*** ljxiash has joined #openstack-nova | 00:00 | |
*** Fdaisuke has quit IRC | 00:01 | |
*** dave-mccowan has joined #openstack-nova | 00:01 | |
*** ljxiash has quit IRC | 00:04 | |
*** alex_xu has joined #openstack-nova | 00:05 | |
*** READ10 has joined #openstack-nova | 00:16 | |
*** TravT has joined #openstack-nova | 00:33 | |
*** greghaynes has quit IRC | 00:33 | |
*** WillUK2010 has joined #openstack-nova | 00:33 | |
*** TravT has quit IRC | 00:37 | |
*** READ10 has quit IRC | 00:38 | |
*** gmann has joined #openstack-nova | 00:39 | |
*** greghaynes has joined #openstack-nova | 00:41 | |
*** irenab has joined #openstack-nova | 00:46 | |
*** WillUK2010 has quit IRC | 00:49 | |
*** claudiub|4 has quit IRC | 00:49 | |
*** WillUK2010 has joined #openstack-nova | 00:49 | |
*** WillUK2010 has quit IRC | 00:50 | |
*** irenab has quit IRC | 00:50 | |
*** hfu has joined #openstack-nova | 00:51 | |
*** hfu has quit IRC | 00:51 | |
*** hfu has joined #openstack-nova | 00:52 | |
*** jamesdenton has joined #openstack-nova | 00:57 | |
*** jamesdenton has quit IRC | 01:03 | |
*** hfu has quit IRC | 01:04 | |
*** ducnc has joined #openstack-nova | 01:15 | |
*** grumpycatt has quit IRC | 01:16 | |
*** grumpycatt has joined #openstack-nova | 01:16 | |
*** kiran-r has joined #openstack-nova | 01:18 | |
*** ducnc has quit IRC | 01:20 | |
*** dimtruck is now known as zz_dimtruck | 01:20 | |
*** zhangjl has joined #openstack-nova | 01:20 | |
openstackgerrit | Tovin Seven proposed openstack/nova: Remove mox in virt/xenapi/image/test_bittorrent.py https://review.openstack.org/378496 | 01:22 |
---|---|---|
*** thorst has joined #openstack-nova | 01:24 | |
openstackgerrit | Rong Han proposed openstack/nova: Make hypervisor-servers support one host has multi-compute-nodes. https://review.openstack.org/398661 | 01:24 |
*** kiran-r has quit IRC | 01:27 | |
openstackgerrit | Tovin Seven proposed openstack/nova: Remove mox in virt/xenapi/image/test_bittorrent.py https://review.openstack.org/378496 | 01:32 |
*** hfu has joined #openstack-nova | 01:34 | |
*** hfu has quit IRC | 01:37 | |
openstackgerrit | Tovin Seven proposed openstack/nova: Remove mox in virt/xenapi/image/test_bittorrent.py https://review.openstack.org/378496 | 01:39 |
*** tovin07 has joined #openstack-nova | 01:40 | |
*** irenab has joined #openstack-nova | 01:40 | |
*** zhangyufei has joined #openstack-nova | 01:41 | |
*** yuntongjin has joined #openstack-nova | 01:43 | |
*** irenab has quit IRC | 01:45 | |
*** zz_dimtruck is now known as dimtruck | 01:48 | |
*** mnaser has quit IRC | 01:52 | |
*** zhs_ has joined #openstack-nova | 01:54 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_instance_info_cache https://review.openstack.org/307155 | 01:54 |
*** ljxiash has joined #openstack-nova | 01:54 | |
*** huanxie has quit IRC | 01:59 | |
*** zhs_ has quit IRC | 02:00 | |
*** mnaser has joined #openstack-nova | 02:07 | |
*** huanxie has joined #openstack-nova | 02:10 | |
*** moshele has joined #openstack-nova | 02:28 | |
*** thorst has quit IRC | 02:31 | |
*** thorst has joined #openstack-nova | 02:31 | |
*** hongbin has joined #openstack-nova | 02:32 | |
*** TravT has joined #openstack-nova | 02:34 | |
*** TravT has quit IRC | 02:35 | |
*** wxy__ has quit IRC | 02:36 | |
*** TravT has joined #openstack-nova | 02:36 | |
*** thorst has quit IRC | 02:40 | |
*** TravT has quit IRC | 02:41 | |
*** diga has quit IRC | 02:42 | |
*** shuquan has joined #openstack-nova | 02:44 | |
*** RuiChen has joined #openstack-nova | 02:46 | |
*** kaisers__ has joined #openstack-nova | 02:48 | |
*** Jeffrey4l has quit IRC | 02:51 | |
*** kaisers_ has quit IRC | 02:51 | |
*** harlowja has quit IRC | 02:57 | |
openstackgerrit | gecong proposed openstack/nova-specs: Fix a grammar mistake https://review.openstack.org/400038 | 03:07 |
*** shuquan has quit IRC | 03:08 | |
*** moshele has quit IRC | 03:10 | |
*** rmstar has quit IRC | 03:14 | |
*** efried has joined #openstack-nova | 03:14 | |
*** efried1 has quit IRC | 03:15 | |
*** rmstar has joined #openstack-nova | 03:15 | |
*** Sukhdev has joined #openstack-nova | 03:15 | |
*** thorst has joined #openstack-nova | 03:24 | |
*** irenab has joined #openstack-nova | 03:28 | |
*** thorst has quit IRC | 03:32 | |
*** irenab has quit IRC | 03:32 | |
*** hfu has joined #openstack-nova | 03:33 | |
*** yuntongjin has quit IRC | 03:36 | |
*** shuquan has joined #openstack-nova | 03:38 | |
*** namnh has joined #openstack-nova | 03:40 | |
*** thorst has joined #openstack-nova | 03:43 | |
*** thorst has quit IRC | 03:44 | |
openstackgerrit | Huan Xie proposed openstack/nova: xenapi: support the hotplug of a neutron port https://review.openstack.org/365505 | 03:45 |
*** links has joined #openstack-nova | 03:46 | |
*** hfu has quit IRC | 03:53 | |
*** mdrabe has joined #openstack-nova | 03:59 | |
*** gongysh2 has quit IRC | 04:01 | |
*** shuquan has quit IRC | 04:02 | |
*** fragatina has quit IRC | 04:07 | |
*** cubix_ has joined #openstack-nova | 04:07 | |
*** fragatina has joined #openstack-nova | 04:07 | |
*** dave-mccowan has quit IRC | 04:12 | |
*** gmann_ has joined #openstack-nova | 04:17 | |
*** links has quit IRC | 04:17 | |
*** irenab has joined #openstack-nova | 04:22 | |
*** gongysh2 has joined #openstack-nova | 04:27 | |
*** irenab has quit IRC | 04:27 | |
*** shuquan has joined #openstack-nova | 04:27 | |
*** huanxie has quit IRC | 04:29 | |
*** shuquan has quit IRC | 04:32 | |
*** TravT has joined #openstack-nova | 04:37 | |
*** TravT has quit IRC | 04:42 | |
*** Jeffrey4l has joined #openstack-nova | 04:42 | |
*** Jeffrey4l has quit IRC | 04:43 | |
*** Jeffrey4l has joined #openstack-nova | 04:44 | |
*** cubix_ is now known as kubix | 04:44 | |
*** hongbin has quit IRC | 04:54 | |
*** mdrabe has quit IRC | 05:01 | |
*** prateek has joined #openstack-nova | 05:03 | |
*** zhangyufei has quit IRC | 05:13 | |
*** avolkov has joined #openstack-nova | 05:14 | |
*** shuquan has joined #openstack-nova | 05:14 | |
*** sandanar has joined #openstack-nova | 05:14 | |
*** irenab has joined #openstack-nova | 05:16 | |
*** yamahata has joined #openstack-nova | 05:16 | |
*** huanxie has joined #openstack-nova | 05:18 | |
openstackgerrit | Alex Xu proposed openstack/nova: Remove the description of compute_api_class option https://review.openstack.org/399386 | 05:20 |
*** irenab has quit IRC | 05:21 | |
*** claudiub|4 has joined #openstack-nova | 05:26 | |
*** yamahata has quit IRC | 05:31 | |
*** GB21 has joined #openstack-nova | 05:32 | |
*** dikonoor has joined #openstack-nova | 05:32 | |
*** gcb has joined #openstack-nova | 05:32 | |
*** shuquan has quit IRC | 05:35 | |
*** kevinz has quit IRC | 05:35 | |
*** shuquan has joined #openstack-nova | 05:38 | |
*** TravT has joined #openstack-nova | 05:38 | |
openstackgerrit | NITIN GUPTA proposed openstack/python-novaclient: Added a new nova CLI command "nova delete-all" https://review.openstack.org/393923 | 05:40 |
*** TravT has quit IRC | 05:43 | |
*** yuntongjin has joined #openstack-nova | 05:44 | |
*** ljxiash has quit IRC | 05:45 | |
*** ljxiash has joined #openstack-nova | 05:45 | |
*** ljxiash has quit IRC | 05:49 | |
*** ljxiash has joined #openstack-nova | 05:49 | |
*** mdnadeem has joined #openstack-nova | 05:50 | |
*** mdnadeem has quit IRC | 05:51 | |
*** mdnadeem has joined #openstack-nova | 05:53 | |
*** unicell1 has quit IRC | 05:53 | |
*** unicell has joined #openstack-nova | 05:53 | |
*** Sukhdev has quit IRC | 05:54 | |
*** ratailor has joined #openstack-nova | 05:56 | |
*** GB21 has quit IRC | 05:56 | |
*** kaisers__ has quit IRC | 05:56 | |
*** sridharg has joined #openstack-nova | 05:59 | |
*** I has joined #openstack-nova | 06:08 | |
*** sudipto_ has joined #openstack-nova | 06:08 | |
*** I is now known as Guest56972 | 06:08 | |
*** lpetrut has joined #openstack-nova | 06:10 | |
*** tojuvone has joined #openstack-nova | 06:13 | |
*** zhangyufei has joined #openstack-nova | 06:13 | |
*** shuquan has quit IRC | 06:17 | |
*** GB21 has joined #openstack-nova | 06:19 | |
*** shuquan has joined #openstack-nova | 06:25 | |
*** jaosorior has joined #openstack-nova | 06:25 | |
*** moshele has joined #openstack-nova | 06:30 | |
*** moshele has quit IRC | 06:31 | |
*** moshele has joined #openstack-nova | 06:31 | |
*** moshele has quit IRC | 06:32 | |
*** moshele has joined #openstack-nova | 06:33 | |
*** shuquan has quit IRC | 06:33 | |
*** GB21 has quit IRC | 06:33 | |
*** yuntongjin has quit IRC | 06:34 | |
*** TravT has joined #openstack-nova | 06:39 | |
*** yuntongjin has joined #openstack-nova | 06:43 | |
*** TravT has quit IRC | 06:44 | |
*** Alagar has joined #openstack-nova | 06:46 | |
*** Alagar has quit IRC | 06:46 | |
*** jaosorior has quit IRC | 06:46 | |
*** jaosorior has joined #openstack-nova | 06:47 | |
*** bhagyashris has joined #openstack-nova | 06:49 | |
*** tojuvone_ has joined #openstack-nova | 06:51 | |
*** tojuvone__ has joined #openstack-nova | 06:51 | |
*** tojuvone__ has quit IRC | 06:52 | |
*** tojuvone has quit IRC | 06:52 | |
*** tojuvone_ has quit IRC | 06:52 | |
*** shuquan has joined #openstack-nova | 06:54 | |
*** mjura has joined #openstack-nova | 06:55 | |
*** shuquan has quit IRC | 07:00 | |
*** GB21 has joined #openstack-nova | 07:01 | |
*** tojuvone has joined #openstack-nova | 07:01 | |
*** mjura has quit IRC | 07:02 | |
*** mjura has joined #openstack-nova | 07:03 | |
*** shuquan has joined #openstack-nova | 07:05 | |
*** gmann_ has quit IRC | 07:05 | |
*** lpetrut has quit IRC | 07:10 | |
*** shuquan has quit IRC | 07:10 | |
*** Oku_OS-away is now known as Oku_OS | 07:14 | |
*** pcaruana has joined #openstack-nova | 07:17 | |
*** tuanluong has joined #openstack-nova | 07:21 | |
*** ekuris has joined #openstack-nova | 07:23 | |
*** shuquan has joined #openstack-nova | 07:26 | |
*** kevinz has joined #openstack-nova | 07:28 | |
*** swamireddy_ has quit IRC | 07:28 | |
*** andreas_s has joined #openstack-nova | 07:33 | |
*** lpetrut has joined #openstack-nova | 07:34 | |
*** rdo has quit IRC | 07:36 | |
*** rdo has joined #openstack-nova | 07:37 | |
*** belmoreira has joined #openstack-nova | 07:38 | |
*** kevinz has quit IRC | 07:38 | |
*** kevinz has joined #openstack-nova | 07:39 | |
*** lpetrut has quit IRC | 07:41 | |
*** nkrinner_afk is now known as nkrinner | 07:42 | |
*** kaisers_ has joined #openstack-nova | 07:44 | |
*** GB21 has quit IRC | 07:45 | |
*** shuquan has quit IRC | 07:47 | |
*** shuquan_ has joined #openstack-nova | 07:47 | |
*** matrohon has joined #openstack-nova | 07:55 | |
*** GB21 has joined #openstack-nova | 07:57 | |
*** irenab_ has joined #openstack-nova | 07:59 | |
*** kaisers__ has joined #openstack-nova | 07:59 | |
*** kaisers_ has quit IRC | 08:01 | |
*** matrohon has quit IRC | 08:02 | |
*** matrohon has joined #openstack-nova | 08:02 | |
*** irenab_ has quit IRC | 08:03 | |
*** boris-42 has quit IRC | 08:06 | |
*** edand has joined #openstack-nova | 08:07 | |
*** boris-42 has joined #openstack-nova | 08:08 | |
*** bhagyashris has quit IRC | 08:09 | |
*** shuquan_ has quit IRC | 08:10 | |
*** shuquan has joined #openstack-nova | 08:21 | |
*** bhagyashris has joined #openstack-nova | 08:21 | |
*** ralonsoh has joined #openstack-nova | 08:25 | |
*** WillUK2010 has joined #openstack-nova | 08:25 | |
*** shuquan has quit IRC | 08:27 | |
*** TommyLikeHu has joined #openstack-nova | 08:29 | |
TommyLikeHu | hey guys does olso.privsep has a irc channel, what's the of it? | 08:29 |
*** GB21 has quit IRC | 08:31 | |
*** nandal has joined #openstack-nova | 08:32 | |
*** paul-carlton has joined #openstack-nova | 08:33 | |
*** oanson has joined #openstack-nova | 08:36 | |
*** markus_z has joined #openstack-nova | 08:39 | |
*** TravT has joined #openstack-nova | 08:40 | |
*** shuquan has joined #openstack-nova | 08:42 | |
*** gongysh2 has quit IRC | 08:43 | |
*** TravT has quit IRC | 08:45 | |
*** GB21 has joined #openstack-nova | 08:47 | |
*** NobodyCam has quit IRC | 08:49 | |
bauzas | TommyLikeHu: probably #openstack-oslo | 08:51 |
*** NobodyCam has joined #openstack-nova | 08:52 | |
*** irenab_ has joined #openstack-nova | 08:53 | |
TommyLikeHu | bauzas: thanks | 08:55 |
*** shuquan has quit IRC | 08:56 | |
*** irenab_ has quit IRC | 08:57 | |
*** lpetrut has joined #openstack-nova | 08:58 | |
*** zzzeek has quit IRC | 09:00 | |
*** paul-carlton has quit IRC | 09:01 | |
*** zzzeek has joined #openstack-nova | 09:03 | |
*** shuquan has joined #openstack-nova | 09:03 | |
*** lpetrut has quit IRC | 09:05 | |
*** shuquan has quit IRC | 09:08 | |
*** lucas-afk is now known as lucasagomes | 09:09 | |
*** pilgrimstack has joined #openstack-nova | 09:16 | |
*** lpetrut has joined #openstack-nova | 09:21 | |
*** shuquan has joined #openstack-nova | 09:21 | |
*** gouthamr has joined #openstack-nova | 09:24 | |
*** GnomeKing has joined #openstack-nova | 09:26 | |
*** GnomeKing is now known as BobBall | 09:27 | |
*** shuquan has quit IRC | 09:27 | |
*** shuquan has joined #openstack-nova | 09:28 | |
*** paul-carlton has joined #openstack-nova | 09:29 | |
*** GB21 has quit IRC | 09:33 | |
*** nherciu has joined #openstack-nova | 09:35 | |
*** mhickey has joined #openstack-nova | 09:35 | |
*** ljxiash has quit IRC | 09:39 | |
*** ljxiash has joined #openstack-nova | 09:40 | |
*** zhangyufei has left #openstack-nova | 09:40 | |
*** yassine has joined #openstack-nova | 09:40 | |
*** zhangyufei has joined #openstack-nova | 09:40 | |
*** zhangyufei is now known as yufei | 09:40 | |
*** yassine is now known as Guest25976 | 09:41 | |
*** kevinz has quit IRC | 09:42 | |
*** kevinz has joined #openstack-nova | 09:43 | |
*** vassie has joined #openstack-nova | 09:44 | |
*** ljxiash has quit IRC | 09:45 | |
*** roeyc has joined #openstack-nova | 09:45 | |
*** GB21 has joined #openstack-nova | 09:45 | |
*** yufei has left #openstack-nova | 09:46 | |
*** yufei has joined #openstack-nova | 09:47 | |
*** irenab_ has joined #openstack-nova | 09:47 | |
*** derekh has joined #openstack-nova | 09:48 | |
*** shuquan has quit IRC | 09:49 | |
*** takashin has left #openstack-nova | 09:50 | |
*** takashin has joined #openstack-nova | 09:52 | |
*** irenab_ has quit IRC | 09:52 | |
*** moshele has quit IRC | 09:53 | |
*** edand has quit IRC | 09:54 | |
*** owalsh_ has joined #openstack-nova | 09:56 | |
*** owalsh_ has quit IRC | 09:57 | |
*** owalsh_ has joined #openstack-nova | 09:58 | |
*** owalsh_ has quit IRC | 09:58 | |
*** owalsh has joined #openstack-nova | 09:58 | |
*** alexpilotti has quit IRC | 09:58 | |
*** yuntongjin has quit IRC | 09:59 | |
*** ihrachys has joined #openstack-nova | 10:04 | |
*** shuquan has joined #openstack-nova | 10:04 | |
*** zhangjl has quit IRC | 10:04 | |
*** moshele has joined #openstack-nova | 10:05 | |
lyarwood | mdbooth: re https://bugs.launchpad.net/nova/+bug/1643017 , I'm looking into using the xml ready flag as a workaround now | 10:07 |
openstack | Launchpad bug 1643017 in OpenStack Compute (nova) "libvirtError: block copy still active: disk 'vdb' not ready for pivot yet" [Undecided,In progress] - Assigned to Lee Yarwood (lyarwood) | 10:07 |
mdbooth | lyarwood: Ok | 10:07 |
lyarwood | mdbooth: did you look into switching this code over to using the events that everyone keeps talking about btw? | 10:07 |
mdbooth | lyarwood: Remind me, which xml contains the ready flag? | 10:08 |
mdbooth | lyarwood: Only at a very high level. | 10:08 |
lyarwood | mdbooth: the device https://www.redhat.com/archives/libvir-list/2016-October/msg00227.html | 10:08 |
openstackgerrit | Daniel Berrange proposed openstack/nova: console: introduce basic framework for security proxying https://review.openstack.org/345396 | 10:08 |
openstackgerrit | Daniel Berrange proposed openstack/nova: console: introduce framework for RFB authentication https://review.openstack.org/345397 | 10:08 |
openstackgerrit | Daniel Berrange proposed openstack/nova: console: introduce the VeNCrypt RFB authentication scheme https://review.openstack.org/345398 | 10:08 |
openstackgerrit | Daniel Berrange proposed openstack/nova: console: provide an RFB security proxy implementation https://review.openstack.org/345399 | 10:08 |
mdbooth | lyarwood: I still think it's a tossup between this ready flag and attempting to backport the events thing, btw. | 10:09 |
lyarwood | mdbooth: yeah I was going to do the ready flag first as it should be easier and then look into the events stuff | 10:09 |
lyarwood | mdbooth: we are going to need this back to Liberty downstream | 10:09 |
mdbooth | Because although events will be more disruptive to the code, it is the libvirt codepath which has obviously had the most historic love. | 10:09 |
kashyap | lyarwood: See also the upstream openstack-dev discussion of that - http://lists.openstack.org/pipermail/openstack-dev/2016-October/105188.html | 10:10 |
mdbooth | kashyap: Did we determine the historic quirks of the ready flag in libvirt? | 10:10 |
mdbooth | i.e. how did it behave in our most ancient supported version? | 10:11 |
kashyap | mdbooth: I don't think we did an excavation of that, but fairly good discussion on upstream libvirt list thread | 10:11 |
*** shuquan has quit IRC | 10:11 | |
*** shuquan has joined #openstack-nova | 10:11 | |
kashyap | Hm, let me recall / search if there was a discussion of that | 10:11 |
*** dikonoor has quit IRC | 10:11 | |
mdbooth | lyarwood kashyap: That was the root of my slight preference for backporting events, btw. AFAICT it has always worked, so backports should be more reliable. | 10:13 |
*** GB21 has quit IRC | 10:13 | |
kashyap | mdbooth: Found it, a small public IRC discussion with Eric Blake: | 10:13 |
kashyap | 20:46 <@eblake> if I understand it right, ready:true was a reactionary commit (something else introduced the race, so ready: was proposed to tell if the race exists) | 10:13 |
kashyap | 20:46 <@eblake> prior to the race, cur==end was reliable | 10:13 |
kashyap | 20:47 <@eblake> ready:true was introduced in qemu 2.2 | 10:13 |
kashyap | 20:47 < kashyap> eblake: This is what the commit message says: http://git.qemu.org/?p=qemu.git;a=commitdiff;h=ef6dbf1e4 | 10:13 |
kashyap | 20:47 <@eblake> so I suspect that in qemu 2.1 was where the race was introduced | 10:13 |
kashyap | 20:48 <@eblake> and that qemu 2.0 or 1.x probably had a synchronous setup where you could never observer cur==end on a non-ready job | 10:13 |
*** tuanluong has quit IRC | 10:13 | |
kashyap | 20:48 <@eblake> but I don't remember enough history to point to when qemu switched jobs to be a bit more asynchronous | 10:13 |
mdbooth | kashyap lyarwood: Right. Clear as mud :) | 10:13 |
*** sahid has joined #openstack-nova | 10:13 | |
*** sambetts_ is now known as sambetts | 10:14 | |
kashyap | :P | 10:15 |
mdbooth | lyarwood: So it's a tradeoff between disruption to Nova and historic bugs in libvirt. | 10:18 |
*** dikonoor has joined #openstack-nova | 10:20 | |
lyarwood | mdbooth: hmmmmmm | 10:23 |
lyarwood | kashyap: do we know the libvirt version that introduced the ready flag into the XML? | 10:23 |
kashyap | lyarwood: Afraid, not top off my head, without doing the `git blame` thing | 10:24 |
mdbooth | lyarwood: Which xml contains the ready flag, btw? | 10:24 |
mdbooth | Assume you've got to fetch the job xml? | 10:24 |
mdbooth | It's not in the data structure we've already got in that code, is it? | 10:25 |
lyarwood | mdbooth: it's against the device AFAICT, so we'd just need to extend LibvirtConfigGuestDisk to handle it | 10:25 |
*** gabor_antal has quit IRC | 10:25 | |
lyarwood | mdbooth: https://www.redhat.com/archives/libvir-list/2016-October/msg00227.html | 10:26 |
lyarwood | within <mirror></mirror> | 10:26 |
*** mdovgal has joined #openstack-nova | 10:28 | |
kashyap | Yes, it's part of the device attribute, as Peter quotes the snippet in the above thread | 10:28 |
*** GB21 has joined #openstack-nova | 10:30 | |
*** yufei has quit IRC | 10:35 | |
*** shuquan has quit IRC | 10:39 | |
*** TravT has joined #openstack-nova | 10:41 | |
*** irenab_ has joined #openstack-nova | 10:41 | |
*** dikonoor has quit IRC | 10:44 | |
*** TravT has quit IRC | 10:45 | |
*** irenab_ has quit IRC | 10:46 | |
*** nandal has quit IRC | 10:47 | |
*** huanxie has quit IRC | 10:53 | |
*** dikonoor has joined #openstack-nova | 11:03 | |
*** namnh has quit IRC | 11:06 | |
*** rfolco has joined #openstack-nova | 11:07 | |
*** amoralej has joined #openstack-nova | 11:09 | |
amoralej | hi nova, https://bugs.launchpad.net/nova/+bug/1643511 is impacting tripleo and puppet jobs | 11:10 |
openstack | Launchpad bug 1643511 in tripleo "nova-api commands fail when running behind httpd" [Undecided,New] | 11:10 |
*** tesseract has joined #openstack-nova | 11:10 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/nova: Revert "[PY3] byte/string conversions and enable PY3 test" https://review.openstack.org/400193 | 11:10 |
*** tesseract is now known as Guest93740 | 11:11 | |
*** ljxiash has joined #openstack-nova | 11:16 | |
*** GB21 has quit IRC | 11:16 | |
*** prashkre has joined #openstack-nova | 11:20 | |
*** TTris has quit IRC | 11:22 | |
*** rwmjones is now known as rwmjones_holiday | 11:28 | |
*** mvk has quit IRC | 11:29 | |
mdbooth | johnthetubaguy: Can you have a poke at this one: https://review.openstack.org/#/c/383859/ | 11:30 |
mdbooth | Looking for a second +2 on this and the one before it | 11:31 |
*** irenab_ has joined #openstack-nova | 11:31 | |
mdbooth | We discussed it briefly at summit | 11:31 |
johnthetubaguy | mdbooth: yeah, I keep meaning to get to that, I have hope that will happen today | 11:31 |
mdbooth | johnthetubaguy: Thanks | 11:31 |
openstackgerrit | Lee Yarwood proposed openstack/nova: WIP libvirt: Use the mirror ready flag if available https://review.openstack.org/399735 | 11:34 |
openstackgerrit | Lee Yarwood proposed openstack/nova: libvirt: Wait for abort_job to complete when not resizing https://review.openstack.org/382449 | 11:34 |
*** TTris has joined #openstack-nova | 11:35 | |
lyarwood | mdbooth: ^ erm, not sure why git-review just pushed that, did you rebase it this morning? | 11:35 |
*** gszasz has joined #openstack-nova | 11:35 | |
lyarwood | mdbooth: ah I rebased it, sorry! | 11:35 |
mdbooth | lyarwood: I did not | 11:35 |
mdbooth | lyarwood: hehe | 11:35 |
mdbooth | lyarwood: You're not old enough for senior moments | 11:36 |
lyarwood | mdbooth: hehe, I rebased it on Friday, just forgot about it when working on this today | 11:36 |
lyarwood | mdbooth: anyway, that's just a WIP change to check this works in the gate, I'll tidy it up for review after lunch | 11:37 |
openstackgerrit | int32bit proposed openstack/nova: Fix wrong disk size for sparse flat files https://review.openstack.org/396873 | 11:37 |
*** GB21 has joined #openstack-nova | 11:39 | |
prashkre | alex_xu: Hi. Could you please take a look at Mattew's comments on https://review.openstack.org/391113 | 11:39 |
prashkre | johnthetubaguy: Hi John. Could you please review latest patch 12 on https://review.openstack.org/391113. | 11:41 |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in nova/tests/unit/virt/disk https://review.openstack.org/322778 | 11:41 |
*** tojuvone has quit IRC | 11:42 | |
*** lpetrut has quit IRC | 11:42 | |
*** Jeffrey4l has quit IRC | 11:45 | |
*** tojuvone has joined #openstack-nova | 11:46 | |
alex_xu | prashkre: yea, let me check | 11:49 |
*** sudipto_ has quit IRC | 11:50 | |
*** Guest56972 has quit IRC | 11:50 | |
*** vassie has quit IRC | 11:50 | |
*** lpetrut has joined #openstack-nova | 11:51 | |
alex_xu | prashkre: just done | 11:55 |
*** Guest25976 has quit IRC | 11:55 | |
*** nicolasbock has joined #openstack-nova | 11:58 | |
*** dave-mccowan has joined #openstack-nova | 12:04 | |
*** ratailor has quit IRC | 12:08 | |
moshele | johnthetubaguy: hi, I know we are passed the spec freeze, so I was wonder if now is a good time to send Spec freeze exception request in the ML?. I would really like to https://review.openstack.org/#/c/182242/ spec to be approved. It had +2 from danpb before | 12:10 |
johnthetubaguy | moshele: unsure what the plan is for those, I would try catch mriedem if he is online later | 12:11 |
johnthetubaguy | moshele: we probably have double the specs approved for the amount of review bandwidth we have right now | 12:11 |
johnthetubaguy | (double is optimistic) | 12:11 |
*** rtheis has joined #openstack-nova | 12:13 | |
moshele | johnthetubaguy: ok thanks | 12:13 |
*** pilgrimstack has quit IRC | 12:14 | |
*** bvanhav has joined #openstack-nova | 12:18 | |
-openstackstatus- NOTICE: We are currently having capacity issues with our ubuntu-xenial nodes. We have addressed the issue but will be another few hours before new images have been uploaded to all cloud providers. | 12:19 | |
*** jaypipes has joined #openstack-nova | 12:26 | |
*** lucasagomes is now known as lucas-hungry | 12:28 | |
*** gongysh has joined #openstack-nova | 12:29 | |
*** catintheroof has joined #openstack-nova | 12:31 | |
*** ayogi has joined #openstack-nova | 12:33 | |
*** takashin has left #openstack-nova | 12:39 | |
*** mvk has joined #openstack-nova | 12:39 | |
*** Jeffrey4l has joined #openstack-nova | 12:41 | |
*** Jeffrey4l has quit IRC | 12:42 | |
*** Jeffrey4l has joined #openstack-nova | 12:45 | |
*** ayogi has quit IRC | 12:45 | |
*** Jeffrey4l has quit IRC | 12:47 | |
*** kaisers has quit IRC | 12:47 | |
*** Jeffrey4l has joined #openstack-nova | 12:49 | |
*** gcb has quit IRC | 12:49 | |
*** gongysh has quit IRC | 12:50 | |
*** Jeffrey4l has quit IRC | 12:52 | |
*** Jeffrey4l has joined #openstack-nova | 12:52 | |
*** prashkre has quit IRC | 12:54 | |
*** yassine has joined #openstack-nova | 12:55 | |
*** yassine is now known as Guest13439 | 12:56 | |
*** shuquan_ has joined #openstack-nova | 12:58 | |
*** prashkre has joined #openstack-nova | 13:00 | |
*** kaisers has joined #openstack-nova | 13:03 | |
jaypipes | wznoinsk: morning sir! so... has the Intel NFV CI been flaky over the last five days or so, or is it just me? :) | 13:04 |
*** prashkre has quit IRC | 13:05 | |
*** avolkov has quit IRC | 13:09 | |
*** jaosorior is now known as jaosorior_brb | 13:09 | |
*** ekuris has quit IRC | 13:09 | |
jaypipes | moshele: morning sir! :) | 13:10 |
*** avolkov has joined #openstack-nova | 13:10 | |
jaypipes | moshele: not *quite* done with the nesteds resource providers patches. (3/4 done, working on last) | 13:10 |
jaypipes | moshele: I should have them pushed shortly for you to review. | 13:10 |
jaypipes | moshele: I removed the root_provider_uuid from the public REST API as we discussed on Friday. | 13:11 |
moshele | jaypipes: hi :) , ok cool | 13:11 |
jaypipes | moshele: and modified the REST API to have a GET /resource_providers?tree=<UUID> | 13:11 |
jaypipes | moshele: call that gets all provider records in the provider tree of the provider identified by <UUID> | 13:12 |
jaypipes | moshele: does that REST API format/query string parameter work for you? any suggestion to make it clearer? | 13:12 |
*** prateek has quit IRC | 13:14 | |
*** GB21 has quit IRC | 13:15 | |
*** matrohon has quit IRC | 13:15 | |
moshele | aypipes | 13:16 |
moshele | jaypipes: it fine by me | 13:16 |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Rewrite _test_finish_migration https://review.openstack.org/337159 | 13:16 |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Test disk creation in test_hard_reboot https://review.openstack.org/338993 | 13:16 |
*** matrohon has joined #openstack-nova | 13:16 | |
jaypipes | moshele: ok dokey :) | 13:17 |
*** edmondsw has joined #openstack-nova | 13:18 | |
*** diana_clarke has joined #openstack-nova | 13:22 | |
*** erlon-airlong has joined #openstack-nova | 13:22 | |
*** tbachman has joined #openstack-nova | 13:23 | |
*** lamt has joined #openstack-nova | 13:24 | |
*** amoralej is now known as amoralej|lunch | 13:24 | |
*** Jeffrey4l has quit IRC | 13:26 | |
*** pradk has joined #openstack-nova | 13:26 | |
*** mvk has quit IRC | 13:29 | |
*** Cristina__ has joined #openstack-nova | 13:29 | |
*** lucas-hungry is now known as lucasagomes | 13:29 | |
*** CristinaPauna has quit IRC | 13:32 | |
*** raildo has joined #openstack-nova | 13:35 | |
*** mriedem has joined #openstack-nova | 13:36 | |
*** thorst has joined #openstack-nova | 13:38 | |
*** alexpilotti has joined #openstack-nova | 13:38 | |
*** clenimar has joined #openstack-nova | 13:39 | |
openstackgerrit | sahid proposed openstack/nova: libvirt: Acquire TCP ports for console during live migration https://review.openstack.org/275801 | 13:41 |
openstackgerrit | sahid proposed openstack/nova: libvirt: do not return serial address if disabled on destination https://review.openstack.org/397276 | 13:41 |
openstackgerrit | sahid proposed openstack/nova: libvirt: clean serial listen addr to always deal with string https://review.openstack.org/398389 | 13:41 |
*** alexpilotti has quit IRC | 13:42 | |
edleafe | Scheduler subteam meeting in 15 minutes in #openstack-meeting-alt | 13:45 |
*** mdnadeem has quit IRC | 13:47 | |
openstackgerrit | Andrey Volkov proposed openstack/nova: PoC: add hierarchy quota driver https://review.openstack.org/391072 | 13:50 |
*** baoli has joined #openstack-nova | 13:52 | |
*** baoli has quit IRC | 13:54 | |
*** baoli has joined #openstack-nova | 13:54 | |
*** baoli has quit IRC | 13:57 | |
*** tlian has joined #openstack-nova | 13:57 | |
*** baoli has joined #openstack-nova | 13:57 | |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Test disk creation in test_hard_reboot https://review.openstack.org/338993 | 13:58 |
*** tbachman has quit IRC | 13:58 | |
edleafe | Scheduler subteam meeting running now in #openstack-meeting-alt | 14:00 |
*** alexpilotti has joined #openstack-nova | 14:01 | |
*** alexpilotti has quit IRC | 14:01 | |
*** alexpilotti has joined #openstack-nova | 14:01 | |
*** hongbin has joined #openstack-nova | 14:03 | |
*** bvanhav_ has joined #openstack-nova | 14:05 | |
openstackgerrit | Lee Yarwood proposed openstack/nova: WIP libvirt: Use the mirror ready flag if available https://review.openstack.org/399735 | 14:05 |
*** bvanhav has quit IRC | 14:05 | |
*** BlackDex has quit IRC | 14:08 | |
*** oanson has quit IRC | 14:08 | |
openstackgerrit | int32bit proposed openstack/nova: Add index on instances table across project_id and updated_at https://review.openstack.org/399672 | 14:10 |
*** shuquan_ has quit IRC | 14:11 | |
*** thorst has quit IRC | 14:11 | |
*** thorst has joined #openstack-nova | 14:12 | |
*** ekuris has joined #openstack-nova | 14:16 | |
*** sudipto_ has joined #openstack-nova | 14:17 | |
*** Guest56972 has joined #openstack-nova | 14:17 | |
*** mdrabe has joined #openstack-nova | 14:20 | |
*** thorst has quit IRC | 14:20 | |
*** tbachman has joined #openstack-nova | 14:21 | |
lyarwood | paul-carlton: afternoon, are you still looking to take https://review.openstack.org/#/c/389608 forward? | 14:21 |
*** BlackDex has joined #openstack-nova | 14:23 | |
*** sridharg has quit IRC | 14:24 | |
alex_xu | mriedem: jaypipes, morning, johnthetubaguy give a great roadmap how the server list query parameters is going at https://review.openstack.org/#/c/393205/, hope to get some feedback from you, thanks in advance | 14:24 |
*** mriedem has quit IRC | 14:27 | |
paul-carlton | lyarwood, yes, waiting on a manish to fix https://review.openstack.org/#/c/357131/ | 14:29 |
openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.trigger_crash_dump notification https://review.openstack.org/396225 | 14:29 |
paul-carlton | My testings of live migration shows it works with my patch but hard reboot does not even before live migration | 14:30 |
* lyarwood looks | 14:30 | |
*** belmorei_ has joined #openstack-nova | 14:32 | |
*** belmoreira has quit IRC | 14:33 | |
openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.trigger_crash_dump notification https://review.openstack.org/396225 | 14:33 |
jaypipes | alex_xu: done. | 14:35 |
alex_xu | jaypipes: thanks | 14:35 |
lyarwood | paul-carlton: yeah that doesn't strike me as the best way to fix this, we don't reconnect the volume so I don't think we should be reattaching the encryptors. | 14:37 |
*** amoralej|lunch is now known as amoralej | 14:37 | |
*** dikonoor has quit IRC | 14:37 | |
lyarwood | paul-carlton: shouldn't it just be a noop when we restart the domain? | 14:38 |
lyarwood | paul-carlton: actually the encryptor code just needs to move out of _create_domain_and_network to where ever it is that we connect the volumes in the first place | 14:39 |
*** novafire has joined #openstack-nova | 14:39 | |
*** nherciu has quit IRC | 14:41 | |
paul-carlton | lyarwood, agreed, please add your wisdom to that patch, Manish told me he was going to start looking at it last week | 14:42 |
paul-carlton | Once this is fixed, we can progress the live migration issue | 14:42 |
*** burt has joined #openstack-nova | 14:43 | |
*** TravT has joined #openstack-nova | 14:43 | |
*** superdan is now known as dansmith | 14:43 | |
*** mvk has joined #openstack-nova | 14:45 | |
*** beagles is now known as beagles|mtg | 14:46 | |
*** jaosorior_brb has quit IRC | 14:46 | |
markus_z | Is danpb in vacation, does anybody know? | 14:47 |
*** jaosorior_brb has joined #openstack-nova | 14:47 | |
*** TravT has quit IRC | 14:47 | |
*** anush has joined #openstack-nova | 14:47 | |
*** novafire99 has joined #openstack-nova | 14:47 | |
markus_z | @cores: mriedem's change https://review.openstack.org/#/c/395067/4 resolves the blocking issue in our s390x CI. Would be awesome to get that merged. | 14:48 |
*** gcb has joined #openstack-nova | 14:48 | |
johnthetubaguy | jaypipes: I am being dense when looking at the nested stuff (https://review.openstack.org/#/c/386710/) | 14:49 |
johnthetubaguy | jaypipes: when we query the placement API for NUMA resources, how do we select the correct resource provider (the host vs the NUMA node)? | 14:50 |
*** ociuhandu has joined #openstack-nova | 14:50 | |
johnthetubaguy | jaypipes: do we just not report for the host, or are NUMA node resources different, like PINNED_VCPU vs VCPU? | 14:50 |
jaypipes | johnthetubaguy: the root_provider_uuid attribute of the resource provider is the compute node. | 14:50 |
*** novafire has quit IRC | 14:50 | |
jaypipes | johnthetubaguy: yeah, we need different resource provider records for the NUMA nodes and SRIOV PFs in order to tag them with different qualitative attributes. | 14:51 |
jaypipes | johnthetubaguy: as well as differentiate certain quantitative resources (like SRIOV_NET_VF or VCPU or EMULATOR_THREAD_CPU | 14:51 |
johnthetubaguy | ah, so we ask for a NUMA node resource provider with some CPU, and it tells us which host resource provider can deliver that? | 14:52 |
jaypipes | johnthetubaguy: in other words, we need to say that NUMA node 0 should have emulator threads and not NUMA node 1. If we added an inventory record of EMULATOR_THREAD_CPU of some amount, say 4, to the *compute node resource provider*, there's be no way to indicate which NUMA node to pin the emaultor threads to. | 14:53 |
jaypipes | johnthetubaguy: yep, zactly. | 14:53 |
migi | jamielennox: hi | 14:54 |
johnthetubaguy | jaypipes: honestly, emulator threads are not really that special, as I understand it, for a two node VM, 2 vCPUs and 2vCPUs (i.e. 4 in total), we need 3 vCPUs on one NUMA node, and 2vCPUs on another to account for the emulator thread | 14:54 |
johnthetubaguy | that all seems to work just using vCPU resources I think | 14:54 |
migi | jamielennox: I have question regarding your patch, let me know when you will have few so I can ask: https://review.openstack.org/#/c/381498/ | 14:54 |
*** sandanar has quit IRC | 14:57 | |
jaypipes | johnthetubaguy: no, because for realtime workloads on libvirt/QEMU you need to ensure that the NUMA node that houses emulator threads does not have vCPU (qemu-kvm I/O threads) assigned to it. | 14:57 |
jaypipes | johnthetubaguy: so you need a separate resource class for representing the "physical CPU for emulator threads" versus the vCPU. | 14:58 |
johnthetubaguy | jaypipes: oh, so I understood the opposite, you want it on the same NUMA node, but a different pCPU :S | 14:58 |
johnthetubaguy | before we used to have a pCPU shared the emulator thread and vCPU, which is clearly a worry | 14:58 |
jaypipes | johnthetubaguy: no, from what I understand from sahid and danpb, you want the emulator threads pinned to a different NUMA node than is serving vCPU. | 14:58 |
jaypipes | johnthetubaguy: again, for realtime workloads... | 14:59 |
johnthetubaguy | jaypipes: hmm, that seems odd, given the whole shared memory structures used for I/O with the emulator threads | 14:59 |
johnthetubaguy | anyways, I get where you are coming from on that now | 14:59 |
dansmith | johnthetubaguy: yeah, agreed | 14:59 |
openstackgerrit | Stephen Finucane proposed openstack/nova: Fix backwards compatibility for InstanceNUMACell https://review.openstack.org/396184 | 15:00 |
*** mriedem has joined #openstack-nova | 15:00 | |
*** prateek has joined #openstack-nova | 15:00 | |
*** hongbin has quit IRC | 15:01 | |
*** ociuhandu has quit IRC | 15:01 | |
*** paul-carlton has quit IRC | 15:02 | |
*** Dong_Zhi has quit IRC | 15:02 | |
*** thorst has joined #openstack-nova | 15:04 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/nova: Revert "[PY3] byte/string conversions and enable PY3 test" https://review.openstack.org/400193 | 15:04 |
jaypipes | johnthetubaguy: answered your questions on that spec... | 15:04 |
jaypipes | dansmith, johnthetubaguy: from what danpb explained to me, libvirt/QEMU realtime workloads require that the workloads' I/O threads be on a NUMA node that is not used for any emulator threads, otherwise the emulator thread's CPU usage may interfere with the realtime scheduling constraints. sahid, am I saying that correctly? | 15:07 |
dansmith | jaypipes: seems counterintuitive to split the node to me, obviously we have to have a separate cpu.. anyway, I have no idea, just saying it seems counterintuitive | 15:08 |
*** ociuhandu has joined #openstack-nova | 15:09 | |
markus_z | dansmith: I don't quite get your concern in https://review.openstack.org/#/c/397276/3/nova/virt/libvirt/driver.py@5453 | 15:10 |
markus_z | Could you help me out here please? | 15:10 |
*** roeyc has quit IRC | 15:11 | |
*** mlavalle has joined #openstack-nova | 15:11 | |
dansmith | markus_z: because the field is nullable so that if we have no serial configured we can say so | 15:11 |
dansmith | markus_z: passing an empty string to signal that is kinda silly, right? | 15:11 |
*** paul-carlton has joined #openstack-nova | 15:11 | |
dansmith | I mean, we could use "true" and "" instead of booleans, but we don't :) | 15:11 |
*** gabor_antal has joined #openstack-nova | 15:12 | |
*** thorst has quit IRC | 15:12 | |
markus_z | dansmith: I have no strong opinion on '' vs None, to be honest | 15:12 |
markus_z | I can live with using None | 15:12 |
dansmith | well, I do :) | 15:12 |
dansmith | markus_z: tbh, I don't understand why we're making the change at all, since it's also trivial to check if it's unset: | 15:13 |
dansmith | markus_z: "if 'serial_listen_addr' in migate_data: do_thing()" | 15:13 |
openstackgerrit | Danil Akhmetov proposed openstack/nova: Use proper user and tenant in the owner section of libvirt.xml. https://review.openstack.org/399679 | 15:13 |
dansmith | maybe for older clients, which is fine | 15:14 |
markus_z | The value of "serial_listen_addr" is also used to update the domain XML, but yeah, I agree that we can do that. | 15:14 |
dansmith | but anyway, just passing '' is not cool, IMHO | 15:14 |
dansmith | markus_z: in fact, the code being removed that checks for the set-or-not-ness does exactly the same.. converts it to None if unset! | 15:16 |
*** rama_y has joined #openstack-nova | 15:16 | |
* jaypipes adds "set-or-not-ness" to personal dictionary | 15:16 | |
*** vladikr has joined #openstack-nova | 15:17 | |
*** ekuris has quit IRC | 15:18 | |
markus_z | dansmith: OK, I get that. What about the absence of the attribute "serial_listen_addr" in migration data? IIUC this should be checked? | 15:19 |
dansmith | markus_z: it's being checked right now, but this series is removing it | 15:19 |
dansmith | jaypipes: that one is free | 15:19 |
jaypipes | :) | 15:19 |
openstackgerrit | Danil Akhmetov proposed openstack/nova: Use proper user and tenant in the owner section of libvirt.xml. https://review.openstack.org/399679 | 15:20 |
markus_z | dansmith: According to sahid this attribute will always be available, do you disagree? | 15:20 |
dansmith | markus_z: I think it will from newer code but I'm expecting not from older code based on the stuff he's removing | 15:20 |
*** kevinz has quit IRC | 15:21 | |
dansmith | markus_z: I don't understand why we wouldn't keep that filtering function, check for is-set and pass through the none and call it good?\ | 15:21 |
markus_z | It won't hurt to check for the existence/absence of that attribute I guess, agreed. | 15:22 |
*** victorhe has joined #openstack-nova | 15:23 | |
sahid | jaypipes: we don't have the requierement to pin the emulator threads on a different numa node than the vCPU used for realtime is placed | 15:24 |
sahid | (i have well understood the whole discussion) | 15:24 |
markus_z | dansmith: Okay, summarized: keep None and the check for set-or-not-ness. I see no issue with that. | 15:25 |
dansmith | markus_z: yep | 15:25 |
*** eharney has joined #openstack-nova | 15:25 | |
*** oanson has joined #openstack-nova | 15:26 | |
*** bnemec has joined #openstack-nova | 15:26 | |
markus_z | dansmith: Thanks. If you could also have a look at https://review.openstack.org/#/c/373264/ , that would improve the situation with consoles and live-migration, too. | 15:26 |
jaypipes | sahid: we do *not* have that requirement? hmm, I must have badly misunderstood danpb then :( | 15:27 |
dansmith | markus_z: just dropped some supporting comments on those | 15:27 |
jaypipes | sahid: I thought cpu_realtime_policy=isolated was just that. | 15:27 |
markus_z | Thanks a lot. Would be great to get this situation fixed in Ocata. | 15:27 |
dansmith | jaypipes: cpu, not note, AFAIK | 15:28 |
dansmith | heh, node | 15:29 |
jaypipes | mdbooth: you working on rebasing https://review.openstack.org/#/c/337160/? | 15:29 |
*** ducttape_ has joined #openstack-nova | 15:29 | |
mdbooth | jaypipes: Yes | 15:29 |
mdbooth | Can't remember exactly where in the pipeline it is, but yes | 15:30 |
* mdbooth was away all last week | 15:30 | |
* mdbooth is responding to review comments from the first rebase tranche | 15:30 | |
mdbooth | Thanks for those, btw | 15:31 |
*** prateek has quit IRC | 15:31 | |
jaypipes | mdbooth: np | 15:32 |
sahid | jaypipes: yes, perhaps it's better to confirm with him but from my notes and understading we may want to make running emulator threads and RT on the same NUMA node (on different pCPUs) | 15:33 |
*** ekuris has joined #openstack-nova | 15:33 | |
*** ducttape_ has quit IRC | 15:35 | |
*** thorst has joined #openstack-nova | 15:35 | |
*** gcb has quit IRC | 15:37 | |
*** siva_krish has joined #openstack-nova | 15:39 | |
*** yamahata has joined #openstack-nova | 15:42 | |
*** ekuris has quit IRC | 15:42 | |
*** ekuris has joined #openstack-nova | 15:42 | |
*** harlowja_at_home has joined #openstack-nova | 15:48 | |
*** TravT has joined #openstack-nova | 15:50 | |
*** dimtruck is now known as zz_dimtruck | 15:52 | |
*** claudiub has joined #openstack-nova | 15:53 | |
*** unicell has quit IRC | 15:53 | |
*** unicell has joined #openstack-nova | 15:54 | |
*** nic has joined #openstack-nova | 15:54 | |
*** claudiub|4 has quit IRC | 15:55 | |
*** nic has quit IRC | 15:55 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Remove few TODO's from cells.py https://review.openstack.org/385140 | 15:58 |
*** Sukhdev has joined #openstack-nova | 15:58 | |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Cleanup test_create_configdrive https://review.openstack.org/339114 | 15:59 |
*** prashkre has joined #openstack-nova | 16:01 | |
*** WillUK2010 has quit IRC | 16:01 | |
*** eharney has quit IRC | 16:01 | |
jaypipes | dansmith, mriedem, melwitt, bauzas: would be great to get reviews on this series if you have some time: https://review.openstack.org/#/c/396963/. thx in advance. | 16:01 |
*** ljxiash has quit IRC | 16:02 | |
markus_z | oh man, all powers combined ;) | 16:03 |
markus_z | http://bit.ly/2gv5K5B | 16:03 |
jaypipes | mdbooth: so we have ddt as a dependency now? | 16:03 |
*** jdurgin1 has joined #openstack-nova | 16:03 | |
mdbooth | jaypipes: We agreed at the contributor meetup that ddt was ok, not worth a special effort. | 16:04 |
dansmith | *some* people agreed | 16:04 |
mdbooth | This looked like a good use, as we've got 1 test with multiple inputs. | 16:04 |
mdbooth | It's already in global requirements, anyway | 16:04 |
*** armax has joined #openstack-nova | 16:05 | |
jaypipes | mdbooth: ok, just wanted to double-check. I believe I left the room to talk with Toby Ford around the time this was being discussed. | 16:05 |
*** pcaruana has quit IRC | 16:06 | |
*** jwcroppe has joined #openstack-nova | 16:07 | |
jaypipes | mdbooth: what does ddt.unpack do again? | 16:07 |
mdbooth | jaypipes: Um... | 16:08 |
mdbooth | things | 16:08 |
mdbooth | very important things | 16:08 |
mdbooth | that must be done | 16:08 |
jaypipes | lol | 16:08 |
* mdbooth copied the recipe out of the examples :) | 16:08 | |
*** mjura has quit IRC | 16:08 | |
*** nic has joined #openstack-nova | 16:09 | |
openstackgerrit | int32bit proposed openstack/nova: Trivial fix of indentation with 8 spaces in libvirt conf https://review.openstack.org/396863 | 16:09 |
*** RuiChen has quit IRC | 16:10 | |
*** Guest93740 has quit IRC | 16:11 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 16:12 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: allow filter providers in tree https://review.openstack.org/377215 | 16:12 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: update client to set parent provider https://review.openstack.org/385693 | 16:12 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: add nested resource providers https://review.openstack.org/377138 | 16:12 |
jaypipes | moshele: ^^ | 16:12 |
*** zz_dimtruck is now known as dimtruck | 16:12 | |
moshele | jaypipes: will take a look | 16:13 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: priorities: Populate Ocata priorities https://review.openstack.org/400326 | 16:14 |
*** hongbin has joined #openstack-nova | 16:15 | |
*** eharney has joined #openstack-nova | 16:17 | |
siva_krish | Hi mdbooth! Trying to reproduce the following bug https://bugs.launchpad.net/nova/+bug/1605016 confirmed by you. Tried using the following setup https://assafmuller.com/2015/04/06/multinode-dvr-devstack/ still not able to replicate the scenario | 16:17 |
openstack | Launchpad bug 1605016 in OpenStack Compute (nova) "Post copy live migration interrupts network connectivity" [High,Confirmed] - Assigned to Sarafraj Singh (sarafraj-singh) | 16:17 |
*** unicell has quit IRC | 16:17 | |
*** moshele has quit IRC | 16:17 | |
*** unicell has joined #openstack-nova | 16:17 | |
mdbooth | siva_krish: How long is your instance staying in post-copy mode before completion? | 16:17 |
mdbooth | siva_krish: Unless you specifically push on it, this phase tends to be very fast. | 16:18 |
mdbooth | Or, in fact, you don't get into post-copy mode anyway | 16:18 |
mdbooth | So there's a chance that you are seeing the issue, but that it doesn't last very long. | 16:19 |
mdbooth | Which is why I ended up just frigging the code to artificially extend it. | 16:19 |
mdbooth | Anyway, for the above reasons I don't think that this is a huge problem in practise. | 16:19 |
EmilienM | mriedem: we found a critical bug when running nova in wsgi, see https://review.openstack.org/#/c/400193/ | 16:21 |
EmilienM | mriedem: sounds like devstack is not testing this case | 16:21 |
mdbooth | I didn't know Nova supported mod_wsgi | 16:22 |
mriedem | "supports" | 16:22 |
mriedem | it's not CI tested in the normal jobs as noted | 16:22 |
siva_krish | mdbooth I tried pinging the instance but there was no drop of packets and tried running stress command that dirtes the memory before doing migrate as well | 16:22 |
mriedem | EmilienM: can just the wsgi module be fixed instead of the entire revert? | 16:23 |
mdbooth | siva_krish: Yeah, I suspect you're hitting it but not noticing it. | 16:23 |
EmilienM | mriedem: that's what I'm trying to figure with the revert's author | 16:23 |
mriedem | EmilienM: seems a simple six.py2 conditional or something could work | 16:24 |
*** belmorei_ has quit IRC | 16:24 | |
mdbooth | siva_krish: Did I document some code changes in that bug? If so, try making those. | 16:24 |
siva_krish | mdbooth but I do see the few pings take more time to respond which I believe happens with live migration usually | 16:24 |
mdbooth | They will artificially add a pause before completing the migration in post-copy mode. | 16:24 |
mdbooth | post-copy is really fast, which is why it's a good thing. | 16:24 |
*** liangy_ has joined #openstack-nova | 16:25 | |
mdbooth | How big is your instance? | 16:25 |
siva_krish | mdbooth mine is m1.small - 1CPU and 2 GB RAM | 16:27 |
siva_krish | Should I try using bigger flavor | 16:27 |
siva_krish | ? | 16:27 |
edleafe | mriedem: in the scheduler meeting this morning we noted that there are 3 specs that didn't make it by the freeze. There is no disagreement on any of them. What incantation do we need to chant to get exceptions? | 16:27 |
*** jgriffith has quit IRC | 16:28 | |
edleafe | They are: | 16:28 |
mriedem | goats blood at midnight | 16:28 |
edleafe | Scheduler filters in DB: https://review.openstack.org/#/c/300178/ | 16:28 |
edleafe | Custom Resource Classes: https://review.openstack.org/#/c/312696/ | 16:28 |
edleafe | Nested Resource Providers: https://review.openstack.org/#/c/386710/ | 16:28 |
mriedem | the 2nd is just process, as the code is already being worked | 16:28 |
mriedem | i'll take a look at that one at some point here | 16:28 |
mriedem | i'm mostly heads down on internal things at | 16:28 |
mriedem | *atm | 16:28 |
edleafe | ok, thanks | 16:29 |
edleafe | understand that internal thing :) | 16:29 |
mriedem | johnthetubaguy: i'm going to tell you right now that i'm a terrible person and haven't gone through the latest updates on https://review.openstack.org/#/c/373203/ yet | 16:30 |
*** penick has joined #openstack-nova | 16:31 | |
*** dave-johnston has joined #openstack-nova | 16:32 | |
*** bvanhav_ has quit IRC | 16:33 | |
*** markus_z has quit IRC | 16:33 | |
*** andreas_s has quit IRC | 16:35 | |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Rename Backend snapshot and image https://review.openstack.org/333272 | 16:39 |
*** Jizhaoxuan_ has joined #openstack-nova | 16:40 | |
*** weshay has joined #openstack-nova | 16:41 | |
*** liuyulong_ has quit IRC | 16:41 | |
*** Jizhaoxuan_ has quit IRC | 16:42 | |
*** jaosorior_brb is now known as jaosorior | 16:42 | |
openstackgerrit | Matthew Edmonds proposed openstack/nova: Separate CRUD policy for server_groups https://review.openstack.org/391113 | 16:43 |
*** AAR has joined #openstack-nova | 16:44 | |
*** paul-carlton has quit IRC | 16:44 | |
AAR | hi, could anyone help me in pinging between 2 VMs? | 16:45 |
*** burt has quit IRC | 16:45 | |
AAR | I am not able to ping them | 16:45 |
*** AAR has quit IRC | 16:46 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/nova: Use byte string or utf8 depending on python version for wsgi https://review.openstack.org/400336 | 16:46 |
*** AAR has joined #openstack-nova | 16:46 | |
*** mtanino has joined #openstack-nova | 16:46 | |
*** AAR has quit IRC | 16:47 | |
openstackgerrit | Matthew Booth proposed openstack/nova: libvirt: Rename Backend snapshot and image https://review.openstack.org/333272 | 16:47 |
*** nkrinner is now known as nkrinner_afk | 16:49 | |
*** tojuvone has quit IRC | 16:50 | |
siva_krish | mdbooth: Is it possible for you to share with me DVR setup that you had for reproducing the bug ? | 16:52 |
*** thorst has quit IRC | 16:53 | |
*** thorst has joined #openstack-nova | 16:54 | |
*** bvanhav_ has joined #openstack-nova | 16:55 | |
*** josecastroleon has quit IRC | 16:56 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Remove few TODO's from cells.py https://review.openstack.org/385140 | 16:57 |
*** mnaser has quit IRC | 16:57 | |
*** Zer0Byte__ has joined #openstack-nova | 16:58 | |
mdbooth | siva_krish: I don't still have it. It was stock, though. | 16:58 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/nova: Use byte string or utf8 depending on python version for wsgi https://review.openstack.org/400336 | 16:59 |
*** Sukhdev has quit IRC | 16:59 | |
*** thorst has quit IRC | 17:00 | |
*** harlowja_at_home has quit IRC | 17:00 | |
*** xyang1 has joined #openstack-nova | 17:01 | |
*** mnaser has joined #openstack-nova | 17:01 | |
*** sahid has quit IRC | 17:01 | |
*** penick has quit IRC | 17:01 | |
siva_krish | tdurakov: is there anyway to abort live-migration apart from using 'nova live-migration-abort' ? | 17:02 |
mdbooth | siva_krish: You mean any supported way? | 17:03 |
mdbooth | If you're looking for any way, you could interfere with the guest on the destination prior to the switch to post-copy | 17:03 |
mdbooth | That obviously requires root access on the host, though | 17:03 |
*** GB21 has joined #openstack-nova | 17:04 | |
*** penick has joined #openstack-nova | 17:05 | |
openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 17:06 |
*** rmstar has quit IRC | 17:06 | |
*** liverpooler has quit IRC | 17:07 | |
*** liverpooler has joined #openstack-nova | 17:07 | |
siva_krish | mdbooth: I was looking any API calls within nova to do it. I just came across this scenario while doing live-migration https://bugs.launchpad.net/nova/+bug/1643623. Just wanted to make sure I haven't missed any possibilities | 17:07 |
openstack | Launchpad bug 1643623 in OpenStack Compute (nova) "Not able to abort live migration" [Undecided,New] - Assigned to Sivasathurappan Radhakrishnan (siva-radhakrishnan) | 17:07 |
*** BobBall_AWOL has quit IRC | 17:09 | |
mdbooth | siva_krish: That bug has nothing to do with abort | 17:09 |
mdbooth | It's simply asserting that abort doesn't fix the problem | 17:10 |
mdbooth | But as there's no migration in progress, you wouldn't expect it to | 17:10 |
mdbooth | The bug is that it gets into a wedged state when an admin specifies an invalid destination | 17:10 |
mdbooth | It's possible we may also want an additional admin api for forcibly cancelling wedged migrations, similar to reset-state | 17:11 |
mdbooth | But really you just want to fix the necessity for that | 17:11 |
openstackgerrit | sean mooney proposed openstack/os-vif: remove use of contextlib and with nested https://review.openstack.org/382725 | 17:13 |
siva_krish | mdbooth yeah you are right since it can't migrate at all it is not related to abort but to end user it would look like it is stuck in migrating state. Should we need an admin api to do that else can't we validate it with available hosts even before migrating as we do for instances ? | 17:14 |
mdbooth | siva_krish: Does a similar bug exist for cold migrations? | 17:16 |
mdbooth | If not, how is it handled there? | 17:16 |
*** oanson has quit IRC | 17:18 | |
siva_krish | mdbooth: I haven't checked it with cold migrate. Will go and check it out | 17:19 |
*** mhickey has quit IRC | 17:20 | |
*** rmstar has joined #openstack-nova | 17:20 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 17:20 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: update client to set parent provider https://review.openstack.org/385693 | 17:20 |
*** matrohon has quit IRC | 17:21 | |
*** lpetrut has quit IRC | 17:21 | |
openstackgerrit | Sujitha proposed openstack/nova: Improve help text for libvirt options https://review.openstack.org/372722 | 17:24 |
*** Guest13439 has quit IRC | 17:26 | |
*** tjones has joined #openstack-nova | 17:27 | |
wznoinsk | jaypipes, yes, guilty, we went thru a couple of issues with Intel NFV CI that accidentally started to happen around maintenance tasks (both things were impacting the results back time), it's fine as of last night (GMT) so please recheck | 17:32 |
*** TravT has quit IRC | 17:33 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: implement ResourceClass.create() https://review.openstack.org/386844 | 17:35 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: REST API for resource classes https://review.openstack.org/391959 | 17:35 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.get_by_name() https://review.openstack.org/391917 | 17:35 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: implement destroy() resource class https://review.openstack.org/391918 | 17:35 |
jaypipes | wznoinsk: coolio, thanks for the update! :) | 17:35 |
jaypipes | wznoinsk: remember... when all else fails, blame sean-k-mooney ;P | 17:36 |
wznoinsk | you know... it's not actually that funny when it's actually true... joking :-) | 17:36 |
sean-k-mooney | what did i do now :)# | 17:37 |
*** baoli has quit IRC | 17:39 | |
*** derekh has quit IRC | 17:42 | |
jaypipes | mriedem: good day, sir. mind if I bother you about the remaining nova-specs around resource providers that we're aiming for in Ocata? they are scheduler integration with placement (https://review.openstack.org/300178) custom resource classes (https://review.openstack.org/312696) and nested resource providers (https://review.openstack.org/386710). We didn't want to bother you last week because we | 17:43 |
jaypipes | knew you were focused more on the non-priority things. | 17:43 |
*** anush has quit IRC | 17:44 | |
openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.create.error notification https://review.openstack.org/397715 | 17:44 |
*** ihrachys has quit IRC | 17:47 | |
edleafe | jaypipes: already bugged him :) | 17:51 |
edleafe | jaypipes: we need goat's blood to get an exception | 17:51 |
*** markvoelker_ has joined #openstack-nova | 17:52 | |
*** sudipto_ has quit IRC | 17:52 | |
*** Guest56972 has quit IRC | 17:52 | |
*** nic has quit IRC | 17:52 | |
*** nic has joined #openstack-nova | 17:53 | |
*** markvoelker has quit IRC | 17:53 | |
openstackgerrit | int32bit proposed openstack/nova: Add index on instances table across project_id and updated_at https://review.openstack.org/399672 | 17:54 |
*** markvoelker has joined #openstack-nova | 17:55 | |
*** markvoelker_ has quit IRC | 17:57 | |
*** penick has quit IRC | 17:57 | |
*** GB21 has quit IRC | 17:57 | |
*** unicell has quit IRC | 17:57 | |
jaypipes | edleafe: do you have any goats handy? | 17:58 |
edleafe | jaypipes: No, but some of my neighbors have roosters. Maybe that will satisfy mriedem. | 17:59 |
*** eharney has quit IRC | 17:59 | |
*** beagles|mtg is now known as beagles | 18:00 | |
*** ralonsoh has quit IRC | 18:01 | |
*** penick has joined #openstack-nova | 18:02 | |
*** prashkre_ has joined #openstack-nova | 18:05 | |
*** prashkre has quit IRC | 18:05 | |
*** lucasagomes is now known as lucas-afk | 18:07 | |
*** prashkre_ has quit IRC | 18:09 | |
*** prashkre_ has joined #openstack-nova | 18:09 | |
*** yamahata has quit IRC | 18:11 | |
*** bvanhav_ has quit IRC | 18:11 | |
openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.add_fixed_ip notification https://review.openstack.org/332876 | 18:13 |
*** harlowja has joined #openstack-nova | 18:14 | |
*** burt has joined #openstack-nova | 18:14 | |
cfriesen | is it possible to change the availability zone of an instance post-boot? I'm envisioning someone that wants to split an existing availability zone. | 18:14 |
*** ekuris has quit IRC | 18:16 | |
jaypipes | cfriesen: hmm, not easily. you'd need to change the AZ of the compute node the instance is on, then manually set the instance metadata key for availability zone to the new one and even then I might be missing something that would need to happen. | 18:16 |
*** jgriffith has joined #openstack-nova | 18:16 | |
*** jgriffith has quit IRC | 18:16 | |
jaypipes | cfriesen: and of course, you'd need to do that for all instances on that host. | 18:16 |
*** prashkre_ has quit IRC | 18:17 | |
mriedem | jaypipes: comments inline https://review.openstack.org/#/c/312696/ | 18:17 |
*** prashkre_ has joined #openstack-nova | 18:17 | |
mriedem | jaypipes: let me know if you plan to respin based on those, i'd just +W if my comments are addressed rather than wait for another round of +2s | 18:18 |
*** jdurgin1 has quit IRC | 18:21 | |
jlk | Couple questions around snapshots. If I take a snapshot of a running instance, and that instance is a boot from volume instance, does it make sense that the resulting snapshot image in glance is size 0? | 18:24 |
jlk | Does it even make sense to take a snapshot in this way of an instance that's boot from volume? | 18:24 |
*** jaosorior has quit IRC | 18:24 | |
*** Swami has joined #openstack-nova | 18:24 | |
openstackgerrit | Jay Pipes proposed openstack/nova-specs: Resource providers: custom resource classes https://review.openstack.org/312696 | 18:26 |
jaypipes | mriedem, edleafe, johnthetubaguy: addressed your review comments ^ | 18:26 |
jaypipes | thx | 18:27 |
*** Apoorva has joined #openstack-nova | 18:27 | |
*** lpetrut has joined #openstack-nova | 18:27 | |
*** bvanhav_ has joined #openstack-nova | 18:29 | |
*** unicell has joined #openstack-nova | 18:29 | |
openstackgerrit | int32bit proposed openstack/nova: Add index on instances table across project_id and updated_at https://review.openstack.org/399672 | 18:31 |
*** mtanino has quit IRC | 18:32 | |
*** nherciu has joined #openstack-nova | 18:32 | |
*** tbachman has quit IRC | 18:33 | |
*** tbachman has joined #openstack-nova | 18:34 | |
jlk | oh, weird. Doing a snapshot of a boot from volume appears to 1) make a snapshot volume, and 2) make a 0 sized image. | 18:36 |
cfriesen | jaypipes: thanks, I was thinking it'd be something like that. | 18:37 |
*** sambetts is now known as sambetts|afk | 18:37 | |
*** prashkre_ has quit IRC | 18:37 | |
openstackgerrit | Ed Leafe proposed openstack/nova-specs: Resource providers: custom resource classes https://review.openstack.org/312696 | 18:41 |
edleafe | jaypipes: fixed the red square of death for you ^^ | 18:41 |
*** swamireddy has joined #openstack-nova | 18:42 | |
*** mhickey has joined #openstack-nova | 18:43 | |
*** amoralej is now known as amoralej|off | 18:44 | |
jaypipes | edleafe: cheers :) | 18:45 |
openstackgerrit | Kaitlin Farr proposed openstack/nova-specs: Add addendum to Castellan integration specification https://review.openstack.org/338518 | 18:46 |
openstackgerrit | Gábor Antal proposed openstack/nova: Remove mox in unit/cells/test_cells_manager.py https://review.openstack.org/399036 | 18:46 |
*** knasim-wrs has joined #openstack-nova | 18:50 | |
*** belmoreira has joined #openstack-nova | 18:51 | |
*** jamielennox is now known as jamielennox|away | 18:52 | |
*** lpetrut has quit IRC | 18:52 | |
*** gszasz has quit IRC | 18:54 | |
*** jgriffith has joined #openstack-nova | 18:55 | |
*** lpetrut has joined #openstack-nova | 18:57 | |
*** jamielennox|away is now known as jamielennox | 19:00 | |
*** yamahata has joined #openstack-nova | 19:03 | |
*** rama_y has quit IRC | 19:06 | |
*** eharney has joined #openstack-nova | 19:06 | |
*** eharney has quit IRC | 19:06 | |
*** gouthamr has quit IRC | 19:07 | |
*** mtanino has joined #openstack-nova | 19:07 | |
*** jamesdenton has joined #openstack-nova | 19:11 | |
*** Sukhdev has joined #openstack-nova | 19:13 | |
*** stream10 has joined #openstack-nova | 19:13 | |
*** jamesden_ has joined #openstack-nova | 19:15 | |
*** jamesdenton has quit IRC | 19:15 | |
*** anush has joined #openstack-nova | 19:20 | |
*** diogogmt has joined #openstack-nova | 19:26 | |
alaski | jlk: I believe the intent is that the glance metadata for that 0 size image contains info about the volume that should be attached. However I'm not sure that works in practice | 19:28 |
*** novafire99 has quit IRC | 19:28 | |
jlk | yeah, it looks like when I go to boot from the (volume) snapshot, no image is involved. | 19:28 |
jlk | but maybe something behind the scenes is referencing the image? It just doesn't look like it from the surface. | 19:29 |
alaski | jlk: only if you reference it in the boot command. I expect it was trying to setup the workflow of taking an instance snapshot and then booting from the resulting image, which would actually use a volume behind the scenes | 19:30 |
alaski | this is just my guess from seeing this in the past | 19:30 |
jlk | oh let me see what happens if I select booting from that image. | 19:31 |
openstackgerrit | Lee Yarwood proposed openstack/nova: WIP libvirt: Reconnect volumes and encryptors during a hard reboot https://review.openstack.org/400384 | 19:31 |
alaski | My guess is that it doesn't work, because it's not tested. But I believe that's what someone thought should happen at some point | 19:31 |
jlk | Well that seems to result in a horizon error :) | 19:32 |
jlk | ah, in Mitaka. In Newton via horizon it seems to automatically adjust things so that a volume is created from the image. | 19:33 |
mriedem | the volume snapshot meta should be stored in hte image | 19:33 |
mriedem | there is a test for this in tempest i think | 19:33 |
mriedem | it's wonky | 19:33 |
alaski | mriedem: oh, nice | 19:33 |
jlk | yeah, much confuse | 19:33 |
mriedem | this is the test https://github.com/openstack/tempest/blob/master/tempest/scenario/test_volume_boot_pattern.py#L166 | 19:33 |
mriedem | so yeah there is bdm meta in the image snapshot from the volue | 19:34 |
mriedem | *bolume | 19:34 |
mriedem | damn | 19:34 |
mriedem | volume | 19:34 |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2682 is the 0 size | 19:35 |
jlk | but that fails in newton | 19:35 |
jlk | "Volume is smaller than the minimum size specified in image metadata. Volume size is 1073741824 bytes, minimum size is 10737418240 bytes. (HTTP 400) (Request-ID: req-e0fefaa3-3828-472c-91ea-2c6ea1405b0f)" | 19:35 |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2741 | 19:35 |
jlk | the image says it's 10G, the volume is set to 11 G | 19:35 |
jlk | Looks like something is counting off by a few. | 19:36 |
jlk | something is adding an extra 0 where it shouldn't be. | 19:36 |
*** belmoreira has quit IRC | 19:38 | |
mriedem | this is where the compute api pulls the snapshot volume bdms out of the image meta https://github.com/openstack/nova/blob/master/nova/compute/api.py#L679 | 19:38 |
mriedem | on server create | 19:38 |
mriedem | i only know most of this because test_create_ebs_image_and_check_boot used to trigger a race fail in the gate which gave me ulcers | 19:39 |
jlk | lol | 19:39 |
jlk | mriedem: I think something is wrong in this. I'm getting an error out of Nova, and it appears to be off by an order of magnitude. | 19:43 |
jlk | the image min_disk is listed at "10" | 19:44 |
jlk | image_min_disk is image.get(min_disk) which is 10, * units.Gi | 19:45 |
*** rama_y has joined #openstack-nova | 19:46 | |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L622 ? | 19:47 |
jlk | right, I'm trying to figure out if it's that math that is wrong, or the volume size later on | 19:47 |
mriedem | ^ is master mind you | 19:47 |
jlk | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L655 | 19:47 |
jlk | 10737418240 is being reported for the min size | 19:48 |
jlk | 1073741824 as the volume size | 19:48 |
jlk | looks like volume size | 19:49 |
*** moshele has joined #openstack-nova | 19:49 | |
jlk | maybe this is a bug in horizon that it's not updating the size of the volume it's going to request | 19:50 |
mriedem | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1205 ? | 19:50 |
mriedem | no math in there | 19:51 |
*** dimtruck is now known as zz_dimtruck | 19:51 | |
mriedem | idk, would have to have a reproduce scenario on the command line so it could be tested against mitaka and master | 19:52 |
mriedem | and i'm off to a meeting | 19:52 |
*** baoli has joined #openstack-nova | 19:52 | |
jlk | Here is a debug http://paste.openstack.org/show/589949/ | 19:53 |
jlk | volume size is 15 in the initial request. | 19:53 |
jlk | but the error message is still acting as if the volume size is 1G not 15 | 19:54 |
*** eharney has joined #openstack-nova | 19:55 | |
*** josecastroleon has joined #openstack-nova | 19:56 | |
*** zz_dimtruck is now known as dimtruck | 19:57 | |
jlk | mriedem: okay, here's a short reproducer http://paste.openstack.org/show/589950/ | 19:59 |
jlk | mriedem: and I think the main issue is that the request is looking at this special image, seeing in it the block device mapping which has '"volume_size": 1', and using that as the source information for what the instance will boot up. | 20:00 |
jlk | which may or may not be related to an attached boot from volume request. | 20:00 |
*** yamahata has quit IRC | 20:01 | |
*** josecastroleon has quit IRC | 20:01 | |
*** bvanhav_ has quit IRC | 20:01 | |
*** kfarr has joined #openstack-nova | 20:02 | |
*** yamahata has joined #openstack-nova | 20:02 | |
*** swamireddy has quit IRC | 20:05 | |
*** dimtruck is now known as zz_dimtruck | 20:06 | |
*** hwoarang has quit IRC | 20:09 | |
*** hwoarang has joined #openstack-nova | 20:10 | |
*** hwoarang has quit IRC | 20:10 | |
*** hwoarang has joined #openstack-nova | 20:10 | |
*** belmoreira has joined #openstack-nova | 20:16 | |
*** slaweq_ has joined #openstack-nova | 20:17 | |
*** bvanhav_ has joined #openstack-nova | 20:18 | |
EmilienM | mriedem: Juan proposed a fix when running nova in wsgi, can you please add it in your list of reviews? https://review.openstack.org/#/c/400336/ | 20:20 |
EmilienM | the jobs are all passing (except one still running) | 20:21 |
*** swamireddy has joined #openstack-nova | 20:21 | |
*** liusheng has quit IRC | 20:24 | |
*** Swami has quit IRC | 20:24 | |
*** belmoreira has quit IRC | 20:24 | |
*** brault has quit IRC | 20:25 | |
*** tjones has quit IRC | 20:26 | |
*** oanson has joined #openstack-nova | 20:27 | |
*** zz_dimtruck is now known as dimtruck | 20:27 | |
*** knasim-wrs has quit IRC | 20:27 | |
*** liusheng has joined #openstack-nova | 20:31 | |
jlk | mriedem: I filed https://bugs.launchpad.net/nova/+bug/1643684 to keep track of this better. | 20:32 |
openstack | Launchpad bug 1643684 in OpenStack Compute (nova) "_check_requested_image uses incorrect bdm info to validate target volume when booting from boot from volume instance snapshot image" [Undecided,New] | 20:32 |
*** Zer0Byte__ has quit IRC | 20:35 | |
*** rfolco has quit IRC | 20:38 | |
*** Swami has joined #openstack-nova | 20:40 | |
*** bvanhav_ has quit IRC | 20:42 | |
*** WillUK2010 has joined #openstack-nova | 20:42 | |
*** alexpilotti has quit IRC | 20:44 | |
*** nherciu has quit IRC | 20:44 | |
*** WillUK2010 has quit IRC | 20:44 | |
*** WillUK2010 has joined #openstack-nova | 20:45 | |
dansmith | EmilienM: what patch made the change? | 20:46 |
dansmith | EmilienM: normally it would be good to have that in the commit message so we can examine the intent there | 20:46 |
dansmith | last change to that file is 0f152eb66153e16d4d8aba0f8afebaab046aea35 which is old | 20:47 |
EmilienM | dansmith: https://review.openstack.org/#/c/333212/ | 20:47 |
EmilienM | 3 days ago I think | 20:47 |
*** WillUK2010 has quit IRC | 20:48 | |
dansmith | ah, I saw the original date | 20:48 |
dansmith | oomichi: around? | 20:48 |
dansmith | EmilienM: do you know that the py3 code is right? I don't know what specifies "headers should be {bytes|strings}" | 20:50 |
dansmith | EmilienM: also, "utf-8 strings" is not really a thing, right? it's either a "unicode string" or "bytes, encoding a unicode string with utf-8" ... ? | 20:51 |
*** oanson has quit IRC | 20:51 | |
EmilienM | dansmith: right, I think Juan re-used what was written before | 20:52 |
dansmith | EmilienM: yeah I know, but I'm not sure what is really expected | 20:52 |
*** slaweq_ has quit IRC | 20:52 | |
*** slaweq_ has joined #openstack-nova | 20:54 | |
EmilienM | dansmith: Juan was pointing to this code in the bug report https://github.com/GrahamDumpleton/mod_wsgi/blob/develop/src/server/wsgi_convert.c#L27 | 20:57 |
*** lpetrut has quit IRC | 21:01 | |
*** moshele has quit IRC | 21:02 | |
*** ijw has joined #openstack-nova | 21:02 | |
*** anush has quit IRC | 21:04 | |
*** tjones has joined #openstack-nova | 21:05 | |
*** bvanhav_ has joined #openstack-nova | 21:09 | |
mriedem | EmilienM: we have tripleo in the nova experimental queue now right? can't we run that on https://review.openstack.org/#/c/400336/ ? | 21:11 |
*** jamesden_ has quit IRC | 21:11 | |
EmilienM | mriedem: this patch is one of a long list of problems we are currently dealing with | 21:12 |
EmilienM | mriedem: i come back from PTO I don't have all details yet but iirc there is a lot of blockers atm | 21:12 |
dansmith | mriedem: I expect it's probably fine, but I don't really know | 21:12 |
dansmith | mriedem: so I'm okay with it if you are I guess, but I wish oomichi was around as he may know more | 21:13 |
mriedem | EmilienM: ok so even if we run the job it's likely to fail you're saying | 21:13 |
EmilienM | I'm curious to know why we don't test wsgi setup in the gate | 21:13 |
*** penick has quit IRC | 21:13 | |
mriedem | EmilienM: patches welcome? | 21:13 |
EmilienM | mriedem: i'm happy to work on it, I wasn't sure if this thing would be welcome :) | 21:13 |
mriedem | EmilienM: honestly i have a hard time keeping tabs on all of the snowflakes that we support but don't test | 21:14 |
mriedem | we try to get a decent representation of features in CI | 21:14 |
mriedem | it could be added to the placement job that only runs on nova | 21:14 |
EmilienM | mriedem: no worries, /me just trying to help | 21:14 |
mriedem | and is kind of our dumping ground for special nova-only things | 21:14 |
EmilienM | mriedem: that is one of the reasons why I think having jobs like tripleo or other installers is useful, it tests some "snowflakes" | 21:15 |
EmilienM | I had to google it btw | 21:15 |
*** penick has joined #openstack-nova | 21:16 | |
mriedem | dansmith: +2 | 21:17 |
*** bvanhav_ has quit IRC | 21:18 | |
EmilienM | thanks for your help folks | 21:20 |
*** alexpilotti has joined #openstack-nova | 21:21 | |
*** jwcroppe has quit IRC | 21:22 | |
*** jwcroppe has joined #openstack-nova | 21:23 | |
*** dave-mccowan has quit IRC | 21:24 | |
openstackgerrit | Ed Leafe proposed openstack/nova: placement: allow filter providers in tree https://review.openstack.org/377215 | 21:24 |
*** alexpilotti has quit IRC | 21:26 | |
*** jwcroppe has quit IRC | 21:27 | |
openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove 409 from "delete floating-ip" https://review.openstack.org/400413 | 21:28 |
*** swamireddy has quit IRC | 21:30 | |
*** WillUK2010 has joined #openstack-nova | 21:32 | |
*** stream10 has quit IRC | 21:36 | |
*** jamesdenton has joined #openstack-nova | 21:44 | |
openstackgerrit | jichenjc proposed openstack/nova: remove additional param https://review.openstack.org/400416 | 21:47 |
*** dimtruck is now known as zz_dimtruck | 21:48 | |
*** slaweq_ has quit IRC | 21:53 | |
*** fragatina has quit IRC | 21:55 | |
*** slaweq_ has joined #openstack-nova | 21:55 | |
*** jwcroppe has joined #openstack-nova | 21:56 | |
*** jwcroppe_ has joined #openstack-nova | 21:57 | |
*** zz_dimtruck is now known as dimtruck | 21:57 | |
*** jwcroppe has quit IRC | 22:00 | |
*** kfarr has quit IRC | 22:00 | |
*** jwcroppe_ has quit IRC | 22:01 | |
*** lpetrut has joined #openstack-nova | 22:05 | |
*** jamesdenton has quit IRC | 22:06 | |
*** WillUK2010 has quit IRC | 22:11 | |
*** pilgrimstack has joined #openstack-nova | 22:18 | |
*** eliqiao has quit IRC | 22:18 | |
*** swamireddy has joined #openstack-nova | 22:18 | |
*** fragatina has joined #openstack-nova | 22:18 | |
mriedem | jaypipes: looks like you rebased something here and broke it https://review.openstack.org/#/c/386844/ | 22:20 |
*** eliqiao has joined #openstack-nova | 22:20 | |
mriedem | probably dropped ed's changes from last week | 22:21 |
mriedem | i've got a pretty long but easy series to review here https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/use-neutron-by-default | 22:25 |
mriedem | working on fixing up unit tests to work with CONF.use_neutron=True | 22:25 |
mriedem | for the eventual switch | 22:25 |
mriedem | starts at https://review.openstack.org/#/c/396359/ | 22:25 |
*** rtheis has quit IRC | 22:26 | |
*** WillUK2010 has joined #openstack-nova | 22:26 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: handle os-brick InvalidConnectorProtocol on init https://review.openstack.org/395067 | 22:27 |
*** edleafe- has joined #openstack-nova | 22:33 | |
*** pilgrimstack has quit IRC | 22:33 | |
*** edleafe has quit IRC | 22:34 | |
*** edmondsw has quit IRC | 22:34 | |
mriedem | dansmith: melwitt: what do you think about just dropping this fallback to lookup rbd creds in nova.conf when using volumes and they aren't in the cinder connection dict https://review.openstack.org/#/c/389399/3/nova/virt/libvirt/volume/net.py | 22:35 |
mriedem | today we use what's in nova.conf if not configured in cinder but that's pretty busted i think | 22:36 |
mriedem | b/c they could be different backing stores | 22:36 |
mriedem | so we could log a warning for the fallback and drop it in pike, or just rip it out now | 22:36 |
nic | I've always been of the mind that the current behaviour only works "by accident" | 22:41 |
mriedem | it looks that way yeah | 22:43 |
*** eharney has quit IRC | 22:45 | |
*** lpetrut has quit IRC | 22:46 | |
jaypipes | mriedem: yeah, I'm workin' on it... | 22:49 |
openstackgerrit | STEW TY proposed openstack/nova: Complete verification for os-floating-ips https://review.openstack.org/329704 | 22:50 |
*** victorhe_ has joined #openstack-nova | 22:53 | |
*** victorhe has quit IRC | 22:56 | |
*** victorhe_ has quit IRC | 22:56 | |
*** victorhe has joined #openstack-nova | 22:57 | |
*** victorhe has quit IRC | 22:59 | |
*** victorhe_ has joined #openstack-nova | 22:59 | |
*** jwcroppe has joined #openstack-nova | 23:03 | |
*** victorhe_ has quit IRC | 23:04 | |
*** tbachman has quit IRC | 23:05 | |
*** burt has quit IRC | 23:08 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: cleanup network volume driver auth config https://review.openstack.org/389389 | 23:08 |
openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: prefer cinder rbd auth values over nova.conf https://review.openstack.org/389399 | 23:08 |
mriedem | nic: cburgess: i went with the straddle approach for ocata, remove in pike | 23:08 |
*** lamt has quit IRC | 23:08 | |
cburgess | mriedem Yeah I can get behind that. | 23:09 |
mriedem | i asked myself what mc hammer would do | 23:10 |
mriedem | and he said don't hurt'em | 23:10 |
mriedem | i model most of life's decisions after mc hammer, including finances and dress | 23:10 |
cburgess | mriedem LOL | 23:10 |
cburgess | mriedem This seems... unwise. | 23:11 |
mriedem | hey, have you ever had your own breakfast cereal?! | 23:11 |
cburgess | no, do you? | 23:11 |
mriedem | i might be thinking of urkel | 23:11 |
mriedem | yup http://media.tumblr.com/tumblr_lqcy4ogCrm1qkzq2g.jpg | 23:11 |
mriedem | mc hammer had his own cartoon, it was urkel that had a cereal | 23:12 |
mriedem | on that note, i'm going to shuffle dance on out of here | 23:12 |
*** mriedem has quit IRC | 23:12 | |
*** lamt has joined #openstack-nova | 23:13 | |
WillUK2010 | anyone know whats causing this, Host openstack has more disk space than database expected (196 GB > 134 GB) seeing it in my nova-scheduler logs | 23:23 |
*** baoli_ has joined #openstack-nova | 23:23 | |
openstackgerrit | STEW TY proposed openstack/nova: Complete verification of servers-action-fixed-ip.inc https://review.openstack.org/338606 | 23:25 |
*** baoli has quit IRC | 23:27 | |
*** claudiub has quit IRC | 23:28 | |
*** victorhe has joined #openstack-nova | 23:34 | |
*** xyang1 has quit IRC | 23:37 | |
*** victorhe has quit IRC | 23:38 | |
*** ntpttr has quit IRC | 23:39 | |
*** lamt has quit IRC | 23:40 | |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.create() https://review.openstack.org/386844 | 23:40 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: REST API for resource classes https://review.openstack.org/391959 | 23:40 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.get_by_name() https://review.openstack.org/391917 | 23:40 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.destroy() https://review.openstack.org/391918 | 23:40 |
openstackgerrit | Jay Pipes proposed openstack/nova: placement: adds ResourceClass.save() https://review.openstack.org/400448 | 23:40 |
*** baoli_ has quit IRC | 23:41 | |
*** baoli has joined #openstack-nova | 23:41 | |
*** ntpttr has joined #openstack-nova | 23:41 | |
*** jaypipes has quit IRC | 23:41 | |
*** openstack has joined #openstack-nova | 23:44 | |
*** mlavalle has quit IRC | 23:45 | |
*** liangy_ has quit IRC | 23:51 | |
*** victorhe has joined #openstack-nova | 23:55 | |
*** irenab_ is now known as irenab | 23:56 | |
oomichi | dansmith: EmilienM: I know that now. My bad. Sorry about that. | 23:56 |
oomichi | dansmith: EmilienM: The reason I could not find that is all nova gate jobs are based on xenial(py3) | 23:57 |
*** nicolasbock has quit IRC | 23:58 | |
*** unicell has quit IRC | 23:59 | |
*** victorhe has quit IRC | 23:59 | |
*** unicell1 has joined #openstack-nova | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!