*** su_zhang has quit IRC01:24
*** whenry_ has joined #openstack-nova01:25
*** nelsnels_ has joined #openstack-nova01:30
*** nelsnelson has quit IRC01:33
*** rushil has joined #openstack-nova01:38
*** shuquan has joined #openstack-nova01:39
*** jerrygb has joined #openstack-nova01:44
*** jerrygb has quit IRC01:45
*** jerrygb has joined #openstack-nova01:45
*** Apoorva has quit IRC01:46
alex_xumorning nova01:46
*** salv-orlando has joined #openstack-nova01:46
*** Daisy has joined #openstack-nova01:47
*** rushil has quit IRC01:48
*** whenry_ has quit IRC01:51
*** shuquan has quit IRC01:54
*** shuquan has joined #openstack-nova01:56
*** salv-orlando has quit IRC01:56
*** shuquan has quit IRC01:57
*** dims_ has quit IRC01:57
*** baoli has joined #openstack-nova01:57
*** dims_ has joined #openstack-nova01:57
*** haomaiwang has joined #openstack-nova01:58
*** rushil has joined #openstack-nova01:58
*** shuquan has joined #openstack-nova01:58
*** rushil_ has joined #openstack-nova01:59
*** haomaiwang has quit IRC02:01
eliqiaomorning.02:01
*** haomaiwang has joined #openstack-nova02:01
*** kiranr has joined #openstack-nova02:01
*** yuntongjin has joined #openstack-nova02:02
*** rushil has quit IRC02:03
openstackgerritZhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding  https://review.openstack.org/21756702:04
*** shyu has joined #openstack-nova02:05
*** yamahata has joined #openstack-nova02:08
*** dims_ has quit IRC02:10
*** asettle has quit IRC02:10
*** dims_ has joined #openstack-nova02:13
*** dims__ has joined #openstack-nova02:18
*** dims_ has quit IRC02:18
*** bswartz has joined #openstack-nova02:19
*** mwagner_lap has joined #openstack-nova02:21
*** darrenc is now known as darrenc_afk02:21
*** dims__ has quit IRC02:23
*** dave-mccowan has quit IRC02:26
openstackgerritJeffrey Zhang proposed openstack/nova: Remove useless code in _poll_volume_usage function  https://review.openstack.org/22394702:29
*** yamahata has quit IRC02:29
*** unicell has quit IRC02:33
*** baoli has quit IRC02:37
*** ferest has joined #openstack-nova02:38
*** baoli has joined #openstack-nova02:38
*** ferest has quit IRC02:46
*** asettle has joined #openstack-nova02:49
*** darrenc_afk is now known as darrenc02:49
*** whenry_ has joined #openstack-nova02:50
*** eliqiao has left #openstack-nova02:56
*** eliqiao has joined #openstack-nova02:56
*** haomaiwang has quit IRC03:01
*** markvoelker has quit IRC03:01
*** haomaiwa_ has joined #openstack-nova03:01
openstackgerritRafael Folco proposed openstack/nova: libvirt: check min required qemu/libvirt versions on ppc64/ppc64le  https://review.openstack.org/22655503:07
*** nelsnels_ has quit IRC03:08
*** boris-42 has quit IRC03:09
openstackgerritRafael Folco proposed openstack/nova: libvirt: check min required qemu/kvm versions on ppc64/ppc64le  https://review.openstack.org/22655503:09
*** shuquan has quit IRC03:10
*** shuquan has joined #openstack-nova03:11
*** david-lyle has joined #openstack-nova03:13
*** Daisy has quit IRC03:14
*** nelsnelson has joined #openstack-nova03:14
*** Daisy has joined #openstack-nova03:14
*** markvoelker has joined #openstack-nova03:16
*** ssurana has joined #openstack-nova03:18
*** zhangjn_ has joined #openstack-nova03:18
*** su_zhang has joined #openstack-nova03:18
*** EinstCrazy has quit IRC03:19
*** zhangjn has quit IRC03:20
*** otter768 has quit IRC03:21
*** shuquan has quit IRC03:25
*** shuquan has joined #openstack-nova03:28
*** ssurana has quit IRC03:28
*** ssurana has joined #openstack-nova03:28
*** fawadkhaliq has joined #openstack-nova03:32
*** ssurana has quit IRC03:32
*** baoli has quit IRC03:32
*** Daisy has quit IRC03:33
*** Daisy has joined #openstack-nova03:33
*** Daisy has quit IRC03:38
*** ijw has joined #openstack-nova03:44
*** armax has quit IRC03:44
*** shuquan has quit IRC03:45
*** Marga__ has quit IRC03:46
*** Marga_ has joined #openstack-nova03:46
*** dims_ has joined #openstack-nova03:47
*** diogogmt has quit IRC03:48
*** ijw has quit IRC03:49
*** Marga_ has quit IRC03:50
*** shuquan has joined #openstack-nova03:53
*** rushil_ has quit IRC03:53
*** dims_ has quit IRC03:54
*** jerrygb has quit IRC03:56
*** mdrabe has joined #openstack-nova03:56
*** shuquan has quit IRC03:58
*** haomaiwa_ has quit IRC04:01
*** haomaiwang has joined #openstack-nova04:01
*** Marga_ has joined #openstack-nova04:02
*** changbl has quit IRC04:04
*** yuntongjin has quit IRC04:05
*** changbl has joined #openstack-nova04:09
*** ijw has joined #openstack-nova04:09
*** ijw has quit IRC04:09
*** ijw has joined #openstack-nova04:09
openstackgerritYingxin proposed openstack/nova: Fix MetricWeigher to use MonitorMetricList  https://review.openstack.org/22162004:10
*** mdrabe has quit IRC04:12
*** pixelb has joined #openstack-nova04:13
*** jwcroppe has joined #openstack-nova04:14
*** whenry_ has quit IRC04:16
*** achanda has quit IRC04:17
*** salv-orlando has joined #openstack-nova04:17
openstackgerritRajesh Tailor proposed openstack/nova: Set migration status to 'error' on live-migration failure  https://review.openstack.org/21548304:18
*** pixelb has quit IRC04:19
*** hitalia has joined #openstack-nova04:19
*** zhangjn_ has quit IRC04:24
*** zhangjn has joined #openstack-nova04:25
*** zhangjn_ has joined #openstack-nova04:28
*** zhangjn has quit IRC04:29
*** kiranr has quit IRC04:32
*** zhangjn_ has quit IRC04:32
*** zhangjn has joined #openstack-nova04:33
*** fawadk has joined #openstack-nova04:33
*** salv-orlando has quit IRC04:35
*** fawadkhaliq has quit IRC04:35
*** Marga_ has quit IRC04:36
*** Marga_ has joined #openstack-nova04:37
*** zhangjn_ has joined #openstack-nova04:39
*** yuntongjin has joined #openstack-nova04:40
*** deepthi has joined #openstack-nova04:40
*** zhangjn has quit IRC04:42
*** zhangjn_ has quit IRC04:42
*** zhangjn has joined #openstack-nova04:43
*** Sree has joined #openstack-nova04:44
*** Sree_ has joined #openstack-nova04:46
*** Sree_ is now known as Guest1466204:46
*** achanda has joined #openstack-nova04:48
*** yuntongjin2 has joined #openstack-nova04:48
*** Sree has quit IRC04:49
*** yuntongjin has quit IRC04:49
*** asettle has quit IRC04:55
*** deepakcs has joined #openstack-nova04:56
*** zhangjn has quit IRC04:57
*** Sree has joined #openstack-nova04:59
*** zhangjn has joined #openstack-nova04:59
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-nova05:01
*** Guest14662 has quit IRC05:02
*** ildikov has quit IRC05:21
*** otter768 has joined #openstack-nova05:22
*** otter768 has quit IRC05:26
*** sahid has joined #openstack-nova05:29
*** sahid_ has joined #openstack-nova05:35
*** salv-orlando has joined #openstack-nova05:45
*** boris-42 has joined #openstack-nova05:48
*** dims_ has joined #openstack-nova05:50
*** salv-orlando has quit IRC05:52
*** su_zhang has quit IRC05:54
*** kiran-r has joined #openstack-nova05:54
*** dims_ has quit IRC05:55
*** kiranr has joined #openstack-nova05:58
*** ildikov has joined #openstack-nova06:00
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-nova06:01
*** kiran-r has quit IRC06:02
*** kiranr has quit IRC06:02
openstackgerritHiroyuki Eguchi proposed openstack/nova: Cannot rebuild a instance booted from volume  https://review.openstack.org/20145806:06
*** Sree has quit IRC06:07
*** jerrygb has joined #openstack-nova06:08
*** Sree has joined #openstack-nova06:08
*** jerrygb has quit IRC06:12
*** fawadk has quit IRC06:14
*** fawadkhaliq has joined #openstack-nova06:17
*** scheuran has joined #openstack-nova06:17
*** bswartz has quit IRC06:18
*** Guest35886 has quit IRC06:18
eliqiaoalex_xu ping06:20
alex_xueliqiao: pong06:20
eliqiaore https://review.openstack.org/#/c/224438/6/nova/utils.py06:20
eliqiaowhat cases you expecting test.06:20
*** bswartz has joined #openstack-nova06:21
alex_xueliqiao: dnsmasq in linux and windows the max length of hostname supported06:21
*** nkrinner has joined #openstack-nova06:21
alex_xueliqiao: do you have windows virtual matchine?06:22
*** stevemar has quit IRC06:23
eliqiaoalex_xu:  I don't06:23
*** jwcroppe has quit IRC06:23
alex_xueliqiao: ok, let me try to find someone have one06:23
*** jwcroppe has joined #openstack-nova06:23
eliqiao alex_xu:  I am think we are over-kill on this case.06:25
alex_xueliqiao: why?06:25
*** jwcroppe has quit IRC06:28
*** yuntongjin2 has quit IRC06:30
eliqiaoit is hacking we do this in nova, (I mean for the truncation stuff), metadata service /config-drive will grab hostname from nova's database and will set it to dnsmasq service or Operatation System (/etc/hostname)06:31
*** eliqiao has left #openstack-nova06:31
*** vilobhmm11 has quit IRC06:31
*** eliqiao has joined #openstack-nova06:31
*** fawadk has joined #openstack-nova06:32
*** EinstCrazy has joined #openstack-nova06:32
alex_xueliqiao: if we don't truncate it, then the dns name for host won't works06:33
eliqiaomaybe I think it too simple. but as sean side, what if it's a very long host name like "aaa.bbb.ccc.ddd" , after truncation , we get "aaa.b" which can not be resolved?06:33
*** Murali has joined #openstack-nova06:33
*** fawadkhaliq has quit IRC06:33
eliqiaofor this case, how can we do the truncation?06:33
alex_xueliqiao: Actually Sean suggest warning log for truncation06:33
eliqiaosean's words : "Also, truncation shouldn't be silent, because it could produce things like foo.c which won't resolve."06:33
eliqiaoalex_xu:  give a log warnning makes some sense06:34
eliqiaoalex_xu: would you please advise how to testing the maximum length of an vm's hostname could be?06:37
alex_xueliqiao: actually I'm try to test now06:37
*** romainh has joined #openstack-nova06:40
*** abhishekk has joined #openstack-nova06:46
*** rdopiera has joined #openstack-nova06:47
*** fawadkhaliq has joined #openstack-nova06:51
*** kiranr has joined #openstack-nova06:51
*** fawadk has quit IRC06:52
*** karimb has joined #openstack-nova06:54
*** sahid_ has quit IRC06:54
*** matrohon has joined #openstack-nova06:55
*** alex_klimov has joined #openstack-nova06:55
*** alex_klimov has quit IRC06:56
*** alex_klimov has joined #openstack-nova06:56
*** sahid_ has joined #openstack-nova06:56
eliqiaoalex_xu not sure if this is what you want http://www.freesoft.org/CIE/RFC/1035/9.htm06:57
*** browne has quit IRC06:57
*** fawadk has joined #openstack-nova06:58
alex_xueliqiao: rfc said 255, but actually situation may not06:58
*** snikitin has joined #openstack-nova06:59
*** fawadkhaliq has quit IRC07:00
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-nova07:01
eliqiaoalex_xu: free to come to my cube?07:02
alex_xueliqiao: yea07:02
*** lpetrut has joined #openstack-nova07:04
*** fawadkhaliq has joined #openstack-nova07:09
*** fawadk has quit IRC07:10
*** abhishekk_ has joined #openstack-nova07:10
*** rajesht__ has joined #openstack-nova07:10
*** kodokuu has joined #openstack-nova07:11
*** gszasz has joined #openstack-nova07:11
*** abhishekk has quit IRC07:12
*** rajesht_ has quit IRC07:13
*** sahid_ has quit IRC07:14
*** sahid has quit IRC07:14
*** sahid has joined #openstack-nova07:14
*** achanda has quit IRC07:14
*** karimb has quit IRC07:15
*** markvoelker has quit IRC07:15
*** karimb has joined #openstack-nova07:16
ndipanovalex_xu, found a bug in that last patch you already reviewed. Will push the change if you could take a look that would be awesome07:20
*** otter768 has joined #openstack-nova07:23
*** stevemar has joined #openstack-nova07:23
*** jwcroppe has joined #openstack-nova07:23
*** jichen has joined #openstack-nova07:25
*** ptm_ has joined #openstack-nova07:25
*** karimb has quit IRC07:27
*** stevemar has quit IRC07:27
*** otter768 has quit IRC07:28
*** ptm__ has joined #openstack-nova07:28
*** cfriesen__ has quit IRC07:28
alex_xundipanov: sure, will do after few minutes07:29
*** ptm_ has quit IRC07:31
*** acc_ has quit IRC07:31
*** karimb has joined #openstack-nova07:32
*** sayali has quit IRC07:34
*** lpetrut has quit IRC07:36
*** sayali has joined #openstack-nova07:38
ndipanovalex_xu, still not posted :(07:40
*** salv-orlando has joined #openstack-nova07:40
alex_xundipanov: yea...I just read the comment07:40
alex_xundipanov: basically, 'resize' in 'resize' works, but yes, it is wrong. Maybe it is ok for now?07:41
* alex_xu didn't get the meaning of ' this should not work for old flavors' yet, thinking07:42
*** Sree_ has joined #openstack-nova07:42
*** Sree_ is now known as Guest8906307:43
ndipanovalex_xu, trying to figure that out07:45
*** Sree has quit IRC07:46
*** jlanoux has joined #openstack-nova07:46
*** salv-orl_ has joined #openstack-nova07:47
*** salv-orlando has quit IRC07:48
ndipanovalex_xu, so it should actually be new flavors07:48
ndipanovwe will never get the new flavor this way07:49
*** fawadkhaliq has quit IRC07:49
*** fawadkhaliq has joined #openstack-nova07:50
alex_xundipanov: sorry...I didn't get why that related to new/old flavor..07:52
*** dims_ has joined #openstack-nova07:52
alex_xundipanov: maybe we just need migration.migration_type == 'resize', we needn't 'in'07:53
ndipanovalex_xu, ah you are right resize in resize works07:53
ndipanovok never mind07:53
alex_xundipanov: yea, never mind :)07:53
ndipanovalex_xu, only one cup of coffee07:54
openstackgerritEli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty  https://review.openstack.org/22443807:56
*** danpb has joined #openstack-nova07:57
*** salv-orl_ has quit IRC07:59
*** salv-orlando has joined #openstack-nova07:59
*** dims_ has quit IRC07:59
bauzasmorning Nova08:00
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-nova08:01
*** kodokuu_ has joined #openstack-nova08:01
*** kodokuu has quit IRC08:04
*** BobBall_AWOL is now known as BobBall08:04
alex_xueliqiao: https://review.openstack.org/224438 there are some typo problem on the user-face message08:05
alex_xueliqiao: and some suggestion about NOTE08:05
eliqiaofare enough alex_xu, thx08:06
alex_xueliqiao: np08:06
*** markus_z has joined #openstack-nova08:06
alex_xueliqiao: and missing one use-case for truncate_hostname on default hostname08:07
alex_xueliqiao: sorry, s/use-case/test-case/08:08
*** haomaiwang has quit IRC08:10
*** haomaiwa_ has joined #openstack-nova08:12
*** derekh has joined #openstack-nova08:13
eliqiaoalex_xu: ya, I get you , and we though it on same case.08:15
*** aix has joined #openstack-nova08:15
alex_xueliqiao: cool08:15
*** yassine__ has joined #openstack-nova08:15
*** markvoelker has joined #openstack-nova08:15
*** jwcroppe has quit IRC08:19
*** lucas-dinner has joined #openstack-nova08:19
*** lucasagomes has joined #openstack-nova08:19
openstackgerritEli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty  https://review.openstack.org/22443808:20
eliqiaoalex_xu: ^^^08:20
*** markvoelker has quit IRC08:20
openstackgerritjichenjc proposed openstack/nova: Remove duplicate VALID_NAME_REGEX  https://review.openstack.org/21848908:20
alex_xueliqiao: cool, thanks08:20
*** ptm__ has quit IRC08:23
*** Kevin_Zheng has joined #openstack-nova08:24
*** kiran-r has joined #openstack-nova08:29
*** kiran-r has quit IRC08:34
*** atuvenie has joined #openstack-nova08:35
*** alex_klimov has quit IRC08:36
*** alex_klimov has joined #openstack-nova08:36
*** Daisy has joined #openstack-nova08:39
*** kiran-r has joined #openstack-nova08:41
ndipanovjohnthetubaguy, https://review.openstack.org/#/c/226411/08:44
ndipanovalthough that one can be backported - but it has 2x +2s08:45
*** sudipto has joined #openstack-nova08:46
openstackgerritAlex Xu proposed openstack/nova: Give instance default hostname if hostname is empty  https://review.openstack.org/22443808:54
*** Sree has joined #openstack-nova08:57
*** Guest89063 has quit IRC09:00
*** haomaiwa_ has quit IRC09:01
*** _RuiChen has quit IRC09:01
*** 16WAAB8UT has joined #openstack-nova09:01
*** _RuiChen has joined #openstack-nova09:02
*** kiranr has quit IRC09:02
*** lucas-dinner has quit IRC09:07
johnthetubaguyndipanov: sweet, so I how do you want to deal with this bug, do you want to create a new one for the remaining bits, and close out this one for liberty?09:07
johnthetubaguyalex_xu: how are we doing with that bug patch?09:08
*** jerrygb has joined #openstack-nova09:08
johnthetubaguyndipanov: we can just leave it open, but that seems less satisfying some how09:08
*** lpetrut has joined #openstack-nova09:09
*** dims_ has joined #openstack-nova09:11
*** jerrygb has quit IRC09:13
*** atuvenie has quit IRC09:14
*** kiranr has joined #openstack-nova09:15
*** e0ne has joined #openstack-nova09:15
*** jwcroppe has joined #openstack-nova09:16
*** otter768 has joined #openstack-nova09:24
*** stevemar has joined #openstack-nova09:25
*** RuiChen has joined #openstack-nova09:28
openstackgerritJohn Garbutt proposed openstack/nova: Filter leading/trailing spaces for name field in v2.1 compat mode  https://review.openstack.org/22203209:28
*** stevemar has quit IRC09:28
*** otter768 has quit IRC09:28
*** _RuiChen has quit IRC09:29
*** alexpilotti has joined #openstack-nova09:32
* danpb wonders when master will open for mitaka merges09:34
*** ociuhandu has quit IRC09:36
johnthetubaguydanpb: once we get all the RC1 bugs closed09:37
* danpb untags all bugs from RC1 ;-P09:37
johnthetubaguyI hoped it would have been yesterday, currently gunning for thursday, unless we get done sooner09:37
johnthetubaguydanpb: honestly, I am tempted to do that at this point09:37
*** Daisy has quit IRC09:39
*** Daisy has joined #openstack-nova09:40
alex_xujohnthetubaguy: I think this is ready, today I and eliqiao test windows,linxu and dnsmasq https://review.openstack.org/22443809:42
*** redbeard has joined #openstack-nova09:42
*** kodokuu_ has quit IRC09:43
*** Daisy has quit IRC09:44
*** Daisy has joined #openstack-nova09:44
johnthetubaguyalex_xu: wow, limit of 63, sounds like the best choice though09:44
alex_xujohnthetubaguy: yea09:44
johnthetubaguyalex_xu: is our default name larger than that?09:44
*** dims_ has quit IRC09:44
alex_xujohnthetubaguy: no, actually not, but I saw eliqiao add that, I think it's fine, if someday we make default name longer and the thing won't be broken09:45
johnthetubaguyyeah, just thinking it through, I think we are good09:45
alex_xucool09:45
johnthetubaguyalex_xu: so thinking about to the "difficult" patch09:46
johnthetubaguyI restored your patchset 509:47
johnthetubaguyI am tempted to say we go for that09:47
*** jwcroppe_ has joined #openstack-nova09:47
alex_xujohnthetubaguy: yea...09:47
johnthetubaguyso I have one comment so far, let me send it09:47
johnthetubaguyalex_xu: got a question here: https://review.openstack.org/#/c/222032/7/nova/api/openstack/compute/flavor_manage.py,cm09:47
alex_xujohnthetubaguy: actually I commented also, about another tradeoff, anyway I don't think we match legacy_v2 api behaviour totally is right09:47
* alex_xu is checking09:48
alex_xujohnthetubaguy: actually it's done at underlayer https://github.com/openstack/nova/blob/master/nova/compute/flavors.py#L10909:49
*** e0ne has quit IRC09:49
*** hitalia has quit IRC09:49
*** jwcroppe has quit IRC09:49
alex_xumaybe I should use normailize_name in underlayer also09:49
johnthetubaguyalex_xu: ah, OK, its tempting to drop that one, and add it in the API for consistency09:50
johnthetubaguyalex_xu: but honestly, that could be a follow up patch09:50
*** e0ne has joined #openstack-nova09:50
alex_xujohnthetubaguy: yea, that is good way, but need check ec2, legacy v2 and v2.109:50
alex_xujohnthetubaguy: yea, thanks09:50
johnthetubaguyalex_xu: yeah, totally09:50
*** Sree has quit IRC09:52
bauzasjohnthetubaguy: how can I communicate that one RC1 bug is okay to be merged ?09:53
bauzasie. https://review.openstack.org/#/c/221620/509:53
johnthetubaguybauzas: not sure what you are asking09:54
johnthetubaguybauzas: I guess its RC1-potential09:54
johnthetubaguyoh, wait, thats the really bad regression right?09:54
bauzasjohnthetubaguy: I mean, there are 21 bugs in https://bugs.launchpad.net/nova/+bugs?field.tag=liberty-rc-potential09:54
johnthetubaguyyeah, there are09:54
bauzasjohnthetubaguy: but I reviewed https://bugs.launchpad.net/nova/+bug/1493680 and I consider the bugfix acceptable09:55
openstackLaunchpad bug 1493680 in OpenStack Compute (nova) "HostState.metrics is no longer a dict" [High,In progress] - Assigned to Yingxin (cyx1231st)09:55
bauzasjohnthetubaguy: so I just wonder how I could send the bat-signal that it's worth reviewing it09:55
bauzasof course I can ping you guys :)09:56
bauzasbut are we still using the etherpad of doom ?09:56
bauzasthat's my wonders09:56
bauzasjohnthetubaguy: anyhow, I don't want to overthink09:58
bauzasjohnthetubaguy: pinging should be enough09:58
johnthetubaguyyeah, we don't have an etherpad for this right now09:59
*** claudiub has joined #openstack-nova10:00
*** 16WAAB8UT has quit IRC10:01
*** lan has joined #openstack-nova10:01
*** Daisy has quit IRC10:01
*** 6JTAB3MXI has joined #openstack-nova10:01
*** Kennan_Vacation has quit IRC10:01
*** Daisy has joined #openstack-nova10:02
*** Kennan_Vacation has joined #openstack-nova10:02
bauzasfair enough :)10:02
*** atuvenie has joined #openstack-nova10:03
*** Daisy has quit IRC10:05
*** Daisy has joined #openstack-nova10:05
*** dims_ has joined #openstack-nova10:09
*** alex_klimov has quit IRC10:09
*** ndipanov has quit IRC10:09
*** eglynn has joined #openstack-nova10:10
*** kiran-r has quit IRC10:10
*** ndipanov has joined #openstack-nova10:11
alex_xusdague: hope you can take a look at https://review.openstack.org/224438 and https://review.openstack.org/222032 when you wakeup, thanks10:13
*** IanGovett has joined #openstack-nova10:16
*** markvoelker has joined #openstack-nova10:17
*** markvoelker has quit IRC10:21
openstackgerritHans Lindgren proposed openstack/nova: Remove unused refresh_security_group_members() call  https://review.openstack.org/18263210:22
openstackgerritHans Lindgren proposed openstack/nova: Make secgroup rules refresh with refresh_instance_security_rules()  https://review.openstack.org/18263310:22
*** alex_klimov has joined #openstack-nova10:24
*** Daisy has quit IRC10:25
*** Daisy has joined #openstack-nova10:26
openstackgerritJian LI proposed openstack/nova: Sync vms az with the host when using aggregate  https://review.openstack.org/22668310:27
*** ociuhandu has joined #openstack-nova10:29
*** Daisy has quit IRC10:30
*** Daisy has joined #openstack-nova10:30
*** jwcroppe_ has quit IRC10:30
*** jwcroppe has joined #openstack-nova10:31
*** Daisy has quit IRC10:31
*** Daisy has joined #openstack-nova10:32
*** jwcroppe has quit IRC10:35
*** Daisy has quit IRC10:36
sdaguealex_xu: ok, just waking, let me see10:36
alex_xusdague: thanks10:37
*** alex_klimov has quit IRC10:39
*** alex_klimov has joined #openstack-nova10:40
*** alex_klimov has quit IRC10:40
*** alex_klimov has joined #openstack-nova10:40
openstackgerritHans Lindgren proposed openstack/nova: Remove and deprecate conductor provider_fw_rule_get_all()  https://review.openstack.org/21624110:43
openstackgerritHans Lindgren proposed openstack/nova: Remove unused provider firewall rules functionality in nova  https://review.openstack.org/18402710:43
openstackgerritHans Lindgren proposed openstack/nova: Remove provider_fw_rules tables and associated db api methods  https://review.openstack.org/21624210:43
openstackgerritAnkit Agrawal proposed openstack/nova: Remove unnecessary call to info_cache.delete  https://review.openstack.org/22669010:44
openstackgerritSergey Nikitin proposed openstack/nova: WIP: Use legacy engineFacade from oslo_db.enginefacade  https://review.openstack.org/22131010:44
openstackgerritSergey Nikitin proposed openstack/nova: WIP: use oslo_db.enginefacade  https://review.openstack.org/22506510:45
*** jichen has quit IRC10:46
*** lucasagomes has quit IRC10:48
*** exploreshaifali has joined #openstack-nova10:51
*** EinstCrazy has quit IRC10:51
*** zhangjn has quit IRC10:51
*** kiran-r has joined #openstack-nova10:53
*** lucasagomes has joined #openstack-nova10:53
*** lucasagomes_ has joined #openstack-nova11:00
*** lucasagomes_ has left #openstack-nova11:00
*** 6JTAB3MXI has quit IRC11:01
*** haomaiwang has joined #openstack-nova11:01
*** shyu has quit IRC11:02
*** aix has quit IRC11:03
alex_xujohnthetubaguy: bauzas, looks like 'target' have chance to be 'None' in https://review.openstack.org/#/c/221620/5/nova/scheduler/weights/metrics.py line 10311:04
johnthetubaguyalex_xu: yep, you are right, good spot11:05
alex_xujohnthetubaguy: thanks, so looks like that patch missing an testcase also11:06
*** Daisy has joined #openstack-nova11:06
*** salv-orlando has quit IRC11:12
*** salv-orlando has joined #openstack-nova11:13
*** lan has quit IRC11:13
*** Daisy has quit IRC11:14
*** Daisy has joined #openstack-nova11:15
*** lucasagomes is now known as lucas-hungry11:15
*** Daisy has quit IRC11:17
*** markvoelker has joined #openstack-nova11:17
*** Daisy has joined #openstack-nova11:19
*** jwcroppe has joined #openstack-nova11:21
*** markvoelker has quit IRC11:22
*** haomaiwang has quit IRC11:22
*** baoli has joined #openstack-nova11:23
*** otter768 has joined #openstack-nova11:25
*** shakamunyi has quit IRC11:25
rajesht__johnthetubaguy: if u hv time, could you please look at this backport https://review.openstack.org/#/c/223968/11:26
rajesht__johnthetubaguy: thanks :)11:26
johnthetubaguyrajesht__: I am not on stable-core I am afraid11:26
rajesht__johnthetubaguy: ok, but you can give your suggestion, if you want :)11:27
*** otter768 has quit IRC11:29
dims_rajesht__: stable folks hangout on #openstack-stable11:32
sudiptoalex_xu, https://review.openstack.org/#/c/221620/5/nova/scheduler/weights/metrics.py - for your comment on this - w.r.t None.value - do you ever see that happening?11:32
*** Sree has joined #openstack-nova11:32
rajesht__dims_: thanks for info :)11:32
*** gongysh has joined #openstack-nova11:32
sudiptoalex_xu, since if not target has a if/else inside it - will not always return before ever reaching there?11:33
*** baoli_ has joined #openstack-nova11:33
gongyshhi, who can tell me where to find the document about nova aggregator and cell document? thanks11:33
sudiptoalex_xu, the if raises an exception and the else returns.11:34
*** aix has joined #openstack-nova11:34
*** jerrygb has joined #openstack-nova11:35
*** baoli has quit IRC11:36
*** mwagner_lap has quit IRC11:37
*** Daisy has quit IRC11:39
*** jerrygb has quit IRC11:39
openstackgerritJohn Garbutt proposed openstack/nova-specs: Locking resources by service token  https://review.openstack.org/20388011:42
openstackgerritJohn Garbutt proposed openstack/nova-specs: Locking servers by service token  https://review.openstack.org/20388011:43
openstackgerritHiroyuki Eguchi proposed openstack/nova: Cannot rebuild a instance booted from volume  https://review.openstack.org/20145811:43
openstackgerritAnkit Agrawal proposed openstack/nova: Detach volume after deleting instance with no host  https://review.openstack.org/19406311:47
*** jerrygb has joined #openstack-nova11:48
*** Daisy has joined #openstack-nova11:49
*** cristicalin has joined #openstack-nova11:49
*** EinstCrazy has joined #openstack-nova11:49
alex_xusudipto: I didn't see that, but from the code, I guess the code means the scheduler configure want to skip that metric and that metric didn't report from compute node also11:50
*** sayali has quit IRC11:50
alex_xus/scheduler configure/scheduler configuration/11:50
*** Daisy has quit IRC11:51
sudiptoalex_xu, yeah it sounds like that.11:51
*** dave-mccowan has joined #openstack-nova11:51
sudiptoalex_xu, but i feel your comment is valid anyway - since there's a chance that if block won't be hit - and it will eventually come down to None.value.11:52
alex_xusudipto: yea, there is a chance11:53
*** sayali has joined #openstack-nova11:55
*** exploreshaifali has quit IRC11:56
*** thorst has joined #openstack-nova11:58
*** EinstCrazy has quit IRC11:59
*** krtaylor has quit IRC12:01
*** Daisy has joined #openstack-nova12:02
*** achanda has joined #openstack-nova12:03
*** zhangjn has joined #openstack-nova12:04
*** EinstCrazy has joined #openstack-nova12:04
*** ankit_ag has joined #openstack-nova12:04
*** EinstCrazy has quit IRC12:06
*** kiran-r has quit IRC12:10
*** krtaylor has joined #openstack-nova12:13
*** e0ne has quit IRC12:13
*** aysyd has joined #openstack-nova12:15
*** deepakcs has quit IRC12:15
*** pixelb has joined #openstack-nova12:17
*** achanda has quit IRC12:18
openstackgerritTomi Juvonen proposed openstack/nova-specs: Get valid server state  https://review.openstack.org/19224612:18
*** lucas-hungry is now known as lucasagomes12:19
*** Sree has quit IRC12:20
*** markvoelker has joined #openstack-nova12:20
*** fawadkhaliq has quit IRC12:21
*** kiranr has quit IRC12:22
*** raildo-afk is now known as raildo12:22
*** fawadkhaliq has joined #openstack-nova12:22
*** jwcroppe_ has joined #openstack-nova12:22
*** mwagner_lap has joined #openstack-nova12:23
*** zhenguo has quit IRC12:24
bauzasalex_xu: well, technically, it will never go to L103 if target is None12:24
bauzasjohnthetubaguy: ^12:24
ankit_agjohnthetubaguy: hi, you around ?12:25
bauzasalex_xu: johnthetubaguy: I should have commented that I was agreeing to leave that since we're very close to RC112:25
bauzasalex_xu: johnthetubaguy: and since I don't know the proposer, I preferred to leave +112:25
*** jwcroppe has quit IRC12:25
bauzasoh sec, yeah there is another if12:26
bauzasalex_xu: good point then, providing a new iteration12:26
*** takedakn has joined #openstack-nova12:27
*** zzzeek has joined #openstack-nova12:28
*** jwcroppe_ is now known as jwcroppe12:29
*** tpeoples has joined #openstack-nova12:29
*** Sree has joined #openstack-nova12:29
*** abhijeetm has joined #openstack-nova12:31
*** bswartz has quit IRC12:32
*** edmondsw has joined #openstack-nova12:32
*** diogogmt has joined #openstack-nova12:32
*** diogogmt has quit IRC12:33
alex_xubauzas: :)12:34
alex_xujohnthetubaguy: bauzas we need merge https://review.openstack.org/#/c/221620/ before firday, right?12:35
alex_xuoops, s/firday/friday/12:36
*** thorst has left #openstack-nova12:38
bauzasalex_xu: well, it's possible to merge it after and backport it as a candidate for a RC2, but I'd prefer to spin off a new change now12:39
bauzasalex_xu: that's why I'm working on a new PS12:40
bauzasalex_xu: should be done by the next mins12:40
alex_xubauzas: oh, got it, cool12:40
alex_xubauzas: thanks12:40
alex_xubauzas: I'm just thinking the author is from my team, I can notice him the time, but you will respin the new change, so it's fine now12:41
bauzasalex_xu: yeah, it seems he's Chinese, right?12:41
*** takedakn has quit IRC12:41
alex_xubauzas: yea, at ShangHai12:41
bauzasalex_xu: so I guess he already left12:41
*** takedakn has joined #openstack-nova12:42
alex_xubauzas: yes12:42
*** leitan has joined #openstack-nova12:42
bauzasalex_xu: that's not a problem, I was leaving a few comments that I can fix now :)12:42
alex_xubauzas: yea12:43
*** takedakn has quit IRC12:44
openstackgerritClaudiu Belu proposed openstack/nova-specs: Adds spec for instance live resize  https://review.openstack.org/14121912:44
claudiubjohnthetubaguy: hi, if you have some time, can you take a look at the host capabilities spec? https://review.openstack.org/#/c/222200/12:46
*** Daisy has quit IRC12:46
*** sayali has quit IRC12:47
*** Sree has quit IRC12:49
*** Sree has joined #openstack-nova12:49
*** takedakn has joined #openstack-nova12:49
*** sayali has joined #openstack-nova12:50
ankit_agsdague, dansmith: Hi, I have addressed your review comments on https://review.openstack.org/#/c/217579/ could you please take a look12:55
*** mdrabe has joined #openstack-nova12:58
*** thorst1 has joined #openstack-nova13:00
*** kiran-r has joined #openstack-nova13:01
*** rlrossit has joined #openstack-nova13:02
*** artom has joined #openstack-nova13:02
openstackgerritSylvain Bauza proposed openstack/nova: Fix MetricWeigher to use MonitorMetricList  https://review.openstack.org/22162013:04
bauzasalex_xu: johnthetubaguy: ^ please see the above respined13:04
*** deepthi has quit IRC13:07
*** salv-orlando has quit IRC13:07
*** haomaiwang has joined #openstack-nova13:08
*** jecarey has joined #openstack-nova13:11
bauzasalex_xu: /me needs some caffeine13:13
alex_xubauzas: :)13:13
*** ijuwang has joined #openstack-nova13:15
sdagueankit_ag: done13:17
ankit_agsdague: thanks :)13:17
*** ctrath has joined #openstack-nova13:17
*** sudipto has quit IRC13:18
*** dims_ has quit IRC13:21
*** dims_ has joined #openstack-nova13:21
*** alexpilotti has quit IRC13:22
*** gongysh has quit IRC13:22
*** ankit_ag has quit IRC13:23
*** burgerk has joined #openstack-nova13:24
openstackgerritSylvain Bauza proposed openstack/nova: Fix MetricWeigher to use MonitorMetricList  https://review.openstack.org/22162013:24
bauzasalex_xu: johnthetubaguy: ^ that (much better)13:24
*** otter768 has joined #openstack-nova13:25
*** takedakn has quit IRC13:25
*** jwcroppe has quit IRC13:26
*** stevemar has joined #openstack-nova13:26
*** dustins has joined #openstack-nova13:28
*** stevemar has quit IRC13:29
*** ildikov has quit IRC13:30
*** otter768 has quit IRC13:30
bauzasjohnthetubaguy: markus_z: so we have 72 new bugs, hard to tell whether we have regressions or not13:30
openstackgerritsahid proposed openstack/nova: libvirt: add cpu schedular priority config  https://review.openstack.org/19756713:30
openstackgerritsahid proposed openstack/nova: libvirt: rework membacking config to support future features  https://review.openstack.org/19756613:30
openstackgerritsahid proposed openstack/nova: virt: add contraint to handle realtime policy  https://review.openstack.org/19756813:30
openstackgerritsahid proposed openstack/nova: libvirt: add realtime support  https://review.openstack.org/19756913:30
openstackgerritsahid proposed openstack/nova: hardware: improve parse_cpu_spec to handle exclusion range  https://review.openstack.org/22678313:30
bauzasmarkus_z: I can see that you made huge work by triaging those and provide a tag13:31
mriedembauzas: many of those 'new' bugs are old bugs13:31
mriedemi was going through some yesterday13:31
bauzasmriedem: okay, my take is trying to identify over those 72 bugs if we missed something spotted which was crucial13:32
mriedemsome are actually in progres13:32
mriedemcan we filter on when they were reported?13:33
mriedemlike in the last 2 weeks?13:33
mriedemi guess not13:33
mriedemthanks launchpad13:33
bauzaslemme see13:33
sdaguemriedem: you can sort by it13:33
sdagueyou have to add another field13:33
bauzassdague: right13:34
sdaguemriedem: it's "age"13:34
bauzasyeah, can also see the tags now13:34
sdaguehttps://bugs.launchpad.net/nova/+bugs?orderby=-datecreated&start=013:34
mriedemyeah i see now13:35
mriedemthanks13:35
*** bradjones has quit IRC13:35
mriedemheh https://bugs.launchpad.net/nova/+bug/148422313:35
openstackLaunchpad bug 1484223 in OpenStack Compute (nova) " Revert "Revert "Add VIF_DELETED notification event to Nova""" [Undecided,New]13:35
*** jaypipes has joined #openstack-nova13:36
*** Murali has quit IRC13:36
sdagueoh, that's the docimpact auto bug flapping back our way13:36
mriedemyeah, i closed it13:36
mriedemit's already in the release notes13:37
*** rdopiera has quit IRC13:37
mriedemwe could maybe put something into the nova devref about the event stuff that happens between nova and neutron13:37
mriedembut i don't have the energy to write that up13:37
doffmsdague: mriedem: I'm happy to help go through them. Should I tag with 'regression' or just post here if I think something is important.13:37
*** abhishekk_ has quit IRC13:37
mriedemdoffm: just post any critical sounding things here13:38
mriedemno vmware people around13:39
mriedemthere are some new vmware bugs13:39
sdaguehttps://bugs.launchpad.net/nova/+bug/1494688 - isn't really a vmware bug, this is just the general requests gorp we get all the time13:39
*** Sree has quit IRC13:39
openstackLaunchpad bug 1494688 in oslo.vmware "[VMWARE] Connection pool is full, discarding connection" [Undecided,New]13:39
*** xyang1 has joined #openstack-nova13:39
sdaguewho's on irc that works on requests, it would be good to figure out how to handle stuff like that better13:40
mriedemsigmavirus?13:40
mriedemno idea really13:40
mriedemoh works on, idk13:41
dims_sdague: mriedem: sigmavirus24 i think13:41
openstackgerritPavel Kholkin proposed openstack/nova: WIP: use oslo_db.enginefacade  https://review.openstack.org/22678813:41
*** cfriesen__ has joined #openstack-nova13:41
*** lsmola has quit IRC13:41
sdaguebauzas: https://bugs.launchpad.net/nova/+bug/1495946 seems very related to your fixes around AZs a couple weeks ago13:41
openstackLaunchpad bug 1495946 in OpenStack Compute (nova) "Nova boot does not respect zone name if node is specified" [Undecided,New] - Assigned to Raga lahari (raga-lahari)13:41
sdaguemostly, you might have fixed it by accident, would be interesting to see what your take is on that one13:42
bauzassdague: actually, that's the force_hosts hack13:42
bauzassdague: there is one trick when you want to specify a destination on the boot command13:42
bauzassdague: then, you provide an az hint like 'az:host:node'13:43
bauzassdague: but that's not really an AZ check13:43
mriedemi thought most people that used vcenter also used neutron?13:43
mriedemhttps://bugs.launchpad.net/nova/+bug/149886513:43
openstackLaunchpad bug 1498865 in OpenStack Compute (nova) "VMware: nova-network creates a port group only on one ESXi in cluster in case with VlanManager" [Undecided,Confirmed]13:43
sdaguemriedem: we don't even support esxi any more right?13:43
bauzassdague: so, that's a bit unrelated to the AZ issues we discussed in the past13:43
sdaguebauzas: sure, I'm curious if given you going through that code you have a pointer on fixing that13:44
mriedemsdague: i think they are talking about the backend esxi hosts in the cluster13:44
bauzassdague: see the hack here https://github.com/openstack/nova/blob/master/nova/compute/api.py#L589-L60213:44
sdaguebecause, I agree, it should reject13:44
bauzassdague: I really hate that one, honestly, but that's in the API contract13:44
mriedembauzas: is there anything that you don't hate? :P13:44
bauzasmriedem: heh, I'm French, remember ?13:45
mriedemoh, right13:45
bauzasfortunately, we have a policy for that13:45
bauzassdague: https://github.com/openstack/nova/blob/master/etc/nova/policy.json#L1113:46
*** mpaolino has joined #openstack-nova13:46
*** burt has joined #openstack-nova13:48
doffmhttps://bugs.launchpad.net/nova/+bug/148766813:48
openstackLaunchpad bug 1487668 in OpenStack Compute (nova) "nova evacuate doesn't work with neutron dvr architecture" [Undecided,New]13:48
doffmhttps://bugs.launchpad.net/nova/+bug/149227413:49
openstackLaunchpad bug 1492274 in OpenStack Compute (nova) "nova evacuate does not update instance's neutron port location in the DB" [Undecided,New]13:49
*** jwcroppe has joined #openstack-nova13:49
doffmTwo bugs, both recent, could be the same issue.13:49
*** lsmola has joined #openstack-nova13:50
*** e0ne has joined #openstack-nova13:51
bauzasmriedem: valid point on putting the TrustedFilter experimental13:51
bauzasmriedem: that's barely the same level of expectation that I was wanting with asking for deprecating the filter13:52
bauzasmriedem: so I'm fine with your proposal :)13:52
bauzasmriedem: replying on it13:52
mriedembauzas: i think it's too late for deprecation in liberty, but i'd +2 a change that logs a warning saying it's experimental13:52
mriedemcool13:52
bauzasmriedem: yeah, I just wanted to send a bat-signal that this filter is somehow wrong, and should be rather considered as a POC13:53
bauzasrather than something good for production13:53
mriedemthat reminds me, i was wondering if you posted to the ops list asking if anyone was using it?13:53
*** xgsa1 has quit IRC13:53
bauzasmriedem: nope13:54
bauzasmriedem: but given all the Intel replies I had, I guess it's something which is quite in use13:54
*** stevemar has joined #openstack-nova13:55
mriedemwell, that's not surprising since i think they added the filter and own that external service13:55
bauzasyup13:55
mriedemjust yell at alex_xu whenever that filter angers you13:55
*** VW has joined #openstack-nova13:55
*** su_zhang has joined #openstack-nova13:55
bauzaswell, it's more a design problem as I spotted in my email13:56
*** xgsa has joined #openstack-nova13:56
*** haomaiwang has quit IRC14:01
sdaguemriedem: https://bugs.launchpad.net/nova/+bug/1491307 that's just a dup right?14:01
openstackLaunchpad bug 1491307 in OpenStack Security Advisory "secgroup rules doesn't work for instance immediately" [Undecided,Triaged]14:01
sdaguewhich you fixed14:01
*** haomaiwang has joined #openstack-nova14:01
mriedemyes....14:01
bauzassdague: FWIW, trying to reproduce https://bugs.launchpad.net/nova/+bug/149594614:01
openstackLaunchpad bug 1495946 in OpenStack Compute (nova) "Nova boot does not respect zone name if node is specified" [Undecided,New] - Assigned to Raga lahari (raga-lahari)14:01
mriedemsdague: it was fixed under bug 148473814:02
openstackbug 1484738 in OpenStack Compute (nova) kilo "keyerror when refreshing instance security groups" [High,Fix committed] https://launchpad.net/bugs/1484738 - Assigned to Matt Riedemann (mriedem)14:02
mriedemsdague: bug 1491307 is an ossa thing14:02
openstackbug 1491307 in OpenStack Security Advisory "secgroup rules doesn't work for instance immediately" [Undecided,Triaged] https://launchpad.net/bugs/149130714:02
mriedemso i think that's why they left it open?14:02
mriedemtristanC: how to we resolve that? ^14:02
*** jerrygb has quit IRC14:03
*** smatzek has joined #openstack-nova14:07
*** jerrygb has joined #openstack-nova14:07
mriedemjroll: can you triage this? https://bugs.launchpad.net/nova/+bug/149875614:07
openstackLaunchpad bug 1498756 in OpenStack Compute (nova) "Power state inconsistence after using Ironic node-set-power-state" [Undecided,New]14:07
*** takedakn has joined #openstack-nova14:08
sdaguebauzas: https://bugs.launchpad.net/nova/+bug/1488986 can you determine what state that should end up in?14:08
openstackLaunchpad bug 1488986 in OpenStack Compute (nova) "nova scheduler for race condition" [Undecided,New]14:08
*** exploreshaifali has joined #openstack-nova14:08
* bauzas looking14:08
mriedemjroll: nevermind, i know the issue14:09
*** breitz has quit IRC14:10
*** breitz has joined #openstack-nova14:10
bauzassdague: well, it's rather more a design discussion, because yes, there are some possibly some races14:12
*** obedmr has joined #openstack-nova14:12
*** mtanino has joined #openstack-nova14:13
*** abhijeetm has quit IRC14:13
bauzassdague: I'm tempted to triage it as invalid, as we're following that in https://review.openstack.org/#/c/192260/7/doc/source/scheduler_evolution.rst,cm L14114:13
bauzas(and btw. the scheduler no longer calls the compute node directly)14:13
openstackgerritClaudiu Belu proposed openstack/nova-specs: Adds spec for exposing host capabilities  https://review.openstack.org/22220014:14
bauzassdague: to be clear, I'm not seeing a clear actionable item like an usual bugfix and we already have those kind of bugs which are more improvement requests14:14
mriedembauzas: you can always mark as wishlist14:15
bauzassdague: the story could be, yes there could be some races but we do claims on the compute manager that allow retries14:15
*** fawadkhaliq has quit IRC14:15
bauzasmriedem: yup, I know14:15
mriedem'i wish there weren't races in the code'14:15
sdaguebauzas: sure, I'm fine with that14:15
sdagueOpinion / Wishlist is often good for stuff like that14:15
sdaguelike yes, there are issues, this isn't really actionable14:16
bauzasoh yeah, good point14:16
mriedemmarkus_z: for things like this https://bugs.launchpad.net/nova/+bug/149875214:17
openstackLaunchpad bug 1498752 in OpenStack Compute (nova) "VMware VM's vmdk files can't be deleted if in different datastore than VM" [Undecided,Incomplete]14:17
jrollmriedem: now I'm curious about this lifecycle event stuff14:17
mriedemi mark those as incomplete if they don't say what version they are on or provide any kind of logs14:17
mriedemjroll: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L122214:18
mriedemhttps://github.com/openstack/nova/blob/master/nova/virt/driver.py#L127514:18
mriedemand emit_event below it14:18
markus_zmriedem: Yeah, I have to be more rigid when such information is missing.14:18
mriedembaoli_: can you triage this? https://bugs.launchpad.net/nova/+bug/149868414:20
openstackLaunchpad bug 1498684 in OpenStack Compute (nova) "Fail to use SR-IOV ports when VFIO is loaded" [Undecided,New]14:20
mriedemdanpb: ^ or maybe you14:20
jrollmriedem: seems like a weird (impossible?) thing for an external service to hook into, unless I'm missing something :)14:21
mriedemjroll: it's not impossible if the libvirt and hyper-v drivers are implementing it14:21
mriedemthe libvirt driver has a thread which listens for events from libvirt itself14:21
jrollmmmm14:21
mriedemwhich are power state changes on a guest14:21
mriedemjroll: so i guess if i were to implement this in the ironic driver in nova, i'd be listening for events coming out of the ironic service for power state changes on a node14:22
baoli_mriedem: will take a look once I have a chance.14:22
jrollcould get interesting, could get weird14:22
mriedemjroll: could get freaky?14:22
jrollmriedem: yeah, I'm thinking rabbit notifications14:22
jrollheh14:22
openstackgerritdane-fichter proposed openstack/nova-specs: Nova Support of Glance Image Signing  https://review.openstack.org/18887414:22
*** browne has joined #openstack-nova14:22
*** rushil has joined #openstack-nova14:22
jrollman, I wonder if we could get rid of a bunch of polling this way14:23
*** Sree has joined #openstack-nova14:23
danpbjroll: bunch of polling for what ?14:23
danpboh i see, you mean in the ironic driver14:24
jrolldanpb: yeah, the polling for active etc14:24
*** electrocucaracha has joined #openstack-nova14:25
*** burgerk has quit IRC14:26
*** eharney has joined #openstack-nova14:26
*** tonytan4ever has joined #openstack-nova14:28
*** tpeoples has quit IRC14:32
mriedemandrewbogott: did you figure anything out with this? https://bugs.launchpad.net/nova/+bug/149819714:36
openstackLaunchpad bug 1498197 in OpenStack Compute (nova) "No longer able to delete security group rules in kilo" [Undecided,New]14:36
mriedemandrewbogott: i don't know what to do with that one14:36
*** READ10 has joined #openstack-nova14:36
andrewbogottmriedem: I haven’t looked at it much since filing.  I was hoping that whoever worked on the db migration would see my stack trace and immediately see the mistake :)14:37
*** jecarey has left #openstack-nova14:37
*** takedakn has quit IRC14:37
mriedemandrewbogott: there isn't a db migration related to this in kilo that i'm aware of14:37
*** ildikov has joined #openstack-nova14:37
*** atuvenie has quit IRC14:38
andrewbogottmaybe there should be :)14:38
andrewbogottBut, I’ll dig into the code later on today14:38
*** kiran-r has quit IRC14:39
*** salv-orlando has joined #openstack-nova14:40
*** stevemar has quit IRC14:40
bauzassdague: FYI, providing force_hosts means that you don't run filters, ie. you don't verify your AZ https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L516-L51814:41
bauzassdague: because AZFilter isn't called14:41
bauzassdague: so bug 1495946 is rather a Wishlist14:42
openstackbug 1495946 in OpenStack Compute (nova) "Nova boot does not respect zone name if node is specified" [Undecided,New] https://launchpad.net/bugs/1495946 - Assigned to Raga lahari (raga-lahari)14:42
sdaguebauzas: ok, it does seem weird that the use behavior is to let it through14:42
sdagueyeh, confirmed / wishlist, I'm fine with that14:42
bauzassdague: yup, agreed, that's something I would like to change14:42
sdaguewell, if you want to change it, I'd set it to low instead of wishlist14:42
bauzassdague: following the same patterns as for http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/check-destination-on-migrations.html14:42
sdaguewishlist is kind of "would be nice, doubt we'll ever get to it"14:43
bauzassdague: well, it would require a spec etc.14:43
*** r-daneel has joined #openstack-nova14:43
bauzassdague: and technically, it's a behavioural change, not a bug14:43
*** aarefiev has quit IRC14:43
mriedemandrewbogott: do you have the full strack trace rather than just the tail?14:44
andrewbogottmriedem: yep, it’s here:  https://phabricator.wikimedia.org/T11249214:44
*** exploreshaifali has quit IRC14:44
andrewbogottI can copy/paste if you think that’s useful14:44
mriedemomg pointers14:45
*** salv-orlando has quit IRC14:45
mriedemnope, i have the trace now, thanks14:46
*** stevemar has joined #openstack-nova14:46
*** haomaiwang has quit IRC14:47
*** aarefiev has joined #openstack-nova14:49
*** kiranr has joined #openstack-nova14:50
mriedemandrewbogott: can you dump that row from the security_group_rules table?14:50
andrewbogottsure, stay tuned14:50
mriedemit doesn't really make sense14:51
mriedemself.deleted = self.id14:51
mriedemboth of those columns are integers14:51
*** emagana has joined #openstack-nova14:53
andrewbogottmriedem: is this the row you’re after?  https://dpaste.de/xvff14:55
*** mahatic has joined #openstack-nova14:55
*** claudiub has quit IRC14:56
mriedemlooks right14:56
mriedemthose are the values you provided to the cli14:56
*** Sree has quit IRC14:56
mriedemid=2769 should be fine14:56
*** mahatic has left #openstack-nova14:56
mriedemwtf14:56
*** kiran-r has joined #openstack-nova14:56
*** Sree has joined #openstack-nova14:56
*** scheuran has quit IRC14:58
*** kiran-r has quit IRC14:58
andrewbogottisn’t the code trying to set deleted=2769?  And isn’t deleted a bool?14:58
*** tpeoples has joined #openstack-nova14:58
*** Daisy has joined #openstack-nova14:58
mriedemno, deleted is an integer14:58
mriedemb/c this is soft delete14:58
mriedemcan you dump the schema for your security_group_rules table?14:59
markus_zAFAIK if the int column "deleted" has a value != 0 it is considered as deleted. And usually the id of the row is used for that value.14:59
andrewbogotthttps://dpaste.de/F7CT14:59
mriedemmarkus_z: right: https://github.com/openstack/oslo.db/blob/master/oslo_db/sqlalchemy/models.py#L14414:59
mriedemandrewbogott: aha15:00
andrewbogottElsewhere in the table I see that value as only 0, 1, and 12715:00
mriedemyour id column is int and deleted is tinyint15:00
andrewbogottyep15:00
mriedemso that's the out of range issue15:00
andrewbogottso the old code only set it to 0,1,-115:00
mriedemold code being?15:00
mriedemjuno?15:00
andrewbogottUm...15:00
andrewbogottMy install is a continuous upgrade from at least folsom, possibly cactus.15:01
andrewbogott:)15:01
mriedembut you're currently upgrading from juno to kilo right?15:01
andrewbogottI’m sure this worked in icehouse, I suspect that it also worked in juno15:01
*** Sree has quit IRC15:01
andrewbogott(I was only running Juno for about three weeks, it’s vaguely possible that no one tried to delete a security rule during that time)15:02
mriedemi guess i'm not aware of anything changing on the type definition for the deleted column in the last few releases15:02
*** Daisy has quit IRC15:02
mriedemzzzeek: rpodolyaka: ^ heard of anything like that?15:02
mriedemif deleted was a boolean way back in the day then tinyint might make sense, but there would have had to have been a db migratoin in nova that altered that column along the way to make it an int15:04
bauzasmriedem: isn't related to the soft-delete effort ?15:04
andrewbogottIt could be that the type changed way back but the code didn’t actually make use of the larger int type until kilo15:04
mriedemandrewbogott: the soft delete thing like this has been around a long time15:04
andrewbogottmriedem: meanwhile, I’ll ask my dba to change the type of that field to an int — sounds like that’ll do it.15:04
*** Daisy has joined #openstack-nova15:04
andrewbogottis tinyint 8 bits or 16?15:04
mriedemeven back in havana https://github.com/openstack/nova/blob/havana-eol/nova/openstack/common/db/sqlalchemy/models.py#L10615:04
mriedemhttps://dev.mysql.com/doc/refman/5.0/en/numeric-types.html15:04
*** zhangjn has quit IRC15:05
andrewbogotthuh15:05
andrewbogottWe would have to be /very/ lucky to have our users only ever try to delete rules with id < 12715:06
*** Piet has quit IRC15:06
mriedemandrewbogott: 152 was the db migration that converted the type way back when https://github.com/openstack/nova/blob/havana-eol/nova/db/sqlalchemy/migrate_repo/versions/152_change_type_of_deleted_column.py15:07
sdaguemriedem: which is folsom, right?15:08
sdaguesorry, looks like grizzly15:08
sdagueandrewbogott: honestly, I bet you just got lucky15:08
andrewbogottIt’s totally possible that my table has been broken since then.  I would expect to have noticed, but...15:08
mriedemyeah grizzly15:09
sdaguedeleting sec group rules is a thing not a ton of people do15:09
*** karimb has quit IRC15:09
andrewbogottwell, users will add security group rules to solve a failed connection — deleting rules doesn’t solve a problem other than one’s one compulsiveness15:09
andrewbogottso, yeah15:09
*** Daisy has quit IRC15:09
*** karimb has joined #openstack-nova15:09
*** ZZelle has quit IRC15:09
mriedemalright, marked https://bugs.launchpad.net/nova/+bug/1498197 as invalid15:09
openstackLaunchpad bug 1498197 in OpenStack Compute (nova) "No longer able to delete security group rules in kilo" [Undecided,Invalid]15:09
*** Daisy has joined #openstack-nova15:10
openstackgerritDaniel Berrange proposed openstack/nova-specs: Libvirt driver emulator threads placement policy  https://review.openstack.org/22589315:10
*** ZZelle has joined #openstack-nova15:10
mriedemandrewbogott: i can provide the address to send consulting fees etc15:10
mriedem:)15:10
andrewbogottThis doesn’t explain why  my schema didn’t get upgraded, though, does it?15:10
markus_zmriedem: Does that mean the db migration script was missing something?15:10
*** claudiub has joined #openstack-nova15:10
mriedempossibly 152 was busted15:10
mriedem152 was rolled into 216 in havana15:11
*** Daisy has quit IRC15:11
andrewbogottIt’s definitely moot, since I’m surely the last user with this use case :)15:11
*** Daisy has joined #openstack-nova15:11
openstackgerritDaniel Berrange proposed openstack/nova-specs: Libvirt real time instances  https://review.openstack.org/13968815:11
mriedem*216 in icehouse15:11
andrewbogottBut also probably a real bug, if anyone cares15:11
*** burgerk has joined #openstack-nova15:11
mriedemwell, not really a bug we can fix15:11
mriedemsince this looks right in 216 https://github.com/openstack/nova/blob/icehouse-eol/nova/db/sqlalchemy/migrate_repo/versions/216_havana.py#L88315:12
andrewbogottah, fair enough.15:12
andrewbogottAnyway, the fix for me is straightforward, so I’m happy.  Thank you!15:12
mriedem\o/15:13
*** yamahata has joined #openstack-nova15:13
markus_zcool cool15:13
*** Apoorva has joined #openstack-nova15:13
*** penick has joined #openstack-nova15:14
openstackgerritAlexis Lee proposed openstack/nova-specs: Add list-server-invalid-option-as-error.rst  https://review.openstack.org/21975515:15
*** amotoki has quit IRC15:17
*** amotoki has joined #openstack-nova15:20
johnthetubaguymriedem: markus_z: just reaching out to ask about RC1 blockers, have you spotted anything in the bug queue?15:21
markus_zjohnthetubaguy: Nothing which seems fishy to me15:22
mriedemnothing new in the last 2 days that looks like a blocker15:22
johnthetubaguyalso, anyone else got anything major they spotted, generally15:22
*** salv-orlando has joined #openstack-nova15:22
johnthetubaguymarkus_z: mriedem: cool, thanks15:22
*** salv-orlando has quit IRC15:22
mriedemi'm just looking at 'new' bugs as of this morning though, not things that have been confirmed or triaged in the last 2 days15:23
johnthetubaguylooking close to RC1 then, looking at this: https://launchpad.net/nova/+milestone/liberty-rc115:23
*** salv-orlando has joined #openstack-nova15:23
johnthetubaguycool, we need an RC2 for translations anyways, so we can always backport things15:23
*** tsekiyama has joined #openstack-nova15:24
johnthetubaguySo if folks agree about cutting for RC1 now, please do vote on here: https://review.openstack.org/#/c/226340/115:24
openstackgerritsahid proposed openstack/nova: libvirt: add cpu schedular priority config  https://review.openstack.org/19756715:25
openstackgerritsahid proposed openstack/nova: libvirt: rework membacking config to support future features  https://review.openstack.org/19756615:25
openstackgerritsahid proposed openstack/nova: virt: add contraint to handle realtime policy  https://review.openstack.org/19756815:25
openstackgerritsahid proposed openstack/nova: libvirt: add realtime support  https://review.openstack.org/19756915:25
openstackgerritsahid proposed openstack/nova: hardware: improve parse_cpu_spec to handle exclusion range  https://review.openstack.org/22678315:25
johnthetubaguyonce https://review.openstack.org/#/c/226340/1 lands, we hit mitaka!15:25
mriedemhttps://bugs.launchpad.net/nova/+bugs?field.tag=liberty-rc-potential ?15:26
johnthetubaguyyeah, there don't seem to be any urgent ones that can't be backported15:26
*** otter768 has joined #openstack-nova15:26
*** tonytan4ever has quit IRC15:26
*** n0ano has joined #openstack-nova15:26
*** jhesketh has quit IRC15:26
johnthetubaguybut please do help double check that15:26
dansmithjohnthetubaguy: I'm +2 on the rc115:27
mriedemi'm going through them now15:27
dansmithI'll go through the tagged ones again too15:27
dansmithlast I checked, nothing stood out to me15:27
johnthetubaguydansmith: cool, thanks15:27
*** jhesketh has joined #openstack-nova15:27
*** step6829 has joined #openstack-nova15:28
dansmithjohnthetubaguy: er, we landed ndipanov's patch right?15:28
ndipanovyes15:28
ndipanovboth of them15:28
dansmithcool15:28
johnthetubaguycool, I thought so, yey!15:28
*** tonytan4ever has joined #openstack-nova15:28
dansmiththere is this one: https://review.openstack.org/#/c/221620/15:29
bauzasjohnthetubaguy: yeah, just spotted a last-minute issue with some RC1 bug, but we can cut RC115:29
dansmithbut it's not even passing unit tests15:29
dansmithand it could be backported easily15:30
bauzasdansmith: that's the one I was just referring15:30
dansmithah, cool15:30
*** claudiub has quit IRC15:30
bauzas+1 to cut RC1 now and see how to backport that one to RC215:30
johnthetubaguyoh, yeah, it failed them15:30
bauzasjohnthetubaguy: yeah, a NoneType error, classic15:30
johnthetubaguyyeah, we can backport15:30
*** Apoorva has quit IRC15:30
johnthetubaguyso turns out we need an RC2 for the translations anyways15:30
bauzasso, +1'ing for mitaka opening15:30
mriedemthis is still new and marked as liberty rc potential, but there haven't been any updates https://bugs.launchpad.net/nova/+bug/149351215:31
openstackLaunchpad bug 1493512 in OpenStack Compute (nova) "n-cpu misreports shared storage when attempting to block migrate" [Undecided,New]15:31
*** rushil has quit IRC15:31
mriedemso whatever the fix is could be backported probably, just the comments about regression were troubling15:31
*** jdurgin1 has joined #openstack-nova15:31
*** otter768 has quit IRC15:31
johnthetubaguyah, yeah, regression sounds bad15:31
*** alejandrito has joined #openstack-nova15:33
*** Piet has joined #openstack-nova15:34
mriedemndipanov: dansmith: so i guess https://bugs.launchpad.net/nova/+bug/1417667 isn't liberty-rc-potential anymore right?15:34
openstackLaunchpad bug 1417667 in OpenStack Compute (nova) "migration/evacuation/rebuild/resize of instance with dedicated cpus needs to recalculate cpus on destination" [Medium,In progress] - Assigned to Nikola Đipanov (ndipanov)15:34
*** browne has quit IRC15:35
johnthetubaguymriedem: oh, true, I did wonder about splitting the bug in half, and marking one half as merged15:35
dansmithmriedem: I thought that was a pretty invasive change, so I'm not sure it was ever really rc potential15:36
*** matrohon has quit IRC15:36
mriedemsure, i just remove the potential tag so it's not cluterring up https://bugs.launchpad.net/nova/+bugs?field.tag=liberty-rc-potential15:36
mriedemwhat are we doing with https://review.openstack.org/#/c/189544/ ?15:37
mriedembackport the schema check once it's fixed?15:37
dansmithwe should probably mark things that are rc-potential as backport-potential unless we know they're not right?15:37
mriedemyeah probably15:37
dansmithmriedem: it's -W for a couple weeks, so yeah, backport if and when appropriate15:38
*** tonytan4ever has quit IRC15:38
ndipanovmriedem, well15:38
ndipanovwe still have the whole live migration thing15:38
mriedemndipanov: yeah but we arne't holding up rc for it right?15:38
ndipanovwe got the patches we wanted for rc15:39
mriedemi tend to think that the liberty-rc-potential list should be empty when we do rc115:39
dansmithmriedem: except for the scheduler thing we know we're shooting for rc215:39
dansmithwell, unless we target it or whatever15:39
mriedemdansmith: that metrics thing?15:39
BobBallmriedem: You just removed rc-potential from https://bugs.launchpad.net/nova/+bug/1496557 - Was that the bug you intended to remove?  Don't see it discussed here, so just checking? :)15:39
openstackLaunchpad bug 1496557 in OpenStack Compute (nova) "xenapi boot from volume is broken after move to ImageMeta object if volume passed in" [High,Fix committed] - Assigned to Andrew Laski (alaski)15:39
dansmithmriedem: yeah15:39
mriedemBobBall: b/c it's fixed15:40
mriedemBobBall: it's not rc potential if it's fixed15:40
BobBallAh fair enough15:40
* BobBall is clearly not understanding the branching strategy15:40
mriedemmarkus_z: i'm marking https://bugs.launchpad.net/nova/+bug/1455252 as liberty-backport-potential - since it was reported against juno i don't think it should hold up the liberty rc115:40
openstackLaunchpad bug 1455252 in OpenStack Compute (nova) "enabling serial console breaks live migration" [High,In progress] - Assigned to Alexander Schmidt (alexs-h)15:40
markus_z+1 to what dan said about using "backport-potential" and removing "rc-potential".15:41
openstackgerritDaniel Berrange proposed openstack/nova-specs: Virtual guest device role tagging  https://review.openstack.org/19566215:41
markus_zmriedem: yeah, john did also discover a possible issue with rolling upgrades in the implementation.15:41
johnthetubaguywe have another tag for that, let me go dig that up15:41
*** browne has joined #openstack-nova15:41
markus_z+1 to what mriedem said about having an empty list for "rc-potential".15:42
johnthetubaguyoh, I think it is liberty-backport-potential15:42
mriedemyeah, already been marking hose15:42
markus_zjohnthetubaguy: Yes, that one sounds right ^15:42
mriedem*those15:42
*** cristicalin has quit IRC15:42
johnthetubaguyso, I think normally we use rc-potential, becuase its stuff that when fixed, might trigger an RC release15:42
mriedemit'll be super fun when we don't have 6 month release cycles anymore15:42
mriedemrc potential is fine after rc115:43
mriedemto trigger rc215:43
johnthetubaguyyeah15:43
mriedembut rc potential o na fixed bug before rc1 is not necessary15:43
*** mdrabe has quit IRC15:43
johnthetubaguyoh, right, totally15:43
johnthetubaguyall the bugs fixes before RC1 are just normal bug fixes, nothing extra required15:44
johnthetubaguyanyways, if people are happy with cutting RC1 now, please to vote on this patch: https://review.openstack.org/#/c/226340/15:45
johnthetubaguywell, vote on there if you are unhappy is important too I guess15:45
*** rushil has joined #openstack-nova15:45
mriedemthere, -115:45
mriedem:)15:45
mriedemshould be done with the rc potentials in a few minutes15:46
johnthetubaguycools15:46
*** penick has quit IRC15:46
openstackgerritChristian Berendt proposed openstack/nova: Disable the in tree EC2 API by default  https://review.openstack.org/22685615:46
*** e0ne has quit IRC15:47
*** rushil has quit IRC15:47
*** e0ne has joined #openstack-nova15:48
zzzeekmriedem: “that” == deleted used to be a tinyint ?  havent seen that..15:49
mriedemzzzeek: pre-grizzly15:50
zzzeekmriedem: way before my time15:50
*** su_zhang has quit IRC15:50
*** su_zhang has joined #openstack-nova15:51
*** pixelb has quit IRC15:52
*** pixelb has joined #openstack-nova15:52
dansmithmriedem: that live migration bug has no patch right?15:52
dansmithhttps://bugs.launchpad.net/nova/+bug/127621415:52
openstackLaunchpad bug 1276214 in OpenStack Compute (nova) "Live migration failure in API doesn't revert task_state to None" [Medium,In progress] - Assigned to Bartosz Fic (bartosz-fic)15:52
dansmithand I expect it's not a regression15:52
mriedemthere is a patch here https://review.openstack.org/#/c/168916/15:53
mriedemdansmith: the bug was reported over 18 months ago, so no, not a regression15:53
dansmithah, I missed that15:54
dansmithyeah15:54
dansmithso not worth holding rc over15:54
mriedemnope, updated the bug15:54
mriedemjust looking at https://bugs.launchpad.net/nova/+bug/148364515:54
openstackLaunchpad bug 1483645 in OpenStack Compute (nova) "Ephemeral disk size in volume can be bypassed when booting instance" [High,In progress] - Assigned to jichenjc (jichenjc)15:54
dansmithokay15:54
*** sayali has quit IRC15:55
*** su_zhang has quit IRC15:56
mriedemthere is a patch https://review.openstack.org/#/c/213762/ it has -1 from ndipanov, and there is a workaround i think, so not an rc blocker imo15:56
*** ijw has quit IRC15:56
*** dulek_ has joined #openstack-nova15:57
*** dulek_ has quit IRC15:57
*** nkrinner has quit IRC15:57
mriedemdansmith: johnthetubaguy: alright, +2 on https://review.openstack.org/#/c/226340/15:58
mriedemshould i just approve?15:58
dansmithI'd let mr. tuba guy do it, personally :)15:58
mriedemhe proposed it15:58
johnthetubaguyI don't mind doing it15:58
dansmithyeah, but two +2s and being the PTL, I think it's safe :)15:58
*** mdrabe has joined #openstack-nova15:58
mriedemoutrageous15:59
mriedemthere is only one thing more outrageous15:59
mriedemhttps://www.youtube.com/watch?v=YSLhrIuAZak15:59
markus_z:D seriously, you have a list on your desktop for each occasion, right?16:00
johnthetubaguymarkus_z: I was just thinking the same thing16:00
*** alex_klimov has quit IRC16:02
dansmithheh16:03
markus_zJust so that I get it, we tag the bugs again with "liberty-rc-potential" when we want to consider them for rc2?16:03
mriedemmarkus_z: yeah16:03
*** Apoorva has joined #openstack-nova16:03
*** jlanoux has quit IRC16:06
markus_zOK, get it.16:06
mriedembtw, why do we have rc2 for translations now? b/c of the soft string freeze they only do translations after rc1 now?16:06
*** leeantho has joined #openstack-nova16:07
*** sayali has joined #openstack-nova16:08
*** karimb has quit IRC16:10
*** atuvenie has joined #openstack-nova16:11
Daisystable/liberty is created ?16:11
mriedemDaisy: not yet16:18
mriedemDaisy: soon https://review.openstack.org/#/c/226340/16:18
*** dane-fichter_ has joined #openstack-nova16:19
*** jdurgin1 has quit IRC16:19
*** romainh has left #openstack-nova16:20
*** whenry_ has joined #openstack-nova16:21
*** jwcroppe has quit IRC16:25
Daisymriedem: is this the last patch ? Will you merge other patches ? I wonder if I could create the branch for Liberty translations now. If you don't have string changes, I think I could.16:27
*** sahid has quit IRC16:27
DaisyI mean, create a branch for Liberty translation in the translation website.16:27
mriedemDaisy: there is one bug that still needs fixing which was targeted for rc216:27
mriedemi have no idea how the branching thing works in zanata16:28
mriedemthat's a johnthetubaguy or ttx question16:28
*** yassine__ has quit IRC16:28
*** tpeoples has quit IRC16:29
DaisyWill you merge patches for rc1 except the one you mentioned ? Will you have patches with string changes for rc1 ?16:29
*** nic has joined #openstack-nova16:30
DaisyIf no string changes for rc1 now, only waiting for that patch merge, I think I could make a liberty branch in translation website. johnthetubaguy, hope you could see that.16:30
clarkbyes I think you can create it then hopefully patches to keep in sync land today16:30
clarkbthey are very close16:30
*** kiran-r has joined #openstack-nova16:31
*** nic1 has joined #openstack-nova16:31
Daisyok. then I will create a branch named as "stable-liberty"16:31
*** armax has joined #openstack-nova16:32
johnthetubaguyDaisy: yes, our open mitaka patch should merge soon16:33
markus_zHow much release candidate do we usually have until the final release? Just to get a feeling.16:34
*** nic has quit IRC16:34
markus_z*candidates16:35
johnthetubaguyclarkb: once we have stable/liberty for Nova, does the bot start merging translations over in the stable branch, or is that a backport task?16:35
johnthetubaguyI remember running a script manually in the past16:35
clarkbit will start once my changes to do it merge16:35
johnthetubaguyclarkb: sweet16:36
clarkbI thibk we are waiting on a fix from me and branches to be cut16:36
*** tjones has joined #openstack-nova16:37
clarkbI plan to push that fix in a bit currently busy at with family16:37
bauzasmarkus_z: that really depends on the stability, no rule of thumb16:38
johnthetubaguymarkus_z: two or three is normal-ish16:39
bauzasmarkus_z: if you want to get a taste of what was there for the last releases https://launchpad.net/nova/+series16:39
openstackgerritSylvain Bauza proposed openstack/nova: Fix MetricWeigher to use MonitorMetricList  https://review.openstack.org/22162016:39
johnthetubaguymarkus_z: I plan on cutting an RC in about 10 days for translations, it would be something very scary to make me cut an RC2 before then16:39
markus_zbauzas: Yessss, graphs, thanks! :)16:39
bauzasmarkus_z: but you won't get the list of bugfixes per RC since all of them are modified to be in the last release GA16:39
*** diogogmt has joined #openstack-nova16:40
bauzasjohnthetubaguy: so I made a last spinoff of https://review.openstack.org/22162016:40
bauzasjohnthetubaguy: should fix the last py27 failure16:40
bauzasjohnthetubaguy: for some reason, the CachingScheduler testcase was not catching that the metrics field was set to none16:40
markus_zjohnthetubaguy: Yes, makes sense to include the translations in RC2.16:42
markus_zbauzas: OK, I don't think that I would need such a list. "Liberty" is still my first "real" cycle and I wanted to get a better feeling how things work.16:43
mriedemsmatzek: are you working on a fix for this? https://bugs.launchpad.net/nova/+bug/148811116:44
openstackLaunchpad bug 1488111 in OpenStack Compute (nova) "Boot from volumes that fail in initialize_connection are not rescheduled" [Low,Triaged] - Assigned to Samuel Matzek (smatzek)16:44
smatzekmriedem:  I plan to but have not yet started.16:45
*** eglynn has quit IRC16:47
*** derekh has quit IRC16:49
markus_zmriedem: You are not happy with the additional tags I use on the bugs?16:49
mriedemmarkus_z: like what?16:49
markus_zhttps://bugs.launchpad.net/nova/+bug/148811116:49
openstackLaunchpad bug 1488111 in OpenStack Compute (nova) "Boot from volumes that fail in initialize_connection are not rescheduled" [Low,Triaged] - Assigned to Samuel Matzek (smatzek)16:49
*** rdopiera has joined #openstack-nova16:49
mriedemmarkus_z: i removed the spawn tag yeah16:50
mriedemi'm not sure what that is meant to convey16:50
mriedemit seems too granular16:50
mriedemlike, a spawn could fail for a ton of reasons16:50
mriedemso i wouldn't expect there to be a sub-team that looks at spawn failures16:50
*** salv-orlando has quit IRC16:51
*** salv-orlando has joined #openstack-nova16:52
markus_zIt wasn't for the subteams. I wanted to start an analysis of the bugs, to see which features are impaired the most, over all layers.16:52
mriedemstill16:53
mriedemspawn as a tag is way too generic for any useful analysis imo16:53
mriedemi can see doing that for live-migrate or resize, things like that being useful16:53
*** aix has quit IRC16:53
*** MarcF has quit IRC16:53
*** unicell has joined #openstack-nova16:54
johnthetubaguyI do like the idea of identifying groups of bugs, like "live-migrate" and "quotas"16:54
markus_zright, alone it does nothing. My goal was to measure the co-existence with other tags on a bug to see how the spawn in e.g. libvirt is doing.16:54
*** jwcroppe has joined #openstack-nova16:55
*** tonytan4ever has joined #openstack-nova16:55
*** Daisy has quit IRC16:55
*** tpeoples has joined #openstack-nova16:56
*** Daisy has joined #openstack-nova16:56
markus_zmriedem: I'll prepare something in the next weeks to show what my end-goal here is. Then we can discuss if we would benefit from that. deal?16:56
*** tonytan_brb has joined #openstack-nova16:56
johnthetubaguymarkus_z: yeah, not sure spawn is a good place to start, I would look at something like "ceph" or "bdm" maybe?16:57
*** exploreshaifali has joined #openstack-nova16:57
mriedemwell we have a tag 'volumes' that would cover bdms16:57
mriedemanyway, bikeshedding16:57
*** mpaolino has quit IRC16:57
mriedemi wasn't aware these were part of a larger effort16:57
markus_zI didn't communicate that good enough.16:57
johnthetubaguymarkus_z: loving the idea though, maybe a quick ML posting about "hey I am doing this bug thing"?16:58
*** rushil has joined #openstack-nova16:58
johnthetubaguyI know folks who have done lots of triage in the past might be able to give some pointers (thats not really me, I have to be honest!)16:58
markus_zjohnthetubaguy: Yeah, I'll do that.16:58
*** emagana has quit IRC16:59
*** armax has quit IRC16:59
*** salv-orl_ has joined #openstack-nova16:59
*** tonytan4ever has quit IRC16:59
*** penick has joined #openstack-nova17:00
*** kiran-r has quit IRC17:00
*** Daisy has quit IRC17:00
*** lucasagomes is now known as lucas-dinner17:00
markus_zjohnthetubaguy: Sure, I welcome every input I can get.17:00
*** tonytan_brb has quit IRC17:01
*** salv-orlando has quit IRC17:01
*** salv-orlando has joined #openstack-nova17:02
*** su_zhang has joined #openstack-nova17:03
*** salv-orl_ has quit IRC17:04
*** nic1 has quit IRC17:06
johnthetubaguymarkus_z: I guess I am thinking sdague might have some good ideas around that17:06
*** electrocucaracha has quit IRC17:06
markus_zjohnthetubaguy: He made a very good post on the ML on that last September: http://lists.openstack.org/pipermail/openstack-dev/2014-September/046517.html17:07
markus_zjohnthetubaguy: If it is that what you mean.17:08
johnthetubaguyah, yeah, that must be what I was remembering!17:08
markus_zI have read it but need to find time to discuss my implementation ideas for that. I have a brain dump text file which contains some things but that is not consumable by others...17:09
johnthetubaguytotally makes sense17:09
*** achanda has joined #openstack-nova17:10
*** kiranr has quit IRC17:12
*** VW has quit IRC17:12
mriedemmarkus_z: here is a good example of tags i think https://bugs.launchpad.net/nova/+bug/149623517:13
openstackLaunchpad bug 1496235 in OpenStack Compute (nova) "Boot from volume faild with availability_zone option, in case of cinder do not have availability_zone" [Undecided,New] - Assigned to Ikuo Kumagai (i-kumagai)17:13
mriedemwith availability-zones17:13
mriedemmarkus_z: since AZs are the special wrinkle there17:13
*** VW has joined #openstack-nova17:13
*** tpeoples has quit IRC17:14
*** ociuhandu has quit IRC17:15
markus_zmriedem: agreed17:15
*** ijw has joined #openstack-nova17:15
*** e0ne has quit IRC17:16
*** danpb has quit IRC17:16
*** VW has quit IRC17:17
*** nic has joined #openstack-nova17:17
openstackgerritJohn Garbutt proposed openstack/nova-specs: Make consolidate-console-api match impl  https://review.openstack.org/19903317:19
*** ijw has quit IRC17:20
*** su_zhang_ has joined #openstack-nova17:21
*** su_zhang has quit IRC17:21
*** raildo is now known as raildo-afk17:25
dansmithls17:25
dansmithoops17:25
*** smatzek has quit IRC17:26
*** ijw has joined #openstack-nova17:26
*** smatzek has joined #openstack-nova17:26
*** ankit has joined #openstack-nova17:27
melwittalaski: did you see my question on https://review.openstack.org/#/c/176925/ ? just wanted to confirm whether the changes to the fake instances get function are still needed17:27
*** otter768 has joined #openstack-nova17:27
ankithi all, Can someone please review https://review.openstack.org/#/c/185549/17:28
*** yamahata has quit IRC17:29
*** otter768 has quit IRC17:31
*** ociuhandu has joined #openstack-nova17:32
mriedembauzas: this is a fun AZ one https://bugs.launchpad.net/nova/+bug/149623517:33
openstackLaunchpad bug 1496235 in OpenStack Compute (nova) "Boot from volume faild with availability_zone option, in case of cinder do not have availability_zone" [Undecided,New] - Assigned to Ikuo Kumagai (i-kumagai)17:33
mriedembauzas: it's basically invalid17:33
*** rajesht has joined #openstack-nova17:33
mriedemif you try to boot from volume with source=image into a nova az and that same az isn't in cinder, the volume create fails17:34
alaskimelwitt: saw your question.  I don't remember that code well enough to answer, let me test real quick17:35
*** su_zhang_ has quit IRC17:35
*** su_zhang has joined #openstack-nova17:35
*** atuvenie has quit IRC17:39
*** hitalia has joined #openstack-nova17:41
*** raildo-afk is now known as raildo17:41
rajeshthi cores, need one more +2 on code cleanup https://review.openstack.org/#/c/220142/17:42
rajeshthttps://review.openstack.org/#/c/221271/17:42
*** vilobhmm11 has joined #openstack-nova17:43
*** sudipto has joined #openstack-nova17:44
*** emagana has joined #openstack-nova17:45
*** rajesht has quit IRC17:45
*** markus_z has quit IRC17:45
*** mriedem has quit IRC17:46
*** garyk has joined #openstack-nova17:47
alaskimelwitt: removing the marker edition cause the test runner to use all memory and crash something on my system forcing me to reboot.  So I think it's still necessary :)17:48
alaskis/edition/addition/17:48
*** rlrossit has quit IRC17:48
melwittalaski: oh, sorry :(17:49
*** mriedem has joined #openstack-nova17:49
*** ankit has quit IRC17:50
alaskimelwitt: heh, no worries.  it's my fault for tabbing away while the test ran and then I couldn't get back to it to kill it17:51
*** atuvenie has joined #openstack-nova17:51
alaskimelwitt: I can fix the list comprehension if you'd like17:52
*** VW has joined #openstack-nova17:52
*** rlrossit has joined #openstack-nova17:53
*** yamahata has joined #openstack-nova17:54
melwittalaski: nah, that's okay. just wrote it in the case of a respin. I didn't get how the existing test (not your new one) ends up with marker being not None (looking again)17:54
*** kiran-r has joined #openstack-nova17:54
*** mriedem1 has joined #openstack-nova17:54
*** mriedem has quit IRC17:54
mriedem1doffm: https://blueprints.launchpad.net/nova/+spec/refresh-quotas-usage17:55
*** sudipto has quit IRC17:55
*** e0ne has joined #openstack-nova17:56
alaskimelwitt: the first time _get_paginated_instances is run it will pass marker=None to that method.  After that marker is set to the uuid of the last instance returned17:57
melwittalaski: I guess with pagination, there's no chance you could get all of the instances in one call? it has to be two minimum? I was thinking of when you get all the first time17:57
*** krtaylor has quit IRC17:58
alaskimelwitt: you could get them all in one call, but we don't know if that's the case or not17:58
melwittalaski: meaning... there's nothing that can tell you that you got them all, the first time17:58
melwittokay17:58
*** diogogmt has quit IRC17:58
alaskimelwitt: nothing I can think of.  If the results are less than limit we would know, but if we get exactly limit results we won't know17:59
melwittalaski: right..17:59
*** diogogmt has joined #openstack-nova17:59
dansmithjohnthetubaguy: presume you want the /topic changed once the mitaka patch lands?18:00
*** electrocucaracha has joined #openstack-nova18:01
*** hemna is now known as hemnafk18:01
*** andrewbogott has left #openstack-nova18:05
*** e0ne has quit IRC18:06
doffmmriedem1: ta.18:07
*** jwcroppe has quit IRC18:08
*** ijuwang has quit IRC18:08
*** mriedem1 has quit IRC18:10
*** READ10 has quit IRC18:10
*** su_zhang has quit IRC18:11
*** su_zhang has joined #openstack-nova18:11
*** krtaylor has joined #openstack-nova18:11
ndipanovdansmith, got a nice little fix for you to -2^H^H review18:11
dansmithwell, now I'm in a good mood.. hit me18:12
*** e0ne has joined #openstack-nova18:13
*** annegentle has joined #openstack-nova18:13
*** mriedem has joined #openstack-nova18:13
*** zhangjn has joined #openstack-nova18:14
*** rushil has quit IRC18:15
*** ctrath has quit IRC18:18
*** zhangjn has quit IRC18:19
*** whenry_ has quit IRC18:20
*** gszasz has quit IRC18:20
*** ctrath has joined #openstack-nova18:21
bauzasmriedem: okay, I'll look at the bug18:22
*** rushil has joined #openstack-nova18:22
*** rushil has quit IRC18:23
*** achanda has quit IRC18:26
mriedembauzas: i posted to the ML on it also18:26
mriedemnot a priority though18:26
openstackgerritNikola Dipanov proposed openstack/nova: rebuild: Apply migration context before calling the driver  https://review.openstack.org/22692318:27
ndipanovdansmith, ^18:27
*** dims__ has joined #openstack-nova18:28
*** vishwanathj has joined #openstack-nova18:29
openstackgerritMatt Riedemann proposed openstack/nova-specs: Re-propose spec for refresh-quotas-usage  https://review.openstack.org/22692518:30
*** achanda has joined #openstack-nova18:30
bauzasmriedem: ack18:30
*** jichen has joined #openstack-nova18:31
mriedemrlrossit: something to work on if you're looking for work https://review.openstack.org/#/c/226925/18:31
*** dims_ has quit IRC18:31
openstackgerritMark Doffman proposed openstack/nova: DO NOT MERGE: Extra logging for lxc teardown fail.  https://review.openstack.org/22312518:32
bauzasmriedem: so, MHO is that I wonder why Nova and Cinder would need to have the same AZs18:32
kashyapmake -j818:32
kashyapUgh, sorry18:32
*** aix has joined #openstack-nova18:33
*** kiran-r has quit IRC18:33
ndipanovbah pep8 !18:33
* bauzas needs to dig into cinder code...18:34
*** rushil has joined #openstack-nova18:35
openstackgerritNikola Dipanov proposed openstack/nova: rebuild: Apply migration context before calling the driver  https://review.openstack.org/22692318:35
*** amotoki has quit IRC18:35
kashyapmriedem: Hi, did you get to the end of -- https://review.openstack.org/#/c/218355/18:37
mriedemkashyap: nope18:37
mriedemi've lost hope on that one18:37
mriedemat least until kilo-eol happens18:37
kashyapSorry, I couldn't spend time to understand the issue more, I'm buried with a few things for a couple of weeks.18:38
mriedemi also couldn't reproduce it manually using novaclient in stable/kilo devstack18:38
mriedemthat's fine18:38
mriedemat this point it's passing for liberty, even though i didn't change anything18:38
mriedembut the initial volume attach in juno and kilo doesn't result in a partition showing up on the guest18:38
mriedemfor whatever magical crazy reason18:38
*** diogogmt has quit IRC18:39
vilobhmm11alaski : ping18:40
alaskivilobhmm11: hi18:40
vilobhmm11regarding https://review.openstack.org/#/c/222744/18:41
kashyapmriedem: Okay.  I've subscribed to it, and will keep an eye on it.18:41
vilobhmm11alaski : was thinking of modeling quota by az and quota by az (and flavor) in the way flavor classes will be modelled…18:43
*** redbeard has quit IRC18:43
vilobhmm11so how will the flavor classes have the information about the AZ ?18:44
vilobhmm11or is there a plan to introduce something like az groups or something ?18:44
alaskivilobhmm11: flavors/flavor classes don't have knowledge of AZs18:45
alaskiI'm not sure why you need quotas on AZs though, rather than on flavors.  If you set it up so that only flavor X (or flavor class X) can build in AZs 1 & 2 then you can apply quotas to the flavors rather than the AZ18:46
mriedemdansmith: alaski: what do we do about moving liberty specs from approved to implemented in the nova-specs repo? just needs someone to go and do that?18:46
vilobhmm11alaski : that was my gut feeling18:47
alaskimriedem: that's my recollection of what happened last time18:47
mriedemalaski: ok18:47
bauzaswhat happens with AZs suddently with all these bugs and discussions coming in ? o_O18:48
mriedembauzas: people love AZs now18:48
PietThe OpenStack UX project team would like to invite you to participate in a survey to help drive product direction for the Neutron project.  The survey should take no more than fifteen minutes to complete.18:48
PietWe are specifically looking for folks with the following roles; Architect, Cloud Operator, System Administrator, Network Architect / Engineer / Administrator, Developer, or IT Manager / Director.  In addition, they should be familiar with either Nova Networks and/or Neutron.18:48
PietYou will also be entered in a raffle for one of two $100 US Amazon gift cards at the end of the survey.   As always, the results from the survey will be shared with the OpenStack community.18:48
PietPlease click on the following link to begin the survey.18:48
Piethttps://www.surveymonkey.com/r/osnetworking18:48
*** e0ne has quit IRC18:49
alaskiwe should record each cycles "hip new thing" somewhere.  first NFV then containers now AZs18:50
*** su_zhang has quit IRC18:51
*** penick has quit IRC18:51
bauzasalaski: like Spiderman, let's do a reboot for AZs ?18:52
bauzasthat's nolanesque18:53
bauzasuh, nolanish rather18:53
mriedemomg we got volume bugs under 100 https://bugs.launchpad.net/nova/+bugs?field.tag=volumes&orderby=status&start=018:55
*** lpetrut has quit IRC18:57
*** rdopiera has quit IRC18:57
*** diogogmt has joined #openstack-nova18:58
openstackgerritAndrew Laski proposed openstack/nova: Neutron: update cells when saving info_cache  https://review.openstack.org/22694119:00
*** diogogmt_ has joined #openstack-nova19:02
*** alex_klimov has joined #openstack-nova19:03
*** diogogmt has quit IRC19:04
*** diogogmt_ is now known as diogogmt19:04
*** penick has joined #openstack-nova19:06
*** Marga_ has quit IRC19:06
*** Marga_ has joined #openstack-nova19:08
*** Piet has quit IRC19:08
*** e0ne has joined #openstack-nova19:12
*** Marga_ has quit IRC19:13
*** rushil has quit IRC19:13
*** hitalia has quit IRC19:14
*** exploreshaifali has quit IRC19:15
*** rushil has joined #openstack-nova19:18
*** hodos has joined #openstack-nova19:19
dansmithndipanov: so I know the answer is probably "no", but would it be safer to stash/restore locally the current numa_topology instead of restoring the old one from the context?19:19
*** garyk has quit IRC19:20
dansmithndipanov: like, I think that the assumption your context manager implies is that it will set the new one temporarily and then revert19:20
ndipanovdansmith, well yes19:20
dansmithndipanov: but if the caller has instance.numa_topology set to something crazy, this will not restore that when the context manager exits19:20
ndipanovhmmm19:20
ndipanovwhat do you mean?19:20
dansmithit would be trivial to just store that in local scope and restore right?19:20
dansmithndipanov: let's say I have set instance.numa_topology = None just before making this call with the mutated context manager19:21
dansmithwhen it exits, it will set instance.numa_topology = migration_context.old_, not what I had it before19:21
dansmithso I could go from something -> new -> old19:22
*** tonytan4ever has joined #openstack-nova19:22
dansmithwhen I think the caller has the expectation that you will *restore* it19:22
ndipanovdansmith, yar!19:23
ndipanov-1 it and I'll fix it tomorrow19:23
dansmithokay19:23
dansmithhttp://pastebin.com/DH4bnqhz19:23
dansmithis what I mean19:23
ndipanovyes19:24
ndipanovI will just copy that :)19:24
*** dims__ has quit IRC19:24
*** e0ne has quit IRC19:24
*** dims_ has joined #openstack-nova19:25
*** garyk has joined #openstack-nova19:25
*** mriedem has quit IRC19:26
dansmithcool19:26
*** dane-fichter_ has quit IRC19:26
*** otter768 has joined #openstack-nova19:28
*** whenry_ has joined #openstack-nova19:28
*** salv-orlando has quit IRC19:29
openstackgerritMark Doffman proposed openstack/nova: libvirt: Don't attempt to convert initrd images.  https://review.openstack.org/22528919:29
*** salv-orlando has joined #openstack-nova19:29
*** penick has quit IRC19:31
*** annegentle has quit IRC19:32
*** mriedem has joined #openstack-nova19:32
*** annegentle has joined #openstack-nova19:32
*** hemnafk is now known as hemna19:33
*** otter768 has quit IRC19:33
*** derekh has joined #openstack-nova19:36
*** garyk has quit IRC19:36
*** Marga_ has joined #openstack-nova19:37
*** penick has joined #openstack-nova19:41
*** achanda has quit IRC19:41
*** garyk has joined #openstack-nova19:41
*** artom has quit IRC19:42
*** mriedem has quit IRC19:43
*** mriedem has joined #openstack-nova19:44
*** ccarmack has quit IRC19:46
*** ccarmack has joined #openstack-nova19:47
*** Piet has joined #openstack-nova19:48
*** su_zhang has joined #openstack-nova19:48
*** Marga_ has quit IRC19:50
*** matrohon has joined #openstack-nova19:51
*** rlrossit1 has joined #openstack-nova19:52
*** rlrossit has quit IRC19:52
*** mwagner_lap has quit IRC19:54
*** VW has quit IRC19:55
*** VW has joined #openstack-nova19:55
*** rlrossit1 has quit IRC19:57
*** ctrath has quit IRC19:57
*** ccarmack has quit IRC19:57
openstackgerritSamuel Cassiba proposed openstack/nova-specs: Add support for FreeBSD as a compute node  https://review.openstack.org/22499819:59
*** VW has quit IRC19:59
*** hodos|2 has joined #openstack-nova20:00
*** ctrath has joined #openstack-nova20:00
*** rlrossit has joined #openstack-nova20:01
*** ccarmack has joined #openstack-nova20:01
*** salv-orlando has quit IRC20:01
*** fawadkhaliq has joined #openstack-nova20:02
*** hodos has quit IRC20:02
*** fawadkhaliq has quit IRC20:05
*** garyk has quit IRC20:06
openstackgerritMark Doffman proposed openstack/nova: libvirt: Don't attempt to convert initrd images.  https://review.openstack.org/22528920:06
*** jwcroppe has joined #openstack-nova20:08
*** stevemar has quit IRC20:11
*** stevemar has joined #openstack-nova20:11
*** angdraug has joined #openstack-nova20:13
*** stevemar has quit IRC20:13
*** salv-orlando has joined #openstack-nova20:14
*** derekh has quit IRC20:16
*** rushil has quit IRC20:17
*** whenry_ has quit IRC20:18
*** smatzek has quit IRC20:19
*** hitalia has joined #openstack-nova20:21
*** VW has joined #openstack-nova20:24
openstackgerritMatt Riedemann proposed openstack/nova: Deprecate cinder.cross_az_attach option  https://review.openstack.org/22697720:25
mriedemjgriffith: bauzas: alaski: scottda: ^20:25
bauzasmriedem: \o/20:25
bauzasjaypipes: ^ too20:25
scottdamriedem: I am trying to figure out with hemna how to introduce returning VolumeIsBusy exception to Nova....20:26
bauzasjaypipes: long story here, we tried to understand why AZs neded to be provided to Cinder20:26
scottdaWe're thinking with api-microversions, a new cinderclient would make a new call that gets the exception...20:26
scottdabut I see that Nova does not upper-pin the cinderclient.20:26
hemnahaven't exactly wrapped my head around dealing with new cinderclient and cinder that has support for new microversion, with an old nova, say L.20:26
mriedemscottda: novaclient allows the caller to specify the microversion20:27
mriedemand defaults to latest20:27
mriedemscottda: so nova would just have a minimum required cinderclient, and then opt into microversions as needed20:27
openstackgerritSylvain Bauza proposed openstack/nova: Set TrustedFilter as experimental  https://review.openstack.org/19459220:27
hemnaso the caller of cinderclient is forced to know the microversion # and ask for it.20:28
scottdaSo, the novaclient microversion would take care of insuring that we used the minimum required cinderclient? sounds ok20:28
hemnanot cinderclient itself ?20:28
mriedemsorry, on a call20:29
scottdaok, we can figure this out later20:29
dansmithhemna: I think the point is, you can tell novaclient that you're making a call foo() and that the microversion should be X, if you want to depend on certain semantics that you only get running >=X20:30
hemnadansmith, so in our case, when nova calls cinderclient to take some action, it passes in a newer microversion that it knows about?20:31
hemnathen I presume cinderclient has to tell the caller it got X version20:31
dansmithhemna: I wasn't following the full discussion, so I'm not sure exactly what you were shooting for20:31
hemnanp20:31
dansmithhemna: are you trying to determine if you're running with a new-enough nova?20:31
bauzasmriedem: jaypipes: johnthetubaguy: so I made the TrustedFilter experimental https://review.openstack.org/22697720:31
scottdadansmith: no20:31
hemnawe are looking to add microversioning support to cinder20:32
hemnaand hence cinderclient20:32
scottdaWe are going to implement microversions in Cinder20:32
dansmithoh, heh20:32
hemnaand one of the new microversions will return a new exception that nova doesn't know about currently20:32
scottdaThis will allow us to change the cinder <-> nova API, we hope20:32
scottdain a safe way20:32
dansmithokay, so the way nova handles this,20:32
hemnaso we are trying to figure out how to deal with an L nova talking to an M cinder and not break :)20:32
dansmithis that if the caller didn't tell the client that it knows about the microversion where that new exception was added, the client behaves like it did originally20:33
scottdaok, that is good. And easy for a caller of cinder api going directly to cinder api...20:33
dansmithlike, we assume "no version" means "the behavior before the versioning was added"20:33
scottdaBut what about, say 'nova volume-attach', which uses cinderclient, which then calls cinder api?20:33
hemnaI guess I need to look at how the caller of novaclient asks for a version20:33
dansmithscottda: so, to clarify there are two cases: the client as a python client and the client as a CLI20:34
hemnadansmith, I think we are concerned with the python client20:34
dansmithscottda: I believe that as a python client, it acts as if "no version" means "original version" and if operating as a CLI, then "no version" means "the latest version"20:34
dansmithright20:35
dansmithso in that case,20:35
*** tjones1 has joined #openstack-nova20:35
dansmithL nova would be not specifying the version,20:35
mriedemhemna: e.g. https://review.openstack.org/#/c/190346/20:35
dansmithwhich means cinderclient needs to behave like it does today20:35
mriedemyeah, what dan said20:36
mriedemthen eventually nova starts asking for a certain cinder microversion20:36
hemnaok20:36
dansmithyeah, once we're new enough to pass the version, then cinderclient will start exposing the new behavior20:36
hemnaI'm starting to poke around in the novaclient code to see where there is an api for asking for a specific version of a call20:36
hemnagah...have to run to a meeting.....bbl20:37
*** hemna is now known as hemnafk20:37
*** tjones1 has left #openstack-nova20:37
scottdaOK, I'll have to write some code and go from there. Thanks.20:40
*** Apoorva has quit IRC20:41
*** achanda has joined #openstack-nova20:42
*** step6829 has quit IRC20:43
*** su_zhang_ has joined #openstack-nova20:45
*** su_zhang has quit IRC20:45
*** hemnafk is now known as hemna20:45
*** su_zhang_ has quit IRC20:45
*** IanGovett has quit IRC20:45
*** salv-orlando has quit IRC20:45
*** su_zhang has joined #openstack-nova20:45
*** lucas-dinner has quit IRC20:46
*** achanda has quit IRC20:47
hemnahttps://github.com/openstack/python-novaclient/blob/master/novaclient/api_versions.py20:47
hemnaso I guess that's the magic sauce then20:47
mriedemhemna: yeah, see how https://review.openstack.org/#/c/190346/ does that20:48
mriedemthat change is pretty straight forward20:48
hemnayah20:48
hemnathe caveat there though is the caller doesn't know which one he got ?20:48
hemnaI guess it doesn't really matter20:49
*** achanda has joined #openstack-nova20:49
melwittthis is the novaclient doc showing how to request the version when you create the client object http://docs.openstack.org/developer/python-novaclient/api.html20:49
hemnamelwitt, great thank you20:49
bauzasmelwitt: hah, I love the warning there20:49
bauzas'could shoot you in the foot"20:50
*** su_zhang has quit IRC20:50
scottdaBut the novaclient will request an API microversion from nova API. Doesn't that mean the Nova API will need a new function that will call something new via the cinderclient to make all of this work?20:50
*** dustins has quit IRC20:50
melwittbauzas: heh, yeah20:51
*** su_zhang has joined #openstack-nova20:51
*** garyk has joined #openstack-nova20:52
dansmithscottda: the rest api you mean?20:52
dansmithoh, no20:52
*** raildo is now known as raildo-afk20:53
dansmithnova-api will need to change to request a specific version when it makes some cinderclient call, correct20:53
hemnadansmith, well it will just ask for a version when it instantiates the cinderclient object no ?20:53
melwittyeah, I was thinking, do we add a microversion to the nova api that will request the new version from cinder I guess?20:54
mriedemalaski: it's a good thing you approved that change already :) http://lists.openstack.org/pipermail/openstack-operators/2015-September/008224.html20:54
dansmithhemna: it's usually per-call I think20:54
dansmithmelwitt: only if we have to change our behavior, right?20:54
dansmithso it depends20:54
scottdaOK. So this is a bit complex because to the way calls to cinder for volume attach are intertwined with the nova manage code...Has this been done before for any other microversion change.20:54
hemnadansmith, but this http://docs.openstack.org/developer/python-novaclient/api.html  looks like it's per object20:54
hemna?20:54
melwittdansmith: hm, yeah20:54
*** d34dh0r53 has quit IRC20:55
dansmithhemna: yeah, it does20:55
*** achanda has quit IRC20:55
*** d34dh0r53 has joined #openstack-nova20:55
alaskimriedem: heh.  it's a harmless revert if necessary20:55
dansmithhemna: I defer to melwitt who *actually* knows, as opposed to myself who just knows someone that knows :)20:55
*** achanda has joined #openstack-nova20:55
alaskimriedem: I can remove the +W if you want20:55
bauzasmriedem: alaski: that's a deprecation, not exactly a removal :)20:56
dansmithhemna: my point is, I think, that you might make one call at 2.1 and another at 2.320:56
mriedemalaski: i know, just thought it was funny20:56
melwitthemna: yeah, it's per object, the version specified20:56
mriedem'feel free to comment, oh, nvm, f you!'20:56
hemnaok, that makes sense20:57
alaskimriedem: heh, I was so hyped I missed that thread20:57
hemnaso I suppose after cinder microversions land and the updated version that has VolumeIsBusy, we'll patch nova to ask for the new microversion20:57
mriedemi'm glad i could help hype you up today20:57
hemnaand add all the try: blocks around the calls to cinderclient and deal w/ the new exception20:57
dansmithhemna: yar20:57
hemnathe old nova won't ask the new cinderclient for the newer microversion and cinderclient/cinder has to behave as it does today20:58
bauzasso I wonder how AZs can be helpful with containers for NFV20:58
mriedembauzas: great question20:59
hemnascottda, follow that ?20:59
*** VW has quit IRC20:59
*** VW has joined #openstack-nova21:00
scottdaYes,  I get it at the conceptual level. I just ported the nova microversion code to cinder, so I get all that....21:00
hemnaok coolio21:00
hemnaI 'think' I get it21:00
hemnaheh21:00
scottdaI'm just concerned with putting the microversion wrappers in the right place in the Nova code.21:00
dansmithscottda: meaning the try..except to handle the new exception?21:01
hemnaI don't think there is a heck of a lot to do on the nova side21:01
hemnaother than adding the try blocks and asking for a specific version when creating cinderclient object21:01
scottdadansmith: yeah. Should those go in the manager.py code around calls to volume_api.attach()?21:01
*** penick has quit IRC21:01
dansmithscottda: well, I dunno what you're specifically trying to do, but .. sounds right ;)21:02
scottdaWith a microversion > x wrapper?21:02
hemnascottda, the microversion gets set at cinderclient creation time21:02
*** VW_ has joined #openstack-nova21:02
dansmithscottda: no microversion comparison there, just request the one you want, and then try..except around the call21:02
hemnaso that would just go here: https://github.com/openstack/nova/blob/master/nova/volume/cinder.py#L12821:03
scottdaWell, that's right. If there are try blocks in the new code, it won't matter if you get an old Cinder server. Duh.21:03
dansmithalaski: doesn't all this mean that you guys will have to finally deploy a newer-than-diablo cinder? :P21:03
*** VW has quit IRC21:03
*** penick has joined #openstack-nova21:04
alaskidansmith: hah.  I do believe we're up to Havana now sir21:04
dansmithalaski: ooh, fancy pants21:04
*** hemna is now known as hemnafk21:05
scottdaWell, if you've Havana Cinder, it won't recognize the microversion header, and you'll get the old behaviour. That's the point of all this, right?21:06
alaskibut yes, we would need to move more towards the middle of the alphabet21:06
dansmithscottda: well, not really21:06
dansmithscottda: they will have to be on a newer cinderclient to even take the parameter21:06
*** smatzek has joined #openstack-nova21:06
dansmithscottda: and then cinderclient should not agree to talk to an old cinder if the caller requested at least version X21:06
dansmithscottda: otherwise nova is saying "I NEED a version that throws this exception if this happens"21:07
dansmithand cinderclient would be lying if it agreed but didn't deliver21:07
* dansmith likes to anthropomorphize python code21:07
alaskiwe do run up to date cinderclients21:08
dansmithyeah, but once we roll to this, you'll be stuck i think21:08
alaskiyeah, if Nova starts requiring the use of microversions21:08
alaskiwhen they're in the client21:08
dansmith"Cinder microversion 2.3: Eff-up RAX"21:09
*** penick has quit IRC21:09
alaskiheh21:10
jrollHEY21:10
*** thorst1 has quit IRC21:10
*** rlrossit has left #openstack-nova21:10
jrollalaski: also we're on icehouse \o/21:10
alaskiwoo!21:10
dansmithicehouse cinder?21:10
jrollyes21:11
mriedemthat's getting dangerously close to a stable branch maintained upstream21:11
dansmithhot damn, party like it's 2013!21:11
*** salv-orlando has joined #openstack-nova21:11
jrollscottda: this is how we're dealing with adding microversions to nova-ironic interactions, if it helps https://review.openstack.org/#/c/221848/721:12
jrollscottda: think of pre-microversions as 1.0 or whatever21:12
jrolldefault to that21:12
jrollif it's configured to use newer... do it21:12
*** bswartz has joined #openstack-nova21:13
scottdathanks21:15
jrollscottda: oh, one more thing, require a newer cinderclient in nova that does that the version param21:17
scottdaYeah, it looks like an older Novaclient will work with newer (microversioned Cinder) but a newer (microversioned) Novaclient will not work with older cinder.21:18
*** ZZelle_ has joined #openstack-nova21:18
dansmithscottda: ideally it would21:18
dansmithscottda: it should, if possible, fall back to the other behavior if the cinder is too old21:19
jrollwhy does novaclient come into the picture? does novaclient talk to cinder directly?21:19
dansmithscottda: so in nova, we can detect if the server supports this stuff by probing21:19
jrollI though it would be novaserver talking to cinder21:19
dansmithjroll: in some cases yeah21:19
jrollfor real? :|21:19
mriedemdansmith: we removed that...21:20
dansmithI don't remember why, nor do I remember people being thrilled about it :)21:20
dansmithah21:20
* jroll adds ironic support to novaclient21:20
mriedemor deprecated for removal, it was my todo after YVR21:20
melwittyes21:20
*** tpeoples has joined #openstack-nova21:20
mriedembtw, dibs on removal in mitaka21:20
mriedemwhich i guess is nowish21:20
*** smatzek has quit IRC21:20
melwittjroll: nooo21:21
melwittnovaclient: the everything client21:21
*** Marga_ has joined #openstack-nova21:21
*** tjones has left #openstack-nova21:21
jrollhehe21:21
mriedemexcept for translations21:21
melwittit was that and we're making it not that. or at least mriedem is21:21
mriedemwe aren't allowed to translate novaclient21:21
melwittoh, right21:22
* jroll backs away slowly21:22
mriedemhey! stop him!21:23
jrollmriedem: do I actually want to know why that is not allowed?21:24
jrollI feel like probably not, but curiousity and all21:24
dansmithjroll: are you itching to translate more crap?21:24
mriedemthe argument is that there are higher priorities21:24
mriedemwhich there are21:24
jrollso don't bother.21:24
mriedembut i don't see how that means you can't make it possible to translate novaclient21:24
jrollhuh.21:24
mriedemyeah, can't win, don't try21:24
*** ndipanov has quit IRC21:25
mriedemplus, horizon and osc still proxy through novaclient21:25
mriedemso it's not like it's deprecated21:25
melwittthe ML thread said we can translate openstackclient, to just focus the efforts over there21:25
jrollthat means we can delete all of the low-priority blueprints, then, right?21:25
mriedemmelwitt: yeah, but there are things not using osc21:25
jrolllike all the people21:25
* jroll ducks21:25
mriedembut i agree osc is higher priority21:25
melwittyeah21:25
mriedembut osc is not a python sdk21:26
mriedemand we aren't supposed to use the sdk in the server projects21:26
mriedemso we use the clients, or write our own rest service like tempest does21:26
mriedemah damn21:28
mriedem# NOTE(mriedem): Remove this along with the deprecated commands in the first21:28
mriedem# python-novaclient release AFTER the nova server 13.0.0 'M' release.21:28
mriedemcan't remove that volume stuff from novaclient until after mitaka21:28
*** jichen has quit IRC21:28
melwitthells bells21:29
*** otter768 has joined #openstack-nova21:29
dansmithlol21:29
dansmithmelwitt: now you're just trying21:29
*** VW_ has quit IRC21:29
mriedemvery curmudgeonly, well done21:30
melwittI was going to say "darn it" and mriedem name reminded me there are other options21:30
*** VW has joined #openstack-nova21:31
*** Apoorva has joined #openstack-nova21:32
*** achanda has quit IRC21:32
*** otter768 has quit IRC21:34
*** vladikr has quit IRC21:35
*** matrohon has quit IRC21:36
*** achanda has joined #openstack-nova21:36
*** obedmr_ has joined #openstack-nova21:40
*** hemnafk is now known as hemna21:40
*** burgerk has quit IRC21:42
*** obedmr has quit IRC21:43
openstackgerritChuck Carmack proposed openstack/nova: WIP: Disable security group tests in tempest-dsvm-cells-rc using a config option  https://review.openstack.org/22604321:43
melwittdansmith: I've been looking at this patch https://review.openstack.org/#/c/173723/, it seems correct but I had a hard time tracing how to verify that FixedIP 1.1 goes with Network 1.021:45
melwittdansmith: then I was thinking about how we could test for these. I wasn't clear if these will all be caught by the test_obj_relationships from now on and if there are errors we just have to fix them manually and detecting them would be hard?21:46
melwitt(detecting old errors, I mean)21:47
openstackgerritBrian Haley proposed openstack/nova: Disable IPv6 on bridge devices  https://review.openstack.org/19805421:47
dansmithmelwitt: the easy thing to do is to punt until mitaka and then we can delete all the object relationship crap21:47
*** burt has quit IRC21:47
dansmithmelwitt: and the problem goes away because the client declares all its versions21:47
melwittdansmith: okay. I missed that we won't need object relationships anymore21:49
dansmithmelwitt: yeah.. you're welcome :)21:49
melwitthaha21:50
bauzasdansmith: I thought we needed a V2 for each object?21:50
dansmithbauzas: nope21:51
dansmithbauzas: instancev2 depends on that functionality,21:51
dansmithbut once we're sure that we're cool, we can drop all relationship stuff from mitaka21:51
dansmithbecause liberty clients will always ask for backports with a full list of their supported objects21:51
bauzasdansmith: oh okay21:51
bauzasdansmith: yeah saw the conductor change21:52
*** jaypipes has quit IRC21:52
dansmithwe just couldn't remove that from liberty because kilo clients won't, so liberty has to continue to use the relationship maps21:52
dansmithbut no kilo nodes will send instance v2, so we're cool21:52
dansmithit's a complex dance21:52
dansmithso I forgive you all for not keeping it all in your heads :)21:52
melwittwhew! :)21:53
bauzasdansmith: gotcha21:54
bauzasonce mitaka opens, we're free to say that liberty and mitaka know the new stuff21:54
bauzasso we can drop the old stufuf21:54
bauzaswell, mitaka opened actually21:55
* bauzas needs to update his mindset21:55
mriedemtonyb: are you going to mark the nova part of this bug as invalid? https://bugs.launchpad.net/nova/+bug/141508721:55
openstackLaunchpad bug 1415087 in OpenStack Compute (nova) "[OSSA 2015-011] Format-guessing and file disclosure in image convert (CVE-2015-1850, CVE-2015-1851)" [High,Triaged] - Assigned to Tony Breeds (o-tony)21:55
tonybmriedem: Hmmm I s'pose.21:56
mriedemthe thing is, people are freaking about publishing security advisories21:56
tonybmriedem: I suspect it's an issue but I'm not enough of a black hat to get to it :/21:56
*** achanda has quit IRC21:57
dansmithbauzas: yeah, but I don't want to actually drop things until the grenade job is doing L->M21:57
bauzasoh yeah21:58
tonybmriedem: I thought that was dealt with :(21:58
mriedemso cinder had a change here https://review.openstack.org/#/c/191785/22:00
*** achanda has joined #openstack-nova22:00
mriedemwhere qemu-img convert was called22:00
mriedemand so does nova https://github.com/openstack/nova/blob/master/nova/virt/images.py#L7122:01
*** eharney has quit IRC22:01
*** rfolco has quit IRC22:02
tonybmriedem: Sure I had a similar fix but was never able to show it helped22:03
*** flaper87 has quit IRC22:03
*** nihilifer1 has joined #openstack-nova22:03
*** nihilifer has quit IRC22:03
*** alexpilotti has joined #openstack-nova22:04
tonybmriedem: So how about we hold off on closeing the bug, I'll polish my patch and add it and then we can talk through it at the meeting ~20hours22:04
*** ccarmack has left #openstack-nova22:05
*** flaper87 has joined #openstack-nova22:07
openstackgerritChuck Carmack proposed openstack/nova: WIP: Disable security group tests in tempest-dsvm-cells-rc using a config option  https://review.openstack.org/22604322:08
*** penick has joined #openstack-nova22:10
*** mdrabe has quit IRC22:11
*** VW has quit IRC22:11
*** VW has joined #openstack-nova22:12
*** stevemar has joined #openstack-nova22:13
*** annegent_ has joined #openstack-nova22:14
*** stevemar has quit IRC22:15
*** openstackgerrit has quit IRC22:16
*** otter768 has joined #openstack-nova22:16
*** VW has quit IRC22:16
*** openstackgerrit has joined #openstack-nova22:16
*** VW has joined #openstack-nova22:17
*** annegentle has quit IRC22:18
*** smatzek has joined #openstack-nova22:19
*** smatzek has quit IRC22:19
*** salv-orlando has quit IRC22:20
*** mestery has quit IRC22:21
*** diogogmt has quit IRC22:21
*** mestery has joined #openstack-nova22:22
*** salv-orlando has joined #openstack-nova22:22
openstackgerritChristine Wang proposed openstack/nova: Update name validation with expanded characters  https://review.openstack.org/22703922:26
*** mriedem is now known as mriedem_away22:27
*** ctrath has quit IRC22:28
*** otter768 has quit IRC22:31
*** jamielennox is now known as jamielennox|away22:35
*** su_zhang has quit IRC22:35
*** su_zhang has joined #openstack-nova22:35
*** alejandrito has quit IRC22:36
*** otter768 has joined #openstack-nova22:36
*** VW_ has joined #openstack-nova22:37
*** leitan has quit IRC22:40
*** VW has quit IRC22:41
*** VW_ has quit IRC22:42
*** dims__ has joined #openstack-nova22:43
*** tjones1 has joined #openstack-nova22:44
*** alexpilotti has quit IRC22:44
*** dims_ has quit IRC22:46
*** su_zhang has quit IRC22:46
*** su_zhang has joined #openstack-nova22:46
*** annegent_ has quit IRC22:49
*** ZZelle_ has quit IRC22:52
*** obedmr_ has quit IRC22:54
*** alex_klimov has quit IRC22:54
*** jerrygb has quit IRC22:55
*** jwcroppe has quit IRC22:56
*** jwcroppe has joined #openstack-nova22:57
*** jwcroppe has quit IRC23:01
*** asettle has joined #openstack-nova23:03
*** annegentle has joined #openstack-nova23:04
*** annegentle has quit IRC23:05
*** tjones1 has left #openstack-nova23:07
*** tonytan4ever has quit IRC23:09
*** tpeoples has quit IRC23:09
*** hemna is now known as hemnafk23:10
*** xyang1 has quit IRC23:12
*** stevemar has joined #openstack-nova23:13
*** emagana has quit IRC23:18
*** electrocucaracha has quit IRC23:19
*** penick has quit IRC23:21
*** zhangjn has joined #openstack-nova23:24
*** zhangjn has quit IRC23:24
*** penick has joined #openstack-nova23:28
*** otter768 has quit IRC23:29
*** achanda has quit IRC23:33
*** tpeoples has joined #openstack-nova23:34
*** takedakn has joined #openstack-nova23:35
*** hodos|2 has quit IRC23:36
*** jwcroppe has joined #openstack-nova23:37
*** nic has quit IRC23:41
*** shyu has joined #openstack-nova23:42
*** jwcroppe has quit IRC23:45
*** baoli_ has quit IRC23:51
*** jamielennox|away is now known as jamielennox23:52
*** takedakn1 has joined #openstack-nova23:53
*** takedakn1 has quit IRC23:55
*** takedakn1 has joined #openstack-nova23:55
*** takedakn has quit IRC23:56
*** takedakn1 is now known as takedakn23:56
*** takedakn has quit IRC23:59
*** zhangjn has joined #openstack-nova23:59

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