| *** haomaiwang has quit IRC | 00:01 | |
| dansmith | melwitt: if it's a serialized object, you use obj_from_primitive() | 00:01 |
|---|---|---|
| *** haomaiwang has joined #openstack-nova | 00:02 | |
| dansmith | melwitt: if it's a dict-like thing that looks like the object itself, then you use class(context, **primitive) | 00:02 |
| dansmith | melwitt: so, it depends on what you've got | 00:02 |
| openstackgerrit | Danny Wilson proposed openstack/nova: libvirt: add discard support for attached volumes https://review.openstack.org/205726 | 00:02 |
| alex_xu | melwitt: hope you can take a look this when you have time https://review.openstack.org/#/q/status:open+project:openstack/python-novaclient+branch:master+topic:bp/keypair-x509-certificates,n,z it already got one +2, thanks in advance! | 00:02 |
| melwitt | dansmith: okay, thanks. I'm reviewing the RequestSpec object patch, is how my question came up. on creating a Flavor object from a primitive | 00:03 |
| dansmith | melwitt: yeah, I think in that one, he has a thing that looks like a flavor (a legacy thing), not a serialized objects.Flavor object | 00:03 |
| melwitt | dansmith: yeah, I went in scheduler/utils.py to see build_request_spec, it does primitivize a Flavor object if it encounters one. I suppose there's a chance of a legacy thing being possible too. I'll pose it as a question in the review | 00:05 |
| dansmith | melwitt: meaning Flavor.obj_to_primitive()? or nova.objects.obj_to_primitive() ? | 00:06 |
| openstackgerrit | xing-yang proposed openstack/nova: Add os-brick based LibvirtVolumeDriver for ScaleIO https://review.openstack.org/194454 | 00:06 |
| melwitt | dansmith: instance_type = obj_base.obj_to_primitive(instance_type) | 00:07 |
| *** hemna is now known as hemnafk | 00:07 | |
| dansmith | melwitt: yeah, that's not a serialized object, that's a dicty thing | 00:07 |
| dansmith | which means you use Flavor(context, **primitive) | 00:07 |
| melwitt | dansmith: ohh. didn't know that | 00:08 |
| dansmith | I mean, ffs, how are you not getting this.. it's SO OBVIOUS | 00:08 |
| *** annashen has joined #openstack-nova | 00:08 | |
| dansmith | (totally not) | 00:08 |
| melwitt | dansmith: I know!! I am ashamed | 00:08 |
| * dansmith is the one that should be ashamed | 00:08 | |
| melwitt | :P nah | 00:08 |
| *** pixelbeat_ has joined #openstack-nova | 00:09 | |
| dansmith | yes, I dunno how we ended up with two things called primitive and thought it was a good idea | 00:09 |
| dansmith | well, I do, but.. | 00:09 |
| melwitt | one of these days, I'll remember the base.to_primitive vs class.to_primitive | 00:09 |
| *** rmcall has quit IRC | 00:10 | |
| dansmith | well, | 00:10 |
| dansmith | soon enough the silly one will go away | 00:10 |
| dansmith | well, maybe not soon enough, but eventually | 00:11 |
| *** penick has joined #openstack-nova | 00:12 | |
| *** annashen has quit IRC | 00:12 | |
| melwitt | true | 00:13 |
| gzhai | dansmith: Do you have time to take a look at https://review.openstack.org/#/c/203573/ ? | 00:14 |
| dansmith | gzhai: sorry, I'm just about to run off | 00:14 |
| gzhai | dansmith: it's ok. just a simple fix. you can try if have time in future. Thanks. | 00:15 |
| *** mtanino has quit IRC | 00:15 | |
| *** tsekiyama has quit IRC | 00:16 | |
| openstackgerrit | Danny Wilson proposed openstack/nova: libvirt: add discard support for attached volumes https://review.openstack.org/205726 | 00:17 |
| *** achanda has quit IRC | 00:18 | |
| *** patrickeast has quit IRC | 00:18 | |
| *** tjones has quit IRC | 00:18 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add DiskNotFound and VolumeNotFound test https://review.openstack.org/181455 | 00:19 |
| *** Sukhdev_ has quit IRC | 00:19 | |
| *** penick has quit IRC | 00:20 | |
| *** tjones has joined #openstack-nova | 00:21 | |
| *** thorst1 has joined #openstack-nova | 00:22 | |
| *** achanda has joined #openstack-nova | 00:22 | |
| *** jasondotstar has quit IRC | 00:23 | |
| *** thorst1 has quit IRC | 00:24 | |
| *** marcusvrn has quit IRC | 00:25 | |
| *** annashen has joined #openstack-nova | 00:29 | |
| *** Sukhdev_ has joined #openstack-nova | 00:30 | |
| *** salv-orlando has quit IRC | 00:31 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add Support for json-home content_type in wsgi https://review.openstack.org/207261 | 00:31 |
| *** tellesnobrega has joined #openstack-nova | 00:33 | |
| *** edmondsw has quit IRC | 00:34 | |
| *** annashen has quit IRC | 00:34 | |
| *** tellesnobrega has quit IRC | 00:34 | |
| *** dannywilson has quit IRC | 00:36 | |
| *** Sukhdev_ has quit IRC | 00:37 | |
| *** dboik has joined #openstack-nova | 00:37 | |
| *** tjones has quit IRC | 00:38 | |
| *** meghal has quit IRC | 00:39 | |
| *** vishwanathj has joined #openstack-nova | 00:41 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add secgroup param checks for Neutron https://review.openstack.org/192043 | 00:43 |
| *** vishwanathj has quit IRC | 00:46 | |
| *** vishwanathj has joined #openstack-nova | 00:46 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add secgroup param checks for Neutron https://review.openstack.org/192043 | 00:46 |
| *** MentalRay_ has quit IRC | 00:47 | |
| openstackgerrit | enikanorov proposed openstack/nova: Fix forming a message for the exception raised if port limit is hit https://review.openstack.org/204738 | 00:48 |
| *** jwcroppe has joined #openstack-nova | 00:49 | |
| *** otter768 has joined #openstack-nova | 00:54 | |
| *** vishwanathj has quit IRC | 00:55 | |
| *** vilobhmm has quit IRC | 00:55 | |
| *** vilobhmm has joined #openstack-nova | 00:57 | |
| *** jasondotstar has joined #openstack-nova | 00:58 | |
| *** h00327910__ has joined #openstack-nova | 00:59 | |
| *** mriedem has joined #openstack-nova | 00:59 | |
| *** vilobhmm1 has joined #openstack-nova | 00:59 | |
| *** haomaiwang has quit IRC | 01:01 | |
| *** vilobhmm has quit IRC | 01:02 | |
| *** haomaiwang has joined #openstack-nova | 01:02 | |
| *** vilobhmm1 has quit IRC | 01:04 | |
| *** leeantho has quit IRC | 01:06 | |
| *** angdraug has quit IRC | 01:06 | |
| *** tjones has joined #openstack-nova | 01:08 | |
| *** davideagnello has quit IRC | 01:12 | |
| mriedem | xyang: a couple more comments | 01:14 |
| mriedem | pretty close | 01:14 |
| *** wanghua has joined #openstack-nova | 01:15 | |
| mriedem | still got a few changes in this series that need another +2 https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/consolidate-libvirt-fs-volume-drivers,n,z | 01:17 |
| mriedem | and this is ready to go i think https://review.openstack.org/#/c/194412/ | 01:18 |
| xyang | mriedem: hi, thanks, let me check | 01:24 |
| xyang | mriedem: I thought I have to set it in os brick connector first, otherwise I won't have it? Other driver like ceph also doesn't have it yet, right? | 01:27 |
| xyang | mriedem: you are saying this is actually set in nova, so every driver should already have it? | 01:29 |
| *** browne has quit IRC | 01:29 | |
| *** annashen has joined #openstack-nova | 01:30 | |
| *** annashen has quit IRC | 01:35 | |
| *** jasondotstar has quit IRC | 01:35 | |
| openstackgerrit | xing-yang proposed openstack/nova: Add os-brick based LibvirtVolumeDriver for ScaleIO https://review.openstack.org/194454 | 01:35 |
| *** tpeoples has joined #openstack-nova | 01:38 | |
| xyang | mriedem: fixed them. Thanks! | 01:39 |
| *** humble_ has joined #openstack-nova | 01:41 | |
| *** wanghua has quit IRC | 01:45 | |
| *** Jeffrey4l has joined #openstack-nova | 01:47 | |
| *** vilobhmm has joined #openstack-nova | 01:47 | |
| *** dan has quit IRC | 01:50 | |
| *** vilobhmm1 has joined #openstack-nova | 01:51 | |
| *** vilobhmm has quit IRC | 01:51 | |
| *** MentalRay_ has joined #openstack-nova | 01:52 | |
| *** meghal has joined #openstack-nova | 01:55 | |
| *** haomaiwang has quit IRC | 02:01 | |
| *** baoli has joined #openstack-nova | 02:02 | |
| *** haomaiwang has joined #openstack-nova | 02:02 | |
| *** pixelbeat_ has quit IRC | 02:02 | |
| *** Kennan has quit IRC | 02:02 | |
| *** Kennan has joined #openstack-nova | 02:02 | |
| openstackgerrit | xing-yang proposed openstack/nova: Add os-brick based LibvirtVolumeDriver for ScaleIO https://review.openstack.org/194454 | 02:04 |
| openstackgerrit | WenPing Zhang proposed openstack/nova: Add support for yfs storage protocol https://review.openstack.org/202459 | 02:04 |
| *** annashen has joined #openstack-nova | 02:05 | |
| *** dboik has quit IRC | 02:05 | |
| *** annegentle has quit IRC | 02:06 | |
| *** dims has quit IRC | 02:19 | |
| *** unicell has quit IRC | 02:19 | |
| *** dims has joined #openstack-nova | 02:19 | |
| mriedem | xyang: the nova libvirt driver calls connect_volume on the volume driver which updates the connection_info dict with the device_path, so that when the nova libvirt driver calls volume_driver.get_config it passes in that connection_info, | 02:19 |
| mriedem | so we can just get the device_path from that | 02:19 |
| mriedem | the hgst driver was doing it that way | 02:19 |
| xyang | mriedem: got it. Thanks! | 02:20 |
| mriedem | yeah, here it connects the volume http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/driver.py#n1054 | 02:21 |
| *** agireud has joined #openstack-nova | 02:21 | |
| mriedem | next line it gets the config | 02:21 |
| mriedem | it's a bit wonky, i mean, connect_volume could just return the config and then we wouldn't need the pass by reference stuff and separate call to get_config | 02:21 |
| mriedem | not sure why it was written that way | 02:21 |
| xyang | mriedem: I see | 02:21 |
| xyang | mriedem: let me know if anything else needs to be changed | 02:23 |
| mriedem | oh it's that way b/c of live migration | 02:23 |
| mriedem | so no, i think we're good, let me check the latest patch | 02:23 |
| xyang | mriedem: ok | 02:24 |
| mriedem | +2 | 02:24 |
| mriedem | nice work :) | 02:24 |
| xyang | mriedem: thanks a lot for all your help! | 02:25 |
| mriedem | np | 02:25 |
| *** vilobhmm1 has quit IRC | 02:26 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge "versions" tests between v2 and v2.1 https://review.openstack.org/201439 | 02:27 |
| *** cfriesen_ has quit IRC | 02:27 | |
| *** jswarren has quit IRC | 02:28 | |
| *** mriedem1 has joined #openstack-nova | 02:29 | |
| *** jasondotstar has joined #openstack-nova | 02:31 | |
| *** annashen_ has joined #openstack-nova | 02:31 | |
| *** tjones has quit IRC | 02:31 | |
| *** mriedem has quit IRC | 02:31 | |
| *** eliqiao1 has left #openstack-nova | 02:31 | |
| *** agireud has quit IRC | 02:32 | |
| *** annashen_ has quit IRC | 02:36 | |
| *** earlephilhower_ has joined #openstack-nova | 02:39 | |
| *** davideagnello has joined #openstack-nova | 02:39 | |
| *** agireud has joined #openstack-nova | 02:40 | |
| *** earlephilhower has quit IRC | 02:41 | |
| *** READ10 has quit IRC | 02:41 | |
| *** jasondotstar has quit IRC | 02:41 | |
| *** nelsnelson has quit IRC | 02:45 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Add rootwrap daemon mode support https://review.openstack.org/180695 | 02:46 |
| *** changbl has quit IRC | 02:47 | |
| *** annegentle has joined #openstack-nova | 02:50 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Add rootwrap daemon mode support https://review.openstack.org/180695 | 02:51 |
| *** apmelton has quit IRC | 02:51 | |
| *** hakimo_ has joined #openstack-nova | 02:52 | |
| *** meghal has quit IRC | 02:52 | |
| *** nelsnelson has joined #openstack-nova | 02:53 | |
| *** apmelton has joined #openstack-nova | 02:54 | |
| *** hakimo has quit IRC | 02:54 | |
| *** gzhai has quit IRC | 02:57 | |
| *** dims has quit IRC | 02:57 | |
| *** liusheng has joined #openstack-nova | 03:00 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** haomaiwang has joined #openstack-nova | 03:01 | |
| *** slong_ has joined #openstack-nova | 03:01 | |
| *** slong has quit IRC | 03:01 | |
| *** browne has joined #openstack-nova | 03:02 | |
| *** earlephilhower_ is now known as earlephilhower | 03:04 | |
| *** changbl has joined #openstack-nova | 03:06 | |
| *** cpschult has joined #openstack-nova | 03:11 | |
| *** baoli has quit IRC | 03:13 | |
| *** tjones has joined #openstack-nova | 03:15 | |
| *** rfolco has quit IRC | 03:17 | |
| *** kevinbenton has quit IRC | 03:18 | |
| *** annegentle has quit IRC | 03:20 | |
| *** markvoelker_ has quit IRC | 03:20 | |
| *** kevinbenton has joined #openstack-nova | 03:20 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Skip additionalProperties checks when LegacyV2CompatibleWrapper enabled https://review.openstack.org/193858 | 03:21 |
| openstackgerrit | Alex Xu proposed openstack/nova: Fix the wrong file name for legacy v2 compatible wrapper functional test https://review.openstack.org/206326 | 03:21 |
| openstackgerrit | Alex Xu proposed openstack/nova: Return v2 version info with v2 legacy compatible wrapper https://review.openstack.org/206930 | 03:21 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add scenario for API sample tests with legacy v2 compatible wrapper https://review.openstack.org/203485 | 03:21 |
| oomichi | alex_xu: hi, the timing is conflict on ^^^ | 03:22 |
| *** markvoelker_ has joined #openstack-nova | 03:22 | |
| oomichi | alex_xu: so can you check my comment on https://review.openstack.org/#/c/193858 ? | 03:22 |
| alex_xu | oomichi: thanks, let me check | 03:23 |
| oomichi | alex_xu: very simple point | 03:23 |
| alex_xu | oomichi: yea, thanks, I think that is reasonable. will update soon | 03:24 |
| *** annegentle has joined #openstack-nova | 03:26 | |
| *** annegentle has quit IRC | 03:26 | |
| *** gzhai has joined #openstack-nova | 03:27 | |
| *** gzhai has quit IRC | 03:28 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Skip additionalProperties checks when LegacyV2CompatibleWrapper enabled https://review.openstack.org/193858 | 03:29 |
| openstackgerrit | Alex Xu proposed openstack/nova: Fix the wrong file name for legacy v2 compatible wrapper functional test https://review.openstack.org/206326 | 03:29 |
| openstackgerrit | Alex Xu proposed openstack/nova: Return v2 version info with v2 legacy compatible wrapper https://review.openstack.org/206930 | 03:29 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add scenario for API sample tests with legacy v2 compatible wrapper https://review.openstack.org/203485 | 03:29 |
| alex_xu | oomichi: ^ it's done :) | 03:29 |
| oomichi | alex_xu: cool, thanks :) | 03:30 |
| *** annashen_ has joined #openstack-nova | 03:32 | |
| *** lucas__ has joined #openstack-nova | 03:33 | |
| *** jwcroppe has quit IRC | 03:33 | |
| *** jwcroppe has joined #openstack-nova | 03:34 | |
| *** dave-mccowan has quit IRC | 03:34 | |
| *** annashen_ has quit IRC | 03:37 | |
| *** lucas__ has quit IRC | 03:37 | |
| *** cpschult has quit IRC | 03:38 | |
| *** jwcroppe has quit IRC | 03:38 | |
| *** annegentle has joined #openstack-nova | 03:40 | |
| *** jamielennox|away is now known as jamielennox | 03:47 | |
| *** otter768 has quit IRC | 03:50 | |
| *** rajesht has joined #openstack-nova | 03:53 | |
| *** shyama has joined #openstack-nova | 03:53 | |
| *** deepthi has joined #openstack-nova | 03:55 | |
| *** kevinbenton has quit IRC | 03:56 | |
| bharath | Hi, may I know why are we creating a linux bridge every time when we launch an instance? I am using libvirt for nova and ovs as neutron plugin | 03:56 |
| bharath | we are creating per vif every time, referring this code https://github.com/openstack/nova/blob/6b4146e944611685b593deabaa1d83d410e90318/nova/virt/libvirt/vif.py#L445 | 03:56 |
| *** gzhai has joined #openstack-nova | 03:58 | |
| *** coolsvap|away is now known as coolsvap | 03:59 | |
| *** gzhai has quit IRC | 04:00 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** kevinbenton has joined #openstack-nova | 04:01 | |
| *** 16WAAEKLZ has joined #openstack-nova | 04:02 | |
| *** annashen has quit IRC | 04:03 | |
| *** slong__ has joined #openstack-nova | 04:03 | |
| *** slong_ has quit IRC | 04:04 | |
| *** aluria has quit IRC | 04:05 | |
| *** lpetrut has joined #openstack-nova | 04:14 | |
| *** mdrabe has quit IRC | 04:20 | |
| *** salv-orlando has joined #openstack-nova | 04:21 | |
| *** salv-orlando has quit IRC | 04:25 | |
| *** achanda has quit IRC | 04:26 | |
| *** fawadkhaliq has joined #openstack-nova | 04:27 | |
| *** agireud has quit IRC | 04:31 | |
| *** jlvillal has quit IRC | 04:32 | |
| *** annashen has joined #openstack-nova | 04:33 | |
| *** jamielennox is now known as jamielennox|away | 04:33 | |
| *** annashen has quit IRC | 04:38 | |
| *** garyk has quit IRC | 04:42 | |
| *** sakaYK has joined #openstack-nova | 04:42 | |
| *** bharath has quit IRC | 04:43 | |
| *** annegentle has quit IRC | 04:44 | |
| *** chhavi has joined #openstack-nova | 04:47 | |
| *** deepakcs has joined #openstack-nova | 04:48 | |
| *** annashen has joined #openstack-nova | 04:50 | |
| *** fawadkhaliq has quit IRC | 04:52 | |
| *** Marga_ has quit IRC | 04:52 | |
| *** Marga_ has joined #openstack-nova | 04:53 | |
| *** tjones has quit IRC | 04:54 | |
| *** oomichi has quit IRC | 05:00 | |
| *** davideagnello has quit IRC | 05:00 | |
| *** 16WAAEKLZ has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-nova | 05:02 | |
| *** oomichi has joined #openstack-nova | 05:05 | |
| *** oomichi has quit IRC | 05:05 | |
| *** fawadkhaliq has joined #openstack-nova | 05:05 | |
| *** shyama has quit IRC | 05:06 | |
| *** yamahata has joined #openstack-nova | 05:10 | |
| *** shyama has joined #openstack-nova | 05:11 | |
| *** sakaYK has quit IRC | 05:14 | |
| *** unicell has joined #openstack-nova | 05:15 | |
| *** annashen has quit IRC | 05:18 | |
| *** meghal has joined #openstack-nova | 05:20 | |
| *** Eric_HOU has joined #openstack-nova | 05:29 | |
| *** achanda has joined #openstack-nova | 05:29 | |
| *** meghal has quit IRC | 05:30 | |
| *** deepthi has quit IRC | 05:33 | |
| *** annashen has joined #openstack-nova | 05:34 | |
| *** vilobhmm has joined #openstack-nova | 05:36 | |
| *** whenry has quit IRC | 05:36 | |
| *** rotbeard has joined #openstack-nova | 05:37 | |
| *** annashen has quit IRC | 05:39 | |
| *** lpetrut has quit IRC | 05:41 | |
| openstackgerrit | John Hua proposed openstack/nova: Support extra tgz images that only has a single VHD file in it https://review.openstack.org/205831 | 05:44 |
| *** annashen has joined #openstack-nova | 05:45 | |
| *** garyk has joined #openstack-nova | 05:49 | |
| *** dannywilson has joined #openstack-nova | 05:49 | |
| *** shyama has quit IRC | 05:49 | |
| *** otter768 has joined #openstack-nova | 05:51 | |
| *** lpetrut has joined #openstack-nova | 05:52 | |
| *** lsmola has joined #openstack-nova | 05:54 | |
| *** salv-orlando has joined #openstack-nova | 05:54 | |
| *** salv-orlando has quit IRC | 05:55 | |
| *** otter768 has quit IRC | 05:55 | |
| *** Murali has joined #openstack-nova | 05:56 | |
| *** salv-orlando has joined #openstack-nova | 05:56 | |
| *** josecastroleon has joined #openstack-nova | 05:58 | |
| *** lpetrut has quit IRC | 05:59 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:02 | |
| *** Kennan2 has joined #openstack-nova | 06:03 | |
| *** Kennan has quit IRC | 06:03 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: add support for cores per socket https://review.openstack.org/187942 | 06:04 |
| *** nkrinner has joined #openstack-nova | 06:06 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: add in net_io resource limitations https://review.openstack.org/180202 | 06:06 |
| openstackgerrit | garyk proposed openstack/nova: VMware: create common object for limits, reservations and shares https://review.openstack.org/179059 | 06:06 |
| openstackgerrit | garyk proposed openstack/nova: VMware: Resource limits for memory https://review.openstack.org/179060 | 06:06 |
| openstackgerrit | garyk proposed openstack/nova: VMware: store extra_specs object https://review.openstack.org/179061 | 06:06 |
| openstackgerrit | garyk proposed openstack/nova: VMware: add resource limits for disk https://review.openstack.org/179062 | 06:06 |
| *** Eric_HOU has left #openstack-nova | 06:09 | |
| *** dannywilson has quit IRC | 06:11 | |
| *** dannywilson has joined #openstack-nova | 06:12 | |
| *** artom has quit IRC | 06:14 | |
| garyk | johnthetubaguy: ^ - they all needed to be rebased - not sure why gerrit had a conflict | 06:25 |
| *** sakaYK has joined #openstack-nova | 06:25 | |
| openstackgerrit | liaonanhai proposed openstack/nova: Add user_id info: add 'user_id' to REST os-simple-tenant-usage output https://review.openstack.org/151522 | 06:26 |
| *** ankit_ag has joined #openstack-nova | 06:29 | |
| *** rmcall has joined #openstack-nova | 06:30 | |
| *** deepthi has joined #openstack-nova | 06:31 | |
| *** rmcall has quit IRC | 06:33 | |
| *** annashen_ has joined #openstack-nova | 06:35 | |
| *** annashen_ has quit IRC | 06:39 | |
| *** rdopiera has joined #openstack-nova | 06:39 | |
| *** oomichi has joined #openstack-nova | 06:40 | |
| oomichi | johnthetubaguy: hi, morning | 06:42 |
| *** david-lyle has quit IRC | 06:43 | |
| *** annashen has quit IRC | 06:46 | |
| *** shyama has joined #openstack-nova | 06:46 | |
| *** Vikas___ has joined #openstack-nova | 06:47 | |
| *** scheuran has joined #openstack-nova | 06:48 | |
| *** vilobhmm has quit IRC | 06:49 | |
| *** david-lyle has joined #openstack-nova | 06:49 | |
| *** mpavone has joined #openstack-nova | 06:50 | |
| *** bkopilov has quit IRC | 06:52 | |
| *** bkopilov has joined #openstack-nova | 06:55 | |
| *** vilobhmm has joined #openstack-nova | 06:56 | |
| *** Nikolay_St has joined #openstack-nova | 06:57 | |
| *** _gryf has quit IRC | 07:00 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** davideagnello has joined #openstack-nova | 07:01 | |
| *** rubasov has quit IRC | 07:02 | |
| *** haomaiwang has joined #openstack-nova | 07:02 | |
| *** rubasov has joined #openstack-nova | 07:04 | |
| *** davideagnello has quit IRC | 07:05 | |
| *** abhishekk has joined #openstack-nova | 07:05 | |
| *** rubasov has left #openstack-nova | 07:05 | |
| *** doude has joined #openstack-nova | 07:05 | |
| *** unicell has quit IRC | 07:08 | |
| *** unicell has joined #openstack-nova | 07:09 | |
| *** belmoreira has joined #openstack-nova | 07:10 | |
| *** browne has quit IRC | 07:10 | |
| *** rubasov has joined #openstack-nova | 07:12 | |
| *** slong__ has quit IRC | 07:13 | |
| *** salv-orlando has quit IRC | 07:15 | |
| *** markus_z has joined #openstack-nova | 07:16 | |
| *** dannywilson has quit IRC | 07:18 | |
| *** achanda has quit IRC | 07:20 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Show 'locked' information in server details https://review.openstack.org/202391 | 07:22 |
| *** achanda has joined #openstack-nova | 07:23 | |
| *** vishwanathj has joined #openstack-nova | 07:23 | |
| *** vilobhmm1 has joined #openstack-nova | 07:24 | |
| *** unicell has quit IRC | 07:24 | |
| *** haypo has joined #openstack-nova | 07:25 | |
| *** vilobhmm has quit IRC | 07:26 | |
| *** jcoufal has joined #openstack-nova | 07:26 | |
| haypo | morning. i have a patch to fix nova unit tests in nova kilo for openssh 6.8 (fedora 22): https://review.openstack.org/#/c/191847/ - it already got a +2, but it lacks a second +2 ;-) | 07:26 |
| *** unicell has joined #openstack-nova | 07:27 | |
| *** vishwanathj has quit IRC | 07:28 | |
| *** vishwanathj has joined #openstack-nova | 07:29 | |
| *** jwcroppe has joined #openstack-nova | 07:29 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Register JSON-Home data for v2.1 resources https://review.openstack.org/207323 | 07:30 |
| *** vilobhmm1 has quit IRC | 07:31 | |
| *** shyama has quit IRC | 07:34 | |
| *** annashen has joined #openstack-nova | 07:35 | |
| *** sahid has joined #openstack-nova | 07:39 | |
| *** annashen has quit IRC | 07:40 | |
| -openstackstatus- NOTICE: Our CI system is broken again today, jobs are not getting processed at all. | 07:41 | |
| *** mpaolino has joined #openstack-nova | 07:41 | |
| *** ChanServ changes topic to "Our CI system is broken again today, jobs are not getting processed at all." | 07:41 | |
| *** matrohon has joined #openstack-nova | 07:42 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Register JSON-Home data for resources with parent https://review.openstack.org/205905 | 07:43 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Register JSON-Home data for resources with inherits https://review.openstack.org/205944 | 07:44 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Register JSON-Home data for resources actions https://review.openstack.org/206360 | 07:44 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add Support for json-home content_type in wsgi https://review.openstack.org/207261 | 07:44 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Register JSON-Home data for v2.1 resources https://review.openstack.org/207323 | 07:45 |
| *** rmcall has joined #openstack-nova | 07:45 | |
| *** ndipanov has quit IRC | 07:46 | |
| *** moshele has joined #openstack-nova | 07:47 | |
| -openstackstatus- NOTICE: CI system is broken and very far behind. Please do not approve any changes for a while. | 07:51 | |
| *** ChanServ changes topic to "CI system is broken and very far behind. Please do not approve any changes for a while." | 07:51 | |
| *** otter768 has joined #openstack-nova | 07:52 | |
| *** jasonpan has joined #openstack-nova | 07:52 | |
| *** alex_klimov has joined #openstack-nova | 07:52 | |
| *** alex_klimov has quit IRC | 07:53 | |
| *** alex_klimov has joined #openstack-nova | 07:53 | |
| johnthetubaguy | oomichi: morning | 07:55 |
| *** achanda has quit IRC | 07:56 | |
| *** otter768 has quit IRC | 07:56 | |
| johnthetubaguy | garyk: seems like gerrit is in a bad way right now, but it should be OK for anyone to reapprove those ones now (i.e. you have met the criteria to pass the freeze, and get merged) | 07:56 |
| johnthetubaguy | mikal: are you around by any chance for a quick catch up | 07:57 |
| *** jamielennox|away is now known as jamielennox | 07:57 | |
| oomichi | johnthetubaguy: maybe today is very busy for you;) I also have a qestion about l-2 deadline. | 07:58 |
| *** achanda has joined #openstack-nova | 07:58 | |
| oomichi | johnthetubaguy: can we consider json-home feature is a part of v2.1 feature or not? | 07:59 |
| johnthetubaguy | oomichi: I am about to go offline, unfortunately, but I hope I can help | 07:59 |
| johnthetubaguy | oomichi: I think I made it higher than low priority, maybe medium? | 07:59 |
| oomichi | johnthetubaguy: yeah, that is right on current bp priority of json-home | 07:59 |
| oomichi | johnthetubaguy: that is medium | 08:00 |
| johnthetubaguy | oomichi: so basically, I am going to defer the low ones, then double check the rest, if that makes sense, so it should be safe | 08:00 |
| *** haomaiwang has quit IRC | 08:01 | |
| johnthetubaguy | oomichi: its really up to the api subteam to decide what they want in their "priority" though, within reason. I figured that one was important, but not really important. But I am willing to be told otherwise. Does sound OK? | 08:01 |
| *** haomaiwang has joined #openstack-nova | 08:02 | |
| oomichi | johnthetubaguy: ah, I see the point. after re-reading your mail, I got it. thanks :) | 08:02 |
| oomichi | johnthetubaguy: yeah, I will try to get agreements from the subteam | 08:03 |
| oomichi | johnthetubaguy: thanks for your time | 08:03 |
| johnthetubaguy | oomichi: cool, no problems | 08:03 |
| johnthetubaguy | oomichi: if I defer something I shouldn't have, for example, we can bring it back | 08:04 |
| johnthetubaguy | oomichi: one thing about bugs with API changes, I am still cool trying to push those forward at this point, we will need to slow that down if it gets too distracting, but it seems like a small trickle that we can deal with at this point, but yeah, let me know if the API wants to change direction there, going with my gut instinct on that really | 08:06 |
| *** achanda has quit IRC | 08:07 | |
| bauzas | morning Nova | 08:08 |
| *** shyama has joined #openstack-nova | 08:09 | |
| *** markus_z has quit IRC | 08:09 | |
| *** lucasagomes has joined #openstack-nova | 08:12 | |
| *** rmcall has quit IRC | 08:12 | |
| *** humble__ has joined #openstack-nova | 08:12 | |
| *** humble_ has quit IRC | 08:14 | |
| *** kmARC has joined #openstack-nova | 08:14 | |
| *** sahid has quit IRC | 08:17 | |
| johnthetubaguy | bauzas: did you seem my email on the exception process? did it make sense to you? | 08:18 |
| bauzas | johnthetubaguy: yup, I'm fine with it | 08:19 |
| *** shuquan has joined #openstack-nova | 08:19 | |
| bauzas | johnthetubaguy: an etherpad helps keeping the ML quiet | 08:19 |
| johnthetubaguy | bauzas: cool, I am offline for much of today and tomorrow, so just wanting to make sure most of the details are clear | 08:19 |
| johnthetubaguy | bauzas: you are out tomorrow I guess? | 08:19 |
| johnthetubaguy | or was that next week? | 08:19 |
| bauzas | johnthetubaguy: yeah, till 13th included :( | 08:19 |
| bauzas | johnthetubaguy: family time | 08:20 |
| johnthetubaguy | bauzas: I hope you have a good relax! | 08:20 |
| bauzas | even if that's really time because of L3 | 08:20 |
| johnthetubaguy | bauzas: oh, and that! | 08:20 |
| bauzas | bad time | 08:20 |
| *** _gryf has joined #openstack-nova | 08:20 | |
| bauzas | you know that I'm a workaholic | 08:20 |
| bauzas | but that's fair to leave my computer out for the next 2 weeks and just do hiking | 08:20 |
| johnthetubaguy | true, I find having a break can make you go faster, as odd as that sounds! | 08:20 |
| bauzas | erm | 08:21 |
| winston-d_zZZ | dannywilson, beagles /win 4 | 08:21 |
| winston-d_zZZ | oops | 08:21 |
| bauzas | johnthetubaguy: thanks | 08:21 |
| bauzas | johnthetubaguy: I'll still be there for the Nova meeting today | 08:21 |
| bauzas | johnthetubaguy: so I could +1/-1 any discussion there | 08:22 |
| johnthetubaguy | I am in a field sadly, so I think mikal is doing the honours, but I should be online tomorrow morning to catch up with things a little | 08:22 |
| *** sahid has joined #openstack-nova | 08:24 | |
| *** kiran-r has joined #openstack-nova | 08:25 | |
| *** jistr has joined #openstack-nova | 08:25 | |
| bauzas | johnthetubaguy: sure thing | 08:26 |
| garyk | johnthetubaguy: not sure that i understand. do we need to wait for the approval again? | 08:26 |
| *** romainh has joined #openstack-nova | 08:26 | |
| johnthetubaguy | garyk: it was just my way of saying, I was going to hold off approve that, due to the above message from infra, but since you have been approved once, you don't need an exception to merge after the freeze | 08:27 |
| *** matrohon has quit IRC | 08:28 | |
| bauzas | johnthetubaguy: I guess you also saw mriedem1's thread ? | 08:28 |
| johnthetubaguy | bauzas: not yet | 08:29 |
| bauzas | johnthetubaguy: k, take time | 08:29 |
| *** matrohon has joined #openstack-nova | 08:29 | |
| garyk | johnthetubaguy: ok, thanks. | 08:31 |
| johnthetubaguy | garyk: basically, we are not aligned with release management for this freeze, so we can be looser with the rules to make our lives a bit easier | 08:31 |
| garyk | johnthetubaguy: honetsly i am really confused with the relase at the moment. my understanding is that there was talk about tags and in parallel with the the liberty milestones. not sure if these are one and the same or orthogonal. | 08:33 |
| *** afazekas has joined #openstack-nova | 08:33 | |
| garyk | all i know is that the gate is choking :) | 08:33 |
| johnthetubaguy | garyk: yeah, in the ML post I noted we have decoupled the liberty-2 tag and the Non-priority Feature Freeze, to make everyones lives a bit easier, and still meet all our process goals | 08:34 |
| *** annashen has joined #openstack-nova | 08:36 | |
| *** BobBall_AWOL is now known as BobBaool | 08:36 | |
| *** BobBaool is now known as BobBall | 08:36 | |
| garyk | ok, thanks for the clarification | 08:37 |
| *** bfic has joined #openstack-nova | 08:37 | |
| *** jwcroppe has quit IRC | 08:37 | |
| johnthetubaguy | garyk: no problem | 08:37 |
| *** salv-orlando has joined #openstack-nova | 08:37 | |
| *** jwcroppe has joined #openstack-nova | 08:38 | |
| *** mpaolino has quit IRC | 08:38 | |
| *** ZZelle has joined #openstack-nova | 08:38 | |
| *** danpb has joined #openstack-nova | 08:39 | |
| *** mpaolino has joined #openstack-nova | 08:39 | |
| BobBall | XenProject CI has been disabled because it's run out of disk space again and failing all changes | 08:40 |
| *** annashen has quit IRC | 08:41 | |
| BobBall | Second time this week so we'll be digging in more to try and understand why the logs are filling up so fast before re-enabling | 08:41 |
| *** pixelbeat_ has joined #openstack-nova | 08:42 | |
| *** jwcroppe has quit IRC | 08:42 | |
| johnthetubaguy | BobBall: same issue gerrit had at one point this week, if that helps | 08:43 |
| johnthetubaguy | BobBall: I mean, infra | 08:43 |
| *** yassine has joined #openstack-nova | 08:43 | |
| johnthetubaguy | BobBall: at least it sounds similar | 08:44 |
| BobBall | Looks like several people have found this recently | 08:44 |
| BobBall | https://issues.jenkins-ci.org/browse/JENKINS-10160 | 08:44 |
| BobBall | https://issues.jenkins-ci.org/browse/JENKINS-29490 is the latest version of the issue AFAICT | 08:45 |
| *** lpetrut has joined #openstack-nova | 08:47 | |
| johnthetubaguy | interesting | 08:47 |
| *** aix has joined #openstack-nova | 08:47 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: support limits via image meta data https://review.openstack.org/180942 | 08:48 |
| openstackgerrit | garyk proposed openstack/nova: Objects: add limits, reservations and shares to ImageMetaProps object https://review.openstack.org/188321 | 08:48 |
| *** salv-orlando has quit IRC | 08:49 | |
| BobBall | Assuming that's the issue it should be fixed now. Fingers crossed eh? | 08:52 |
| *** fawadkhaliq has quit IRC | 08:53 | |
| *** ChanServ changes topic to "This channel is for Nova development. For support of Nova deployments, please use #openstack. -=- master is open for Lemming development" | 09:00 | |
| -openstackstatus- NOTICE: CI is back online but has a huge backlog. Please be patient and if possible delay approving changes until it has caught up. | 09:00 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** ndipanov has joined #openstack-nova | 09:01 | |
| *** jlanoux has joined #openstack-nova | 09:01 | |
| *** haomaiwang has joined #openstack-nova | 09:02 | |
| *** achanda has joined #openstack-nova | 09:08 | |
| openstackgerrit | Zoltan Arnold Nagy proposed openstack/nova: libvirt: add encrypted rbd volume support https://review.openstack.org/206576 | 09:10 |
| *** achanda has quit IRC | 09:13 | |
| bauzas | johnthetubaguy: OH "Bad timing, but I am mostly away from my computer over the next few days" | 09:18 |
| bauzas | johnthetubaguy: travel or holidays ? | 09:19 |
| *** jwcroppe has joined #openstack-nova | 09:20 | |
| *** jwcroppe has quit IRC | 09:21 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-novaclient: Refactor and update test_instances and test_volumes_api https://review.openstack.org/207355 | 09:21 |
| *** jwcroppe has joined #openstack-nova | 09:21 | |
| *** flwang1 has joined #openstack-nova | 09:22 | |
| *** jwcroppe_ has joined #openstack-nova | 09:23 | |
| *** jwcroppe_ has quit IRC | 09:23 | |
| *** jwcroppe_ has joined #openstack-nova | 09:24 | |
| *** fawadkhaliq has joined #openstack-nova | 09:25 | |
| *** jamielennox is now known as jamielennox|away | 09:25 | |
| *** fawadkhaliq has quit IRC | 09:26 | |
| *** fawadkhaliq has joined #openstack-nova | 09:26 | |
| *** oomichi has quit IRC | 09:27 | |
| *** jwcroppe has quit IRC | 09:27 | |
| *** jwcroppe_ has quit IRC | 09:28 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Fix overloading of block device on boot by device name https://review.openstack.org/190324 | 09:30 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Add documentation for block device mapping https://review.openstack.org/192651 | 09:30 |
| *** ip0353 has joined #openstack-nova | 09:30 | |
| *** ig0r_ has joined #openstack-nova | 09:31 | |
| *** sakaYK has quit IRC | 09:34 | |
| *** jcoufal has quit IRC | 09:35 | |
| *** ishant has joined #openstack-nova | 09:35 | |
| *** annashen has joined #openstack-nova | 09:37 | |
| *** salv-orlando has joined #openstack-nova | 09:38 | |
| *** jwcroppe has joined #openstack-nova | 09:38 | |
| openstackgerrit | Shuquan Huang proposed openstack/nova: nova.utils._get_root_helper() should be public https://review.openstack.org/207366 | 09:40 |
| *** jwcroppe has quit IRC | 09:40 | |
| *** Nikolay_St has quit IRC | 09:41 | |
| *** annashen has quit IRC | 09:42 | |
| *** Nikolay_St has joined #openstack-nova | 09:43 | |
| *** dims has joined #openstack-nova | 09:50 | |
| bauzas | flavor or instance_type ? | 09:51 |
| bauzas | good question from melwitt here https://review.openstack.org/#/c/145528/37/nova/objects/request_spec.py,cm | 09:52 |
| *** jcoufal has joined #openstack-nova | 09:52 | |
| bauzas | who could give a clear view on that ? | 09:52 |
| bauzas | I don't want to add a badly named field now which would need some upgrade stuff to fix that :( | 09:52 |
| bauzas | dansmith: ^ | 09:52 |
| *** otter768 has joined #openstack-nova | 09:53 | |
| bauzas | (pings at very indecent hours, I know) | 09:53 |
| *** coolsvap is now known as coolsvap|away | 09:55 | |
| lxsli | ndipanov: good morning | 09:55 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Add documentation for block device mapping https://review.openstack.org/192651 | 09:56 |
| sdague | alex_xu: you still up? I'd like to get the schema relaxation stuff landed, but it seemed kind of light on test cases, and it's really important we don't get that one wrong | 09:56 |
| lxsli | ndipanov: can we discuss resource-objects please? | 09:56 |
| ndipanov | lxsli, yeah sure | 09:57 |
| *** marzif_ has joined #openstack-nova | 09:57 | |
| *** otter768 has quit IRC | 09:57 | |
| ndipanov | johnthetubaguy, would be interested in your take on this https://review.openstack.org/#/c/197063/ | 09:58 |
| lxsli | ndipanov: so imo Inventory is only responsible for tracking the amount available+used, not for fitting, do you agree? | 09:58 |
| ndipanov | lxsli, well it's your object model not mine :) | 09:59 |
| ndipanov | but kidding aside | 09:59 |
| ndipanov | I am not sure | 09:59 |
| ndipanov | some resources do need the fitting | 09:59 |
| ndipanov | it's just how they work | 09:59 |
| lxsli | yes absolutely, but that is something else's job | 10:00 |
| ndipanov | so not sure it should be forcefully taken out of inventory | 10:00 |
| lxsli | Today, the host_manager calls hw.instance_topology_from_instance to do the fitting | 10:00 |
| ndipanov | (which is not a bad name I have to say) | 10:00 |
| ndipanov | hmmm does it? | 10:00 |
| lxsli | it does! and then hw.get_host_numa_usage_from_instance to do the actual consumption | 10:00 |
| *** haomaiwang has quit IRC | 10:01 | |
| *** alexpilotti has joined #openstack-nova | 10:01 | |
| ndipanov | argh | 10:01 |
| ndipanov | so hw.instance_topology_from_instance is really just a wrapper around the fact that we don't use objects everywhere | 10:01 |
| *** haomaiwang has joined #openstack-nova | 10:01 | |
| lxsli | Sure, the main point is that fitting and consumption are already distinct steps | 10:01 |
| ndipanov | it's sole purpose is to have a single method to get the topology OBJECT from anything that says it's an instance | 10:01 |
| ndipanov | yes they are | 10:01 |
| ndipanov | but fitting has to be done | 10:02 |
| ndipanov | for resources such as PCI/numa | 10:02 |
| *** jcoufal has quit IRC | 10:02 | |
| ndipanov | it's not an optional step, right? | 10:02 |
| lxsli | So providing we can get the NUMATopology out of the Inventory and supply it to the fitting step, we can later consume from the Inventory with pre-fitted resources (IE, specific ones) | 10:02 |
| lxsli | No absolutely | 10:02 |
| ndipanov | well fitting goes first, and then consumption happens with the result of fitting | 10:03 |
| lxsli | Yes and the Inventory is really only concerned with that second step | 10:03 |
| ndipanov | ok I can see it that way | 10:03 |
| lxsli | So it doesn't need an interface to return the results of fitting, because it doesn't do fitting | 10:03 |
| ndipanov | ok | 10:03 |
| openstackgerrit | lyanchih proposed openstack/nova: libvirt: serial console ports count upper limit needs to be checked https://review.openstack.org/207373 | 10:03 |
| ndipanov | but we still do need that interface right?\ | 10:03 |
| lxsli | Yes, somewhere | 10:03 |
| lxsli | Currently it's provided by hw.instance_topology_from_instance | 10:04 |
| ndipanov | no it's provided by hw.numa_fit_instance_to_host | 10:04 |
| lxsli | oh right | 10:05 |
| lxsli | where did I get the other one from? ... nevermnind | 10:05 |
| lxsli | Anyway, does that answer your comment? | 10:05 |
| ndipanov | but my point is - if you are writing a generic framework for how resources are consumed in nova, you can't just skip this step - you should model it and then make it a noop for the ones that don't need it | 10:06 |
| ndipanov | I mean | 10:06 |
| ndipanov | I guess you could say that you only pass already fitted stuff to inventory methods | 10:06 |
| lxsli | You seem to be conflating fitting and consumption again | 10:06 |
| lxsli | Yes exactly! | 10:06 |
| openstackgerrit | Shuquan Huang proposed openstack/nova: The API unit tests for serial console use http instead of ws https://review.openstack.org/207375 | 10:06 |
| ndipanov | ok | 10:06 |
| ndipanov | it just seems to me that those should belong in the same object | 10:07 |
| openstackgerrit | Jeffrey Zhang proposed openstack/nova: Cleanup the the test_api_validation unittest https://review.openstack.org/207376 | 10:07 |
| ndipanov | I guess you guys don't agree | 10:07 |
| ndipanov | but that's ok | 10:07 |
| *** shuquan has quit IRC | 10:07 | |
| lxsli | I'm not saying that's a bad idea | 10:07 |
| ndipanov | without actually working out the code that uses Inventory it's hard to be sure | 10:08 |
| ndipanov | but I have a feeling that it is going to end up with a set of special cases somewhere | 10:08 |
| lxsli | I'd prefer to get this merged and then perhaps do that as a next step, with no-op for IntegerInventory as you suggested | 10:08 |
| ndipanov | anyway - I don't need to block on this - it can be fixed later | 10:09 |
| ndipanov | are those otherwise ready to go in? | 10:09 |
| lxsli | Yes I think so | 10:09 |
| ndipanov | ok | 10:09 |
| ndipanov | will respond now | 10:09 |
| lxsli | They have been widely and extensively reviewed :) | 10:09 |
| _gryf | ndipanov, hi, thanks on your feedback on https://review.openstack.org/#/c/197063/ | 10:09 |
| lxsli | Thanks | 10:10 |
| *** deepthi has quit IRC | 10:10 | |
| *** shyama has quit IRC | 10:10 | |
| ndipanov | _gryf, np - would be awesome to get some more thoughts on it - there's some work being discussed in that area of code so it's best to keep ppl on the same page about where things are going | 10:11 |
| *** exploreshaifali has joined #openstack-nova | 10:11 | |
| *** jcoufal has joined #openstack-nova | 10:12 | |
| _gryf | ndipanov, I'm just wondering, how much sense it would be in removing zk driver from SG - as a resolution for the bug I filled the other day (https://bugs.launchpad.net/nova/+bug/1443910) | 10:12 |
| openstack | Launchpad bug 1443910 in OpenStack Compute (nova) "Zookeeper servicegroup driver crashes" [Low,Confirmed] | 10:12 |
| *** hshiina has joined #openstack-nova | 10:12 | |
| ndipanov | _gryf, THIS IS NOVA!!!! we never remove things! | 10:12 |
| ndipanov | just kidding | 10:13 |
| _gryf | ndipanov, ndipanov oh… ;) | 10:13 |
| ndipanov | well it's set to be replaced by a compeltely different implementation | 10:13 |
| ndipanov | so that's kind of the same thing really | 10:13 |
| _gryf | ndipanov, so the other option is to rewrite the zk driver with kazoo as a backend | 10:13 |
| _gryf | or | 10:13 |
| ndipanov | _gryf, kinda - it's to use tooz which is a python wrapper around some basic distributed primitves | 10:14 |
| _gryf | wait for tooz service groups | 10:14 |
| ndipanov | and does use kazoo internally | 10:14 |
| ndipanov | tooz | 10:14 |
| _gryf | yes | 10:14 |
| ndipanov | is the answer | 10:14 |
| ndipanov | afaik (right johnthetubaguy ) | 10:14 |
| ndipanov | ? | 10:14 |
| ndipanov | that's the reason why I don't think blocking your patch is reasonable | 10:15 |
| *** jasondotstar has joined #openstack-nova | 10:15 | |
| ndipanov | at this point | 10:15 |
| _gryf | right. | 10:16 |
| lxsli | johnthetubaguy: check around the 1min mark: https://www.youtube.com/watch?v=LZY9_Xr5XPA&index=9&list=RDRve03u7oEvI | 10:16 |
| *** kbyrne has quit IRC | 10:21 | |
| *** kbyrne has joined #openstack-nova | 10:23 | |
| *** romainh has quit IRC | 10:23 | |
| bauzas | ndipanov: I disagree with you about ZK | 10:25 |
| bauzas | ndipanov: re: _gryf https://review.openstack.org/#/c/197063/11//COMMIT_MSG,cm | 10:25 |
| ndipanov | bauzas, obivously you do | 10:26 |
| bauzas | ndipanov: because even if we know that we'll be removing ZK, it's there until the tooz BP is done | 10:26 |
| ndipanov | oh comeon | 10:26 |
| ndipanov | it's garbage | 10:26 |
| ndipanov | and this is a useful feature people need | 10:26 |
| bauzas | ndipanov: right, but that's not a reason for not having the check within the base method IMHO | 10:26 |
| ndipanov | what base method? | 10:27 |
| bauzas | ndipanov: ie. like I said to _gryf I'm +1 if he adds the if branch directly in the base module | 10:27 |
| ndipanov | ok sure whatever | 10:27 |
| bauzas | ndipanov: https://github.com/openstack/nova/blob/master/nova/servicegroup/api.py#L87 | 10:27 |
| bauzas | ndipanov: here ^ | 10:28 |
| bauzas | ndipanov: that way, we know that any driver would then be not checked | 10:28 |
| ndipanov | yeah woks for me | 10:28 |
| bauzas | ndipanov: and it wouldn't need to change ZK | 10:28 |
| bauzas | _gryf: ^ | 10:28 |
| *** yamahata has quit IRC | 10:30 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Fix cinder volume stuck in swap_volume https://review.openstack.org/207385 | 10:32 |
| *** jwcroppe has joined #openstack-nova | 10:32 | |
| ndipanov | bauzas, +1 from me - pls comment on the patch though so we have the right context there | 10:32 |
| bauzas | ndipanov: just did :) | 10:32 |
| *** josecastroleon has quit IRC | 10:34 | |
| *** claudiub has joined #openstack-nova | 10:35 | |
| *** tpeoples has quit IRC | 10:38 | |
| *** annashen has joined #openstack-nova | 10:38 | |
| *** shyama has joined #openstack-nova | 10:39 | |
| *** aix has quit IRC | 10:39 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: default_schedule_zone should be a list of AZ(s) https://review.openstack.org/207221 | 10:39 |
| *** annashen has quit IRC | 10:42 | |
| alex_xu | sdague: hi, I'm here | 10:47 |
| alex_xu | sdague: just saw the xen server ci failed, I just need recheck? | 10:48 |
| alex_xu | sdague: oops, all the tests didn't run | 10:49 |
| sdague | alex_xu: did you see my comments from yesterday on that patch? | 10:50 |
| alex_xu | sdague: yes, I updated it also | 10:50 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Store correct VirtCPUTopology https://review.openstack.org/197125 | 10:50 |
| sdague | ok, cool, I'll take a look again after I get another cup of coffee. | 10:51 |
| alex_xu | sdague: cool, thanks :) | 10:51 |
| sdague | some of the unit tests for ec2 are failing on a json assert, and I just realized our assertJsonEqual is not very good at really telling you want failed | 10:54 |
| *** Jeffrey4l has quit IRC | 10:56 | |
| lxsli | ndipanov: you made me check the spec again and I think I'm finally seeing what you were getting at. One of the goals of the resource-objects spec is to provide uniform resource comparison and consumption. If you have to pre-fit some resources beforehand, it kinda obviates the comparison check. So can_provide needs to take unfitted resources and do the fitting. Is that a fair summary? | 10:59 |
| ndipanov | yes | 10:59 |
| lxsli | darn. | 10:59 |
| ndipanov | well | 10:59 |
| ndipanov | or have a separate method | 10:59 |
| ndipanov | up to you | 11:00 |
| lxsli | What'd the usecase for can_provide(fitted) be? | 11:00 |
| ndipanov | I think we don't do nearly enough API design for a project written in a language that is basically all about nice programming interfaces :) | 11:00 |
| *** haomaiwang has quit IRC | 11:01 | |
| ndipanov | lxsli, what do you mean - what usecase? | 11:01 |
| ndipanov | you have to do it whenever you want to do a placement on a new host (so either boot or migrate) | 11:01 |
| lxsli | If we had fit_resources(unfitted) and can_provide(fitted) rather than a single can_provide(unfitted) - when would we ever call them singly? | 11:02 |
| ndipanov | we could turn it on it's head and say | 11:02 |
| *** dims_ has joined #openstack-nova | 11:02 | |
| ndipanov | a resource knows how to fit itself given the context of an inventory | 11:02 |
| *** jwcroppe has quit IRC | 11:02 | |
| *** dims has quit IRC | 11:02 | |
| ndipanov | so inside consume() | 11:02 |
| *** 7GHAAT9JB has joined #openstack-nova | 11:02 | |
| ndipanov | we would do | 11:02 |
| *** davideagnello has joined #openstack-nova | 11:03 | |
| ndipanov | res = resource.fit_onto(inventory) | 11:03 |
| ndipanov | and the same thing inside consume | 11:03 |
| lxsli | I think I prefer having the container do it | 11:04 |
| ndipanov | but the problem with that is that it obfuscates the fact that there has to be a DB call in between | 11:04 |
| ndipanov | so yes I agree | 11:04 |
| *** zhenguo has quit IRC | 11:04 | |
| *** jlvillal has joined #openstack-nova | 11:04 | |
| ndipanov | we would only call the fitting method in a claim for example | 11:04 |
| lxsli | It's lunchtime here but I'll work a patch afterwards | 11:04 |
| ndipanov | lxsli, here too | 11:04 |
| *** mwagner_lap has quit IRC | 11:05 | |
| lxsli | thanks for your help! | 11:06 |
| *** davideagnello has quit IRC | 11:07 | |
| dims_ | lxsli: i had not heard them before (youtube link)! sound good :) | 11:07 |
| *** romainh has joined #openstack-nova | 11:10 | |
| *** achanda has joined #openstack-nova | 11:10 | |
| openstackgerrit | Sean Dague proposed openstack/nova: return more details on assertJsonEqual fail https://review.openstack.org/207403 | 11:12 |
| sdague | we've got at least 2 patches failing in the gate because of a non obvious race in our unit tests in the ec2 path | 11:12 |
| sdague | that patch should give us enough debug to figure those out | 11:12 |
| *** aix has joined #openstack-nova | 11:13 | |
| *** shyama has quit IRC | 11:13 | |
| *** josecastroleon has joined #openstack-nova | 11:14 | |
| *** achanda has quit IRC | 11:15 | |
| *** lucasagomes has quit IRC | 11:19 | |
| *** tonyb has quit IRC | 11:19 | |
| *** jasondotstar has quit IRC | 11:19 | |
| *** tonyb has joined #openstack-nova | 11:20 | |
| openstackgerrit | Sean Dague proposed openstack/nova: return more details on assertJsonEqual fail https://review.openstack.org/207403 | 11:24 |
| sdague | danpb: quick respin on that based on dmitry's suggestion | 11:24 |
| openstackgerrit | Moshe Levi proposed openstack/nova: Enhance PCI passthrough whitelist to support regex https://review.openstack.org/199488 | 11:25 |
| *** deepakcs has quit IRC | 11:25 | |
| *** salv-orl_ has joined #openstack-nova | 11:25 | |
| *** jcoufal has quit IRC | 11:26 | |
| *** salv-orlando has quit IRC | 11:26 | |
| *** jcoufal has joined #openstack-nova | 11:26 | |
| *** gszasz has joined #openstack-nova | 11:28 | |
| *** gszasz has quit IRC | 11:31 | |
| *** gszasz has joined #openstack-nova | 11:32 | |
| *** annashen has joined #openstack-nova | 11:39 | |
| *** annashen has quit IRC | 11:43 | |
| *** thorst1 has joined #openstack-nova | 11:44 | |
| *** rotbeard has quit IRC | 11:45 | |
| *** marcusvrn has joined #openstack-nova | 11:47 | |
| nagyz_ | mriedem1, it's not yet public - it's something we want to integrate first into ICOS delivery | 11:49 |
| nagyz_ | mriedem1, by the summit I'm hoping Nikola (who is the main author) can get it out there on github | 11:50 |
| *** rajesht has quit IRC | 11:53 | |
| *** otter768 has joined #openstack-nova | 11:53 | |
| *** otter768 has quit IRC | 11:58 | |
| *** 7GHAAT9JB has quit IRC | 12:01 | |
| *** haomaiwang has joined #openstack-nova | 12:02 | |
| *** exploreshaifali has quit IRC | 12:02 | |
| *** fawadk has joined #openstack-nova | 12:02 | |
| *** fawadkhaliq has quit IRC | 12:03 | |
| *** lucasagomes has joined #openstack-nova | 12:05 | |
| *** lucasagomes is now known as lucas-hungry | 12:05 | |
| *** _kiran_ has joined #openstack-nova | 12:06 | |
| *** cpschult has joined #openstack-nova | 12:06 | |
| *** shyama has joined #openstack-nova | 12:07 | |
| *** kiran-r has quit IRC | 12:09 | |
| openstackgerrit | Alexis Lee proposed openstack/nova: Prevent (un)pinning unknown CPUs https://review.openstack.org/207040 | 12:10 |
| *** fawadk has quit IRC | 12:10 | |
| *** ishant has quit IRC | 12:11 | |
| lxsli | bauzas: afternoon | 12:16 |
| bauzas | lxsli: 'lo, on a call | 12:17 |
| *** vladikr has joined #openstack-nova | 12:18 | |
| lxsli | OK | 12:18 |
| *** cpschult has quit IRC | 12:18 | |
| *** edmondsw has joined #openstack-nova | 12:19 | |
| *** _kiran_ has quit IRC | 12:20 | |
| *** eharney has quit IRC | 12:21 | |
| *** jasondotstar has joined #openstack-nova | 12:22 | |
| *** shuquan has joined #openstack-nova | 12:23 | |
| *** jwcroppe has joined #openstack-nova | 12:23 | |
| *** ericksonsantos has joined #openstack-nova | 12:25 | |
| *** samueldmq has joined #openstack-nova | 12:26 | |
| *** hshiina has quit IRC | 12:28 | |
| *** lucas-hungry is now known as lucasagomes | 12:30 | |
| *** shuquan has quit IRC | 12:33 | |
| *** aysyd has joined #openstack-nova | 12:33 | |
| *** jamielennox|away is now known as jamielennox | 12:35 | |
| *** liusheng has quit IRC | 12:37 | |
| *** liusheng has joined #openstack-nova | 12:37 | |
| *** flwang1 has quit IRC | 12:38 | |
| *** flwang1 has joined #openstack-nova | 12:38 | |
| *** romainh has left #openstack-nova | 12:40 | |
| *** annashen has joined #openstack-nova | 12:40 | |
| *** hshiina has joined #openstack-nova | 12:41 | |
| *** baoli has joined #openstack-nova | 12:41 | |
| *** baoli has quit IRC | 12:43 | |
| *** baoli has joined #openstack-nova | 12:44 | |
| *** annashen has quit IRC | 12:44 | |
| *** raildo has joined #openstack-nova | 12:46 | |
| *** dustins has joined #openstack-nova | 12:48 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Fix IronicHostManager to skip get_by_host() call. https://review.openstack.org/206736 | 12:48 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down https://review.openstack.org/184086 | 12:49 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: Service group drivers forced_down flag utilization https://review.openstack.org/197063 | 12:49 |
| *** salv-orl_ has quit IRC | 12:54 | |
| *** nkrinner has quit IRC | 12:55 | |
| *** htruta has joined #openstack-nova | 12:55 | |
| *** kylek3h has quit IRC | 12:55 | |
| *** kylek3h has joined #openstack-nova | 12:55 | |
| *** rlrossit has joined #openstack-nova | 12:57 | |
| lxsli | bauzas: ping? | 12:57 |
| bauzas | lxsli: ping | 12:57 |
| *** kylek3h has quit IRC | 12:57 | |
| bauzas | lxsli: still on call | 12:57 |
| lxsli | bauzas: ok, I'll type and reply when you can please | 12:58 |
| *** kylek3h has joined #openstack-nova | 12:58 | |
| lxsli | bauzas: looking at your RequestSpec, you have an InstanceNUMATopology there, which is already placed. I'm looking at ImageMetaProps which is as far as I've been able to trace the request object back in the current code and that has 'hw_numa_cpus': fields.ListOfSetsOfIntegersField() | 12:59 |
| lxsli | which again is already placed, so it seems like the instance request must already be placed? Does that match your knowledge please? | 13:00 |
| *** haomaiwang has quit IRC | 13:01 | |
| bauzas | lxsli: ending the call, will reply to you in a few | 13:01 |
| *** haomaiwang has joined #openstack-nova | 13:02 | |
| *** kylek3h has quit IRC | 13:03 | |
| bauzas | lxsli: I don't get your problem, you mean why I didn't provided an Instance object? | 13:03 |
| bauzas | (call ended) | 13:03 |
| *** flwang1 has quit IRC | 13:03 | |
| *** davideagnello has joined #openstack-nova | 13:03 | |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Add logging for dm-crypt error conditions https://review.openstack.org/198774 | 13:03 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Replace stubs with mocks for test_dmcrypt https://review.openstack.org/198752 | 13:03 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Disconnect dm-crypt on instance suspend/stop https://review.openstack.org/141485 | 13:03 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Add unit tests for methods https://review.openstack.org/198880 | 13:03 |
| bauzas | lxsli: what do you mean by "placed" ? | 13:03 |
| lxsli | I thought that NUMA requests came in in an abstract form, EG "two CPUs in one cell and two in another, with X mem" | 13:04 |
| lxsli | And the scheduler "placed" or "fitted" that to each hosts NUMATopology to discover specific CPUs which can fulfill that | 13:05 |
| *** markus_z has joined #openstack-nova | 13:05 | |
| lxsli | in fact I found the fitting logic earlier | 13:05 |
| lxsli | so I'm really confused about this ImageMetaProps thing | 13:06 |
| *** jamielennox is now known as jamielennox|away | 13:06 | |
| *** eharney has joined #openstack-nova | 13:06 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: Enhance PCI passthrough whitelist to support regex https://review.openstack.org/199488 | 13:08 |
| *** browne has joined #openstack-nova | 13:08 | |
| *** davideagnello has quit IRC | 13:08 | |
| *** flwang1 has joined #openstack-nova | 13:08 | |
| bauzas | lxsli: the ImageMetaProps object comes from the fact that we're passing the image in the req_spec | 13:10 |
| bauzas | lxsli: and that's there is a filter reading it | 13:10 |
| *** ig0r_ has quit IRC | 13:10 | |
| lxsli | bauzas: damnit I got it now | 13:11 |
| *** fawadkhaliq has joined #openstack-nova | 13:11 | |
| lxsli | bauzas: the ImageMetaProps is just specifying how many CPUs per node, I read it wrong | 13:11 |
| *** markus_z has quit IRC | 13:12 | |
| lxsli | bauzas: in that case... -1 on requestspec, sorry. You're storing an InstanceNUMATopology and I think you need to store the # of cells per node instead | 13:13 |
| bauzas | lxsli: I don't get you again... | 13:13 |
| bauzas | lxsli: the InstanceNUMATopology object is coming from instance_props which is given in req_spec | 13:14 |
| bauzas | lxsli: please read from_primitives() before commenting... | 13:14 |
| *** exploreshaifali has joined #openstack-nova | 13:15 | |
| *** flwang1 has quit IRC | 13:15 | |
| *** MentalRay_ has quit IRC | 13:15 | |
| *** ig0r_ has joined #openstack-nova | 13:16 | |
| *** fawadkhaliq has quit IRC | 13:16 | |
| *** rushiagr_away is now known as rushiagr | 13:17 | |
| *** dims_ has quit IRC | 13:20 | |
| *** dims has joined #openstack-nova | 13:20 | |
| *** marzif_ has quit IRC | 13:21 | |
| *** marzif_ has joined #openstack-nova | 13:22 | |
| *** marzif_ has quit IRC | 13:22 | |
| *** cpschult has joined #openstack-nova | 13:22 | |
| *** dustins_ has joined #openstack-nova | 13:22 | |
| *** dustins has quit IRC | 13:22 | |
| *** dustins_ is now known as dustins | 13:22 | |
| *** sahid has quit IRC | 13:25 | |
| lxsli | bauzas: OK so it looks like from_primitives is passed a request_spec. It takes the instance_properties off that and passes to _from_instance, which pulls the numa_topology off and copies it to the RequestSpec instance | 13:26 |
| *** kbyrne has quit IRC | 13:26 | |
| lxsli | all of which is irrelevant because the basic point is that if you have an InstanceNUMATopology object, it's necessarily been fitted | 13:26 |
| *** jwcroppe has quit IRC | 13:26 | |
| bauzas | lxsli: yep, if you don't feel comfortable, please read the whole series before commenting | 13:26 |
| *** rfolco has joined #openstack-nova | 13:27 | |
| *** jwcroppe has joined #openstack-nova | 13:27 | |
| *** sahid has joined #openstack-nova | 13:27 | |
| bauzas | lxsli: I don't want to change the behaviour, I'm just changing the proxy | 13:27 |
| lxsli | unless you're going to tell me an InstanceNUMATopology is unfitted, I'm pretty comfortable | 13:27 |
| *** kbyrne has joined #openstack-nova | 13:27 | |
| bauzas | lxsli: seen how it's fitted? | 13:27 |
| bauzas | lxsli: see the HostManager patch | 13:27 |
| bauzas | lxsli: sounds like you're yakshaving honestly | 13:28 |
| lxsli | So you agree you're putting a fitted placement into the requestspec? | 13:28 |
| bauzas | noooooooooooooo | 13:28 |
| lxsli | Please don't accuse me of that before you understand my point :P | 13:28 |
| bauzas | lxsli: your point is maybe valid, but out of concern from the change | 13:28 |
| lxsli | Can you explain why? | 13:29 |
| *** sahid has quit IRC | 13:29 | |
| bauzas | lxsli: the change is only modifying the placeholder, not how the consumer | 13:29 |
| *** sahid has joined #openstack-nova | 13:29 | |
| bauzas | not *the* | 13:29 |
| bauzas | honestly, I prefer to work on fixing the existing comments before moving on vacation :) | 13:29 |
| *** jasondotstar has quit IRC | 13:30 | |
| lxsli | well if you don't want to discuss, I have to leave the -1 | 13:30 |
| bauzas | np | 13:30 |
| *** mpaolino has quit IRC | 13:30 | |
| lxsli | getting cake then ^^ | 13:30 |
| *** ig0r_ has quit IRC | 13:31 | |
| bauzas | lxsli: just before eating, could you please follow my advice and review my whole series ? | 13:31 |
| bauzas | lxsli: if so, you'd have noticed https://review.openstack.org/#/c/191251/20/nova/scheduler/host_manager.py,cm | 13:31 |
| bauzas | lxsli: and particularly L283 and follows | 13:32 |
| *** jwcroppe has quit IRC | 13:32 | |
| scheuran | mriedem1, are you around? | 13:35 |
| *** VW_ has joined #openstack-nova | 13:36 | |
| bauzas | ndipanov: around ? | 13:39 |
| ndipanov | yea | 13:40 |
| *** mjturek1 has joined #openstack-nova | 13:40 | |
| bauzas | ndipanov: you okay with me removing all numa the helpers of the hell in https://review.openstack.org/#/c/191251/20/nova/scheduler/host_manager.py,cm ? | 13:40 |
| bauzas | ndipanov: I thought it would have made you pleased | 13:40 |
| *** annashen has joined #openstack-nova | 13:40 | |
| bauzas | ndipanov: but apparently, I weren't too far | 13:41 |
| bauzas | :) | 13:41 |
| ndipanov | bauzas, yes I want those gone | 13:41 |
| bauzas | coolness, let the magic do | 13:41 |
| ndipanov | I think some of them can go away with your change | 13:41 |
| bauzas | I already said bye bye to one | 13:41 |
| ndipanov | and some of them will go away with PaulMurray's patches to objectify rt | 13:41 |
| ndipanov | has that landed btw | 13:42 |
| bauzas | ndipanov: yeah, that's the thing | 13:42 |
| bauzas | ndipanov: I can't remove all of them because of the host_state unpredictability | 13:42 |
| *** jecarey has joined #openstack-nova | 13:42 | |
| ndipanov | crap | 13:42 |
| *** mwagner_lap has joined #openstack-nova | 13:42 | |
| ndipanov | then leave it in | 13:42 |
| bauzas | ndipanov: that's what I did | 13:43 |
| bauzas | ndipanov: if you see the change, I removed the useless call to hardware.get_host_numa_usage_from_instance() since I was 100% sure that it was an object | 13:43 |
| *** annegentle has joined #openstack-nova | 13:43 | |
| bauzas | ndipanov: but since it was calling another one... | 13:43 |
| bauzas | ndipanov: so you'd suggest to keep hardware.get_host_numa_usage_from_instance ? | 13:44 |
| bauzas | ndipanov: and not do the copy/paste ? | 13:44 |
| bauzas | ndipanov: well, by thinking of that, makes sense | 13:44 |
| *** smatzek has joined #openstack-nova | 13:44 | |
| bauzas | ndipanov: should put a big note explaining the crap | 13:44 |
| *** annashen has quit IRC | 13:45 | |
| *** kylek3h has joined #openstack-nova | 13:45 | |
| *** zzzeek has joined #openstack-nova | 13:45 | |
| *** abhishekk has quit IRC | 13:46 | |
| mriedem1 | scheuran: i am now | 13:46 |
| openstackgerrit | garyk proposed openstack/nova: Compute: enable admin to decide action for conflicting states https://review.openstack.org/190047 | 13:47 |
| *** humble_ has joined #openstack-nova | 13:47 | |
| scheuran | mriedem1, cool. Good morning! | 13:47 |
| dims | mriedem1: ndipanov: nova cores - added a few new to the quick hit list - https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 13:47 |
| bauzas | dims: hola | 13:47 |
| *** lucas__ has joined #openstack-nova | 13:47 | |
| dims | bauzas: good afternoon | 13:48 |
| bauzas | dims: you didn't answered my question, why did you dumped the existing proposals , :) | 13:48 |
| bauzas | ? :) | 13:48 |
| scheuran | mriedem1, just wanted to clarify the exact requirements to get my nova code merged | 13:48 |
| scheuran | mriedem1, so talking about macvtap-vif | 13:48 |
| scheuran | mriedem1, is it sufficient to have the code somewhere out on github? | 13:48 |
| *** humble__ has quit IRC | 13:48 | |
| *** humble__ has joined #openstack-nova | 13:48 | |
| dims | bauzas: sorry i had not seen your question | 13:48 |
| bauzas | dims: np | 13:48 |
| bauzas | dims: again, not a big deal | 13:48 |
| bauzas | dims: just wondering why | 13:49 |
| dims | bauzas: i moved the ones with -1 to below | 13:49 |
| bauzas | dims: oh ok | 13:49 |
| dims | bauzas: and removed items that were already fixed | 13:49 |
| bauzas | dims: all of them ? huh | 13:49 |
| *** salv-orlando has joined #openstack-nova | 13:49 | |
| scheuran | mriedem1, there is a prototype available, I updated it again this morning to be compatible to the changed vif interface | 13:49 |
| bauzas | dims: okay, I appreciate the spring cleaning :) | 13:49 |
| bauzas | dims: I just thought some of the existing list were not doomed :) | 13:49 |
| mriedem1 | scheuran: link to the public code? | 13:49 |
| dims | bauzas: anything with either a jenkins -1 or -1 from someone is not clean | 13:49 |
| *** lykinsbd has joined #openstack-nova | 13:49 | |
| bauzas | dims: agree | 13:50 |
| scheuran | mriedem1 , https://github.com/scheuran/networking-macvtap/tree/networking-macvtap-poc | 13:50 |
| dims | bauzas: if you see something ready, just move them back please | 13:50 |
| scheuran | mriedem1, it's still in a separate branch | 13:50 |
| bauzas | dims: anyway, not a big deal :) | 13:50 |
| scheuran | mriedem1, (not in master) | 13:50 |
| *** lucas__ has quit IRC | 13:50 | |
| *** lucas__ has joined #openstack-nova | 13:50 | |
| dims | bauzas: just to be sure, the only things that i delete is stuff that is already merged | 13:50 |
| bauzas | dims: ack, no worries again | 13:51 |
| scheuran | mriedem1, it's a prototype but fully functional | 13:51 |
| bauzas | dims: I really appreciate you having time for that :) | 13:51 |
| mriedem1 | scheuran: ok, but i'm still a bit concerned about the direction with the neutron team | 13:51 |
| mriedem1 | and where this fits given the confusion over the governance change | 13:51 |
| mriedem1 | mestery said yesterday it'd be sorted out in the next neutron meeting | 13:51 |
| mriedem1 | or to start a ML thread on this | 13:52 |
| *** rotbeard has joined #openstack-nova | 13:52 | |
| *** humble_ has quit IRC | 13:52 | |
| dims | bauzas: with the freeze today, hoping cores will get time to check the list.. | 13:52 |
| bauzas | dims: bugs are not impacted by the freeze :) | 13:53 |
| bauzas | dims: that's why I deferred my quest for bugs until FF, ie. L3 :) | 13:53 |
| dims | bauzas: y, it's the question of time | 13:53 |
| *** jaypipes has joined #openstack-nova | 13:53 | |
| bauzas | dims: oh you mean post-freeze ? | 13:53 |
| bauzas | dims: sure | 13:53 |
| bauzas | dims: I thought you were expressing some urgency | 13:54 |
| bauzas | :) | 13:54 |
| scheuran | mriedem1, yes it already has been discussed - but then a new proposal came up which I had to evaluate. So my comeback would be in next weeks neutron meeting. | 13:54 |
| *** chuckcarmack has joined #openstack-nova | 13:54 | |
| scheuran | mriedem1, I was holding back the ML thread, to clarify with you first the exact issue | 13:54 |
| *** otter768 has joined #openstack-nova | 13:54 | |
| mriedem1 | danpb: what are your thoughts on this wrt https://review.openstack.org/#/c/182283/ ^ ? | 13:54 |
| mriedem1 | scheuran: i guess i'd like to see mestery or sc68cal from neutron +1 the nova change before we approve it | 13:55 |
| *** krtaylor has quit IRC | 13:55 | |
| *** annegentle has quit IRC | 13:55 | |
| mriedem1 | more or less as a way of saying they are ok with this general direction | 13:55 |
| mestery | mriedem1: link to nova change? | 13:55 |
| mriedem1 | https://review.openstack.org/#/c/182283/ | 13:55 |
| mestery | thanks mriedem1 | 13:55 |
| mriedem1 | scheuran: so if the macvtap stuff doesn't live as a separate repo/agent, how much of this code has to change? | 13:56 |
| mriedem1 | or you wouldn't know until that happens | 13:56 |
| mriedem1 | ? | 13:56 |
| scheuran | mriedem1, on nova side nothing would change | 13:56 |
| mriedem1 | ok, that helps | 13:57 |
| scheuran | mriedem1, the nova part just extracts the essential arguments out of the vif api response and generates the xml | 13:58 |
| *** otter768 has quit IRC | 13:59 | |
| dims | bauzas: i found one of yours in the -1 bucket that was ready, so moved it up :) | 13:59 |
| mestery | mriedem1 scheuran: Replied on both https://review.openstack.org/#/c/182283/ and https://review.openstack.org/#/c/195907/. | 14:00 |
| *** exploreshaifali has quit IRC | 14:00 | |
| *** haomaiwang has quit IRC | 14:01 | |
| bauzas | dims: no worries, just following that from very far | 14:01 |
| *** haomaiwang has joined #openstack-nova | 14:02 | |
| scheuran | mestery, thanks for this clear statement. | 14:03 |
| *** artom has joined #openstack-nova | 14:03 | |
| mestery | scheuran: yw :) | 14:03 |
| mriedem1 | mestery: thanks | 14:03 |
| mestery | mriedem1: Happy to help | 14:03 |
| *** agireud has joined #openstack-nova | 14:03 | |
| *** ctrath has joined #openstack-nova | 14:04 | |
| *** ankit_ag has quit IRC | 14:04 | |
| *** bfic has left #openstack-nova | 14:06 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/nova: Use urllib.urlencode instead of dict_to_query_str function https://review.openstack.org/207468 | 14:06 |
| *** markvoelker_ has quit IRC | 14:06 | |
| nagyz_ | mriedem1, if I move my rbd encryption patch to depend-on on your patch and put it nicely under a volume/rbd.py will that get your vote? :-) | 14:07 |
| mriedem1 | nagyz_: i haven't even looked at it yet | 14:07 |
| *** thangp has joined #openstack-nova | 14:07 | |
| nagyz_ | mriedem1, I just had to rebase it after some of the refactor work you've been doing ended up merged in | 14:07 |
| *** krtaylor has joined #openstack-nova | 14:07 | |
| nagyz_ | so that prompted my question | 14:07 |
| mriedem1 | nagyz_: today is feature freeze so those changes are probably going to get the most attention | 14:09 |
| mriedem1 | we have time for bug fixes later | 14:09 |
| nagyz_ | I hope nobody will start to argue that it's a new feature not a bigfix then :-) | 14:09 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Fix test_relationships() for subobject versions https://review.openstack.org/198904 | 14:10 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Change List objects to use obj_relationships https://review.openstack.org/198730 | 14:10 |
| rlrossit | dansmith: ^ | 14:10 |
| dansmith | rlrossit: I didn't even check.. what was the result of the run? | 14:11 |
| sdague | dansmith / mriedem1 : https://review.openstack.org/#/c/207403/ - actually make assertJsonEqual dump enough info to figure out why some of the unit tests are racey | 14:12 |
| *** ajayaa has joined #openstack-nova | 14:12 | |
| dansmith | rlrossit: hmm, no jenkins results on PS18? | 14:13 |
| sdague | dansmith: the check queue is > 500 | 14:13 |
| dansmith | sdague: this was yesterday at like 3pm my time | 14:14 |
| sdague | > 600 now | 14:14 |
| sdague | yep | 14:14 |
| *** achanda has joined #openstack-nova | 14:14 | |
| mriedem1 | sdague: yeah, just reviewed it | 14:14 |
| dansmith | longest thing in the queue I see is 5h | 14:14 |
| mriedem1 | sdague: +2 but added lifeless to see if he had any input | 14:14 |
| sdague | dansmith: there was another zuul crash and restart | 14:14 |
| dansmith | ah | 14:14 |
| sdague | so 5h is just from zuul restart | 14:14 |
| mriedem1 | it'd be nice if we could get assertJsonEqual up into testtools | 14:15 |
| dansmith | ffs | 14:15 |
| sdague | also, a lot of projects, including nova, have racey unit tests which is gumming up gate | 14:15 |
| *** Kennan2 has quit IRC | 14:16 | |
| sdague | so my guess is check is really that we're 24 - 36 hours queue time atm | 14:16 |
| *** Kennan has joined #openstack-nova | 14:16 | |
| *** achanda has quit IRC | 14:18 | |
| dansmith | oye | 14:18 |
| *** jecarey has quit IRC | 14:18 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: libvirt: don't add filesystem disk to parallels containers unconditionally https://review.openstack.org/186858 | 14:19 |
| sdague | yeh, well, so it is | 14:19 |
| mriedem1 | yeah, i've kind of given up on trying to keep tabs on a lot of the project functional test jobs | 14:19 |
| mriedem1 | http://status.openstack.org/elastic-recheck/data/uncategorized.html | 14:19 |
| mriedem1 | like neutron functional | 14:19 |
| scheuran | mriedem1, thanks Matt! - just got the mail that the patch got queued at the gate | 14:20 |
| sdague | mriedem1: yeh, the neutron functional job is pretty much a disaster from predicability | 14:20 |
| mriedem1 | not only that but the failures are not terribly easy to fingerprint since there are no service logs | 14:22 |
| mriedem1 | well, stepping back, neutron service logs.... :) | 14:22 |
| *** tpeoples has joined #openstack-nova | 14:23 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-novaclient: Refactor and update test_instances and test_volumes_api https://review.openstack.org/207355 | 14:24 |
| *** annegentle has joined #openstack-nova | 14:24 | |
| mriedem1 | sdague: mind slapping 1479780 into your change? | 14:25 |
| sdague | mriedem1: not if it means a +A :) | 14:25 |
| mriedem1 | you mean not if it means losing the +2s? | 14:26 |
| openstackgerrit | Sean Dague proposed openstack/nova: return more details on assertJsonEqual fail https://review.openstack.org/207403 | 14:26 |
| mriedem1 | ah | 14:26 |
| sdague | well, honestly, we can't make any progress with one of our major races until that lands | 14:26 |
| sdague | major unit test races, that is | 14:26 |
| sdague | you'll notice every uncategorized nova unit test race is that test, with this issue | 14:27 |
| *** MentalRay_ has joined #openstack-nova | 14:27 | |
| *** markvoelker has joined #openstack-nova | 14:27 | |
| *** nagyz_ has quit IRC | 14:27 | |
| mriedem1 | i'm running some logstash queries now | 14:27 |
| mriedem1 | 24 hits in 7 days | 14:28 |
| mriedem1 | a bit spiky in the last 24 hours | 14:28 |
| *** flwang1 has joined #openstack-nova | 14:28 | |
| mriedem1 | http://goo.gl/yB0I71 | 14:28 |
| *** markvoelker_ has joined #openstack-nova | 14:28 | |
| sdague | yeh, well last 24 hrs is just because there are basically always a ton of nova jobs running | 14:28 |
| *** eharney has quit IRC | 14:29 | |
| vladikr | dansmith, Hi, any chance you could look at the keypairs stuff again.. so it wont miss the deadline https://review.openstack.org/#/c/70485/ ? | 14:29 |
| *** tonytan4ever has joined #openstack-nova | 14:30 | |
| dansmith | vladikr: okay | 14:31 |
| vladikr | dansmith, thanks | 14:31 |
| *** ig0r_ has joined #openstack-nova | 14:31 | |
| *** ddaskal has joined #openstack-nova | 14:32 | |
| *** markvoelker has quit IRC | 14:32 | |
| *** Vikas___ has quit IRC | 14:33 | |
| lxsli | jaypipes: got a second? | 14:35 |
| mnestratov | mriedem1: good morning, wanted to let you know that I added volume test coverage https://review.openstack.org/#/c/186858/ | 14:35 |
| openstackgerrit | Jeffrey Zhang proposed openstack/nova: remove deprecated noauth auth strategy https://review.openstack.org/207486 | 14:35 |
| mriedem1 | rlrossit: ctrath: chuckcarmack: if you're looking for something to do, categorizing gate bugs is always helpful http://status.openstack.org/elastic-recheck/data/uncategorized.html | 14:35 |
| mriedem1 | http://docs.openstack.org/infra/elastic-recheck/readme.html | 14:36 |
| lxsli | jaypipes: I've been on a whole voyage this morning but I finally agree with Nikola, we should remove the "is not None" from L292 of https://review.openstack.org/#/c/128992/24/nova/objects/resources.py | 14:36 |
| mriedem1 | mnestratov: looking | 14:36 |
| *** artom has quit IRC | 14:36 | |
| mriedem1 | danpb: https://review.openstack.org/#/c/194412/ is an easy bp win | 14:37 |
| mriedem1 | danpb: and https://review.openstack.org/#/c/194454/ | 14:37 |
| ctrath | mriedem1: is there a doc that explains how to categorize gate bugs? | 14:37 |
| mriedem1 | libvirt volume drivers that talk to os-brick, just standard shims | 14:37 |
| mriedem1 | ctrath: yeah, see the doc link i just put up above | 14:38 |
| ctrath | oh.. didn't see that.. thanks | 14:38 |
| *** zsais has joined #openstack-nova | 14:39 | |
| *** kylek3h has quit IRC | 14:39 | |
| *** kylek3h has joined #openstack-nova | 14:39 | |
| *** Jeffrey4l has joined #openstack-nova | 14:41 | |
| *** masayukig has quit IRC | 14:41 | |
| *** Murali has quit IRC | 14:41 | |
| *** annashen has joined #openstack-nova | 14:41 | |
| *** masayukig has joined #openstack-nova | 14:42 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Split instance resize operation into 2 phases https://review.openstack.org/195088 | 14:42 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Create class hierarchy for tasks in conductor https://review.openstack.org/184495 | 14:42 |
| *** salv-orlando has quit IRC | 14:42 | |
| *** bapalm_ has joined #openstack-nova | 14:43 | |
| *** kylek3h has quit IRC | 14:43 | |
| danpb | mriedem1: we've created a nasty problem with rootwrap compute.filters with os-brick | 14:44 |
| danpb | because we've tied changes to os-brick to a need to update nova's compute.filters | 14:44 |
| *** eharney has joined #openstack-nova | 14:44 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-novaclient: Refactor and update test_instances and test_volumes_api https://review.openstack.org/207355 | 14:45 |
| danpb | we need to solve that by getting an include mechanism so compute.filters can just reference an os-brick.filters file | 14:45 |
| *** sahid has quit IRC | 14:45 | |
| mriedem1 | danpb: yes | 14:45 |
| danpb | so i'm not approving those extra vol drivers right now | 14:45 |
| danpb | as i think this is a major design flaw we need to solve first | 14:46 |
| mriedem1 | danpb: well, we've already opened pandora's box there with the os-brick integration change in nova | 14:46 |
| *** annashen has quit IRC | 14:46 | |
| mnestratov | danpb: mriedem1: review https://review.openstack.org/#/c/202442/ lost your +2 and +A because gate rejected it. Could you please take a look once again? | 14:46 |
| *** tpeoples has quit IRC | 14:48 | |
| mriedem1 | ndipanov: i think you should review this https://review.openstack.org/#/c/186858/ | 14:49 |
| mriedem1 | i'm lost in the test cases | 14:49 |
| *** mtanino has joined #openstack-nova | 14:50 | |
| jaypipes | lxsli, ndipanov: looking... and reading. | 14:50 |
| *** kylek3h has joined #openstack-nova | 14:50 | |
| *** achanda has joined #openstack-nova | 14:51 | |
| rlrossit | dansmith: uhhh ok I'm confused.... I could've sworn when I looked last night Jenkins gave it a thumbs up... regardless, I added a unit test in there to catch my mistake so I think we're good | 14:51 |
| dansmith | rlrossit: okay | 14:52 |
| *** kylek3h_ has joined #openstack-nova | 14:54 | |
| *** kylek3h has quit IRC | 14:54 | |
| *** Jeffrey4l has quit IRC | 14:55 | |
| *** afazekas has quit IRC | 14:55 | |
| *** Piet_ has joined #openstack-nova | 14:57 | |
| dansmith | jaypipes: https://review.openstack.org/#/c/70485/21 <-- I believe he got all your feedback addressed | 14:57 |
| jaypipes | lxsli, ndipanov: so, in the original design of the resource object's can_provide() and consume() interface, the can_provide() method was deliberately NOT a reservation. It was a *check*. The consume() method is the reservation. | 14:57 |
| xyang | mriedem1: danpb , do we have a problem with the libvirt volume drivers now? Just saw danpb's comments | 14:57 |
| *** tpeoples has joined #openstack-nova | 14:57 | |
| jaypipes | dansmith: k, reviewing now. | 14:57 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: rootwrap: remove obsolete filters for baremetal https://review.openstack.org/207492 | 14:57 |
| *** BobBall is now known as BobBall_AWOL | 14:58 | |
| lxsli | jaypipes: we can do it that way too but then we must refit in consume | 14:58 |
| dansmith | jaypipes: also, when you're done we need to circle back on sudipto's thing, so just ping me when you have a few minutes to chat | 14:58 |
| *** masayukig has quit IRC | 14:59 | |
| lxsli | jaypipes: I think numa_usage_from_instances requires a fitted InstanceNUMATopology. We also need to return the fit from consume | 14:59 |
| *** burt has joined #openstack-nova | 14:59 | |
| *** masayukig has joined #openstack-nova | 15:00 | |
| danpb | mriedem1: so looking at it, it should just require two things to fix the rootwrap mess | 15:00 |
| danpb | mriedem1: 1. os-brick git tree should contain rootwrap filters for stuff it needs to run | 15:00 |
| *** rmcall has joined #openstack-nova | 15:00 | |
| danpb | mriedem1: 2. devstack should copy the os-brickk rootwrap filters into /etc/nova/rootwrap.d | 15:00 |
| *** haomaiwang has quit IRC | 15:01 | |
| *** krtaylor has quit IRC | 15:01 | |
| mriedem1 | and /etc/cinder/rootwrap.d | 15:01 |
| mriedem1 | since cinder is in the same boat | 15:01 |
| danpb | yep, | 15:02 |
| lxsli | I'm basing that on line 1112 of hardware.py which looks at the instancecell ID, if it's unfitted I don't think that's set | 15:02 |
| mriedem1 | note that multipathd was added here https://review.openstack.org/#/c/175569/32/etc/nova/rootwrap.d/compute.filters | 15:02 |
| jaypipes | lxsli: we *do* refit in consume()... it calls hw.numa_usage_from_instances(...) | 15:02 |
| *** haomaiwang has joined #openstack-nova | 15:02 | |
| danpb | that's even more reason to actually have them maintained in os-brick itself | 15:02 |
| *** angdraug has joined #openstack-nova | 15:02 | |
| danpb | as we don't want to duplicate those lines in cinder and nova | 15:02 |
| mriedem1 | yes, i agree with that | 15:02 |
| mriedem1 | i guess the question is, can we handle this as a bug fix after feature freeze? | 15:02 |
| mriedem1 | since it has to happen regardless for multipathd | 15:02 |
| *** ig0r__ has joined #openstack-nova | 15:03 | |
| *** ig0r_ has quit IRC | 15:03 | |
| jaypipes | lxsli: the thing that is missing is actually for the consume() methods of IntegerInventory and NUMAInventory to raise something like exc.NoMoreResources() if there is a race between can_provide() and consume() that causes the total used to exceed the available. | 15:03 |
| danpb | when is feature freeze | 15:03 |
| mriedem1 | danpb: today | 15:03 |
| jaypipes | lxsli: I will add that functionality right now... gimme a sec. | 15:03 |
| *** eglynn has joined #openstack-nova | 15:03 | |
| *** rmcall has quit IRC | 15:03 | |
| mriedem1 | danpb: and i'd consider the bug fix a blocker for liberty | 15:03 |
| *** achanda has quit IRC | 15:03 | |
| danpb | yeah, if we agree that this is a release blocker for liberty i'd be ok with it | 15:04 |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-novaclient: Update pick_image function https://review.openstack.org/207497 | 15:05 |
| bauzas | bah, zuul is in the weeds, I have a 17+ patch series to upload and I'm leaving in 2 hours | 15:05 |
| bauzas | smart | 15:05 |
| mriedem1 | danpb: ok - added the same comments in the scaleio change | 15:05 |
| mriedem1 | danpb: i can get hemnafk caught up on that later too, and open the bug now | 15:05 |
| *** zsais has quit IRC | 15:05 | |
| danpb | yeah, guess we need bugs against os-brick, nova and cinder | 15:05 |
| danpb | and devstack | 15:05 |
| earlephilhower | mrieden1: Can you add same comment on the hgst drvr, too? | 15:06 |
| openstackgerrit | Marian Horban proposed openstack/nova: Added newline at end of files https://review.openstack.org/207498 | 15:07 |
| earlephilhower | mriedem1: (sorry typo, no coffee yet) Can you add same comment on the hgst drvr, too? | 15:07 |
| xyang | danpb: mriedem1 , I can help hemnafk with those changes | 15:07 |
| dansmith | bauzas: you've got a couple -1s on your base patch.. | 15:08 |
| *** ig0r_ has joined #openstack-nova | 15:08 | |
| bauzas | dansmith: yup, hence my attempt to upload that before tonight :) | 15:08 |
| dansmith | bauzas: okay | 15:08 |
| *** logan2 has quit IRC | 15:08 | |
| *** ig0r__ has quit IRC | 15:09 | |
| ctrath | mriedem: so, I'm trying to understand the elasticrecheck stuff… I'm looking at http://logs.openstack.org/49/206249/3/gate/gate-nova-python27/00e9a9d/console.html#_2015-07-30_09_56_39_827 | 15:09 |
| mriedem1 | danpb: https://bugs.launchpad.net/nova/+bug/1479842 | 15:09 |
| openstack | Launchpad bug 1479842 in OpenStack Compute (nova) "os-brick needs to provide it's own rootwrap filters file" [Undecided,New] | 15:09 |
| lxsli | jaypipes: sorry I'm sure I'm misunderstanding something but how does the instancecell.id get set by numa_usage_from_instances please? | 15:09 |
| ctrath | do I create a new bug and filter for this | 15:09 |
| *** MentalRay__ has joined #openstack-nova | 15:09 | |
| ctrath | mriedem1: ^ | 15:10 |
| lxsli | jaypipes: I can see that being set in _numa_fit_instance_cell which is called by numa_fit_instance_to_host, but numa_fit_instance_to_host isn't called by numa_usage_from_instances and I don't see anything else which would set the instancecell.id | 15:10 |
| danpb | mriedem1: ok, i approved the changes and added that bug as a comment | 15:10 |
| lxsli | I'm looking at lines 1112 and 803 in virt/hardware.py | 15:11 |
| *** fawadkhaliq has joined #openstack-nova | 15:11 | |
| ndipanov | lxsli, I think it only gets set by fit_instance_... | 15:11 |
| *** MentalRay_ has quit IRC | 15:11 | |
| xyang | danpb: mriedem1 thanks! | 15:11 |
| ndipanov | that's the only thing that really should be setting that and cpu pinning iformation | 15:12 |
| *** ig0r__ has joined #openstack-nova | 15:12 | |
| lxsli | ndipanov: OK so numa_usage_from_instances doesn't do fitting? | 15:12 |
| ndipanov | nope | 15:12 |
| ndipanov | it expects it to be set | 15:12 |
| mriedem1 | ctrath: already done for that one https://review.openstack.org/#/c/207484/ | 15:12 |
| lxsli | then we have a bug, consume should also call numa_fit_instance_to_host | 15:12 |
| lxsli | jaypipes: ^^ | 15:12 |
| ndipanov | lxsli, well not necessarily | 15:13 |
| * lxsli sobs quietly. | 15:13 | |
| ndipanov | lxsli, well I mean | 15:13 |
| ndipanov | yeah... | 15:13 |
| *** krtaylor has joined #openstack-nova | 15:14 | |
| lxsli | I'm actually laughing, just been through the wringer on this today | 15:14 |
| ctrath | mriedem: so, how do I know if there was a filter already created? | 15:14 |
| ctrath | mriedem1: | 15:14 |
| *** unicell has quit IRC | 15:14 | |
| *** ig0r__ has quit IRC | 15:15 | |
| jaypipes | lxsli: I really don't like the idea of coupling the resource consumption of the host NUMA topology with the fitting of a particular instance NUMA topology to one or more host NUMA cells. | 15:15 |
| *** ig0r_ has quit IRC | 15:15 | |
| ndipanov | jaypipes, and they need not be coupled | 15:15 |
| ndipanov | lxsli, https://github.com/openstack/nova/blob/master/nova/compute/claims.py#L201-L230 | 15:15 |
| jaypipes | lxsli: because that is not how the host NUMA topology is stored in the database. | 15:15 |
| ndipanov | this does only fitting and stashes the topology it claimed | 15:16 |
| jaypipes | lxsli: the host numa topology does not store the instance cell to host cell ID mapping at all. | 15:16 |
| ndipanov | so it can be consumed later from the instance | 15:16 |
| jaypipes | ndipanov: right, which is why I don't think this is something we need to change about the patch proposed. | 15:16 |
| ndipanov | jaypipes, which is why I +2ed it | 15:16 |
| ndipanov | although I do think that eventually you will want to add a fitting call on the Inventory | 15:16 |
| *** busterswt has joined #openstack-nova | 15:17 | |
| lxsli | jaypipes: you want the consume method to be the reservation though right? | 15:17 |
| *** diogogmt has joined #openstack-nova | 15:17 | |
| *** dannywilson has joined #openstack-nova | 15:17 | |
| jaypipes | ndipanov: perhaps, when claims are pulled into the scheduler... | 15:17 |
| *** jcoufal has quit IRC | 15:17 | |
| jaypipes | lxsli: the reservations of the resource != the fitting of one instance to a particular cell. | 15:18 |
| lxsli | jaypipes: and you can't reserve without fitting first, and you don't want a gap between fitting and claiming | 15:18 |
| dansmith | bauzas: are you going to s/instance_type/flavor/ ? | 15:18 |
| jaypipes | lxsli: yes you can. | 15:18 |
| lxsli | jaypipes: aha | 15:18 |
| bauzas | dansmith: that's something I was waiting you to be awake :) | 15:18 |
| bauzas | and other US peersq | 15:18 |
| bauzas | hence my comment in the review | 15:18 |
| bauzas | so | 15:18 |
| bauzas | flavor vs. instance_type, who wins ? | 15:18 |
| jaypipes | instance_type. | 15:18 |
| dansmith | bauzas: personally, I prefer flavor, and we previously had a bp to convert as many as possible (mriedem did that work). So I'd prefer you be consistent | 15:18 |
| dansmith | but it makes jaypipes mad | 15:19 |
| lxsli | jaypipes: well... that brings me back to how instancecell.id gets set | 15:19 |
| dansmith | which may be a bonus, I dunno | 15:19 |
| bauzas | tbh, I don't care, I just want to be consistent :) | 15:19 |
| bauzas | and I would like to avoid some field duplication with upgrade path | 15:19 |
| bauzas | for any possible renaming | 15:19 |
| lxsli | jaypipes: because numa_usage_from_instances relies on that being set to consume from the host_topology and ndipanov+I seem to agree it only gets set when an InstanceNUMATopology is fitted | 15:19 |
| bauzas | so, flavor ? | 15:19 |
| jaypipes | dansmith: lots of things make me mad. I eventually submit to your allmighty omniscience, though. | 15:19 |
| mriedem1 | ctrath: if there is already an e-r query, then the thing isn't in the uncategorized bugs page | 15:20 |
| dansmith | jaypipes: <3 | 15:20 |
| mriedem1 | ctrath: b/c the e-r query categorizes the bug | 15:20 |
| jaypipes | lxsli, ndipanov: unfortunately, I need to speak at a town hall meeting inside mirantis in about 8 minutes... may we talk abotu this in a little while (maybe 45 mintes? | 15:20 |
| bauzas | what is the right phrase to say in that circumstance? | 15:20 |
| mriedem1 | ah yes, instance_type -> flavor | 15:20 |
| mriedem1 | that was a real fun one that was half baked | 15:21 |
| lxsli | jaypipes: sure thing | 15:21 |
| *** vishwanathj has quit IRC | 15:21 | |
| dansmith | jaypipes: "I need to speak at a town-hall meeting" is not a thing I'll ever get to say | 15:21 |
| jaypipes | dansmith: :P | 15:21 |
| *** dannywil_ has joined #openstack-nova | 15:22 | |
| *** dannywil_ has quit IRC | 15:22 | |
| ctrath | mriedem1: that doesn't solve the issue that I was asking.. The bug that I was looking at was uncategorized | 15:22 |
| bauzas | oh, got the english translation | 15:22 |
| *** dannywil_ has joined #openstack-nova | 15:22 | |
| mriedem1 | ctrath: because the e-r query change wasn't merged yet, | 15:22 |
| bauzas | Speak now or forever hold your peace ! | 15:22 |
| bauzas | so I count one strong opinion about flavor | 15:23 |
| mriedem1 | and there is a window of time between when new queries are merged and the uncategorized bugs page is refreshed | 15:23 |
| *** dannywi__ has joined #openstack-nova | 15:23 | |
| ctrath | mriedem1: right, so, how do I know if somebody already looked at the bug and it hasn't been merged? | 15:23 |
| bauzas | okay, given the result, long live flavor | 15:23 |
| *** dannywi__ has quit IRC | 15:24 | |
| mriedem1 | ctrath: well, there is the open review dashboard for e-r https://review.openstack.org/#/q/status:open+project:openstack-infra/elastic-recheck,n,z | 15:24 |
| *** dannywilson has quit IRC | 15:24 | |
| mnestratov | ndipanov: hi. could you please take a look https://review.openstack.org/#/c/186858/ ? | 15:24 |
| mriedem1 | ctrath: the workflow is generally finding a decent representative error in the logs and then plugging that into logstash to see how it hits (100% failure in the gate on multiple change sets) | 15:25 |
| mriedem1 | then find an existing bug or open a new one | 15:25 |
| ctrath | ok | 15:25 |
| mriedem1 | ctrath: sometimes they are just merge failures in the gate and we don't really care about those, like on unit test jobs | 15:26 |
| *** dannywi__ has joined #openstack-nova | 15:26 | |
| *** dannywil_ has quit IRC | 15:27 | |
| *** moshele has quit IRC | 15:27 | |
| mriedem1 | ctrath: unit test jobs also suck because a lot of times they aren't 100% failures due to negative tests | 15:28 |
| mriedem1 | e.g. http://logs.openstack.org/51/205551/3/gate/gate-neutron-dsvm-functional/4de39db/console.html#_2015-07-30_13_36_30_757 is a failure, but the error is not 100% failure http://goo.gl/uhcJO3 | 15:29 |
| *** aweeks has quit IRC | 15:29 | |
| *** angular_mike_ has quit IRC | 15:29 | |
| *** BadCub has quit IRC | 15:29 | |
| mriedem1 | probably b/c it also shows up in passing tests that are doing negative scenarios | 15:29 |
| mriedem1 | if you click on the build_status link in kibana it shows that hits on 75% successful runs | 15:30 |
| mriedem1 | so it's not reliable | 15:30 |
| bauzas | dansmith: given your review on https://review.openstack.org/#/c/145528/37 is there a specific rule for hydrating an object orphaned or not ? | 15:30 |
| ctrath | gotcha | 15:30 |
| dansmith | bauzas: you mean about context? | 15:31 |
| bauzas | dansmith: I mean, I thought that since we were having a context, it'd be good to make it non-orphaned, but sounds like it's better not | 15:31 |
| bauzas | dansmith: yup | 15:31 |
| *** zul has quit IRC | 15:31 | |
| dansmith | bauzas: well, it just depends, like I said, on whether you want it to be able to lazy-load things | 15:31 |
| bauzas | dansmith: right, I thought it was an opportunity | 15:31 |
| dansmith | bauzas: if you don't, then setting a context will make a lazy-load call to conductor, which may succeed or fail depending | 15:31 |
| bauzas | dansmith: I don't have a specific opinion on that, was just opportunity | 15:32 |
| dansmith | bauzas: I don't *think* there is any reason you'd need to lazy-load things out of any of those sub-objects, right? | 15:32 |
| bauzas | dansmith: nope, don't think so too | 15:32 |
| dansmith | either way, I don't think this case demands a strong decision either way | 15:32 |
| *** dannywi__ has quit IRC | 15:32 | |
| bauzas | dansmith: yeah... | 15:32 |
| bauzas | dansmith: let's remove the context | 15:32 |
| dansmith | I was just defending you against the evil melwitt :D | 15:32 |
| bauzas | hah | 15:32 |
| bauzas | no, but melwitt's point was fair, I wasn't consistent | 15:33 |
| dansmith | yep, definitely | 15:33 |
| dansmith | she'll catch you every time | 15:33 |
| bauzas | :) | 15:33 |
| *** zsais has joined #openstack-nova | 15:33 | |
| bauzas | okay, removing the context, it will enforce the idea that I hydrated the object by hand and could probably miss some unset attrs | 15:34 |
| *** dannywilson has joined #openstack-nova | 15:34 | |
| mriedem1 | danpb: have you seen this? http://logs.openstack.org/46/191046/12/gate/gate-tempest-dsvm-neutron-full/2a7910f/logs/screen-n-cpu.txt.gz?level=TRACE | 15:34 |
| dansmith | bauzas: yep | 15:34 |
| bauzas | I like the idea to hydrate the context only if we get that from either a DB or from another object correctly hydrated | 15:34 |
| *** hshiina has quit IRC | 15:34 | |
| dansmith | bauzas: no context will make those pop, instead of either silently lazy-loading (expensive) or failing after burning RPC time | 15:35 |
| *** dannywilson has quit IRC | 15:35 | |
| bauzas | yup | 15:35 |
| bauzas | zactly | 15:35 |
| *** rotbeard has quit IRC | 15:35 | |
| bauzas | I'd prefer an Exception now than a timeout | 15:35 |
| bauzas | or using an expensive RPC call | 15:35 |
| *** dannywilson has joined #openstack-nova | 15:36 | |
| dansmith | a timeout in unit tests, yeah | 15:36 |
| bauzas | (which could be a layer isolation violation, since the scheduler doesn't target to call the DB anyway) | 15:36 |
| bauzas | for flavors and other objects | 15:36 |
| bauzas | ok | 15:36 |
| bauzas | moving on | 15:36 |
| dansmith | oh, this is in the scheduler | 15:38 |
| *** sahid has joined #openstack-nova | 15:38 | |
| dansmith | yeah, so .. a bit weirder, but.. consistency over anything, and probably no context anyway | 15:38 |
| *** ijw has joined #openstack-nova | 15:39 | |
| mriedem1 | sahid: so on https://review.openstack.org/#/c/191046/12/nova/virt/libvirt/host.py - that looks totally busted http://logs.openstack.org/46/191046/12/gate/gate-tempest-dsvm-neutron-full/2a7910f/logs/screen-n-cpu.txt.gz?level=TRACE | 15:41 |
| *** jasondotstar has joined #openstack-nova | 15:42 | |
| *** annashen has joined #openstack-nova | 15:42 | |
| *** matrohon has quit IRC | 15:44 | |
| *** hemnafk is now known as hemna | 15:45 | |
| *** annashen has quit IRC | 15:47 | |
| *** alexpilotti has quit IRC | 15:48 | |
| *** obondarev has quit IRC | 15:48 | |
| *** obondarev has joined #openstack-nova | 15:49 | |
| openstackgerrit | Dmitry Guryanov proposed openstack/nova: libvirt: add vzstorage volumes suppport https://review.openstack.org/190843 | 15:49 |
| sahid | mriedem1: responded to you, this as nothing to do with libvirt, it's a mistake on the code i should to use libvirt_driver | 15:49 |
| sahid | btw good catch | 15:50 |
| *** annashen has joined #openstack-nova | 15:50 | |
| mriedem1 | well, the gate jobs caught it | 15:50 |
| mriedem1 | i'm surprised it actually got to the gate queue | 15:50 |
| *** unicell has joined #openstack-nova | 15:50 | |
| sahid | it's wierd no unit tests are using this pass | 15:51 |
| *** tdurakov has quit IRC | 15:51 | |
| sahid | s/pass/path | 15:52 |
| *** tdurakov has joined #openstack-nova | 15:52 | |
| *** ig0r_ has joined #openstack-nova | 15:52 | |
| *** franz_geiser has joined #openstack-nova | 15:53 | |
| franz_geiser | hi all, i have a question of a fresh install on centos 7, can anyone help me please? | 15:54 |
| *** otter768 has joined #openstack-nova | 15:55 | |
| franz_geiser | After telemetry module install, openstack-nova-api and openstack-nova-cert are not starting | 15:55 |
| *** ramishra has quit IRC | 15:55 | |
| *** DuncanT has quit IRC | 15:55 | |
| *** h00327910__ has quit IRC | 15:55 | |
| *** marcusvrn has quit IRC | 15:55 | |
| *** jswarren has joined #openstack-nova | 15:58 | |
| *** lucas__ has quit IRC | 15:59 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Prepare for unicode enums from Oslo https://review.openstack.org/207087 | 15:59 |
| *** otter768 has quit IRC | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwa_ has joined #openstack-nova | 16:02 | |
| *** vishwanathj has joined #openstack-nova | 16:04 | |
| *** thumpba has joined #openstack-nova | 16:04 | |
| *** vishwana_ has joined #openstack-nova | 16:05 | |
| *** salv-orl_ has joined #openstack-nova | 16:06 | |
| *** jlanoux has quit IRC | 16:08 | |
| *** vishwanathj has quit IRC | 16:08 | |
| *** lsmola has quit IRC | 16:09 | |
| *** zul has joined #openstack-nova | 16:10 | |
| openstackgerrit | Kyrylo Romanenko proposed openstack/python-novaclient: Specify NIC option for nova boot https://review.openstack.org/207523 | 16:11 |
| dansmith | this is the top longest-waiting thing that is ready to go on the list: https://review.openstack.org/#/c/70485/ | 16:12 |
| dansmith | according to L200 on https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 16:12 |
| dansmith | has one +2 needs +W | 16:12 |
| *** scheuran has quit IRC | 16:12 | |
| *** exploreshaifali has joined #openstack-nova | 16:14 | |
| *** davideagnello has joined #openstack-nova | 16:14 | |
| *** zul has quit IRC | 16:16 | |
| *** franz_geiser has quit IRC | 16:17 | |
| *** leeantho has joined #openstack-nova | 16:17 | |
| *** logan2 has joined #openstack-nova | 16:19 | |
| *** alex_klimov has quit IRC | 16:22 | |
| *** lpetrut has quit IRC | 16:22 | |
| *** jasondotstar has quit IRC | 16:23 | |
| *** jistr has quit IRC | 16:24 | |
| *** marcusvrn has joined #openstack-nova | 16:26 | |
| *** aweeks has joined #openstack-nova | 16:26 | |
| *** zul has joined #openstack-nova | 16:28 | |
| *** ip0353 has quit IRC | 16:28 | |
| openstackgerrit | xin wu proposed openstack/nova: bug 1478938, improve exception message https://review.openstack.org/207533 | 16:29 |
| openstack | bug 1478938 in OpenStack Compute (nova) "Maximum number of ports exceeded log" [Undecided,New] https://launchpad.net/bugs/1478938 | 16:29 |
| *** claudiub has quit IRC | 16:31 | |
| *** aix has quit IRC | 16:31 | |
| *** BadCub has joined #openstack-nova | 16:31 | |
| *** sahid has quit IRC | 16:31 | |
| *** deepthi has joined #openstack-nova | 16:32 | |
| *** josecastroleon has quit IRC | 16:32 | |
| *** rlrossit has left #openstack-nova | 16:32 | |
| *** dims has quit IRC | 16:33 | |
| *** zul has quit IRC | 16:33 | |
| *** chuckcarmack has left #openstack-nova | 16:34 | |
| *** angular_mike_ has joined #openstack-nova | 16:36 | |
| *** salv-orl_ has quit IRC | 16:39 | |
| *** smatzek has quit IRC | 16:41 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: objects: add hw_vif_multiqueue_enabled image property https://review.openstack.org/202168 | 16:42 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: change assertEqual args to expected, observed for VIF https://review.openstack.org/204190 | 16:42 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: enable virtio-net multiqueue https://review.openstack.org/128829 | 16:42 |
| *** zul has joined #openstack-nova | 16:42 | |
| *** ramishra has joined #openstack-nova | 16:43 | |
| *** DuncanT has joined #openstack-nova | 16:43 | |
| *** annashen_ has joined #openstack-nova | 16:43 | |
| *** READ10 has joined #openstack-nova | 16:43 | |
| *** deepthi has quit IRC | 16:45 | |
| *** alexpilotti has joined #openstack-nova | 16:46 | |
| mriedem1 | danpb: you might want to jump in the -cinder channel where we're discussing the os-brick rootwrap thing | 16:47 |
| *** annashen_ has quit IRC | 16:47 | |
| *** doude has quit IRC | 16:49 | |
| *** doude has joined #openstack-nova | 16:49 | |
| *** doude has quit IRC | 16:49 | |
| *** doude has joined #openstack-nova | 16:49 | |
| *** lpetrut has joined #openstack-nova | 16:50 | |
| *** kylek3h_ has quit IRC | 16:52 | |
| *** lucas__ has joined #openstack-nova | 16:52 | |
| *** kmARC has quit IRC | 16:52 | |
| *** mpavone has quit IRC | 16:53 | |
| *** rdopiera has quit IRC | 16:55 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: libvirt: make snapshot use RBD snapshot/clone when available https://review.openstack.org/205282 | 16:55 |
| *** dims has joined #openstack-nova | 16:57 | |
| *** nic has joined #openstack-nova | 16:57 | |
| openstackgerrit | Anthony Lee proposed openstack/nova: Fix live-migrations usage of the wrong connector information https://review.openstack.org/202770 | 16:58 |
| *** belmoreira has quit IRC | 16:58 | |
| *** pixelbeat_ has quit IRC | 16:58 | |
| *** Nikolay_St has quit IRC | 16:58 | |
| *** lpetrut has quit IRC | 16:59 | |
| mnestratov | danpb: could you please look again https://review.openstack.org/#/c/186858/ | 16:59 |
| *** vishwana_ has quit IRC | 17:00 | |
| *** garyk has quit IRC | 17:00 | |
| *** haomaiwa_ has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-nova | 17:02 | |
| *** angdraug has quit IRC | 17:02 | |
| *** lucasagomes is now known as lucas-dinner | 17:04 | |
| *** meghal has joined #openstack-nova | 17:06 | |
| *** busterswt has quit IRC | 17:08 | |
| *** h00327910__ has joined #openstack-nova | 17:09 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: libvirt: don't add filesystem disk to parallels containers unconditionally https://review.openstack.org/186858 | 17:09 |
| *** danpb has quit IRC | 17:10 | |
| *** vilobhmm has joined #openstack-nova | 17:16 | |
| *** artom has joined #openstack-nova | 17:16 | |
| *** raildo has quit IRC | 17:18 | |
| *** lpetrut has joined #openstack-nova | 17:18 | |
| *** raildo has joined #openstack-nova | 17:18 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests() https://review.openstack.org/199205 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Create RequestSpec object https://review.openstack.org/145528 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Consider that all scheduler calls are IO Ops https://review.openstack.org/188486 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler https://review.openstack.org/188487 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec https://review.openstack.org/202678 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object https://review.openstack.org/197912 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object https://review.openstack.org/203078 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec https://review.openstack.org/202676 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler https://review.openstack.org/199051 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec https://review.openstack.org/202677 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec https://review.openstack.org/202674 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec https://review.openstack.org/202675 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add RequestSpec methods for primitiving into dicts https://review.openstack.org/187197 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec https://review.openstack.org/202673 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in HostManager https://review.openstack.org/191251 | 17:20 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use the same pci_requests field for all filters and HostManager https://review.openstack.org/191250 | 17:20 |
| *** kmARC has joined #openstack-nova | 17:21 | |
| *** zsais has quit IRC | 17:21 | |
| bauzas | alaski: you should be unhappy with ^ | 17:22 |
| bauzas | alaski: s/instance_type/flavor | 17:22 |
| bauzas | ask dansmith if you are angry :p | 17:22 |
| dansmith | what? alaski prefers instance_type? | 17:23 |
| bauzas | dansmith: 2 hours for doing s/instance_type/flavor eh eh | 17:23 |
| bauzas | dansmith: no, he's just working on persisting the Spec object | 17:23 |
| dansmith | ah | 17:23 |
| bauzas | dansmith: so I guess he would have to rework on its DB calls :) | 17:23 |
| bauzas | his | 17:24 |
| bauzas | even | 17:24 |
| *** bapalm_ has quit IRC | 17:25 | |
| *** achanda has joined #openstack-nova | 17:25 | |
| *** smatzek has joined #openstack-nova | 17:25 | |
| *** achanda has quit IRC | 17:26 | |
| *** dims has quit IRC | 17:26 | |
| *** annashen has quit IRC | 17:28 | |
| *** achanda has joined #openstack-nova | 17:29 | |
| *** lpetrut has quit IRC | 17:30 | |
| *** jecarey has joined #openstack-nova | 17:30 | |
| *** meghal has quit IRC | 17:31 | |
| *** meghal has joined #openstack-nova | 17:31 | |
| *** browne has quit IRC | 17:33 | |
| *** ajayaa has quit IRC | 17:33 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add libvirt-python to test-requirements.txt to enforce min version https://review.openstack.org/183706 | 17:34 |
| *** erlon has joined #openstack-nova | 17:35 | |
| alaski | bauzas: eh, that's an easy change at least | 17:36 |
| bauzas | alaski: speak of you then :p | 17:36 |
| *** ssurana has joined #openstack-nova | 17:36 | |
| bauzas | alaski: I had to review all my 17 changes to see where I was checking it :) | 17:37 |
| *** boris-42 has joined #openstack-nova | 17:37 | |
| alaski | yeah, I meant for me :) just a couple of renames | 17:37 |
| bauzas | alaski: nah, jk | 17:37 |
| bauzas | alaski: it was a boring but necessary stuff | 17:37 |
| alaski | dansmith: I have no preference really, but agree that we should stick to flavor for consistency | 17:38 |
| dansmith | okay | 17:38 |
| dansmith | I care about consistency more than anything, but I feel like the previous blueprints and efforts have us headed in the direction of standardizing on flavor | 17:39 |
| dansmith | whether starting down that path was a good idea or not is certainly debatable | 17:39 |
| alaski | yep. as part of moving flavors from the cell to the api db we're renaming the tables to flavor* | 17:41 |
| bauzas | tbh, I'm okay with providing a 'flavor' field, because if not, it would be difficult to change the field name later | 17:41 |
| bauzas | it's quite difficult to provide an upgrade path for just renamings | 17:42 |
| *** chuckcarmack has joined #openstack-nova | 17:42 | |
| bauzas | so, since flavor sounds the better option, I prefer to take 2 hours on my own and fix my series | 17:42 |
| bauzas | even if jaypipes could be grumbling :p | 17:42 |
| *** annashen has joined #openstack-nova | 17:44 | |
| *** Sukhdev_ has joined #openstack-nova | 17:44 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Fix DeprecationWarning when using BaseException.message https://review.openstack.org/207560 | 17:46 |
| *** cfriesen has joined #openstack-nova | 17:47 | |
| cfriesen | ndipanov: I see in numa_get_constraints() that we currently set the pagesize the same for all cells. what's the reason for this limitation? | 17:49 |
| *** annashen has quit IRC | 17:49 | |
| ndipanov | cfriesen, I assume that we don't really have a way to expose it to the user | 17:51 |
| ndipanov | not sure there is any other reason deeper than - we didn't implement it because we needed to have resonable scope :) | 17:52 |
| *** tonytan4ever has quit IRC | 17:52 | |
| ndipanov | why would that be useful though? | 17:52 |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Fix the usage output of the nova-idmapshift command https://review.openstack.org/207562 | 17:53 |
| *** meghal has quit IRC | 17:55 | |
| *** annashen has joined #openstack-nova | 17:56 | |
| *** otter768 has joined #openstack-nova | 17:56 | |
| *** igordcard has joined #openstack-nova | 17:56 | |
| *** alexpilotti has quit IRC | 17:56 | |
| *** smatzek has quit IRC | 17:57 | |
| *** igordcard_ has quit IRC | 17:57 | |
| *** igordcard_ has joined #openstack-nova | 17:57 | |
| *** busterswt has joined #openstack-nova | 17:58 | |
| *** busterswt has quit IRC | 17:58 | |
| *** zsais has joined #openstack-nova | 17:59 | |
| *** coolsvap|away is now known as coolsvap | 17:59 | |
| openstackgerrit | Danny Wilson proposed openstack/nova: libvirt: add discard support for attached volumes https://review.openstack.org/205726 | 18:00 |
| *** haomaiwang has quit IRC | 18:01 | |
| *** otter768 has quit IRC | 18:01 | |
| mriedem1 | get-vmware-people returns a 404 | 18:01 |
| mriedem1 | :( | 18:02 |
| *** haomaiwang has joined #openstack-nova | 18:02 | |
| *** salv-orlando has joined #openstack-nova | 18:05 | |
| *** ZZelle_ has joined #openstack-nova | 18:06 | |
| mriedem1 | jaypipes: it'd be nice to have you review https://review.openstack.org/#/c/207087/ as an ovo core | 18:06 |
| *** kmARC has quit IRC | 18:06 | |
| *** meghal has joined #openstack-nova | 18:07 | |
| *** meghal has quit IRC | 18:07 | |
| dansmith | mriedem1: I can link you the ovo change that is pending if you want it for context | 18:09 |
| jaypipes | dansmith: approved vladikr's keypair patch. vladikr thanks for pushing the fixes on that. | 18:09 |
| *** ijw has quit IRC | 18:09 | |
| mriedem1 | dansmith: sure | 18:09 |
| vladikr | jaypipes, thanks :) | 18:09 |
| *** meghal has joined #openstack-nova | 18:10 | |
| *** meghal has quit IRC | 18:10 | |
| *** armax has quit IRC | 18:10 | |
| ndipanov | mriedem1, do you remember that bug with live migration and not sving bdm data | 18:10 |
| ndipanov | did we ever fix that or just talked about fixing it? | 18:11 |
| jaypipes | mriedem1: done. | 18:11 |
| *** mdrabe has joined #openstack-nova | 18:11 | |
| jaypipes | vladikr: np. sorry for the delays in reviews. | 18:11 |
| *** browne has joined #openstack-nova | 18:12 | |
| mriedem1 | ndipanov: i do and we didn't | 18:12 |
| ndipanov | mriedem1, if you could dig up the link that would be uber helpful | 18:12 |
| mriedem1 | ndipanov: https://bugs.launchpad.net/nova/+bug/1419577 | 18:12 |
| openstack | Launchpad bug 1419577 in OpenStack Compute (nova) "when live-migrate failed, lun-id couldn't be rollback in havana" [High,Confirmed] | 18:12 |
| dansmith | ndipanov: https://bugs.launchpad.net/nova/+bug/1365804? | 18:12 |
| openstack | Launchpad bug 1365804 in OpenStack Compute (nova) "Did not find the volume after live_migration" [Low,Incomplete] - Assigned to Eli Qiao (taget-9) | 18:12 |
| mriedem1 | there were 2 related | 18:13 |
| *** meghal has joined #openstack-nova | 18:13 | |
| ByPasS | Hi everyone, is there a way to sync instances count from database and the hypervisor when they are out of sync ? (without directly mod the database itself) | 18:13 |
| mriedem1 | https://bugs.launchpad.net/nova/+bug/1416314 was a duplicate | 18:13 |
| openstack | Launchpad bug 1419577 in OpenStack Compute (nova) "duplicate for #1416314 when live-migrate failed, lun-id couldn't be rollback in havana" [High,Confirmed] | 18:13 |
| mriedem1 | ByPasS: yes | 18:13 |
| mriedem1 | the sync_instance_power_state periodic task does that every 10 minutes by default | 18:13 |
| *** agireud has quit IRC | 18:14 | |
| ByPasS | mriedem1 : thats where the WARNING comes from : While synchronizing instance power states, found 14 instances in the database and 12 instances on the hypervisor. | 18:14 |
| mriedem1 | ByPasS: http://git.openstack.org/cgit/openstack/nova/tree/nova/compute/manager.py#n5720 | 18:14 |
| dansmith | mriedem1: that bug involves a failed live migration and not rolling back properly, yes? | 18:16 |
| dansmith | I mean, it says that, but the code you're pointing to.. | 18:16 |
| mriedem1 | dansmith: yes | 18:16 |
| mriedem1 | did a red hat customer just explode? | 18:17 |
| *** agireud has joined #openstack-nova | 18:17 | |
| mnestratov | ndipanov: do you have time to look at this https://review.openstack.org/#/c/186858/ ? | 18:17 |
| *** meghal has quit IRC | 18:17 | |
| *** chhavi has quit IRC | 18:22 | |
| dansmith | mriedem1: https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L278-278 | 18:22 |
| mriedem1 | dansmith: i know | 18:22 |
| dansmith | mriedem1: since driver.attach_volume() can modify connection_info, I would think we should move that save to before the "if detached" check | 18:22 |
| *** dboik has joined #openstack-nova | 18:23 | |
| dansmith | er, I guess we should be running that anyway | 18:23 |
| dansmith | since we're going from detached to attached | 18:23 |
| dansmith | maybe? | 18:23 |
| mriedem1 | https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L228 | 18:23 |
| mriedem1 | once we exit the method, if connection_info changed, the bdm.save() is called regardless | 18:23 |
| dansmith | ah | 18:23 |
| mriedem1 | we do it explicity before calling volume_api.attach to avoid the race | 18:24 |
| dansmith | I see | 18:24 |
| dansmith | so why did you say we didn't fix this? | 18:24 |
| openstackgerrit | Raildo Mascena de Sousa Filho proposed openstack/nova: Nested Quota Driver: Finding parent_id and immediate child list https://review.openstack.org/200941 | 18:24 |
| mriedem1 | fix which thing? the live migration rollback? | 18:24 |
| dansmith | mriedem1: yeah | 18:24 |
| *** ig0r__ has joined #openstack-nova | 18:24 | |
| mriedem1 | time? people? resources? | 18:24 |
| dansmith | no, I mean.. | 18:25 |
| dansmith | seems like we *are* doing the updates now | 18:25 |
| dansmith | mriedem1: by the way, this is the ovo context patch: https://review.openstack.org/#/c/196041/ | 18:25 |
| *** ndipanov has quit IRC | 18:25 | |
| *** kmARC has joined #openstack-nova | 18:25 | |
| dansmith | mriedem1: see his query on PS12 | 18:25 |
| dansmith | mriedem1: we discussed the fix (which is this) on irc | 18:25 |
| mriedem1 | when ndipanov and i talked about it before, we talked about a hack to stash the connection_info in the bdm table before the new connection is made to the dest host and then on rollback from failure we restore that stashed connection_info | 18:25 |
| mriedem1 | like old_/new_ flavor | 18:25 |
| *** ndipanov has joined #openstack-nova | 18:26 | |
| *** eglynn has quit IRC | 18:26 | |
| *** ddaskal has quit IRC | 18:26 | |
| *** patrickeast has joined #openstack-nova | 18:26 | |
| *** dims has joined #openstack-nova | 18:26 | |
| dansmith | oh, you think this is not rolling back because we are updating it | 18:26 |
| dansmith | I see | 18:26 |
| dansmith | well, not sure this is really the thing I'm trying to chase anyway, | 18:26 |
| dansmith | but it was related | 18:26 |
| *** jasondotstar has joined #openstack-nova | 18:30 | |
| *** wanghao_ has joined #openstack-nova | 18:30 | |
| *** thumpba_ has joined #openstack-nova | 18:30 | |
| *** obondarev_ has joined #openstack-nova | 18:31 | |
| *** zzz_pczesno has joined #openstack-nova | 18:31 | |
| *** josecastroleon has joined #openstack-nova | 18:31 | |
| *** briancurtin_ has joined #openstack-nova | 18:31 | |
| *** coolsvap is now known as coolsvap|away | 18:32 | |
| dannywilson | mriedem1: hello, with regards to https://review.openstack.org/#/c/205726/ and creating a spec should I create it in specs/liberty or specs/mitaka? | 18:32 |
| mriedem1 | dansmith: also, the bdm.save() before volume_api.attach() is purely for a race that you really only see in the gate because of how fast the detach call comes after the attach | 18:33 |
| *** marcusvrn has quit IRC | 18:33 | |
| mriedem1 | b/c tempest is waiting for in-use to happen on the volume status and immediately detaches, within 1 second | 18:33 |
| dansmith | ah | 18:33 |
| mriedem1 | so not a real customer issue, unless your customer is an asshole | 18:34 |
| dansmith | okay | 18:34 |
| mriedem1 | dannywilson: mitaka | 18:34 |
| mriedem1 | dannywilson: nova feature freeze is today | 18:34 |
| *** meghal has joined #openstack-nova | 18:34 | |
| *** jlvillal_ has joined #openstack-nova | 18:34 | |
| *** aweeks_ has joined #openstack-nova | 18:34 | |
| *** briancurtin has quit IRC | 18:34 | |
| *** pczesno has quit IRC | 18:34 | |
| *** wanghao has quit IRC | 18:34 | |
| *** aweeks has quit IRC | 18:34 | |
| *** rushiagr has quit IRC | 18:34 | |
| *** igordcard_ has quit IRC | 18:34 | |
| *** jmccrory has quit IRC | 18:34 | |
| *** burt has quit IRC | 18:34 | |
| *** ericksonsantos has quit IRC | 18:34 | |
| *** jlvillal has quit IRC | 18:34 | |
| *** kbyrne has quit IRC | 18:34 | |
| *** obondarev has quit IRC | 18:34 | |
| *** ildikov has quit IRC | 18:34 | |
| *** thumpba has quit IRC | 18:34 | |
| *** mjturek1 has quit IRC | 18:34 | |
| *** flwang1 has quit IRC | 18:34 | |
| *** ericksonfgds has joined #openstack-nova | 18:34 | |
| *** jmccrory has joined #openstack-nova | 18:34 | |
| *** briancurtin_ is now known as briancurtin | 18:34 | |
| *** wanghao_ is now known as wanghao | 18:34 | |
| *** ildikov has joined #openstack-nova | 18:34 | |
| *** burt has joined #openstack-nova | 18:35 | |
| *** kbyrne has joined #openstack-nova | 18:35 | |
| openstackgerrit | Raildo Mascena de Sousa Filho proposed openstack/nova: Nested Quota Driver: Finding parent_id and immediate child list https://review.openstack.org/200941 | 18:35 |
| dannywilson | mriedem1: okay, thank you | 18:35 |
| *** jlvillal_ is now known as jlvillal | 18:35 | |
| *** mjturek1 has joined #openstack-nova | 18:35 | |
| *** aweeks_ is now known as aweeks | 18:36 | |
| scottda | mriedem1: I signed myself up as the Cinder -> Nova liaison and I put you down as Nova -> Cinder liaison, if that is OK? | 18:39 |
| scottda | https://wiki.openstack.org/wiki/CrossProjectLiaisons#Inter-project_Liaisons | 18:39 |
| mriedem1 | scottda: sure | 18:39 |
| mriedem1 | at some point if i get enough badges i think i win a prize | 18:39 |
| dansmith | and that prize is "never able to make progress on any one thing" | 18:39 |
| mriedem1 | huzzah! | 18:40 |
| *** salv-orlando has quit IRC | 18:40 | |
| dannywilson | mriedem1: or anyone for that matter, one more question, I have a Cinder spec for this and this Nova spec is going to be very similar, it seems silly to copy most of the Cinder spec into a Nova spec, any way to re-use that Cinder spec? Sorry if this is a silly question | 18:41 |
| mriedem1 | dannywilson: well the cinder spec probably says 'provide x' and the nova spec says 'consume x' right? | 18:42 |
| *** rushiagr has joined #openstack-nova | 18:42 | |
| mriedem1 | dannywilson: i saw the cinder spec had a +2 | 18:42 |
| dannywilson | mriedem1: yes, but all the reasons are the same | 18:42 |
| dannywilson | mriedem1: yep | 18:43 |
| mriedem1 | personally i don't want to have to do a goto in the nova spec, | 18:43 |
| mriedem1 | and it's not hard to copy/paste if it's fine in the cinder spec | 18:43 |
| *** sudipto has joined #openstack-nova | 18:43 | |
| mriedem1 | obviously add a link to the cinder spec in the references section of the nova spec | 18:43 |
| *** changbl has quit IRC | 18:43 | |
| dannywilson | mriedem1: agreed. my worry was they might get out of sync with each other but in this case that is probably not going to happen | 18:44 |
| dannywilson | mriedem1: thank you | 18:44 |
| mriedem1 | yeah, since the cinder spec looks pretty close to being approved i wouldn't worry about that | 18:44 |
| *** lyarwood has quit IRC | 18:44 | |
| *** tonytan4ever has joined #openstack-nova | 18:44 | |
| *** exploreshaifali has quit IRC | 18:46 | |
| *** ig0r_ has quit IRC | 18:50 | |
| *** igordcard has quit IRC | 18:52 | |
| *** eglynn has joined #openstack-nova | 18:53 | |
| sudipto | jaypipes, hi, can i get your quick view on the comment that dansmith made here: https://review.openstack.org/#/c/198972/10/nova/objects/monitor_metric.py the first comment. | 18:53 |
| dansmith | sudipto: I already pinged him about that this morning | 18:53 |
| *** flwang1 has joined #openstack-nova | 18:54 | |
| dansmith | he's a busy dude | 18:54 |
| sudipto | dansmith, thanks :) | 18:54 |
| sudipto | dansmith, yeah super busy :) | 18:54 |
| sudipto | dansmith, so should i go ahead with what you suggested? | 18:54 |
| dansmith | sudipto: so, let me explain a little | 18:55 |
| sudipto | dansmith, sure | 18:55 |
| dansmith | sudipto: if you change something in version 1.5 you have to be able to backport the object to 1.4, etc | 18:55 |
| dansmith | if you make something nullable in 1.5 | 18:55 |
| cfriesen | ndipanov: sorry, was out for lunch. I was looking at something that wants to use 2MB hugepages for networking. I was thinking it would be convenient to represent it as an instance numa topology with no CPUs and hugepages, regardless of whether the guest itself is using hugepages. That way we'd get all the scheduling "for free". | 18:55 |
| dansmith | sudipto: and then have to convert it to non-nullable in 1.4, what do you do? where does the value come from? | 18:55 |
| dansmith | sudipto: that's why I assert that it's not a backwards-compatible change | 18:55 |
| dansmith | sudipto: also, I don't think we *need* it to be none, because unset is what we actually want.. "there is no value for this field" not "this field has a value and it is none" | 18:56 |
| *** changbl has joined #openstack-nova | 18:56 | |
| sudipto | dansmith, yeah - got it from your comment actually...makes sense to me. I guess the attr_is_set function can assert the rest. | 18:56 |
| dansmith | sudipto: right | 18:57 |
| sudipto | dansmith, i will go ahead and do another patchset | 18:57 |
| dansmith | sudipto: okay, do that, I'll +2 and hopefully poke jaypipes to do the same | 18:57 |
| sudipto | dansmith, hopefully jaypipes will be in agreement :) | 18:57 |
| sudipto | dansmith, sure. | 18:58 |
| *** bapalm_ has joined #openstack-nova | 19:00 | |
| *** haomaiwang has quit IRC | 19:01 | |
| *** josecastroleon has quit IRC | 19:01 | |
| *** Marga_ has quit IRC | 19:01 | |
| *** haomaiwang has joined #openstack-nova | 19:02 | |
| *** Marga_ has joined #openstack-nova | 19:03 | |
| *** cpschult has quit IRC | 19:04 | |
| *** Marga_ has quit IRC | 19:04 | |
| *** ig0r_ has joined #openstack-nova | 19:04 | |
| *** Marga_ has joined #openstack-nova | 19:04 | |
| *** bapalm_ has quit IRC | 19:05 | |
| *** annashen has quit IRC | 19:08 | |
| *** eglynn has quit IRC | 19:09 | |
| *** annashen has joined #openstack-nova | 19:09 | |
| *** MentalRay__ has quit IRC | 19:09 | |
| jaypipes | dansmith: agree with you. "there is no value for this field" is correct. not "the value for this field is None" | 19:09 |
| jaypipes | sudipto: ^^ | 19:10 |
| dansmith | jaypipes: okay cool | 19:10 |
| *** annashen has quit IRC | 19:10 | |
| sudipto | jaypipes, ok great. | 19:10 |
| *** annashen has joined #openstack-nova | 19:10 | |
| *** marcusvrn has joined #openstack-nova | 19:14 | |
| *** MentalRay__ has joined #openstack-nova | 19:16 | |
| *** mwagner_lap has quit IRC | 19:16 | |
| *** lucas__ has quit IRC | 19:18 | |
| *** flwang1 has quit IRC | 19:20 | |
| *** cpschult has joined #openstack-nova | 19:20 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: libvirt: make snapshot use RBD snapshot/clone when available https://review.openstack.org/205282 | 19:22 |
| openstackgerrit | Raildo Mascena de Sousa Filho proposed openstack/nova: Nested Quota Driver: Adding v2 and v3(v2.1) support https://review.openstack.org/149828 | 19:24 |
| *** READ10 has quit IRC | 19:24 | |
| *** pixelbeat_ has joined #openstack-nova | 19:25 | |
| *** annashen has quit IRC | 19:26 | |
| *** annashen has joined #openstack-nova | 19:26 | |
| *** nagyz has joined #openstack-nova | 19:28 | |
| *** lpetrut has joined #openstack-nova | 19:28 | |
| *** jasondotstar has quit IRC | 19:29 | |
| *** salv-orlando has joined #openstack-nova | 19:29 | |
| *** annashen_ has joined #openstack-nova | 19:29 | |
| *** lpetrut has quit IRC | 19:30 | |
| nagyz | mriedem1, when you have the time can you answer my os-brick question that I left as a comment on the rbd encryption review? | 19:31 |
| *** armax has joined #openstack-nova | 19:31 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Handle instance_destroy_at_top failure https://review.openstack.org/207587 | 19:32 |
| *** annashen_ has quit IRC | 19:34 | |
| *** lucas__ has joined #openstack-nova | 19:34 | |
| *** nic has quit IRC | 19:34 | |
| *** nic has joined #openstack-nova | 19:35 | |
| *** armax has quit IRC | 19:36 | |
| bauzas | _gryf: saw my comment on https://review.openstack.org/#/c/197063/ ? | 19:38 |
| bauzas | _gryf: I don't want to be yakshaving, but I think it's not good to do your way | 19:38 |
| lxsli | hey jaypipes did you get a chance to think after your townhall? | 19:40 |
| *** annashen has quit IRC | 19:40 | |
| *** annashen has joined #openstack-nova | 19:41 | |
| *** pixelbeat_ has quit IRC | 19:41 | |
| *** fawadkhaliq has quit IRC | 19:42 | |
| *** fawadkhaliq has joined #openstack-nova | 19:43 | |
| *** pixelbeat_ has joined #openstack-nova | 19:43 | |
| *** lpetrut has joined #openstack-nova | 19:43 | |
| *** Sukhdev_ has quit IRC | 19:44 | |
| mriedem1 | nagyz: replied | 19:46 |
| *** openstackgerrit has quit IRC | 19:46 | |
| *** openstackgerrit has joined #openstack-nova | 19:47 | |
| mriedem1 | nagyz: there is an rbd driver in os-brick already http://git.openstack.org/cgit/openstack/os-brick/tree/os_brick/initiator/connector.py?id=0.3.2#n1263 | 19:47 |
| mriedem1 | nagyz: so i think you should work the encryption support into os-brick and then integrate that into nova | 19:47 |
| mriedem1 | nagyz: honestly i'm not sure what the work effort is to switch the current rbd volume driver support in nova over to using the rbd connector in os-brick, that'd be a question for hemna most likely | 19:48 |
| *** lpetrut has quit IRC | 19:48 | |
| *** achanda has quit IRC | 19:49 | |
| nagyz | cheers | 19:49 |
| *** jasondotstar has joined #openstack-nova | 19:50 | |
| mriedem1 | nagyz: i guess if we can show that the encrypted volume test in tempest is passing with your change then i'd be open to getting that into nova in liberty as a bug fix and then lift that out of nova into os-brick, and swap those in mitaka | 19:50 |
| nagyz | how about I refactor the code to go into it's own module under volume/, reenable the tempest test now as a depends-on change, we see if it passes properly, and if so then I'll make the os-brick changes in M? | 19:50 |
| mriedem1 | jinx | 19:50 |
| nagyz | ok we're thikning the same :-) | 19:50 |
| mriedem1 | plus, you know, unit tests | 19:50 |
| nagyz | right I just had no idea what's th ebest way to unit-test linux commands that actually modify the hypervisor state | 19:51 |
| nagyz | be back in around 5 | 19:51 |
| mriedem1 | nagyz: you mock out the calls to the other modules | 19:51 |
| mriedem1 | you make sure that connect_volume mutates connection_info, i.e. connection_info['data']['device_path'] = symlink_dev | 19:52 |
| mriedem1 | etc etc | 19:52 |
| mriedem1 | symlink_dev is mocked | 19:52 |
| openstackgerrit | Vikhyat Umrao proposed openstack/nova: RBD: Reading rbd_default_features from ceph.conf https://review.openstack.org/186537 | 19:53 |
| *** lpetrut has joined #openstack-nova | 19:54 | |
| *** mnestratov has quit IRC | 19:54 | |
| *** lpetrut has quit IRC | 19:55 | |
| *** cfriesen_ has joined #openstack-nova | 19:56 | |
| *** cfriesen has quit IRC | 19:56 | |
| *** annashen has quit IRC | 19:56 | |
| *** ig0r_ has quit IRC | 19:56 | |
| *** meghal has quit IRC | 19:56 | |
| *** tpatil has joined #openstack-nova | 19:57 | |
| *** annashen has joined #openstack-nova | 19:57 | |
| *** otter768 has joined #openstack-nova | 19:57 | |
| *** belmoreira has joined #openstack-nova | 19:57 | |
| *** cfriesen__ has joined #openstack-nova | 19:59 | |
| *** haomaiwang has quit IRC | 20:01 | |
| *** pixelbeat_ has quit IRC | 20:01 | |
| *** otter768 has quit IRC | 20:01 | |
| *** haomaiwang has joined #openstack-nova | 20:02 | |
| *** browne has quit IRC | 20:02 | |
| *** cfriesen_ has quit IRC | 20:02 | |
| *** browne has joined #openstack-nova | 20:03 | |
| mriedem1 | browne: can you review this https://review.openstack.org/#/c/165060/? | 20:04 |
| mriedem1 | i'd like to see some vmware devs review this kind of stuff | 20:04 |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: Allow NUMA based reporting for Monitors https://review.openstack.org/198972 | 20:04 |
| mriedem1 | browne: i reviewed it quite a bit and then got some concerns about backwards compatibility which garyk said wouldn't be a problem but i'm skeptical | 20:05 |
| *** cpschult has quit IRC | 20:05 | |
| *** dboik has quit IRC | 20:06 | |
| *** dboik has joined #openstack-nova | 20:06 | |
| sudipto | dansmith, jaypipes have updated the patchset, after struggling with net connectivity. Now will hit the bed! Thx! | 20:06 |
| dansmith | sudipto: thanks | 20:06 |
| *** sudipto has quit IRC | 20:07 | |
| *** cfriesen has joined #openstack-nova | 20:09 | |
| *** armax has joined #openstack-nova | 20:09 | |
| *** mwagner_lap has joined #openstack-nova | 20:09 | |
| *** armax has quit IRC | 20:09 | |
| hemna | mriedem1, so the rbd libvirt volume driver knows that it's going to a libvirt vm, so it uses the libvirt built in capabilities for rbd. | 20:10 |
| hemna | the os-brick connector is a generic system level volume discovery that doesn't know about libvirt. | 20:10 |
| mriedem1 | hemna: ok, i thought it looked like it was missing all of the libvirt net volume driver stuff | 20:11 |
| hemna | eventually, we could add another connector that is a libvirt aware connector | 20:11 |
| mriedem1 | hemna: do we then move that into os-brick? | 20:11 |
| mriedem1 | right | 20:11 |
| *** cfriesen_ has joined #openstack-nova | 20:11 | |
| hemna | I think there is another connector that has that same issue | 20:11 |
| hemna | that could benefit from a libvirt based connector parent. | 20:11 |
| mriedem1 | so i guess nagyz is on the right track with just trying to get it working as-is in nova in liberty | 20:11 |
| hemna | yup | 20:12 |
| mriedem1 | alright, well i'm out for a few hours | 20:12 |
| *** mriedem1 is now known as mriedem_away | 20:12 | |
| *** cfriesen__ has quit IRC | 20:12 | |
| *** annegentle has quit IRC | 20:13 | |
| *** annegentle has joined #openstack-nova | 20:13 | |
| *** gszasz has quit IRC | 20:14 | |
| *** cfriesen has quit IRC | 20:14 | |
| *** baoli has quit IRC | 20:15 | |
| nagyz | great. :) | 20:15 |
| nagyz | thanks for the comments for both of you | 20:15 |
| *** achanda has joined #openstack-nova | 20:16 | |
| *** openstackgerrit has quit IRC | 20:16 | |
| *** openstackgerrit has joined #openstack-nova | 20:17 | |
| *** chuckcarmack has left #openstack-nova | 20:17 | |
| *** ijw has joined #openstack-nova | 20:18 | |
| hemna | np | 20:22 |
| *** meghal has joined #openstack-nova | 20:27 | |
| *** Piet_ has quit IRC | 20:27 | |
| *** maurosr has quit IRC | 20:28 | |
| *** ekhugen has quit IRC | 20:29 | |
| *** timello has quit IRC | 20:29 | |
| *** lucas__ has quit IRC | 20:29 | |
| *** rfolco has quit IRC | 20:29 | |
| *** rfolco has joined #openstack-nova | 20:30 | |
| *** agireud has quit IRC | 20:30 | |
| *** meghal has quit IRC | 20:31 | |
| *** tonytan4ever has quit IRC | 20:31 | |
| *** cpschult has joined #openstack-nova | 20:33 | |
| *** Piet has joined #openstack-nova | 20:33 | |
| *** lucas__ has joined #openstack-nova | 20:33 | |
| *** lucas__ has quit IRC | 20:35 | |
| *** vladikr has quit IRC | 20:35 | |
| *** lucas__ has joined #openstack-nova | 20:35 | |
| *** meghal has joined #openstack-nova | 20:36 | |
| *** busterswt has joined #openstack-nova | 20:36 | |
| *** tonytan4ever has joined #openstack-nova | 20:37 | |
| openstackgerrit | Danny Wilson proposed openstack/nova-specs: Add ability to support discard/unmap/trim for Cinder backend https://review.openstack.org/207615 | 20:37 |
| *** baoli has joined #openstack-nova | 20:38 | |
| *** baoli has quit IRC | 20:39 | |
| *** bauzas has quit IRC | 20:39 | |
| *** baoli has joined #openstack-nova | 20:39 | |
| *** ccesario has joined #openstack-nova | 20:40 | |
| *** annegentle has quit IRC | 20:40 | |
| *** mnestratov has joined #openstack-nova | 20:42 | |
| *** ijw has quit IRC | 20:43 | |
| *** bauzas has joined #openstack-nova | 20:43 | |
| *** browne has quit IRC | 20:43 | |
| *** thangp has quit IRC | 20:43 | |
| *** browne has joined #openstack-nova | 20:44 | |
| *** VW_ has quit IRC | 20:44 | |
| *** VW_ has joined #openstack-nova | 20:44 | |
| *** openstackgerrit has quit IRC | 20:46 | |
| *** openstackgerrit has joined #openstack-nova | 20:47 | |
| *** raildo has quit IRC | 20:48 | |
| *** VW_ has quit IRC | 20:49 | |
| *** ijw has joined #openstack-nova | 20:50 | |
| ccesario | hello, could someone know how "nova" choose the datastore on initial placement of instance when using Vmware driver ? Is it automatic!? | 20:50 |
| *** lucas__ has quit IRC | 20:50 | |
| *** lpetrut has joined #openstack-nova | 20:51 | |
| bauzas | ccesario: ask VMWare | 20:52 |
| tonyb | ccesario: garyk would be your best bet. | 20:52 |
| *** ddaskal has joined #openstack-nova | 20:52 | |
| tonyb | bauzas: hey! | 20:52 |
| bauzas | ccesario: I mean, there is not really a placement done | 20:52 |
| tonyb | bauzas: how are you doing? | 20:52 |
| bauzas | ccesario: the main vmware cloud is behind one compute | 20:53 |
| bauzas | tonyb: good good :) | 20:53 |
| bauzas | tonyb: as a French guy, in holiday by tomorrow :) | 20:53 |
| bauzas | respectful for my country | 20:53 |
| tonyb | bauzas: :) recover from your virtual Rochester trip? | 20:53 |
| bauzas | tonyb: yup, the beer was good | 20:54 |
| tonyb | bauzas: Good news (about the holiday) | 20:54 |
| bauzas | tonyb: and the jetlag was not too huge | 20:54 |
| tonyb | bauzas: hehe. good to hear. | 20:54 |
| bauzas | yeah, a flight around 5 min | 20:55 |
| bauzas | :) | 20:55 |
| bauzas | and you ? :) | 20:55 |
| *** flwang1 has joined #openstack-nova | 20:55 | |
| mikal | Nova meeting in 5 minutes? | 20:55 |
| tonyb | bauzas: sure the flying time was short but still the sleep disruption was real | 20:55 |
| *** aysyd has quit IRC | 20:55 | |
| browne | mriedem1: sorry for the late reply, thank for looking at garyk's patch, I'll review that patch today. | 20:56 |
| bauzas | tonyb: ha | 20:56 |
| bauzas | oh right, meeting | 20:56 |
| bauzas | and then, bye for 10 days ! | 20:56 |
| tonyb | bauzas: I'm good. | 20:56 |
| *** jasondotstar has quit IRC | 20:56 | |
| bauzas | tonyb: coolness :) | 20:57 |
| *** jamielennox|away is now known as jamielennox | 20:57 | |
| *** hshiina has joined #openstack-nova | 20:57 | |
| ccesario | bouzas, maybe my question is wrong..... well, how explain ... I have two datastores in my Vmware Cluster ... each time that I "launch Instance" the instance it is stored in a different datastore. | 20:57 |
| ccesario | I mean this as 'initial placement' | 20:58 |
| bauzas | ccesario: ah that | 20:59 |
| tonyb | ccesario: I'm very far from a vmware expert but I think that's vcenter making the decision and then tellign nova. | 20:59 |
| ccesario | sorry but incorrect terms.... this is new for me | 20:59 |
| bauzas | ccesario: so yeah, each datastore is seen as a compute host | 20:59 |
| browne | ccesario: there is a datastore_regex option in the vmware driver that affects placement of instances | 20:59 |
| *** ijw has quit IRC | 20:59 | |
| bauzas | ccesario: the vmware driver is reporting the datastore metrics as a whole | 20:59 |
| bauzas | contiguous if you prefer | 21:00 |
| *** tpatil has quit IRC | 21:00 | |
| *** thumpba_ has quit IRC | 21:00 | |
| bauzas | like if it was only one host | 21:00 |
| *** haomaiwang has quit IRC | 21:01 | |
| *** lpetrut has quit IRC | 21:01 | |
| *** openstack has joined #openstack-nova | 21:09 | |
| ccesario | browne, yes, like this... or based in config (nova.conf) | 21:09 |
| *** agireud has joined #openstack-nova | 21:09 | |
| *** lucas-dinner has quit IRC | 21:11 | |
| browne | ccesario: I'll have to confirm with garyk, but believe you can only decide placement on the datastore for all instances, not specific ones. there's no image property i could find for that | 21:13 |
| ccesario | sure..... If you could check.... I appreciate this :) | 21:14 |
| *** annashen has quit IRC | 21:15 | |
| ccesario | thanks for these infos :) | 21:15 |
| browne | ccesario: np | 21:16 |
| *** annegentle has joined #openstack-nova | 21:16 | |
| *** annashen has joined #openstack-nova | 21:18 | |
| *** timello has joined #openstack-nova | 21:18 | |
| *** zul has joined #openstack-nova | 21:18 | |
| *** ekhugen has joined #openstack-nova | 21:19 | |
| *** maurosr has joined #openstack-nova | 21:19 | |
| *** achanda has quit IRC | 21:26 | |
| *** marzif_ has joined #openstack-nova | 21:26 | |
| *** lucas__ has joined #openstack-nova | 21:26 | |
| *** agireud has quit IRC | 21:29 | |
| *** annashen_ has joined #openstack-nova | 21:30 | |
| *** lucas__ has quit IRC | 21:31 | |
| *** marzif_ has quit IRC | 21:32 | |
| *** marzif_ has joined #openstack-nova | 21:34 | |
| *** moshele has joined #openstack-nova | 21:34 | |
| *** baoli has quit IRC | 21:35 | |
| *** annashen_ has quit IRC | 21:35 | |
| *** marzif_ has quit IRC | 21:35 | |
| *** marzif_ has joined #openstack-nova | 21:36 | |
| *** VW_ has joined #openstack-nova | 21:36 | |
| *** maurosr has quit IRC | 21:36 | |
| openstackgerrit | Anthony Lee proposed openstack/nova: Fix live-migrations usage of the wrong connector information https://review.openstack.org/202770 | 21:36 |
| *** ekhugen has quit IRC | 21:36 | |
| *** timello has quit IRC | 21:37 | |
| *** tjones has joined #openstack-nova | 21:40 | |
| *** VW_ has quit IRC | 21:41 | |
| *** achanda has joined #openstack-nova | 21:41 | |
| *** lykinsbd has quit IRC | 21:43 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: libvirt: make snapshot use RBD snapshot/clone when available https://review.openstack.org/205282 | 21:43 |
| *** yassine has joined #openstack-nova | 21:44 | |
| *** moshele has quit IRC | 21:44 | |
| edleafe | jroll: did you see bauzas' comment on https://review.openstack.org/#/c/206736/? | 21:45 |
| edleafe | jroll: are there *any* possible ways that an ironic deployment would need instance information for its filters? | 21:46 |
| *** jecarey has quit IRC | 21:46 | |
| edleafe | jroll: e.g., does the SameHost filter even make sense for ironic? | 21:46 |
| *** markvoelker_ has quit IRC | 21:46 | |
| *** cpschult has quit IRC | 21:48 | |
| *** jasondotstar has joined #openstack-nova | 21:48 | |
| *** annegentle has quit IRC | 21:49 | |
| *** busterswt has quit IRC | 21:49 | |
| *** rushiagr is now known as rushiagr_away | 21:50 | |
| *** ig0r__ has quit IRC | 21:51 | |
| *** thorst1 has joined #openstack-nova | 21:51 | |
| openstackgerrit | Scott DAngelo proposed openstack/nova-specs: Support force detach volume to nova https://review.openstack.org/84048 | 21:52 |
| *** angdraug has joined #openstack-nova | 21:52 | |
| bauzas | edleafe: I know that's silly | 21:53 |
| bauzas | edleafe: that's certainly not deployed | 21:53 |
| bauzas | edleafe: I only care about making it very verbose | 21:54 |
| *** marzif_ has quit IRC | 21:54 | |
| bauzas | edleafe: the thing is, I wonder how to make that clear, in a doc or elsewhere | 21:54 |
| bauzas | edleafe: like a known limitation for Ironic | 21:54 |
| edleafe | bauzas: I'm not up-to-date with ironic docs | 21:55 |
| bauzas | edleafe: yep, me too :( | 21:55 |
| bauzas | edleafe: so I just wonder how to properly document it | 21:55 |
| bauzas | edleafe: a DocImpact tag is certainly lazy | 21:55 |
| edleafe | bauzas: but I got the impression from jroll that filtering on existing instance info makes no sense | 21:55 |
| edleafe | since all "instances" are on the same host | 21:56 |
| bauzas | edleafe: yup that I agree hence my "its silly" ^ | 21:56 |
| edleafe | bauzas: ah, I thought you meant just the SameHostFilter | 21:56 |
| bauzas | edleafe: well, providing instances to the scheduler is only needed by SameHostFilter, right? | 21:56 |
| edleafe | bauzas: no, there are the instance_type affinities | 21:57 |
| bauzas | edleafe: well, instance_type, not instance? | 21:57 |
| bauzas | edleafe: lemme look again | 21:57 |
| *** otter768 has joined #openstack-nova | 21:58 | |
| edleafe | bauzas: hosts that have or don't have instances with the same instance_type | 21:58 |
| edleafe | bauzas: there are only a few that need instance info | 21:58 |
| edleafe | bauzas: and none of those make sense for ironic | 21:58 |
| mikal | So, my random thought for the day | 21:58 |
| mikal | I wonder how much of our review bandwidth is consumed by gate issues? | 21:59 |
| *** dboik has quit IRC | 21:59 | |
| *** kmARC has quit IRC | 21:59 | |
| bauzas | mikal: I'm not that impacted | 21:59 |
| melwitt | in the case of https://review.openstack.org/#/c/202605/, what happens when part of the series merges but the rest of it doesn't make the deadline? wait until M opens? | 22:00 |
| *** annegentle has joined #openstack-nova | 22:00 | |
| bauzas | mikal: I mean, changes that are missing Jenkins are today's ones, which means that either someone provided a new change or he was reviewed and did the review limbo dance | 22:00 |
| *** tr3buchet has quit IRC | 22:00 | |
| *** haomaiwang has quit IRC | 22:01 | |
| bauzas | edleafe: any link I could take a look ? | 22:01 |
| dansmith | melwitt: I dunno, it would probably be a good candidate for exception | 22:01 |
| dansmith | melwitt: smallish, half-merged, etc | 22:01 |
| melwitt | dansmith: nod | 22:01 |
| *** rodolfo-alonso has joined #openstack-nova | 22:01 | |
| bauzas | edleafe: looking at the type_affinity filters, can't find some using the host_state.instances attr | 22:02 |
| *** jwcroppe has joined #openstack-nova | 22:02 | |
| *** annashen has quit IRC | 22:02 | |
| *** haomaiwang has joined #openstack-nova | 22:02 | |
| rodolfo-alonso | This filter is obligatoriness | 22:02 |
| bauzas | edleafe: I only find host_state.instances check within the SameHost filter | 22:02 |
| *** zzzeek has quit IRC | 22:02 | |
| bauzas | rodolfo-alonso: so | 22:02 |
| rodolfo-alonso | The filter proposed is not a must | 22:02 |
| edleafe | bauzas: https://review.openstack.org/#/c/160512/19/nova/scheduler/filters/type_filter.py,cm | 22:02 |
| bauzas | rodolfo-alonso: you speaking of https://review.openstack.org/#/c/189279/8/doc/source/filter_scheduler.rst,cm ? | 22:03 |
| edleafe | bauzas: that was the change to the type_filter | 22:03 |
| *** otter768 has quit IRC | 22:03 | |
| rodolfo-alonso | no | 22:03 |
| rodolfo-alonso | https://review.openstack.org/#/c/189279/8/doc/source/filter_scheduler.rst | 22:03 |
| *** mnestratov has quit IRC | 22:03 | |
| rodolfo-alonso | Sorry for the confusion | 22:03 |
| bauzas | edleafe: well, it checks the filter_props.instance, not the host_state.instances, right? | 22:03 |
| bauzas | edleafe: I haven't yet looked at the helper | 22:04 |
| edleafe | bauzas: no, it uses the host_state.instances | 22:04 |
| bauzas | rodolfo-alonso: yeah, I'm just in a middle of a second convo :) | 22:04 |
| rodolfo-alonso | sorry | 22:04 |
| bauzas | edleafe: okay, fair to say | 22:04 |
| *** ctrath has quit IRC | 22:05 | |
| bauzas | edleafe: then, I'm a bit more worried, since Ironic is using flavors for their own purpose | 22:05 |
| bauzas | rodolfo-alonso: so, back to our convo | 22:05 |
| bauzas | rodolfo-alonso: let me emphasize again what I said in the review | 22:05 |
| edleafe | bauzas: why? The filter has to do with same/diff flavor on the same host | 22:05 |
| edleafe | bauzas: all ironic instances are on the same host | 22:06 |
| bauzas | rodolfo-alonso: "'m more in favor of modifying the existing AggregateInstanceExtraSpecsFilter to keep its behaviour but also have a more stricter enforcement logic (ie. the opposite logic)" | 22:06 |
| rodolfo-alonso | bauzas: I think thats the point | 22:06 |
| bauzas | edleafe: an Ironic node is responsible for managing a certain number of physical hosts - ie. instances | 22:06 |
| *** moshele has joined #openstack-nova | 22:07 | |
| *** sajeesh has joined #openstack-nova | 22:07 | |
| bauzas | edleafe: IIUC, Ironic is using ironic conductors for talking to those instances | 22:07 |
| bauzas | edleafe: and an Ironic node is actually an Ironic API node | 22:07 |
| bauzas | edleafe: I'd prefer to wait for jroll | 22:07 |
| edleafe | bauzas: but there is no meaning for an affinity, as there is for virt instances | 22:08 |
| edleafe | bauzas: +1 on getting a better explanation from jroll | 22:08 |
| *** smatzek has joined #openstack-nova | 22:08 | |
| *** markvoelker has joined #openstack-nova | 22:08 | |
| bauzas | edleafe: well, ironic asks the user to create flavors | 22:08 |
| bauzas | edleafe: so we could segregate per flavor maybe | 22:09 |
| jroll | here now. | 22:09 |
| bauzas | edleafe: I mean, *someone* could use that filter for segregating his cloud | 22:09 |
| jroll | should I read the scrollback or just say there's only one compute host ever so it doesn't matter | 22:09 |
| edleafe | jroll: reading the scrollback? | 22:09 |
| edleafe | jinx | 22:09 |
| jroll | like | 22:10 |
| jroll | there's only one host to choose | 22:10 |
| edleafe | jroll: I think bauzas might want more details | 22:10 |
| jroll | and all nodes can have one and only one instance | 22:10 |
| bauzas | jroll: okay, my take is that either we say it's unsupported to run filters in an Ironic env, or we find a way to workaround that | 22:10 |
| jroll | 1 host, many nodes, nodes have a max capacity of one. | 22:10 |
| jroll | bauzas: no, we need filters... just instance tracking doesn't make sense | 22:10 |
| bauzas | jroll: okay, which filters N? | 22:11 |
| *** zsais has quit IRC | 22:11 | |
| jroll | let me rephrase: it doesn't ever make sense to schedule instance X based on any other instances. | 22:11 |
| bauzas | jroll: well, instance tracking makes sense for the filters using that :) | 22:11 |
| jroll | bauzas: https://github.com/openstack/nova/blob/master/nova/scheduler/ironic_host_manager.py#L33-49 | 22:11 |
| jroll | bauzas: well, it doesn't make sense to use filters that scheduler based on instance tracking | 22:12 |
| jroll | that schedule** | 22:12 |
| bauzas | jroll: okay, what if a weirdo guy explicitely adds the filters that are using the instance tracking ? | 22:12 |
| bauzas | jroll: that list is just a default opt | 22:12 |
| edleafe | bauzas: if the instance info was there, they would be boned anyway | 22:13 |
| bauzas | edleafe: right, but they wouldn't understand | 22:13 |
| edleafe | bauzas: your assumption is that they don't understand what they're doing in the first place | 22:13 |
| jroll | I mean | 22:13 |
| jroll | edleafe: most people don't, to be clear, but then again those people use the defaults | 22:14 |
| jroll | so I'm confused why e.g. setting host_state.instances = {} and using those filters would break things | 22:14 |
| jroll | I guess it would break SameHost? even if there's only one host? | 22:14 |
| bauzas | okay, I don't want to nitpick folks, I just think we could just say that Ironic doesn't support other than the default filters | 22:15 |
| *** jamielennox is now known as jamielennox|away | 22:15 | |
| jroll | hmm | 22:15 |
| bauzas | jroll: is it totally non-sense to run 2 computes ? | 22:15 |
| jroll | bauzas: yes (though 2 computes with same host name is somewhat supported) | 22:16 |
| jroll | each compute host tracks *all* ironic nodes | 22:16 |
| jroll | so you'd have two hosts tracking the same nodes | 22:16 |
| bauzas | jroll: what if I shard my ironic cloud into two ? :) | 22:16 |
| jroll | (host1, node1) (host2, node1) etc | 22:16 |
| jroll | bauzas: sure, you could have two hosts talking to two ironic clusters | 22:17 |
| bauzas | jroll: since nova gives the wrapper level | 22:17 |
| jroll | hmm | 22:17 |
| * jroll thinks about implications | 22:17 | |
| *** hshiina has quit IRC | 22:17 | |
| jroll | I was actually wondering about this the other day... | 22:17 |
| jroll | I also might be okay with saying we don't support non-defaults, as long as we don't actively block non-default (for people that write their own or whatever) | 22:18 |
| jroll | but would want to talk to ironic folks | 22:18 |
| bauzas | jroll: well, maybe a WARN log when starting the Ironic HostManager | 22:18 |
| bauzas | ? | 22:18 |
| jroll | bauzas: could be, yeah, or docs and help text | 22:19 |
| bauzas | something like "using other than the default filters is somehow possible but either discouraged or unsupported" | 22:19 |
| jroll | yeah | 22:19 |
| *** VW_ has joined #openstack-nova | 22:20 | |
| bauzas | rodolfo-alonso: since I'm going to drop off ( on EU TZ) | 22:20 |
| edleafe | jroll: or "using the virt filters with ironic is not a good thing" | 22:20 |
| bauzas | 00:20 here | 22:20 |
| jroll | yeah, something like that | 22:20 |
| rodolfo-alonso | ok.... | 22:20 |
| bauzas | rodolfo-alonso: and since I'm going to be off for the next 2 weeks | 22:21 |
| rodolfo-alonso | me too | 22:21 |
| rodolfo-alonso | ufffff | 22:21 |
| edleafe | bauzas: before you go, can you remove the -1? | 22:21 |
| jroll | bauzas: edleafe: today has been kind of crazy for me, let me think about this more tonight/tomorrow | 22:21 |
| edleafe | bauzas: either update the commit message, or I'll do it | 22:21 |
| jroll | I need to read more code before making a decision :) | 22:21 |
| bauzas | rodolfo-alonso: I'd suggest you to investigate porting your code to the existing AggregareExtraSpecs filter and do the necessary modifications to have the default behaviour be kept | 22:22 |
| *** thorst1 has quit IRC | 22:22 | |
| edleafe | jroll: the basic question, though, is: does it ever make sense to filter an ironic request based on the instances on a given host? | 22:22 |
| rodolfo-alonso | bauzas: but this is the same reply.... and the problem is still there | 22:22 |
| edleafe | jroll: the other filters are a different question | 22:22 |
| bauzas | edleafe: well, I defer that to jroll | 22:22 |
| jroll | edleafe: IMO it does not. however, I need to think about the "two ironic clusters" situation more | 22:22 |
| jroll | edleafe: with one ironic cluster it certainly does not. | 22:23 |
| rodolfo-alonso | bauzas: it's not possible to use this kind of filter in this class. | 22:23 |
| bauzas | rodolfo-alonso: no, because I'd suggest you to use a config flag for changing the behaviour | 22:23 |
| jroll | edleafe: though now I'm thinking about near/far so maybe I'm wrong :) | 22:23 |
| bauzas | on demand | 22:23 |
| rodolfo-alonso | bauzas: to be necessaty or not? | 22:23 |
| bauzas | rodolfo-alonso: well, I was suggesting a config opt defaulting to the current behaviour, but which would provide the behaviour you would like if not default | 22:24 |
| rodolfo-alonso | bauzas: what I mean: if this extra_spec is not in the aggregate, should pass | 22:24 |
| *** nic has quit IRC | 22:24 | |
| bauzas | rodolfo-alonso: or you could extend the existing semantics to match your needs | 22:24 |
| bauzas | rodolfo-alonso: I'd rather be more in favor of the latter | 22:25 |
| bauzas | rodolfo-alonso: speaking of the aggregate metadata semantics of course | 22:25 |
| rodolfo-alonso | bauzas: i'll think about this.... | 22:25 |
| *** VW__ has joined #openstack-nova | 22:25 | |
| edleafe | jroll: ok, I have to run out for a bit. If you could add your thoughts to the review, that would be excellent | 22:25 |
| bauzas | rodolfo-alonso: ie. the current semantics would keep the existing behaviour | 22:25 |
| jroll | edleafe: yeah, I'll probably defer to tomorrow morning though. | 22:26 |
| edleafe | jroll: cool | 22:26 |
| bauzas | rodolfo-alonso: but providing some extra semantics which would change the behaviour | 22:26 |
| bauzas | folks, have a good night | 22:26 |
| bauzas | I'm dropping off | 22:26 |
| rodolfo-alonso | bye | 22:26 |
| edleafe | bauzas: please don't leave the -1 hanging while you're off enjoying yourself. :) | 22:26 |
| jroll | bauzas: have a great vacation :) | 22:27 |
| bauzas | edleafe: why ? am I so terrible ?:) | 22:27 |
| dansmith | edleafe: he should leave a -1 if he wants to | 22:27 |
| dansmith | it's totally fine | 22:27 |
| jroll | now I feel like he should leave a -2 just for fun | 22:27 |
| bauzas | jroll: eh, can't :) | 22:27 |
| dansmith | I can | 22:27 |
| dansmith | don't tempt me | 22:27 |
| edleafe | jroll: ok, then dansmith can do it | 22:27 |
| edleafe | ha! | 22:27 |
| dansmith | bauzas: go to sleep, I'll defend your -1 | 22:27 |
| bauzas | anyway, bye folks and see you later | 22:28 |
| *** bauzas is now known as bauzas_off | 22:28 | |
| edleafe | bauzas: enjoy! | 22:28 |
| jroll | bauzas: I'll have dan give you core for the 5 seconds it takes to -2 :P | 22:28 |
| dansmith | boy, wouldn't that be a thing | 22:28 |
| *** VW_ has quit IRC | 22:28 | |
| dansmith | temporary core provisions to disagree with someone extra hard | 22:28 |
| jroll | it'd be so fun | 22:29 |
| * jroll proposes to the mailing list | 22:29 | |
| edleafe | how about temporary super-core? | 22:29 |
| dansmith | heh | 22:29 |
| jroll | maybe you get like, 5 per cycle | 22:29 |
| *** lucas__ has joined #openstack-nova | 22:29 | |
| edleafe | -10 power! | 22:29 |
| jroll | and use them at your discretion | 22:29 |
| *** VW__ has quit IRC | 22:30 | |
| *** vilobhmm has quit IRC | 22:30 | |
| *** vilobhmm has joined #openstack-nova | 22:30 | |
| *** apuimedo has quit IRC | 22:31 | |
| *** moshele has quit IRC | 22:32 | |
| *** markvoelker_ has joined #openstack-nova | 22:32 | |
| *** lucas__ has quit IRC | 22:34 | |
| *** markvoelker has quit IRC | 22:35 | |
| *** vilobhmm has quit IRC | 22:35 | |
| *** apuimedo has joined #openstack-nova | 22:36 | |
| *** smatzek has quit IRC | 22:36 | |
| *** belmoreira has quit IRC | 22:37 | |
| *** ddaskal has quit IRC | 22:38 | |
| jroll | edleafe: another case to think about: clouds that use ironic and virt drivers, with host aggregates | 22:38 |
| jroll | I assume the aggregate filter would filter first but I'm not sure | 22:38 |
| *** eharney has quit IRC | 22:38 | |
| *** Sukhdev_ has joined #openstack-nova | 22:39 | |
| *** tpeoples has quit IRC | 22:41 | |
| *** annegentle has quit IRC | 22:44 | |
| *** harlowja has quit IRC | 22:44 | |
| *** harlowja has joined #openstack-nova | 22:44 | |
| *** jswarren has quit IRC | 22:45 | |
| *** MentalRay__ has quit IRC | 22:47 | |
| *** dustins has quit IRC | 22:52 | |
| *** eglynn has joined #openstack-nova | 22:58 | |
| *** boris-42 has quit IRC | 23:00 | |
| *** haomaiwang has quit IRC | 23:01 | |
| *** meghal has quit IRC | 23:01 | |
| *** haomaiwa_ has joined #openstack-nova | 23:02 | |
| *** meghal has joined #openstack-nova | 23:02 | |
| *** rodolfo-alonso has quit IRC | 23:03 | |
| *** zhenguo has joined #openstack-nova | 23:03 | |
| *** eglynn has quit IRC | 23:06 | |
| *** lucas__ has joined #openstack-nova | 23:06 | |
| *** slong has joined #openstack-nova | 23:06 | |
| *** lucas__ has quit IRC | 23:11 | |
| *** lucas__ has joined #openstack-nova | 23:11 | |
| *** apuimedo has quit IRC | 23:12 | |
| *** lucas____ has joined #openstack-nova | 23:13 | |
| *** VW_ has joined #openstack-nova | 23:14 | |
| *** markvoelker_ has quit IRC | 23:14 | |
| *** cfriesen__ has joined #openstack-nova | 23:14 | |
| *** cfriesen_ has quit IRC | 23:15 | |
| *** SourabhP has joined #openstack-nova | 23:15 | |
| *** lucas__ has quit IRC | 23:15 | |
| *** meghal has quit IRC | 23:16 | |
| *** meghal has joined #openstack-nova | 23:16 | |
| *** ijw has joined #openstack-nova | 23:17 | |
| *** ijw has quit IRC | 23:17 | |
| *** vilobhmm has joined #openstack-nova | 23:17 | |
| *** lucas____ has quit IRC | 23:17 | |
| *** ijw has joined #openstack-nova | 23:18 | |
| *** Jeffrey4l has joined #openstack-nova | 23:20 | |
| *** meghal has quit IRC | 23:23 | |
| *** VW_ has quit IRC | 23:24 | |
| *** VW_ has joined #openstack-nova | 23:25 | |
| *** htruta has quit IRC | 23:26 | |
| *** diogogmt has quit IRC | 23:28 | |
| *** VW_ has quit IRC | 23:29 | |
| *** Piet has quit IRC | 23:30 | |
| *** yassine has quit IRC | 23:31 | |
| *** dims_ has joined #openstack-nova | 23:31 | |
| *** dims_ has quit IRC | 23:31 | |
| *** annashen has joined #openstack-nova | 23:31 | |
| *** meghal has joined #openstack-nova | 23:33 | |
| *** Jeffrey4l has quit IRC | 23:34 | |
| *** ijw has quit IRC | 23:35 | |
| *** dims has quit IRC | 23:35 | |
| *** vilobhmm has quit IRC | 23:36 | |
| *** annashen has quit IRC | 23:36 | |
| *** armax has joined #openstack-nova | 23:38 | |
| openstackgerrit | Scott DAngelo proposed openstack/nova-specs: Support force detach volume to nova https://review.openstack.org/84048 | 23:39 |
| *** armax has quit IRC | 23:39 | |
| *** jaypipes has quit IRC | 23:39 | |
| *** marcusvrn has quit IRC | 23:43 | |
| *** vipul has quit IRC | 23:46 | |
| *** Jeffrey4l has joined #openstack-nova | 23:46 | |
| *** pixelbeat_ has joined #openstack-nova | 23:46 | |
| *** vipul has joined #openstack-nova | 23:46 | |
| *** ijw has joined #openstack-nova | 23:46 | |
| *** armax has joined #openstack-nova | 23:47 | |
| *** ZZelle_ has quit IRC | 23:52 | |
| *** Piet has joined #openstack-nova | 23:54 | |
| *** patrickeast has quit IRC | 23:54 | |
| *** Piet has quit IRC | 23:54 | |
| *** dannywilson has quit IRC | 23:55 | |
| *** nelsnelson has quit IRC | 23:56 | |
| gmann | alaski: ping | 23:56 |
| *** salv-orlando has quit IRC | 23:57 | |
| *** pixelbeat_ has quit IRC | 23:57 | |
| *** otter768 has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!