| *** thurloat_isgone is now known as thurloat | 00:00 | |
| *** johnbelamaric has joined #openstack-neutron | 00:02 | |
| *** yfujioka_ has joined #openstack-neutron | 00:02 | |
| *** ijw_ has joined #openstack-neutron | 00:02 | |
| *** dannywilson has quit IRC | 00:02 | |
| *** yfujioka has quit IRC | 00:03 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Check for 'removed' in port_info before reference https://review.openstack.org/187819 | 00:05 |
|---|---|---|
| *** ijw has quit IRC | 00:06 | |
| *** padkrish has joined #openstack-neutron | 00:07 | |
| *** vivek-ebay has quit IRC | 00:10 | |
| *** padkrish has quit IRC | 00:12 | |
| *** shakamunyi has joined #openstack-neutron | 00:12 | |
| *** barra204 has joined #openstack-neutron | 00:12 | |
| openstackgerrit | Merged openstack/neutron: Add route to metadata IP by default https://review.openstack.org/187431 | 00:16 |
| *** dane_leblanc has quit IRC | 00:16 | |
| *** ian_brown has joined #openstack-neutron | 00:21 | |
| *** padkrish has joined #openstack-neutron | 00:21 | |
| *** aranjan has quit IRC | 00:24 | |
| *** vivek-ebay has joined #openstack-neutron | 00:25 | |
| *** ijw_ has quit IRC | 00:27 | |
| *** ijw has joined #openstack-neutron | 00:27 | |
| *** padkrish has quit IRC | 00:27 | |
| *** thurloat is now known as thurloat_isgone | 00:28 | |
| *** sbalukoff has quit IRC | 00:29 | |
| openstackgerrit | enikanorov proposed openstack/neutron: Catch broad exception in methods used in FixedIntervalLoopingCall https://review.openstack.org/185722 | 00:30 |
| *** padkrish has joined #openstack-neutron | 00:31 | |
| enikanorov_ | kevinbenton: could take a look at ^^ | 00:31 |
| *** zhhuabj has quit IRC | 00:31 | |
| *** zhhuabj has joined #openstack-neutron | 00:32 | |
| enikanorov_ | please look over the discussion in previos patchset, folks tried to push me to refactor looping call, which I'd prefer to do afterwards | 00:32 |
| enikanorov_ | because otherwise it would complicate fix backporting | 00:32 |
| *** zhenguo has joined #openstack-neutron | 00:33 | |
| *** tiswanso has joined #openstack-neutron | 00:33 | |
| *** Sukhdev has quit IRC | 00:35 | |
| *** sripriya has quit IRC | 00:35 | |
| *** shwetaap has quit IRC | 00:35 | |
| openstackgerrit | Aaron Rosen proposed openstack/neutron: Remove extra indent in testcases https://review.openstack.org/187827 | 00:36 |
| *** aepifanov has joined #openstack-neutron | 00:36 | |
| *** yeungp has quit IRC | 00:37 | |
| *** tiswanso has quit IRC | 00:38 | |
| openstackgerrit | Sridhar Ramaswamy proposed openstack/neutron-specs: Add Dynamic Multipoint VPN to Neutron VPNaaS https://review.openstack.org/181563 | 00:38 |
| *** breitz has quit IRC | 00:45 | |
| *** breitz1 has joined #openstack-neutron | 00:45 | |
| *** shashankhegde has quit IRC | 00:47 | |
| *** fawadkhaliq has quit IRC | 00:48 | |
| *** salv-orlando has joined #openstack-neutron | 00:48 | |
| *** nlahouti has quit IRC | 00:49 | |
| *** padkrish has quit IRC | 00:52 | |
| *** padkrish has joined #openstack-neutron | 00:52 | |
| *** Swami has quit IRC | 00:53 | |
| *** salv-orlando has quit IRC | 00:55 | |
| *** csoukup has joined #openstack-neutron | 00:55 | |
| *** irenab_ has joined #openstack-neutron | 00:57 | |
| *** irenab has quit IRC | 00:58 | |
| *** irenab_ is now known as irenab | 00:58 | |
| *** johnbelamaric has quit IRC | 00:58 | |
| *** vivek-ebay has quit IRC | 01:02 | |
| *** padkrish has quit IRC | 01:03 | |
| *** Aish has quit IRC | 01:05 | |
| *** _cjones_ has quit IRC | 01:06 | |
| *** madhu_ak_ has quit IRC | 01:06 | |
| *** a_le has quit IRC | 01:07 | |
| sc68cal | kevinbenton: reading scrollback, back from dinner. Yes I'd say that dougwig's statement that a provider network is an admin setting the vlan tag rather than allowing auto allocation is a big part of the provider networking API extension use case | 01:07 |
| sc68cal | kevinbenton: you still need to use the Neutron DHCP agent - but yes you can skip the L3 agent | 01:07 |
| sc68cal | kevinbenton: since the DHCP agent is the only way you can guarantee that the IP addr that you see in the API is the one the guest instance actually gets, unless you're writing some custom code | 01:08 |
| sc68cal | also RIP Sam-I-Am's appendix | 01:09 |
| sc68cal | and amuller's lol | 01:09 |
| *** mlavalle has quit IRC | 01:09 | |
| *** superfly_ has joined #openstack-neutron | 01:12 | |
| *** spandhe has quit IRC | 01:13 | |
| *** superfly_ is now known as superflyy | 01:14 | |
| *** jbell8 has joined #openstack-neutron | 01:16 | |
| *** aepifanov has quit IRC | 01:19 | |
| *** aepifanov has joined #openstack-neutron | 01:19 | |
| *** amotoki has quit IRC | 01:21 | |
| *** sridhar_ram has quit IRC | 01:22 | |
| *** barra204 has quit IRC | 01:24 | |
| *** shakamunyi has quit IRC | 01:24 | |
| *** superflyy has quit IRC | 01:24 | |
| *** tiswanso has joined #openstack-neutron | 01:25 | |
| *** tiswanso has quit IRC | 01:27 | |
| *** tiswanso has joined #openstack-neutron | 01:27 | |
| *** yalie has joined #openstack-neutron | 01:33 | |
| *** yalie has left #openstack-neutron | 01:33 | |
| *** hitalia has quit IRC | 01:34 | |
| *** ijw has quit IRC | 01:34 | |
| *** ijw has joined #openstack-neutron | 01:35 | |
| *** stanzgy has joined #openstack-neutron | 01:36 | |
| *** erkules_ has joined #openstack-neutron | 01:38 | |
| *** ian_brown has quit IRC | 01:39 | |
| *** s3wong has quit IRC | 01:40 | |
| *** amotoki has joined #openstack-neutron | 01:41 | |
| *** baoli has quit IRC | 01:41 | |
| *** erkules has quit IRC | 01:41 | |
| *** vivek-ebay has joined #openstack-neutron | 01:42 | |
| *** sridhar_ram has joined #openstack-neutron | 01:44 | |
| *** amotoki has quit IRC | 01:45 | |
| *** erkules_ has quit IRC | 01:50 | |
| *** erkules_ has joined #openstack-neutron | 01:52 | |
| *** shwetaap has joined #openstack-neutron | 01:52 | |
| *** tflynn_ has joined #openstack-neutron | 01:53 | |
| *** jyuso has joined #openstack-neutron | 01:53 | |
| *** tflynn has quit IRC | 01:56 | |
| *** tflynn_ has quit IRC | 01:58 | |
| *** Networkn3rd has joined #openstack-neutron | 01:58 | |
| *** yuanying has joined #openstack-neutron | 02:03 | |
| *** ianbrown has joined #openstack-neutron | 02:04 | |
| *** amotoki has joined #openstack-neutron | 02:04 | |
| *** shashankhegde has joined #openstack-neutron | 02:06 | |
| *** shashankhegde has quit IRC | 02:06 | |
| *** vivek-ebay has quit IRC | 02:06 | |
| *** salv-orlando has joined #openstack-neutron | 02:08 | |
| openstackgerrit | Merged openstack/neutron: Check for 'removed' in port_info before reference https://review.openstack.org/187819 | 02:11 |
| *** a_le has joined #openstack-neutron | 02:11 | |
| *** sdake has quit IRC | 02:11 | |
| *** salv-orlando has quit IRC | 02:14 | |
| *** nlahouti has joined #openstack-neutron | 02:15 | |
| *** aepifanov has quit IRC | 02:16 | |
| *** yamahata has quit IRC | 02:17 | |
| *** sridhar_ram has quit IRC | 02:19 | |
| *** mohitsharma has joined #openstack-neutron | 02:20 | |
| *** shwetaap has quit IRC | 02:21 | |
| *** baoli has joined #openstack-neutron | 02:23 | |
| *** circ-user-KKe7g has quit IRC | 02:23 | |
| *** haigang has quit IRC | 02:26 | |
| *** haigang has joined #openstack-neutron | 02:27 | |
| *** madhu_ak has joined #openstack-neutron | 02:28 | |
| *** yuanying_ has joined #openstack-neutron | 02:28 | |
| *** annegentle has joined #openstack-neutron | 02:28 | |
| *** haigang has quit IRC | 02:30 | |
| *** yuanying has quit IRC | 02:31 | |
| *** haigang has joined #openstack-neutron | 02:32 | |
| *** achanda has joined #openstack-neutron | 02:32 | |
| *** tomoe_ has joined #openstack-neutron | 02:32 | |
| *** haigang has quit IRC | 02:33 | |
| *** haigang has joined #openstack-neutron | 02:34 | |
| *** ivar-laz_ has joined #openstack-neutron | 02:37 | |
| *** ivar-laz_ has quit IRC | 02:37 | |
| *** haigang has quit IRC | 02:38 | |
| *** Networkn3rd has quit IRC | 02:40 | |
| *** ivar-lazzaro has quit IRC | 02:40 | |
| *** nlahouti has quit IRC | 02:44 | |
| *** annegentle has quit IRC | 02:44 | |
| *** mchestr has joined #openstack-neutron | 02:44 | |
| *** haigang has joined #openstack-neutron | 02:46 | |
| *** aranjan has joined #openstack-neutron | 02:47 | |
| *** ashishjain has joined #openstack-neutron | 02:48 | |
| *** mchestr has quit IRC | 02:49 | |
| *** aranjan has quit IRC | 02:49 | |
| *** baoli has quit IRC | 02:49 | |
| *** ijw has quit IRC | 02:50 | |
| *** annegentle has joined #openstack-neutron | 02:51 | |
| *** annegentle has quit IRC | 02:51 | |
| *** X-Rob_ is now known as X-Rob | 02:54 | |
| *** oda-g has quit IRC | 02:55 | |
| *** dims__ has quit IRC | 02:56 | |
| *** achanda has quit IRC | 03:02 | |
| *** achanda has joined #openstack-neutron | 03:03 | |
| *** vivek-ebay has joined #openstack-neutron | 03:04 | |
| *** achanda has quit IRC | 03:08 | |
| *** tflynn has joined #openstack-neutron | 03:13 | |
| *** mohitsharma has quit IRC | 03:13 | |
| *** tflynn_ has joined #openstack-neutron | 03:14 | |
| *** yuanying has joined #openstack-neutron | 03:15 | |
| *** zhhuabj has quit IRC | 03:15 | |
| *** zhhuabj_ has joined #openstack-neutron | 03:16 | |
| *** yuanying_ has quit IRC | 03:16 | |
| *** mohitsharma has joined #openstack-neutron | 03:17 | |
| *** tflynn has quit IRC | 03:17 | |
| *** yuanying has quit IRC | 03:22 | |
| *** dboik has joined #openstack-neutron | 03:23 | |
| *** dsanders has quit IRC | 03:23 | |
| *** dboik has quit IRC | 03:27 | |
| *** achanda has joined #openstack-neutron | 03:34 | |
| *** spandhe has joined #openstack-neutron | 03:36 | |
| *** kiran-r has joined #openstack-neutron | 03:38 | |
| *** arosen1 has quit IRC | 03:41 | |
| *** spandhe_ has joined #openstack-neutron | 03:42 | |
| *** spandhe has quit IRC | 03:43 | |
| *** spandhe_ is now known as spandhe | 03:43 | |
| *** tiswanso has quit IRC | 03:43 | |
| *** markvoelker has quit IRC | 03:44 | |
| *** yfried|afk is now known as yfried_ | 03:51 | |
| *** debedb has joined #openstack-neutron | 03:51 | |
| *** yuanying has joined #openstack-neutron | 03:52 | |
| *** Longgeek_ has quit IRC | 03:53 | |
| *** alexpilotti has joined #openstack-neutron | 03:54 | |
| *** Longgeek has joined #openstack-neutron | 03:54 | |
| *** Aish has joined #openstack-neutron | 03:55 | |
| *** Aish has left #openstack-neutron | 03:55 | |
| *** vivek-ebay has quit IRC | 03:56 | |
| *** Longgeek_ has joined #openstack-neutron | 03:57 | |
| *** achanda has quit IRC | 03:58 | |
| *** achanda has joined #openstack-neutron | 03:58 | |
| *** fedexo has joined #openstack-neutron | 03:58 | |
| *** madhu_ak has quit IRC | 03:59 | |
| *** changbl_ has joined #openstack-neutron | 04:00 | |
| *** Longgeek has quit IRC | 04:00 | |
| *** Longgeek has joined #openstack-neutron | 04:02 | |
| *** yfried_ is now known as yfried|afk | 04:02 | |
| *** yatin has joined #openstack-neutron | 04:03 | |
| *** a_le has quit IRC | 04:03 | |
| *** achanda has quit IRC | 04:03 | |
| *** Longgeek_ has quit IRC | 04:04 | |
| *** debedb has quit IRC | 04:06 | |
| *** yfried|afk is now known as yfried_ | 04:06 | |
| *** a_le has joined #openstack-neutron | 04:07 | |
| *** Longgeek_ has joined #openstack-neutron | 04:08 | |
| *** Swami has joined #openstack-neutron | 04:09 | |
| *** yuanying has quit IRC | 04:10 | |
| *** salv-orlando has joined #openstack-neutron | 04:10 | |
| *** Longgeek has quit IRC | 04:10 | |
| kevinbenton | sc68cal: thanks. so what do you think someone means when they say "I'm using provider networks" ? | 04:11 |
| salv-orlando | kevinbenton: so you're saying we do multiple deletes because "better safe than sorry"? | 04:11 |
| *** yuanying has joined #openstack-neutron | 04:12 | |
| kevinbenton | salv-orlando: yeah, we're extra secure that way :) | 04:12 |
| salv-orlando | because you know, database are unpredictable and sometimes do not honour ON DELETE CASCADE ;) | 04:12 |
| *** Longgeek_ has quit IRC | 04:12 | |
| salv-orlando | personally I just did that (the orm.relationship cascade) out of ignorance, but others might have been doing that on purpose | 04:13 |
| salv-orlando | kevinbenton: so you think that's a feature and not a bug? | 04:14 |
| kevinbenton | salv-orlando: i highly doubt it. i've done it just based on mimicking other code :) | 04:14 |
| salv-orlando | kevinbenton: same here... I actually thought that it was a way to reinforce the foreign key constrain you already applied | 04:15 |
| kevinbenton | salv-orlando: "neutron offers secure™ record deletion" | 04:15 |
| *** Swami_ has joined #openstack-neutron | 04:15 | |
| kevinbenton | salv-orlando: so basically sqlalchemy is offering a way to do cascading without actually having a db that supports it? | 04:16 |
| kevinbenton | that's what i understood from your bug report at least | 04:16 |
| salv-orlando | kevinbenton: it can be read in that way. The doc say it's application-level cascading | 04:17 |
| salv-orlando | so it's not enforced with DB constraint but within ORM. | 04:17 |
| *** Swami has quit IRC | 04:17 | |
| kevinbenton | salv-orlando: i see | 04:17 |
| *** madhu_ak has joined #openstack-neutron | 04:17 | |
| salv-orlando | Now, if you have a relational DB, I don't see a reason for which you'd like to not use db-level cascading | 04:17 |
| *** Longgeek has joined #openstack-neutron | 04:18 | |
| salv-orlando | and that's what sqlalchemy docs say too | 04:18 |
| *** Longgeek has quit IRC | 04:18 | |
| salv-orlando | when I filed the bug zzeek was not available for comment ;) | 04:18 |
| *** yuanying has quit IRC | 04:18 | |
| *** Longgeek has joined #openstack-neutron | 04:19 | |
| *** yuanying has joined #openstack-neutron | 04:19 | |
| *** yuanying has quit IRC | 04:20 | |
| kevinbenton | salv-orlando: which side do you define the cascade behavior on for sqlalchemy? | 04:20 |
| *** Longgeek has quit IRC | 04:20 | |
| kevinbenton | salv-orlando: could you use it to have it go clean something out of a global table based on UUIDs? | 04:20 |
| *** tomoe_ has quit IRC | 04:20 | |
| *** tomoe__ has joined #openstack-neutron | 04:20 | |
| *** Longgeek has joined #openstack-neutron | 04:20 | |
| *** jbell8 has quit IRC | 04:21 | |
| kevinbenton | salv-orlando: since the global table could have entries for all sorts of objects | 04:21 |
| *** Swami_ has quit IRC | 04:21 | |
| *** Longgeek_ has joined #openstack-neutron | 04:23 | |
| *** yuanying_ has joined #openstack-neutron | 04:24 | |
| *** banix has quit IRC | 04:24 | |
| *** sthillma has joined #openstack-neutron | 04:24 | |
| *** Longgee__ has joined #openstack-neutron | 04:25 | |
| *** Longgeek has quit IRC | 04:25 | |
| *** yuanying_ has quit IRC | 04:25 | |
| *** sthillma_ has joined #openstack-neutron | 04:25 | |
| *** dsanders has joined #openstack-neutron | 04:26 | |
| *** Longgeek_ has quit IRC | 04:26 | |
| *** Longgeek has joined #openstack-neutron | 04:26 | |
| *** yuanying has joined #openstack-neutron | 04:27 | |
| *** Longgeek_ has joined #openstack-neutron | 04:28 | |
| *** madhu_ak has quit IRC | 04:29 | |
| *** sthillma has quit IRC | 04:29 | |
| *** Longge___ has joined #openstack-neutron | 04:29 | |
| *** Longgee__ has quit IRC | 04:30 | |
| *** coolsvap|afk is now known as coolsvap | 04:30 | |
| *** coolsvap is now known as coolsvap|afk | 04:31 | |
| *** Longgeek has quit IRC | 04:32 | |
| *** coolsvap|afk is now known as coolsvap | 04:32 | |
| *** Longgeek_ has quit IRC | 04:32 | |
| *** Swami has joined #openstack-neutron | 04:38 | |
| *** changbl_ has quit IRC | 04:40 | |
| *** krtaylor has quit IRC | 04:40 | |
| salv-orlando | do you mean having a single table called "properties" whose record have uuids associated with several tables, like, for instance, "toast" and "yoghurt"? | 04:40 |
| salv-orlando | kevinbenton: ^ ^ | 04:40 |
| salv-orlando | sorry about the example, having breakfast | 04:40 |
| kevinbenton | salv-orlando: precisely | 04:40 |
| salv-orlando | yeah I guess that would be a case for orm-level cascade | 04:41 |
| *** spandhe has quit IRC | 04:41 | |
| kevinbenton | salv-orlando: i've been looking at the ACL code and was trying to figure out how to have a generic ACL entries table that gets cleaned up when the corresponding item gets deleted | 04:43 |
| *** yfried_ has quit IRC | 04:44 | |
| *** markvoelker has joined #openstack-neutron | 04:45 | |
| *** abregman has joined #openstack-neutron | 04:45 | |
| *** yuanying has quit IRC | 04:45 | |
| *** _cjones_ has joined #openstack-neutron | 04:48 | |
| *** yuanying has joined #openstack-neutron | 04:48 | |
| *** lori|away is now known as lori | 04:49 | |
| *** markvoelker has quit IRC | 04:50 | |
| *** Swami has quit IRC | 04:50 | |
| *** oda-g has joined #openstack-neutron | 04:51 | |
| *** ildikov has quit IRC | 04:54 | |
| *** _cjones_ has quit IRC | 04:55 | |
| *** jbell8 has joined #openstack-neutron | 04:56 | |
| *** kiran-r has quit IRC | 04:59 | |
| *** mohitsharma has quit IRC | 05:05 | |
| *** ashishjain has quit IRC | 05:07 | |
| *** ildikov has joined #openstack-neutron | 05:13 | |
| *** dannywilson has joined #openstack-neutron | 05:13 | |
| salv-orlando | kevinbenton: are you already implementing ACLs for read/write rights on resources? | 05:16 |
| *** yamahata has joined #openstack-neutron | 05:19 | |
| *** dannywilson has quit IRC | 05:25 | |
| *** dannywilson has joined #openstack-neutron | 05:25 | |
| *** ekuris has joined #openstack-neutron | 05:26 | |
| *** krtaylor has joined #openstack-neutron | 05:27 | |
| *** apuimedo has quit IRC | 05:27 | |
| kevinbenton | salv-orlando: just starting on it | 05:28 |
| kevinbenton | salv-orlando: well the network sharing thing (assuming that's what you're talking about) | 05:28 |
| *** sthillma_ has quit IRC | 05:28 | |
| kevinbenton | salv-orlando: but i'm still not sure whether or not to jam most of everything into one table | 05:28 |
| *** yfried_ has joined #openstack-neutron | 05:28 | |
| kevinbenton | salv-orlando: or just to have a full ACL table for each type | 05:29 |
| *** dsanders has quit IRC | 05:30 | |
| lizk | A basic question about LBaaS, do we really need members are from the same subnet with pool's, can anybody give some strong enough explanation? | 05:33 |
| gsagie | lizk : just wondering, can you give the opposite explanation why it needs to be from different subnet? (i think you would most likely want to pick members from the same subnet with a similar VIP) | 05:36 |
| *** ildikov has quit IRC | 05:38 | |
| *** moshele has joined #openstack-neutron | 05:42 | |
| *** fedexo has quit IRC | 05:42 | |
| lizk | gsagie : hi, first, I don't know, I just found that there are just members' IPs, not subnet_ids in HAProxy configure file, and some simple test I did also shown me that in case members are from different subnet from pool's can still work. So I come to ask for help : ) | 05:44 |
| openstackgerrit | Oleg Bondarev proposed openstack/neutron: Wrap ML2 delete_port with db retry decorator https://review.openstack.org/180466 | 05:44 |
| lizk | and another question comes to me, when and where will pool's subnet_id works in neutron-lbaas, I did a simple quick scan in neutron-lbaas, but didn't find anything helpful... | 05:45 |
| rm_work | lizk: ask in #openstack-lbaas channel | 05:46 |
| lizk | rm_work: cool, thx, I will | 05:46 |
| rm_work | lizk: the subnet_id on the pool controls plugging networks, as i said before | 05:46 |
| *** dsanders has joined #openstack-neutron | 05:46 | |
| rm_work | so, if you have members from another subnet, it MIGHT work -- if the other network happens to be plugged for some other reason | 05:47 |
| lizk | rm_work: yeah, I remember that, but I'm still confusing on that.. | 05:47 |
| rm_work | but the pool's subnet_id is what actually forces additional networks to be plugged | 05:47 |
| rm_work | ok so, imagine you have a nova VM | 05:47 |
| rm_work | you spin it up on one network, "public-net" | 05:47 |
| rm_work | and it runs HAProxy | 05:48 |
| rm_work | you want to loadbalance to nodes that are on the network "my-private-net" | 05:48 |
| rm_work | somehow that network needs to be plugged to the VM | 05:48 |
| salv-orlando | kevinbenton: that's a good question, but if you are doing it for networks only you can reserve the right to make a call at later stage. | 05:49 |
| rm_work | so, the way the lbaas system knows it needs to plug that network is because of the subnet_id on the pool | 05:49 |
| *** jyuso has quit IRC | 05:49 | |
| rm_work | lizk: if you wanted to loadbalance nodes that are on the public internet, obviously it would work regardless of what subnet_id you put in the pool | 05:49 |
| *** veena has joined #openstack-neutron | 05:50 | |
| rm_work | because the VM already has the public-net pkugged | 05:50 |
| rm_work | *plugged | 05:50 |
| *** sudipto has joined #openstack-neutron | 05:50 | |
| *** ihrachyshka has joined #openstack-neutron | 05:50 | |
| rm_work | it doesn't really ENFORCE that the members of a pool belong to that subnet, but we only guarantee routing if they are | 05:50 |
| *** yuanying has quit IRC | 05:51 | |
| rm_work | lizk: i don't know if i actually managed to explain any better or not :/ | 05:51 |
| rm_work | this is a question we have gotten quite a few times | 05:51 |
| *** itzikb has joined #openstack-neutron | 05:51 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Revert "Add VIF_DELETED notification event to Nova" https://review.openstack.org/187868 | 05:52 |
| lizk | greet, thanks for the better explanation you given this time : ) | 05:53 |
| kevinbenton | salv-orlando: makes sense. i best not waste too much time trying to prematurely abstract | 05:53 |
| kevinbenton | salv-orlando: assuming that's a saying :) | 05:54 |
| lizk | rm_work: I get some imagine now, | 05:54 |
| *** vthapar has joined #openstack-neutron | 05:54 | |
| openstackgerrit | watanabe isao proposed openstack/neutron: Bug-Fix for unexpected DHCP agent redundant https://review.openstack.org/138659 | 05:55 |
| *** dguerri`away has quit IRC | 05:55 | |
| *** mohitsharma has joined #openstack-neutron | 05:56 | |
| *** itzikb has quit IRC | 05:56 | |
| *** dguerri`away has joined #openstack-neutron | 05:57 | |
| *** dguerri`away is now known as dguerri | 05:57 | |
| *** itzikb has joined #openstack-neutron | 05:59 | |
| *** josecastroleon has joined #openstack-neutron | 05:59 | |
| *** armax has quit IRC | 06:01 | |
| *** SridharG has joined #openstack-neutron | 06:01 | |
| *** ihrachyshka has quit IRC | 06:02 | |
| *** dsanders has quit IRC | 06:03 | |
| *** ildikov has joined #openstack-neutron | 06:04 | |
| *** zhhuabj has joined #openstack-neutron | 06:05 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Revert "Revert "Add VIF_DELETED notification event to Nova"" https://review.openstack.org/187871 | 06:06 |
| *** dsanders has joined #openstack-neutron | 06:07 | |
| *** stefanb has quit IRC | 06:08 | |
| *** zhhuabj_ has quit IRC | 06:09 | |
| openstackgerrit | Ian Wienand proposed openstack/neutron: Ensure netfilter is enabled for bridges https://review.openstack.org/187873 | 06:10 |
| *** kiran-r has joined #openstack-neutron | 06:10 | |
| *** Bhavya has joined #openstack-neutron | 06:16 | |
| *** shakamunyi has joined #openstack-neutron | 06:16 | |
| *** barra204 has joined #openstack-neutron | 06:17 | |
| *** csoukup has quit IRC | 06:17 | |
| *** yamamoto has joined #openstack-neutron | 06:17 | |
| *** dannywilson has quit IRC | 06:18 | |
| *** smallbig has quit IRC | 06:19 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: ovs agent resync may miss port remove event https://review.openstack.org/165023 | 06:20 |
| *** stefanb has joined #openstack-neutron | 06:22 | |
| *** Bhavya has quit IRC | 06:24 | |
| *** bjornar_ has quit IRC | 06:25 | |
| *** moshele is now known as moshele_ | 06:28 | |
| *** moshele_ has quit IRC | 06:28 | |
| *** moshele has joined #openstack-neutron | 06:29 | |
| *** mohitsharma has quit IRC | 06:29 | |
| *** mohitsharma has joined #openstack-neutron | 06:29 | |
| *** a_le has quit IRC | 06:29 | |
| *** smallbig has joined #openstack-neutron | 06:30 | |
| *** dsanders has quit IRC | 06:32 | |
| *** markvoelker has joined #openstack-neutron | 06:34 | |
| *** scheuran has joined #openstack-neutron | 06:34 | |
| *** ianbrown has quit IRC | 06:35 | |
| *** Longgeek has joined #openstack-neutron | 06:35 | |
| *** lizk has quit IRC | 06:37 | |
| *** yuanying has joined #openstack-neutron | 06:37 | |
| *** Longge___ has quit IRC | 06:37 | |
| *** markvoelker has quit IRC | 06:38 | |
| *** numan has joined #openstack-neutron | 06:39 | |
| *** dsanders has joined #openstack-neutron | 06:48 | |
| *** ronis has joined #openstack-neutron | 06:49 | |
| *** anshul has joined #openstack-neutron | 06:50 | |
| *** moha_hunt has joined #openstack-neutron | 06:50 | |
| *** anshul has quit IRC | 06:51 | |
| *** anshul has joined #openstack-neutron | 06:51 | |
| *** pradipta has joined #openstack-neutron | 06:55 | |
| *** neeti has joined #openstack-neutron | 06:56 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: Make MockFixedIntervalLoopingCall class as a helper class https://review.openstack.org/187885 | 06:59 |
| *** yfried_ has quit IRC | 06:59 | |
| *** yfried has joined #openstack-neutron | 06:59 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: Make MockFixedIntervalLoopingCall class as a helper class https://review.openstack.org/187885 | 07:02 |
| *** kawa2014 has joined #openstack-neutron | 07:02 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: Make MockFixedIntervalLoopingCall class as a helper class https://review.openstack.org/187885 | 07:02 |
| *** yfried is now known as yfried|afk | 07:10 | |
| *** fzdarsky has joined #openstack-neutron | 07:12 | |
| *** ushkalim has joined #openstack-neutron | 07:14 | |
| *** vikram has joined #openstack-neutron | 07:15 | |
| *** neeti_ has joined #openstack-neutron | 07:15 | |
| *** Nakato has quit IRC | 07:16 | |
| *** Nakato has joined #openstack-neutron | 07:18 | |
| *** neeti has quit IRC | 07:18 | |
| *** abregman has quit IRC | 07:18 | |
| *** haigang has quit IRC | 07:19 | |
| *** alex_klimov has joined #openstack-neutron | 07:19 | |
| *** bradjones has quit IRC | 07:20 | |
| *** jpena has joined #openstack-neutron | 07:20 | |
| *** yfried|afk is now known as yfried | 07:20 | |
| *** dsanders has quit IRC | 07:21 | |
| openstackgerrit | shihanzhang proposed openstack/neutron: Send 'security_groups_member_updated' when port'ip changes https://review.openstack.org/177159 | 07:21 |
| *** rotbeard has joined #openstack-neutron | 07:22 | |
| *** bradjones has joined #openstack-neutron | 07:22 | |
| *** yuanying has quit IRC | 07:22 | |
| *** dsanders has joined #openstack-neutron | 07:23 | |
| *** mohitsharma has quit IRC | 07:27 | |
| *** matrohon has joined #openstack-neutron | 07:27 | |
| *** mohitsharma has joined #openstack-neutron | 07:27 | |
| *** smallbig has quit IRC | 07:28 | |
| *** jyuso1 has joined #openstack-neutron | 07:29 | |
| *** abregman has joined #openstack-neutron | 07:31 | |
| *** neeti_ has quit IRC | 07:32 | |
| *** neeti has joined #openstack-neutron | 07:33 | |
| *** moha_hunt has quit IRC | 07:35 | |
| *** chlong has quit IRC | 07:37 | |
| *** coolsvap is now known as coolsvap|afk | 07:41 | |
| openstackgerrit | Yuuichi Fujioka proposed openstack/neutron: Introduce an API test for specified floating ip address https://review.openstack.org/181606 | 07:45 |
| *** coolsvap|afk is now known as coolsvap | 07:46 | |
| *** nmagnezi has joined #openstack-neutron | 07:47 | |
| openstackgerrit | Moshe Levi proposed openstack/neutron-specs: ML2/SR-IOV QoS support with bandwidth limiting https://review.openstack.org/187895 | 07:48 |
| *** devvesa has joined #openstack-neutron | 07:50 | |
| *** jistr has joined #openstack-neutron | 07:51 | |
| *** itzikb has quit IRC | 07:53 | |
| *** ygbo has joined #openstack-neutron | 07:56 | |
| *** dims_ has joined #openstack-neutron | 07:58 | |
| *** rossella_s has joined #openstack-neutron | 08:02 | |
| *** dims_ has quit IRC | 08:03 | |
| *** jlanoux has joined #openstack-neutron | 08:05 | |
| *** stanzgy has quit IRC | 08:05 | |
| *** stanzgy__ has joined #openstack-neutron | 08:05 | |
| *** tomoe__ has quit IRC | 08:05 | |
| *** tomoe_ has joined #openstack-neutron | 08:06 | |
| *** mkolesni has joined #openstack-neutron | 08:07 | |
| *** itzikb has joined #openstack-neutron | 08:07 | |
| ajo | mkolesni <-> hanzhang | 08:07 |
| ajo | :) | 08:07 |
| ajo | hanzhang was willing to talk with you about distributed DHCP :) | 08:07 |
| *** tflynn_ has quit IRC | 08:10 | |
| ajo | mkolesni, as far as I understood from hanzhang , he has a demo | 08:12 |
| ajo | hanzhang++ | 08:12 |
| mkolesni | ajo: well that spec wasn't approved for kilo and had some work to be done | 08:13 |
| *** Nakato has quit IRC | 08:13 | |
| mkolesni | ajo: so i'm not sure how much work is still needed just to get the spec right | 08:13 |
| *** Nakato has joined #openstack-neutron | 08:14 | |
| *** tomoe_ has quit IRC | 08:14 | |
| *** itzikb_ has joined #openstack-neutron | 08:14 | |
| ajo | mkolesni, discuss with hanzhang ;) he's a surprising guy :) | 08:15 |
| *** itzikb has quit IRC | 08:15 | |
| hanzhang | mkolesni: hi | 08:16 |
| *** ArchiFleKs has quit IRC | 08:16 | |
| mkolesni | hanzhang: hi | 08:16 |
| *** ArchiFleKs has joined #openstack-neutron | 08:17 | |
| hanzhang | mkolesni: I am very glad to see you on IRC :) | 08:17 |
| mkolesni | hanzhang: thanks :) | 08:17 |
| hanzhang | what's your plan for this BP:https://blueprints.launchpad.net/neutron/+spec/distributed-dhcp in L cycle | 08:18 |
| openstackgerrit | Ian Wienand proposed openstack/neutron: Ensure netfilter is enabled for bridges https://review.openstack.org/187873 | 08:18 |
| mkolesni | hanzhang: i don't know if i'll have time to work on it | 08:19 |
| mkolesni | hanzhang: but i can help with reviews | 08:19 |
| hanzhang | mkolesni: thx very much, your idea for distributed-dhcp is very cool, so I want to implement it in L cycle | 08:21 |
| *** apuimedo has joined #openstack-neutron | 08:21 | |
| ajo | hanzhang, if you have time, may be now we can follow the RFE process and the slimmed down specs... :) | 08:21 |
| mkolesni | hanzhang: sure, you can submit it for review to liberty cycle, just make sure to address comments from kilo if theres anything needs changing | 08:22 |
| ajo | specially if you have some demo/poc, that should help illustrate what are the plans on the low level. | 08:22 |
| *** nyechiel_ has joined #openstack-neutron | 08:22 | |
| mkolesni | ajo: im not sure how much can be slimmed down, it's a pretty extensive change | 08:22 |
| mkolesni | hanzhang, ajo: depends if you plan to conserve ips or not | 08:22 |
| *** markvoelker has joined #openstack-neutron | 08:22 | |
| *** dsanders has quit IRC | 08:23 | |
| ajo | hanzhang: https://github.com/openstack/neutron/blob/master/doc/source/policies/blueprints.rst#rfe-submission-guidelines | 08:23 |
| *** moha_hunt has joined #openstack-neutron | 08:24 | |
| ajo | mkolesni, may be IP conservation can be done iteratively, since, tenant-net ips are cheap... may be it's just a documentation issue | 08:24 |
| ajo | and that can be worked out in a second iteration | 08:24 |
| mkolesni | ajo: thats what i was proposing initially but im not sure the core team would approve of such an approach | 08:24 |
| ajo | mkolesni, what are the complexities of saving IPs? | 08:25 |
| hanzhang | mkolesni, ajo, I would like to do this work, if you agree, I can add you to co-author | 08:25 |
| mkolesni | ajo: u need to make changes at the agent level, so for ovs it means extra flows | 08:25 |
| ajo | mkolesni, to prevent the DHCP traffic flying out of the node, right? | 08:25 |
| ajo | mkolesni, and same with iptables for LB | 08:25 |
| ajo | but then I wonder how to handle that for all the other implementations (SR-IOV, and all the other plugins...) | 08:26 |
| ajo | I guess it may only work with plugins able to isolate ports into "node only" | 08:26 |
| ajo | may be we need a property on the port first | 08:26 |
| mkolesni | ajo: well it would be configurable if youre running it like that or not | 08:26 |
| *** markvoelker has quit IRC | 08:27 | |
| *** jbell8 has quit IRC | 08:27 | |
| mkolesni | ajo: i dunno if it's something i'd put on the port, seems to me like a deployment decision | 08:27 |
| ajo | mkolesni it could be a port extension "node_only_port" or something like that... | 08:27 |
| ajo | mkolesni, so the plugins not implementing such extension won't be able to do distributed-dhcp | 08:28 |
| ajo | because port creation would fail | 08:28 |
| *** ZZelle has quit IRC | 08:28 | |
| mkolesni | ajo: i think even if they dont support it they should allow distributing without ip conservation | 08:28 |
| *** itzikb_ has quit IRC | 08:28 | |
| *** ZZelle has joined #openstack-neutron | 08:28 | |
| mkolesni | ajo: thats why initially i thought of config values.. | 08:29 |
| *** itzikb_ has joined #openstack-neutron | 08:29 | |
| ajo | mkolesni, that sounds reasonable to me, | 08:29 |
| ajo | mkolesni since... tenant network ips are cheap.. | 08:30 |
| hanzhang | I agree with mkolesni that use a config values | 08:30 |
| ajo | in that case it's important to recommend tenants not using /24 CIDRs | 08:30 |
| ajo | on big deployments | 08:30 |
| mkolesni | hanzhang: anyway you can submit the spec to liberty and add me as co author | 08:30 |
| ajo | mkolesni, hanzhang , now new liberty specs need to go via RFE | 08:31 |
| ajo | I'd do that first | 08:31 |
| mkolesni | ajo: well u know it can also be mitigated a bit if scheduling is integrated nicely | 08:31 |
| ajo | and current specs under evaluation have a 1-month deadline, otherwise they need to jump back to rfe | 08:31 |
| mkolesni | ajo: currently we don't know when a vm is down and not using a port.. | 08:32 |
| ajo | mkolesni, true | 08:32 |
| mkolesni | ajo: if we knew that we could shutdown the dhcp on inactive compute nodes | 08:32 |
| *** txomon|fon has joined #openstack-neutron | 08:32 | |
| mkolesni | but then again, i'm not sure if that's worth the hassle :) | 08:32 |
| hanzhang | mkolesni: do you meant a vm migrate from one node to another? | 08:33 |
| ajo | mkolesni, also, I guess we may need better rpc mechanisms to distribute net updates | 08:33 |
| ajo | like topics | 08:33 |
| ajo | otherwise you need to replicate info to every agent | 08:33 |
| mkolesni | i.e. probably best to spend that time developing the ip concerving approach | 08:33 |
| mkolesni | hanzhang: no that whould be handled | 08:33 |
| mkolesni | hanzhang: when vm is shut down the port is still up and running | 08:34 |
| mkolesni | from neutron's perspective | 08:34 |
| *** heyongli has quit IRC | 08:34 | |
| mkolesni | again this might be not aorthwhile persuing as im not sure it's cost effective | 08:34 |
| *** stanzgy__ has quit IRC | 08:35 | |
| *** tomoe_ has joined #openstack-neutron | 08:38 | |
| openstackgerrit | Merged openstack/neutron: Revert "Add VIF_DELETED notification event to Nova" https://review.openstack.org/187868 | 08:39 |
| *** ianbrown has joined #openstack-neutron | 08:41 | |
| hanzhang | mkolesni: I dont't get a idea for this case, but it's not a common case :) | 08:42 |
| mkolesni | hanzhang: what case? | 08:43 |
| hanzhang | mkolesni: when all VMs which belongs to a network shutdown on a compute node, then we shudown the dhcp server | 08:46 |
| mkolesni | hanzhang: well i dont know how common it is or not, but i think it doesn't matter much anyway since it's better to put the effort into conserving ips for example | 08:47 |
| mkolesni | hanzhang: or just not doing anything in phase 1 and just let there be as many ips consumed as there are compute nodes | 08:47 |
| mkolesni | hanzhang: sounds scary but if the community can be convinced that tenant ip's are cheap (which is mostly true) then great | 08:48 |
| mkolesni | hanzhang: btw i had some email thread about it a while ago, and operators seemed to insist on the one ip solution | 08:48 |
| *** heyongli has joined #openstack-neutron | 08:49 | |
| *** blues-man has joined #openstack-neutron | 08:49 | |
| *** jlibosva has joined #openstack-neutron | 08:49 | |
| hanzhang | mkolesni: I also agree the one ip solution | 08:50 |
| *** heyongli has quit IRC | 08:51 | |
| mkolesni | hanzhang: https://openstack.nimeyo.com/36185/openstack-operators-request-for-feedback-on-dhcp-ip-usage | 08:51 |
| mkolesni | hanzhang: yeah ok but it would be much easier to push to liberty without it | 08:51 |
| mkolesni | hanzhang: in terms of sheer amount of work to do | 08:52 |
| mkolesni | hanzhang: and then best effor it to liberty and if not then M cycle | 08:52 |
| mkolesni | hanzhang: just my perspective on how likely it is to push this to liberty | 08:53 |
| hanzhang | let me check it, I have made a simple demo for one ip solution: https://review.openstack.org/#/c/184423/, you can read it firstly | 08:53 |
| *** ianbrown has quit IRC | 08:56 | |
| *** nicktgn has joined #openstack-neutron | 08:56 | |
| *** safchain has joined #openstack-neutron | 08:57 | |
| *** kawa2014 has quit IRC | 09:00 | |
| *** moha_hunt has quit IRC | 09:01 | |
| *** nicktgn has quit IRC | 09:03 | |
| mkolesni | hanzhang: well these flows seem very broad, blocking any dhcp traffic from crossing the node boundary | 09:06 |
| *** heyongli has joined #openstack-neutron | 09:06 | |
| mkolesni | hanzhang: obviously in the real proposal it will have to be more precise flows | 09:06 |
| *** saggi has joined #openstack-neutron | 09:07 | |
| mkolesni | hanzhang: also i'm not sure that another dvr style change is gonna fly.. | 09:07 |
| hanzhang | mkolesni: yes, this is simple demo :) | 09:07 |
| mkolesni | hanzhang: it's better to help push first a solution to support distributed ports in neutron, than clone dvr | 09:07 |
| *** heyongli has quit IRC | 09:07 | |
| *** dguitarbite has quit IRC | 09:09 | |
| *** vthapar has quit IRC | 09:10 | |
| *** vthapar has joined #openstack-neutron | 09:11 | |
| *** txomon|fon has quit IRC | 09:12 | |
| hanzhang | mkolesni: I agree the current solution for distributed port in dvr is not good | 09:14 |
| *** pavel_bondar has quit IRC | 09:14 | |
| *** stanzgy has joined #openstack-neutron | 09:14 | |
| *** kawa2014 has joined #openstack-neutron | 09:14 | |
| *** nyt1066 has joined #openstack-neutron | 09:15 | |
| openstackgerrit | Wim De Clercq proposed openstack/neutron: Add Nuage plugin configuration parameters. https://review.openstack.org/187922 | 09:15 |
| *** claudiub has joined #openstack-neutron | 09:17 | |
| *** oreillyd has joined #openstack-neutron | 09:21 | |
| *** afazekas has joined #openstack-neutron | 09:22 | |
| *** txomon|fon has joined #openstack-neutron | 09:27 | |
| *** pavel_bondar has joined #openstack-neutron | 09:28 | |
| *** lazy_prince has joined #openstack-neutron | 09:28 | |
| *** yamamoto has quit IRC | 09:29 | |
| *** ianbrown has joined #openstack-neutron | 09:47 | |
| *** pradipta has quit IRC | 09:48 | |
| *** txomon|fon has quit IRC | 09:49 | |
| *** sudipto has quit IRC | 09:51 | |
| *** veena has quit IRC | 09:53 | |
| *** ianbrown has quit IRC | 09:56 | |
| *** ronis has quit IRC | 09:58 | |
| *** dguitarbite has joined #openstack-neutron | 09:59 | |
| *** txomon|fon has joined #openstack-neutron | 10:02 | |
| *** yfujioka_ has quit IRC | 10:02 | |
| *** erkules_ is now known as erkules | 10:02 | |
| *** erkules has joined #openstack-neutron | 10:02 | |
| openstackgerrit | Merged openstack/neutron: Catch broad exception in methods used in FixedIntervalLoopingCall https://review.openstack.org/185722 | 10:05 |
| *** dims_ has joined #openstack-neutron | 10:06 | |
| *** kawa2014 has quit IRC | 10:06 | |
| *** kawa2014 has joined #openstack-neutron | 10:07 | |
| *** mohitsharma has quit IRC | 10:08 | |
| *** saggi has quit IRC | 10:08 | |
| *** gildub has quit IRC | 10:08 | |
| *** tomoe_ has quit IRC | 10:09 | |
| *** saggi has joined #openstack-neutron | 10:09 | |
| *** markvoelker has joined #openstack-neutron | 10:11 | |
| *** dr_gogeta86 has joined #openstack-neutron | 10:14 | |
| *** markvoelker has quit IRC | 10:15 | |
| dr_gogeta86 | hi guys | 10:16 |
| dr_gogeta86 | i got a lil problem with neutron policy.json | 10:16 |
| dr_gogeta86 | i wanna deny router updates from certain users | 10:17 |
| *** pc_m has joined #openstack-neutron | 10:19 | |
| openstackgerrit | Oleg Bondarev proposed openstack/neutron: Fix callback registry notification for security group rule https://review.openstack.org/187586 | 10:20 |
| ajo | dr_gogeta86, AFAIK, that's yet not possible | 10:20 |
| ajo | you can allow/disallow admin / owner / general | 10:21 |
| ajo | but I'm not completely sure | 10:21 |
| ajo | that's what the RBAC proposal is chasing | 10:21 |
| *** lazy_prince has quit IRC | 10:22 | |
| ajo | dr_gogeta86: https://review.openstack.org/#/c/132661/ | 10:24 |
| ajo | still this spec targets networks only, not routers yet | 10:24 |
| *** merooney has joined #openstack-neutron | 10:24 | |
| *** mwagner_afk has quit IRC | 10:26 | |
| *** nyt1066 has quit IRC | 10:29 | |
| *** qianli_ has quit IRC | 10:39 | |
| *** amotoki has quit IRC | 10:40 | |
| *** sudipto has joined #openstack-neutron | 10:41 | |
| *** sudipto has quit IRC | 10:42 | |
| *** sudipto has joined #openstack-neutron | 10:43 | |
| *** sudipto has quit IRC | 10:44 | |
| *** amotoki has joined #openstack-neutron | 10:44 | |
| *** sudipto has joined #openstack-neutron | 10:45 | |
| *** amotoki has quit IRC | 10:48 | |
| *** abregman has quit IRC | 10:49 | |
| *** stanzgy has quit IRC | 10:49 | |
| *** sudipto has quit IRC | 10:49 | |
| *** abregman has joined #openstack-neutron | 10:50 | |
| *** jyuso1 has quit IRC | 10:52 | |
| *** itzikb_ has quit IRC | 10:56 | |
| *** dsneddon has quit IRC | 10:57 | |
| *** yfried is now known as yfried|afk | 10:58 | |
| *** ronis has joined #openstack-neutron | 11:00 | |
| *** itzikb_ has joined #openstack-neutron | 11:00 | |
| *** sdake has joined #openstack-neutron | 11:02 | |
| *** skraynev_ has joined #openstack-neutron | 11:04 | |
| *** julim has joined #openstack-neutron | 11:06 | |
| *** itzikb_ has quit IRC | 11:06 | |
| *** sudipto has joined #openstack-neutron | 11:07 | |
| openstackgerrit | Gal Sagie proposed openstack/neutron: Add Geneve type driver support to ML2 https://review.openstack.org/187945 | 11:11 |
| *** yfried|afk is now known as yfried | 11:13 | |
| *** merooney has quit IRC | 11:16 | |
| *** aix has quit IRC | 11:25 | |
| *** merooney has joined #openstack-neutron | 11:27 | |
| *** itzikb has joined #openstack-neutron | 11:30 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 11:30 |
| *** txomon|fon has quit IRC | 11:30 | |
| openstackgerrit | Dmitry Ratushnyy proposed openstack/neutron: Add more API tests for port-security extension: https://review.openstack.org/175399 | 11:31 |
| *** yalie has joined #openstack-neutron | 11:31 | |
| *** ferest has joined #openstack-neutron | 11:31 | |
| *** ronis has quit IRC | 11:31 | |
| *** gongysh has joined #openstack-neutron | 11:32 | |
| *** yamamoto has joined #openstack-neutron | 11:32 | |
| *** johnbelamaric has joined #openstack-neutron | 11:36 | |
| *** ferest_ has joined #openstack-neutron | 11:36 | |
| *** yalie has quit IRC | 11:38 | |
| *** ferest has quit IRC | 11:39 | |
| *** ferest_ has quit IRC | 11:39 | |
| *** markvoelker has joined #openstack-neutron | 11:42 | |
| *** johnbelamaric has quit IRC | 11:44 | |
| *** saggi has quit IRC | 11:44 | |
| *** txomon|fon has joined #openstack-neutron | 11:45 | |
| *** markvoelker has quit IRC | 11:46 | |
| *** baoli has joined #openstack-neutron | 11:47 | |
| *** baoli has quit IRC | 11:49 | |
| *** yamamoto_ has joined #openstack-neutron | 11:50 | |
| *** baoli has joined #openstack-neutron | 11:50 | |
| *** merooney has quit IRC | 11:50 | |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Decompose db_base_plugin_v2.py https://review.openstack.org/187961 | 11:51 |
| *** vthapar_ has joined #openstack-neutron | 11:53 | |
| *** yamamoto has quit IRC | 11:53 | |
| *** vthapar has quit IRC | 11:56 | |
| *** irenab has quit IRC | 11:56 | |
| *** irenab has joined #openstack-neutron | 11:57 | |
| *** mohitsharma has joined #openstack-neutron | 11:57 | |
| *** merooney has joined #openstack-neutron | 11:57 | |
| *** sudipto has quit IRC | 11:58 | |
| *** blues-man has quit IRC | 11:59 | |
| *** aix has joined #openstack-neutron | 12:00 | |
| *** markvoelker has joined #openstack-neutron | 12:01 | |
| *** ronis has joined #openstack-neutron | 12:04 | |
| *** lennyb has joined #openstack-neutron | 12:09 | |
| *** yamamoto_ has quit IRC | 12:09 | |
| *** amuller has joined #openstack-neutron | 12:10 | |
| *** jpena is now known as jpena|lunch | 12:10 | |
| *** yamamoto has joined #openstack-neutron | 12:13 | |
| *** mkolesni has quit IRC | 12:14 | |
| *** lennyb_ has quit IRC | 12:14 | |
| *** avozza is now known as zz_avozza | 12:16 | |
| *** zz_avozza is now known as avozza | 12:17 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: Extra indent in test_ovs_neutron_agent https://review.openstack.org/187975 | 12:17 |
| *** amotoki has joined #openstack-neutron | 12:18 | |
| *** jpena|lunch has quit IRC | 12:19 | |
| *** vikram has quit IRC | 12:19 | |
| *** rotbeard has quit IRC | 12:22 | |
| *** yamamoto has quit IRC | 12:23 | |
| *** tkatarki has joined #openstack-neutron | 12:24 | |
| *** avozza has left #openstack-neutron | 12:25 | |
| openstackgerrit | Merged openstack/neutron-vpnaas: Python 3: use six.iteritems instead of dict.items https://review.openstack.org/187356 | 12:31 |
| openstackgerrit | Oleg Bondarev proposed openstack/neutron: Fix callback registry notification for security group rule https://review.openstack.org/187586 | 12:31 |
| *** rkukura_ has joined #openstack-neutron | 12:33 | |
| *** kiranr has joined #openstack-neutron | 12:34 | |
| *** kiran-r has quit IRC | 12:34 | |
| *** yamamoto has joined #openstack-neutron | 12:34 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Ensure non-overlapping cidrs in subnetpools with galera (WIP) https://review.openstack.org/187985 | 12:34 |
| *** baoli has quit IRC | 12:34 | |
| *** rkukura has quit IRC | 12:35 | |
| *** rkukura_ is now known as rkukura | 12:35 | |
| *** merooney has quit IRC | 12:37 | |
| *** yuanying has joined #openstack-neutron | 12:38 | |
| *** kiranr has quit IRC | 12:39 | |
| openstackgerrit | Oleg Bondarev proposed openstack/neutron: Fix confusing parameters names https://review.openstack.org/187987 | 12:40 |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: Adds base in-tree functional testing of the ovs_neutron_agent https://review.openstack.org/140042 | 12:42 |
| *** tomoe_ has joined #openstack-neutron | 12:43 | |
| *** yuanying has quit IRC | 12:46 | |
| *** merooney has joined #openstack-neutron | 12:50 | |
| *** kbringard has joined #openstack-neutron | 12:50 | |
| *** yfried has quit IRC | 12:51 | |
| *** yatin has quit IRC | 12:52 | |
| *** rwsu has joined #openstack-neutron | 12:54 | |
| openstackgerrit | Dmitry Ratushnyy proposed openstack/neutron: Add more API tests for port-security extension: https://review.openstack.org/175399 | 12:56 |
| *** dane_leblanc has joined #openstack-neutron | 12:57 | |
| *** jistr is now known as jistr|mtg | 12:57 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Fullstack testing devref follow up https://review.openstack.org/187990 | 12:58 |
| *** merooney has quit IRC | 12:59 | |
| *** rpothier has joined #openstack-neutron | 12:59 | |
| *** yamamoto has quit IRC | 13:01 | |
| *** dane_leblanc has quit IRC | 13:02 | |
| *** tkatarki has quit IRC | 13:04 | |
| *** yamamoto has joined #openstack-neutron | 13:05 | |
| *** tkatarki has joined #openstack-neutron | 13:07 | |
| *** johnbelamaric has joined #openstack-neutron | 13:08 | |
| *** saggi has joined #openstack-neutron | 13:08 | |
| *** shwetaap has joined #openstack-neutron | 13:11 | |
| openstackgerrit | Elena Ezhova proposed openstack/neutron: Handle SIGTERM in linuxbridge_neutron_agent https://review.openstack.org/187996 | 13:12 |
| *** shwetaap1 has joined #openstack-neutron | 13:12 | |
| *** blues-man has joined #openstack-neutron | 13:13 | |
| *** blues-man has quit IRC | 13:13 | |
| *** blues-man has joined #openstack-neutron | 13:13 | |
| openstackgerrit | Elena Ezhova proposed openstack/neutron: Handle SIGTERM in linuxbridge_neutron_agent https://review.openstack.org/187996 | 13:13 |
| *** shwetaap has quit IRC | 13:16 | |
| *** jpena has joined #openstack-neutron | 13:18 | |
| *** dave-mcc_ has joined #openstack-neutron | 13:19 | |
| *** rossella_s has quit IRC | 13:19 | |
| *** rossella_s has joined #openstack-neutron | 13:20 | |
| *** itzikb has quit IRC | 13:20 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Fullstack testing devref follow up https://review.openstack.org/187990 | 13:22 |
| *** dave-mccowan has quit IRC | 13:23 | |
| *** tiswanso has joined #openstack-neutron | 13:23 | |
| *** zzzeek has joined #openstack-neutron | 13:23 | |
| *** itzikb has joined #openstack-neutron | 13:23 | |
| *** yamamoto has quit IRC | 13:27 | |
| *** yamamoto has joined #openstack-neutron | 13:29 | |
| *** neeti has quit IRC | 13:29 | |
| *** rushil has joined #openstack-neutron | 13:31 | |
| *** jhova has joined #openstack-neutron | 13:31 | |
| dr_gogeta86 | ajo, policy.json got an order ? | 13:33 |
| *** moshele has quit IRC | 13:33 | |
| dr_gogeta86 | this and that | 13:33 |
| dr_gogeta86 | this wins on that ? | 13:34 |
| ajo | dr_gogeta86 what do you mean?, if the entries are processed in order? | 13:34 |
| dr_gogeta86 | yep | 13:34 |
| ajo | I do not know, probably salv-orlando know better than me, | 13:34 |
| ajo | or sc68cal | 13:34 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron-vpnaas: Enable random hash seeds https://review.openstack.org/187107 | 13:34 |
| ajo | know->knows | 13:34 |
| openstackgerrit | Ann Kamyshnikova proposed openstack/neutron: Juno_initial migration https://review.openstack.org/188005 | 13:34 |
| *** baoli has joined #openstack-neutron | 13:35 | |
| *** moshele has joined #openstack-neutron | 13:36 | |
| *** jpena has quit IRC | 13:36 | |
| *** itzikb has quit IRC | 13:37 | |
| *** itzikb has joined #openstack-neutron | 13:38 | |
| dr_gogeta86 | salv-orlando, if can help me | 13:39 |
| openstackgerrit | Ann Kamyshnikova proposed openstack/neutron: Juno_initial migration https://review.openstack.org/188005 | 13:40 |
| kbringard | quick question if has a moment: can anyone tell me what the thought process behind allowing tenants to mark a rule on a firewall as "shared" was? | 13:41 |
| *** sudipto has joined #openstack-neutron | 13:41 | |
| *** abregman has quit IRC | 13:42 | |
| *** vikram has joined #openstack-neutron | 13:42 | |
| *** dan_ has joined #openstack-neutron | 13:43 | |
| *** dan_ is now known as Guest83679 | 13:43 | |
| *** changbl_ has joined #openstack-neutron | 13:43 | |
| *** abregman has joined #openstack-neutron | 13:46 | |
| *** rushil has quit IRC | 13:47 | |
| *** apuimedo has quit IRC | 13:47 | |
| *** phil_h has quit IRC | 13:48 | |
| *** ihrachyshka has joined #openstack-neutron | 13:49 | |
| *** jpena has joined #openstack-neutron | 13:50 | |
| *** banix_ has joined #openstack-neutron | 13:51 | |
| * sc68cal reads scrollback | 13:51 | |
| *** mwagner_afk has joined #openstack-neutron | 13:52 | |
| *** jprovazn has joined #openstack-neutron | 13:52 | |
| *** cbouch has joined #openstack-neutron | 13:53 | |
| salv-orlando | dr_gogeta86: all the matching policies must be satisfied, in no specific order | 13:54 |
| *** yamamoto has quit IRC | 13:54 | |
| *** claudiub has quit IRC | 13:55 | |
| sc68cal | kbringard: for the FwaaS api? | 13:55 |
| kbringard | sc68cal: yea, I think I'm understanding it a bit better now | 13:56 |
| *** yfried has joined #openstack-neutron | 13:57 | |
| dr_gogeta86 | salv-orlando, the goal is to deny all network modification to user who have a specified role | 13:58 |
| *** jistr|mtg is now known as jistr | 13:59 | |
| *** coolsvap is now known as coolsvap|afk | 14:00 | |
| *** ildikov has quit IRC | 14:00 | |
| *** chem has joined #openstack-neutron | 14:00 | |
| dr_gogeta86 | for example "create_network": "rule:admin_or_network_owner or !role:member_no_network", | 14:01 |
| dr_gogeta86 | member_no_network is the role | 14:01 |
| dr_gogeta86 | i did it wrong ? | 14:01 |
| *** merooney has joined #openstack-neutron | 14:01 | |
| *** abregman has quit IRC | 14:01 | |
| *** csoukup has joined #openstack-neutron | 14:02 | |
| *** dboik has joined #openstack-neutron | 14:02 | |
| openstackgerrit | Merged openstack/neutron-fwaas: Python 3: use six.iteritems instead of dict.items https://review.openstack.org/187364 | 14:02 |
| openstackgerrit | Merged openstack/neutron: Extra indent in test_ovs_neutron_agent https://review.openstack.org/187975 | 14:03 |
| *** moshele has quit IRC | 14:03 | |
| *** ildikov has joined #openstack-neutron | 14:03 | |
| *** yamamoto has joined #openstack-neutron | 14:04 | |
| *** moshele has joined #openstack-neutron | 14:04 | |
| *** yamamoto has quit IRC | 14:07 | |
| *** lazy_prince has joined #openstack-neutron | 14:07 | |
| salv-orlando | dr_gogeta86: I think policy.json support negated rules - something like create_network: not role:<banned_role> | 14:07 |
| dr_gogeta86 | i'm using !role:member_no_network | 14:07 |
| dr_gogeta86 | but sometimes using admin i've got some error | 14:08 |
| *** armax has joined #openstack-neutron | 14:08 | |
| *** johnbelamaric has quit IRC | 14:11 | |
| *** sudipto has quit IRC | 14:12 | |
| *** abregman has joined #openstack-neutron | 14:13 | |
| *** triggerz is now known as tobasco | 14:14 | |
| salv-orlando | dr_gogeta86: sometimes? or always? and can you trace the error back to policy evaluation? | 14:16 |
| pavel_bondar | ihrachyshka, hi | 14:18 |
| *** dane_leblanc has joined #openstack-neutron | 14:19 | |
| *** Networkn3rd has joined #openstack-neutron | 14:19 | |
| ihrachyshka | pavel_bondar, hi | 14:19 |
| *** nmagnezi has quit IRC | 14:19 | |
| pavel_bondar | ihrachyshka: what is the right thing to do with copyright here, https://review.openstack.org/#/c/187961/1/neutron/db/db_base_plugin_common.py | 14:19 |
| pavel_bondar | ihrachyshka: It is just method moving, so shoud copyright be updated to Infoblox or left as is 'Openstack Foundation'? | 14:21 |
| pavel_bondar | s/shoud/should/ | 14:21 |
| ihrachyshka | pavel_bondar, unless your employer requires you to attach that meaningless copyright attribution, you may just kill both. Quoting https://wiki.openstack.org/wiki/Documentation/Copyright, "If you update a page, you can add the entity you represent (self or organization) to the list of Copyright holders, but do not remove any listed Copyright headings." | 14:22 |
| ihrachyshka | pavel_bondar, I don't personally think moving code justifies authorship attribution at all | 14:22 |
| ihrachyshka | oh, sorry, you should not kill both I guess, leave foundation at least | 14:22 |
| dr_gogeta86 | salv-orlando, how can I trace evaluation ? | 14:22 |
| ihrachyshka | you can append another one, but I wouldn't do it unless forced | 14:23 |
| openstackgerrit | Russell Bryant proposed openstack/neutron: Add sub-project lieutenants https://review.openstack.org/187733 | 14:23 |
| *** numan has quit IRC | 14:24 | |
| pavel_bondar | ihrachyshka, thanks, got it, will leave foundation copyright | 14:24 |
| *** zz_jgrimm is now known as jgrimm | 14:25 | |
| *** david-lyle has quit IRC | 14:25 | |
| Sam-I-Am | mornings | 14:28 |
| frickler | can someone confirm that https://bugs.launchpad.net/tempest/+bug/1182883 has been fixed in nova some time during the last two years? | 14:29 |
| openstack | Launchpad bug 1182883 in neutron "List servers matching a regex fails with Neutron" [Medium,Confirmed] | 14:29 |
| Sam-I-Am | amuller: hello fellow appendix-less being | 14:29 |
| amuller | Sam-I-Am: Hello fellow evolved human | 14:29 |
| *** thurloat_isgone is now known as thurloat | 14:29 | |
| frickler | at least something like nova list --ip 10.53.* seems to work fine | 14:29 |
| Sam-I-Am | amuller: although i've been reading some stuff that says the appendix has some sort of purpose | 14:29 |
| sc68cal | HenryG: I just marked https://bugs.launchpad.net/neutron/+bug/1358297 as invalid, since the last comment from the reporter says he's going to open a different bug | 14:29 |
| openstack | Launchpad bug 1358297 in neutron "Port doesn't receive IP SLAAC in subnets with Router advertisements without dnsmasq" [Medium,Invalid] - Assigned to Henry Gessau (gessau) | 14:29 |
| amuller | Sam-I-Am: Nonsense, we are clearly superior human beings | 14:30 |
| Sam-I-Am | however, it does not involve digesting bacon, donuts, or beer and therefore is not important | 14:30 |
| frickler | so I would like to go ahead and re-enable the tempest test that has been skipped because of that bug | 14:30 |
| amuller | mestery: ihrachyshka: Can we merge this full stack fix? https://review.openstack.org/#/c/180882/ | 14:31 |
| amuller | mestery: ihrachyshka: Just verified it manually on my system, works great and fixes a serious issue | 14:31 |
| mestery | amuller: looking | 14:31 |
| *** rwsu has quit IRC | 14:31 | |
| ihrachyshka | amuller, in a meeting, will check in 30 mins | 14:31 |
| amuller | mestery: ihrachyshka: Thanks | 14:31 |
| *** tkatarki has quit IRC | 14:31 | |
| *** rushil has joined #openstack-neutron | 14:34 | |
| *** numan_ has joined #openstack-neutron | 14:35 | |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Decompose db_base_plugin_v2.py https://review.openstack.org/187961 | 14:35 |
| *** barra204 has quit IRC | 14:36 | |
| *** shakamunyi has quit IRC | 14:36 | |
| *** raildo has joined #openstack-neutron | 14:39 | |
| *** TrevorV_ has joined #openstack-neutron | 14:40 | |
| amotoki | mestery: I just noticed neutron-*aas stable branches have different ACL from neutron stable branches. It seems neutron-*aas core can +2 to stable branches. | 14:42 |
| amotoki | mestery: I am not sure it is intended or not. | 14:42 |
| *** afazekas has quit IRC | 14:43 | |
| mestery | amotoki: https://review.openstack.org/#/c/186820/ | 14:43 |
| mestery | amotoki: That needs to merge then all will be good with the world ? | 14:43 |
| mestery | I'll ping infra again now | 14:43 |
| amotoki | mestery: ah... you beat me :-) | 14:44 |
| *** carl_baldwin has joined #openstack-neutron | 14:44 | |
| mestery | amotoki: :) | 14:45 |
| *** tomoe_ has quit IRC | 14:45 | |
| *** yamamoto has joined #openstack-neutron | 14:45 | |
| *** yamamoto has quit IRC | 14:46 | |
| *** tkatarki has joined #openstack-neutron | 14:46 | |
| *** tkatarki has quit IRC | 14:48 | |
| *** killer_prince has joined #openstack-neutron | 14:48 | |
| *** lazy_prince has quit IRC | 14:49 | |
| *** ronis has quit IRC | 14:49 | |
| *** pc_m has quit IRC | 14:51 | |
| *** tkatarki has joined #openstack-neutron | 14:52 | |
| *** lazy_prince has joined #openstack-neutron | 14:53 | |
| *** killer_prince has quit IRC | 14:53 | |
| *** catohornet has joined #openstack-neutron | 14:55 | |
| *** johnbelamaric has joined #openstack-neutron | 14:56 | |
| *** daneyon has joined #openstack-neutron | 14:57 | |
| *** agireud has joined #openstack-neutron | 14:58 | |
| *** dsneddon has joined #openstack-neutron | 14:59 | |
| *** dguitarbite has quit IRC | 14:59 | |
| openstackgerrit | Carl Baldwin proposed openstack/neutron: Use a single method to remove an address with its conntrack state https://review.openstack.org/186150 | 15:00 |
| *** anshul has quit IRC | 15:01 | |
| *** sadasu_ has joined #openstack-neutron | 15:01 | |
| *** daneyon_ has joined #openstack-neutron | 15:01 | |
| sadasu_ | ajo: you here? | 15:01 |
| ajo | hi sadasu_ , yes :) | 15:01 |
| *** pc_m has joined #openstack-neutron | 15:01 | |
| sadasu_ | I signed up for a task in the qos work items | 15:02 |
| *** ekuris has quit IRC | 15:02 | |
| sadasu_ | will need some time to understand the details...might ping you for that during the week | 15:02 |
| *** tomoe_ has joined #openstack-neutron | 15:03 | |
| *** daneyon has quit IRC | 15:04 | |
| *** kiran-r has joined #openstack-neutron | 15:05 | |
| ajo | sadasu_, sure, feel free to ping me | 15:05 |
| ajo | I'm generally available around CEST time #openstack-neutron-qos , but otherwise feel free to reach me on mangelajo@rehdat.com, | 15:06 |
| *** ollivier has joined #openstack-neutron | 15:06 | |
| *** vivek-ebay has joined #openstack-neutron | 15:09 | |
| *** changbl_ has quit IRC | 15:10 | |
| *** zhipeng has joined #openstack-neutron | 15:10 | |
| *** jgrimm is now known as zz_jgrimm | 15:10 | |
| sadasu_ | ajo: thanks! | 15:11 |
| ajo | thank you too sadasu_ :) | 15:11 |
| *** changbl_ has joined #openstack-neutron | 15:11 | |
| carl_baldwin | gsagie: ping | 15:12 |
| *** kiran-r has quit IRC | 15:13 | |
| moshele | ajo, irenab: Qos the agent extension mgr should be the same as in sec 1.1 in https://drive.google.com/file/d/0B2XATqL7DxHFRHNjU3k1UFNYRjQ/view right? | 15:14 |
| ajo | moshele, similar, AFAIK there will be some changes | 15:14 |
| ajo | it needs to be updated | 15:15 |
| ajo | I need to run ;) | 15:15 |
| moshele | ajo: ok I will start POC the code | 15:15 |
| moshele | vikrim: hi | 15:16 |
| moshele | vikram: hi | 15:16 |
| *** marun has joined #openstack-neutron | 15:16 | |
| vikram | hi | 15:17 |
| openstackgerrit | Merged openstack/neutron: Python3: use six.iteritems() instead of dict.iteritems() https://review.openstack.org/184066 | 15:17 |
| *** killer_prince has joined #openstack-neutron | 15:17 | |
| *** mestery_ has joined #openstack-neutron | 15:17 | |
| *** Aish has joined #openstack-neutron | 15:18 | |
| moshele | vikram: I will start POC the Qos the agent extension mgr see section 1.1 in https://drive.google.com/file/d/0B2XATqL7DxHFRHNjU3k1UFNYRjQ/view you can join me if you want | 15:18 |
| vikram | That will be great | 15:18 |
| *** tflynn has joined #openstack-neutron | 15:18 | |
| openstackgerrit | Merged openstack/neutron: Remove extra indent in testcases https://review.openstack.org/187827 | 15:19 |
| openstackgerrit | Merged openstack/neutron: Remove extra indent in testcases https://review.openstack.org/187827 | 15:19 |
| *** rwsu has joined #openstack-neutron | 15:19 | |
| *** mestery has quit IRC | 15:20 | |
| *** mlavalle has joined #openstack-neutron | 15:20 | |
| *** lazy_prince has quit IRC | 15:20 | |
| *** mchestr has joined #openstack-neutron | 15:21 | |
| *** claudiub has joined #openstack-neutron | 15:22 | |
| *** zhipeng has quit IRC | 15:22 | |
| *** alex_klimov has quit IRC | 15:23 | |
| *** merooney has quit IRC | 15:24 | |
| *** alexpilotti has quit IRC | 15:24 | |
| *** ollivier has quit IRC | 15:25 | |
| *** tflynn has quit IRC | 15:25 | |
| *** numan_ has quit IRC | 15:25 | |
| *** gridinv_ has joined #openstack-neutron | 15:26 | |
| *** vivek-ebay has quit IRC | 15:27 | |
| *** tomoe_ has quit IRC | 15:27 | |
| *** RawInfoSec has joined #openstack-neutron | 15:28 | |
| *** dsanders has joined #openstack-neutron | 15:29 | |
| *** anilvenkata has joined #openstack-neutron | 15:29 | |
| *** andrelucas has joined #openstack-neutron | 15:30 | |
| *** mestery_ is now known as mestery | 15:30 | |
| *** RawInfoSec has left #openstack-neutron | 15:33 | |
| *** rushil has quit IRC | 15:34 | |
| *** killer_prince has quit IRC | 15:34 | |
| *** anilvenkata has left #openstack-neutron | 15:34 | |
| *** killer_prince has joined #openstack-neutron | 15:34 | |
| *** a_le has joined #openstack-neutron | 15:34 | |
| *** killer_prince has quit IRC | 15:35 | |
| *** kiran-r has joined #openstack-neutron | 15:35 | |
| *** mohitsharma has quit IRC | 15:35 | |
| *** Networkn3rd has quit IRC | 15:37 | |
| *** numan_ has joined #openstack-neutron | 15:37 | |
| *** david-lyle has joined #openstack-neutron | 15:39 | |
| *** Sukhdev has joined #openstack-neutron | 15:39 | |
| *** zz_jgrimm is now known as jgrimm | 15:40 | |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Refactor and decompose db_base_plugin_v2.py https://review.openstack.org/153236 | 15:41 |
| *** claudiub|2 has joined #openstack-neutron | 15:42 | |
| *** claudiub has quit IRC | 15:43 | |
| *** ushkalim has quit IRC | 15:43 | |
| *** andrelucas has quit IRC | 15:44 | |
| *** yamamoto has joined #openstack-neutron | 15:46 | |
| *** luqas has quit IRC | 15:46 | |
| *** david-lyle has quit IRC | 15:47 | |
| *** alexpilotti has joined #openstack-neutron | 15:48 | |
| *** nmagnezi has joined #openstack-neutron | 15:50 | |
| openstackgerrit | Merged openstack/neutron: Wrap ML2 delete_port with db retry decorator https://review.openstack.org/180466 | 15:50 |
| *** sbalukoff has joined #openstack-neutron | 15:50 | |
| *** jbell8 has joined #openstack-neutron | 15:51 | |
| ihrachyshka | amuller, I won't be able to review full stack patch today. tomorrow for sure. sorry. | 15:51 |
| *** yamamoto has quit IRC | 15:52 | |
| *** jlibosva has quit IRC | 15:52 | |
| *** mohitsharma has joined #openstack-neutron | 15:52 | |
| *** david-lyle has joined #openstack-neutron | 15:53 | |
| *** devvesa has quit IRC | 15:53 | |
| amuller | ihrachyshka: ack | 15:54 |
| *** busterswt has joined #openstack-neutron | 15:54 | |
| *** nmagnezi has quit IRC | 15:55 | |
| *** jbell8 has quit IRC | 15:55 | |
| *** shivharis has joined #openstack-neutron | 15:56 | |
| *** moha_hunt has joined #openstack-neutron | 15:56 | |
| *** dannywilson has joined #openstack-neutron | 15:57 | |
| *** nlahouti has joined #openstack-neutron | 15:57 | |
| HenryG | sc68cal: belated thanks | 15:59 |
| *** yeungp has joined #openstack-neutron | 15:59 | |
| *** numan_ has quit IRC | 15:59 | |
| *** Aish has quit IRC | 16:00 | |
| *** shwetaap1 has quit IRC | 16:00 | |
| *** shwetaap has joined #openstack-neutron | 16:00 | |
| *** matrohon has quit IRC | 16:01 | |
| *** tiswanso has quit IRC | 16:01 | |
| *** tiswanso has joined #openstack-neutron | 16:02 | |
| *** vthapar_ has quit IRC | 16:02 | |
| *** davidchep has joined #openstack-neutron | 16:02 | |
| *** SridharG has left #openstack-neutron | 16:04 | |
| *** kawa2014 has quit IRC | 16:05 | |
| *** jgrimm is now known as zz_jgrimm | 16:07 | |
| *** jlanoux has quit IRC | 16:07 | |
| *** _cjones_ has joined #openstack-neutron | 16:07 | |
| *** zz_jgrimm is now known as jgrimm | 16:09 | |
| openstackgerrit | Merged openstack/neutron: Fix confusing parameters names https://review.openstack.org/187987 | 16:10 |
| *** a_le has quit IRC | 16:10 | |
| sc68cal | mestery: just posted a bit of a rant on this bug - https://bugs.launchpad.net/neutron/+bug/1460177 | 16:10 |
| openstack | Launchpad bug 1460177 in neutron "Support metadata service with IPv6-only tenant network" [Undecided,Triaged] | 16:10 |
| *** moshele has quit IRC | 16:10 | |
| mestery | sc68cal: :) | 16:11 |
| *** jistr has quit IRC | 16:11 | |
| *** afazekas has joined #openstack-neutron | 16:12 | |
| sc68cal | mestery: I'll need to get two tatoos on my forehead - "Does it work with IPv6" - and "DON'T TRY AND FIX THE EC2 APIs" | 16:12 |
| *** moha_hunt has quit IRC | 16:12 | |
| *** gridinv_ has quit IRC | 16:13 | |
| aveiga_ | sc68cal: I prefer a sign, but if tats are your thing... | 16:13 |
| sc68cal | or possibly "Don't try and 'improve' the EC2 API' | 16:13 |
| *** SridharGaddam has quit IRC | 16:13 | |
| mestery | lol | 16:14 |
| *** kiranr has joined #openstack-neutron | 16:14 | |
| *** kiran-r has quit IRC | 16:14 | |
| *** anshul has joined #openstack-neutron | 16:14 | |
| *** vikram has left #openstack-neutron | 16:15 | |
| *** dims_ has quit IRC | 16:18 | |
| *** dims_ has joined #openstack-neutron | 16:18 | |
| *** catohornet has left #openstack-neutron | 16:19 | |
| *** dims__ has joined #openstack-neutron | 16:20 | |
| *** jamielennox is now known as jamielennox|away | 16:21 | |
| *** barra204 has joined #openstack-neutron | 16:21 | |
| *** fzdarsky has quit IRC | 16:22 | |
| *** david-lyle has quit IRC | 16:22 | |
| *** anshul has quit IRC | 16:23 | |
| *** sriatom has joined #openstack-neutron | 16:23 | |
| *** vivek-ebay has joined #openstack-neutron | 16:24 | |
| *** itzikb has quit IRC | 16:24 | |
| *** dims_ has quit IRC | 16:24 | |
| *** barra204 has quit IRC | 16:28 | |
| *** ihrachyshka has quit IRC | 16:28 | |
| *** ygbo has quit IRC | 16:29 | |
| kevinbenton | mestery, amotoki: i don't like the current drivers team meeting time. the first half overlaps with a meeting I have at work | 16:30 |
| *** marun_ has joined #openstack-neutron | 16:30 | |
| mestery | kevinbenton: We'll sort this all out. | 16:30 |
| *** abregman has quit IRC | 16:31 | |
| amotoki | kevinbenton: mestery: fair enough. Aligning other meeting time sounds good :-) | 16:31 |
| salv-orlando | sc68cal: are you sure the guy was suggesting to "fix" the EC2 metadata service? It seems he proposes a cloud-init change (wrong, out of scope), but beyond that it seems that he's not advocating for any change in the segment between the metadata proxy and the nova metadata endpoint | 16:31 |
| amotoki | the latter half overlaps Ml2 meeting ;-( | 16:31 |
| *** ronis has joined #openstack-neutron | 16:32 | |
| *** nmagnezi has joined #openstack-neutron | 16:32 | |
| *** scheuran has quit IRC | 16:33 | |
| *** marun has quit IRC | 16:33 | |
| *** yatin has joined #openstack-neutron | 16:34 | |
| *** marun has joined #openstack-neutron | 16:35 | |
| *** marun_ has quit IRC | 16:35 | |
| *** kiranr has quit IRC | 16:36 | |
| *** _kiran_ has joined #openstack-neutron | 16:36 | |
| *** jhova has quit IRC | 16:37 | |
| *** madhu_ak has joined #openstack-neutron | 16:37 | |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: L2 agent RPC add new RPC calls https://review.openstack.org/147435 | 16:37 |
| *** a_le has joined #openstack-neutron | 16:38 | |
| dougwig | sc68cal: if you get that first tattoo, the rest of us will have to get one that just says "No". | 16:40 |
| *** Swami has joined #openstack-neutron | 16:40 | |
| dougwig | :) | 16:40 |
| *** safchain has quit IRC | 16:41 | |
| *** _kiran_ has quit IRC | 16:41 | |
| *** jbell8 has joined #openstack-neutron | 16:42 | |
| *** hanzhang has quit IRC | 16:47 | |
| openstackgerrit | John Schwarz proposed openstack/neutron: Fullstack testing devref follow up https://review.openstack.org/187990 | 16:47 |
| *** rushil has joined #openstack-neutron | 16:48 | |
| *** hanzhang has joined #openstack-neutron | 16:48 | |
| kevinbenton | dougwig: what tattoo? | 16:51 |
| *** oreillyd has quit IRC | 16:51 | |
| *** amuller is now known as amuller_afk | 16:51 | |
| *** mohitsharma has quit IRC | 16:52 | |
| sc68cal | salv-orlando: possibly, but there are also proposals floating around for adding new addresses the metadata proxy wil listen on | 16:53 |
| sc68cal | like ::169.254.169.254 | 16:53 |
| *** sriatom is now known as _sri | 16:53 | |
| kevinbenton | sc68cal: am i misunderstanding, or is the real thing that defines the address cloud-init rather than EC2's API? | 16:54 |
| sc68cal | kevinbenton: cloud-init uses 169.254.169.254 because that's the IPv4 link local that Amazon's EC2 metadata service defines | 16:55 |
| *** jpena has quit IRC | 16:55 | |
| *** afazekas has quit IRC | 16:55 | |
| sc68cal | kevinbenton: I think that there is some stuff in cloud-init to query other addresses | 16:55 |
| *** Longgeek has quit IRC | 16:55 | |
| sc68cal | kevinbenton: if i'm not mistaken | 16:55 |
| *** claudiub|2 has quit IRC | 16:55 | |
| *** merooney has joined #openstack-neutron | 16:56 | |
| kevinbenton | sc68cal: so was your concern that we just might listen on a different address than what EC2 decides to listen on when they do ipv6? | 16:56 |
| sc68cal | kevinbenton: yes | 16:56 |
| *** kutija has joined #openstack-neutron | 16:57 | |
| openstackgerrit | Armando Migliaccio proposed openstack/neutron: Revert "Revert "Add VIF_DELETED notification event to Nova"" https://review.openstack.org/187871 | 16:57 |
| kevinbenton | sc68cal: but why would it matter as long as cloud-init is asking? | 16:57 |
| sc68cal | kevinbenton: not sure what you mean, expand? | 16:57 |
| kevinbenton | sc68cal: it doesn't matter if we are listening on a different address than EC2 as long as cloud-init tries the address we are listening on | 16:58 |
| openstackgerrit | Carl Baldwin proposed openstack/neutron: Enhance utils.ensure_dir to be resilient to concurrent workers https://review.openstack.org/188083 | 16:59 |
| sc68cal | kevinbenton: I disagree, since the addresses that an HTTP based metadata API is part of the API contract that the AWS API has | 16:59 |
| sc68cal | *listens on is part of the | 16:59 |
| kevinbenton | sc68cal: but isn't it cloud-init that is defining the contract, not EC2? | 16:59 |
| sc68cal | kevinbenton: no, cloud-init just accesses the metadata service | 17:00 |
| sc68cal | kevinbenton: so cloud-init is a consumer of the API, and the contract | 17:00 |
| sc68cal | as I understand it, but I could be wrong | 17:00 |
| kevinbenton | sc68cal: cloud-init chooses the address is contacts with | 17:00 |
| kevinbenton | sc68cal: EC2 is not in control of that, right? | 17:00 |
| *** jhova has joined #openstack-neutron | 17:01 | |
| sc68cal | kevinbenton: but the addresses that cloud-init uses, is an address defined by the EC2 metadata API | 17:01 |
| *** dannywilson has quit IRC | 17:01 | |
| arosen | kevinbenton what address do you want to use instead? | 17:01 |
| sc68cal | http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instance-metadata.html | 17:01 |
| kevinbenton | sc68cal: i thought you implied that it already contacts addresses other than the 169 one? | 17:01 |
| Swami | armax: hi | 17:02 |
| sc68cal | kevinbenton: I think cloud-init has some logic in it for "metadata_urls" that can be used to get metadata out | 17:02 |
| *** sthillma has joined #openstack-neutron | 17:02 | |
| sc68cal | kevinbenton: but frankly I don't know if that was a great idea to start with | 17:03 |
| *** merooney has quit IRC | 17:03 | |
| *** dannywilson has joined #openstack-neutron | 17:03 | |
| kevinbenton | sc68cal: what i'm getting at is that cloud-init determines the addresses we can select from, not EC2. so if cloud-init has support for an IPv6 address, we can use whatever that is | 17:03 |
| kevinbenton | arosen: an ipv6 addr | 17:03 |
| *** ivar-lazzaro has joined #openstack-neutron | 17:03 | |
| arosen | gotcha | 17:03 |
| sc68cal | kevinbenton: I still don't agree that cloud-init makes that choice - the EC2 API itself specifies that there's a well known IP address to get the metadata from. It just so happens that someone extended cloud-init to support metadata_urls for finding alternate endpoints for the metadata | 17:04 |
| sc68cal | since that's not defined in the metadata API | 17:05 |
| *** merooney has joined #openstack-neutron | 17:05 | |
| *** yatin has quit IRC | 17:05 | |
| sc68cal | so we've already had a deviation, thankfully just in cloud-init | 17:05 |
| kevinbenton | sc68cal: does EC2 not use cloud-init? | 17:06 |
| *** tkatarki has quit IRC | 17:06 | |
| sc68cal | kevinbenton: it's the other way around | 17:06 |
| *** nmagnezi has quit IRC | 17:06 | |
| sc68cal | kevinbenton: cloud-init uses the EC2 API | 17:06 |
| *** dannywilson has quit IRC | 17:06 | |
| *** tkatarki has joined #openstack-neutron | 17:06 | |
| sc68cal | kevinbenton: meaning cloud-init will attempt to query 169.254.169.264 | 17:06 |
| sc68cal | for any HTTP server, then tries to get a couple key/values | 17:07 |
| kevinbenton | sc68cal: i know how it works | 17:07 |
| *** ildikov has quit IRC | 17:07 | |
| kevinbenton | sc68cal: but if EC2 expects people to use cloud-init | 17:07 |
| kevinbenton | sc68cal: then cloud-init is in charge of address discovery | 17:07 |
| sc68cal | kevinbenton: EC2 doesn't expect people to use cloud init | 17:07 |
| sc68cal | kevinbenton: the docs have examples where they use cURL | 17:07 |
| kevinbenton | sc68cal: what do the majority of the AMIs use? | 17:08 |
| kevinbenton | sc68cal: when i boot a standard ubuntu image on EC2, what populates my ssh key? | 17:08 |
| *** s3wong has joined #openstack-neutron | 17:09 | |
| *** akamyshnikova_ has joined #openstack-neutron | 17:09 | |
| sc68cal | kevinbenton: but cloud init is just a client, it doesn't determine the API | 17:09 |
| *** amotoki has quit IRC | 17:09 | |
| *** afazekas has joined #openstack-neutron | 17:09 | |
| sc68cal | that's like saying just because we can add things to python-neutronclient for additional APIs, we can extend the security group API | 17:09 |
| *** shivharis has quit IRC | 17:09 | |
| kevinbenton | sc68cal: that's relative. if amazon isn't in charge of cloud-init, they certainly can't change the API | 17:10 |
| kevinbenton | sc68cal: in a way that breaks cloud-init | 17:10 |
| sc68cal | kevinbenton: that's subjective. I'd argue that if Amazon makes an API change, cloud-init will change to reflect | 17:10 |
| sc68cal | not the reverse | 17:10 |
| kevinbenton | sc68cal: it can't without invalidating every AMI out there with cloud-init | 17:10 |
| kevinbenton | sc68cal: whatever amazon listens on, it's up to cloud-init to choose the correct address to connect to | 17:11 |
| *** mchestr has quit IRC | 17:12 | |
| aveiga_ | kevinbenton: I think the issue is that changing metadata service to add v6 in a way that doesn't exist on Amazon runs the risk of Amazon implementing v6 access differently, and having then forked the API, cloud-init would have to choose which to support | 17:12 |
| aveiga_ | which more likely than not would force a change on our metadata service to conform to the EC2 way | 17:12 |
| *** mchestr has joined #openstack-neutron | 17:12 | |
| kevinbenton | aveiga_: that's not a different API, it wouldn't require a fork | 17:12 |
| *** merooney has quit IRC | 17:12 | |
| kevinbenton | if it's just a different address, then cloud-init would need to try that as well | 17:13 |
| *** david-lyle has joined #openstack-neutron | 17:13 | |
| aveiga_ | assuming well-known-addr is the way they'd implement it | 17:13 |
| kevinbenton | but the larger point i'm making is actually that amazon is at the behest of cloud-init | 17:13 |
| *** merooney has joined #openstack-neutron | 17:13 | |
| kevinbenton | unless they are in charge of cloud-init, they have to follow whatever it's address discovery mechanism is | 17:13 |
| sc68cal | kevinbenton: I guess the difference we have is that I believe the address that it listens on is part of the actual API contract. | 17:13 |
| *** david-lyle has quit IRC | 17:13 | |
| *** blues-man has quit IRC | 17:14 | |
| *** josecastroleon has quit IRC | 17:14 | |
| kevinbenton | sc68cal: I would call it API discovery | 17:14 |
| *** marun has quit IRC | 17:14 | |
| kevinbenton | sc68cal: but even if you say it's part of the contract, it doesn't change the main point | 17:14 |
| kevinbenton | sc68cal: unless amazon is in control of cloud-init, they aren't in charge of the address | 17:14 |
| *** david-lyle has joined #openstack-neutron | 17:15 | |
| aveiga_ | I think there's a cause-effect swap here | 17:15 |
| *** dguerri is now known as dguerri`away | 17:15 | |
| aveiga_ | didn't cloud-init get written to use that address that EC2 selected? | 17:15 |
| *** marun has joined #openstack-neutron | 17:15 | |
| *** anshul has joined #openstack-neutron | 17:15 | |
| *** amuller_afk is now known as amuller | 17:15 | |
| kevinbenton | harlowja: does cloud-init exist solely to use the EC2 API? | 17:15 |
| aveiga_ | I don't think Amazon picked that addr because cloud-init had been using it | 17:16 |
| *** anshul has quit IRC | 17:16 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 17:16 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-fwaas: Updated from global requirements https://review.openstack.org/188096 | 17:16 |
| sc68cal | aveiga_: ++ | 17:16 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/188098 | 17:17 |
| kevinbenton | but it doesn't matter at this point | 17:17 |
| kevinbenton | cloud-init is now in charge | 17:17 |
| kevinbenton | because amazon doesn't want to break all of the AMIs | 17:17 |
| aveiga_ | that's a bold assumption | 17:17 |
| *** shwetaap has quit IRC | 17:18 | |
| kevinbenton | aveiga_: how so? breaking cloud-init would be idiotic for EC2 | 17:18 |
| kevinbenton | aveiga_: every single AMI developed by the community using cloud-init would be trash | 17:18 |
| aveiga_ | breaking it only on features that EC2 had never supported by introducing those features in a different way doesn't seem like something they'd have an issue with | 17:19 |
| aveiga_ | it's still an assumption, either way | 17:19 |
| aveiga_ | and making assumptions against a closed entities future decisions is something I'm not very comfortable with | 17:19 |
| aveiga_ | it's akin to gambling | 17:19 |
| kevinbenton | aveiga_: ok, let me put this differently. do you think amazon will not support cloud-init for any IPv6 metadata in the future? | 17:20 |
| kevinbenton | or will they rely on support from cloud-init? | 17:20 |
| aveiga_ | it already doesn't, so why should it mold EC2 to the way OpenStack does it? No, I don't automatically assume they will | 17:20 |
| kevinbenton | aveiga_: ok, so none of the thousands of AMIs out there will support ipv6-only | 17:20 |
| aveiga_ | they don't already | 17:21 |
| aveiga_ | or did I miss something? | 17:21 |
| *** fzdarsky has joined #openstack-neutron | 17:21 | |
| *** yamahata has quit IRC | 17:21 | |
| kevinbenton | aveiga_: if EC2 adds ipv6 connectivity and cloud-init has a discovery mechanism for IPv6 addresses, they would | 17:21 |
| aveiga_ | sure, assuming the cloud-init discovery mechanism supports finding whatever is selected for EC2 | 17:22 |
| *** jbell8 has quit IRC | 17:22 | |
| sc68cal | meaning hopefully cloud-init saw into the future as to how AWS was going to deploy v6 | 17:22 |
| aveiga_ | so either we gamble and get lucky and it matches, or there's cloud-init rework to support two different methods | 17:23 |
| sc68cal | more likely Amazon just announces v6 availability then cloud-init adds support | 17:23 |
| kevinbenton | aveiga_: right. so from EC2's perspective, they either add a listener on a v6 address cloud-init can find, or they tell everyone to jump in a lake | 17:23 |
| sc68cal | kevinbenton: they've been telling people to jump in the lake for years | 17:24 |
| *** yeungp has quit IRC | 17:24 | |
| *** mchestr has quit IRC | 17:24 | |
| *** mchestr has joined #openstack-neutron | 17:24 | |
| *** yeungp has joined #openstack-neutron | 17:24 | |
| aveiga_ | it's still a gamble on assumptions around a closed entity making a decision | 17:24 |
| kevinbenton | aveiga_: it's not really a big gamble. for neutron it's just a single iptables rule | 17:25 |
| kevinbenton | but my point still stands, unless cloud-init never gains support for EC2 IPv6, cloud-init is who we need to cater to | 17:26 |
| kevinbenton | not EC2 | 17:26 |
| *** dboik has quit IRC | 17:26 | |
| aveiga_ | but I thought the discussion was around metadata service existing to provide EC2 compatibility | 17:27 |
| *** dboik has joined #openstack-neutron | 17:27 | |
| aveiga_ | not about whether cloud-init has a capability | 17:27 |
| *** reed_ has joined #openstack-neutron | 17:27 | |
| kevinbenton | the discussion is whether or not it's safe to have our metadata service lisent on an ipv6 address | 17:27 |
| marun | 18/query kevinbenton | 17:28 |
| *** david-lyle has quit IRC | 17:28 | |
| marun | oops | 17:28 |
| kevinbenton | and i'm arguing that if cloud-init has support for a discovery mechanism, we use that | 17:28 |
| kevinbenton | because unless EC2 is willing to throw out cloud-init as a supported client, we should be relatively safe | 17:29 |
| Swami | kevinbenton: hi | 17:30 |
| *** vikram has joined #openstack-neutron | 17:31 | |
| kevinbenton | Swami: in a meeting for the next half-hour so will be unresponsive for a bit | 17:31 |
| Swami | ok, I suspect that one of your patches might have broken the dvr job. | 17:31 |
| Swami | https://review.openstack.org/#/c/187819/ | 17:31 |
| openstackgerrit | John Davidge proposed openstack/neutron: Introduce pluggable framework for IPv6 Prefix Delegation https://review.openstack.org/158697 | 17:31 |
| *** shwetaap has joined #openstack-neutron | 17:31 | |
| *** bjornar_ has joined #openstack-neutron | 17:32 | |
| *** s3wong has quit IRC | 17:33 | |
| vikram | #openstack-meeting-3 | 17:34 |
| salv-orlando | sc68cal, kevinbenton what are we bikeshedding about today? | 17:34 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 17:34 |
| *** jbell8 has joined #openstack-neutron | 17:34 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-fwaas: Updated from global requirements https://review.openstack.org/188096 | 17:34 |
| aveiga_ | salv-orlando: what color to paint the yak | 17:34 |
| *** Sukhdev has quit IRC | 17:34 | |
| openstackgerrit | Merged openstack/neutron: Move full-stack logs post-tests https://review.openstack.org/180882 | 17:34 |
| *** aveiga_ is now known as aveiga | 17:35 | |
| *** sridhar_ram has joined #openstack-neutron | 17:35 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/188098 | 17:35 |
| salv-orlando | aveiga: rainbow, of course | 17:35 |
| aveiga | salv-orlando: wrong answer. Right answer: doesn't matter, since you'll shave the yak tomorrow, anyway | 17:36 |
| salv-orlando | aveiga: you're so nihilist... | 17:36 |
| aveiga | hehe, you got me | 17:36 |
| sc68cal | the yak shed needs to be more shaggy | 17:37 |
| aveiga | it's because I had to leave the beautiful mountains of Vancouver behind | 17:37 |
| salv-orlando | sc68cal: so what should we pledge allegiance to? cloud-init or the EC2 metadata spec? | 17:38 |
| *** fzdarsky has quit IRC | 17:38 | |
| *** vivek-ebay has quit IRC | 17:38 | |
| sc68cal | salv-orlando: I preferred your comment at the summit that we all pledge ourselves to money | 17:38 |
| *** hitalia has joined #openstack-neutron | 17:39 | |
| sc68cal | salv-orlando: in all seriousness, we bypass this specific bikeshed by moving to use config drive | 17:39 |
| * sc68cal points to the bike shed that is his favorite color | 17:40 | |
| *** yamahata has joined #openstack-neutron | 17:40 | |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: OVS agent react to events instead of polling https://review.openstack.org/164880 | 17:41 |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: Add get_events to OVSDB monitor https://review.openstack.org/161806 | 17:41 |
| sc68cal | at least with config drive, it's an openstack "api" (although it's just a virtual CDROM drive really) that we control the destiny of | 17:41 |
| *** kiran-r has joined #openstack-neutron | 17:41 | |
| sc68cal | and then we wait to see what AWS does on the ipv6 side for the metadata API service - when they implement something, we'll implement it | 17:42 |
| salv-orlando | sc68cal: I think that's the kind of thing an architect of indolence would do. | 17:42 |
| *** Networkn3rd has joined #openstack-neutron | 17:44 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 17:44 |
| salv-orlando | sc68cal: your conclusion seems very reasonable to me. While cloud-init can be tuned, it's mainly a EC2 client. Making it work with ipv6 implies us making assumptions on how the metadata service will behave in IPv6 | 17:45 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Revert "Check for 'removed' in port_info before reference" https://review.openstack.org/188121 | 17:45 |
| sc68cal | salv-orlando: yes, I wish i had a better solution rather than appearing to just concern-troll | 17:48 |
| salv-orlando | sc68cal: everyone of us is somebody | 17:49 |
| salv-orlando | else's troll | 17:49 |
| *** hitalia has quit IRC | 17:50 | |
| jaypipes | salv-orlando: getting quite metaphysical this evening :) | 17:50 |
| *** nyechiel_ has quit IRC | 17:50 | |
| marun | away | 17:52 |
| *** reed_ has quit IRC | 17:52 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Remove reference to non-existent fullstack fixture https://review.openstack.org/188124 | 17:52 |
| *** agireud has quit IRC | 17:53 | |
| *** sripriya has joined #openstack-neutron | 17:53 | |
| *** vikram has quit IRC | 17:54 | |
| *** mchestr has quit IRC | 17:55 | |
| *** mchestr has joined #openstack-neutron | 17:55 | |
| *** catohornet has joined #openstack-neutron | 17:56 | |
| *** busterswt has quit IRC | 17:57 | |
| sc68cal | salv-orlando: I need that on a plaque | 17:58 |
| aveiga | sc68cal: why not a tattoo? | 17:58 |
| *** dguerri`away is now known as dguerri | 17:59 | |
| sc68cal | aveiga: lol | 17:59 |
| *** hitalia has joined #openstack-neutron | 17:59 | |
| *** nlahouti has quit IRC | 17:59 | |
| *** nyechiel_ has joined #openstack-neutron | 18:01 | |
| *** mchestr has quit IRC | 18:01 | |
| *** mchestr has joined #openstack-neutron | 18:01 | |
| *** s3wong has joined #openstack-neutron | 18:01 | |
| *** kiran-r has quit IRC | 18:02 | |
| *** yeungp1 has joined #openstack-neutron | 18:02 | |
| *** yeungp has quit IRC | 18:02 | |
| *** jbell8 has quit IRC | 18:02 | |
| *** nlahouti has joined #openstack-neutron | 18:03 | |
| kevinbenton | sc68cal, salv-orlando: hang on a sec. does cloud-init not already have support for ipv6? | 18:04 |
| kevinbenton | because if it doesn't then my entire argument is irrelevant | 18:04 |
| aveiga | not explicitly | 18:04 |
| kevinbenton | is it via dns discovery or somethign? | 18:05 |
| sc68cal | kevinbenton: no, I think it just tries the link local ipv4 address, and if it doesn't get anything back it moves on | 18:06 |
| sc68cal | to mounting the config drive | 18:06 |
| *** dboik has quit IRC | 18:07 | |
| *** dboik has joined #openstack-neutron | 18:07 | |
| *** afazekas has quit IRC | 18:07 | |
| russellb | otherwiseguy: weren't you looking at ipv6 support in cloud-init? | 18:07 |
| kevinbenton | aveiga: so what did you mean then? | 18:08 |
| kevinbenton | is there any way now to get cloud-init to try ipv6 now? | 18:08 |
| *** catohornet has left #openstack-neutron | 18:08 | |
| kevinbenton | if not, this was just an epic waste of time | 18:08 |
| aveiga | I don't think so. I mean I think it could be recoded to use an IPv6 address without change to the code other than the hardcoded address (i.e. not storing a 32-bit int for the addr, etc) | 18:09 |
| sc68cal | kevinbenton: possibly through populating metadata_urls | 18:09 |
| *** andrelucas has joined #openstack-neutron | 18:09 | |
| sc68cal | kevinbenton: but otherwise like aveiga says | 18:09 |
| *** apuimedo has joined #openstack-neutron | 18:09 | |
| *** mingy has joined #openstack-neutron | 18:10 | |
| sc68cal | although cloud-init is python I believe, no worries about 32 bit ints | 18:10 |
| kevinbenton | sc68cal: how do those metadata_urls get populated? | 18:10 |
| *** dane_leblanc has quit IRC | 18:10 | |
| *** tflynn has joined #openstack-neutron | 18:10 | |
| *** ildikov has joined #openstack-neutron | 18:11 | |
| kevinbenton | aveiga: but the distributed version of cloud-init is what we need to work with, we can't assume an image will have a tweaked version | 18:11 |
| *** tflynn has quit IRC | 18:11 | |
| sc68cal | kevinbenton: I don't know, I'd have to look | 18:11 |
| aveiga | kevinbenton: I'm not entirely sure, but I think those would break | 18:11 |
| *** tflynn has joined #openstack-neutron | 18:12 | |
| sc68cal | I think it's a config file baked into the AMI/image | 18:12 |
| *** ivar-lazzaro has quit IRC | 18:12 | |
| *** nyechiel_ has quit IRC | 18:12 | |
| kevinbenton | ugh | 18:12 |
| kevinbenton | then yeah, this was a waste of time :) | 18:12 |
| kevinbenton | cloud-init has no ipv6 auto-discovery | 18:13 |
| kevinbenton | so there is nothing neutron could even do | 18:13 |
| kevinbenton | it's adding support for something that doesn't exist | 18:13 |
| aveiga | that's what sc68cal and I were trying to say | 18:13 |
| aveiga | nothing would actively prevent this, but we'd have to provide the new, currently undefined IPv6 addr somehow | 18:14 |
| *** ivar-lazzaro has joined #openstack-neutron | 18:15 | |
| *** angela-s has quit IRC | 18:15 | |
| *** sridhar_ram has quit IRC | 18:16 | |
| *** agireud has joined #openstack-neutron | 18:16 | |
| *** angela-s has joined #openstack-neutron | 18:16 | |
| kevinbenton | aveiga: you mean cloud-init would | 18:17 |
| kevinbenton | right? | 18:17 |
| aveiga | no, I mean we'd have to provide it to cloud-init, since cloud-init currently has no way of knowing it | 18:18 |
| *** sridhar_ram has joined #openstack-neutron | 18:18 | |
| *** sballe has quit IRC | 18:18 | |
| aveiga | or, in the end, yes as cloud-init would have to be updated to include that new value | 18:18 |
| kevinbenton | but if i understand what you both are saying is that cloud-init has no network discovery mechanism | 18:18 |
| aveiga | which is what I was getting at when I was talking about EC2 ever implementing it | 18:18 |
| *** ivar-lazzaro has quit IRC | 18:19 | |
| *** ivar-lazzaro has joined #openstack-neutron | 18:19 | |
| kevinbenton | EC2 doesn't matter. if cloud-init were to implement a discovery mechanism of some sort, we just need to be compatible with that | 18:20 |
| aveiga | kevinbenton: according to http://cloudinit.readthedocs.org/en/latest/topics/datasources.html#what-is-a-datasource no, there is no URL discovery. It just iterates through known methods | 18:21 |
| *** akamyshnikova_ has quit IRC | 18:22 | |
| *** andrelucas has quit IRC | 18:23 | |
| aveiga | the only reason neturon metadata works is that it literally emulates the EC2 configuration | 18:23 |
| *** dannywilson has joined #openstack-neutron | 18:23 | |
| *** annegentle has joined #openstack-neutron | 18:23 | |
| aveiga | which ironically, is why EC2 actually would matter | 18:23 |
| aveiga | from the docs, it seems that cloud-init's discovery is designed to match against the EC2 stuff | 18:24 |
| *** yamahata has quit IRC | 18:25 | |
| *** dannywilson has quit IRC | 18:25 | |
| *** yamahata has joined #openstack-neutron | 18:25 | |
| *** dannywilson has joined #openstack-neutron | 18:25 | |
| kevinbenton | aveiga: no, the only reason neutron metadata works is because it's compatible with cloud-init :) | 18:26 |
| *** aranjan has joined #openstack-neutron | 18:26 | |
| aveiga | kevinbenton: semantics. It's compatible because it emulates EC2. But at this point, we're splitting hairs. It's the same thign | 18:26 |
| kevinbenton | aveiga: well no, the important difference would be that if cloud-init had a feature not yet supported by EC2, we could support it | 18:27 |
| kevinbenton | which is what the whole argument was about | 18:27 |
| kevinbenton | i was under the impression that cloud-init had an address discovery mechanism, | 18:27 |
| aveiga | so you'd create something else, that's not cloud-init metadata, that could be dual-stacked with discovery | 18:27 |
| kevinbenton | aveiga: what do you mean? | 18:28 |
| *** tomoe_ has joined #openstack-neutron | 18:28 | |
| aveiga | well, the "metadata" section that the well-known address works with in cloud-init, is LITERACLLY called EC2 in the docs. So if we made something else, it would be a new discovery method in cloud-init | 18:28 |
| aveiga | either way, you have to change cloud-init | 18:29 |
| kevinbenton | aveiga: oh, right | 18:29 |
| kevinbenton | my argument is that if cloud-init supports something, we should support it | 18:29 |
| kevinbenton | but it's become pointless because cloud-init doesn't support the relevant something i thought it supported | 18:30 |
| aveiga | kevinbenton: hold up there. Just because cloud-init supports, say, RHEVm, we should implement floppy-drive data support? | 18:30 |
| *** andrelucas has joined #openstack-neutron | 18:31 | |
| aveiga | I'd rather focus on getting the config-drive stuff to work, because it's network-agnostic | 18:31 |
| aveiga | and just do ONE thing right | 18:31 |
| aveiga | the metadata service is mostly there so that t ools that work in AWS can be moved over with less rewrite | 18:32 |
| salv-orlando | what's preventing config-drive from working with IPv6? | 18:32 |
| aveiga | salv-orlando: I wasn't aware that anything was | 18:32 |
| kevinbenton | aveiga: if someone takes an image built for amazon, do you not care to make it work on openstack? | 18:32 |
| *** merooney has quit IRC | 18:32 | |
| *** slaweq has joined #openstack-neutron | 18:32 | |
| *** tomoe_ has quit IRC | 18:32 | |
| salv-orlando | aveiga: afaict we even gate upstream with configdrive | 18:33 |
| aveiga | kevinbenton: in order for that image to work on openstack, it would (by default!) require IPv4 support | 18:33 |
| aveiga | so it's a moot point | 18:33 |
| aveiga | just run metadata on the IPv4 addr and call it a day | 18:33 |
| kevinbenton | aveiga: the whole "do one thing right" is misleading since we are basically talking about one iptables rule | 18:33 |
| *** david-lyle has joined #openstack-neutron | 18:33 | |
| kevinbenton | aveiga: the time we have wasted here could have had the support added | 18:33 |
| aveiga | you can add the support all you want, but you still have to change cloud-init | 18:34 |
| aveiga | unless you're talking about a proxy | 18:34 |
| kevinbenton | aveiga: no, i understand that cloud-init has to be changed | 18:34 |
| kevinbenton | what i'm saying is that if cloud-init has some sort of address discovery for ipv6 support i would like to add support in neutron | 18:34 |
| aveiga | well, it doesn't | 18:35 |
| aveiga | so, chicken or egg? | 18:35 |
| kevinbenton | saying, "let the metadata service bitrot because cloudrive" isn't interesting to me | 18:35 |
| *** ijw has joined #openstack-neutron | 18:35 | |
| aveiga | ok | 18:35 |
| kevinbenton | aveiga: well it depends on if the person that filed that RFE is successful in getting the support in cloud-init | 18:35 |
| kevinbenton | aveiga: if the cloud-init community is willing to support something, i don't see why we couldn't add it even if EC2 doesn't support it | 18:36 |
| aveiga | go for it | 18:36 |
| *** padkrish has joined #openstack-neutron | 18:37 | |
| *** Networkn3rd has quit IRC | 18:37 | |
| kevinbenton | well i have to wait to see if cloud-init support is ever added, which is why this conversation was mainly pointless. I thought we were already at that step | 18:37 |
| *** jprovazn has quit IRC | 18:38 | |
| *** mingy has quit IRC | 18:39 | |
| *** aix has quit IRC | 18:39 | |
| *** elmiko has joined #openstack-neutron | 18:40 | |
| elmiko | salv-orlando, HenryG, hi, i'm wondering if you might have a minute or two to talk about the api working group? | 18:40 |
| salv-orlando | elmiko: I have 19 minutes before the next meeting | 18:41 |
| elmiko | salv-orlando: ok, i'll be quick ;) | 18:41 |
| salv-orlando | or plenty of time in 109 minute s;) | 18:41 |
| elmiko | no worries, this won't take long | 18:41 |
| elmiko | my name is michael mccune and i'm just doing some outreach from the wg to share a couple of links and touch base with the projects | 18:42 |
| elmiko | we have a set of merge guidelines that up for review now which talk about how we're going to improve the process https://review.openstack.org/#/c/186836/ | 18:42 |
| elmiko | and the nova folks have put together a set of liaison responsibilities that they are experimenting with, https://wiki.openstack.org/wiki/Nova/APIWGLiaisons | 18:43 |
| elmiko | so i'm just passing these around and talking with the listed liaisons in hopes of spreading the word =) | 18:43 |
| elmiko | i imagine in the next few weeks we'll be increasing our communication about upcoming guidelines, posting to the ml, and with luck getting some input from the liaisons before we do final merges | 18:44 |
| salv-orlando | elmiko: I can prepare a similar draft for Neutron and publish it on the wiki, but perhaps it does not make sense a lot because I expect the same responsibilities to apply to every project | 18:44 |
| elmiko | salv-orlando: yea, and imo, i think we should incorporate these responsibilities into the api-wg site. but they are under test currently | 18:44 |
| salv-orlando | elmiko: I'll start reviewing openstack/api-wg patches again, then ;) I've taken a bit of a break to complete some other project | 18:45 |
| salv-orlando | elmiko: starting with patch #186836 | 18:45 |
| elmiko | cool! part of our new process will be to freeze guidelines once we are happy with the language, and then during the freeze we will post the reviews to the openstack-dev ml, and talk about them in the cross project meeting | 18:45 |
| *** mlavalle has quit IRC | 18:46 | |
| elmiko | ideally, liaisons will be able to review the frozen guidelines and if necessary add comments. | 18:46 |
| elmiko | i think this will help reduce the volume of reviews to the ones that are ready for merging | 18:46 |
| salv-orlando | makes sense to me. btw, api-wg reviews have never been that many. | 18:47 |
| elmiko | and if you have any questions, comments, concerns, we usually hang out in openstack-api | 18:47 |
| elmiko | yea, it's pretty low volume but we are starting to get more throughput | 18:48 |
| *** hitalia has quit IRC | 18:48 | |
| salv-orlando | nah just one thing that it was puzzling me... would it be possible to add something to the commit message to identify which patches contain a guideline? or the branch topic? | 18:48 |
| elmiko | that's a good idea, i'll add it to the agenda for our meeting this week | 18:48 |
| *** andrelucas has quit IRC | 18:49 | |
| salv-orlando | elmiko: it's not a major thing, but this will help liasions to focus only on the things that matter to them. (no big deal anyway, I can usually immediately get if a patch has a guideline or amends something else!) | 18:50 |
| elmiko | salv-orlando: yea, and this is partially why we are instituting the freeze for larger guidelines. it will help with visibility on the big ones. | 18:50 |
| openstackgerrit | Merged openstack/neutron-vpnaas: Enable random hash seeds https://review.openstack.org/187107 | 18:51 |
| elmiko | salv-orlando: thanks for your time, good luck with the meeting =) | 18:52 |
| *** tkatarki has quit IRC | 18:52 | |
| *** baoli has quit IRC | 18:52 | |
| openstackgerrit | Merged openstack/neutron: Make MockFixedIntervalLoopingCall class as a helper class https://review.openstack.org/187885 | 18:52 |
| Swami | mestery, armax : here is the tempest patch that causes the DVR job failure rate to spike up. https://review.openstack.org/#/c/188139/ | 18:52 |
| mestery | Swami: nice work! | 18:53 |
| *** pradipta has joined #openstack-neutron | 18:53 | |
| Swami | mestery: we have been in the race of chasing and reverting patches for dvr. It would be great if we could make is voting before another such event occurs. Because this might get triggered both from tempest as well as from neutron. | 18:54 |
| *** tkatarki has joined #openstack-neutron | 18:54 | |
| mestery | Swami: Indeed. Lets see it stabilize once this merges and then get it voting. | 18:54 |
| *** nlahouti has quit IRC | 18:54 | |
| Swami | mestery: thanks | 18:54 |
| *** merooney has joined #openstack-neutron | 18:54 | |
| *** mordred has quit IRC | 18:55 | |
| *** mordred has joined #openstack-neutron | 18:55 | |
| Swami | kevinbenton: sorry to bother your patch. I just abandoned the revert | 18:56 |
| *** merooney has quit IRC | 18:56 | |
| *** Networkn3rd has joined #openstack-neutron | 18:56 | |
| russellb | sc68cal: so, i need to do some devstack neutron work ... should i hack neutron-legacy? or something else? specifically, there is some pre-req work i need to do to be able to do a grenade job that tests rolling upgrades | 18:57 |
| *** barra204 has joined #openstack-neutron | 18:57 | |
| russellb | i need to be able to control starting/stopping the l2 agent separately from the other agents | 18:58 |
| russellb | since in this test, we'll leave it running and upgrade / restart everything else | 18:58 |
| russellb | but basically, just don't want to waste time hacking on part of devstack that is frozen or something | 18:58 |
| *** mlavalle has joined #openstack-neutron | 18:59 | |
| *** hitalia has joined #openstack-neutron | 18:59 | |
| *** Networkn3rd has quit IRC | 19:00 | |
| *** Networkn3rd has joined #openstack-neutron | 19:00 | |
| *** ijw has quit IRC | 19:00 | |
| *** barra204 has quit IRC | 19:02 | |
| *** tkatarki has quit IRC | 19:03 | |
| *** padkrish has quit IRC | 19:03 | |
| openstackgerrit | Adolfo Duarte proposed openstack/neutron: HA network requires port_security_enabled to be set https://review.openstack.org/188146 | 19:03 |
| *** sadasu_ has quit IRC | 19:04 | |
| *** nlahouti has joined #openstack-neutron | 19:06 | |
| *** padkrish_ has joined #openstack-neutron | 19:06 | |
| *** agireud_ has joined #openstack-neutron | 19:08 | |
| *** agireud has quit IRC | 19:09 | |
| *** agireud_ is now known as agireud | 19:09 | |
| *** aranjan has quit IRC | 19:11 | |
| *** slaweq has quit IRC | 19:15 | |
| *** abregman has joined #openstack-neutron | 19:16 | |
| *** tkatarki has joined #openstack-neutron | 19:17 | |
| *** Networkn3rd has quit IRC | 19:17 | |
| *** apuimedo has quit IRC | 19:19 | |
| *** pradipta has quit IRC | 19:20 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Add RPC command to delete FloatingIP Agent GW Port https://review.openstack.org/159317 | 19:22 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Delete Floatingip agent gateway port through rpc https://review.openstack.org/187769 | 19:23 |
| *** jbell8 has joined #openstack-neutron | 19:23 | |
| *** tholtzen has joined #openstack-neutron | 19:24 | |
| *** tholtzen has left #openstack-neutron | 19:24 | |
| *** ronis has quit IRC | 19:24 | |
| *** mingy has joined #openstack-neutron | 19:24 | |
| russellb | sc68cal: nm, got dtroyer in #openstack-dev | 19:24 |
| *** mingy has left #openstack-neutron | 19:25 | |
| *** jgrimm is now known as zz_jgrimm | 19:26 | |
| *** merooney has joined #openstack-neutron | 19:28 | |
| openstackgerrit | Ritesh Anand proposed openstack/neutron: Added 44 test cases for DVR L3 schedulers. 11 cases listed below are to run for Chance and Least-routers scheduler, for both auto and explicit scheduling. https://review.openstack.org/188157 | 19:31 |
| *** amuller is now known as amuller_afk | 19:35 | |
| *** Swami has quit IRC | 19:35 | |
| *** rushil has quit IRC | 19:36 | |
| *** tkatarki has quit IRC | 19:36 | |
| *** igordcard_ has quit IRC | 19:40 | |
| *** claudiub has joined #openstack-neutron | 19:40 | |
| *** padkrish_ has quit IRC | 19:44 | |
| *** padkrish has joined #openstack-neutron | 19:45 | |
| *** tkatarki has joined #openstack-neutron | 19:46 | |
| *** nlahouti has quit IRC | 19:47 | |
| *** merooney has quit IRC | 19:47 | |
| *** hitalia has quit IRC | 19:49 | |
| *** padkrish has quit IRC | 19:50 | |
| *** mchestr has quit IRC | 19:50 | |
| *** elmiko has left #openstack-neutron | 19:50 | |
| *** mchestr has joined #openstack-neutron | 19:50 | |
| *** crc32 has joined #openstack-neutron | 19:51 | |
| *** Networkn3rd has joined #openstack-neutron | 19:51 | |
| *** _sri has quit IRC | 19:53 | |
| *** armax has quit IRC | 19:53 | |
| *** jpena has joined #openstack-neutron | 19:55 | |
| *** andrelucas has joined #openstack-neutron | 19:55 | |
| *** mchestr has quit IRC | 19:57 | |
| *** mchestr has joined #openstack-neutron | 19:57 | |
| *** padkrish has joined #openstack-neutron | 19:58 | |
| *** barra204 has joined #openstack-neutron | 19:58 | |
| *** mingy has joined #openstack-neutron | 20:01 | |
| *** dave-mcc_ has quit IRC | 20:02 | |
| *** barra204 has quit IRC | 20:03 | |
| *** marun_ has joined #openstack-neutron | 20:03 | |
| *** marun has quit IRC | 20:03 | |
| *** sballe has joined #openstack-neutron | 20:04 | |
| *** Aish has joined #openstack-neutron | 20:04 | |
| *** ajmiller has joined #openstack-neutron | 20:05 | |
| *** merooney has joined #openstack-neutron | 20:05 | |
| *** dave-mccowan has joined #openstack-neutron | 20:05 | |
| *** daneyon_ has quit IRC | 20:08 | |
| *** daneyon has joined #openstack-neutron | 20:08 | |
| openstackgerrit | Doug Wiegley proposed openstack/neutron: WIP - DO NOT MERGE - Test that lbaas job is using the right neutron https://review.openstack.org/188166 | 20:08 |
| *** rkukura has quit IRC | 20:08 | |
| *** salv-orlando has quit IRC | 20:09 | |
| *** rushil has joined #openstack-neutron | 20:09 | |
| *** salv-orlando has joined #openstack-neutron | 20:09 | |
| *** rkukura has joined #openstack-neutron | 20:09 | |
| *** baoli has joined #openstack-neutron | 20:10 | |
| *** thurloat is now known as thurloat_isgone | 20:10 | |
| *** daneyon_ has joined #openstack-neutron | 20:11 | |
| *** baoli has quit IRC | 20:13 | |
| *** baoli has joined #openstack-neutron | 20:13 | |
| *** jpena has quit IRC | 20:14 | |
| *** hitalia has joined #openstack-neutron | 20:14 | |
| *** daneyon has quit IRC | 20:14 | |
| *** alexpilotti has quit IRC | 20:17 | |
| *** gridinv_ has joined #openstack-neutron | 20:19 | |
| *** gridinv_ has quit IRC | 20:24 | |
| *** busterswt has joined #openstack-neutron | 20:25 | |
| *** merooney has quit IRC | 20:27 | |
| *** hitalia has quit IRC | 20:27 | |
| *** Sukhdev has joined #openstack-neutron | 20:29 | |
| *** tkatarki has quit IRC | 20:29 | |
| *** matrohon has joined #openstack-neutron | 20:29 | |
| *** tkatarki has joined #openstack-neutron | 20:30 | |
| *** jbell8 has quit IRC | 20:31 | |
| *** padkrish has quit IRC | 20:33 | |
| *** padkrish has joined #openstack-neutron | 20:33 | |
| *** padkrish_ has joined #openstack-neutron | 20:35 | |
| *** padkrish has quit IRC | 20:35 | |
| russellb | dougwig: nice (last gerrit patch), glad that's getting fixed :) | 20:36 |
| *** davidchep has quit IRC | 20:41 | |
| *** banix_ has quit IRC | 20:43 | |
| *** nyechiel_ has joined #openstack-neutron | 20:43 | |
| *** tiswanso has quit IRC | 20:44 | |
| *** madhu_ak_ has joined #openstack-neutron | 20:44 | |
| *** matrohon has quit IRC | 20:46 | |
| *** baoli has quit IRC | 20:46 | |
| *** madhu_ak has quit IRC | 20:47 | |
| *** nlahouti has joined #openstack-neutron | 20:51 | |
| *** rushil has quit IRC | 20:52 | |
| *** apuimedo has joined #openstack-neutron | 20:54 | |
| *** davidchep has joined #openstack-neutron | 20:54 | |
| *** cbouch has quit IRC | 20:54 | |
| *** hitalia has joined #openstack-neutron | 20:54 | |
| *** madhu_ak_ is now known as madhu_ak | 20:55 | |
| kevinbenton | mestery: that patch swami reverted is interesting | 20:55 |
| *** nyechiel_ has quit IRC | 20:55 | |
| kevinbenton | mestery: it sounds like nuage is going to break | 20:55 |
| kevinbenton | mestery: with the revert | 20:55 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 20:56 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/188098 | 20:57 |
| *** baoli has joined #openstack-neutron | 20:57 | |
| mestery | kevinbenton: heh, while, with the patch, the DVR job was headed to 100% failure rate | 20:58 |
| *** marun_ has quit IRC | 20:58 | |
| kevinbenton | mestery: Yeah, it sounds like nuage is the less compliant one by not allowing a network to exist without a router | 20:59 |
| *** davidchep has quit IRC | 21:00 | |
| mestery | kevinbenton: Agreed, lets unblock the rest of the world at the expense of nuage in this case. | 21:00 |
| kevinbenton | mestery: so I prefer the revert, but it makes me wonder if we need to define a tempest test that doesn't depend on a router | 21:00 |
| *** Networkn3rd has quit IRC | 21:00 | |
| mestery | kevinbenton: I think you're on to something there :) | 21:01 |
| *** Networkn3rd has joined #openstack-neutron | 21:01 | |
| *** Networkn3rd has quit IRC | 21:01 | |
| kevinbenton | mestery: i think they've been getting by on the coincidence that tempest tests use floating IPs to access everything | 21:01 |
| kevinbenton | mestery: or they just skip the broken ones :) | 21:02 |
| mestery | heh | 21:02 |
| kevinbenton | Do we have an @skipIfFails decorator? | 21:03 |
| kevinbenton | :) | 21:03 |
| *** davidchep has joined #openstack-neutron | 21:04 | |
| *** padkrish_ has quit IRC | 21:04 | |
| *** padkrish has joined #openstack-neutron | 21:05 | |
| *** annegentle has quit IRC | 21:06 | |
| *** amuller_afk is now known as amuller | 21:09 | |
| *** padkrish has quit IRC | 21:09 | |
| *** padkrish has joined #openstack-neutron | 21:12 | |
| *** rushil has joined #openstack-neutron | 21:12 | |
| *** rossella_s has quit IRC | 21:13 | |
| salv-orlando | kevinbenton, mestery: I'm kind of sure that I and mtreinish defined several tests where routers are not created about 16 months ago | 21:14 |
| salv-orlando | so perhaps those are skipped in nuage CI | 21:14 |
| kevinbenton | salv-orlando: yeah, i imagine that's the case | 21:14 |
| mestery | ++ | 21:14 |
| salv-orlando | kevinbenton: like if they're running only tempest.api.network | 21:15 |
| *** ivar-lazzaro has quit IRC | 21:16 | |
| *** dboik_ has joined #openstack-neutron | 21:17 | |
| *** ivar-lazzaro has joined #openstack-neutron | 21:17 | |
| *** tkatarki has quit IRC | 21:17 | |
| *** padkrish has quit IRC | 21:17 | |
| *** dboik has quit IRC | 21:20 | |
| *** padkrish_ has joined #openstack-neutron | 21:20 | |
| mingy | I recently pushed a commit to python-neutronclient on gerrit. However, the Jenkins tests currently fail due to an unrelated issue (which I suspect is https://bugs.launchpad.net/openstack-gate/+bug/1455102) . What should I do in this case? (I am new to openstack) | 21:21 |
| openstack | Launchpad bug 1455102 in python-novaclient "some test jobs broken by tox 2.0 not passing env variables" [High,In progress] - Assigned to Sean Dague (sdague) | 21:21 |
| *** dboik_ has quit IRC | 21:21 | |
| *** alex_klimov has joined #openstack-neutron | 21:21 | |
| *** TrevorV_ has quit IRC | 21:23 | |
| *** padkrish_ has quit IRC | 21:25 | |
| *** baoli has quit IRC | 21:28 | |
| kevinbenton | mingy: you will have to wait until it's fixed | 21:28 |
| kevinbenton | mingy: or you can make your patch dependent on the fix if the fix is already up for review | 21:29 |
| *** baoli has joined #openstack-neutron | 21:29 | |
| kevinbenton | mingy: it looks like the fix for neutron has already merged https://review.openstack.org/#/q/Ifd00abed7bf0a68d4d46d12230118022fa2292ef,n,z | 21:29 |
| kevinbenton | mingy: in which case you can just click the rebase button on your patch or issue a "recheck" comment | 21:30 |
| *** rpothier has quit IRC | 21:30 | |
| *** hitalia has quit IRC | 21:30 | |
| kevinbenton | mingy: oh sorry, those patches were for neutron, not the client | 21:33 |
| *** banix has joined #openstack-neutron | 21:34 | |
| kevinbenton | mingy: so you'll have to wait for the fix | 21:34 |
| dougwig | i am pleasantly surprised to see that all the third-party CIs failed on this one: https://review.openstack.org/#/c/188166/ | 21:37 |
| kevinbenton | dougwig: reviewed | 21:38 |
| dougwig | kevinbenton: lol, smartass. | 21:39 |
| *** bjornar_ has quit IRC | 21:39 | |
| *** dontalton has joined #openstack-neutron | 21:39 | |
| mingy | kevinbenton: cool, how do I make my patch dependent on the fix? | 21:40 |
| *** patrickeast has joined #openstack-neutron | 21:40 | |
| openstackgerrit | Paul Michali proposed openstack/neutron-vpnaas: VPNaaS: And devref doc infrastructure https://review.openstack.org/188185 | 21:40 |
| *** busterswt has quit IRC | 21:40 | |
| *** madhu_ak_ has joined #openstack-neutron | 21:40 | |
| kevinbenton | mingy: so once the commit is available, you can check it out and then cherry-pick your patch on top of it and re-run "git review" | 21:41 |
| kevinbenton | mingy: http://docs.openstack.org/infra/manual/developers.html | 21:41 |
| mingy | kevinbenton: gotcha, will do :) thanks! | 21:41 |
| kevinbenton | mingy: search the page for "adding a dependency" | 21:41 |
| *** madhu_ak has quit IRC | 21:43 | |
| kevinbenton | russellb: ping | 21:44 |
| *** alexpilotti has joined #openstack-neutron | 21:45 | |
| openstackgerrit | Kahou Lei proposed openstack/python-neutronclient: Invalid fixed-ip value should return meaningful message https://review.openstack.org/186985 | 21:48 |
| *** padkrish has joined #openstack-neutron | 21:48 | |
| *** mwagner_afk has quit IRC | 21:49 | |
| *** padkrish has quit IRC | 21:52 | |
| *** ianbrown has joined #openstack-neutron | 21:52 | |
| *** mwagner_afk has joined #openstack-neutron | 21:53 | |
| *** johnbelamaric has quit IRC | 21:53 | |
| kevinbenton | salv-orlando: ping | 21:58 |
| *** abregman has quit IRC | 22:00 | |
| *** ianbrown has quit IRC | 22:00 | |
| *** agireud has quit IRC | 22:01 | |
| *** armax has joined #openstack-neutron | 22:03 | |
| *** ianbrown has joined #openstack-neutron | 22:05 | |
| *** crc32 has quit IRC | 22:05 | |
| *** thurloat_isgone is now known as thurloat | 22:05 | |
| salv-orlando | hi kevinbenton | 22:08 |
| amuller | HenryG: heya, got a question about port_security | 22:08 |
| *** wigwam_work has quit IRC | 22:09 | |
| *** rkukura has quit IRC | 22:16 | |
| *** rkukura has joined #openstack-neutron | 22:16 | |
| *** mwagner_afk has quit IRC | 22:16 | |
| amuller | HenryG: salv-orlando: Any idea why port_security for ML2 was implemented via the external extensions option thingie, and it's not loaded by default? | 22:17 |
| *** dave-mccowan has quit IRC | 22:18 | |
| patrickeast | hi everyone, i’m working on my cinder ci system and trying to add some compute nodes, we are set up with neutron with openvswitch and vxlan tunnels, intances on the new compute nodes are unable to get ip addresses and i traced through some of the network setup.. the dhcp broadcasts are getting into br-int and look like they are being dropped in br-tun, my br-tun flows here: http://paste.openstack.org/show/260700/ | 22:18 |
| patrickeast | if i am interpreting that correctly all roads lead to actions=drop | 22:18 |
| patrickeast | which seems wrong… | 22:18 |
| patrickeast | so, is it safe to modify that by hand? or does this point to a misconfiguration of my openvswitch agent? | 22:18 |
| patrickeast | apologies if this is a dumb question, i know more about block storage than networking :( | 22:19 |
| *** dave-mccowan has joined #openstack-neutron | 22:20 | |
| *** jamielennox|away is now known as jamielennox | 22:20 | |
| *** padkrish has joined #openstack-neutron | 22:23 | |
| kevinbenton | patrickeast: it usually points to a misconfiguration | 22:23 |
| kevinbenton | patrickeast: anything you add manually will be wiped out from that flow table on agent restart | 22:23 |
| patrickeast | kevinbenton: ok cool, thats good to know | 22:24 |
| *** dave-mccowan has quit IRC | 22:24 | |
| kevinbenton | patrickeast: if you do a neutron port-show on the port that belongs to one of those instances, what does it show? | 22:24 |
| patrickeast | kevinbenton: http://paste.openstack.org/show/260714/ | 22:26 |
| patrickeast | kevinbenton: anything i should be checking for there? or look out of the ordinary? | 22:27 |
| *** padkrish has quit IRC | 22:27 | |
| kevinbenton | patrickeast: the status | 22:27 |
| kevinbenton | patrickeast: and the binding fields to look for "binding_failed" | 22:28 |
| *** kbringard has quit IRC | 22:28 | |
| kevinbenton | patrickeast: but the status is ACTIVE so that means the agent thinks it has wired everything up correctly | 22:28 |
| kevinbenton | patrickeast: for that VM at least | 22:28 |
| patrickeast | kevinbenton: hmm yea, looks ok from there | 22:28 |
| *** sridhar_ram1 has joined #openstack-neutron | 22:28 | |
| *** Sukhdev has quit IRC | 22:29 | |
| *** dguerri is now known as dguerri`away | 22:29 | |
| *** rushil has quit IRC | 22:29 | |
| openstackgerrit | Carl Baldwin proposed openstack/neutron-specs: Introduce address scopes https://review.openstack.org/180267 | 22:30 |
| kevinbenton | patrickeast: what about the port of the dhcp server | 22:30 |
| kevinbenton | patrickeast: can you do the port show there? | 22:30 |
| patrickeast | kevinbenton: yep, one sec | 22:31 |
| *** sridhar_ram has quit IRC | 22:31 | |
| *** banix has quit IRC | 22:31 | |
| *** banix has joined #openstack-neutron | 22:32 | |
| patrickeast | kevinbenton: this one also looks ok afaik http://paste.openstack.org/show/260715/ | 22:32 |
| *** sridhar_ram1 has quit IRC | 22:33 | |
| *** chem has quit IRC | 22:34 | |
| openstackgerrit | Ming Yang proposed openstack/python-neutronclient: This change allows the bash completion script to work with BSD sed https://review.openstack.org/186870 | 22:34 |
| amuller | patrickeast: can you net-show on the network the instances are connected to? I'd bet it's of type local | 22:35 |
| amuller | patrickeast: where it should be of type vxlan | 22:35 |
| amuller | patrickeast: also can you check if you have l2pop enabled? in the mechanism_drivers list of the ml2 conf file | 22:35 |
| *** Zemeio has quit IRC | 22:36 | |
| patrickeast | amuller: its vxlan http://paste.openstack.org/show/260716/ | 22:36 |
| patrickeast | (i think) | 22:36 |
| patrickeast | lemmie check on the l2pop | 22:36 |
| kevinbenton | amuller: i was just looking at l2pop, if l2_population is enabled in the client but the md is disabled, will that result in this issue? | 22:36 |
| amuller | kevinbenton: yes it would | 22:36 |
| amuller | kevinbenton: it would cause exactly this issue | 22:36 |
| *** sridhar_ram has joined #openstack-neutron | 22:37 | |
| kevinbenton | patrickeast: so check your ml2 server conf and see if the l2 population mechanism driver is there | 22:37 |
| kevinbenton | patrickeast: if it's not, you need to have l2_population disabled on each agent | 22:37 |
| amuller | patrickeast: btw what deployment tool are you using? devstack I guess? | 22:38 |
| *** banix has quit IRC | 22:38 | |
| patrickeast | kevinbenton: ok cool, i’ll check that… i vaugely recall seeing that and wondering if i needed to enable it or not | 22:38 |
| patrickeast | amuller: heh well.. its an RDO juno install that we have since modified a bunch and are manually adding a few more compute nodes to | 22:39 |
| *** padkrish has joined #openstack-neutron | 22:39 | |
| patrickeast | we do weird things like bridges to 10Gb iscsi connections so our devstack VM’s run on this can do tempest tests against our flash array and cinder driver | 22:39 |
| amuller | patrickeast: with packstack? | 22:40 |
| patrickeast | yea | 22:40 |
| patrickeast | initially | 22:40 |
| amuller | ahh ok | 22:40 |
| amuller | because off the bat packstack should be fine | 22:40 |
| *** shwetaap has quit IRC | 22:40 | |
| kevinbenton | patrickeast, amuller: i need to step away for a bit. i'll be back on in 20 mins or so | 22:40 |
| amuller | if you modify the confs after that that's on you =D | 22:40 |
| patrickeast | haha yea, we are definitely off the beaten path | 22:40 |
| *** shwetaap has joined #openstack-neutron | 22:41 | |
| patrickeast | amuller: ok so i see l2_population is false right now | 22:42 |
| amuller | patrickeast: can you paste the ml2 conf on your server and the confs you pass to the OVS agents on all nodes? | 22:43 |
| patrickeast | amuller: can do, one sec | 22:43 |
| *** madhu_ak_ is now known as madhu_ak | 22:44 | |
| patrickeast | here is the ovs plugin config for the compute node being added: http://paste.openstack.org/show/260718/ | 22:46 |
| patrickeast | one for the ovs plugin on the controller/compute/network node we are adding it to (and is working correctly): http://paste.openstack.org/show/260719/ | 22:47 |
| patrickeast | amuller: ^ | 22:47 |
| * amuller looking | 22:47 | |
| patrickeast | ok so heres a confusing part, which im assuming i just missed something | 22:48 |
| *** barra204 has joined #openstack-neutron | 22:48 | |
| *** rkukura has quit IRC | 22:48 | |
| *** shwetaap has quit IRC | 22:48 | |
| patrickeast | i dont see any services using the /etc/neutron/plugin.ini (and in turn ml2_conf.ini) which has the ml2 section, is that supposed to be the case? | 22:49 |
| patrickeast | or have i gone terribly wrong somewhere | 22:49 |
| *** csoukup has quit IRC | 22:49 | |
| amuller | patrickeast: neutron-server systemd init file is supposed to be using plugin.ini | 22:49 |
| amuller | patrickeast: according to my memory :) | 22:49 |
| patrickeast | ooh yea | 22:50 |
| patrickeast | ok | 22:50 |
| patrickeast | just none of the agents | 22:50 |
| *** rkukura has joined #openstack-neutron | 22:50 | |
| amuller | patrickeast: hey l2_population is True on the node you're adding | 22:50 |
| patrickeast | oh | 22:50 |
| amuller | patrickeast: you should disable that =D | 22:50 |
| patrickeast | i just set that | 22:50 |
| * patrickeast is playing with things | 22:51 | |
| patrickeast | ok so it should be disabled | 22:51 |
| patrickeast | its back to false | 22:51 |
| amuller | patrickeast: and restart that agent :) | 22:52 |
| amuller | patrickeast: also how's the plugin.ini on the controller looking? | 22:52 |
| patrickeast | oh oops, http://paste.openstack.org/show/260732/ | 22:52 |
| patrickeast | thought i linked it already | 22:52 |
| amuller | patrickeast: can you try restarting the server and all agents, delete instances, delete networks, re-create, and see if it works now? | 22:53 |
| *** barra204 has quit IRC | 22:53 | |
| *** apuimedo has quit IRC | 22:53 | |
| patrickeast | :o all of them? | 22:53 |
| *** shwetaap has joined #openstack-neutron | 22:54 | |
| patrickeast | i didn’t actually load it with the l2_population = True config though | 22:54 |
| patrickeast | amuller: or even without it, thats the next course of action? | 22:54 |
| amuller | patrickeast: ahh if you never actually had it configured with True I don't actually know what's wrong =D | 22:55 |
| patrickeast | haha whew | 22:55 |
| *** johnbelamaric has joined #openstack-neutron | 22:55 | |
| amuller | patrickeast: you're supposed to have broadcast flows in table 22 | 22:55 |
| amuller | patrickeast: if everything is configured correctly that's what happens | 22:55 |
| patrickeast | yea the flow on the other node (thats working) looks much nicer for br-tun | 22:55 |
| *** mlavalle has quit IRC | 22:56 | |
| patrickeast | things go through the vxlan business and back into the system instead of being dropped | 22:56 |
| amuller | patrickeast: I would try turning off l2_population on that new OVS agent and restarting though and see if br-tun gets populated with the correct flows | 22:56 |
| *** sridhar_ram has quit IRC | 22:56 | |
| patrickeast | amuller: sounds good | 22:57 |
| openstackgerrit | Merged openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/188098 | 22:57 |
| *** sridhar_ram has joined #openstack-neutron | 22:57 | |
| *** a_le has quit IRC | 22:58 | |
| openstackgerrit | Sourabh Patwardhan proposed openstack/neutron: Plug port into Cisco Nexus1000V switch https://review.openstack.org/159306 | 22:58 |
| *** Sukhdev has joined #openstack-neutron | 22:59 | |
| *** baoli has quit IRC | 23:01 | |
| patrickeast | amuller: no luck, still the same | 23:01 |
| patrickeast | well, i’m going to go sift around a bit in the docs and see if there was any step i missed, maybe try re-installing the compute node | 23:02 |
| patrickeast | amuller: kevinbenton: thanks for your help with this, i really appreciate it! | 23:02 |
| patrickeast | i might be back with more questions later in the week | 23:02 |
| *** aranjan has joined #openstack-neutron | 23:03 | |
| amuller | patrickeast: can you check the new OVS agent log? It should have Assigning %(vlan_id)s as local vlan for... log | 23:04 |
| amuller | with %(vlan_id) being some randon number | 23:04 |
| *** carl_baldwin has quit IRC | 23:05 | |
| amuller | patrickeast: then if you're up for it you can edit the ovs agent py and add a log in "def provision_local_vlan" and confirm that it's not adding the flood flow, then try to figure out why | 23:05 |
| *** padkrish has quit IRC | 23:08 | |
| *** pc_m has quit IRC | 23:09 | |
| *** rkukura has quit IRC | 23:09 | |
| openstackgerrit | Merged openstack/neutron: Refactor type_tunnel/gre/vxlan to reduce duplicate code https://review.openstack.org/187011 | 23:10 |
| openstackgerrit | Merged openstack/neutron: Add sub-project lieutenants https://review.openstack.org/187733 | 23:10 |
| *** padkrish has joined #openstack-neutron | 23:11 | |
| *** wigwam_work has joined #openstack-neutron | 23:14 | |
| *** johnbelamaric has quit IRC | 23:15 | |
| *** yeungp has joined #openstack-neutron | 23:17 | |
| *** yeungp1 has quit IRC | 23:17 | |
| *** chlong has joined #openstack-neutron | 23:19 | |
| *** gildub has joined #openstack-neutron | 23:20 | |
| *** thurloat is now known as thurloat_isgone | 23:22 | |
| *** rkukura has joined #openstack-neutron | 23:23 | |
| patrickeast | amuller: yep, will do | 23:23 |
| *** banix has joined #openstack-neutron | 23:25 | |
| *** sballe has quit IRC | 23:28 | |
| *** sridhar_ram has quit IRC | 23:31 | |
| *** Swami has joined #openstack-neutron | 23:33 | |
| openstackgerrit | Aaron Rosen proposed openstack/neutron: Add stt type driver to ML2 https://review.openstack.org/180331 | 23:34 |
| *** sdake has quit IRC | 23:35 | |
| *** dontalton has quit IRC | 23:37 | |
| *** mchestr has quit IRC | 23:37 | |
| *** yamamoto has joined #openstack-neutron | 23:40 | |
| *** rwsu has quit IRC | 23:41 | |
| *** tidwellr has left #openstack-neutron | 23:41 | |
| *** a_le has joined #openstack-neutron | 23:41 | |
| *** a_le has quit IRC | 23:45 | |
| *** johnbelamaric has joined #openstack-neutron | 23:45 | |
| openstackgerrit | Kahou Lei proposed openstack/neutron: net-list should return gracefully for existing non port security enabled network https://review.openstack.org/188209 | 23:45 |
| *** dims_ has joined #openstack-neutron | 23:51 | |
| *** yfujioka has joined #openstack-neutron | 23:52 | |
| *** jbell8 has joined #openstack-neutron | 23:53 | |
| *** dims__ has quit IRC | 23:55 | |
| *** padkrish has quit IRC | 23:57 | |
| *** aranjan has quit IRC | 23:57 | |
| *** claudiub has quit IRC | 23:57 | |
| *** padkrish has joined #openstack-neutron | 23:57 | |
| *** aranjan has joined #openstack-neutron | 23:58 | |
| *** ajmiller has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!