Monday, 2017-03-13

*** adrian_otto has quit IRC00:00
*** adrian_otto has joined #openstack-nova00:09
*** gfhellma has joined #openstack-nova00:10
*** mnaser has left #openstack-nova00:12
*** baoli has quit IRC00:13
*** adrian_otto has quit IRC00:18
*** jamielennox is now known as jamielennox|away00:20
*** catintheroof has joined #openstack-nova00:26
*** catintheroof has quit IRC00:26
*** catintheroof has joined #openstack-nova00:26
*** jamielennox|away is now known as jamielennox00:27
*** baoli has joined #openstack-nova00:28
*** baoli has quit IRC00:28
*** yingjun has joined #openstack-nova00:31
openstackgerritAlex Xu proposed openstack/nova master: placement: Enable attach traits to ResourceProvider  https://review.openstack.org/37620100:31
openstackgerritAlex Xu proposed openstack/nova master: placement: Add Traits API to placement service  https://review.openstack.org/37620000:31
openstackgerritAlex Xu proposed openstack/nova master: placement: Add resource provider traits API  https://review.openstack.org/37620200:31
openstackgerritAlex Xu proposed openstack/nova master: placement: Add Trait and TraitList objects  https://review.openstack.org/37619900:31
*** raunak has quit IRC00:33
*** yamamoto has joined #openstack-nova00:39
*** namnh has joined #openstack-nova00:42
*** fragatina has joined #openstack-nova00:43
*** dave-mccowan has joined #openstack-nova00:59
*** phuongnh has joined #openstack-nova01:06
*** zhurong has joined #openstack-nova01:11
*** tbachman_ has joined #openstack-nova01:12
*** tbachman has quit IRC01:13
*** tbachman_ is now known as tbachman01:13
*** Shunli has joined #openstack-nova01:21
*** yamamoto has quit IRC01:25
*** baoli has joined #openstack-nova01:29
*** amotoki has quit IRC01:30
*** zz_dimtruck is now known as dimtruck01:32
*** mriedem has quit IRC01:33
*** armax has joined #openstack-nova01:34
*** wangqun has joined #openstack-nova01:34
*** baoli has quit IRC01:34
*** kevinz has joined #openstack-nova01:39
*** mtanino has joined #openstack-nova01:40
*** salv-orlando has joined #openstack-nova01:41
*** Kevin_Zheng has joined #openstack-nova01:41
*** dimtruck is now known as zz_dimtruck01:42
*** amotoki has joined #openstack-nova01:43
*** gcb has joined #openstack-nova01:44
openstackgerritLi Yingjun proposed openstack/nova-specs master: Spec for support set vgamem for video  https://review.openstack.org/43878701:45
*** salv-orlando has quit IRC01:45
*** yonglihe has joined #openstack-nova01:46
*** amotoki has quit IRC01:49
*** baoli has joined #openstack-nova01:50
*** amotoki has joined #openstack-nova01:55
*** baoli has quit IRC01:55
openstackgerritLi Yingjun proposed openstack/nova-specs master: Spec for support set vgamem for video  https://review.openstack.org/43878701:56
*** armax has quit IRC01:57
*** youzp has joined #openstack-nova01:57
*** siva_krishnan has left #openstack-nova01:57
*** youzp has quit IRC01:58
*** youzp has joined #openstack-nova01:58
*** yuntongjin has joined #openstack-nova01:59
*** iceyao has joined #openstack-nova02:05
*** yamamoto has joined #openstack-nova02:05
*** kaisers_ has joined #openstack-nova02:10
*** kaisers__ has quit IRC02:14
*** gouthamr has quit IRC02:17
*** catintheroof has quit IRC02:18
*** amotoki has quit IRC02:19
*** amotoki has joined #openstack-nova02:29
*** amotoki has quit IRC02:30
*** amotoki has joined #openstack-nova02:30
*** gongysh has joined #openstack-nova02:33
*** zz_dimtruck is now known as dimtruck02:33
openstackgerritMaho Koshiya proposed openstack/nova master: Add interfaces functional negative tests  https://review.openstack.org/44289202:38
*** takashin has joined #openstack-nova02:41
*** dimtruck is now known as zz_dimtruck02:43
*** irenab_ is now known as irenab02:43
*** catintheroof has joined #openstack-nova02:45
*** catintheroof has quit IRC02:45
*** catintheroof has joined #openstack-nova02:45
*** baoli has joined #openstack-nova02:51
*** yuntongjin has quit IRC02:56
*** baoli has quit IRC02:56
*** ircuser-1 has joined #openstack-nova02:57
*** carthaca_ has quit IRC03:00
*** sapcc-bot has quit IRC03:01
*** databus23_ has quit IRC03:01
*** sapcc-bot has joined #openstack-nova03:01
*** databus23_ has joined #openstack-nova03:01
*** carthaca_ has joined #openstack-nova03:01
*** tuanluong has joined #openstack-nova03:06
*** carthaca_ has quit IRC03:11
*** sapcc-bot has quit IRC03:11
*** databus23_ has quit IRC03:11
*** carthaca_ has joined #openstack-nova03:11
*** sapcc-bot has joined #openstack-nova03:11
*** databus23_ has joined #openstack-nova03:11
*** raunak has joined #openstack-nova03:15
*** sapcc-bot3 has joined #openstack-nova03:17
*** databus23_ has quit IRC03:17
*** sapcc-bot has quit IRC03:17
*** carthaca_ has quit IRC03:17
*** carthaca_1 has joined #openstack-nova03:17
*** databus23_1 has joined #openstack-nova03:17
*** nicolasbock has quit IRC03:21
*** zz_dimtruck is now known as dimtruck03:25
*** vishwanathj has quit IRC03:30
*** adrian_otto has joined #openstack-nova03:31
*** dave-mccowan has quit IRC03:32
*** bswartz has joined #openstack-nova03:32
*** catintheroof has quit IRC03:37
*** sree has joined #openstack-nova03:39
*** mtanino has quit IRC03:40
*** claudiub has quit IRC03:40
*** adrian_otto has quit IRC03:41
*** salv-orlando has joined #openstack-nova03:42
*** salv-orlando has quit IRC03:46
*** adrian_otto has joined #openstack-nova03:50
*** adrian_otto has quit IRC03:55
*** nmathew has joined #openstack-nova03:57
*** nmathew- has joined #openstack-nova03:59
*** links has joined #openstack-nova04:02
*** nmathew has quit IRC04:02
openstackgerritTakashi NATSUME proposed openstack/nova master: api-ref: Fix parameters and examples in aggregate API  https://review.openstack.org/42551204:04
openstackgerritTakashi NATSUME proposed openstack/nova master: Avoid forcing translation on logging calls  https://review.openstack.org/41387604:04
openstackgerritTakashi NATSUME proposed openstack/nova master: api-ref: Fix parameters in os-hosts.inc  https://review.openstack.org/41644404:05
openstackgerritTakashi NATSUME proposed openstack/nova-specs master: Abort Cold Migration  https://review.openstack.org/33473204:05
*** fragatina has quit IRC04:10
*** mdnadeem has joined #openstack-nova04:25
*** psachin has joined #openstack-nova04:28
*** iceyao has quit IRC04:29
*** iceyao has joined #openstack-nova04:31
*** raunak has quit IRC04:33
*** markvoelker has quit IRC04:35
*** raunak has joined #openstack-nova04:41
*** sudipto has joined #openstack-nova04:45
*** nmathew- has quit IRC04:49
*** ayogi has joined #openstack-nova04:49
*** sridharg has joined #openstack-nova04:49
*** baoli has joined #openstack-nova04:53
*** baoli has quit IRC04:57
*** kaisers_ has quit IRC04:59
*** takashin has quit IRC05:01
*** ssurana has joined #openstack-nova05:11
*** kornicameister has joined #openstack-nova05:12
*** lpetrut has joined #openstack-nova05:12
*** crusio has quit IRC05:15
*** nmathew has joined #openstack-nova05:15
*** crusio has joined #openstack-nova05:15
*** jbernard has quit IRC05:16
*** jbernard has joined #openstack-nova05:17
*** sridharg has quit IRC05:19
*** thorst_afk has joined #openstack-nova05:19
*** thorst_afk has quit IRC05:19
*** trinaths has joined #openstack-nova05:23
*** thorst_afk has joined #openstack-nova05:28
*** kaisers_ has joined #openstack-nova05:29
*** thorst_afk has quit IRC05:32
*** kaisers_ has quit IRC05:34
*** markvoelker has joined #openstack-nova05:36
*** markvoelker has quit IRC05:41
*** salv-orlando has joined #openstack-nova05:43
*** thorst_afk has joined #openstack-nova05:43
*** lpetrut has quit IRC05:47
*** salv-orlando has quit IRC05:47
*** thorst_afk has quit IRC05:47
*** baoli has joined #openstack-nova05:54
*** unicell has joined #openstack-nova05:57
*** baoli has quit IRC05:58
*** unicell has quit IRC05:59
*** tovin07 has joined #openstack-nova06:00
*** salv-orlando has joined #openstack-nova06:02
*** unicell has joined #openstack-nova06:04
*** tovin07_ has joined #openstack-nova06:13
*** unicell has quit IRC06:20
*** raunak has quit IRC06:25
*** raunak has joined #openstack-nova06:26
*** raunak has quit IRC06:26
*** raunak has joined #openstack-nova06:27
*** raunak has quit IRC06:27
*** raunak has joined #openstack-nova06:27
*** raunak has quit IRC06:28
*** raunak has joined #openstack-nova06:28
*** raunak has quit IRC06:29
*** raunak has joined #openstack-nova06:29
*** raunak has quit IRC06:29
*** kaisers_ has joined #openstack-nova06:31
*** dimtruck is now known as zz_dimtruck06:39
*** mnestratov has joined #openstack-nova06:40
*** ducnc has joined #openstack-nova06:42
*** sridharg has joined #openstack-nova06:42
*** ekuris has joined #openstack-nova06:46
openstackgerritLi Yingjun proposed openstack/nova-specs master: Spec for service create and destroy notification  https://review.openstack.org/44473106:47
*** baoli has joined #openstack-nova06:54
openstackgerritLi Yingjun proposed openstack/nova-specs master: Spec for service create and destroy notification  https://review.openstack.org/44473106:56
*** baoli has quit IRC06:59
*** sree has quit IRC07:00
*** _ioni is now known as ioni07:01
*** Oku_OS-away is now known as Oku_OS07:02
*** kaisers_ has quit IRC07:03
*** Shunli has quit IRC07:06
*** Shunli has joined #openstack-nova07:07
*** nkrinner_afk is now known as nkrinner07:08
*** satyar has joined #openstack-nova07:09
*** markus_z has joined #openstack-nova07:14
*** lpetrut has joined #openstack-nova07:15
*** salv-orl_ has joined #openstack-nova07:24
*** salv-orlando has quit IRC07:26
*** kevinz has quit IRC07:30
*** kevinz has joined #openstack-nova07:32
*** John341_ has quit IRC07:32
*** salv-orl_ has quit IRC07:33
*** andreas_s has joined #openstack-nova07:38
*** karimb has joined #openstack-nova07:42
*** iceyao has quit IRC07:47
*** claudiub has joined #openstack-nova07:48
*** ekuris has quit IRC07:50
*** iceyao has joined #openstack-nova07:50
*** ralonsoh has joined #openstack-nova07:51
*** belmoreira has joined #openstack-nova07:54
*** iceyao has quit IRC07:54
*** kaisers_ has joined #openstack-nova07:58
*** zzzeek has quit IRC08:00
*** zzzeek has joined #openstack-nova08:00
*** hshiina has joined #openstack-nova08:00
*** salv-orlando has joined #openstack-nova08:04
*** tovin07 has quit IRC08:07
*** zioproto has joined #openstack-nova08:11
*** ssurana has quit IRC08:11
*** tovin07 has joined #openstack-nova08:12
*** tesseract has joined #openstack-nova08:13
*** tesseract has quit IRC08:17
*** openstackgerrit has quit IRC08:18
*** kaisers_ has quit IRC08:25
*** pcaruana has joined #openstack-nova08:25
Qimingalex_xu, there?08:27
robcresswellQuick question; someone's added a patch to Horizon that hides the Soft Reboot button if the Instance state is anything other than Active; is this correct? Had a look at the API docs but they only seem to show how to form the request, not any of the conditions around its usage.08:28
*** openstackgerrit has joined #openstack-nova08:31
openstackgerritSergey Nikitin proposed openstack/nova master: Replace diagnostics objects with Nova diagnostics objects  https://review.openstack.org/39448008:31
openstackgerritSergey Nikitin proposed openstack/nova master: Added nova objects for intance diagnostics  https://review.openstack.org/43474508:31
openstackgerritSergey Nikitin proposed openstack/nova master: Standardization of VM diagnostics info API.  https://review.openstack.org/35554008:31
openstackgerritSergey Nikitin proposed openstack/nova master: Completed implementation of instance diagnostics for Xen  https://review.openstack.org/39961308:31
*** tesseract has joined #openstack-nova08:32
*** jpena|off is now known as jpena08:38
*** abalutoiu_ has quit IRC08:41
*** jpena is now known as jpena|off08:51
*** VAhl has joined #openstack-nova08:54
*** jaosorior has joined #openstack-nova08:57
*** abalutoiu_ has joined #openstack-nova08:58
*** jpena|off is now known as jpena08:58
bauzasgood morning Nova09:00
alex_xuQiming: hi, i'm here09:02
*** VAhl has quit IRC09:03
*** ekuris has joined #openstack-nova09:04
*** trinaths1 has joined #openstack-nova09:04
*** trinaths has quit IRC09:05
Qimingalex_xu, I was checking admin password09:08
*** tuanluong has quit IRC09:09
*** yamamoto has quit IRC09:10
Qimingalex_xu, it turns out that I will need to use "/servers/{server_id}/os-server-password" for password GET, and DELETE, use POST /servers/{server_id}/action for password update, and the password can be created only via POST /servers09:11
*** aarefiev_afk is now known as aarefiev09:19
*** lucas-afk is now known as lucasagomes09:20
*** Guest91536 has quit IRC09:21
openstackgerritGhanshyam Mann proposed openstack/nova master: Typo fix in releasenotes: deprecate network options  https://review.openstack.org/44477809:24
*** Guest91536 has joined #openstack-nova09:25
gmannsfinucan: alex_xu oomichi simple one - https://review.openstack.org/#/c/444778/09:25
*** dakhmetov has joined #openstack-nova09:26
sfinucangmann: Done09:26
* sfinucan didn't want to lose +2s :)09:26
gmannsfinucan: :) thanks09:27
openstackgerritLee Yarwood proposed openstack/nova master: compute: Move detach logic from manager into driver BDM  https://review.openstack.org/43952009:27
openstackgerritLee Yarwood proposed openstack/nova master: compute: Use detach_volume during default rebuild  https://review.openstack.org/44393209:27
openstackgerritLee Yarwood proposed openstack/nova master: objects: Add attachment_id to BlockDeviceMapping  https://review.openstack.org/43766509:27
openstackgerritLee Yarwood proposed openstack/nova master: compute: Only destroy BDMs after successful detach call  https://review.openstack.org/44069309:27
openstackgerritLee Yarwood proposed openstack/nova master: db: Add attachment_id to block_device_mapping  https://review.openstack.org/43759709:27
* alex_xu have same thinking with sfinucan09:28
alex_xugmann: done09:28
alex_xuQiming: i'm thinking whether all of those password are same thing09:28
Qimingalex_xu, they are, IMO09:28
*** efoley_ has joined #openstack-nova09:29
gmannalex_xu: thanks09:29
gmannalex_xu: Qiming admin one is different(updated using changePassword) and os-server-password is metadata one09:30
*** efoley__ has joined #openstack-nova09:31
gmannbecause delete call just remvoe from metadata which is encrypted one, it does not change in server09:31
Qiminggmann, I'm no nova expert, should I care if the password is stored in metadata or somewhere else?09:32
Qiminggmann, the doc tells me I can specify password for admin, I can change it, I can get it, and I can wipe it out09:33
alex_xuI remember the admin one is for xen driver09:33
* gmann checking09:33
*** efoley_ has quit IRC09:34
Qiminggmann, the strange thing is that they are located in different locations, probably fine for command line tools, but not so good for SDK modeling09:34
*** Shunli has quit IRC09:35
lyarwoodmdbooth: morning, https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/cinder-new-attach-apis should be ready for reviews once you have time today, the focus is really on the db and object changes first if you wouldn't mind focusing on them.09:37
openstackgerritThomas Bechtold proposed openstack/nova master: Log exception when hypervisor version check fails  https://review.openstack.org/43492709:38
*** tovin07 has quit IRC09:39
*** derekh has joined #openstack-nova09:42
dakhmetovGreetings, guys!09:45
dakhmetovCould you please review this patch: https://review.openstack.org/#/c/399679/09:45
*** yingjun has quit IRC09:47
dakhmetovI know a couple of people who are really affected by the inconsistency in instance's libvirt.xml. It will be great if you help me reviewing it.09:50
*** baoli has joined #openstack-nova09:52
*** VAhl has joined #openstack-nova09:53
johnthetubaguydakhmetov: so I love half of your change, and worry about the other half09:54
johnthetubaguydakhmetov: using instance.user_id and instance.project_id totally makes sense, we should have done that for the beginning09:54
johnthetubaguyusername and projectname... thats feels wrong, what about we don't add that any more?09:55
dakhmetov2 johnthetubaguy please share your concern09:55
johnthetubaguyI will add a note on the patch09:55
gmannQiming: admin pass can be injected while creating server and change password using action API is for xen and libvirt as alex_xu mentioned09:55
johnthetubaguydakhmetov: a related comment came up on friday at a HPC meetup I went to, so agreed it needs fixing09:55
openstackgerritMichael  Henkel proposed openstack/nova master: Implements: blueprint opencontrail-dpdk-support  https://review.openstack.org/44118309:55
*** baoli has quit IRC09:56
gmann@Qiming  not sure about exact SDK requirement on that and if any issue on nova side?09:56
*** gszasz has joined #openstack-nova09:57
johnthetubaguydakhmetov: I am worried about how usernames and project names change over time in keystone, but we keep those names just matching what they were went the instance booted09:58
johnthetubaguydakhmetov: maybe thats OK for your use case I guess?09:58
dakhmetov2 johnthetubaguy: yes, I also thought about it. The only option to avoid this issue is open Keystone session there each time10:00
johnthetubaguydakhmetov: my idea was to only have the ids in there, from the instance object, and leave the others blank for now?10:01
johnthetubaguythat would ensure they are "correct" in some sense, but I feel bad breaking the folks that want the names there10:02
*** satyar has quit IRC10:04
*** andreas_s has quit IRC10:04
dakhmetovjohnthetubaguy: sure, there are a lot of folks who needs names there, as it described in nova xmlns 1.010:04
dakhmetovso, what we have in current patch seems to be best solution, despite the fact that the stored user and project names can change10:07
mdboothdakhmetov: I've reviewed a change related to this recently. Do you have a link?10:08
dakhmetovmdbooth: exactly! https://review.openstack.org/#/c/39967910:08
*** iceyao has joined #openstack-nova10:09
mdboothAh, ok. You fishing for a second +2?10:09
dakhmetovmdbooth: yes, a kind of :)10:09
* mdbooth sees comments from johnthetubaguy and reads10:09
*** hshiina has quit IRC10:10
mdboothjohnthetubaguy: My take on this is that we break things by rebuilding the libvirt xml. We shouldn't be doing that, and this is a necessary workaround.10:11
mdboothi.e. the conversation isn't so much about what these values mean, whether they're correct, or what they really should be10:11
mdboothBut more: are they the same as before10:11
mdboothThe other conversation is also interesting, though.10:12
johnthetubaguymdbooth: my worry is the bug that says I updated the project or user name in keystone, and now my lib virt xml is out of sync, for me its either we don't include the names or we do what dakhmetov has proposed10:12
johnthetubaguytotally agreed grabbing it from the context was exactly the wrong thing to do10:13
* mdbooth reads the bug again10:13
johnthetubaguyI was meaning the bug that is raised next time, after we fix this bit10:14
mdboothAh, ok.10:14
mdboothSo, that bug was always present10:14
johnthetubaguyyep10:15
mdboothThis information is static, so storing dynamic information there is suspect at best.10:15
johnthetubaguyyeah, thats my take10:15
mdboothAnd I agree, and probably that we should remove username and projectname10:15
mdboothHowever, that's tangential to this bug, which is that we break our existing contract10:15
johnthetubaguycool, I was wondering if I was going down a crazy thought path10:15
*** rmart04 has joined #openstack-nova10:16
mdboothdakhmetov: Incidentally, if you're close to users who are actually relying on this, I'd be interested to know what your take is on how to fix the changing username issue.10:17
johnthetubaguy+1 thats what I was fishing for really10:17
*** gszasz has quit IRC10:17
mdboothdakhmetov: Is the issue that people have scripted this, and scripts will require updating?10:18
johnthetubaguyFWIW, this change breaks those folks too10:19
* mdbooth suspects that nobody actually checks the xml namespace in use, so bumping it would be academic.10:19
*** andreas_s has joined #openstack-nova10:19
mdboothjohnthetubaguy: Really? I missed that.10:19
johnthetubaguyall existing instance, once modified will now have no username or projectname, we are missing a unit test for that10:19
*** kaisers_ has joined #openstack-nova10:20
mdboothjohnthetubaguy: Ah, yes. It's a fix only for future instances, I'd glossed over that.10:20
mdboothWe could save the info before deleting it...10:21
mdboothMeh, lets get this done before gold plating it.10:21
*** zhurong has quit IRC10:21
*** zhurong has joined #openstack-nova10:22
*** sree has joined #openstack-nova10:23
*** owalsh_ is now known as owalsh10:23
*** nicolasbock has joined #openstack-nova10:24
*** alexpilotti has joined #openstack-nova10:24
johnthetubaguydakhmetov: I updated my comments to try summarise what I am thinking now, basically I would split this into two patches, and the second patch needs an extra unit test.10:25
mdboothdakhmetov: Can you quickly write the test case johnthetubaguy has asked for? Should be simple enough and probably just works, but we should definitely be testing that.10:25
*** kaisers has quit IRC10:26
*** kevinz has quit IRC10:26
*** shaohe_feng has quit IRC10:26
*** sree has quit IRC10:27
*** rmart04 has quit IRC10:28
*** phuongnh has quit IRC10:29
*** tovin07_ has quit IRC10:29
alex_xuQiming: gmann, yea, i mean it even worse, they have similar name, but doing different thing10:32
*** hwoarang_ is now known as hwoarang10:33
*** gongysh has quit IRC10:36
*** lifeless has quit IRC10:37
dakhmetovmdbooth: johnthetubaguy: reading your comments10:40
*** nicolasbock has quit IRC10:41
*** ralonsoh_ has joined #openstack-nova10:41
*** markvoelker has joined #openstack-nova10:41
dakhmetovas workaround for outdated names in instance's system_metadata: names can be updated in mysql10:41
dakhmetovit's dirty hack, but anyway solves the possible issue10:42
mdboothdakhmetov: I think the safer option is to test that having nothing there at least doesn't break existing instances. That means that we're not regressing. Saving the info before rewriting the xml is better, but also a game of whack-a-mole, so best be safe first I think.10:44
*** ralonsoh has quit IRC10:44
*** lifeless has joined #openstack-nova10:45
*** wangqun has quit IRC10:45
*** ralonsoh_ is now known as ralonsoh10:45
*** iceyao has quit IRC10:45
*** markvoelker has quit IRC10:46
*** namnh has quit IRC10:47
kashyaplyarwood: Isn't melwitt right here -- https://review.openstack.org/#/c/441204/5/nova/tests/unit/virt/libvirt/test_guest.py10:49
dakhmetovmdbooth: So, what I need is to handle the situation with missing owner_user_name and owner_project_name in instance's metadata, as well as add test for this. Right?10:50
kashyapSpecifically her question about 'live=False'10:50
*** kaisers has joined #openstack-nova10:50
johnthetubaguydakhmetov: as long as we test for that, thats my main worry out the way10:50
mdboothdakhmetov: I suspect it already handles that, but we should definitely test as johnthetubaguy says.10:50
mdboothIf it doesn't work, that would be a regression.10:50
johnthetubaguydakhmetov: yeah, I think the get() is OK, but deep down, I would rather we didn't populate the names at all myself, hence me being tempted to do this as two patches, one that just gets user_id and project_id from the instance, sets the others to None, then a second patch that adds in your cache of the names.10:52
*** baoli has joined #openstack-nova10:53
*** iceyao has joined #openstack-nova10:53
mdboothjohnthetubaguy: How about resolving username and projectname from keystone at the time we rewrite the metadata?10:54
mdboothRather than storing it in system meta10:54
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova master: Read NIC features in libvirt  https://review.openstack.org/44441510:54
mdboothWould that be so bad?10:54
*** kaisers_ has quit IRC10:56
openstackgerritStephen Finucane proposed openstack/nova-specs master: Reserve NUMA nodes with PCI devices attached  https://review.openstack.org/36446810:56
*** alexpilotti has quit IRC10:56
dakhmetovmdbooth: if we open Keystone session to resolve ids to names in libvirtxmdriver with token from request context, it may not have permissions to resolve names in keystone10:57
dakhmetovwe may not have permissions to resolve IDs to names in keystone even if we always open admin session10:57
*** baoli has quit IRC10:57
dakhmetovit's how I see it10:57
*** trinaths1 has quit IRC10:57
mdboothdakhmetov: Explain10:57
mdboothdakhmetov: I'm not being funny, I just don't understand that interaction all that well.11:00
*** cdent has joined #openstack-nova11:01
dakhmetovmdbooth: keystone policies are adjustable in a wide range11:01
dakhmetovhttps://github.com/openstack/keystone/blob/master/etc/policy.json11:01
*** alexpilotti has joined #openstack-nova11:01
mdboothdakhmetov: Where did the lookups come from originally?11:02
mdboothAre you saying that the policy context is different between creation and subsequent lookup, or that somebody might have changed the policy?11:02
dakhmetovmdbooth: so, listing users or projects can even be prohibited11:03
mdboothdakhmetov: It would be a pretty odd policy which denied an authenticated user permission to lookup their own username though, right?11:04
mdbooth    "identity:get_project": "rule:admin_required or project_id:%(target.project.id)s",11:05
mdbooth    "identity:get_user": "rule:admin_or_owner",11:05
* mdbooth guesses those would be the relevant permissions11:05
johnthetubaguydakhmetov: I don't think we should call out to keystone in there11:05
mdboothjohnthetubaguy: I know that came up earlier in the review, but are you thinking of performance or robustness when you say that?11:06
johnthetubaguydakhmetov: I would rather we get the patch in that uses instance.project_id and instance.user_id first, thats an important fix (we could use the context)11:06
*** houming has joined #openstack-nova11:07
johnthetubaguymdbooth: a combination of all of those really, plus security around every compute node having those permissions, and setting that up correctly11:07
johnthetubaguymdbooth: matching uuids to current names feels like something the monitoring systems could do11:07
openstackgerritjichenjc proposed openstack/nova master: Trivial Fix  https://review.openstack.org/44231011:08
openstackgerritjichenjc proposed openstack/nova master: Add description for Image location in snapshot  https://review.openstack.org/44037911:09
*** nicolasbock has joined #openstack-nova11:09
*** sapcc-bot3 has quit IRC11:12
*** carthaca_1 has quit IRC11:12
*** databus23_1 has quit IRC11:12
*** carthaca_ has joined #openstack-nova11:12
*** sapcc-bot has joined #openstack-nova11:12
*** mkoderer_ has joined #openstack-nova11:12
*** dgonzalez_ has joined #openstack-nova11:12
*** databus23_ has joined #openstack-nova11:12
*** david_1 has joined #openstack-nova11:12
*** tpatzig_ has joined #openstack-nova11:12
*** mkoderer_ has quit IRC11:14
*** dgonzalez_ has quit IRC11:14
*** david_1 has quit IRC11:14
*** tpatzig_ has quit IRC11:14
*** sdague_ has joined #openstack-nova11:16
openstackgerritjichenjc proposed openstack/nova master: check hw:mem_page_size when change extra_spec  https://review.openstack.org/40861011:16
*** amotoki_ has joined #openstack-nova11:16
*** rodrigods has quit IRC11:18
bauzasjohnthetubaguy: oh, just saw https://review.openstack.org/#/c/442387/1/resolutions/20170307-compute-platform.rst11:18
*** smatzek has joined #openstack-nova11:18
bauzasjohnthetubaguy: I haven't seen any ML discussion around this11:18
*** rodrigods has joined #openstack-nova11:18
bauzasjohnthetubaguy: is this something you discussed in ATL ?11:18
johnthetubaguybauzas: I haven't started any, its something I have been wanting to suggest for a while11:19
*** amotoki has quit IRC11:19
* bauzas should try to attend TC meetings 11:19
bauzasjohnthetubaguy: oh okay11:19
*** andreas_s has quit IRC11:19
bauzasfine by me, just wanted to make sure people were happy with it11:19
*** gongysh has joined #openstack-nova11:19
johnthetubaguybauzas: I suspect people hate it, but I think its needed11:20
cdentjohnthetubaguy: it's gonna rub some people funny, but I agree it is important11:20
johnthetubaguyI think we keep avoiding difficult conversations for two long on this kind of thing11:21
bauzas++11:22
bauzasanyway, /me goes offline11:22
cdentjohnthetubaguy: hell yes11:22
*** rodrigods has quit IRC11:23
*** rodrigods has joined #openstack-nova11:23
johnthetubaguycdent: I like your idea, frame the problem, not the people, I am going to go do that now11:23
*** gongysh has quit IRC11:23
cdentthat's mostly for conflict avoidance. I think everyone is aware there is a problem...11:24
johnthetubaguyyeha11:24
*** mvk has quit IRC11:27
*** moshele has joined #openstack-nova11:27
*** tpatzig_ has joined #openstack-nova11:28
*** sapcc-bot has quit IRC11:28
*** databus23_ has quit IRC11:28
*** carthaca_ has quit IRC11:28
*** dgonzalez_ has joined #openstack-nova11:28
*** david_1 has joined #openstack-nova11:28
*** databus23_2 has joined #openstack-nova11:28
*** sapcc-bot4 has joined #openstack-nova11:28
*** carthaca_2 has joined #openstack-nova11:28
*** tpatzig_ has quit IRC11:30
*** dgonzalez_ has quit IRC11:30
*** david_1 has quit IRC11:30
*** andreas_s has joined #openstack-nova11:32
dakhmetovjohnthetubaguy: mdbooth: guys, thank you for the advice! I'll add unit test for missing keys in system metadata.11:32
*** sdague has quit IRC11:34
*** sdague_ is now known as sdague11:34
*** houming has quit IRC11:34
*** bvanhav has joined #openstack-nova11:34
*** gcb has quit IRC11:35
*** bvanhav has quit IRC11:35
*** bvanhav has joined #openstack-nova11:35
VAhlOpenstack [Newton]: When deploying new instance on Nova. I got the following error. Running privsep helper: ['sudo', 'nova-rootwrap', '/etc/nova/rootwrap.conf', 'privsep-helper', '--config-file', '/etc/nova/nova.conf', '--config-file', '/etc/nova/nova-compute.conf', '--privsep_context', 'os_brick.privileged.default', '--privsep_sock_path', '/tmp/tmpcWGEib/privsep.sock']11:35
VAhl etc/nova/rootwrap.conf is own by root:root. etc/nova/nova.conf + nova-compute.conf is own by nova:nov11:35
*** Guest84425 has joined #openstack-nova11:36
*** psachin has quit IRC11:36
*** nmathew has quit IRC11:37
*** dakhmetov has quit IRC11:37
*** nicolasbock has quit IRC11:37
*** hshiina has joined #openstack-nova11:39
*** dakhmetov has joined #openstack-nova11:41
*** trinaths has joined #openstack-nova11:41
*** nicolasbock has joined #openstack-nova11:42
*** markvoelker has joined #openstack-nova11:42
sfinucangmann: Comments left on https://review.openstack.org/#/c/431038/11:43
sfinucanalex_xu: If you're looking for easy config-opt patches to review, you might look this way --> https://review.openstack.org/#/c/431039/ :)11:44
* sfinucan has a sudden craving for San Antonio-style breakfast tacos and mourns the sorry state of Mexican cuisine in Ireland11:45
alex_xusfinucan: thanks :)11:45
*** markvoelker has quit IRC11:46
*** houming has joined #openstack-nova11:49
*** yingjun has joined #openstack-nova11:51
*** erlon has joined #openstack-nova11:51
*** kaisers_ has joined #openstack-nova11:52
*** avolkov has joined #openstack-nova11:53
*** baoli has joined #openstack-nova11:53
*** sudipto has quit IRC11:56
*** CristinaPauna has quit IRC11:56
*** psachin has joined #openstack-nova11:57
*** baoli has quit IRC11:58
*** mvk has joined #openstack-nova12:01
*** gongysh has joined #openstack-nova12:03
*** b3nt_pin has joined #openstack-nova12:04
*** b3nt_pin is now known as Guest7192412:04
*** rfolco has joined #openstack-nova12:06
*** aysyd has joined #openstack-nova12:06
*** dakhmetov has quit IRC12:06
*** trinaths has quit IRC12:06
*** yingjun has quit IRC12:11
*** markvoelker has joined #openstack-nova12:16
*** Guest71924 is now known as beagles12:17
*** abalutoiu_ has quit IRC12:18
*** figleaf is now known as edleafe12:21
artomsfinucan, thanks for the tagged attach reviews :) I'll get to websocket proxy security, I promise. In my defense, it's harder to get full context for the latter than the former, and I don't feel like just reviewing patch by patch and writing nits12:23
artomI had an idea about removing the tag=None param to the driver interface, let me respin that first12:23
*** kaisers_ has quit IRC12:24
artomBuilding device metadata is probably not a huge deal, optimization-wise, but the DB call hidden in instance.save definitely is. So I'm thinking build the new device metadata, compare it to the old one, and only save if there's a difference12:25
artomGet rid of the driver interface change, still no useless DB queries12:25
artomWait...12:26
*** DuncanT has joined #openstack-nova12:26
artomDoesn't save() call obj_what_changed() or whatever it's called?12:26
artomDoes it still call the database if nothing has changed?12:26
sfinucanartom: Yeah, no (immediate) panic :) the websocket proxy stuff /is/ tougher to review than other stuff, seeing as it's not just glue like other nova code12:28
sfinucanI'd be interested in seeing the approach to removing the tag parameter though, if it works out12:29
sfinucanSteadily working through the rest of the series now12:29
sfinucansahid: RE: https://review.openstack.org/#/c/400875/, this is the issue I have with it https://review.openstack.org/#/c/400875/23/nova/virt/hardware.py@84812:29
artomBased on ^^^ I think we can just remove with no other changes/repercussions?12:29
sfinucanFor N calls of _numa_fit_instance_cell_with_pinning, you will only ever have one where num_cpu_reserved is anything but 012:29
sfinucanThat, to me, is overloading12:30
artomdansmith can confirm when he wakes up, but if instance.save() is smart about how it calls the database, which I think it is, we're good12:30
sfinucanartom: looking12:30
*** edmondsw has joined #openstack-nova12:31
openstackgerritSergey Nikitin proposed openstack/nova master: [WIP] PIC NUMA policy PoC  https://review.openstack.org/39052012:32
*** kevinz has joined #openstack-nova12:34
artomhttps://github.com/openstack/nova/blob/master/nova/objects/instance.py#L721 https://github.com/openstack/nova/blob/master/nova/objects/instance.py#L66312:35
*** liusheng has quit IRC12:35
*** jheroux has joined #openstack-nova12:35
artom^^ that pretty conclusively indicates that if device_metadata has not changed, there will be no useless DB calls12:36
*** liusheng has joined #openstack-nova12:36
*** houming has quit IRC12:38
*** sree has joined #openstack-nova12:38
*** moshele has quit IRC12:42
sfinucanartom: That would be reading of it to12:42
sfinucanGood spot12:42
sfinucanSo12:42
*** sree has quit IRC12:43
sfinucanWe can just call it every single time with negligible performance impact12:43
*** esberglu has joined #openstack-nova12:44
*** esberglu has quit IRC12:44
*** esberglu has joined #openstack-nova12:45
artomsfinucan, pretty sure12:46
artomsfinucan, the little ones are waking up, let me drive them to daycare and I'll continue after12:46
sfinucanartom: (y)12:47
* sfinucan requires food anyway12:47
*** esberglu has quit IRC12:49
*** sree has joined #openstack-nova12:50
*** salv-orlando has quit IRC12:50
*** xyang1 has joined #openstack-nova12:53
*** erol has joined #openstack-nova12:53
*** arne_r has joined #openstack-nova12:54
erolhi everyone, I wonder Is there any way to flush Nova Schedular RetryFilter count? I hit max attempt. What can i do now?12:56
sean-k-mooneysc68cal: if you get a chance can you review https://review.openstack.org/#/c/444788 it will fix our third party ci after https://review.openstack.org/#/c/435575/ added a test requirement that break ovs compilation12:56
sean-k-mooneysc68cal: it should also give a speed up in the ci jobs as less stuff needs to be downloaded which is a nice side effect12:57
sean-k-mooneyerol: usually you delete the instance and try again but if you hit max attempts its often because there was a problem with the request12:58
sean-k-mooneyerol: either no host satifies the request or you have an infra issue that is stoping the cloud from working correctly12:58
*** abalutoiu_ has joined #openstack-nova13:00
*** yingjun has joined #openstack-nova13:01
*** zhurong has quit IRC13:01
*** gszasz has joined #openstack-nova13:02
*** bvanhav has quit IRC13:03
erolsean-k-mooney: Thanks. I try to run Ironic and have an issue on provisioning. After I attempted several times. I came a cross with these output in schedular log http://paste.openstack.org/show/602488/13:03
*** bvanhav has joined #openstack-nova13:03
*** tblakes has joined #openstack-nova13:04
*** beagles is now known as beagles_mtg13:04
*** nicolasbock has quit IRC13:05
*** Guest84425 has quit IRC13:06
*** nicolasbock has joined #openstack-nova13:09
*** liverpooler has joined #openstack-nova13:09
*** liverpooler has quit IRC13:09
*** liverpooler has joined #openstack-nova13:10
*** youzp has quit IRC13:11
*** youzp has joined #openstack-nova13:12
*** vladikr has joined #openstack-nova13:12
openstackgerritRafael Folco proposed openstack/nova master: DELETE all inventory for a resource provider  https://review.openstack.org/41666913:12
openstackgerritRafael Folco proposed openstack/nova master: Raising http codes on old microversion  https://review.openstack.org/44449713:12
*** sudipto has joined #openstack-nova13:13
*** eharney has quit IRC13:14
*** gcb has joined #openstack-nova13:14
*** salv-orlando has joined #openstack-nova13:15
*** youzp has quit IRC13:16
*** esberglu has joined #openstack-nova13:19
*** esberglu has quit IRC13:19
*** esberglu has joined #openstack-nova13:19
*** kevinz has quit IRC13:20
*** kevinz has joined #openstack-nova13:20
*** kaisers_ has joined #openstack-nova13:20
*** salv-orl_ has joined #openstack-nova13:24
*** salv-orlando has quit IRC13:26
*** VAhl has quit IRC13:27
*** smatzek has quit IRC13:29
*** lucasagomes is now known as lucas-hungry13:35
*** eharney has joined #openstack-nova13:40
*** felipemonteiro has joined #openstack-nova13:41
openstackgerritSergey Nikitin proposed openstack/nova master: [WIP] PIC NUMA policy PoC  https://review.openstack.org/39052013:41
*** hongbin has joined #openstack-nova13:45
*** mdnadeem has quit IRC13:46
*** eharney has quit IRC13:47
*** gouthamr has joined #openstack-nova13:47
*** eharney has joined #openstack-nova13:47
*** ayogi has quit IRC13:48
*** mriedem has joined #openstack-nova13:48
mriedemo/13:48
*** sudipto has quit IRC13:50
*** salv-orl_ has quit IRC13:51
*** jpena is now known as jpena|lunch13:51
dansmithalex_xu: you still around?13:51
*** baoli has joined #openstack-nova13:52
*** iceyao has quit IRC13:52
*** gcb has quit IRC13:52
*** iceyao has joined #openstack-nova13:52
bauzasoh, you US folks already had DST shift ?13:55
edleafeScheduler subteam meeting in 5 minutes in #openstack-meeting-alt13:55
leakypipesrfolco: were you working on adding error codes (in addition to error messages) to the placement API? can't remember if that was you or diga...13:55
mriedembauzas: that was this weekend13:55
edleafebauzas: yep - yeaterday13:55
mriedemleakypipes: neither https://review.openstack.org/#/c/418393/13:56
leakypipesmriedem: ah!13:56
leakypipesit was _gryf13:56
*** irenab has quit IRC13:56
bauzasmriedem: edleafe: cool then, France is now closer to the US :)13:57
rfolcoleakypipes, morning. Not me.13:58
leakypipesrfolco: np :) thanks!13:59
*** irenab_ has joined #openstack-nova13:59
*** zz_dimtruck is now known as dimtruck13:59
*** lucasxu has joined #openstack-nova14:00
edleafeleakypipes: time to plug the leak14:00
*** leakypipes is now known as jaypipes14:00
jaypipes:)14:00
*** arne_r has quit IRC14:00
*** baoli has quit IRC14:01
*** dnalezyty has joined #openstack-nova14:02
*** baoli has joined #openstack-nova14:02
*** mlavalle has joined #openstack-nova14:02
*** yingjun has quit IRC14:02
*** ltomasboz is now known as ltomasbo14:02
*** amoralej is now known as amoralej|lunch14:04
_gryfyup, I did that bp14:04
*** amotoki_ has quit IRC14:05
markus_zmriedem: JFYI, the VNC issue we talked about 1w ago will be fixed with: https://review.openstack.org/#/c/444164/14:05
*** awaugama has joined #openstack-nova14:05
*** smatzek has joined #openstack-nova14:05
*** gcb has joined #openstack-nova14:07
markus_zmriedem: I also have to pester you about: https://review.openstack.org/#/c/406898/ All info you asked for is added to that review.14:07
*** baoli has quit IRC14:08
mriedemok14:08
openstackgerritdane-fichter proposed openstack/nova-specs master: Trusted flavors  https://review.openstack.org/31222514:09
sfinucanjaypipes: Might you be so kind as to stick this on your review backlog for this week? https://review.openstack.org/#/c/361140/14:10
*** dane-fichter has joined #openstack-nova14:10
sfinucanjohnthetubaguy too ^, if you could14:10
*** sudipto has joined #openstack-nova14:10
*** sree has quit IRC14:11
*** Jeffrey4l has joined #openstack-nova14:12
* johnthetubaguy nods at sfinucan14:12
* sfinucan nods back at johnthetubaguy 14:12
dane-fichtermriedem , sdague : Could you guys take a look at my change adding the security tempest tests as an experimental check in Nova? https://review.openstack.org/#/c/443167/14:12
mriedemdane-fichter: sure14:13
dane-fichtermriedem: thanks!14:14
*** links has quit IRC14:16
*** Jeffrey4l has quit IRC14:17
*** Jeffrey4l has joined #openstack-nova14:18
jaypipessfinucan: ya14:18
sfinucanjaypipes: Thank you, sir14:19
jaypipessfinucan: np. you know I love my numa. :P14:19
sfinucanjaypipes: I saw numa'y stuff and decided to r̶u̶i̶n̶ make your day14:20
*** baoli has joined #openstack-nova14:22
*** hshiina has quit IRC14:22
*** baoli has quit IRC14:22
mriedemmarkus_z: the multinode vnc fix looks good to me, that's even called out in the devstack multinode docs14:22
jaypipessfinucan: :)14:24
*** burt has joined #openstack-nova14:24
openstackgerritStephen Finucane proposed openstack/nova master: trivial: Add a note about 'cells_api'  https://review.openstack.org/44492614:25
*** beagles_mtg is now known as beagles14:25
dansmithjohnthetubaguy: alex_xu: I'm totally lost.. where does cachingscheduler override things in HostManager?14:26
johnthetubaguydansmith: it uses get_all_hosts() basically14:26
*** vsaienk0 has joined #openstack-nova14:27
johnthetubaguyrather than get_by_uuid14:27
johnthetubaguys14:27
*** markus_z has quit IRC14:27
johnthetubaguydansmith: https://github.com/openstack/nova/blob/master/nova/scheduler/caching_scheduler.py#L7514:27
johnthetubaguydansmith: might be time to deprecate that thing soon, although I was hoping to wait until we do allocations in the conductor really14:28
dansmithjohnthetubaguy: okay the main scheduler uses that purely to prime the data structures, but not for an actual schedule.. wouldn't the caching scheduler only cache hosts that are valid for a particular spec_obj if we made that change?14:28
johnthetubaguydansmith: no, it caches the full list of hosts14:28
johnthetubaguyand bypasses the placement PAI14:28
johnthetubaguyAPI14:28
johnthetubaguydansmith: did we have moving allocations into the conductor pencilled in for this cycle yet?14:29
dansmithjohnthetubaguy: right, but.. I don't understand what you think needs to change.. the _all_cells method takes a spec_obj only if you want to cause it to limit the host list based on a requested cell,14:29
dansmithwhich you'd never want to do with that right?14:29
johnthetubaguydansmith: oh, sorry, I see your point now... it just doesn't work for the caching scheduler14:30
*** udesale has joined #openstack-nova14:30
dansmithright14:30
*** kevinz has quit IRC14:31
dansmithjohnthetubaguy: either we'd have to honor the cell request somewhere else, or just not have this happen if you're using cachingscheduler14:31
dansmithI dunno who all uses it, tbh, so I guess I'm not sure the implications of that14:31
*** kevinz has joined #openstack-nova14:32
johnthetubaguydansmith: large deployments currently suck without using that, like >100 nodes or something like that, so probably only the large ones14:32
dansmithjohnthetubaguy: what about large deployments now that we have placement?14:32
*** baoli has joined #openstack-nova14:32
dansmithjohnthetubaguy: we should be getting the host list waaayyy faster now as of this cycle14:32
johnthetubaguydansmith: they are much worse, if I understand it correctly14:32
dansmithWAT14:33
johnthetubaguyyep14:33
*** satyar has joined #openstack-nova14:33
dansmithwhy?14:33
vsaienk0mriedem: could you please approve https://blueprints.launchpad.net/nova/+spec/ironic-hotplug-interfaces for Pike, the code is ready on both nova and ironic side, and it is very small change which already covered by CI tests. It would be great to have this implemented in Pike14:33
johnthetubaguydansmith: we don't do the allocations soon enough14:33
dansmithjaypipes: ^14:33
johnthetubaguydansmith: so you quickly hit the three retries, because everyone picks the same host14:33
dansmithjohnthetubaguy: placement makes that _worse_ ?14:33
dansmithI thought cachingscheduler was for get_all_hosts() performance...14:34
johnthetubaguydansmith: worse than using caching scheduler, I mean14:34
* jaypipes reads back14:34
johnthetubaguydansmith: its performance, and caching of results between API calls14:34
*** kaisers_ has quit IRC14:34
dansmithjohnthetubaguy: does cachingscheduler affect allocation timing though? I thought it was just about finding the list of hosts quickly, which placement *should*be improving14:34
johnthetubaguyjaypipes: in my head I had us killing the caching scheduler once allocations are written in the conductor14:35
johnthetubaguydansmith: no, its also about that cache being updated with the previous results, thats why it only works if you have one nova-scheduler, and fails badly if you have two caching schedulers14:35
*** dave-mccowan has joined #openstack-nova14:35
dansmithjohnthetubaguy: ugh, okay14:35
johnthetubaguydansmith: yeah, its really non-obvious from the code thats how it works14:36
*** claudiub has quit IRC14:36
johnthetubaguydansmith: that logic that makes when we scheduler a request of 5 builds at once not all pick the same host activates for every request using the same cached list14:37
mriedemvsaienk0: done14:37
*** marst has quit IRC14:37
*** lucas-hungry is now known as lucasagomes14:37
vsaienk0mriedem: thanks!14:37
jaypipesjohnthetubaguy: I'd appreciate a hangout discussion with you about the caching scheduler and the placement service.14:38
jaypipesjohnthetubaguy: do you have some time to discuss it now?14:38
dansmithjohnthetubaguy: I guess I'm surprised that's a real problem with randomizing the result list and sufficient capacity14:38
dansmithjaypipes: I want in on that14:38
dansmithbecause cells.14:38
mriedemcan't you configure the scheduler to spread rather than pack? or does just no one do that in production?14:38
jaypipesdansmith: I figured as much :)14:38
johnthetubaguyjaypipes: dansmith: I have 20 mins now if you are both free?14:38
dansmithI am14:38
jaypipesme too.14:38
mriedemi'm jumping in too14:38
jaypipessec, getting link.14:38
johnthetubaguymriedem: right, no one wants spread, because you run our of large instances14:38
jaypipeshttps://hangouts.google.com/call/s6p6dqm3dfbyxjwu5ppyhj4y2ae14:39
jaypipesmriedem: ^14:39
mriedemyeah, damn firefox doesn't support hangouts anymore on the latest version14:40
* mriedem switches14:40
dansmithjaypipes: https://review.openstack.org/#/c/438025/21/nova/scheduler/host_manager.py14:41
* alex_xu jump in the hangout to train English listening14:43
*** felipemonteiro has quit IRC14:43
sc68calsean-k-mooney: you'll need to reach out to sdague about that issue, I'd like his input14:46
*** tonygunk has quit IRC14:46
*** amoralej|lunch is now known as amoralej14:46
jaypipesjohnthetubaguy: https://github.com/openstack/nova/blob/master/nova/scheduler/client/__init__.py#L7714:47
*** marst has joined #openstack-nova14:48
mriedempkoniszewski: i'm not totally sure about this backport since no one from the live migration subteam reviewed the change on master, can you take a look at this and see if the change itself (even though already merged on master), is OK? https://review.openstack.org/#/c/441037/14:49
*** marst has quit IRC14:53
*** marst has joined #openstack-nova14:53
*** mkoderer has joined #openstack-nova14:53
*** jpena|lunch is now known as jpena14:54
openstackgerritBéla Vancsics proposed openstack/nova master: Transform instance.trigger_crash_dump notification  https://review.openstack.org/39622514:55
*** baoli has quit IRC14:55
*** baoli_ has joined #openstack-nova14:55
*** tonygunk has joined #openstack-nova14:56
*** dimtruck is now known as zz_dimtruck14:57
*** kevinz has quit IRC14:57
*** esberglu has quit IRC14:57
*** kevinz has joined #openstack-nova14:57
*** esberglu has joined #openstack-nova14:57
*** yamahata has joined #openstack-nova14:58
*** gongysh has quit IRC14:59
mriedempkoniszewski: nevermind, it's only for resize, and tests there are all OK, so lgtm.15:00
mriedemlyarwood: can you take a look at these ocata backports for a release this week? https://review.openstack.org/#/c/441171/ https://review.openstack.org/#/c/444374/15:01
*** esberglu has quit IRC15:02
mriedemdansmith: note i haven't posted the pike priorities list to the specs repo yet, but we had a list from the ptg, so this will change that15:03
*** raj_singh_ has joined #openstack-nova15:03
dansmithmriedem: it will a little, but I think it's clear that claims will be a queens thing15:03
*** liangy has joined #openstack-nova15:04
mriedemdansmith: ok that works too - and good to know since at the summit i/we have to present on what's in the works for queens, as annoying as that is to be reporting at this point15:04
dansmithyeah15:04
sean-k-mooneysc68cal: sure will do. thanks for taking a look though15:06
artomjohnthetubaguy, heya - sfinucan mentioned that the the SPICE native proxy spec (https://review.openstack.org/#/c/442040/) might be relevant to XVP in Xen - I'm mostly ignorant of what XVP is, but from a quick read I don't *think* there's any potential crossover between a SPICE native proxy, other than both are proxies, but maybe you could take a look?15:06
*** adrian_otto has joined #openstack-nova15:07
sean-k-mooneyerol: i dont think you are ment to use the RamFilter with ironic. at least in the past it had its own filter that replaced it15:07
*** irenab_ has quit IRC15:08
*** sridharg has quit IRC15:08
*** esberglu has joined #openstack-nova15:09
*** burgerk has joined #openstack-nova15:09
*** moshele has joined #openstack-nova15:10
*** felipemonteiro has joined #openstack-nova15:10
*** kevinz has quit IRC15:10
*** udesale has quit IRC15:13
*** jianghuaw has quit IRC15:14
*** zz_dimtruck is now known as dimtruck15:15
*** dane-fichter has quit IRC15:17
*** yingjun has joined #openstack-nova15:17
BobBallartom: Does this have to be spice-specific?  XenAPI uses vnc consoles and - apart from the references to spice throughout the spec and in the API call - the proxying of the connection in this way should probably work for all console types, not just spice?15:19
artomBobBall, not at all :)15:19
*** armax has joined #openstack-nova15:20
artomThe proxy code itself is completely generic, it's just iptables shovelling packets back and forth15:20
BobBallIndeed - the only thing that concerns me ATM is using os-getSPICEConsole spice-native which is clearly spice specific15:20
*** mriedem has quit IRC15:20
artomWe started with SPICE because we need a user-facing feature to actually drive it, but we'll be happy to make adjustments of other drivers need it for them as well15:20
artomBobBall, well, that's the API part for the user to get a URL for his native SPICE client15:21
*** gfhellma has quit IRC15:21
erolsean-k-mooney: I am using mirantis openstack with default settings. :(15:21
artomBobBall, if xenapi wants a native proxy for their console an API can be added for that as well15:22
BobBallartom: Understood.  I'll comment on the spec.15:23
*** beekneemech is now known as bnemec15:23
artomAnd it can tell nova-nativeproxy which host/port it should start a proxy instance for15:23
artomThe host/port would come from the xenapi hypervisor, presumably15:23
artomBut that's your domain ;)15:23
BobBallIndeed.  That's how our consoles work with novnc currently.15:23
*** mriedem has joined #openstack-nova15:25
*** fragatina has joined #openstack-nova15:25
*** hieulq_ has joined #openstack-nova15:26
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova master: Modify PciPassthroughFilter to accept lists  https://review.openstack.org/44494615:27
*** baoli_ has quit IRC15:27
openstackgerritJuan Antonio Osorio Robles proposed openstack/nova master: Make versioned notifications topics configurable  https://review.openstack.org/44494715:28
*** amotoki has joined #openstack-nova15:28
*** bvanhav has quit IRC15:28
*** satyar has quit IRC15:28
*** bvanhav has joined #openstack-nova15:29
sfinucanartom: Think I've all of https://review.openstack.org/#/q/topic:bp/virt-device-tagged-attach-detach reviewed now15:31
sfinucanThanks dansmith for the cells v1 clarification. I recorded it in the code itself here https://review.openstack.org/44492615:31
*** armax has quit IRC15:31
*** baoli has joined #openstack-nova15:31
artomsfinucan, much thanks :) You were responsible for a cute moment this morning. I saw your "oh boy, here we go" email, laughed, daughter asked why I was laughing and wanted to see, I showed her your email, she did a forced laugh as well15:32
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova master: Read NIC features in libvirt  https://review.openstack.org/44441515:32
*** yingjun has quit IRC15:33
BobBallartom: Just for my understanding, does requesting the spice console generate a unique password per request, or is a fixed password generated and associated with each VM?15:34
sfinucanartom: Ha! cute :D15:34
artomBobBall, the password is set by the libvirt driver on the SPICE console, and returned to the user15:35
artomBobBall, xenapi can do whatever it wants in that regard15:35
BobBallartom: So libvirt assigns a non-predictable password to each console?15:35
artomBobBall, that's the current plan, yeah15:36
artomLibvirt would set a new password on every request15:36
BobBallok thanks15:37
*** bvanhav_ has joined #openstack-nova15:37
*** burgerk_ has joined #openstack-nova15:37
artomSince this is essentially opening up the host's console socket to the entire world, we need to make it a minimum secure15:37
artomSo TLS if possible, and authentication15:37
*** smatzek_ has joined #openstack-nova15:37
BobBallI totally agree.15:37
*** bvanhav__ has joined #openstack-nova15:37
openstackgerritMaciej Szankin proposed openstack/nova-specs master: Add use-local-scheduler spec  https://review.openstack.org/43893615:39
*** bvanhav has quit IRC15:40
*** burgerk has quit IRC15:41
*** smatzek has quit IRC15:41
*** burgerk_ has quit IRC15:41
*** bvanhav_ has quit IRC15:41
*** karthiks is now known as karthiks_afk15:43
*** iceyao has quit IRC15:46
*** snecklifter has quit IRC15:47
mriedemdansmith: replied in https://review.openstack.org/#/c/443403/ - let me know if i need to change things. starting orientation for the rest of the day in about 10 minutes15:48
bauzasdansmith: jroll: jaypipes: do we have some spec that explains our consensus in L66 https://etherpad.openstack.org/p/nova-ptg-pike-placement15:50
bauzas?15:50
bauzastl;dr: ironic flavors need to be updated by adding extra specs15:50
bauzass/explains/reflects15:51
jrollbauzas: that's on my list for this week15:51
jroll(writing that spec)15:51
bauzasjroll: okay, ping me when it's done, because I'm trying to push it downstream15:51
jrollbauzas: will do, thanks15:52
*** psachin has quit IRC15:52
bauzasjroll: FWIW, operators would need to add a custom resource class in their extra spec per type of flavor they have, correct?15:52
openstackgerritCedric Brandily proposed openstack/nova master: Remove VMware driver _get_vm_ref_from_uuid method  https://review.openstack.org/44495915:53
dansmithbauzas: yes15:54
jrollbauzas: yes15:54
* jroll throws a coke at dansmith15:54
openstackgerritCedric Brandily proposed openstack/nova master: VMware: use WithRetrieval in get_network_with_the_name  https://review.openstack.org/32558715:57
openstackgerritCedric Brandily proposed openstack/nova master: VMware: use WithRetrieval in vm_util module  https://review.openstack.org/32560815:59
openstackgerritCedric Brandily proposed openstack/nova master: VMware: use WithRetrieval in ds_util module  https://review.openstack.org/32561015:59
erolsean-k-mooney: Hey, When I added another ironic node, I didt get error from Schedular. I think it is all about RetryFilter.15:59
*** crushil has joined #openstack-nova15:59
*** liusheng has quit IRC16:00
*** liusheng has joined #openstack-nova16:00
sean-k-mooneyerol: interesting. though the pastbin was showing that the ram filter was the filter that eliminated all of the host16:01
sean-k-mooneyerol: Filter results: ['RetryFilter: (start: 4, end: 3)', 'AvailabilityZoneFilter: (start: 3, end: 3)', 'RamFilter: (start: 3, end: 0)']16:01
sean-k-mooneyerol: the retry filter eliminated 1 host which was the host that was tried first16:02
*** marst_ has joined #openstack-nova16:03
erolsean-k-mooney: all of 3 node is kvm node, there is no enough resource that instance need on them. Sorry for the indicate it.16:03
sean-k-mooneyerol: so you have a mixed deployment ironic/kvm?16:04
*** nic has joined #openstack-nova16:04
erolI think Retryfilter eliminate Ironic node, and then When it check ram resource it cant find correct one in kvm nodes.16:04
sean-k-mooneyand the ram filter failed on the kvm nodes when trying to select a baremetal flavor16:04
erolsean-k-mooney: yeah. 3 Nova Compute 2 ironic compute for now.16:05
sean-k-mooneyerol: the retry filter only filters nodes if it tred to scheduler to it and then failed16:05
sean-k-mooneyerol: it previously selected u'ironic-compute', u'f01c9155-7d1e-4351-8225-670e0d48452a' that failed to boot and it retryed the rest - that node16:06
*** claudiub has joined #openstack-nova16:06
*** marst has quit IRC16:06
*** moshele has quit IRC16:06
*** belmoreira has quit IRC16:06
erolsean-k-mooney: I am doing lots of mistake about metadata and resources. Then ironic instances faild on provisioning. Maybe i need the delete ironic-compute history when i reach the max attempt.16:07
*** dtp has joined #openstack-nova16:07
*** raghav has joined #openstack-nova16:08
erolsean-k-mooney: failed*16:08
raghavhi16:08
raghavi am getting error in watcher code..which root causes in nova code16:08
sean-k-mooneyerol: perhaps. i have used ironic but only in a limited fashion. one thing that you should be aware of is when deleteing ironic nodes it does not alway delete the neutron port which can cause issues16:09
raghav+----------------+---------------------+-----------------+----------------+-------+ | ID             | Hypervisor Hostname | Hypervisor Type | Host IP        | State | +----------------+---------------------+-----------------+----------------+-------+ | region!child@1 | praveen-dev         | QEMU            | 10.223.197.217 | up    | +----------------+---------------------+-----------------+----------------+-------+16:09
sean-k-mooney*used ironic much, only in a limited fashion16:09
raghavhere id fiels is non-integer16:10
sean-k-mooneyraghav: the id field is a string16:10
raghavmy nova-api version is 15.0.116:10
sean-k-mooneyraghav: infact its usally a uuid16:10
raghavbut in doc its integer16:10
raghavhttps://github.com/openstack/nova/blob/stable/mitaka/nova/objects/compute_node.py#L5916:11
erolsean-k-mooney: I wish there is a cli command to flush failed attempt for RetryFilter16:12
raghavi am getting parsing error in watcher because of this16:12
sean-k-mooneyraghav: so the id field in the db is an int16:13
sean-k-mooneyhowever i belive the ID filed in the api maps to the UUID field in the db as the id filed in the db is not exposed16:14
sean-k-mooneyraghav: i could be wrong on that but jaypipes or mriedem shoudl be able to confirm16:14
mriedemwhich table?16:15
mriedemwe expose the compute_nodes.id field out of the API16:15
mriedemb/c we're dumb16:15
*** david-lyle has joined #openstack-nova16:15
mriedemhttps://github.com/openstack/nova/blob/master/nova/api/openstack/compute/hypervisors.py#L5216:16
mriedemsean-k-mooney: i plan on fixing that16:16
mriedemhttps://blueprints.launchpad.net/nova/+spec/service-hyper-pci-uuid-in-api16:16
sean-k-mooneyraghav: actully i might be getting confused with the flavor sorry ignore what i previously said.16:17
openstackgerritMikhail Feoktistov proposed openstack/nova master: Cleanup code in create_ploop_image  https://review.openstack.org/44497016:17
sean-k-mooneymriedem: cool16:17
mriedemsean-k-mooney: we expose the flavor.id too out of the API16:17
*** Apoorva has joined #openstack-nova16:18
sean-k-mooneymriedem: yes but for the flavor id that is also user set able as any arbitry string, and i know several people who have reversed the name and id filed on the commandline so that is even more of an issue16:18
*** kaisers_ has joined #openstack-nova16:19
openstackgerritMikhail Feoktistov proposed openstack/nova master: Cleanup code in create_ploop_image  https://review.openstack.org/44497016:19
mriedemsean-k-mooney: well, there is flavor.id and flavor.flavorid16:19
mriedemflavor.flavorid is user-settable16:19
mriedemflavor.id is the primary key16:19
mriedemi guess we don't expose flavor.id here https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/views/flavors.py#L3416:20
mriedemit might be the server get16:20
*** moshele has joined #openstack-nova16:20
*** raunak has joined #openstack-nova16:20
mriedemhttps://github.com/openstack/nova/blob/master/nova/api/openstack/compute/views/servers.py#L259 i guess that's flavorid too16:20
*** nicolasbock has quit IRC16:21
sdaguemriedem: we don't expose the .id anywhere iirc16:21
mriedemsdague: ok i thought we did and that was part of what prompted your original spec to nest the flavor details in the server response16:21
mriedemit also came up when moving flavors from the cell to the api db16:21
mriedemi might be misremembering details though, it was a year ago16:21
openstackgerritArtom Lifshitz proposed openstack/nova-specs master: SPICE native client support  https://review.openstack.org/44204016:23
sdaguemriedem: there were interesting challenges with flavor.id in cells v1, because it is an auto incrementing id16:23
sdagueand, thus could be different in the cells16:23
sdagueit is used as the join column16:24
dansmithsame for cellsv2 of course,16:24
dansmithbut I think we've gotten rid of all those things by now16:24
*** READ10 has joined #openstack-nova16:25
*** nicolasbock has joined #openstack-nova16:25
raghavmriedem: so is it bug in nova code?16:28
mriedemraghav: is what a bug in nova code? sorry i'm on a call so i'm not sure what the earlier discussion is about.16:28
mriedemraghav: are you using cellsv1? i see region!child@116:29
mriedemthat looks like cellsv116:29
*** kaisers_ has quit IRC16:29
*** kaisers_ has joined #openstack-nova16:29
*** moshele has quit IRC16:30
raghavmriedem: i have devstack setup16:30
mriedemraghav: note that you don't want to be using cellsv116:30
raghavmriedem: with ocata release16:30
mriedemraghav: what's your ENABLED_SERVICES variable?16:30
mriedemyou shouldn't have n-cell-region or n-cell-child16:30
raghavmriedem: n-api,n-cpu,n-cond,n-sch,n-novnc,n-cauth16:31
mriedemsorry just n-cell16:31
*** gfhellma has joined #openstack-nova16:32
*** salv-orlando has joined #openstack-nova16:35
*** hieulq_ has quit IRC16:35
mriedemraghav: in general, we don't want int id's exposed out of the API, which is why we have https://blueprints.launchpad.net/nova/+spec/service-hyper-pci-uuid-in-api16:36
mriedemraghav: but that's going to be done in a microversion16:36
sdagueraghav: honestly, I think there is a lot of confusion about how you got to that kind of data, and the only way any of us know that to be possible is if you have a cells v1 setup16:37
sdagueraghav: so, perhaps a reproduce scenario would be useful16:38
mriedemraghav: for now os-hypervisors will return an int for the id https://developer.openstack.org/api-ref/compute/?expanded=show-hypervisor-statistics-detail,show-hypervisor-details-detail16:38
*** pcaruana has quit IRC16:38
mriedemas shown in the docs16:38
mriedemmy blueprint is to change that to return the uuid rather than the int id16:38
melwittsdague: in cells v1, we don't officially support flavors in cell dbs. users like CERN have their own automation that syncs flavors from the api cell db to the child cell dbs. fwiw16:38
mriedemraghav: as for the hypervisor hostname being "region!child@1" that looks like cellsv116:39
mriedemso it seems you have n-cells running somewhere in your devstack setup16:39
sdaguemelwitt: right, it was definitely a goofy thing16:39
raghavmriedem: yes its running16:40
raghavmriedem: i could see in screen16:40
mriedemraghav: yeah, so you don't want that16:40
mriedemraghav: that's going to cause you all kinds of confusion16:41
mriedemso i'd start over w/o n-cell enabled16:41
raghavmriedem: so what should i do now16:41
*** zioproto has quit IRC16:41
mriedemraghav: restack w/o n-cell for one16:41
raghavmriedem: in local.conf i have not enabled n-cell16:41
mriedemraghav: well, it's not enabled by default, so something is running it in your setup16:42
*** hshiina has joined #openstack-nova16:42
*** erol has quit IRC16:42
mriedemraghav: also, be aware that cells v2, which is required in ocata, is not the same as the nova-cells service,16:42
mriedemor n-cell service in devstack16:42
mriedemi know it's confusing16:42
*** yamahata has quit IRC16:43
raghavmriedem: i have n-cells-region and n-cells-child running16:43
raghavmriedem: if i re-stack again those will get enabled16:45
*** mvk has quit IRC16:45
sdagueraghav: why do you have those running?16:45
raghavmriedem: i dont know how they are running, i just stacked without enabling them16:46
melwittraghav: what's in your stack-screenrc? are there any n-cell* in there?16:46
raghavmelwitt: yes n-cell-region and n-cell-chile16:47
raghavmelwitt: yes n-cell-region and n-cell-child16:47
sdagueraghav: so... step one, please get back to a base clean environment16:47
melwittraghav: okay. so it looks like you definitely have remnants of an earlier install where n-cell was part of ENABLED_SERVICES16:48
mriedemdansmith: did you still want a specific unit test for https://review.openstack.org/#/c/443403/11/nova/context.py ?16:48
openstackgerritMaciej Szankin proposed openstack/nova-specs master: Add use-local-scheduler spec  https://review.openstack.org/43893616:49
dansmithmriedem: yes, I'm not sure why we wouldn't16:49
raghavmelwitt: how can i remove it16:49
dansmithmriedem: are you doing both of those in a follow-up?16:49
mriedemdansmith: i can do them in this change16:50
mriedemraghav: maybe posting your local.conf in a gist would help us16:51
mriedemraghav: personally i'd re-clone devstack from stable/ocata to make sure it's completely clean16:51
mriedemin a new vm16:51
openstackgerritJohn Garbutt proposed openstack/nova-specs master: Spec to add per instance timeout for LM  https://review.openstack.org/44280916:53
melwittyeah, I'm not 100% an unstack.sh/stack.sh after killing processes would result in a fresh read of ENABLED_SERVICES16:53
mriedemi would completely start over16:53
sdaguethat should, but a ./clean.sh is even safer16:53
sdagueyeh, completely fresh not a bad idea either16:53
openstackgerritJohn Garbutt proposed openstack/nova-specs master: WIP: Replace live migration steps and delays with memory iteration logic  https://review.openstack.org/44333916:53
sdaguethere are clearly bits left over from some past config16:54
raghavok16:56
raghavthanks16:56
*** lucasagomes has quit IRC16:59
*** armax has joined #openstack-nova16:59
*** lucasagomes has joined #openstack-nova17:00
*** efoley_ has joined #openstack-nova17:01
*** liangy has quit IRC17:03
*** lucasxu has quit IRC17:03
*** efoley__ has quit IRC17:04
*** kaisers_ has quit IRC17:05
*** andreas_s has quit IRC17:05
*** nkrinner is now known as nkrinner_afk17:05
*** gszasz has quit IRC17:06
openstackgerritJuan Antonio Osorio Robles proposed openstack/nova master: Make versioned notifications topics configurable  https://review.openstack.org/44494717:15
*** openstackgerrit has quit IRC17:18
*** sudipto has quit IRC17:19
*** yamahata has joined #openstack-nova17:21
*** gouthamr has quit IRC17:21
*** unicell has joined #openstack-nova17:25
*** unicell has quit IRC17:25
*** unicell has joined #openstack-nova17:25
*** catintheroof has joined #openstack-nova17:25
*** bvanhav_ has joined #openstack-nova17:28
*** ltomasbo is now known as ltomasbo|away17:28
*** bvanhav_ is now known as bvanhav17:29
*** bvanhav__ has quit IRC17:29
*** openstackgerrit has joined #openstack-nova17:30
openstackgerritmelanie witt proposed openstack/nova master: Count resources to check quota for cells  https://review.openstack.org/41652117:30
openstackgerritmelanie witt proposed openstack/nova master: Make Quotas object favor the API database  https://review.openstack.org/41094517:30
openstackgerritmelanie witt proposed openstack/nova master: Add online migration to move quotas to API database  https://review.openstack.org/41094617:30
*** lucasxu has joined #openstack-nova17:30
mriedemdansmith: i'm really confused about this test https://github.com/openstack/nova/blob/master/nova/tests/unit/test_context.py#L307-L31017:35
*** kaisers_ has joined #openstack-nova17:35
mriedemshouldn't the assertion be such that the context db/mq connections within the target_cell context match what's in the mapping?17:35
*** browne has joined #openstack-nova17:35
dansmithmriedem: it does17:36
dansmithmriedem: maybe it should assert that get_context is called with the database url, and same for rpc, but otherwise it's checking that they get switched and restored17:36
*** gouthamr has joined #openstack-nova17:38
*** efoley__ has joined #openstack-nova17:38
mriedemoh nvm17:39
mriedemi was expecting to see it have the fake urls17:39
*** efoley_ has quit IRC17:41
*** salv-orlando has quit IRC17:43
*** salv-orlando has joined #openstack-nova17:43
*** fragatina has quit IRC17:44
*** fragatina has joined #openstack-nova17:44
-openstackstatus- NOTICE: restarting gerrit to address performance problems17:45
openstackgerritMatt Riedemann proposed openstack/nova master: Decrement quota usage when deleting an instance in cell0  https://review.openstack.org/44339517:46
openstackgerritMatt Riedemann proposed openstack/nova master: Temporarily untarget context when deleting from cell0  https://review.openstack.org/44340317:46
*** abalutoiu_ has quit IRC17:48
*** fragatina has quit IRC17:49
*** derekh has quit IRC17:49
*** mdrabe has joined #openstack-nova17:51
*** vladikr has quit IRC17:53
*** jamielennox is now known as jamielennox|away17:53
*** nicolasbock has quit IRC17:54
*** jaosorior has quit IRC17:56
*** adrian_otto has quit IRC17:57
*** nicolasbock has joined #openstack-nova17:57
*** avolkov has quit IRC17:58
*** tesseract has quit IRC17:59
*** ralonsoh has quit IRC18:00
openstackgerritDan Smith proposed openstack/nova master: Get instance availability_zone without hitting the api db  https://review.openstack.org/43975418:01
openstackgerritDan Smith proposed openstack/nova master: Teach HostAPI about cells  https://review.openstack.org/44216218:01
openstackgerritDan Smith proposed openstack/nova master: Make os-availability-zones know about cells  https://review.openstack.org/44216318:01
openstackgerritDan Smith proposed openstack/nova master: Make scheduler target cells to get compute node instance info  https://review.openstack.org/43989118:01
openstackgerritDan Smith proposed openstack/nova master: Sort CellMappingList.get_all() for safety  https://review.openstack.org/44317418:01
openstackgerritDan Smith proposed openstack/nova master: Add get_by_instance_uuids() to InstanceMappingList  https://review.openstack.org/44329218:01
openstackgerritDan Smith proposed openstack/nova master: Clean up ClientRouter debt  https://review.openstack.org/44448718:01
openstackgerritDan Smith proposed openstack/nova master: Make server_groups determine deleted-ness from InstanceMappingList  https://review.openstack.org/44329318:01
openstackgerritDan Smith proposed openstack/nova master: Remove Mitaka-era service version check  https://review.openstack.org/44286118:01
openstackgerritDan Smith proposed openstack/nova master: Make conductor ask scheduler to limit migrates to same cell  https://review.openstack.org/43802518:01
openstackgerritDan Smith proposed openstack/nova master: Add workaround to disable group policy check upcall  https://review.openstack.org/44273618:01
openstackgerritDan Smith proposed openstack/nova master: Add SameCellFilter  https://review.openstack.org/44503518:01
*** aarefiev is now known as aarefiev_afk18:01
*** adrian_otto has joined #openstack-nova18:01
*** lpetrut has quit IRC18:03
*** thorst_afk has joined #openstack-nova18:04
*** thorst_afk has quit IRC18:04
dansmithjohnthetubaguy: have you gone away for the day yet?18:05
*** jpena is now known as jpena|off18:05
melwitthah, my counting quota patch failed mriedem's recent quota regression functional test (the expected fail part) because I forgot it also fixes the problem18:05
johnthetubaguydansmith: about to go18:06
johnthetubaguydansmith: whats up?18:06
dansmithjohnthetubaguy: okay well, just heads up on https://review.openstack.org/#/c/445035/18:06
johnthetubaguydansmith: ah, cool, I will try go through that chain again tomorrow18:08
*** tblakes has quit IRC18:08
johnthetubaguydansmith: I forgot we don't get the cell from the host object already, which kinda sucks, but that cache seems like a good stop gap18:08
*** raunak has quit IRC18:09
dansmithjohnthetubaguy: it should fill up pretty fast, not require too much memory, and only change when we add a host, so not too bad I think18:09
*** raunak has joined #openstack-nova18:10
*** hshiina has quit IRC18:10
*** siva_krishnan has joined #openstack-nova18:14
*** efried has quit IRC18:15
*** tblakes has joined #openstack-nova18:16
siva_krishnanjohnthetubaguy:   can you unabandon thhese patches https://review.openstack.org/#/c/308198/ https://review.openstack.org/#/c/328280/ ?18:16
*** adrian_otto has quit IRC18:17
*** adrian_otto has joined #openstack-nova18:19
*** raunak has quit IRC18:20
*** slaweq_ has quit IRC18:20
*** mdrabe_ has joined #openstack-nova18:27
*** mdrabe has quit IRC18:28
*** READ10 has quit IRC18:29
*** annegentle has joined #openstack-nova18:29
*** nicolasbock has quit IRC18:33
*** mvk has joined #openstack-nova18:33
*** raunak has joined #openstack-nova18:34
*** ociuhandu has joined #openstack-nova18:36
*** priteau has joined #openstack-nova18:38
*** fragatina has joined #openstack-nova18:40
*** fragatina has quit IRC18:40
*** fragatina has joined #openstack-nova18:41
*** nicolasbock has joined #openstack-nova18:41
*** lpetrut has joined #openstack-nova18:46
*** lucasagomes is now known as lucas-afk18:46
mriedemrfolco: do you know who juergen leopold at ibm is? running ci on nova. i'm guessing a z person.18:52
rfolcomriedem, hmm let me check, I knew Arne from zKVM CI from Germany these days...18:52
rfolcomriedem, I guess his on the same team. They were having problems with their CI last week.18:54
mriedemok, i just emailed him and copied markus18:54
mriedemasking to start using a system id rather than his email18:55
mriedemand just asking what he's running18:55
*** efoley__ has quit IRC18:58
*** tblakes has quit IRC18:59
*** tblakes has joined #openstack-nova19:01
*** mnestratov has quit IRC19:01
openstackgerritPushkar Umaranikar proposed openstack/nova master: Remove unused columns from BuildRequest table in nova_api db  https://review.openstack.org/42225919:01
*** dharinic is now known as dharinic|lunch19:02
*** mdrabe_ has quit IRC19:04
*** Jeffrey4l has quit IRC19:05
*** Jeffrey4l has joined #openstack-nova19:05
*** tblakes has quit IRC19:06
*** kfarr has joined #openstack-nova19:10
openstackgerritSujitha proposed openstack/nova master: Adding tags field to InstancePayload  https://review.openstack.org/40722819:13
openstackgerritSujitha proposed openstack/nova master: Change tags to default field in Instance object.  https://review.openstack.org/41529819:13
openstackgerritSujitha proposed openstack/nova master: Adding auto_disk_config field to InstancePayload  https://review.openstack.org/41918519:14
openstackgerritSujitha proposed openstack/nova master: Reduce calls to load_tags() to 0  https://review.openstack.org/43514619:14
*** liangy has joined #openstack-nova19:14
*** abalutoiu_ has joined #openstack-nova19:17
*** liusheng has quit IRC19:18
*** liusheng has joined #openstack-nova19:19
*** salv-orl_ has joined #openstack-nova19:24
cfriesensomeone pointed out to me that InstanceActionEvent.pack_action_event_finish() uses a "message" field, but the InstanceActionEvent model in the DB uses "details".  Is this a bug?19:24
cfriesenand given that it's been in there since 2014, is anyone actually looking at it?19:26
*** salv-orlando has quit IRC19:27
openstackgerritPushkar Umaranikar proposed openstack/nova master: Remove config_drive attribute from build_requests table  https://review.openstack.org/43251119:28
*** dharinic|lunch is now known as dharinic19:30
*** jdillaman has quit IRC19:33
openstackgerritOpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements  https://review.openstack.org/44335719:37
sean-k-mooneysdague: o/19:39
openstackgerritOpenStack Proposal Bot proposed openstack/os-vif master: Updated from global requirements  https://review.openstack.org/44512019:39
sean-k-mooneysdague: not sure if your about currently but could you take a look at https://review.openstack.org/#/c/444788/ if you get a chance19:39
*** bvanhav has quit IRC19:39
sean-k-mooneysdague: i asked sc68cal  earlier and he said he would like you to take a look at it first.19:40
openstackgerritMatthew Booth proposed openstack/nova master: Add InstanceMappingList.get_by_instance_uuids  https://review.openstack.org/44513919:40
openstackgerritMatthew Booth proposed openstack/nova master: Add RequestContext.for_cell()  https://review.openstack.org/44514019:40
openstackgerritMatthew Booth proposed openstack/nova master: Add InstanceList.get_by_instance_mapping_list()  https://review.openstack.org/44514119:40
openstackgerritMatthew Booth proposed openstack/nova master: Make server_external_events use an optimal number of db queries  https://review.openstack.org/44514219:40
mdboothdansmith: I was playing with the cell context thing and came up with this series ^^^19:42
sean-k-mooneysdague: it addresses a sidefect of https://review.openstack.org/#/c/435575/ which due to the fact that devstack installes test-requirements globally effect the test cases of ovs in our ci and any other pep8 tests that are not run in a virtual envirionment19:42
mdboothdansmith: It only affects server_external_events, but the idea's there.19:43
*** salv-orl_ has quit IRC19:45
openstackgerritAnusha Unnam proposed openstack/nova master: Add keypairs field to InstancePayload  https://review.openstack.org/41973019:46
*** rmk has joined #openstack-nova19:47
*** rmk has quit IRC19:47
*** rmk has joined #openstack-nova19:47
*** sgordon` is now known as sgordon19:52
*** amoralej is now known as amoralej|off19:57
*** Jeffrey4l has quit IRC20:00
*** Jeffrey4l has joined #openstack-nova20:01
sdaguesean-k-mooney: wait, this is all to include flake8-import-order?20:04
sdaguecan we just revert that instead?20:04
sdaguesean-k-mooney: or more importantly, please explain to me the actual issue that's currently happening, how we got there, and why this is the fix?20:05
*** jheroux has quit IRC20:05
sean-k-mooneysdague: well the flake8-import-order change is what broke use but if your are running tempest test only there is noe reasong to install all the test requirements20:06
sean-k-mooneysdague: so this change also speed up our test runs20:06
sdagueso, why did flake8-import-order break you?20:06
sean-k-mooneysdague: but basically when flake8-import-order was added to the test requirements devstack installes it systemwide. flake8-import-order enables it test by default20:07
*** tblakes has joined #openstack-nova20:07
sdaguesean-k-mooney: the thing we've been here before, it used to be conditional20:07
sdagueok, so I think that's actually a really good reason to remove flake8-import-order from global requirements20:08
sdagueand not allow it20:08
sean-k-mooneywhen we build ovs from source to test ovs-dpdk there makefile enable pep testing if flake8 is installed20:08
sean-k-mooneyovs does not enforce ordering so it cause the make command to fail and then we broke20:08
sean-k-mooneysdague: well that is one option20:08
sdagueright, flake8-import-order seems to be far more intrusive to everything than previously realized20:09
sdagueand I think the answer is just remove it20:09
*** mdrabe_ has joined #openstack-nova20:09
sean-k-mooneybut why is devstack installing the test dependecies globally?20:09
sdaguebecause everyone wanted that20:09
sean-k-mooneythat seams just wronge20:09
sdaguebecause they wanted devstack to be setting up a dev / test environment20:10
sean-k-mooneywhat why is that better then leaving tox install them20:10
sdaguewe can go dig up that change, the code used to be with a conditional here20:10
sean-k-mooneyyes and when i use devstack to set up a test environment i always use tox to run my test and i creats a python virtual enve and installs the test requirement in it20:10
sdagueyeh, and some of those tox environments use system libs20:11
sean-k-mooneysdague: ok but is there any reson to not make it condional and default to true as we proposed20:11
sdaguebecause it's going to be another odd point of difference here20:11
*** crushil has quit IRC20:12
sdagueand every one of these end up causing confusion20:12
sean-k-mooneyyes but devstack is breaking the seperation of test and real requirements20:12
sdaguesean-k-mooney: there are a bunch of different things that are bleeding across, as far as I can tell20:12
*** amotoki has quit IRC20:13
sdagueflake8-import-order enables true by default, which is unexpected20:13
*** slaweq_ has joined #openstack-nova20:13
sdagueovs-dpdk is running pep8 without isolation to a venv, so it gets the impact of systems20:13
sdaguesystem installed packages20:13
sdaguedevstack is installing test-requirements system level because that is what people previously wanted20:14
sdaguesean-k-mooney: you will also find that many projects use test-requirements to specify optional dependencies20:14
sean-k-mooneysdague: yep20:15
sdaguelike all the db drivers are in the nova test-requirements20:15
sean-k-mooneythat is why installing test requiremetns is a bad thing in my view as we neever test without them currently20:15
edleafebauzas: if a compute node goes offline, where is that caught? Currently we don't have any way of removing its resource provider information in that event.20:15
sdagueIf you would like to get everything working with test-requirements not installed at all, I'm good with that20:15
sdagueI don't want a switch there20:15
sdaguebecause that is just, my 3rd party CI can just barely work based on the contents in different requirements files20:16
sdagueand something else will change, and break you20:16
sean-k-mooneywell devstack installs fine with the switch set to false and teh tempest squite works. im not sure about none tempets jobs though20:17
sean-k-mooneysdague: well neutron are build ovs from source for their fuctional and fullstak jobs currently.20:17
sdaguesean-k-mooney: in some configuration you happen to have20:17
sean-k-mooneysdague: they dont deploy nova for those but it would have broke them if they did when flake8-import-order was introduced in nova20:18
sdaguesean-k-mooney: sure, and there are 2 other options here. One is deal with the fact that flake8-import-order is really pretty suplerfuous20:18
sdaguethe other is get ovs-dpdk to run it's pep8 during build under isolation20:18
sean-k-mooneywell that would require chaging upstream ovs makefile which we coudl explore but it would be up to them to accept that change20:19
sdaguesean-k-mooney: or create and activate a venv before running it20:19
sean-k-mooneysdague: so activate venv and run make as part of our devstack plugin then deactivte20:20
sdaguesean-k-mooney: sure, if that works for the isolation20:20
sean-k-mooneysdague: thats an option that i had not taught of. thats certenly possibel.20:21
*** efried has joined #openstack-nova20:21
sdagueI also, honestly think that flake8-import-order being installed globally seems weird just in general due to fallout20:21
efriedI just joined, so I don't have the scrollback, but +1 to that.20:21
sean-k-mooneysdague: yes but to me that is a direct resutl of the devstack behavor20:21
sdagueand I would tend to pull that from global requirements, as I think that's going to have knock on break effects for lots of people that flake8 from the system (which is not an unreasonable thing to want to do)20:21
*** mnestratov has joined #openstack-nova20:22
sean-k-mooneysdague: generally test-requireemnt.txt are not expected to be installed systemwide or at least i never assume they will be20:22
sdaguesean-k-mooney: I accept that if you make it so that we never install them20:22
*** 07EAAIRVF has joined #openstack-nova20:23
sdagueI don't want to be in a weird middle ground where we have to explain to devs... "oh, well for that use case you have to set this to TRUE, but to get this other thing to work, it has to be FALSE"20:23
sean-k-mooneysdague: ill look into the vnev to wrap make, but would you be open to a patch to never install test requirement in devstack then without the conditional20:24
sdaguesean-k-mooney: yep, if you can validate that it's going to work for people20:24
sdaguegit log -S TESTONLY20:26
sdaguewill give you some of the change history about when this was conditional20:26
sean-k-mooneywell i can ask on the mailing list if people have objection to that change and we could try running the exsitng gate jobs for several projects with the proposed patch and see if they break?20:27
*** adrian_otto has quit IRC20:27
sean-k-mooneygit log -S TESTONLY ? in what repo20:27
sdaguesean-k-mooney: well, the issue is also for developer use case, not just the gate case20:27
sdaguesean-k-mooney: devstack repo20:27
sean-k-mooneyah yes but again when im doing upstream dev i use devstack to set up my enviornment but i have always used the tox workflow so im not sure how many people expect them to be system wide20:28
sean-k-mooneybut that is a vail point20:28
sean-k-mooneyit has to work for both usecases20:29
*** fragatin_ has joined #openstack-nova20:30
sean-k-mooneysdague: any thanks for taking a look at this. ill feedback you suggestion to the rest of my team and we will see if the venv change works. if not we will procuse this as an addtional path. as for flake8-import-order are you going to propose a revert or let it stand for now20:33
*** smatzek_ has quit IRC20:33
*** fragatina has quit IRC20:33
sean-k-mooney*anyway20:33
openstackgerritSTEW TY proposed openstack/nova master: Transform instance.resize_revert notification  https://review.openstack.org/39681120:34
*** jamielennox|away is now known as jamielennox20:37
rfolcocdent, around?20:39
cdentyeah rfolco, for a little while longer, what's up?20:39
sdaguesean-k-mooney: I'll +2 a removal of it if you put it up20:39
sdaguewe can leave the import reordering, just pull the install of it20:39
rfolcocdent, quick question: how to make tox -epy27 run my new created lass under test_microversions.py? where am I noob'ing ?20:40
sdaguesean-k-mooney: then look at removing it from global-requirements20:40
edleaferfolco: what command are you running now?20:40
cdentrfolco: you mean running just one test?20:40
rfolcotox -epy27 test_microversion.TestMicroversionUtility20:40
rfolcoits ignoring my class20:41
sean-k-mooneysdague: of flake8-import-order, ok i can do that tomorow, i am just wrapping up for the evening so ill add you to the review when its up.20:41
sean-k-mooneysdague: and ya i did not want to revert the whole patch as everything else looked good to me.20:41
cdentrfolco: if you do 'tox -epy27 test_microversion' does it get all the tests in the file?20:41
rfolcoall them, except mine :)20:42
edleaferfolco: did you save your changes? :)20:42
cdentrfolco: what's the name of your test method(s)/ in the class?20:42
rfolcoclass TestMicroversionUtility(test.NoDBTestCase): .... and def raise_405_out_of_date_version(self):20:43
cdents/raise/test_raise/20:43
edleaferfolco: test names have to start with 'test_'20:43
rfolcooh boy20:43
rfolcocdent, edleafe: thats it. Thx20:44
cdentfrustratring isn't it?20:44
cdentwow, it's a bad day for typing. it might be time for me to retire for the day20:44
rfolcocdent, silly mistake :-/20:45
edleafecdent: I'll be leaving early to catch this tonight: https://www.tobincenter.org/box-office/2017-03/purple-rain-live20:45
cdent\o/20:46
edleafeOur daughter is in the philharmonic20:46
cdentthat'll be fun20:46
cdentyeah, this cold is killing me, I'm calling it a day20:47
* cdent waves20:47
*** cdent has quit IRC20:47
*** 07EAAIRVF has quit IRC20:52
openstackgerritSTEW TY proposed openstack/nova master: Transform instance.unrescue notifications  https://review.openstack.org/38827520:53
mriedemhmm, functional test in https://review.openstack.org/#/c/443395/ fails in ocata, methinks we need the context untarget thing...20:56
*** catinthe_ has joined #openstack-nova20:56
openstackgerritEric Berglund proposed openstack/nova master: PowerVM Driver: spawn/destroy #2: functional  https://review.openstack.org/43859820:56
*** Sukhdev has joined #openstack-nova20:57
mriedemhmm, but the context should have a cell set on it in ocata20:57
*** catintheroof has quit IRC20:58
*** edmondsw has quit IRC20:58
*** satyar has joined #openstack-nova20:59
*** gfhellma has quit IRC21:00
*** awaugama has quit IRC21:02
satyarjohnthetubaguy: tested the fix https://review.openstack.org/#/c/437222 for bug https://bugs.launchpad.net/nova/+bug/166826721:03
openstackLaunchpad bug 1668267 in OpenStack Compute (nova) "Nova fails to correctly update the quota when deleting lots of VMs, some of which previously failed due to a quota error" [High,In progress] - Assigned to Pushkar Umaranikar (pushkar-umaranikar)21:03
satyarstill able to reproduce the issue21:03
dansmithsatyar: that's not the fix21:04
dansmithsatyar: this is: https://review.openstack.org/#/c/443395/21:04
mriedemwell,21:04
mriedemon master it's that and https://review.openstack.org/#/c/443403/21:04
mriedemon ocata it should just be https://review.openstack.org/#/c/443395/ and i'm working on the backport but hitting some issues21:05
satyarohh ok will be testing it Thanks :)21:06
*** aysyd has quit IRC21:06
satyarthanks dansmith and mriedem21:06
*** gouthamr has quit IRC21:06
mriedemsatyar: i've marked the other bug as a duplicate21:08
satyarto?21:09
satyarok got it21:09
mriedemhttps://bugs.launchpad.net/nova/+bug/167062721:09
openstackLaunchpad bug 1670627 in OpenStack Compute (nova) ocata "quota is always in-use after delete the ERROR instances " [Critical,Confirmed]21:09
*** youzp has joined #openstack-nova21:10
openstackgerritDan Smith proposed openstack/nova master: Add SameCellFilter  https://review.openstack.org/44503521:10
*** youzp has quit IRC21:10
satyarideally it should be opposite that bug should be duplicate of the first one ;) but its fine21:11
satyaraslong as we fix the issue :)21:11
mriedemsatyar: yeah, i guess triage failed us, it went under the one it did because someone showed up in the channel last wednesday and brought it up21:12
mriedemright place right time21:12
satyar:P21:12
*** dimtruck is now known as zz_dimtruck21:13
*** tbachman has quit IRC21:15
*** penick has joined #openstack-nova21:21
openstackgerritSivasathurappan Radhakrishnan proposed openstack/nova-specs master: Live Migration of Rescued Instances  https://review.openstack.org/34716121:23
openstackgerritSujitha proposed openstack/nova master: Add description to policies in agents.py  https://review.openstack.org/44520821:28
*** salv-orlando has joined #openstack-nova21:30
*** catintheroof has joined #openstack-nova21:31
openstackgerritSivasathurappan Radhakrishnan proposed openstack/nova master: Reset instance status on invalid host  https://review.openstack.org/40100921:33
*** Jeffrey4l has quit IRC21:35
*** satyar has quit IRC21:35
mriedemwtf, in ocata, when the instance is in cell0 and i go to delete it, i get a 404, but when i do a GET i get a response21:35
*** catinthe_ has quit IRC21:35
*** Jeffrey4l has joined #openstack-nova21:35
*** zz_dimtruck is now known as dimtruck21:35
*** smatzek_ has joined #openstack-nova21:43
*** felipemonteiro has quit IRC21:43
melwittmriedem: are you using novaclient to delete it or the REST API directly? because I feel like I have seen that before when I had bugs in my own code because of how novaclient does a GET before doing the delete. I can't remember the details off the top of my head21:43
*** eharney has quit IRC21:44
mriedemmelwitt: rest api, this is that functional regression test for the quota patch21:46
mriedemhttps://review.openstack.org/#/c/443395/ but on ocata21:46
mriedemthe functional regression test fails waiting for the server to be gone after the delete request, and it's because the delete request fails with a 404 when looking up the instance.flavor21:47
melwittoh, hm21:47
*** lpetrut has quit IRC21:47
mriedemhttps://gist.github.com/mriedem/cece80e5ab375c2fad6d9d599851c2dc21:47
mriedemi think i know why,21:48
mriedemin master the context is targeted when you get the instance21:48
mriedemin ocata it's not,21:48
mriedemso in this change when it's looking up the flavor from the instance, the context isn't targeted to cell021:48
mriedembut we want the quotas in the main cell db,21:49
mriedemso i'm going to have to do the same helper method that i had on master for getting the flavor from a targeted context before creating the reservation21:49
mriedemugh21:49
openstackgerritAnusha Unnam proposed openstack/nova master: POC: add descriptions to server tags  https://review.openstack.org/43484221:50
*** fragatin_ has quit IRC21:51
*** fragatina has joined #openstack-nova21:52
*** annegentle has quit IRC21:58
dansmithmriedem: how do we not trip over that in tempest?21:58
mriedemdansmith: it's for this patch,21:58
mriedemas a result of this patch i mean21:58
mriedemi've got it sorted out, it's just kind of a mess for the backport,21:58
mriedemsince it's a mix of the 2 changes on master21:58
mriedemyou'll see shortly21:59
dansmithoh, I see I thought you meant the regression test doesn't run on ocata before the change21:59
*** smatzek_ has quit IRC21:59
*** smatzek has joined #openstack-nova21:59
openstackgerritmelanie witt proposed openstack/nova master: Count resources to check quota for cells  https://review.openstack.org/41652122:00
openstackgerritmelanie witt proposed openstack/nova master: Make Quotas object favor the API database  https://review.openstack.org/41094522:00
openstackgerritmelanie witt proposed openstack/nova master: Add online migration to move quotas to API database  https://review.openstack.org/41094622:00
*** burt has quit IRC22:01
mriedemi should have probably backported this first to see how it would work22:01
*** baoli has quit IRC22:01
*** kaisers_ has quit IRC22:02
*** jamielennox is now known as jamielennox|away22:03
*** smatzek has quit IRC22:04
mriedemlooks like we have another ocata regression https://review.openstack.org/#/c/444106/22:04
*** jamielennox|away is now known as jamielennox22:07
*** xyang1 has quit IRC22:07
*** yonglihe has quit IRC22:09
dansmithhmm, I thought we did that _on_ retry22:10
*** priteau has quit IRC22:10
*** priteau has joined #openstack-nova22:10
dansmithpopulate_filter_properties() creates the retry of the selected host22:11
dansmithwhich we are calling already22:12
dansmithoh22:12
dansmithonly if populate_retry has done the initial work first I guess22:12
dansmithhuh22:12
*** catintheroof has quit IRC22:13
mriedemgd how many postgres packages do i really need to install just to run pep822:14
*** catintheroof has joined #openstack-nova22:14
*** catintheroof has quit IRC22:14
*** catintheroof has joined #openstack-nova22:15
mriedemsudo apt-get install -y (*pg*|*pq*)22:15
mriedeminstall -y *everything*22:15
*** priteau has quit IRC22:15
openstackgerritDan Smith proposed openstack/nova master: Add SameCellFilter  https://review.openstack.org/44503522:16
*** mlavalle has quit IRC22:19
*** rfolco has quit IRC22:19
mriedemwe should have a functional regression test for that retry fix,22:20
mriedemshould be relatively simple, run 2 nova-compute services, make the scheduler pick the first one and make that one fail to trigger a retry, and then make sure it's actually retried to the 2nd host22:21
dansmithis that enough? it might do that by itself by chance22:21
openstackgerritSujitha proposed openstack/nova master: Add description to policies in aggregates.py  https://review.openstack.org/44523022:22
*** liangy has quit IRC22:23
openstackgerritAnusha Unnam proposed openstack/nova master: POC: add descriptions to server tags  https://review.openstack.org/43484222:23
openstackgerritSujitha proposed openstack/nova master: Add description to policies in agents.py  https://review.openstack.org/44520822:23
mriedemdansmith: how?22:23
mriedemw/o digging into details i guess i don't know22:23
mriedemi'd also like to be able to force the scheduler to pick a host in the test that we've setup to fail22:24
mriedemi'm not entirely sure how to do that in functional tests w/o stubbing something out22:24
dansmithif you asked the scheduler to pick what would have been its second choice and it retries, it'd choose the other one without this fix22:24
dansmithright exactly22:24
mriedemyeah we have to make the test pick the hosts in the order we need for the recreate22:24
mriedemthat's the tricky part,22:24
mriedembut doesn't seem too hard22:24
dansmithdon't we shuffle the results?22:25
mriedemi'd think a weigher could be used22:25
*** mdrabe_ is now known as mdrabe22:25
dansmithyeah maybe22:25
mriedemthat's scheduler black magic to me though22:25
*** Svilen_ has joined #openstack-nova22:25
dansmithhttps://github.com/openstack/nova/blob/master/nova/scheduler/filter_scheduler.py#L124-L12422:26
mriedemi don't want to make a big deal out of it, but this is a complicated piece of code and we clearly don't have it covered sufficiently in tests22:26
*** Svilen_ has left #openstack-nova22:26
dansmithyou'd have to set the subset size to 122:26
melwittit defaults to 1, so no shuffle by default22:26
dansmithokay22:26
mriedemare either of you interested in writing that functional test? :)22:27
dansmiththat seems like an unfortunate default for regular people22:27
*** tbachman has joined #openstack-nova22:27
mriedemit's only random if the weights are equal isn't it?22:27
*** dimtruck is now known as zz_dimtruck22:27
dansmithwait,22:28
dansmithoh, I see22:28
dansmithhuh, I never knew this picked a single host by default22:28
melwittmriedem: I can give it a go. I've written a weird retry test before https://github.com/openstack/nova/blob/master/nova/tests/functional/test_compute_mgr.py#L4022:28
*** gouthamr has joined #openstack-nova22:28
melwittI agree with you that we need coverage on this stuff22:28
*** mdrabe has quit IRC22:29
melwittI think maybe the subset thing is only a problem if you get parallel requests often (tm) that all pick the same host and retry. we did so we increased it from 122:30
dansmithsure, but that's a common complaint22:31
dansmiththat and running multiple schedulers choosing the same host22:31
melwittyeah. I guess no one thought of changing the default22:32
dansmithI just can't think of a good reason to have the default be 122:32
dansmithif it's larger it doesn't hurt low-traffic clouds, but being =1 does hurt high traffic ones by default22:32
mriedemisn't this exactly what johnthetubaguy was saying this morning, why people don't use the filter scheduler with packing?22:32
dansmithmriedem: yes22:32
mriedembecause we don't randomize anything22:32
dansmithmriedem: but remember I said "but we randomize the result"22:33
melwittmriedem today mentioned a good point in a review somewhere I think (?) that it might be since people want to stack instances rather than spread them usually22:33
dansmithmriedem: and he said "right, but at a certain level it doesn't matter"22:33
melwittwell, I typed that way too slowly22:33
mriedemmelwitt: yes because then you can't place large instances22:34
dansmithmriedem: for RAX-level, I can imagine it's not enough, but for medium sizes I would expect this works fine22:34
melwittyeah. well with subset = 1, that means no randomization. are you saying there's a different randomize somewhere?22:34
dansmithmriedem: but we're not doing anyone favors by defaulting that to 1, IMHO22:34
dansmithmelwitt: no, I'm saying even with that on, it's not enough to satisfy rax's need for parallelism22:35
melwittoh22:35
*** dave-mccowan has quit IRC22:36
*** esberglu has quit IRC22:36
*** baoli has joined #openstack-nova22:36
dansmithif you set it to infinity it defeats all your weighting right? so there's probably some inflection point where it starts hurting you22:36
melwittyeah, I'm not sure how that randomize plays with the weighing part22:37
dansmithI think we get back an ordered list from the weighers,22:38
melwittunless it just shuffles the weighed order of N subset22:38
dansmithand that subset size randomizes the N best-weighed hosts22:38
*** Swami has joined #openstack-nova22:38
melwittyeah22:38
dansmithso if you set it to N=$n_computes, you get no weighing22:38
melwittright22:39
dansmithwhich I hear is bad and stuff22:39
melwitthah22:39
*** catintheroof has quit IRC22:40
*** baoli has quit IRC22:41
*** Swami has quit IRC22:43
melwittso I guess a good-ish reason to default to 1 is because the actual good default will vary so much from deployment to deployment22:44
*** Swami has joined #openstack-nova22:44
*** bmace has joined #openstack-nova22:44
dansmithI dunno, if people have less than say five candidate machines they're probably running a pretty damned small cloud22:44
dansmithshooting for the middle and tweaking up or down depending on your size seems reasonable,22:45
dansmithbut yeah, someone will probably be confused with a two-node test cloud if it's !=122:45
melwittyeah, so I was thinking if you have 10 compute nodes and your subset defaulted to 2, then you could run into problems trying to schedule large instances?22:45
dansmithjust unfortunate22:45
dansmithno the weighed hosts are all candidates22:45
dansmithwe could make that default to -1 and choose a subset that is like 5% of the node count or something22:46
dansmithbut anyway..22:46
dansmithI just wonder how many people don't realize22:46
dansmithmight just be me22:46
jrollthat bit our ironic install, fwiw, because the candidate order is always the same as well22:47
melwittyeah, I was thinking if subset is 2 and you get a random choice of the two highest weights, then that will spread the instances slightly, no?22:47
jrollwe had a QE bot building periodically, chewed through the disks on the first few boxes in that list pretty quick22:48
melwittjroll: ah, that's an interesting point too22:49
mriedemdansmith: melwitt: so here is the quota usage fix backport for ocata, which is a mix of both fixes from pike; i've tried explaining that in the commit message: https://review.openstack.org/#/c/445236/22:50
mriedemlet me know if that's terrible22:50
melwittmriedem: looks okay to me. commit message explanation is good22:54
mriedemok. backporting the other change from pike wouldn't help either b/c the functional test on the 1st change still fails without the tweaks i had to make here22:57
mriedemso the only thing you could really do if backporting both changes is to squash them22:57
*** smatzek has joined #openstack-nova22:58
dansmithyeah, I guess if you want pack first, then anything but =1 spreads them23:00
dansmithlike pack first above anything else23:00
dansmithwhich might've been john's point since I think they do care about pack first, and might be why the randomization doesn't help him23:01
melwittah, yeah23:04
openstackgerritSujitha proposed openstack/nova master: Adding tags field to InstancePayload  https://review.openstack.org/40722823:05
*** esberglu has joined #openstack-nova23:06
openstackgerritSujitha proposed openstack/nova master: Change tags to default field in Instance object.  https://review.openstack.org/41529823:06
*** iceyao has joined #openstack-nova23:06
openstackgerritSujitha proposed openstack/nova master: Reduce calls to load_tags() to 0  https://review.openstack.org/43514623:06
openstackgerritSujitha proposed openstack/nova master: Adding auto_disk_config field to InstancePayload  https://review.openstack.org/41918523:06
*** smatzek has quit IRC23:09
*** iceyao has quit IRC23:10
*** esberglu has quit IRC23:10
*** hongbin has quit IRC23:15
*** efried has quit IRC23:15
openstackgerritSujitha proposed openstack/nova master: Add description to policies in assisted_volume_snapshots.py  https://review.openstack.org/44523823:16
*** tbachman has quit IRC23:16
*** lucasxu has quit IRC23:19
*** rfolco has joined #openstack-nova23:24
*** rfolco has quit IRC23:24
*** erlon has quit IRC23:25
*** tbachman has joined #openstack-nova23:25
*** mtanino has joined #openstack-nova23:26
gmannmriedem: johnthetubaguy sdague alex_xu oomichi api-ref for server sort/filter keys - https://review.openstack.org/#/c/421760/23:27
*** claudiub has quit IRC23:39
*** takashin has joined #openstack-nova23:39
*** karimb has quit IRC23:42
*** tbachman has quit IRC23:47
*** tbachman has joined #openstack-nova23:50

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