Monday, 2015-06-22

*** asettle has quit IRC00:00
*** markvoelker has joined #openstack-nova00:01
*** markvoelker has quit IRC00:05
*** otter768 has joined #openstack-nova00:06
*** thorst has quit IRC00:07
*** bradjones has quit IRC00:22
*** lyarwood has quit IRC00:22
*** lyarwood has joined #openstack-nova00:22
*** arnaud___ has joined #openstack-nova00:23
*** bradjones has joined #openstack-nova00:23
*** bradjones has quit IRC00:23
*** bradjones has joined #openstack-nova00:23
*** arnaud___ has quit IRC00:27
*** dimsum__ has quit IRC00:39
*** ferest has quit IRC00:39
*** dimsum__ has joined #openstack-nova00:41
*** annashen has joined #openstack-nova00:48
*** dims_ has joined #openstack-nova00:50
*** dimsum__ has quit IRC00:50
*** IanGovett has joined #openstack-nova00:53
*** otter768 has quit IRC01:05
*** salv-orlando has joined #openstack-nova01:06
*** annashen has quit IRC01:10
*** salv-orlando has quit IRC01:11
openstackgerritDiana Clarke proposed openstack/nova: Handle port group not found case  https://review.openstack.org/19391001:16
*** browne has joined #openstack-nova01:16
*** annashen has joined #openstack-nova01:17
*** vilobhmm has joined #openstack-nova01:17
*** dimsum__ has joined #openstack-nova01:17
*** dims_ has quit IRC01:17
*** jgrimm_ is now known as zz_jgrimm_01:27
*** dims_ has joined #openstack-nova01:40
*** dimsum__ has quit IRC01:41
openstackgerritFei Long Wang proposed openstack/nova: Fix nova backup for volume-backed instance  https://review.openstack.org/16449401:41
*** annashen has quit IRC01:46
*** annashen has joined #openstack-nova01:49
*** markvoelker has joined #openstack-nova01:50
openstackgerritKen'ichi Ohmichi proposed openstack/nova-specs: Fix error messages on check-flavor-type  https://review.openstack.org/19391901:52
*** Sukhdev has joined #openstack-nova01:52
*** markvoelker has quit IRC01:55
*** annashen has quit IRC02:01
*** linkedinyou has quit IRC02:06
*** cfriesen has joined #openstack-nova02:08
openstackgerritXinyuan Huang proposed openstack/nova: Pass requested_networks to scheduler  https://review.openstack.org/19363502:53
*** IanGovett has quit IRC02:53
*** iamjarvo has joined #openstack-nova03:01
*** tpeoples has joined #openstack-nova03:05
*** vilobhmm has quit IRC03:05
*** otter768 has joined #openstack-nova03:05
*** lyarwood has quit IRC03:08
*** lyarwood has joined #openstack-nova03:09
*** achanda has joined #openstack-nova03:10
*** otter768 has quit IRC03:10
*** sdake_ has quit IRC03:10
*** tpeoples has quit IRC03:12
*** alexpilotti has joined #openstack-nova03:13
openstackgerritDiana Clarke proposed openstack/nova: Register the vnc config options under group 'vnc'  https://review.openstack.org/19392303:14
*** jmccrory has quit IRC03:14
*** jmccrory has joined #openstack-nova03:16
*** salv-orlando has joined #openstack-nova03:19
*** asettle has joined #openstack-nova03:21
*** salv-orlando has quit IRC03:22
*** takedakn has joined #openstack-nova03:24
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py  https://review.openstack.org/18778503:28
*** Marga_ has joined #openstack-nova03:30
*** Marga_ has quit IRC03:30
*** Marga_ has joined #openstack-nova03:31
*** takedakn has quit IRC03:33
*** achanda has quit IRC03:33
*** takedakn has joined #openstack-nova03:34
*** annashen has joined #openstack-nova03:34
*** mriedem1 has joined #openstack-nova03:37
*** mriedem has quit IRC03:38
*** markvoelker has joined #openstack-nova03:38
*** dims_ has quit IRC03:40
*** markvoelker has quit IRC03:43
*** mestery has joined #openstack-nova03:44
*** Longgeek has quit IRC03:48
*** Longgeek has joined #openstack-nova03:49
*** Sukhdev has quit IRC03:52
*** annashen has quit IRC03:54
*** jhesketh has quit IRC03:58
*** jhesketh has joined #openstack-nova04:01
*** takedakn has quit IRC04:02
*** sdake has joined #openstack-nova04:04
*** deepthi has joined #openstack-nova04:06
*** iamjarvo has quit IRC04:09
*** ildikov has quit IRC04:11
*** annashen has joined #openstack-nova04:12
*** avishay_ has joined #openstack-nova04:13
*** unicell1 has joined #openstack-nova04:25
*** garyk has joined #openstack-nova04:26
*** unicell has quit IRC04:27
*** annashen has quit IRC04:31
*** cfriesen has quit IRC04:32
*** cfriesen has joined #openstack-nova04:32
*** n0ano has joined #openstack-nova04:32
*** pradipta has joined #openstack-nova04:35
*** cfriesen has quit IRC04:39
*** dimsum__ has joined #openstack-nova04:41
*** Marga_ has quit IRC04:41
*** iamjarvo has joined #openstack-nova04:43
*** dimsum__ has quit IRC04:46
*** ijw_ has joined #openstack-nova04:46
*** Marga_ has joined #openstack-nova04:49
*** _gryf has joined #openstack-nova04:52
*** mestery has quit IRC04:54
*** jwang has joined #openstack-nova04:55
*** vilobhmm has joined #openstack-nova04:56
*** Nikolay_St has quit IRC04:56
*** jwang_ has quit IRC04:57
*** irenab has quit IRC04:58
*** jwang_ has joined #openstack-nova04:59
*** jwang has quit IRC05:01
*** salv-orlando has joined #openstack-nova05:02
*** vilobhmm has quit IRC05:05
*** nkrinner has joined #openstack-nova05:06
*** otter768 has joined #openstack-nova05:06
*** radek has joined #openstack-nova05:07
*** rajesht has joined #openstack-nova05:07
*** vilobhmm has joined #openstack-nova05:08
*** iamjarvo has quit IRC05:10
*** vilobhmm has quit IRC05:10
*** otter768 has quit IRC05:11
*** vilobhmm has joined #openstack-nova05:12
*** vilobhmm has quit IRC05:12
*** salv-orlando has quit IRC05:13
*** ijw_ has quit IRC05:15
*** radek__ has joined #openstack-nova05:18
*** Jabadia has joined #openstack-nova05:19
*** veena has joined #openstack-nova05:20
*** iamjarvo has joined #openstack-nova05:21
*** radek has quit IRC05:21
JabadiaAssuming I have 10 Computes. 10 VM on Each. now i need to replace compute #10 for some reason ( maintenance ) or it have some problem.05:22
*** ijw_ has joined #openstack-nova05:23
JabadiaEvacuate is not perfect for the job, as it will place all 10 VM's on single compute which will be unbalanced scenario05:23
*** gmann has quit IRC05:24
Jabadiaquestion is , am I missing something, or, evacuate need to be enhanced so that it will use scheduler for each VM migration, and not select single compute/host for all05:24
JabadiaI will be glad to implement, I just want to make sure it's not a waste of time05:25
openstackgerritgaryk proposed openstack/nova: Cells: add in CellMapping get_by_id method  https://review.openstack.org/19393805:25
*** nihilifer has joined #openstack-nova05:26
*** markvoelker has joined #openstack-nova05:27
*** rushiagr_away is now known as rushiagr05:27
*** gmann has joined #openstack-nova05:30
openstackgerritgaryk proposed openstack/nova: Cells: add instance cell registration utility to nova-manage  https://review.openstack.org/19388905:31
*** markvoelker has quit IRC05:32
*** kiran-r has joined #openstack-nova05:33
openstackgerritgaryk proposed openstack/nova: Cells: add in missing unit test for get_by_uuid  https://review.openstack.org/19393905:35
*** sudipto has joined #openstack-nova05:41
*** jikyunkim_ has joined #openstack-nova05:42
*** lpetrut has joined #openstack-nova05:42
*** irenab has joined #openstack-nova05:43
*** Nikolay_St has joined #openstack-nova05:46
*** nihilifer has quit IRC05:46
*** Jabadia has quit IRC05:47
*** rajesht has quit IRC05:50
*** alexpilotti has quit IRC05:50
*** rajesht has joined #openstack-nova05:50
*** jmccrory has quit IRC05:52
*** prashup has joined #openstack-nova05:54
prashuphi guys05:56
prashupi pushed a code for review. however it failed due to pep8 (i.e. which was due to code formatting)05:56
prashupand there is another failure due to hyper-v-ci05:56
prashupi m looking into the logs ; but can not find anything significant which points to failure05:56
prashupcan somebody help05:56
prashuphttp://64.119.130.115/193444/1/05:56
*** jmccrory has joined #openstack-nova05:57
sudiptoprashup, you mean you aren't sure of what failed in hyper-v CI? Or both that and pep8?05:59
prashup only hyper-v-CI06:00
prashupformatting errors are clear06:00
prashupi m not sure what failed in hyper-v CI06:00
sudiptodid you try a recheck?06:01
prashupsure i will06:01
*** coolsvapz is now known as coolsvap06:02
*** rajesht has quit IRC06:03
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex  https://review.openstack.org/19003406:10
*** belmoreira has joined #openstack-nova06:10
*** asettle has quit IRC06:10
*** unicell1 has quit IRC06:13
*** unicell has joined #openstack-nova06:14
*** Marga_ has quit IRC06:17
*** iamjarvo has quit IRC06:18
*** browne has quit IRC06:19
*** Marga_ has joined #openstack-nova06:23
*** oro has joined #openstack-nova06:23
*** gmann has quit IRC06:24
*** avishay_ has quit IRC06:25
*** avishay_ has joined #openstack-nova06:25
*** garyk1 has joined #openstack-nova06:26
*** garyk has quit IRC06:28
*** gmann has joined #openstack-nova06:28
*** abhishekk has joined #openstack-nova06:28
*** garyk1 has quit IRC06:31
*** pradipta has quit IRC06:34
*** ianbrown has quit IRC06:35
*** ttx` is now known as ttx06:40
*** ttx has quit IRC06:41
*** ttx has joined #openstack-nova06:41
*** frickler_ is now known as frickler06:41
*** pkoniszewski has joined #openstack-nova06:41
avishay_Hi all. Do I need /var/lib/nova/instances to be on shared storage if I used only boot-from-volume, and have libvirt transfer the xml files?06:43
*** avishay_ is now known as avishay06:44
*** moshele has joined #openstack-nova06:44
*** pradipta has joined #openstack-nova06:46
*** exploreshaifali has joined #openstack-nova06:47
*** Jabadia has joined #openstack-nova06:47
*** Longgeek has quit IRC06:47
*** Longgeek has joined #openstack-nova06:47
*** sahid has joined #openstack-nova06:50
*** hshiina has joined #openstack-nova06:53
*** jhesketh has quit IRC06:54
openstackgerritsahid proposed openstack/nova-specs: console: convert consoles code to use objects framework  https://review.openstack.org/16583806:55
openstackgerritgaryk proposed openstack/nova: Cells: add in missing unit test for get_by_uuid  https://review.openstack.org/19393906:56
*** lpetrut has quit IRC06:58
openstackgerritgaryk proposed openstack/nova: Cells: add instance cell registration utility to nova-manage  https://review.openstack.org/19388907:02
*** jhesketh has joined #openstack-nova07:04
*** otter768 has joined #openstack-nova07:07
*** salv-orlando has joined #openstack-nova07:10
*** otter768 has quit IRC07:12
*** salv-orlando has quit IRC07:13
openstackgerritDiana Clarke proposed openstack/nova: VMware: Handle port group not found case  https://review.openstack.org/19391007:14
*** markvoelker has joined #openstack-nova07:16
*** doron_afk has joined #openstack-nova07:16
*** salv-orlando has joined #openstack-nova07:17
*** mpaolino has joined #openstack-nova07:19
*** markvoelker has quit IRC07:21
*** exploreshaifali has quit IRC07:23
*** exploreshaifali has joined #openstack-nova07:23
*** nihilifer has joined #openstack-nova07:24
*** foexle has joined #openstack-nova07:24
*** ildikov has joined #openstack-nova07:26
*** markus_z has joined #openstack-nova07:26
*** mpavone has joined #openstack-nova07:26
*** garyk has joined #openstack-nova07:27
*** mpavone has quit IRC07:27
garykjohnthetubaguy: danpb: can you please see https://review.openstack.org/192675. I addressed all of the comments over the weekend07:27
*** aloga_ is now known as aloga07:28
*** matrohon has joined #openstack-nova07:30
*** afazekas has joined #openstack-nova07:31
*** jlanoux has joined #openstack-nova07:32
*** Marga_ has quit IRC07:34
*** moshele has quit IRC07:35
*** doude_ has quit IRC07:37
*** doude has joined #openstack-nova07:38
*** ndipanov has joined #openstack-nova07:39
*** bradjones has quit IRC07:43
*** bradjones has joined #openstack-nova07:45
*** bradjones has quit IRC07:45
*** bradjones has joined #openstack-nova07:45
*** nihilifer has quit IRC07:48
*** nihilifer has joined #openstack-nova07:50
*** moshele has joined #openstack-nova07:56
*** haomaiwa_ has quit IRC07:56
moshelejohnthetubaguy: hi07:59
*** nihilifer has quit IRC07:59
*** bradjones has quit IRC08:02
*** lpetrut has joined #openstack-nova08:03
*** Marga_ has joined #openstack-nova08:03
*** tdurakov has quit IRC08:03
*** lucasagomes has joined #openstack-nova08:03
*** Marga_ has quit IRC08:06
*** Marga_ has joined #openstack-nova08:07
*** Jabadia has quit IRC08:09
*** dsanders has quit IRC08:10
*** doron_afk is now known as doron_08:11
*** bradjones has joined #openstack-nova08:14
*** bradjones has quit IRC08:14
*** bradjones has joined #openstack-nova08:14
*** Marga_ has quit IRC08:17
*** nihilifer has joined #openstack-nova08:17
*** yassine has joined #openstack-nova08:17
*** dsanders has joined #openstack-nova08:17
*** lyarwood has quit IRC08:18
*** lyarwood has joined #openstack-nova08:18
*** alex_klimov has joined #openstack-nova08:20
*** josecastroleon has joined #openstack-nova08:21
*** Marga_ has joined #openstack-nova08:21
*** belmoreira has quit IRC08:22
*** nihilifer has quit IRC08:23
*** ildikov has quit IRC08:23
*** claudiub has joined #openstack-nova08:24
*** matrohon has quit IRC08:25
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/19399308:26
*** moshele has quit IRC08:27
openstackgerritOpenStack Proposal Bot proposed openstack/python-novaclient: Updated from global requirements  https://review.openstack.org/19402708:28
*** moshele has joined #openstack-nova08:28
*** dsanders has quit IRC08:28
openstackgerritJens Rosenboom proposed openstack/nova: Add microversion to allow server search option ip6 for non-admin  https://review.openstack.org/17956908:29
*** claudiub has quit IRC08:30
*** moshele has quit IRC08:31
*** moshele has joined #openstack-nova08:32
*** oomichi has joined #openstack-nova08:32
openstackgerritPetrut Lucian proposed openstack/nova: Hyper-V: Implement nova rescue  https://review.openstack.org/12715908:33
*** danpb has joined #openstack-nova08:34
mikalDoes Vladik Romanovsky hang out on IRC?08:35
*** aix has joined #openstack-nova08:36
ndipanovmikal, he does, but he's in canada so probably not now :)08:36
mikalOh, how selfish08:36
mikalOk, I'll just mutter old man things on this review instead of asking him here08:36
*** sudipto has quit IRC08:36
*** ildikov has joined #openstack-nova08:36
*** jlanoux has quit IRC08:37
*** sudipto has joined #openstack-nova08:37
*** bradjones has quit IRC08:38
*** mnestratov|2 has joined #openstack-nova08:39
*** bradjones has joined #openstack-nova08:40
*** bradjones has quit IRC08:40
*** bradjones has joined #openstack-nova08:40
*** rotbeard has joined #openstack-nova08:41
mikalIs libosinfo in pip anywhere? I can08:41
mikal't see it08:41
*** jcoufal has joined #openstack-nova08:41
*** gszasz has joined #openstack-nova08:43
*** tdurakov has joined #openstack-nova08:43
*** derekh has joined #openstack-nova08:44
kashyapmikal: libosinfo is a C project. . .08:46
danpbmikal: it doesn't work that way08:46
danpbmikal: libosinfo is a C library built using GObject, so there is no need for python bindings08:47
mikaldanpb: keep talking08:47
mikaldanpb: what's gobject?08:47
danpbmikal: the PyGobject dynamically calls into the C library using the magic of GObject Introspection08:47
*** karimb has joined #openstack-nova08:47
danpbmikal: its the low level c object model on wihch gtk is built08:47
mikaldanpb: is that what the bonkers module path is about in the import?08:47
mikaldanpb: have we attempted to communicate to distros that we want to use this thing? It hasn't gone through requirements.txt because its optional, but I feel like most people will want to enable it.08:49
*** doron_ is now known as doron_afk08:50
mikaldanpb: perhaps it would be more efficient if you took a look at the comments I put made to https://review.openstack.org/#/c/14962508:50
*** davideagnello has joined #openstack-nova08:50
*** alex_klimov has quit IRC08:51
danpbmikal: this was discussed in the original spec08:51
danpbbasically, the libosinfo library will be available in any distro that is new enough to include GNOME 308:51
danpbso that's basically anything that's released in the last 2-3 years08:52
danpbRHEL-6 is the main one that won't have it, but that's not a big deal08:52
mikalWhat about server distros with no gnome? How much desktop stuff do they have to pull in to get this library?08:52
danpbthere's no dependancy on "desktop stuff"08:52
danpbgobject is a low level library - its just maintained by gnome08:52
danpbthe same for libosinfo08:53
mikalYeah, it just doesn't feel very pythonic, and hence I worry about confusion08:54
*** davideagnello has quit IRC08:55
openstackgerritSylvain Bauza proposed openstack/nova-specs: Proposes to check destination on migrations  https://review.openstack.org/17922408:55
avishayHi all. Do I need /var/lib/nova/instances to be on shared storage if I used only boot-from-volume, and have libvirt transfer the xml files?08:56
*** prashup has quit IRC08:56
mikaldanpb: are these osinfo lookups particularly expensive? Why are we bothering with a caching layer at all here?08:57
*** jcoufal has quit IRC08:57
danpbmikal: the actual loading of the database can be expensive, but the individual lookups probably less so - but i've not actually timed them09:00
*** oomichi has quit IRC09:01
mikaldanpb: ok, yeah the db is loaded once though, so that's not a big deal09:01
mikaldanpb: I feel like we could drop all of the caching code here and no one would notice09:01
openstackgerritPeter Savage proposed openstack/python-novaclient: Added marker functionality to flavours and images  https://review.openstack.org/19321209:01
*** linkedinyou has joined #openstack-nova09:02
*** markvoelker has joined #openstack-nova09:05
garykdanpb: can you please look at https://review.openstack.org/192675 - i addressed your comments over the weekend09:05
*** deepthi has quit IRC09:05
*** belmoreira has joined #openstack-nova09:07
*** ociuhandu has joined #openstack-nova09:07
danpbmikal: quite possible09:08
*** otter768 has joined #openstack-nova09:08
*** markvoelker has quit IRC09:09
tdurakovjohnthetubaguy: hi John09:12
johnthetubaguytdurakov: hi09:12
johnthetubaguymoshele: I am around09:13
*** romainh has joined #openstack-nova09:13
*** otter768 has quit IRC09:13
tdurakovjohnthetubaguy: could you review my spec? https://review.openstack.org/#/c/184495/ left comment to you09:14
johnthetubaguytdurakov: sorry, I am busying sorting out the liberty-1 release paperwork this week, although hopefully I can get to it at some point09:15
*** jikyunkim_ has quit IRC09:16
tdurakovjohnthetubaguy: is there any possibility that you review it before freeze?09:16
*** deepthi has joined #openstack-nova09:17
johnthetubaguytdurakov: I am unsure at this point, given the backlog of 100 specs that all want me to review them09:18
garyktdurakov: is it possible to split that patch into smaller ones (stacked on on top of the other)09:18
*** ociuhandu has quit IRC09:19
*** jog0 has joined #openstack-nova09:19
*** jog0 is now known as jogo09:19
tdurakovjohnthetubaguy, garyk: wrong link: https://review.openstack.org/#/c/183057/09:19
*** claudiub has joined #openstack-nova09:19
tdurakovgaryk: ok, I'll split patch for resize into smaller parts09:20
garyktdurakov: thanks!09:20
johnthetubaguytdurakov: I am a strong -1 on that API in its current form, I am afraid09:23
bauzasndipanov: around for a question re: to NUMA ?09:23
*** ociuhandu has joined #openstack-nova09:23
openstackgerritRajesh Tailor proposed openstack/nova: Fix quota-update when nova-compute startup finish  https://review.openstack.org/17011809:24
bauzasndipanov: I'm just trying to get rid of https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L1191 in https://review.openstack.org/#/c/191251/4/nova/scheduler/host_manager.py,cm09:24
bauzasndipanov: so just wanted to make sure that this method is actually there for making sure that numa_usage_from_instances is correctly called ?09:25
*** rajesht has joined #openstack-nova09:25
bauzasndipanov: if so, I can directly call numa_usage_from_instances() directly from the HostManager09:25
bauzasndipanov: since I already have an InstanceNUMATopology object09:26
ndipanovbauzas, did you read the docstring?09:28
bauzasndipanov: yup, I did09:28
ndipanovbauzas, no one reads those, I know09:28
tdurakovjohnthetubaguy: checked your comment. approach with others could cause the headache in future - we have to carry this every time. The point of this spec is to remove ugly serialized string from response.  I'm not sure that this info is have to be well defined in enums.09:28
tdurakovjoihthetubaguy: what is the profit of such enums there?09:28
bauzasndipanov: to be clear, you're saying why those methods are there09:29
*** deepthi has quit IRC09:29
johnthetubaguytdurakov: so our users will know what the possible API responses are, and so all the different drivers are forced to return the same set of values for that API, should they support that API09:29
bauzasndipanov: okay, if you think my ping is not good, then leave me try to do this and wait for your comment09:29
ndipanovbauzas, well it's not that09:30
ndipanovI just think it's well explained in the docstring09:30
ndipanovso not sure which part is unclear09:30
tdurakovjohnthetubaguy: and users would be forced to convert 'others' values to defined types09:30
ndipanovit's used so that you don't have to do all the type conversions at call site09:30
johnthetubaguytdurakov: I am happy with some catch all values of "unknown" or "other", so it gives us room to grow as those values evolve over time, in the hope that reduces some of the maintenance burden.09:30
johnthetubaguytdurakov: not sure what you mean? it just means its not in the list the API currently supports right?09:31
ndipanovbauzas, if you want to call numa_usage_from_instances09:31
ndipanovbecause that method uses objects09:31
bauzasndipanov: I mean, your docstring is really cool, I just wanted to be sure because I'm like touching things that are quite difficult to change from my perspective, but that's fine, I can test and see :)09:31
ndipanovbut we don't have objects everywhere09:31
ndipanovso unless all code paths that need this use objects09:32
tdurakovyep, and as it is not in list, it's not defined correctly. it became useless for operator, no?09:32
ndipanov(so hoststate, resource_tracker and claims09:32
ndipanov)09:32
ndipanovwe can't remove it09:32
ndipanovand filters09:32
ndipanovso hostState, filter, claim09:32
ndipanovRT09:32
bauzasndipanov: okay, nevermind my question, will test it and we'll see09:32
johnthetubaguytdurakov: if its not defined, its basically could be "any" value, at which point its not that useful either? I guess the use cases for this API are not clear enough in my head to make that call right now.09:33
ndipanovOK man, I mean I can't explain it better than that...09:33
*** jlanoux has joined #openstack-nova09:33
tdurakovjohnthetubaguy: if you really want enums we could bump api in future, but for now the aim of spec is clear format09:34
tdurakovlet's clean up api in few steps09:34
*** rajesht has quit IRC09:34
bauzasndipanov: that's fine, I probably overthought on that09:34
bauzasndipanov: I mean, your docstrings are really clear09:35
bauzasndipanov: so that's why I can test tho09:35
jogojohnthetubaguy: ping http://imgh.us/nova.svg09:35
johnthetubaguytdurakov: so any change we make to the API we basically have to support forever (where forever might be 5 years, or something like that, we don't know yet), so its really quite expensive to do it in too many steps09:35
johnthetubaguyjogo: nice!09:35
tdurakovjohnthetubaguy: we have miocroversions for that, no?09:36
ndipanovbauzas, see where that method is called... in a perfect world we'd just call the one that works on objects...09:36
ndipanovbut we can't do that09:36
jogojohnthetubaguy: built it in dia, I am going to revise it to be nova-net multihost arch only09:36
jogoand add a legend etc.09:36
ndipanovso this was an attempt to keep all of those conversions in one place\09:36
*** rajesht has joined #openstack-nova09:36
jogojohnthetubaguy: but before doing so wanted to get some preliminary feedback about the direction09:36
johnthetubaguytdurakov: yes, but we still have to keep every microversion supported for a long time09:36
johnthetubaguytdurakov: we could turn it back to a string in a future version, if that turns out to make more sense, for example09:37
bauzasndipanov: yeah, but if I'm working right, consume_from_instance will have a spec_obj as a param, and we'll be 100% sure that spec_obj.numa_topology is instance of InstanceNUMATopology09:37
jogotdurakov: microversions are expensive09:38
tdurakovand we will have 2 extra microversions that we have to support09:38
bauzasndipanov: so the helper will only be needed for the host_topology, but not for the instance object09:38
johnthetubaguytdurakov: so, if its left as a string who defines what values that API will take?09:38
johnthetubaguytdurakov: that needs implementing across several virt drivers, how to they make sure they return consistent values for the all strings?09:39
tdurakovjohnthetubaguy: it's already a string, technically all cpu_info is the string now. long, serialize json string. there is no cr's to change that.09:39
johnthetubaguytdurakov: the current API is totally broken, I couldn't agree more on that front09:40
openstackgerritLuo Gangyi proposed openstack/nova: Correct disk filter in dealing with volume-based instance  https://review.openstack.org/19405909:41
ndipanovbauzas, yes - so you can remove the part of the method that handles instances, if that's the last bit that uses the dict09:41
ndipanovthat would be awesome actually09:41
johnthetubaguytdurakov: I am also agreed that your current proposal is way better than the current API09:41
tdurakovjohnthetubaguy: ok, last point: if we decide to have enums, especially with the others value. we will bump versions much more frequent09:41
*** destiny has joined #openstack-nova09:41
ndipanovbauzas, and it probaly is09:41
bauzasndipanov: nope, the PCI manager is using it :(09:42
bauzasndipanov: but lemme see what I can do09:43
tdurakovjohnthetubaguy: all new cpu models have to be added into enums. all such addition will cause to have one more microversion - it'll become microversion hell09:43
johnthetubaguytdurakov: yes. So the way forward here, I feel, is to document this trade off in the alternatives section?09:43
tdurakovenums in alternative section? ok I'll add them.09:44
johnthetubaguytdurakov: that is what that section of the spec was added for, and you can describe the reasons why you want to keep those as strings09:44
sahidmikal: hello, can you re evaluate https://review.openstack.org/#/c/187509/ please, i'm not sure to understand what should i need to update on this spec09:44
tdurakovjohnthetubaguy: thank you John. I'll ping you in few hours09:45
*** deepthi has joined #openstack-nova09:46
mikalsahid: so, yeah, I think the spec should mention that qemu-utils is required. Do we have any other features that require that at the moment (did quiesce ever land?)09:46
johnthetubaguytdurakov: but we need to discuss in there how multiple drivers will have a consistent API (like different names for each CPU feature). Go look at the libvirt source code is not the correct answer there, APIs should be owned by Nova really.09:46
mikalsahid: my point with the admin user name was more that its an attribute of the image, and therefore should be stored as such09:46
mikalsahid: the user not knowing the username was an aside because I was thinking aloud and requires no action from you09:47
sahidmikal: mikal about qemu-utils - you mean required for the guest?09:48
*** oro has quit IRC09:48
mikalsahid: yeah09:48
sahidhum...09:48
mikalsahid: this feature doesn't work unless the guest has that installed, right?09:48
sahidqemu-guest-agent09:48
mikalYeah, that thing09:49
mikalSo, we need to say that in the spec so the tech writers know to document it09:49
mikalI'm not opposed to the agent, but we need to tell people to run it09:49
sahidmikal: sure i mentioned that on user impact09:49
sahidlet me double check09:49
tdurakovjohnthetubaguy: agree, api should be owned by nova. but as for me api should get this info from hypervisor, and left unchanged - as it does currently.09:50
sahidmikal: line 7809:50
sahidnot enough for your point of view?09:50
*** moshele has quit IRC09:50
johnthetubaguytdurakov: problem is, we support many hypervisors, and need an API that is consistent between them, somehow09:50
johnthetubaguytdurakov: its a massive cost, but its one Nova is having to deal with for every other feature (there are several examples of us screwing that up, but thats no excuse to start doing it on purpose)09:51
mikalsahid: I'd like to see a sentence or two about how the implementation works in the proposed change as well. I think that's something that operators will be looking for when they go to use the feature.09:51
mikalsahid: but I like line 78 too, its one of my favourite lines09:52
johnthetubaguysahid: I wondered if we need an image property to detect if the guest agent is installed or not, certainly thats what we do in the XenAPI driver for that feature09:52
sahidmikal: ahaha ok let me work on this, thanks09:52
sahidjohnthetubaguy: well libvirt will complaint about it if the guest-agent is not running09:53
tdurakovjohnthetubaguy: as for me it's not the problem. We should not add any transformations there. Currently op could check it form hypervisor. If we add such transformation operator would go to docs find table that describe which feature is it. It wouldn't be clear09:53
*** moshele has joined #openstack-nova09:53
sahidso we will raise the exception to the user when he will do the request09:53
openstackgerritAnkit Agrawal proposed openstack/nova: Detach volume after deleting instance with no host  https://review.openstack.org/19406309:53
sahidjohnthetubaguy: not sure it is necessary really09:53
mikalsahid: thanks, I like this feature and I want to approve the spec, I think it just needs to explain some stuff a little more (and perhaps allow the admin username to be configured by an image property)09:53
tdurakovespecially if there is lot of features,  it will become another hell to map them with native values09:54
*** gmann_ has joined #openstack-nova09:54
johnthetubaguysahid: do you have to wait for the guest agent to respond with the password you are setting? or is it quick to detect if the guest agent is present? we found it delayed the VM boot if that was not optional via an image property09:54
*** gmann has quit IRC09:55
johnthetubaguysahid: for example, windows images tend to reboot during their fist boot, due to sysprep, so you end up waiting a long time before you get the password back from the guest agent, in the general case09:55
johnthetubaguysahid: but I can't tell from the spec if thats an issue with the qemu guest agent or not09:56
sahidjohnthetubaguy: hum to be honest i do not know09:56
moshelejohnthetubaguy: hi, regarding https://review.openstack.org/#/c/179577/ are you ok  to change it to regex as danpb suggested?09:57
sahidkashyap: ^^09:57
kashyapsahid: Reading the scroll to get some context09:57
sahidthx09:57
johnthetubaguytdurakov: what do you mean native values? the API user should be thinking about Nova API values when talking to the Nova API, thats the whole point of the API, you don't need to worry about native values, so there is less vendor lockin, and the ecosystem tools work nicely across all hypervisors09:57
sahidabout what johnthetubaguy is asking, if you have any information to give to us09:57
kashyapsahid: I don't want to pretend I've used a whole lot of guest agent, but let me see :-)09:59
kashyapjohnthetubaguy: Guest Agent is a service that runs inside the guest10:00
johnthetubaguysdague: kashyap: honestly, I am quite shocked people want this feature, I was kinda hoping we could remove that API at some point, and just keep the password generation by cloud-init only10:00
johnthetubaguykashyap: right thats how the XenAPI one works too10:00
tdurakovjohnthetubaguy: John,  native - values from hypervisors.  cpu info as for me is totally about hypervisor. if we add anothe data layer that encapsulate native values - we made things more complicated. there is more cons than pros from that approach10:02
*** karimb has quit IRC10:02
sudiptojohnthetubaguy, danpb : Hi, sorry for asking on a busy day, do you think this one is not ready for a merge: https://review.openstack.org/#/c/180983/ ?10:02
sahidjohnthetubaguy: your question is about to know if it makes time for qemu to inform that the agent is not running when a request is sent, right?10:02
*** Longgeek has quit IRC10:04
mikaljohnthetubaguy: I am reading LP right and https://blueprints.launchpad.net/nova/+spec/nova-driver-local-file is unapproved, yes?10:05
johnthetubaguymikal: correct10:05
mikaljohnthetubaguy: I ask because I just -2'ed https://review.openstack.org/#/c/134134/ for not having an approved BP10:06
johnthetubaguymikal: sounds good, I was going to re-run the script for that sometime this week10:06
mikalThere's a script?!?10:06
*** jogo has quit IRC10:06
johnthetubaguyyeah10:06
mikalTo add -2's?10:06
mikalThat's cool10:06
mikalIn an we've-automated-being-difficult kind of way10:07
johnthetubaguyone of jogo 's to list out the reviews for blueprints were the blueprint is not approved10:07
kashyapjohnthetubaguy: sahid: Let me quickly test locally, with direct libvirt's qemu-guest-agent command10:07
mikaljohnthetubaguy: in other news, we need smoking jackets for mid-cycle. Its totally a thing we need.10:07
johnthetubaguymikal: i did a hack to open each review in a browser to check it, but yeah10:07
johnthetubaguymikal: lol10:07
mikaljohnthetubaguy: you know you want a smoking jacket... You know you do.10:07
mikaljohnthetubaguy: small hitch, they're $400 USD on Amazon10:08
johnthetubaguymikal: Nova, sponsored by Harris Tweed, that would be very funny10:08
mikaljohnthetubaguy: so we'll have to expense them,10:08
avishayHi all. Do I need /var/lib/nova/instances to be on shared storage if I used only boot-from-volume, and have libvirt transfer the xml files?10:08
johnthetubaguymikal: https://github.com/jogo/gerrit-bp-lp10:09
kashyapavishay: I noticed your earlier question and was trying to recall if I did such a test, but I'm not _quite_ sure of it10:09
avishaykashyap: i know that not having it will affect my console logs, and i saw that there are file locks but not sure if they are relevant10:10
openstackgerritPetrut Lucian proposed openstack/nova: Hyper-V: Fix virtual hard disk detach  https://review.openstack.org/19407010:11
avishaykashyap: my test worked, but that doesn't mean i covered every scenario :)10:11
kashyapavishay: True. There an umpteen no. boot-from-$blah that's possible10:11
danpbsudipto: please stop pinging me every time you upload a new copy of your spec - it is really unhelpful as I have far more to look at than just that one10:11
danpbjohnthetubaguy: the use case for the set-password command is different from that satisfied by the  cloud-init boot time password10:13
avishaykashyap: i should never have qcows there or things like that, not worried about those.  the locks worry me more.  it would be nice if this requirement could be officially relaxed for such environments - HA NFS is a pain10:13
kashyapavishay: I don't have a setup handy, but your question reminds me to set that up, which I've been meaning to for a while.10:13
danpbits for dealing with the case that the admin (for whatever reason) doesn't remember/know the password for existing running instances10:13
johnthetubaguydanpb: agreed10:13
danpbi could see this happening if, for example, you had just sacked your sysadmin and want to re-gain control of your cloud instances10:13
johnthetubaguydanpb: so it wasn't clear to me that it was only implementing that API, I probably miss-read the spec10:13
*** dimsum__ has joined #openstack-nova10:14
danpbits essentially about getting parity with xenapi for that set-password api10:14
mikaldanpb: it happens all the time in public clouds where the root password is assigned to the instance, displayed once, and then not stored by the cloud provider at all10:14
danpbwe were very annoyed that there was somewhere that XenAPI had more features that libvirt ;-P10:15
mikaldanpb: people click the thing, and then have regrets later about not recording the password10:15
*** nihilifer has joined #openstack-nova10:15
danpbmikal: yeah i can imagine actually10:15
mikaldanpb: so no firings required, although we could still fire someone if we thought it was fun10:15
ndipanovmikal, danpb johnthetubaguy while you're reviewing specs:10:17
ndipanovhttps://review.openstack.org/#/c/191226/10:17
ndipanovhas a +1 from jay the sched lead10:17
sudiptodanpb, sorry about that.10:18
ndipanovand https://review.openstack.org/#/c/193576/10:18
ndipanovdansmith, seems to be onboard10:18
mikalndipanov: I am only reviewing funny specs cause its after work hours10:18
mikalndipanov: is your spec amusing in some way?10:18
ndipanovhow funny do you find horrible bugs that have been in the codebase for a long time10:18
ndipanovI personally find them hilarious :)10:19
mikalndipanov: oh, it depends on if I made them or not...10:19
johnthetubaguyndipanov: I am attempting to sort out liberty-1 instead of just doing spec reviews today, I should get back on that!10:19
mikalndipanov: ones I didn't make? Hillarious!10:19
ndipanovmikal, I think these would be very funny then :)10:19
mikalndipanov: I shall read then!10:20
ndipanovthe latter one being more horrible and thus amusing...10:20
openstackgerritChung Chih, Hung proposed openstack/nova: Use monotonic.monotonic and stopwatches instead of time.time  https://review.openstack.org/19407310:20
kashyapavishay: Why do you say "i should never have qcows there"? Perf reasons? If so, if you create a disk image with `preallocation=falloc`, it's quite the same as raw. Maybe you already are aware of it.10:21
mikalndipanov: so, why does this need a spec though?10:22
mikalndipanov: this looks like internal work which isn't contentious10:22
mikalndipanov: so shouldn't it just go for trivial approval?10:22
ndipanovwhich one?10:23
ndipanovmikal, ^10:23
mikalndipanov: https://review.openstack.org/#/c/19122610:23
*** avishay_ has joined #openstack-nova10:23
mikalndipanov: "Introduce lightwieght transactional model for HostState"10:23
*** avishay has quit IRC10:23
ndipanovmikal, that one yeah - but thought there might be discussion around10:23
ndipanovit10:24
ndipanovthe other one has some DB changes10:24
mikalndipanov: oh, I think the documentation is good. I'm just not sure that we would have required it if you'd said all that stuff to me in person.10:24
mikalndipanov: either way, I +2'ed it10:24
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py  https://review.openstack.org/18778510:24
ndipanovthanks10:24
ndipanovthe other one is more interesting really10:25
*** ociuhandu has quit IRC10:25
*** jamielennox is now known as jamielennox|away10:25
mikalndipanov: ok, dansmith wants Jay to comment on the second one, so it might be worth pinging him when he wakes up as well.10:25
ndipanovyeah10:26
openstackgerritJohn Garbutt proposed openstack/nova: Make live migration create a migration object record  https://review.openstack.org/18333110:26
*** hshiina has quit IRC10:27
openstackgerritTimofey Durakov proposed openstack/nova-specs: nova-api hypervisor.cpu_info change from string to json object  https://review.openstack.org/18305710:27
avishay_kashyap: thanks anyway :)10:27
*** karimb has joined #openstack-nova10:27
*** nihilifer has quit IRC10:33
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py  https://review.openstack.org/18778510:33
*** hshiina has joined #openstack-nova10:34
openstackgerritsahid proposed openstack/nova-specs: libvirt: set admin root password  https://review.openstack.org/18750910:34
*** Longgeek has joined #openstack-nova10:34
*** nihilifer has joined #openstack-nova10:37
*** karmatronic has joined #openstack-nova10:38
mikaldanpb: thanks for commenting on that libosinfo review10:39
*** karimb has quit IRC10:39
*** davideagnello has joined #openstack-nova10:39
*** karmatronic has quit IRC10:39
*** karimb has joined #openstack-nova10:39
*** davideagnello has quit IRC10:43
*** irenab has quit IRC10:43
*** kiran-r has quit IRC10:43
*** n0ano has quit IRC10:43
*** sdake has quit IRC10:43
*** gabriel-bezerra has quit IRC10:43
*** clif_h has quit IRC10:43
*** kcaj has quit IRC10:43
*** scottda has quit IRC10:43
*** dguitarbite has quit IRC10:43
*** Vek has quit IRC10:43
*** lbragstad has quit IRC10:43
*** masayukig has quit IRC10:43
*** markmcclain has quit IRC10:43
*** freyes has quit IRC10:43
*** wasmum has quit IRC10:43
*** _d34dh0r53_ has quit IRC10:43
*** ttx has quit IRC10:43
*** Guest11697 has quit IRC10:43
*** adam_g has quit IRC10:43
*** apmelton has quit IRC10:43
*** gus has quit IRC10:43
*** zigo has quit IRC10:43
*** toan has quit IRC10:43
*** jroll has quit IRC10:43
*** comstud has quit IRC10:43
*** jbernard has quit IRC10:43
*** apmelton has joined #openstack-nova10:43
*** ttx has joined #openstack-nova10:43
*** Vek has joined #openstack-nova10:43
*** comstud has joined #openstack-nova10:43
*** gus has joined #openstack-nova10:43
*** masayukig has joined #openstack-nova10:43
*** n0ano has joined #openstack-nova10:43
*** adam_g has joined #openstack-nova10:43
*** kiran-r has joined #openstack-nova10:43
*** clif_h has joined #openstack-nova10:43
*** markmcclain has joined #openstack-nova10:43
*** dguitarbite has joined #openstack-nova10:43
*** jroll has joined #openstack-nova10:43
*** mgagne has joined #openstack-nova10:43
*** romainh has quit IRC10:43
*** mgagne is now known as Guest8120210:43
*** adam_g has quit IRC10:43
*** adam_g has joined #openstack-nova10:43
*** romainh has joined #openstack-nova10:43
*** kcaj has joined #openstack-nova10:44
*** lbragstad has joined #openstack-nova10:44
*** freyes has joined #openstack-nova10:44
*** toan has joined #openstack-nova10:44
*** zigo has joined #openstack-nova10:44
*** irenab has joined #openstack-nova10:44
*** sdake has joined #openstack-nova10:44
*** d34dh0r53 has joined #openstack-nova10:45
*** wasmum has joined #openstack-nova10:45
garykjohnthetubaguy: not sure if you saw but i addressed the comments posted on https://review.openstack.org/192675. i hope that you and danpb can help get it through prior to the deadline.10:46
*** scottda has joined #openstack-nova10:47
*** dguitarbite has quit IRC10:47
*** dimsum__ is now known as dims10:48
*** gabriel-bezerra has joined #openstack-nova10:48
openstackgerritClaudiu Belu proposed openstack/nova-specs: Adds Hyper-V UEFI Secure Boot spec  https://review.openstack.org/19099710:51
*** dims has quit IRC10:53
*** markvoelker has joined #openstack-nova10:53
*** subscope has joined #openstack-nova10:56
*** markvoelker has quit IRC10:58
*** oro has joined #openstack-nova10:59
*** romainh has quit IRC10:59
openstackgerritPeter Savage proposed openstack/python-novaclient: Added marker functionality to flavours and images  https://review.openstack.org/19321211:01
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests  https://review.openstack.org/13979611:02
*** exploreshaifali has quit IRC11:02
claudiubjohnthetubaguy: hi. I've proposed this blueprint as per your suggestion: https://blueprints.launchpad.net/nova/+spec/hyper-v-test-refactor11:03
*** IanGovett has joined #openstack-nova11:04
*** linkedinyou has quit IRC11:04
*** dims has joined #openstack-nova11:05
*** nihilifer has quit IRC11:05
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests  https://review.openstack.org/13979711:06
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests  https://review.openstack.org/14898011:07
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: removes test_hypervapi tests  https://review.openstack.org/13979811:08
mdboothIs anybody here aware of a parser for the error output of 'testr list-tests' ?11:08
mdboothOr at least a description of what it is?11:08
*** otter768 has joined #openstack-nova11:09
*** deepthi has quit IRC11:12
*** otter768 has quit IRC11:14
openstackgerritVineet Menon proposed openstack/nova-specs: Add flavor tables to API database  https://review.openstack.org/19014711:14
*** ZZelle has joined #openstack-nova11:14
*** nihilifer has joined #openstack-nova11:17
*** Longgeek has quit IRC11:18
snikitinjohnthetubaguy: Hi John, sorry to trouble, but could you please look at my instance tags spec? https://review.openstack.org/#/c/177112/ It has six +1 and you voted +2 some time ago. I added one comment and lost all votes :( It was approved for Juno and Kilo. I just need reapprove to continue my work. I really don't want to move it to the next release.. again11:20
*** alex_klimov has joined #openstack-nova11:21
*** Marga_ has quit IRC11:21
*** raorn_ is now known as raorn11:22
*** sdake_ has joined #openstack-nova11:27
*** romainh has joined #openstack-nova11:27
*** jlvillal has quit IRC11:28
*** ttx has quit IRC11:28
*** ttx has joined #openstack-nova11:28
rajeshthi11:28
rajeshtcan anyone tell me why migration record is not created during live migrating an instance ?11:28
*** sdake has quit IRC11:30
johnthetubaguyrajesht: its a bug, I think dansmith is taking a look at that right now11:31
*** dguitarbite has joined #openstack-nova11:31
rajeshtjohnthetubaguy: is he working on it ?11:31
johnthetubaguysnikitin: its on my list, just having to do some blueprint things for liberty-1 today I am afraid11:31
rajeshtjohnthetubaguy: do you have any reference of that bug ? if there is any11:32
johnthetubaguyrajesht: i think so, see robustify-evacuate spec, I think there is something up for review11:32
rajeshtjohnthetubaguy: thanks :)11:32
johnthetubaguyrajesht: its basically because live-migrate got added in libvirt only, and migrate was used for xenapi, and the live-migrate never reached parity yet :(11:33
rajeshtjohnthetubaguy: thanks you very much john :)11:33
johnthetubaguyno problem11:34
*** IanGovett has quit IRC11:34
* mikal gives up for the night.11:34
johnthetubaguyclaudiub: what was the old one called again?11:34
*** IanGovett has joined #openstack-nova11:35
*** matrohon has joined #openstack-nova11:35
openstackgerritJohn Garbutt proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests  https://review.openstack.org/13979611:36
openstackgerritJohn Garbutt proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests  https://review.openstack.org/13979711:37
*** moshele has quit IRC11:37
openstackgerritJohn Garbutt proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests  https://review.openstack.org/14898011:37
openstackgerritAdelina Tuvenie proposed openstack/nova: Fixes "Hyper-V destroy vm fails on Windows Server 2008R2"  https://review.openstack.org/19408911:38
johnthetubaguyclaudiub: I updated the blueprint name and those patches, and removed the -211:38
openstackgerritJohn Garbutt proposed openstack/nova: Hyper-V: removes test_hypervapi tests  https://review.openstack.org/13979811:39
*** Marga_ has joined #openstack-nova11:40
*** matrohon has quit IRC11:40
*** Marga_ has quit IRC11:41
*** foexle has quit IRC11:41
*** Marga_ has joined #openstack-nova11:41
*** marzif_ has joined #openstack-nova11:42
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py  https://review.openstack.org/18778511:42
openstackgerritJohn Garbutt proposed openstack/nova: Make evacuate leave a record for the source compute host to process  https://review.openstack.org/18335411:44
*** lucasagomes is now known as lucas-hungry11:45
johnthetubaguydansmith: this one looks stuck in an odd way: https://review.openstack.org/#/c/188112/11:46
openstackgerritJohn Garbutt proposed openstack/nova: Make readonly field tests use exception from oslo.versionedobjects  https://review.openstack.org/18811211:46
garykjohnthetubaguy: hi, regarding https://review.openstack.org/#/c/147126/, at the summit i was asked to open a BP. i have done that. this is something that is really blocking - the reason is that the neutron plugin in K does not work without this. what do you suggest.11:46
*** Longgeek has joined #openstack-nova11:46
*** mwagner_afk has quit IRC11:47
johnthetubaguygaryk: it hasn't been submitted for approval / review in the nova meeting, so we can't merge the code yet, all blueprints should be approved before the associated code gets merged, when that doesn't happen, its in error, not because its allowed11:47
garykwhen is the next meeting and how can we try and get this approved?11:48
garykteh code has been in review for over 6 months11:49
*** thorst has joined #openstack-nova11:49
johnthetubaguygaryk: the next meeting is on Thursday, as normal, please add it to the agenda there, as with any blueprint that doesn't need a spec11:50
garykjohnthetubaguy: ok, will do11:50
garykso if bps and specs need to be reviewed by the 23rd this means that the chance of getting this in is highly unlikely11:51
garykthe next meeting is on the 25th11:51
johnthetubaguygaryk: deadline is 25th11:51
garykk - wiki has it written as 23rd11:52
johnthetubaguygaryk: ah, so I should update that11:52
openstackgerritsahid proposed openstack/nova-specs: libvirt: set admin root password  https://review.openstack.org/18750911:52
johnthetubaguygaryk: not quite sure what the deadline will mean yet, I have some ideas to try and relax some of this11:52
garykok, thanks. now i am not sure about something11:54
openstackgerritJohn Garbutt proposed openstack/nova-specs: Add rootwrap-daemon-mode blueprint  https://review.openstack.org/10540411:54
*** markvoelker has joined #openstack-nova11:54
garykfor the vmware limits is is under bps with no specs but last week I was asked to draft a spec.11:54
garyki have done that and would hope to maybe get some reviews for this parity feature11:55
johnthetubaguygaryk: we should remove that from the agenda I guess11:56
johnthetubaguygaryk: just to be clear, there are about 100 specs in review right now11:56
garykjohnthetubaguy: i understand. but 3 or 4 people cannot review all of this stuff…11:56
garykthat is just impractical and illogical11:57
*** willemgf has joined #openstack-nova11:57
garykhow big is the driver team these days?11:57
*** markvoelker has quit IRC11:59
johnthetubaguygaryk: agreed its not working, I suggested the alternative at the summit, and it got voted down, but people have not stepped up to review enough, so I plan to ask the same question again, in the hope there are better ideas, and if not, I will have to try out the original plan11:59
*** dprince has joined #openstack-nova12:00
garyki understand and i know what you are trying to do. i too hope that there are better ideas. problem is the community is growing and the reviews are dwindling. so we are between a rock and a hard place.12:00
fricklerhas anyone else seen issues with Virtuozzo Storage CI? Left a failure on https://review.openstack.org/179569 and the link to the logs gives me a 40412:00
garykcoupled with that we have people actually using this stuff in production and need to make sure that we can deliver a whole working solution. which is becoming practiically impossible - when one is able to get something approved in neutron i K and it will not land in Nova in L.12:01
*** doron_afk has quit IRC12:02
willemgfHI, We have 3 controllers on which glance is active. On our compute nodes, in nova.conf those 3 glance servers are configured. We had a situation in which we brought down 2 of these 3 glance-services, and could see that the compute nodes will still make use of those glance-servers that are down, and finally endup with the error "nova.image.glance CommunicationError: Error finding address ...". Is there a mechanism foreseen that nova will12:03
willemgfonly use the active glance-nodes?12:03
*** sudipto has quit IRC12:03
*** markvoelker has joined #openstack-nova12:03
samueldmqhi, good morning !12:04
samueldmqwhat is the state of microversions on nova ? I saw there was a spec targeted to Kilo12:05
samueldmqI am working on the dynamic policies effort (https://wiki.openstack.org/wiki/DynamicPolicies) driven by keystone12:05
samueldmqand I am interested on how would microversions be represented on the policy files12:05
*** doron_afk has joined #openstack-nova12:07
*** pkoniszewski has quit IRC12:07
garyksamueldmq: i suggest that you speak with sdague or alex_xu_ they are the guys most familiar with this12:09
openstackgerritPetrut Lucian proposed openstack/nova: Hyper-V: Fix virtual hard disk detach  https://review.openstack.org/19407012:09
*** VW_ has joined #openstack-nova12:09
sdaguesamueldmq: so, honestly, I feel like all the policy efforts are stalled right now because we don't have the ability to pull that into code in Nova, which would let us have base policy evolve with microversions12:10
*** gszasz has quit IRC12:10
*** doron_ has joined #openstack-nova12:10
*** raildo has joined #openstack-nova12:10
samueldmqsdague: hi, so .. what did we have two weeks ago and what we have now (at least on keystone side)12:11
*** doron_afk has quit IRC12:11
samueldmqsdague: when we had those messages in the ML, we noticed we still needed to organize our spes and have agreement on keystone internally12:11
johnthetubaguysdague: did we get our policy plans drafted yet, I don't remember now, have a feeling there was a review somewhere?12:11
samueldmqsdague: (I am talking about dynamic policies in general)12:11
samueldmqsdague: so now we have this wiki: https://wiki.openstack.org/wiki/DynamicPolicies12:12
samueldmqsdague: that defines the general idea, and presents some sequence diagrams12:12
sdaguejohnthetubaguy: we did not, because there was still an impass on the first thing we needed which is seed of truth from nova, and the encoding of base policy in code12:12
sdaguesamueldmq: so, that seems to continue to ignore those issues12:12
samueldmqnow I want to work with you guys to see how microversions would look like in the policy, so that may affect our caching strategy when dynamically fetching, etc12:12
samueldmqsdague: ^12:13
sdaguehttps://wiki.openstack.org/w/images/4/41/Dynamic-policies-install.png12:13
sdaguecompletely hand waves seed of truth discovery12:13
samueldmqsdague: no wait .. the way we initially planned it was : the initial source of truth is the policy.json in nova's repository12:14
samueldmqsdague: that's stored and 'Stock policy for nova' in keystone12:14
sdaguesamueldmq: right, and that's a disaster waiting to happen, so you shouldn't do that12:14
samueldmqsdague: when an admin changes form this defualt, he/she gets warnings12:14
*** moshele has joined #openstack-nova12:14
samueldmqsdague: which was one of your usecases12:14
sdaguebecause you'll be digging out of that for a long time12:14
samueldmqsdague: initial source of truth will be uploaded from services to keystone12:15
samueldmqsdague: for now (at least this cycle) I think CMS uploading that from the existing policy.json files are the right scope12:16
sdaguesamueldmq: I fundamentally don't12:16
sdagueI think that's a dead end12:16
samueldmqsdague: we can decide if we go to implement /policy on services later12:16
sdaguethat only generates more work12:16
*** mmedvede has joined #openstack-nova12:16
samueldmqsdague: why? we're not going to unify anymore in this cycle (will discuss that later)12:17
sdagueDefining /policy is the step 1 in all of this12:17
samueldmqsdague: I am trying to get teh right use cases so that we can decide the scoep for L12:17
*** Nikolay_St has quit IRC12:17
sdagueright, and I think the right scope for L is get /policy as a contract put out there to the services, so that policy can be refreshed, on demand from keystone to the services12:18
sdagueso we seem to fundamentally disagree about point 112:18
*** amotoki has joined #openstack-nova12:19
sdagueand without /policy contract, it's not even possible to talk about microversion implications in any sane way12:19
samueldmqsdague: oh, I am expecting that when you upgrade a service (new APIs may appear)12:19
samueldmqsdague: so that CMS would re-upload it to keystone12:19
sdaguewhat is CMS?12:20
samueldmqsdague: we're using CMS to upload/update source of truth on keysotne12:20
samueldmqsdague: Configuration Management System12:20
samueldmqsdague: puppet, et c12:20
sdaguesamueldmq: right, and I think that is the *wrong* architecture12:20
sdaguelike fundamentally flawed12:20
*** pask81 has joined #openstack-nova12:20
sdagueand a ton of extra work12:20
samueldmqsdague: I think that fits in the scope for L12:21
sdaguethat will all be deleted in 2 cycles12:21
samueldmqsdague: we can't implement /policy this cycle :(12:21
sdaguesamueldmq: right, so, I think it's a waste of time.12:21
sdaguesamueldmq: why not?12:21
samueldmqsdague: what says we cannot implement /policy next cycle ? once we have everything in place ?12:21
sdaguesamueldmq: right, so what I'm saying is there is no reasonable conversation about having anything coherent with nova microversions and policies without /policy12:22
sdagueso without that, it's just a giant unmanagable mess12:22
sdagueif you would like to make a bigger unmanagable mess this cycle instead of starting on this other piece of work, I guess that's fine12:22
*** edmondsw has joined #openstack-nova12:22
samueldmqsdague: ok what I need it to really understand microversions, so I will understand what you're saying12:22
samueldmqsdague: I don't have the same background, so not enough arguments to agree/disagree with you now12:22
sdagueright, which is my frustration of the keystone folks going off and making a plan without taking into account project needs here12:23
*** ivasev has joined #openstack-nova12:23
samueldmqsdague: no, now is the time I am reaching you guys12:23
sdagueand why I tried to get ahead of it with the ML thread so we could come up with something that made this all work12:23
samueldmqsdague: we needed to have an internal agreement before12:23
samueldmqsdague: then we agree with you in this iteration12:23
sdaguesamueldmq: so that seems weird, if what the projects need weren't taken into account in the first iteration12:24
samueldmqsdague: I appreaciate your efforts, and I think you should really keep working with us12:24
samueldmqsdague: there are some fundamental parts, like fetching etc, that won't change12:24
sdaguesamueldmq: they will fundamentally change12:24
samueldmqsdague: we can fit your requirements starting from that12:24
sdaguehow will fetching not change?12:25
samueldmqsdague: I mean we will need fetching anyway, we defined the overview structure of the solution12:25
*** romainh has left #openstack-nova12:25
samueldmqsdague: caching strategy will change depending on microversions, I think12:25
samueldmqsdague: that's why I need to udnerstand more about microversions, and what are your requirements12:25
johnthetubaguygaryk: another unapproved BP with code here: https://blueprints.launchpad.net/nova/+spec/vmware-swap-support12:26
sdaguesamueldmq: so, honestly, I feel like there is nothing productive to be done here before defining the /policy contract. And there is a lot of other Nova work to do. So if we're talking about the /policy contract I'm happy to participate, but the rest of it just has seemed like circular conversations of "we need this" "no you don't" "yes, we really do"12:27
samueldmqsdague: sure, but I need to understand why /policy is fundamental from the start, that's the point12:27
sdaguesamueldmq: right, so I feel like I spent that week trying to explain it to folks, and I'm not thrilled about starting that all over again from scratch.12:28
sdaguefor background on microversions in general, have you read this yet - https://dague.net/2015/06/05/the-nova-api-in-kilo-and-beyond-2/ ?12:29
samueldmqsdague: is that the fact people may be running on master, and as soon as a new API is added, it would be reflected on keystone side ?12:29
sdaguesamueldmq: yes12:29
garykjohnthetubaguy: ok, thanks.12:29
samueldmqsdague: and doing that through a CMS would be very expensive12:29
*** mestery has joined #openstack-nova12:29
garykhonestly i am not sure how importnat that is, but the code has been in review for donkeys years12:29
sdaguesamueldmq: not that it's expensive, it's error prone12:30
samueldmqsdague: a lot of wasted effort to then implemnet /policy12:30
sdaguepip install nova doesn't update /etc/nova/policy.json12:30
sdaguewe need base policy to be added and exist the second new code is deployed12:30
sdaguewith sane defaults12:30
openstackgerritPhilip Marc Schwartz proposed openstack/nova: Add flag to force experimental run of db contract  https://review.openstack.org/19289412:30
sdaguethe policy.json files are config files, operators can change them12:31
sdagueand need time to merge their contents12:31
samueldmqsdague: so nova registers itself in keystone once it gets installed/upgraded?12:31
sdagueso we need a base in code which starts as sane defaults so as new things deploy the policy is bound to the api, in the same way that database migrations are12:31
samueldmqsdague: via /policy ?12:31
sdaguesamueldmq: no12:31
samueldmqsdague: how is that triggered ?12:32
johnthetubaguygaryk: same with this blueprint: https://blueprints.launchpad.net/nova/+spec/enable-volume-types12:32
johnthetubaguygaryk: although that looks like it needs a spec12:32
sdaguekeystone calls back to nova /policy on some cache timeline, which can be flushed by the admin12:32
dimsjohnthetubaguy: https://blueprints.launchpad.net/nova/+spec/oslo-config-generator was discussed on Thu meeting there was no opposition (http://eavesdrop.openstack.org/meetings/nova/2015/nova.2015-06-18-20.59.log.html#l-228)12:32
*** tobasco_ is now known as tobasco12:32
sdaguebut /policy is as much about coupling policy to api as db migrations are12:32
sdaguewe need the base out of config files, like yesterday12:33
samueldmqsdague: I got your use case, just trying to see the sequence in there12:33
johnthetubaguydims: cool, that should have been approved then, I will read though that and make sure the paperwork was all done properly12:33
samueldmqsdague: hmm.. so keystone asks endpoints for their /policy, to update them12:33
dimsThanks johnthetubaguy12:33
*** aysyanne has joined #openstack-nova12:33
*** mwagner_afk has joined #openstack-nova12:33
sdaguesamueldmq: right, keystone pulls from the endpoints directly12:34
sdaguewhich is the way components are supposed to interact, over REST apis12:34
samueldmqsdague: ok .. makes sense. besides that, another disagreement point from what is in there ?12:34
sdaguenot over random config files being shuffled around12:34
johnthetubaguydims: thanks for the ping, that should be cleared up now I think12:36
dimsnice, appreciate it johnthetubaguy12:36
garykjohnthetubaguy: that one has a spec proposed. let me try and dig that up12:36
*** rfolco has joined #openstack-nova12:36
*** Marga_ has quit IRC12:36
johnthetubaguygaryk: it probably needs the blueprint xyz tag in the commit12:36
ndipanovjohnthetubaguy, https://review.openstack.org/#/c/84048/12:37
*** alexpilotti has joined #openstack-nova12:37
sdaguesamueldmq: https://wiki.openstack.org/w/images/f/f4/Dynamic-policies-enforce.png - I really don't think we should be using a file cache for the fetched policies, there is no reason for disk persistence, it can all be in memory, and less likely to get corrupted12:38
*** sudipto has joined #openstack-nova12:38
sdagueI also think the whole override calculation is going to be impacted by base starting in code12:38
sdaguewhich is why, /policy should be the work for this cycle12:38
andrearosandipanov: johnthetubaguy: I have an idea for 8404812:38
johnthetubaguyndipanov: so I see that spec as make the best of a bad situation, that interface needs a lot more thought, but we also need to unblock our users at some point, just seems like a way forward, but I might be missing something12:38
garykjohnthetubaguy: this si the spec - https://review.openstack.org/#/c/187812/12:38
johnthetubaguyandrearosa: interested12:38
andrearosaprobably we do not need a spec if we consider that a bug.12:38
sdaguein my opinion12:38
samueldmqsdague: I agree with you, and what's there is not the final solution12:39
samueldmqsdague: I designed that to be reached this cycle, so we still writes in the fiels, so things keep working as today12:39
andrearosaI am going to comment on the spec with an alternative solution12:39
ndipanovjohnthetubaguy, sure... let's go with just crappy instead of insanely crappy12:39
ndipanovby all means12:39
samueldmqsdague: we plan to get ride of files, and middleware pass the retrieved policy to the service somehow12:39
andrearosaI am just busy on a different task but I will comment in the the next few hours12:39
ndipanovit's just sad that there is no way to get to not-crappy that I can think of :(12:39
johnthetubaguygaryk: looks like its not considered in scope for nova at this point12:40
andrearosandipanov: just saw your message, probably there is a lightn at the end of the tunnel :)12:40
samueldmqsdague: we need ot keep Defautl policy (source of trught) + Custom policy (what admin changes) separate, and then do the overrid12:40
*** rakesh_hs has joined #openstack-nova12:40
samueldmqsdague: otherwise we would loose custom changes once the Default policy is upgraded12:41
samueldmqsdague: makes sense?12:41
johnthetubaguygaryk: well, maybe not consensus, but there are some issues that need resolving in the spec at least12:41
garykok, understood12:42
sdaguesamueldmq: right, so what I'm saying is this is rushing to get the surface in keystone without getting the plumbing right12:43
sdaguewhich means that now you'll have to deprecate all the plumbing, and it will take twice as long to get to a final end state12:43
*** lyarwood has quit IRC12:44
sdagueand if instead in liberty the plumbing was done. The oslo.policy changes to support code + config overrides. Easy infrastructure to make a /policy to make that discoverable.12:44
samueldmqsdague: k so we've a kind of agreement on the general steps, and basically if we had /policy + warnings on changing from default, we'd be good12:44
*** lyarwood has joined #openstack-nova12:44
sdagueYou'd set up all the infrastructure in the projects so that the surface bits could be done well in the next cycle for keystone12:44
*** Longgeek has quit IRC12:47
samueldmqsdague: sure, I understand your points12:47
samueldmqsdague: what do you think about a meeting specific for this dynamic policy stuff ?12:48
samueldmqsdague: also, we could talk about it in the keystone meeting tomorrow, if you're ok with it12:48
sdaguesamueldmq: when is the keystone meeting?12:48
samueldmqsdague: 18 UTC12:48
*** rod has joined #openstack-nova12:49
sdaguethat probably works12:49
sdaguebut, if I have to explain everything from scratch again, I might rage quit :)12:49
samueldmqsdague: nice, it is the one just before infra meeting12:49
samueldmqsdague: I am not sure all the people there are synchronized with this policy stuff12:50
samueldmqsdague: I will try to get a background understanding as better as I can12:50
rodhi12:50
samueldmqsdague: I've needed to explaing those dynamic policies from the start over and over for different people12:50
sdagueright, and that's my frustration, I tried to do this on the mailing list, so we could all be on the same page. Then that devolved, then keystone made decisions separate from that, not on the ML, which means it was hidden in a corner12:50
rodI got a problem when I try to reach nova vnc console througn reverse proxy12:50
sdaguesamueldmq: right we should stop that. Do this on the ML so we aren't having a million point conversations12:51
samueldmqsdague: no please don't seee this like that12:51
rodthe path variable from vnc_auto.html is undefined12:51
samueldmqsdague: we didn't make decisions12:51
samueldmqsdague: we just have a first proposal12:51
garykjohnthetubaguy: i am following up with zhangi on the spec12:51
johnthetubaguygaryk: OK, thank you12:51
markus_zjohnthetubaguy: Do you think I need a spec for bp libvirt-log-file-and-serial-console ? In the review you mentioned it should be discussed in the next meeting.12:51
samueldmqsdague: we're very open to listen to you, and implement your requirements as well12:51
*** vladikr has joined #openstack-nova12:52
sdaguesamueldmq: ok, good to hear. That was not my take away from that ML thread12:52
samueldmqsdague: we just needed some organization internally, even the specs were a mess12:52
*** bradjones has quit IRC12:52
*** hshiina has quit IRC12:52
samueldmqsdague: I am working hard on organizing these dynamic policy efforts12:52
samueldmqsdague: so yes, we're open, morganfainberg is as well, listening to you guys and taking into account your requirements in on the plan all the time12:53
sdagueok, so I would suggest that primary communication platform for that is the mailing list, where it will reach multiple teams. Because this is a cross project effort12:53
samueldmqsdague: sure! and a cross-project meeting for that, besides the ML, would help a lot IMO12:53
samueldmqsdague: we can try to arrange that in another ML threadh12:54
sdaguehonestly, meetings don't help that much, because only a small number of people are in them12:54
sdagueand they exclude at least half our community by TZS12:54
*** baoli has joined #openstack-nova12:54
*** bradjones has joined #openstack-nova12:54
*** bradjones has quit IRC12:54
*** bradjones has joined #openstack-nova12:54
*** russellb has joined #openstack-nova12:54
samueldmqsdague: we keep using the ML, but I think we need a sync point to make decisions , etc12:55
samueldmqsdague: let's start with the ML, and keystone meeting tomorrow12:55
samueldmqsdague: sounds good  ? I will add a point in the agenda12:55
sdaguesure, lets see what happens after the keystone meeting tomorroe12:55
sdaguegreat12:55
samueldmqsdague: thanks for hearing :) see you tomorrow o/12:56
haypohi. i proposed a patch to make the py34 check job voting for nova: https://review.openstack.org/#/c/194128/12:56
haypoa voting check job will avoid python3 regressions12:56
sdagueyep, looking forward to it12:57
sdaguehaypo: what's the current pass rate on that job?12:58
garyki have never seen so many -2's12:58
haypoi just checked the py34 gate on 3 issues, i didn't see a failure. i hope that all issues with this gate have been fixed12:58
*** otter768 has joined #openstack-nova12:59
hayposdague: sorry, i don't know how to get this information, except of checking random patches to see the CI result12:59
sdaguehaypo: no, I mean the statistical fails. Like what the pass rate over the last week12:59
sdagueuse logstash.openstack.org or graphite.openstack.org12:59
sdaguethat's kind of important in proposing new jos12:59
sdaguejobs12:59
openstackgerritChris Gacsal proposed openstack/nova-specs: Add REST API impact to the multi-attach spec  https://review.openstack.org/19275512:59
*** dave-mccowan has joined #openstack-nova13:02
garykdims: please see my comments on https://review.openstack.org/#/c/187785/13:02
lpetrutjohnthetubaguy: Hi13:02
garyki am not sure why you are not using six.iteritems ...13:02
*** lucas-hungry is now known as lucasagomes13:03
lpetrutjohnthetubaguy: could you please take a look over this BP? https://blueprints.launchpad.net/nova/+spec/hyperv-assisted-volume-snapshot13:03
johnthetubaguylpetrut: sorry, snowed under, probably best to just add that to the agenda for the next nova meeting please?13:04
openstackgerritChris Gacsal proposed openstack/nova-specs: Add REST API impact to the multi-attach spec  https://review.openstack.org/19275513:07
dimsgaryk: thanks, replied13:08
hayposdague: in graphite, i saw 3 failures in the last week. i don't know how to get the url of my graphic :-p13:08
sdaguehaypo: ok13:08
hayposdague: the number of success is ... it's hard to estimate, it varies a lot during a day. it's around 10 checks per day13:08
openstackgerritsahid proposed openstack/nova: libvirt: introduce GuestBlock to wrap around Block API  https://review.openstack.org/19021013:09
openstackgerritsahid proposed openstack/nova: libvirt: update swap volume to use Guest  https://review.openstack.org/19021113:09
openstackgerritsahid proposed openstack/nova: libvirt: return whether a domain is persistent  https://review.openstack.org/19096213:09
openstackgerritsahid proposed openstack/nova: libvirt: make unit tests concise by setup guest object  https://review.openstack.org/19096313:09
openstackgerritsahid proposed openstack/nova: libvirt: introduce method to wait for block device job  https://review.openstack.org/19021413:09
openstackgerritsahid proposed openstack/nova: libvirt: update live snapshot to use Guest object  https://review.openstack.org/19021213:09
openstackgerritsahid proposed openstack/nova: libvirt: update volume delete snapshot to use Guest  https://review.openstack.org/19021313:09
openstackgerritsahid proposed openstack/nova: libvirt: make host return guest information  https://review.openstack.org/19104613:09
openstackgerritsahid proposed openstack/nova: libvirt: make host return power state of guest  https://review.openstack.org/19104713:09
openstackgerritsahid proposed openstack/nova: libvirt: rename GuestVCPUInfo to VCPUInfo  https://review.openstack.org/19413313:09
*** rushiagr is now known as rushiagr_away13:10
*** park_ has joined #openstack-nova13:11
*** rod has quit IRC13:11
openstackgerritClaudiu Belu proposed openstack/nova-specs: Adds Hyper-V Shielded VMs spec  https://review.openstack.org/19289213:11
hayposdague: do you need more stats on the py34 check?13:13
sdaguehaypo: no, that seems reasonable, as long as you will be looking at failures13:13
*** coolsvap is now known as coolsvap|away13:13
garykdims: thanks!13:13
hayposdague: i don't know how to find these failures. any clue?13:14
*** jmccrory has quit IRC13:14
openstackgerritPetrut Lucian proposed openstack/nova: Hyper-V: Fix virtual hard disk detach  https://review.openstack.org/19407013:15
*** erlon has joined #openstack-nova13:15
openstackgerritMoshe Levi proposed openstack/nova-specs: Add spec to enhance PCI passthrough whitelist to support regex  https://review.openstack.org/17957713:15
*** jaypipes has joined #openstack-nova13:15
*** jmccrory has joined #openstack-nova13:15
sdaguehaypo: http://status.openstack.org//elastic-recheck/data/uncategorized.html these are all the uncategorized failures, that's a good starting point13:15
*** Longgeek has joined #openstack-nova13:16
openstackgerritsahid proposed openstack/nova: libvirt: enhance method to return pointer_model from image prop  https://review.openstack.org/17485413:16
openstackgerritsahid proposed openstack/nova: libvirt: deprecate use_usb_tablet in favor of pointer_model  https://review.openstack.org/17624213:16
lpetrutjohnthetubaguy: sure, no problem13:17
*** lyarwood has quit IRC13:17
*** lyarwood has joined #openstack-nova13:17
hayposdague: i cannot find gate-nova-python34 at http://status.openstack.org//elastic-recheck/data/uncategorized.html13:18
sdaguehaypo: so that means it doesn't have an unknown kind of failure13:19
samueldmqsdague: sure, thanks!13:19
samueldmqsdague: oh I was stopped in an old message :-)13:20
sdague:)13:20
*** karimb has quit IRC13:22
*** park_ has quit IRC13:23
sahidbtw jaypipes thank you for all of the reviews you did with the work on libvirt13:23
*** baoli has quit IRC13:24
openstackgerritPetrut Lucian proposed openstack/nova: Hyper-V: Implement nova rescue  https://review.openstack.org/12715913:24
*** moshele has quit IRC13:24
*** baoli_ has joined #openstack-nova13:26
hayposdague: i'm now looking for >project:"openstack/nova" AND build_status:"FAILURE" AND build_name:"gate-nova-python34" AND NOT build_change:"187785"< at http://logstash.openstack.org/ and i see much more failures!13:26
*** _gryf is now known as _gryf|afk13:26
hayposdague: i checked a few of them, but none are related to tox or py34. ex: "[PostBuildScript] - [Error] - Problems occurs: java.lang.InterruptedException" while uploading results to swift13:26
mriedem1haypo: the py34 job doesn't run in the gate queue13:28
mriedem1the uncategorized bugs page is only for gate queue jobs13:28
haypomriedem1: yes, that's a deliberate choice13:28
mriedem1haypo: right, so you'd just have to search for failures on that job in logstash13:28
haypomriedem1: oh :) so http://status.openstack.org//elastic-recheck/data/uncategorized.html is not very useful :)13:28
sdaguehaypo: so the following is needed to get only 1 hit per job - project:"openstack/nova" AND build_status:"FAILURE" AND build_name:"gate-nova-python34" AND NOT build_change:"187785" AND message:"Finished: FAILURE"13:28
mriedem1haypo: right13:29
sdague7 hits this week13:29
jaypipessahid: no probs. will do another round of reviews this morning. saw you pushed a bunch of new revisions...13:29
bauzasjaypipes: heya13:29
sahidyep13:29
* bauzas just hits someone when he sees him13:29
jaypipesbauzas: hey. want to chat about your select destination host bp?13:30
bauzasjaypipes: oh, you are able to look at my mind :)13:30
sahidthen will probably push a etherpad or something like that with a TODO list of what we have to achieve13:30
bauzasI mean, within my mind :)13:30
*** mestery has quit IRC13:30
jaypipessahid: sounds good.13:30
bauzasjaypipes: so, yeah, it seems we need to sync13:30
jaypipesbauzas: hangout?13:31
bauzasjaypipes: before discussing, I replied on 9.30am13:31
jaypipesbauzas: yeah, I read that already :)13:31
bauzasjaypipes: so by rewiewing my last comment, it sounded harsh13:31
bauzasjaypipes: apologies if you thought about that, it was not my view13:31
jaypipesheh, nah, no worries :)13:31
bauzasjaypipes: so, sure we can hangout13:31
bauzasnow ?13:31
*** veena has quit IRC13:32
jaypipesbauzas: gimme 15 to caffeine myself.13:32
*** kiran-r has quit IRC13:32
bauzasjaypipes: sure thing, me too13:32
*** abhishekk has quit IRC13:32
bauzasjaypipes: ping me when you're done then :)13:34
*** zz_jgrimm_ is now known as jgrimm13:35
ildikovjohnthetubaguy: dansmith: Hi! Could you please take a quick look at this short BP: https://review.openstack.org/#/c/193510/1 ?13:35
*** jroll has quit IRC13:36
*** jroll has joined #openstack-nova13:36
johnthetubaguyildikov: I am sorry, there are 100 in the queue right now, and I am busy with release management stuff today I am afraid13:36
johnthetubaguyits on my queue though, just will not be straight way I am afraid13:37
ildikovjohnthetubaguy: any chance before the deadline?13:37
claudiubjohnthetubaguy: then, is there anyone else who can approve specs for L? :)13:37
ildikovjohnthetubaguy: all your comments are fixed including splitting the BP, so it prolly does not require a long time to check it13:38
hayposdague: ok. i read the log of the 7 failures. 4 of them are unrelated to python at all, 3 are real bugs are come from a single change: 144875 which introduces real new python 3 bugs13:38
*** otter768 has quit IRC13:39
ildikovjohnthetubaguy: would be nice to have a table with reviewers and the queue size, so I wouldn't annoy you by default :)13:39
*** thangp has joined #openstack-nova13:39
hayposdague: for me, it's a good reason to make the job voting :) to ensure that such huge refactoring change breaking python 3 bugs doesn't go into nova ;) https://review.openstack.org/#/c/144875/13:39
openstackgerritRadoslav Gerganov proposed openstack/nova: VMware: map one nova-compute to one VC cluster  https://review.openstack.org/10391613:39
*** jogo has joined #openstack-nova13:41
*** ctrath has joined #openstack-nova13:41
*** Marga_ has joined #openstack-nova13:41
johnthetubaguyildikov: seems like russell's side is not working right now, I have this old version: http://5885fef486164bb8596d-41634d3e64ee11f37e8658ed1b4d12ec.r44.cf3.rackcdn.com/nova-specs-openreviews.html13:42
johnthetubaguyildikov: this is the regular one: http://russellbryant.net/openstack-stats/nova-specs-openreviews.html13:42
bauzasrussellb: ^ see above, sounds like your VM is saying bye bye13:43
sdaguehaypo: that's fine, I +2ed the patch13:44
hayposdague: cool! python3 is alive! be careful13:44
ildikovjohnthetubaguy: thanks, I guess I should practice praying more...13:44
johnthetubaguyildikov: submitting early is the best way, although understood thats not always possible13:45
openstackgerritJohn Garbutt proposed openstack/nova: Add context manager for RequestContext to switch db connection  https://review.openstack.org/16190613:47
ildikovjohnthetubaguy: it is up for review for a while, but if we consider review comments it's a question how early it has to be in order to get a chance13:47
*** zigo has quit IRC13:47
*** jogo has quit IRC13:47
openstackgerritAlexandre Levine proposed openstack/nova-specs: user_data modification  https://review.openstack.org/19179113:47
johnthetubaguyildikov: it was uploaded on June 19, the deadline is June 25, thats really really close to the deadline, our average turn around time for specs is probably near a month right now, sadly13:48
ildikovjohnthetubaguy: it's the original one: https://review.openstack.org/#/c/183033/13:49
*** rushiagr_away is now known as rushiagr13:49
openstackgerritRadoslav Gerganov proposed openstack/nova: VMware: map one nova-compute to one VC cluster  https://review.openstack.org/10391613:49
*** mpavone has joined #openstack-nova13:50
ildikovjohnthetubaguy: we had to split to follow the upgrade path and the first half of the blueprint is captured in the first patch, so it's just the original changeID in the second part of the proposal13:50
*** doron_ is now known as doron_afk13:50
johnthetubaguyildikov: ah, gotcha, yep, its getting about the same review as most right now, I have some ideas to improve this, but there were rejected last time I tried to change things, but will try a slightly different version of that soon13:51
openstackgerritAlexandre Levine proposed openstack/nova-specs: delete_on_termination modification spec  https://review.openstack.org/19414413:51
ildikovjohnthetubaguy: different version for the BP you mean?13:52
johnthetubaguyildikov: I mean for the spec review process13:52
*** r-daneel has joined #openstack-nova13:52
ildikovjohnthetubaguy: a-ha, ok, got it13:52
*** irenab has quit IRC13:52
openstackgerritAlexandre Levine proposed openstack/nova-specs: delete_on_termination modification spec  https://review.openstack.org/19414413:52
*** toure has left #openstack-nova13:53
ildikovjohnthetubaguy: BTW there is another BP that was blocked initially but it seems that the blocker activity still hasn't been started: https://review.openstack.org/#/c/182350/13:53
*** mpaolino has quit IRC13:56
*** lyarwood has quit IRC13:56
*** lyarwood has joined #openstack-nova13:57
johnthetubaguyildikov: its on a list of "blocked" specs I need to resolve, and decide what we are doing, we really want to see notifications tidied up this release, so we could unblock all these really good requests, but that just isn't moving forward right now I am afraid :(13:57
mriedem1notifications tidied up? like standardized?13:59
ildikovjohnthetubaguy: we will might be able to join that activity, but it's something that I need to clarify on our side regarding bandwidth13:59
mriedem1i guess we did talk about that at the summit but i don't remember concrete things coming out of it13:59
*** mpaolino has joined #openstack-nova14:00
openstackgerritRadoslav Gerganov proposed openstack/nova: VMware driver: Increasing speed of downloading image  https://review.openstack.org/13562514:00
ildikovjohnthetubaguy: I agree that it would be very important to happen, although if it will not in this cycle, then it would mean a very long delay for this relatively small fetaure/improvement14:00
*** ijuwang has joined #openstack-nova14:01
johnthetubaguyildikov: agreed14:01
*** nelsnelson has joined #openstack-nova14:01
*** iamjarvo has joined #openstack-nova14:02
edleafemriedem1: yeah, there wasn't a definite plan on that. More like a general consensus that the format of a notification is an API, and should be versioned.14:03
*** iamjarvo has quit IRC14:03
rgerganov__claudiub: please remove your -1 on https://review.openstack.org/#/c/13562514:03
rgerganov__claudiub: there is no longer string freeze14:03
*** iamjarvo has joined #openstack-nova14:03
mriedem1edleafe: yeah, which has been what we've been saying for 2+ years :)14:04
mriedem1and there was talk of emitting a notification for log messages or something, i got lost there14:04
ildikovjohnthetubaguy: can you suggest a way forward with these two bps? I hope we can get closer to some win-win situation14:04
mriedem1or intentionally blocked it from memory14:04
*** dguitarbite has quit IRC14:04
haypodims: https://review.openstack.org/#/c/187785/20/nova/objects/instance.py i don't understand why you prefer to replace list(filter(...)) with a list comprehension in a different patch set14:04
edleafemriedem1: probably the latter :)14:04
*** zigo has joined #openstack-nova14:05
*** willemgf has quit IRC14:05
jaypipesbauzas: ok, man, ready to hangout.14:05
*** lpetrut has quit IRC14:05
haypodims: as you wrote, your patch creates a new copy of the list, which would be avoided by using a list comprehension14:05
*** claudiub has quit IRC14:05
bauzasjaypipes: fair, lemme know the URL14:05
jaypipesbauzas: https://plus.google.com/hangouts/_/g6t5hrjbmcd4oo5z4sw3uqrzzia?hl=en14:05
haypodims: i'm not sure that anyone will write the "second patch" to cleanup your changes14:05
*** cfriesen has joined #openstack-nova14:05
lxslihaypo: changing it in the same patchset is dangerously close to cleaning up code in a patch intended to only enable py34 tests14:06
bauzasjaypipes: okay, gimme 30s and I'm in14:06
jaypipesbauzas: yup, no worries.14:07
lxslihaypo: I'm with dims, safer to split that out. There's nothing stopping you putting it up as a dependent patch right now though14:07
haypolxsli: replacing filter() with a list-comprehension is not a complete rewrite, IMHO it's a simple and obvious change. if there is a regression, it should be easier to catch it during the review14:07
lxslihaypo: it's nothing to do with python34 though, it's a performance/style tweak which has no place in that review14:08
*** eharney has joined #openstack-nova14:08
haypolxsli: dims' change makes py2 "slower", my remark is about python 214:08
*** nihilifer has quit IRC14:09
haypobut i don't have a strong opinion on having a second change14:09
ildikovjaypipes: Hi. Just a quick heads up if you could take a look on this one: https://review.openstack.org/#/c/192755/ ?14:09
openstackgerritEd Leafe proposed openstack/nova-specs: Improve reporting when Scheduler cannot find a host  https://review.openstack.org/18773914:09
*** cfriesen has quit IRC14:10
*** cfriesen_ has joined #openstack-nova14:10
*** jbernard has joined #openstack-nova14:10
*** rakesh_hs has quit IRC14:10
jaypipesildikov: yup, it's in the queue. :)14:10
ildikovjaypipes: cool, thanks :)14:10
mriedem1this has a +2 from mikal before I added some more logging to it https://review.openstack.org/#/c/192348/ <- should help debug a long standing race bug in the gate14:11
ildikovjaypipes: let me know if you wouldn't have enough there ;)14:11
*** mriedem1 is now known as mriedem_afk14:12
*** Nikolay_St has joined #openstack-nova14:13
*** beekneemech is now known as bnemec14:13
*** artom has joined #openstack-nova14:14
openstackgerritsahid proposed openstack/nova-specs: Libvirt real time instances  https://review.openstack.org/13968814:14
*** jecarey has joined #openstack-nova14:17
*** cfriesen_ has quit IRC14:19
*** sudipto has quit IRC14:19
*** artom_ has joined #openstack-nova14:21
*** tellesnobrega_ is now known as tellesnobrega14:22
*** artom has quit IRC14:22
*** tong_ has joined #openstack-nova14:23
*** artom_ has quit IRC14:23
*** ameade_ is now known as ameade14:23
*** artom has joined #openstack-nova14:23
ildikovjohnthetubaguy: Just one more question. Could the blueprint freeze excpetion work for any of those two BPs?14:24
*** artom has quit IRC14:24
*** artom has joined #openstack-nova14:24
*** adrian_otto has joined #openstack-nova14:25
*** tpeoples has joined #openstack-nova14:25
*** dboik has joined #openstack-nova14:27
*** Nikolay_St has quit IRC14:27
*** tongli has joined #openstack-nova14:28
johnthetubaguyildikov: unsure exactly how that will look right now14:28
*** thomasem has quit IRC14:28
*** tong_ has quit IRC14:28
*** thomasem has joined #openstack-nova14:29
openstackgerritStefan Amann proposed openstack/nova-specs: Support host type specific block volume attachment  https://review.openstack.org/19416314:30
*** Guest78977 has joined #openstack-nova14:33
*** ferest has joined #openstack-nova14:34
sahidmikal: just ping about https://review.openstack.org/#/c/187509/ i have prefered to address your comments. thanks14:34
*** ociuhandu has joined #openstack-nova14:35
*** mfisch` is now known as mfisch14:36
ildikovjohnthetubaguy: ok14:36
*** baoli_ has quit IRC14:36
*** mfisch is now known as Guest8229014:36
ildikovjohnthetubaguy: well, it would be good to start at least the on-shared storgae flag one in Liberty, as it requires two cycles to get the full job done14:37
*** artom has quit IRC14:37
ildikovjohnthetubaguy: I understand that the other one is not that good from the notification work PoV14:38
ildikovjohnthetubaguy: who should I ping regarding to the notification tidy up work to see where we are?14:38
johnthetubaguyildikov: agreed thats a good argument for an exception, the needing two releases thing14:38
*** baoli has joined #openstack-nova14:38
ildikovjohnthetubaguy: ok, cool, tnx14:39
*** marzif_ has quit IRC14:39
*** ferest has quit IRC14:39
*** artom has joined #openstack-nova14:39
johnthetubaguyildikov: unsure, it should be on here: https://etherpad.openstack.org/p/YVR-nova-liberty-summit-action-items14:39
*** marzif_ has joined #openstack-nova14:39
*** karimb has joined #openstack-nova14:40
ildikovjohnthetubaguy: ok, thanks for the pointer14:40
openstackgerritJohn Garbutt proposed openstack/nova: Consolidate the APIs for getting consoles  https://review.openstack.org/14850914:41
*** electrocucaracha has joined #openstack-nova14:41
*** Longgeek has quit IRC14:43
*** marcusvrn_ has joined #openstack-nova14:43
openstackgerritAlexis Lee proposed openstack/nova-specs: No more soft delete  https://review.openstack.org/18464514:45
openstackgerritgaryk proposed openstack/nova: Cells: add in missing unit test for get_by_uuid  https://review.openstack.org/19393914:46
*** lyarwood has quit IRC14:47
*** lyarwood has joined #openstack-nova14:47
*** electroc_ has joined #openstack-nova14:47
*** lpetrut has joined #openstack-nova14:47
*** electrocucaracha has quit IRC14:48
*** electroc_ has quit IRC14:48
*** claudiub has joined #openstack-nova14:48
*** toure has joined #openstack-nova14:48
openstackgerritDavanum Srinivas (dims) proposed openstack/nova: [WIP] Kicking the oslo.service tires  https://review.openstack.org/19290014:48
openstackgerritStefan Amann proposed openstack/nova-specs: Support host type specific block volume attachment  https://review.openstack.org/19416314:49
*** toure has quit IRC14:49
*** jlvillal has joined #openstack-nova14:49
claudiubjohnthetubaguy: hi. Very short question. Does this require a spec? It's more of a parity thing: https://blueprints.launchpad.net/nova/+spec/hyper-v-vnuma-enable14:51
*** jlvillal has quit IRC14:51
*** sudipto has joined #openstack-nova14:51
*** tonytan4ever has joined #openstack-nova14:52
*** jlvillal has joined #openstack-nova14:52
openstackgerritAlexis Lee proposed openstack/nova-specs: No more soft delete  https://review.openstack.org/18464514:52
*** ctrath has left #openstack-nova14:52
*** sdake_ has quit IRC14:52
*** browne has joined #openstack-nova14:53
*** Guest78977 has quit IRC14:53
*** rotbeard has quit IRC14:54
*** toure has joined #openstack-nova14:56
*** wverdugo500 has joined #openstack-nova14:56
*** busterswt has joined #openstack-nova14:56
*** thomasem has quit IRC14:57
*** ctrath has joined #openstack-nova14:57
*** ctrath has joined #openstack-nova14:57
openstackgerritSylvain Bauza proposed openstack/nova-specs: Proposes to check destination on migrations  https://review.openstack.org/17922414:58
*** veena has joined #openstack-nova14:58
*** wverdugo500 has left #openstack-nova14:58
*** rotbeard has joined #openstack-nova14:58
*** gszasz has joined #openstack-nova14:59
*** yamahata has joined #openstack-nova14:59
*** patrickeast has joined #openstack-nova15:00
*** jcoufal has joined #openstack-nova15:01
*** wendar_ is now known as wendar15:01
openstackgerritgaryk proposed openstack/nova: Cells: add instance cell registration utility to nova-manage  https://review.openstack.org/19388915:02
garykalaski: ^15:02
*** thomasem has joined #openstack-nova15:02
garyki am thinking that maybe we can do it per host. what do you suggest for the filters?15:02
danpbgaryk: what kind of VIF types can be used with vmware ?15:03
*** thomasem has quit IRC15:04
danpbneutron has a whole load of possible VIF types, but i don't see specific handling for any in vmware15:04
*** thomasem has joined #openstack-nova15:04
danpbso i'm guessing you have some assumptions about which neutron mechanism is configured ?15:04
garykdanpb: correct15:04
*** thomasem has quit IRC15:04
*** pixelb has joined #openstack-nova15:05
*** kiran-r has joined #openstack-nova15:05
*** mtanino has joined #openstack-nova15:05
alaskigaryk: we want all of the instances.  we might want to limit the number done at once, but I don't think we need to filter15:06
*** thomasem has joined #openstack-nova15:06
alaskigaryk: per host could be an optional way to limit as well15:07
danpbgaryk: so what config does vmware expect ?15:07
*** TemporalBeing has joined #openstack-nova15:07
garykalaski: ok, undertood15:07
*** davideagnello has joined #openstack-nova15:07
*** mjturek1 has joined #openstack-nova15:07
garykdanpb: when we are using neutron we expect taht the vif type is 'dvs'15:07
garykhttps://github.com/openstack/nova/blob/master/nova/network/model.py#L3415:07
*** nkrinner has quit IRC15:07
garykdanpb: in the next generationplugin this will be OVS15:08
garyk'ovs'15:08
garykthat is, 'ovs' and 'dvs'15:08
openstackgerritXinyuan Huang proposed openstack/nova: Pass requested_networks to scheduler  https://review.openstack.org/19363515:08
*** ctrath has quit IRC15:08
garykdanpb: is that what you were asking15:08
*** ctrath has joined #openstack-nova15:09
*** Longgeek has joined #openstack-nova15:10
garykalaski: tomorrow i'll add in the limit.15:10
alaskigaryk: great, thanks15:10
danpbgaryk: ah ok, i see what you mean15:11
*** kiran-r has quit IRC15:11
*** oro has quit IRC15:11
*** thomasem has quit IRC15:11
*** thomasem has joined #openstack-nova15:11
*** thomasem has quit IRC15:12
*** dave-mccowan has quit IRC15:12
*** thomasem has joined #openstack-nova15:12
*** thomasem has quit IRC15:12
garykdanpb: in the virt driver we need to connect to the dvs network. this can be a number of different types which we treat in the driver - for example opaque - https://review.openstack.org/16575015:12
*** mdrabe has joined #openstack-nova15:13
*** thomasem has joined #openstack-nova15:13
*** agireud has joined #openstack-nova15:13
*** sudipto has quit IRC15:13
*** sudipto_ has joined #openstack-nova15:13
*** adrian_otto has quit IRC15:13
*** thomasem has quit IRC15:13
*** lyarwood has quit IRC15:15
*** lyarwood has joined #openstack-nova15:15
*** aix has quit IRC15:16
*** thomasem has joined #openstack-nova15:17
*** nihilifer has joined #openstack-nova15:19
*** agentle has joined #openstack-nova15:19
*** lyarwood has quit IRC15:20
*** lyarwood has joined #openstack-nova15:20
*** Longgeek has quit IRC15:20
openstackgerritgaryk proposed openstack/nova: Cells: add in missing unit test for get_by_uuid  https://review.openstack.org/19393915:22
*** afazekas has quit IRC15:23
*** dave-mccowan has joined #openstack-nova15:24
*** keekz_ is now known as keekz15:24
openstackgerritStefan Amann proposed openstack/nova-specs: Support host type specific block volume attachment  https://review.openstack.org/19416315:25
*** Nikolay_St has joined #openstack-nova15:28
*** belmoreira has quit IRC15:28
*** rajesht has quit IRC15:28
openstackgerritAndrey Kurilin proposed openstack/nova-specs: Nova API Microversions support in NovaClient  https://review.openstack.org/18841015:30
*** wwriverrat has quit IRC15:31
*** Redcavalier has joined #openstack-nova15:33
*** rlrossit has joined #openstack-nova15:34
garyksdague: why is https://review.openstack.org/#/c/194192/ reverted?15:35
*** dsanders has joined #openstack-nova15:35
markus_zI get a ValueError when I run "tox -e docs" in nova-specs. It's complaining about the "redirects". Do I do something wrong?15:35
*** Guest81202 is now known as mgagne15:36
*** mgagne has joined #openstack-nova15:36
openstackgerritClaudiu Belu proposed openstack/nova-specs: Adds Hyper-V Shielded VMs spec  https://review.openstack.org/19289215:36
bauzasmarkus_z: try 'tox -r -edocs' to rebuild your venv15:37
claudiubdanpb: hi. I've addressed your comments on the Hyper-V uefi secure boot spec. If you have a little bit of time, could you take a look? :) https://review.openstack.org/#/c/190997/15:37
jrolljaypipes: devananda: let's chat about ironic "ha"/scheduling/resource tracker things today so we can get a spec in. I'm free all day, how about y'all?15:37
*** Nikolay_St has quit IRC15:38
jaypipesjroll: I'm free in about an hour15:38
devanandai have a call in 20 minutes, free after that15:38
jrollI'm also free beginning at 1700 utc15:38
jrollwhich is 80 minutes from now15:38
jrollplan on that?15:39
johnthetubaguyclaudiub: probably doesn't need a spec, assuming its using the same settings as libvirt, and they get added into that BP, but it would need to get approved in a nova meeting, by first adding it on the agenda https://blueprints.launchpad.net/nova/+spec/hyper-v-vnuma-enable15:39
*** agentle has quit IRC15:39
*** otter768 has joined #openstack-nova15:39
markus_zbauzas: I tried this, the result keeps being the same.15:39
*** zzzeek has joined #openstack-nova15:40
danpbjohnthetubaguy: claudiub   actually i think it probably does need a spec15:40
danpbbecause that blueprint only talks about exposing NUMA topology to the guest15:40
*** Sukhdev has joined #openstack-nova15:40
*** agentle has joined #openstack-nova15:40
bauzasjaypipes: jroll: devananda: will lurk at your convo, since I have a meeting conflict15:40
danpbin nova that's only half of how numa is setup - there's all the integration  in the schedular15:41
jaypipesjroll: sure, works for me.15:41
*** marzif_ has quit IRC15:41
bauzasmarkus_z: probably a dependency issue then ?15:41
danpband I think we need  to understand how that will be dealt with by hyperv driver from scheduling pov15:41
jrollbauzas: cool. the main thing is (host, node) -> (host) and also how to do failure domain things instead of one nova-compute process :)15:41
*** idevops has joined #openstack-nova15:41
bauzasjroll: yeah I know15:41
*** marzif_ has joined #openstack-nova15:41
*** idevops has quit IRC15:42
bauzasjroll: I mean, I know your usecase15:42
jrollbauzas: ok :)15:42
markus_zbauzas: Looks like an issue in the "redirect.py". I'll debug it. Thanks for your help!15:42
claudiubdanpb: johnthetubaguy: I see. Will try to spec it, as soon as I can. Thanks for the info!15:42
bauzasjroll: tbc, I would prefer a (host, node) -> (node) translation, but I'm yakshaving :)15:43
jrollbauzas: hah. I just want to be inline with whatever the rest of nova is doing.15:43
danpbclaudiub: basically the fun bit is all about the scheduling picking the right hosts to deploy the guest on based on numa availabiity15:44
*** veena has quit IRC15:44
bauzasjaypipes: jroll: just say my name before the meeting is starting so I can get the logs15:44
danpbmaybe it'll be easy and everything we did for the schedular for libvirt will "just work" with hyperv15:44
jrollbauzas: can do :)15:44
*** otter768 has quit IRC15:45
johnthetubaguymriedem_afk: I think there are some -2s that can go now: https://review.openstack.org/#/q/topic:bp/consolidate-libvirt-fs-volume-drivers,n,z15:45
devanandabauzas: fwiw, I would also prefer (host, node) -> (node) -- BUT the nova folks convinced me that that's not going to work for most other use cases at the summit15:45
devanandabauzas: don't ask me to remember why that is, though :)15:45
devanandabauzas: we've come up with a solution for Ironic that should work with just (host), I think. so I'm good with that now15:46
bauzasdevananda: honestly, I just want to make sure that we're finding a good way to help you, that's it15:46
bauzasdevananda: I see the 'host' reason, that's because on how we identify the related running services15:47
devanandabauzas: "failure domains"15:47
bauzasdevananda: yeah, the service is indeed the failure domain15:47
devanandabauzas: https://review.openstack.org/#/c/193263/15:47
devanandasome light reading before we chat15:48
bauzasdevananda: I won't be able to make the call because of $family but I will read the spec and see the irc logs15:48
johnthetubaguydevananda: yeah, we send a cast to the host once the scheduler has picked the host, we also send along the node to that host, or something like that, I think15:48
*** Longgeek has joined #openstack-nova15:48
*** dsanders has quit IRC15:49
johnthetubaguy(i.e. host gives you the RPC queue to send the message to for the nova-compute to get it)15:49
*** veena has joined #openstack-nova15:50
sdaguegaryk: because it's masking a real failure15:50
bauzasjohnthetubaguy: correct https://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L71415:50
sdagueit's not going to pass15:50
sdaguebecause it's masking a failure15:50
sdaguebut need the results up there for people to look at it15:51
*** claudiub has quit IRC15:51
sdaguegaryk: also, see the last hour of conversation in #openstack-qa15:51
ndipanovjaypipes, https://review.openstack.org/#/c/193576/ dansmith thought you might want to chime in on this one... (not sure if I bothered you about it yet)15:52
*** ildikov has quit IRC15:53
jaypipesndipanov: pushed into my queue.15:53
*** cinerama` is now known as cinerama15:53
ndipanovjaypipes, couldn't ask for more :)15:53
*** baoli has quit IRC15:54
*** dguitarbite has joined #openstack-nova15:54
*** claudiub has joined #openstack-nova15:56
openstackgerritEd Leafe proposed openstack/nova-specs: Add a Distinct Exception for Exceeding Max Retries  https://review.openstack.org/19420415:58
openstackgerritStefan Amann proposed openstack/nova-specs: Support host type specific block volume attachment  https://review.openstack.org/19416315:58
*** afazekas has joined #openstack-nova15:58
*** salv-orl_ has joined #openstack-nova16:00
jaypipesjroll: actually, can we push it back another hour? say, 2pm EST?16:00
openstackgerritAndrew Laski proposed openstack/nova: Cells: cell scheduler anti-affinity filter  https://review.openstack.org/19287516:00
jrolljaypipes: fine with me. devananda ?16:00
*** moshele has joined #openstack-nova16:01
devanandajroll, jaypipes: nope. I have another meeting at 11am PST16:01
garyksdague: tx16:02
jaypipesdevananda: can yuo do 1:30pm EST?16:02
jrolldevananda: jaypipes I have a thing at 2:30-3pm PST, other than that completely free, you two pick whatever works :)16:02
devanandajaypipes: I'm free from 12:30 to 2 EST, and 3 EST onwards16:02
*** salv-orlando has quit IRC16:03
jaypipesjroll, devananda: ok, how about 3pm EST16:03
*** wwriverrat has joined #openstack-nova16:03
jroll+116:03
openstackgerritWalter A. Boring IV (hemna) proposed openstack/nova: Switch to using os-brick  https://review.openstack.org/17556916:03
devanandaworks for me16:03
*** hemna has joined #openstack-nova16:04
moshelejohnthetubaguy: hi16:05
johnthetubaguymoshele: hi16:05
jaypipesdevananda, jroll: k. 3pm EST it is.16:05
jrolljaypipes: thanks16:06
*** iamjarvo has quit IRC16:06
moshelejohnthetubaguy: regarding new vif  type is it all waiting for vif refactoring?16:06
*** jlanoux has quit IRC16:06
*** afazekas has quit IRC16:06
*** electrocucaracha has joined #openstack-nova16:06
moshelejohnthetubaguy: I am asking regarding https://review.openstack.org/#/c/187052/16:06
*** electrocucaracha has quit IRC16:07
*** hemna has quit IRC16:07
johnthetubaguymoshele: unsure, if we had a good plan for the VIF, its likely we would give it priority at this point, as the current situation doesn't work, afraid I don't have an answer for you right now16:07
*** agentle has quit IRC16:08
moshelejohnthetubaguy: also regarding NFV specs in https://etherpad.openstack.org/p/liberty-nova-priorities-tracking16:08
*** MarcF has joined #openstack-nova16:08
*** jcoufal has quit IRC16:09
moshelejohnthetubaguy:  is there still a concern regarding   testing16:09
moshelejohnthetubaguy: for example with https://review.openstack.org/#/c/179577/16:10
*** thedodd has joined #openstack-nova16:10
*** Guest82290 is now known as mfisch16:10
johnthetubaguymoshele: afraid I am snowed under with liberty-1 release paper work right now, so I will have to look into this later16:10
*** mfisch has quit IRC16:10
*** mfisch has joined #openstack-nova16:10
moshelejohnthetubaguy:ok thanks16:11
*** pradipta has quit IRC16:12
*** mpavone has quit IRC16:12
*** unicell1 has joined #openstack-nova16:12
*** hemna has joined #openstack-nova16:14
ndipanovsahid, got a sec or are you off?16:14
*** afazekas has joined #openstack-nova16:14
*** electrocucaracha has joined #openstack-nova16:15
*** garyk has quit IRC16:15
*** unicell has quit IRC16:15
openstackgerritStephen Finucane proposed openstack/nova-specs: Add virt-driver CPU thread pinning  https://review.openstack.org/19422116:15
*** patrickeast has quit IRC16:16
*** arnaud___ has joined #openstack-nova16:17
*** moshele has quit IRC16:17
*** unicell1 has quit IRC16:18
*** husanu has joined #openstack-nova16:18
*** david-ly_ is now known as david-lyle16:18
sahidndipanov: i'm here16:20
*** avishay_ has quit IRC16:21
openstackgerritMarkus Zoeller (markus_z) proposed openstack/nova-specs: libvirt: use "serial console" and "log file" in parallel  https://review.openstack.org/19422216:22
*** yamahata has quit IRC16:22
*** tonytan4ever has quit IRC16:22
*** agentle has joined #openstack-nova16:23
ndipanovsahid, so with regards to that serial console thing16:24
ndipanovbasically what you want is to implement the auto port thing libvirt has for VNC. but in Nova16:24
ndipanovfor serial console16:25
sahidndipanov: i do not know what is the auto port thing16:25
ndipanovhmmm maybe it's not it16:25
sahidfrom what i know we do not have a way to know when a port is connected16:26
sahidso we have a guest we a range of port listening16:26
openstackgerritStefan Amann proposed openstack/nova-specs: Support host type specific block volume attachment  https://review.openstack.org/19416316:27
ndipanovhow come libvirt will allow you a multiple number of vnc connections to the same instance??16:27
sahidand when a console is connected to one of them we do not have his information16:27
ndipanovthat is what I assume drove the requirement to have "more than one" serial console when the original BP was being drafted16:27
ndipanovright16:28
ndipanovso that's why I'm saying - push this to the user16:28
sahidyou mean the user will pass the port number?16:29
ndipanovbecause even if you keep that stuff in the DB - you don't know when the user disconnects16:29
*** kiran-r has joined #openstack-nova16:29
ndipanovso we want to stick a bunch of functionality into nova, only to half implement something we can't really fix16:29
ndipanovthat's basically my problem with that spec in it's current form16:30
ndipanovwell no - user says - get me a serial console16:30
ndipanovand Nova returns instead of one like now16:30
ndipanovall of them16:30
ndipanovand validates all of them16:30
ndipanovso if your instance was booted with 3 serial consoles16:31
ndipanovall 3 get returned and validated (get a token)16:31
*** pask81 has quit IRC16:31
ndipanovso then it's up to the user / application to decide how to deal with that16:31
*** yassine has quit IRC16:31
sahidndipanov: really ??16:31
*** kiran-r has quit IRC16:32
*** nihilifer has quit IRC16:32
sahidndipanov: the default range is 10000-2000016:32
*** kiran-r has joined #openstack-nova16:32
ndipanovwell yes16:33
*** kiranr has joined #openstack-nova16:33
*** nihilifer has joined #openstack-nova16:33
sahidndipanov: then we can know when user is disconnected16:33
ndipanovbut if you ask for an instance to have 3 serial consoles, only 3 ports will be allocated16:33
sahidactually the proxy know that16:33
sahidknows knews...16:33
sahidthat why is good to have this information in database16:33
ndipanovno but we don't care16:34
sahidservice can shares the data16:34
*** husanu has quit IRC16:35
ndipanovso the end goal is to replicate the functionality libvirt allows with vnc consoles?16:36
ndipanovthat's a whole different story than just objects16:36
ndipanovI mean that spec is just way too braod16:36
sahidthe websocket provides some mechanisms, it will update the connect_info to explicitly says that this port NNNN is connected16:36
*** kiran-r has quit IRC16:37
sahidthen when new serial request will be done the driver will pass port already connected16:37
ndipanovbroad*16:37
sahidndipanov: no, that speac give to us more16:37
*** husanu has joined #openstack-nova16:37
sahidmemcached is no a database16:37
*** haypo has quit IRC16:38
sahidthen we will have a code more robust if we use the framework object16:38
*** arnaud___ has quit IRC16:38
*** markus_z has quit IRC16:39
*** veena has quit IRC16:40
*** yamahata has joined #openstack-nova16:41
*** xyang1 has joined #openstack-nova16:42
*** nic has joined #openstack-nova16:42
*** doron_afk has quit IRC16:42
*** mpaolino has quit IRC16:43
ndipanovsahid, ok let me think about it some more16:43
ndipanovbut at least we should be clear about what we are doing16:43
*** asselin has joined #openstack-nova16:44
*** rwsu has joined #openstack-nova16:45
*** tongli has quit IRC16:46
*** teju has joined #openstack-nova16:46
*** tongli has joined #openstack-nova16:47
*** unicell has joined #openstack-nova16:47
sahidndipanov: sure :)16:47
openstackgerritNicolas Simonds proposed openstack/nova: libvirt: Make the guest console configurable  https://review.openstack.org/12252316:48
johnthetubaguydansmith: alaski: a priority spec the API sub team seem to have mostly +1ed now: https://review.openstack.org/#/c/173243/17/specs/liberty/approved/api-relax-validation.rst,cm16:49
*** agentle has quit IRC16:49
*** afazekas has quit IRC16:49
*** agentle has joined #openstack-nova16:50
openstackgerritNicolas Simonds proposed openstack/nova: Use utf8_bin collation on the flavor extra-specs table in MySQL  https://review.openstack.org/19030016:51
*** tongli has quit IRC16:51
openstackgerritNicolas Simonds proposed openstack/nova: imagebackend: Split out resize_image logic from create_image  https://review.openstack.org/18785716:51
openstackgerritNicolas Simonds proposed openstack/nova: Fix/implement revert-resize for RBD-backed images  https://review.openstack.org/18739516:51
alaskijohnthetubaguy: okay, will look shortly16:53
*** rotbeard has quit IRC16:54
openstackgerritJohn Garbutt proposed openstack/nova: Convert RT compute_node to be a ComputeNode object  https://review.openstack.org/16094216:54
*** afazekas has joined #openstack-nova16:54
*** derekh has quit IRC16:55
*** ildikov has joined #openstack-nova16:56
johnthetubaguydanpb: I am curious about if this is complete now? https://blueprints.launchpad.net/nova/+spec/convert-image-meta-into-nova-object16:58
*** cinerama has quit IRC16:58
*** tongli has joined #openstack-nova16:59
*** dboik has quit IRC16:59
*** vilobhmm has joined #openstack-nova16:59
*** cinerama has joined #openstack-nova17:00
*** afazekas has quit IRC17:00
*** dboik has joined #openstack-nova17:00
*** teju has left #openstack-nova17:00
*** lucasagomes is now known as lucas-dinner17:02
*** patrickeast has joined #openstack-nova17:05
vilobhmmjohnthetubaguy : ping17:06
johnthetubaguyvilobhmm: hi17:06
*** lyarwood has quit IRC17:08
*** lyarwood has joined #openstack-nova17:08
vilobhmmthanks for merging the servicegroup spec….although i have a WIP for https://review.openstack.org/#/c/138607/ ready…..if you can provide your inputs on https://review.openstack.org/#/c/190322/ it would be great…(since https://review.openstack.org/#/c/190322/ is a dependancy for https://review.openstack.org/#/c/138607/17:08
vilobhmmjohnthetubaguy : let me give you a summary as in where are we now with respect to https://review.openstack.org/#/c/190322/17:09
*** dims has quit IRC17:09
*** dims has joined #openstack-nova17:09
vilobhmmjohnthetubaguy : melwitt, harlowja, me and few others are discussing whether #1. to keep the services information and service liveliness information both in same datastore or #2. Keep services information in one datastore and service liveliness in another (which will be governed by CONF.servicegroup_driver) …so need your opinion regarding that…once the path is clear putting up the implementation should be starightforward for me.17:11
*** pixelb has quit IRC17:12
*** danpb has quit IRC17:12
*** dboik has quit IRC17:13
*** dsanders has joined #openstack-nova17:13
*** ssurana has joined #openstack-nova17:13
*** garyk has joined #openstack-nova17:13
*** lyarwood has quit IRC17:13
*** lyarwood has joined #openstack-nova17:13
*** sahid has quit IRC17:14
*** hemna has quit IRC17:14
*** hemna has joined #openstack-nova17:15
bauzasvilobhmm: to be clear, it should be opiniated by operators17:17
*** pixelb has joined #openstack-nova17:17
johnthetubaguyvilobhmm: so I am not sure the best approach, I don't like the idea of syncing with the DB, but it depends a lot about what works I guess17:17
bauzasvilobhmm: I mean, we should leave the possibility for both either two distinct datastores or only one, IMHO17:17
*** matrohon has joined #openstack-nova17:17
*** sdake_ has joined #openstack-nova17:18
bauzasvilobhmm: the key thing is, just make sure that if 2 datastores, then we have a trivial way to key btw/ them17:18
johnthetubaguyvilobhmm: my gut tells me store it all in the same place, to avoid the sync issue, no idea if thats practical though17:18
bauzasvilobhmm: in other terms, denormalization is okay, provided we have a single key17:18
*** unicell1 has joined #openstack-nova17:18
vilobhmmjohnthetubaguy : +1 I am also of the same thought17:18
vilobhmmone data store17:19
vilobhmmno more syncronization issues17:19
*** unicell has quit IRC17:19
*** rwsu has quit IRC17:19
bauzasvilobhmm: on that point, I'm not really +1 with that17:19
bauzasvilobhmm: because the services table is not only in use for the SG API, but we have various places where it's needed17:19
bauzasvilobhmm: so, given only one datastore, what happens if that datastore is ZK while the rest of Nova is MySQL DB'd ?17:20
bauzasvilobhmm: you'll still need a way to unique key17:20
*** Longgeek has quit IRC17:20
bauzasvilobhmm: so, my take is to say that should be a configuration detail, but we should care at the application level how to key against that for the rest of Nova17:21
bauzasvilobhmm: I'll reply to your spec as soon as I drop from $family_time17:22
openstackgerritNicolas Simonds proposed openstack/nova: Use utf8_bin collation on the flavor extra-specs table in MySQL  https://review.openstack.org/19030017:22
vilobhmmbauzas : if you can point me to places its used other than checking service liveliness or getting service information details….earlier it was direct db call now its through service objects thats it17:22
bauzasvilobhmm: for example, we have a strong dependency on the compute/service relationship17:23
vilobhmmbauzas : since the spec says that everything *seecific to services* goes through servicegroup API layer…the servicegroup API layer will be aware of the underlying datastore17:23
bauzasvilobhmm: that, I agree17:23
bauzasvilobhmm: I'm just saying you'll still need a application layer sync anyway IMO17:24
*** thangp is now known as thangp_afk17:24
*** marcusvrn_ has quit IRC17:24
vilobhmmbauzas : I would love to hear the use-case and in what case the application layer sync will be needed ?17:24
*** lyarwood has quit IRC17:25
*** lyarwood has joined #openstack-nova17:25
vilobhmmbauzas : thanks if you can review the spec it will be great…won't take anymore of your family time …will wait for your review17:25
*** pixelb has quit IRC17:26
mriedem_afkjohnthetubaguy: those are my -2s on my series :)17:27
mriedem_afkjohnthetubaguy: b/c they potentially conflict with hemna's os-brick patch17:27
*** baoli has joined #openstack-nova17:28
jaypipesjohnthetubaguy: ping. re: this server instance tagging blueprint...17:28
*** harlowja has joined #openstack-nova17:28
* mriedem_afk heads out to see jurassic world17:28
*** electrocucaracha has quit IRC17:28
jaypipesjohnthetubaguy: if this misses Liberty, this will be the 4th (!) release this blueprint has been trying for. :(17:28
jaypipesjohnthetubaguy: at this point, what exactly is wrong with the spec?17:28
*** annashen has joined #openstack-nova17:29
mriedem_afkjaypipes: i have that beat https://review.openstack.org/#/c/69047/17:29
jaypipesmriedem_afk: nothing we should be proud of :(17:29
*** alex_klimov has quit IRC17:30
*** baoli has quit IRC17:31
*** baoli has joined #openstack-nova17:32
*** matrohon has quit IRC17:33
*** MVenesio has joined #openstack-nova17:35
*** iamjarvo has joined #openstack-nova17:35
*** rwsu has joined #openstack-nova17:36
*** q3k is now known as `reboot`17:36
*** dane-fichter has joined #openstack-nova17:36
*** marzif_ has quit IRC17:39
*** baoli has quit IRC17:40
*** baoli has joined #openstack-nova17:40
*** otter768 has joined #openstack-nova17:40
*** Sukhdev has quit IRC17:40
*** agentle has quit IRC17:41
*** agentle has joined #openstack-nova17:41
*** gszasz has quit IRC17:42
openstackgerritDiana Clarke proposed openstack/nova: Register the vnc config options under group 'vnc'  https://review.openstack.org/19392317:42
openstackgerritVladik Romanovsky proposed openstack/nova-specs: libvirt: virtio-net multiqueue  https://review.openstack.org/12882517:43
*** dboik has joined #openstack-nova17:43
*** Sukhdev has joined #openstack-nova17:43
*** otter768 has quit IRC17:45
*** ijw_ has quit IRC17:46
openstackgerritVladik Romanovsky proposed openstack/nova-specs: libvirt: virtio-net multiqueue  https://review.openstack.org/12882517:48
*** Longgeek has joined #openstack-nova17:49
*** agentle has quit IRC17:50
*** achanda has joined #openstack-nova17:54
*** VW_ has quit IRC17:54
*** kiranr has quit IRC17:56
*** kiran-r has joined #openstack-nova17:57
*** Nikolay_St has joined #openstack-nova17:58
*** VW_ has joined #openstack-nova17:59
*** elmiko has joined #openstack-nova17:59
elmikohey nova folks, this change https://review.openstack.org/#/c/179857/ is not included in stable/kilo, is it slated for release in liberty-1?18:00
*** amotoki has quit IRC18:00
sdagueelmiko: it's merged, so yes18:01
elmikosdague: thanks, just wanted to confirm =)18:01
sdagueelmiko: that also seems suitable for backport18:01
*** agireud has quit IRC18:01
sdagueif you were so inclined18:01
elmikosdague: cool, will do!18:02
*** patrickeast_ has joined #openstack-nova18:03
*** patrickeast has quit IRC18:03
*** patrickeast_ is now known as patrickeast18:03
*** kiran-r has quit IRC18:04
*** romainh has joined #openstack-nova18:05
*** thangp_afk is now known as thangp18:06
*** salv-orl_ has quit IRC18:07
*** agireud has joined #openstack-nova18:11
*** afazekas has joined #openstack-nova18:12
*** browne has quit IRC18:13
*** browne has joined #openstack-nova18:13
*** ZZelle_ has joined #openstack-nova18:13
openstackgerritDan Smith proposed openstack/nova: Fix up some bits of resource_tracker to use instance flavors  https://review.openstack.org/19199418:13
openstackgerritDan Smith proposed openstack/nova: Remove compat_instance()  https://review.openstack.org/19270818:13
openstackgerritDan Smith proposed openstack/nova: Fix testing object fields with missing instance rows  https://review.openstack.org/19280518:13
openstackgerritDan Smith proposed openstack/nova: Fix up instance flavor usage in compute and network tests  https://review.openstack.org/19234718:13
openstackgerritDan Smith proposed openstack/nova: Fix up ec2 tests for flavors on instances  https://review.openstack.org/19234618:13
openstackgerritDan Smith proposed openstack/nova: Fix up xenapi tests for instance flavors  https://review.openstack.org/19234518:13
*** pixelb has joined #openstack-nova18:20
*** Sukhdev has quit IRC18:24
*** toure has quit IRC18:25
*** jasondot_ has joined #openstack-nova18:25
*** belmoreira has joined #openstack-nova18:26
*** tongli has quit IRC18:27
elmikosdague: would you mind nominating this bug to kilo and juno for nova, https://bugs.launchpad.net/ossn/+bug/145193118:29
openstackLaunchpad bug 1451931 in OpenStack Security Notes "ironic password config not marked as secret" [Undecided,New] - Assigned to Michael McCune (mimccune)18:29
elmikowell, that's the ossn. i meant the nova portion18:29
*** xyang1 has quit IRC18:30
sdagueelmiko: honestly, just propose the backport18:30
elmikosdague: ack, wasn't sure on the procedure18:30
*** tonytan4ever has joined #openstack-nova18:31
sdaguehonestly, mriedem_afk is the stable maint point person for nova, he can probably make sure you are pointed in the right direction18:31
dansmithalaski: around?18:31
elmikosdague: cool, i pushed both branches for review. thanks for the advice =)18:31
*** toure has joined #openstack-nova18:32
alaskidansmith: yes18:34
*** avishay_ has joined #openstack-nova18:35
dansmithalaski: so, we want to have more information about evacuate operations when they're happening, so we have reasonable information that they're progressing18:36
dansmithalaski: or have an idea if they have stalled or failed so we can restart them18:36
dansmithalaski: I want to do that via notifications so they're async and not requiring polling18:37
*** xyang1 has joined #openstack-nova18:37
dansmithalaski: is that cool as opposed to something like actions?18:37
dansmithcool with you I mean18:37
*** sudipto_ has quit IRC18:39
alaskidansmith: yeah, that would be good for us as well.  longer term that could possibly feed into something visible in a task, but notifications are great for something like that18:39
dansmithalaski: definitely, but I mean for the immediate benefit18:39
alaskigreat for exposing this, not for updating a task I mean18:39
alaskidansmith: yeah, I'm on board with it18:39
dansmithcool18:39
*** electrocucaracha has joined #openstack-nova18:41
*** afazekas has quit IRC18:42
*** kiran-r has joined #openstack-nova18:45
*** afazekas has joined #openstack-nova18:49
*** kylek3h has joined #openstack-nova18:51
*** jasondot_ has quit IRC18:51
*** pixelb has quit IRC18:51
*** agentle has joined #openstack-nova18:58
*** salv-orlando has joined #openstack-nova18:59
devanandajroll, jaypipes - around when ever you are19:00
jrollohai19:00
jaypipeshey19:00
jaypipesone sec, grabbing link to hangout.19:00
jrollbauzas: you said to ping so you can grab logs. but apparently hanging out. so. :/19:01
*** ijw_ has joined #openstack-nova19:01
jaypipesdevananda, jroll, bauzas: https://plus.google.com/hangouts/_/grilkgonwl7sazf5gguk425yxya?hl=en19:01
*** agentle_ has joined #openstack-nova19:02
*** afazekas has quit IRC19:03
devanandajaypipes: https://review.openstack.org/#/c/193263/119:03
*** afazekas has joined #openstack-nova19:04
*** salv-orlando has quit IRC19:04
*** agentle has quit IRC19:05
openstackgerritJesse J. Cook proposed openstack/nova-specs: Create a software seam for Glance API code in Nova  https://review.openstack.org/19430319:06
devanandajroll: dude. that hurt19:07
jroll?19:07
devanandawtf is up with your mic?19:07
jrollidk19:07
jaypipesjroll: machine gun static.19:07
*** rushiagr is now known as rushiagr_away19:08
jroll¯\_(ツ)_/¯19:08
*** artom has quit IRC19:08
*** Longgeek has quit IRC19:09
*** rfolco_ has joined #openstack-nova19:10
*** ijw__ has joined #openstack-nova19:11
bauzasjroll: jaypipes: devananda: arriving19:13
*** ijw_ has quit IRC19:13
*** vladikr has quit IRC19:13
*** rwsu has quit IRC19:15
*** agentle_ has quit IRC19:17
*** agentle has joined #openstack-nova19:18
*** afazekas has quit IRC19:19
*** husanu has quit IRC19:21
*** dramakri has joined #openstack-nova19:22
*** busterswt has quit IRC19:23
*** husanu has joined #openstack-nova19:23
*** jasondot_ has joined #openstack-nova19:24
*** jgrimm is now known as zz_jgrimm19:26
*** VW__ has joined #openstack-nova19:26
*** VW_ has quit IRC19:26
*** agentle has quit IRC19:27
*** annashen has quit IRC19:29
*** ociuhandu has quit IRC19:29
*** annashen has joined #openstack-nova19:31
*** annashen has quit IRC19:32
*** annashen has joined #openstack-nova19:32
*** ijw_ has joined #openstack-nova19:37
*** Longgeek has joined #openstack-nova19:37
*** achanda has quit IRC19:37
*** avishay_ has quit IRC19:38
*** rwsu has joined #openstack-nova19:38
*** avishay_ has joined #openstack-nova19:38
*** marzif_ has joined #openstack-nova19:39
*** rwsu has quit IRC19:40
*** rwsu has joined #openstack-nova19:40
*** ijw__ has quit IRC19:40
*** tonytan4ever has quit IRC19:41
*** otter768 has joined #openstack-nova19:41
*** tonytan4ever has joined #openstack-nova19:42
*** karimb has quit IRC19:42
*** lucas-dinner has quit IRC19:43
openstackgerritJesse J. Cook proposed openstack/nova-specs: Create a software seam for Glance API code in Nova  https://review.openstack.org/19430319:43
*** `reboot` is now known as q3k19:44
*** otter768 has quit IRC19:46
*** tonytan4ever has quit IRC19:46
*** mgarza has joined #openstack-nova19:47
*** tonytan4ever has joined #openstack-nova19:47
*** avishay_ has quit IRC19:52
*** dramakri has quit IRC19:52
*** lyarwood has quit IRC19:52
*** dramakri has joined #openstack-nova19:52
*** lyarwood has joined #openstack-nova19:52
*** agentle has joined #openstack-nova19:52
*** annashen has quit IRC19:54
*** lucas-dinner has joined #openstack-nova19:56
*** pixelb has joined #openstack-nova19:57
*** vladikr has joined #openstack-nova19:57
*** afazekas has joined #openstack-nova19:57
*** toure has quit IRC20:01
*** isd has joined #openstack-nova20:02
*** avishay_ has joined #openstack-nova20:02
*** agentle has quit IRC20:02
*** edmondsw has quit IRC20:04
*** VW__ has quit IRC20:05
*** electrocucaracha has quit IRC20:06
*** marzif_ has quit IRC20:07
*** VW_ has joined #openstack-nova20:07
*** electrocucaracha has joined #openstack-nova20:08
*** kiran-r has quit IRC20:09
*** avishay_ has quit IRC20:09
*** rwsu has quit IRC20:10
*** VW_ has quit IRC20:12
melwittalaski: in #openstack-qa I asked mtreinish about whether cells job should vote on tempest changes, he asks if we consider it critical?20:12
mtreinishmelwitt: I left a comment on the review as to that too20:13
melwittmtreinish: cool, thanks20:14
*** VW_ has joined #openstack-nova20:14
dansmithI wouldn't call it critical20:14
*** Mike_D_laptop has joined #openstack-nova20:14
dansmithit'd be nice not to break us with tempest things, but..20:14
dansmithmaybe we see how it goes and if things are breaking often, we revisit?20:14
dansmithjust MHO20:15
* dansmith realizes he wasn't asked20:15
melwitt:D20:15
mtreinishdansmith: yeah that's what I was thinking too20:15
alaskiheh20:15
alaskiI'm more concerned about tempest than devstack.  but I'm fine with seeing how often we break20:15
*** Marga_ has quit IRC20:15
alaskibased on earlier efforts it would have happened 2-3 times last cycle I think20:16
dansmithyeah, that doesn't seem too bad20:16
mtreinishalaski: well it was just that one big change to how networks were used right? it just took us a couple tries to get right :)20:16
dansmithI mean, it sucks, but..20:16
mtreinishalaski: I mean ideally the api should behave the same regardless of cells...20:17
* mtreinish sticks his head in the sand20:17
alaskimtreinish: hah.  I agree, but also have to live with our reality :)20:17
alaskimtreinish: it was mainly the network one, but also something about adding security groups to a test iirc20:18
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/19432520:19
*** afazekas has quit IRC20:19
jaypipesmelwitt: the patch series from dansmith starting here: https://review.openstack.org/#/c/191994 could use your review if you have a minute or two. thx!20:19
dansmithmelwitt: he meant a minute or 3020:20
melwittjaypipes, dansmith: sure, will do20:20
jaypipeslol20:20
melwitt:)20:20
*** belmoreira has quit IRC20:21
*** afazekas has joined #openstack-nova20:21
*** mriedem_afk is now known as mriedem20:24
*** Sukhdev has joined #openstack-nova20:27
mriedemmelwitt: alaski: mtreinish: doesn't the cells tempest job have a big regex of excluded tests?20:28
mtreinishmriedem: yes I think it still does20:28
alaskiyes20:28
melwittmriedem: yes20:28
mriedem^ is kind of why i wanted cells on tempest changes20:28
mriedemb/c someone adds a new tests that does a thing that one of those excluded tests does and it merges and then nova is f'cked20:28
mtreinishideally it should be using config flags to specify that, but whatever20:28
mriedemmtreinish: right20:28
*** vladikr has quit IRC20:28
mriedemmtreinish: so maybe that's a compromise20:28
*** vladikr has joined #openstack-nova20:29
mriedemif there is a certain config thing that should happen in tempest if you're doing cells, and we can cleanup the blacklist that way, then tempest doesn't need to gate on cells20:29
*** mwagner_afk has quit IRC20:29
mtreinishmriedem: well I think I was outvoted on the regex vs adding feature-enabled options thing20:29
*** Anticime1 is now known as Anticimex20:29
mtreinishmriedem: I think maybe just a cells option is to broad a flag to add to tempest though20:29
mtreinishalthough maybe not, if you think that's something an end user needs to know about a deployment20:30
mtreinishbecause that's what those options basically codify20:30
*** jasondot_ has quit IRC20:33
*** Sukhdev has quit IRC20:33
*** Sukhdev has joined #openstack-nova20:36
*** Redcavalier has quit IRC20:36
mriedemcells would be easy but no one would know when to use it20:37
mriedemas a flag i mean20:37
*** electrocucaracha has quit IRC20:37
*** tongli has joined #openstack-nova20:37
*** electrocucaracha has joined #openstack-nova20:38
*** dprince has quit IRC20:38
*** mestery has joined #openstack-nova20:38
BobBallAny20:38
*** armax has joined #openstack-nova20:38
mriedemhttp://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/devstack-gate.yaml#n100420:38
mriedemthat's a lot of tribal knowledge type shit20:38
*** exploreshaifali has joined #openstack-nova20:39
bauzasmriedem: uh, don't blame the file then20:39
bauzas:)20:39
*** tongli has quit IRC20:40
mriedemso i see like resize tests skipped in there, there is a flag in tempest.conf for not running resize tests20:40
mriedemwhy didn't we just use that?20:41
mriedemthere should soon be a flag for encrypted volume tests: https://review.openstack.org/#/c/193831/20:41
mriedemif i get my way20:41
bauzasmriedem: to be clear, I was not aware of any tempest.conf flag20:42
bauzasmriedem: so we can probably try to change some lines20:42
*** Marga_ has joined #openstack-nova20:43
mriedembauzas: yeah, so i guess what i'd try and do is remove as much of the regex as possible with using existing config flags in tempest20:43
bauzasmriedem: sounds good to me20:43
*** dboik has quit IRC20:43
mriedemthen with what's left, if there are particular features that just don't work with those tests, we could add new config option(s) to tempest to skip those for the cells job20:43
*** Mike_D_laptop has quit IRC20:43
*** dboik has joined #openstack-nova20:43
alaskiso one argument for running cells on tempest would be easy testing for removing/modifying exclusions20:44
mriedemif they are things that cells 'should' be able to do, but are bugs, then we could skip based on the bug until the bug is fixed20:44
alaskibut an experimental job could do that20:44
*** lpetrut has quit IRC20:44
mriedemalaski: mtreinish: there is a cells job on tempest experimental queue now right?20:44
alaskiI believe so20:44
bauzasalaski: I like the idea to use tempest.conf flags for failure domains20:44
bauzasmriedem: nope, I moved it20:44
mriedemmoved it?20:45
bauzasmriedem: from the experimental to the check ppl20:45
bauzasmriedem: unless someone added it back20:45
*** yamahata has quit IRC20:46
mriedembauzas: i see the cells job in tempest's experimental queue here: https://review.openstack.org/#/c/190894/3/zuul/layout.yaml20:46
mriedemin melwitt's change to make it voting on nova20:46
mriedemcells is also in the experimental queue for devstack chagnes20:46
bauzasmmm20:47
mriedemone thing that kind of sucks is removing the regexes is going to be hard for things that we're not sure of, since project-config isn't running changes against a cells job20:47
mriedemso we won't be totally sure of all of these removals20:47
*** Marga_ has quit IRC20:47
jrolljohnthetubaguy: (or whoever else) working on neutron/ironic integration, are we going to need a spec here? see lines 59-79 or so: https://review.openstack.org/#/c/187829/4/specs/liberty/network-provider.rst20:48
jrolldansmith: ^ talked to you a bit about this recently20:48
jrollI can definitely do a blueprint, will do spec if necessary, tl;dr adding a flag to the port-create call20:48
bauzasmriedem: don't see the job in the experimental lines20:48
mriedembauzas: line 3644 https://review.openstack.org/#/c/190894/3/zuul/layout.yaml20:49
*** Sukhdev_ has joined #openstack-nova20:49
bauzasmriedem: see my change when I moved the job https://review.openstack.org/#/c/149410/1/zuul/layout.yaml,cm20:49
*** afazekas has quit IRC20:50
jrolljohnthetubaguy: dansmith something like https://review.openstack.org/#/c/186855/ . I'll just write a spec today unless you tell me otherwise :)20:50
mriedemalaski: bauzas: melwitt: mtreinish: are we generally in agreement that the cells job should be skipping things it doesn't work with using the options that tempest already has, like resize?20:50
mriedembauzas: oh, so that just landed20:50
mriedemoh nvm, jan != jun :)20:50
mriedembauzas: that's nova, i'm talking about tempest20:50
bauzasmriedem: aaaaaaarh20:51
alaskimriedem: yes, let's use the proper methods for skipping when available.20:51
melwitt+120:51
* bauzas should write a book "how to discuss 30 mins about something useless"20:51
mriedemgoing through the list of regexed out tests and what tempest options are available is going to take some time, some of those options might have to be exposed through devstack changes and changes to devstack-gate, not sure until digging in20:51
alaskimriedem: the trick is that things like resize are supported, but the resize test must have been doing something that isn't supported like security groups or AZs or something20:52
mriedemdevstack change to set the option in tempest.conf, changes in devstack-gate to set the env var in devstack to set the option in tempest.conf20:52
alaskiso some resize tests may be running now, I'm not really sure20:52
*** Sukhdev__ has joined #openstack-nova20:52
bauzasalaski: mriedem: yeah, using a tempest.conf for a failure domain is definitely better, in case someone adds a new test or something like that20:52
*** Sukhdev has quit IRC20:52
mriedemso....my thinking is, make cells voting in nova (check and gate) in the short term, and then work on cleaning up the regex list so the options are there in tempest so we hopefully don't regress,20:53
mriedemif a tempest change slips in which completely blocks nova, then i guess we'll find out and fix or add to the exclude list to unblock20:53
dansmithjroll: I want to know *why* you need to do this and why it needs to be different from other drivers20:54
*** VW_ has quit IRC20:54
dansmithjroll: so if you can explain that in the commit message or a bug, then maybe you don't need anything20:54
jrolldansmith: right, the spec I linked discusses that. it's an ironic spec, would you like a nova mirror? :)20:54
jrollok20:54
dansmithjroll: but otherwise, I don't have enough information to make any sort of  call20:54
*** annashen has joined #openstack-nova20:54
*** nihilifer has quit IRC20:54
dansmithjroll: no not a mirror, but something.. sounds like that isn't landed yet.. presumably you're waiting to make sure both sides are okay with it?20:55
dansmithjroll: I would put a little text in the commit message, link to the unmerged spec, and leave it WIP for now20:55
*** salv-orlando has joined #openstack-nova20:56
*** sdake_ has quit IRC20:56
jrolldansmith: yep, that's the plan. can do, ty sir20:56
*** Sukhdev_ has quit IRC20:56
jrolldansmith: I just don't want to come back after spec deadlines and have someone ask for a spec :)20:56
mriedemalaski: for those resize tests, i guess we could manually verify that running them on trunk nova works or not20:56
dansmithjroll: yes, and I appreciate that :)20:56
*** mestery has quit IRC20:56
*** yamahata has joined #openstack-nova20:56
openstackgerritDan Smith proposed openstack/nova: API support for migration_type  https://review.openstack.org/18320020:56
openstackgerritDan Smith proposed openstack/nova: Make live migration create a migration object record  https://review.openstack.org/18333120:56
openstackgerritDan Smith proposed openstack/nova: Make evacuate leave a record for the source compute host to process  https://review.openstack.org/18335420:56
openstackgerritDan Smith proposed openstack/nova: Add some notifications to the evacuate path  https://review.openstack.org/19437220:57
openstackgerritDan Smith proposed openstack/nova: Make evacuate update the migration object along the way  https://review.openstack.org/19437320:57
mtreinishmriedem: yes, that's fine with me. Fwiw, that was actually my original proposal back when the giant regex was first added20:57
mriedemmtreinish: well you and i are the same person so that makes sense20:58
*** afazekas has joined #openstack-nova20:58
melwitthaha20:59
mtreinishhehe20:59
*** jasondotstar has joined #openstack-nova20:59
alaskimriedem: I would like to go back through and submit reviews to remove some exclusions and see what still fails20:59
mriedemalaski: project-config changes do'nt run against cells though right?21:00
*** annashen has quit IRC21:00
alaskimriedem: bah, that's right21:00
mriedemmaybe we could add the cells job to project-config's experimental queue :)21:00
mriedemwhile working through the list21:00
mriedemor does that open a black hole?21:00
*** marcusvrn_ has joined #openstack-nova21:00
mriedemproject-config doesn't currently have anything in the experimental queue but that should work21:01
alaskifeels heavy for this.  maybe a devstack change to set a modified regex there21:01
ildikovHi Nova Cores! If any of you have the bandwidth, please take a look at this spec: https://review.openstack.org/#/c/182350/ Thanks! :)21:02
*** raildo has quit IRC21:02
alaskimriedem: but I'm good with an experimental job there if the reviewers there are21:02
*** mestery has joined #openstack-nova21:03
mriedemasking fungi in -infra21:03
*** VW_ has joined #openstack-nova21:04
morganfainbergmriedem: FYI just reviewed the instance-user spec. it's not really insane, just have a concern with metadata service use there - once that concern is solved/addressed I can safely +1 it (cc jaypipes )21:04
morganfainbergmriedem: that would be https://review.openstack.org/#/c/186617/ btw21:05
*** aix has joined #openstack-nova21:05
*** flwang has quit IRC21:06
mriedemalaski: ok, so i'll push the initial project-config change to add cells + tempest to d-g exp queue,21:06
mriedemalaski: regarding cells job voting on nova changes, melwitt's patch, do we want to bring that up in the nova meeting or ML to get general agreement that we want it voting even though tempest changes won't be tested against it automatically?21:07
mriedemso things will probably break from time to time until regex exodus happens21:07
mriedemregex exodus (tm) btw21:07
alaskimriedem: yes, it would be good to get buyin from Nova reviewers for that21:08
mriedemok, you want to send that to the ML now or i can do it21:08
*** ivasev has quit IRC21:08
alaskiyou've got the momentum on it21:08
melwittmriedem: shall I -W the patch for now?21:09
*** electrocucaracha has quit IRC21:09
*** rwsu has joined #openstack-nova21:10
mriedemmelwitt: yeah probably21:10
mriedemalaski: ok i'll write that up now21:10
*** MVenesio has quit IRC21:10
mriedemmorganfainberg: ok, probably won't get anywhere near that today21:10
morganfainbergno worries just letting you know i've gotten some comments on it21:11
morganfainbergmriedem: i know you are headed towards the deadline(s), but i think it will need an update before it's ready fwiw21:11
melwittmriedem: okay, cool21:11
*** thorst has quit IRC21:13
*** annashen has joined #openstack-nova21:15
openstackgerritVladik Romanovsky proposed openstack/nova: virt: introduce libosinfo library to set hardware policy  https://review.openstack.org/14962521:15
openstackgerritVladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring network model  https://review.openstack.org/14962721:15
openstackgerritVladik Romanovsky proposed openstack/nova: libvirt: adding libosinfo configuration  https://review.openstack.org/14962621:15
openstackgerritVladik Romanovsky proposed openstack/nova: libvirt: use osinfo when configuring the disk bus  https://review.openstack.org/14962821:15
*** tonytan4ever has quit IRC21:16
*** alex_klimov has joined #openstack-nova21:17
*** iamjarvo has quit IRC21:17
*** dane-fichter has quit IRC21:18
*** ctrath has quit IRC21:19
*** flwang1 has joined #openstack-nova21:19
*** aysyanne has quit IRC21:19
*** marzif_ has joined #openstack-nova21:20
mriedemyar http://lists.openstack.org/pipermail/openstack-dev/2015-June/067723.html21:23
bauzasmmm, I suppose I have to modify a nova.conf flag for testing the EC2 API within devstack?21:23
* bauzas digging21:24
*** karimb has joined #openstack-nova21:27
*** jasondotstar has quit IRC21:27
*** ociuhandu has joined #openstack-nova21:27
bauzasooooooh21:27
*** afazekas has quit IRC21:28
*** Nikolay_St has quit IRC21:28
*** baoli has quit IRC21:28
*** eharney has quit IRC21:32
russellbmriedem: i have a thing for your regex exodus21:33
mriedemdo tell21:33
russellbhttp://lists.openstack.org/pipermail/openstack-dev/2015-June/067724.html21:33
russellbso you can use that to put the cells job regex in a file in the nova repo21:34
*** rlrossit has quit IRC21:36
russellbmriedem: ^21:36
mriedemrussellb: yeah, just read through all the changes21:36
mriedemrussellb: that's what i call fancy pants21:36
mriedemrussellb: nice, i'll use that instead21:36
*** romainh has quit IRC21:37
*** jasondot_ has joined #openstack-nova21:37
*** rfolco_ has quit IRC21:37
*** rfolco has quit IRC21:37
*** electrocucaracha has joined #openstack-nova21:38
russellbmriedem: yay, glad it seems to help21:38
*** thangp has quit IRC21:38
*** dave-mccowan has quit IRC21:39
*** dboik has quit IRC21:40
*** Sukhdev__ has quit IRC21:40
*** dboik has joined #openstack-nova21:41
*** Sukhdev has joined #openstack-nova21:41
*** rlrossit has joined #openstack-nova21:41
*** rlrossit has left #openstack-nova21:42
*** otter768 has joined #openstack-nova21:42
*** zigo has quit IRC21:43
*** obondarev has quit IRC21:43
*** Longgeek has quit IRC21:43
*** brad[] has quit IRC21:43
*** otter768 has quit IRC21:47
*** melwitt has quit IRC21:48
*** zigo has joined #openstack-nova21:49
*** romainh has joined #openstack-nova21:51
*** jbernard has quit IRC21:52
*** VW_ has quit IRC21:52
*** kevinbenton has quit IRC21:52
*** obondarev has joined #openstack-nova21:53
*** lan has quit IRC21:54
*** chaochin has quit IRC21:54
*** telemonster has quit IRC21:54
*** nihilifer has joined #openstack-nova21:54
*** kevinbenton has joined #openstack-nova21:54
*** telemonster has joined #openstack-nova21:55
*** ianbrown has joined #openstack-nova21:56
*** mtanino has quit IRC21:57
*** jbernard has joined #openstack-nova21:58
*** fawadkhaliq has joined #openstack-nova21:59
*** jwcroppe_ has joined #openstack-nova21:59
*** melwitt has joined #openstack-nova22:00
*** jwcroppe has quit IRC22:01
openstackgerritMatt Riedemann proposed openstack/nova: cells: add devstack/tempest-dsvm-cells-rc for gating  https://review.openstack.org/19441122:01
mriedemrussellb: alaski: melwitt: bauzas: mtreinish: ^ kablammo22:01
openstackgerritJim Rollenhagen proposed openstack/nova: Don't bind ironic networks immediately  https://review.openstack.org/18685522:02
*** dboik_ has joined #openstack-nova22:02
melwittmriedem: hey, that's cool22:02
*** matrohon has joined #openstack-nova22:03
*** jwcroppe_ has quit IRC22:03
*** mgarza has quit IRC22:04
*** jwcroppe has joined #openstack-nova22:04
*** chaochin has joined #openstack-nova22:05
mtreinishmriedem: one of these days I'm going to get the tempest ostestr patch landed so we can just use the blacklist file option to make that interface a bit cleaner22:05
openstackgerritEarle F. Philhower, III proposed openstack/nova: Add os_brick-based VolumeDriver for HGST connector  https://review.openstack.org/19441222:06
*** agentle has joined #openstack-nova22:06
*** iamjarvo has joined #openstack-nova22:06
*** oro has joined #openstack-nova22:06
mriedemmtreinish: i was already cursing ostestr last week22:06
*** dboik has quit IRC22:06
*** lan has joined #openstack-nova22:06
mtreinishmriedem: yeah I caught part of that in the back log this morning22:06
mtreinishI'm not sure what I screwed up though22:06
mtreinishbesides probably everything22:06
mriedemmtreinish: well, -n22:07
mriedemand docs :)22:07
*** dave-mccowan has joined #openstack-nova22:07
*** dboik_ has quit IRC22:07
mriedemi had to dig into the code to figure out how to run a single test case22:07
*** lyarwood has quit IRC22:07
*** lyarwood has joined #openstack-nova22:08
mtreinishmriedem: well there is a --help/-h that includes some details22:08
mtreinishbut yeah docs are a todo22:08
*** afazekas has joined #openstack-nova22:08
*** lucas-dinner has quit IRC22:08
*** oro has quit IRC22:08
mtreinishmaybe you could help?22:08
*** oro has joined #openstack-nova22:08
*** TemporalBeing has left #openstack-nova22:09
openstackgerritJim Rollenhagen proposed openstack/nova: Allow virt driver to define binding:host_id  https://review.openstack.org/19441322:09
mriedemmtreinish: i knew that was coming :)22:10
*** jwcroppe_ has joined #openstack-nova22:10
mtreinishmriedem: think of the possibilities, 2 Matt's working on the same thing22:11
*** fawadkhaliq has quit IRC22:12
*** Longgeek has joined #openstack-nova22:12
* melwitt ponders22:12
*** jasondot_ has quit IRC22:12
mriedemmore black holes22:12
*** jwcroppe has quit IRC22:13
*** rushiagr_away has quit IRC22:13
*** afazekas has quit IRC22:13
*** zigo has quit IRC22:15
*** marzif__ has joined #openstack-nova22:16
*** agireud has quit IRC22:17
*** zigo has joined #openstack-nova22:17
*** ijuwang is now known as ijuwang_away22:18
*** exploreshaifali has quit IRC22:19
*** marzif_ has quit IRC22:19
*** marzif__ has quit IRC22:21
*** fawadkhaliq has joined #openstack-nova22:24
*** mestery has quit IRC22:25
*** tjones1 has joined #openstack-nova22:27
*** armax has quit IRC22:27
*** mdrabe has quit IRC22:27
*** earlephilhower has joined #openstack-nova22:27
*** rushiagr_away has joined #openstack-nova22:28
*** matrohon has quit IRC22:28
*** dave-mccowan has quit IRC22:31
openstackgerritWalter A. Boring IV (hemna) proposed openstack/nova: Switch to using os-brick  https://review.openstack.org/17556922:31
*** electrocucaracha has quit IRC22:32
*** isd has quit IRC22:34
*** husanu has quit IRC22:38
*** husanu has joined #openstack-nova22:40
*** aix has quit IRC22:41
*** aix has joined #openstack-nova22:41
openstackgerritStanislaw Pitucha proposed openstack/nova: Fix comment typo  https://review.openstack.org/19442622:41
*** salv-orlando has quit IRC22:42
*** jasondotstar has joined #openstack-nova22:43
*** radek__ has quit IRC22:44
openstackgerritEarle F. Philhower, III proposed openstack/nova: Add os_brick-based VolumeDriver for HGST connector  https://review.openstack.org/19441222:49
*** agentle has quit IRC22:50
*** alex_klimov has quit IRC22:51
*** agentle has joined #openstack-nova22:51
*** jecarey has quit IRC22:53
*** electrocucaracha has joined #openstack-nova22:54
*** iamjarvo has quit IRC22:54
*** pixelb has quit IRC22:58
*** romainh has left #openstack-nova22:59
*** rwsu has quit IRC23:00
*** ssurana has quit IRC23:01
*** asettle has joined #openstack-nova23:01
*** jdurgin has quit IRC23:04
*** ZZelle_ has quit IRC23:10
*** asettle has quit IRC23:12
*** thedodd has quit IRC23:13
*** ndipanov has quit IRC23:13
*** marcusvrn_ has quit IRC23:14
*** markvoelker has quit IRC23:16
*** melwitt_ has joined #openstack-nova23:22
*** vladikr has quit IRC23:22
*** oro has quit IRC23:23
*** melwitt has quit IRC23:27
*** melwitt_ is now known as melwitt23:27
*** thorst has joined #openstack-nova23:27
*** jasondotstar has quit IRC23:29
*** thorst has quit IRC23:29
*** fawadkhaliq has quit IRC23:31
*** annashen has quit IRC23:31
*** karimb has quit IRC23:32
openstackgerritWalter A. Boring IV (hemna) proposed openstack/nova: Switch to using os-brick  https://review.openstack.org/17556923:32
*** dguitarbite has quit IRC23:33
*** LukeHinds has joined #openstack-nova23:34
LukeHindshello *23:34
LukeHindsthis is driving me mad, anyone have any ideas https://ask.openstack.org/en/question/67724/kilo-assigning-ipv6-even-though-its-disabled/23:34
*** annashen has joined #openstack-nova23:34
LukeHindsI just redeployed and I keep getting ipv6 addresses assigned, even though its disabled23:35
LukeHindsjust followed this to the T and still have the same https://www.rdoproject.org/Running_an_instance_with_Neutron23:36
*** vilobhmm has quit IRC23:36
*** vilobhmm has joined #openstack-nova23:36
*** cjwei has joined #openstack-nova23:37
*** claudiub has quit IRC23:38
openstackgerritWalter A. Boring IV (hemna) proposed openstack/nova: Add os_brick-based VolumeDriver for HGST connector  https://review.openstack.org/19441223:39
*** Sukhdev has quit IRC23:41
LukeHindssorry all, just noted this is more of a dev channel, sorry for disturbing.23:41
*** Longgeek has quit IRC23:43
*** otter768 has joined #openstack-nova23:43
*** ijw__ has joined #openstack-nova23:43
*** agentle has quit IRC23:45
openstackgerritKevin Fox proposed openstack/nova-specs: Instance Users for Cloud Interaction  https://review.openstack.org/18661723:45
*** jaypipes has quit IRC23:46
*** gmann_ has quit IRC23:46
*** annashen has quit IRC23:46
*** ijw_ has quit IRC23:47
*** otter768 has quit IRC23:48
*** gmann has joined #openstack-nova23:48
*** ijw_ has joined #openstack-nova23:53
*** jasondotstar has joined #openstack-nova23:54
*** husanu has quit IRC23:55
*** dave-mccowan has joined #openstack-nova23:56
*** ijw__ has quit IRC23:56
*** husanu has joined #openstack-nova23:57
*** vilobhmm has quit IRC23:57
*** lyarwood has quit IRC23:58
*** lyarwood has joined #openstack-nova23:58
openstackgerritKevin Fox proposed openstack/nova-specs: Instance Users for Cloud Interaction  https://review.openstack.org/18661723:58

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