| *** gangil has joined #openstack-kuryr | 00:02 | |
| *** gangil has quit IRC | 00:04 | |
| *** banix has quit IRC | 00:12 | |
| *** diogogmt has quit IRC | 00:23 | |
| *** shashank_hegde has quit IRC | 00:23 | |
| *** gangil has joined #openstack-kuryr | 00:24 | |
| *** diogogmt has joined #openstack-kuryr | 00:27 | |
| *** abhiraut has quit IRC | 00:31 | |
| *** diogogmt has quit IRC | 00:36 | |
| *** gangil has quit IRC | 01:06 | |
| *** diogogmt has joined #openstack-kuryr | 01:14 | |
| *** gangil has joined #openstack-kuryr | 01:57 | |
| *** yamamoto_ has joined #openstack-kuryr | 02:06 | |
| *** banix has joined #openstack-kuryr | 02:52 | |
| *** yuanying has quit IRC | 02:54 | |
| *** yamamoto_ has quit IRC | 03:26 | |
| *** vikasc has joined #openstack-kuryr | 03:43 | |
| *** banix has quit IRC | 03:43 | |
| *** yuanying has joined #openstack-kuryr | 03:46 | |
| openstackgerrit | Mike Spreitzer proposed openstack/kuryr: Add devref for kubernetes translate 1 and 2 https://review.openstack.org/290172 | 04:04 |
|---|---|---|
| *** yamamoto_ has joined #openstack-kuryr | 04:09 | |
| *** shashank_hegde has joined #openstack-kuryr | 04:30 | |
| *** shashank_hegde has quit IRC | 04:37 | |
| *** shashank_hegde has joined #openstack-kuryr | 04:39 | |
| *** shashank_hegde has quit IRC | 04:51 | |
| *** shashank_hegde has joined #openstack-kuryr | 04:52 | |
| *** oanson has joined #openstack-kuryr | 05:07 | |
| *** irenab has joined #openstack-kuryr | 05:25 | |
| *** salv-orlando has joined #openstack-kuryr | 05:51 | |
| *** fawadkhaliq has joined #openstack-kuryr | 05:57 | |
| *** salv-orlando has quit IRC | 06:32 | |
| *** shashank_hegde has quit IRC | 06:33 | |
| *** shashank_hegde has joined #openstack-kuryr | 06:34 | |
| *** tfukushima has joined #openstack-kuryr | 06:41 | |
| *** devvesa has joined #openstack-kuryr | 07:35 | |
| *** irenab_ has joined #openstack-kuryr | 07:36 | |
| *** irenab has quit IRC | 07:38 | |
| *** irenab_ is now known as irenab | 07:38 | |
| *** fawadkhaliq has quit IRC | 07:39 | |
| *** fawadkhaliq has joined #openstack-kuryr | 07:39 | |
| openstackgerrit | Merged openstack/kuryr: Trivial doc fixes https://review.openstack.org/317650 | 07:47 |
| *** salv-orl_ has joined #openstack-kuryr | 07:47 | |
| *** salv-orl_ has quit IRC | 07:47 | |
| *** salv-orlando has joined #openstack-kuryr | 07:48 | |
| openstackgerrit | Merged openstack/kuryr: startup: Fix Kuryr initialization https://review.openstack.org/317798 | 07:48 |
| *** gangil1 has joined #openstack-kuryr | 08:08 | |
| *** gangil has quit IRC | 08:10 | |
| *** shashank_hegde has quit IRC | 08:38 | |
| *** gangil has joined #openstack-kuryr | 08:44 | |
| *** gangil1 has quit IRC | 08:45 | |
| *** neiljerram has quit IRC | 08:51 | |
| *** gangil1 has joined #openstack-kuryr | 09:07 | |
| *** gangil has quit IRC | 09:08 | |
| *** gangil1 has quit IRC | 09:14 | |
| *** gangil has joined #openstack-kuryr | 09:14 | |
| *** gangil1 has joined #openstack-kuryr | 09:18 | |
| *** gangil has quit IRC | 09:18 | |
| *** fawadkhaliq has quit IRC | 09:24 | |
| *** fawadkhaliq has joined #openstack-kuryr | 09:24 | |
| *** gangil1 has quit IRC | 09:26 | |
| salv-orlando | good morning apuimedo | 09:32 |
| salv-orlando | I have a quick question for you, are you going to push the CNI plugin into kuryr-kubernetes? | 09:33 |
| salv-orlando | Also, have you or Irena already tried out an implementation? | 09:34 |
| salv-orlando | irenab: ^^ | 09:34 |
| irenab | salv-orlando: hi, how can I help? | 09:34 |
| salv-orlando | irenab: I was wondering if you had a shot at the CNI plugin for Kuryr... more specifically I am curious about whether you do complete logical network setup in the plugin (e.g.: POST /v2.0/port) or only VIF plugging | 09:36 |
| *** neiljerram has joined #openstack-kuryr | 09:36 | |
| irenab | We have is progress implementation of both parts for k8s integration, raven (the API watcher and neutron translator) and the CNI Driver that just does the binding. Due to some internal milestone we will start pushing the code to kuryr in about 2 weeks | 09:37 |
| irenab | the logical network is created by raven, currently k8s namepsace=> neutron network, pod triggerst neutron port creation | 09:38 |
| salv-orlando | irenab: cool. I was looking for confirmation of that, as "readiness" of logical network is an argument that was brought up in a discussion with some CNI folks | 09:38 |
| irenab | CNI gets the neutron port details in annotations and performs local binding | 09:38 |
| irenab | we didn’t want to make every worker node to call neutron API | 09:39 |
| apuimedo | sorry, I was in a meeting | 09:40 |
| apuimedo | salv-orlando: you mean getting network information as a standard cni thing? | 09:40 |
| salv-orlando | irenab: indeed that's very similar to what I did for OVN. But I think you're doing it better because you actually return from CNI when both logical network and VIF plugging are ready | 09:40 |
| apuimedo | Instead of the current "add to network" | 09:40 |
| irenab | salv-orlando: exactly | 09:41 |
| salv-orlando | apuimedo, it was more about synchronization of raven and CNI plugin | 09:41 |
| apuimedo | salv-orlando: yes. We do synchronize | 09:41 |
| irenab | CNI won’t bind till there is port in neutron | 09:41 |
| apuimedo | but I still thing that having to go from CNI to raven is wrong (although at the moment there is not a real alternative) | 09:41 |
| salv-orlando | apuimedo: that's good because there a sort of implict expectation that a container must be ready to send/receive data when CNI returns | 09:41 |
| apuimedo | salv-orlando: yes, we do block on that | 09:41 |
| apuimedo | we should also have a CNI return code that said. This failed beyond repair | 09:42 |
| apuimedo | so that the COE wouldn't try to re-schedule | 09:42 |
| salv-orlando | apuimedo: I am working on that | 09:42 |
| irenab | What I would be happy about if we could have the designtated Host at the port creatiion phase as with nova, but do not see how this be done without hacking the scheduler code or impose some order in the API watchers | 09:42 |
| salv-orlando | both with CNI and k8s support | 09:42 |
| apuimedo | salv-orlando: I'll invite you to something in Barcelona then! | 09:42 |
| apuimedo | heck, I'll get you a llangonissa even! | 09:43 |
| salv-orlando | apuimedo: I am always happy to travel to Barcelona - even if it will probably be me moaning about my 20s | 09:43 |
| apuimedo | :-) | 09:43 |
| apuimedo | irenab: there's some abandoned scheduler plugin point | 09:43 |
| apuimedo | but at this point I don't see it realistic that I can convince the k8s-sig-net that they are doing it all wrong :P | 09:43 |
| irenab | *irenab does not like abandoned stuff | 09:44 |
| apuimedo | and that the plugging point should be at the scheduler level with independent namespaces for the plugins | 09:44 |
| irenab | apuimedo: give them some time | 09:44 |
| apuimedo | well, I guess I should physically push for it instead of bitching about it like a moron | 09:45 |
| *** vikasc has quit IRC | 09:45 | |
| irenab | salv-orlando: the status report you work on is in k8s code base? | 09:45 |
| apuimedo | salv-orlando: would it be helpful for you if I started pushing the CNI driver alone? | 09:45 |
| apuimedo | (documenting what it expects) | 09:46 |
| salv-orlando | irenab: there's been probably some cross-wiring. I am now working on the "permanent failure" issue. | 09:48 |
| apuimedo | salv-orlando: on the CNI or on the k8s front? | 09:49 |
| salv-orlando | As for status reporting I'm looking into that too. So far I have added simply code to return the MAC address at least. | 09:49 |
| salv-orlando | apuimedo: technically on the kubernetes front. But there should be a standardized CNI error code for permanent failure | 09:49 |
| salv-orlando | as for status reporting, do you have in mind something similar to the what is returned by the kubernetes pod network status hook or something different? | 09:51 |
| irenab | salv-orlando: apuimedo : lunch time, will be back soon | 09:52 |
| salv-orlando | irenab: enjoy. and apologies for interrupting you abruptly. | 09:53 |
| *** tfukushima has quit IRC | 09:57 | |
| *** tfukushima has joined #openstack-kuryr | 09:57 | |
| *** vikasc has joined #openstack-kuryr | 09:57 | |
| *** salv-orl_ has joined #openstack-kuryr | 10:14 | |
| *** salv-orlando has quit IRC | 10:17 | |
| *** tfukushima has quit IRC | 10:31 | |
| apuimedo | salv-orl_: I was thinking of adding it to https://github.com/containernetworking/cni/blob/master/SPEC.md | 10:45 |
| salv-orl_ | apuimedo: adding what? the status hook or the error code? | 10:45 |
| apuimedo | meaningul standard error codes | 10:46 |
| salv-orl_ | yeah there is already a very exhaustive list of well-known error codes ;) | 10:46 |
| apuimedo | salv-orl_: the status should probably be in the result array https://github.com/containernetworking/cni/blob/master/SPEC.md#result | 10:46 |
| apuimedo | salv-orl_: where? | 10:46 |
| salv-orl_ | apuimedo: here https://github.com/containernetworking/cni/blob/master/SPEC.md#well-known-error-codes (but I was joking) | 10:47 |
| apuimedo | LOL | 10:47 |
| apuimedo | no wonder I missed it | 10:47 |
| apuimedo | it's long like the bible | 10:47 |
| apuimedo | The white hairs in my beard are rustling in disbelief that this is considered a production ready spec when it doesn't have any richness in error codes | 10:49 |
| apuimedo | :P | 10:49 |
| salv-orl_ | apuimedo: People considered Neutron "Essex" production ready | 10:50 |
| apuimedo | salv-orl_: I don't think they were consuming legal substances | 10:53 |
| *** yamamoto_ has quit IRC | 10:56 | |
| salv-orl_ | apuimedo: I was there. I was the dealer. | 10:58 |
| apuimedo | salv-orl_: they had the good stuff available then :-) | 10:58 |
| apuimedo | t quantum | 10:58 |
| apuimedo | I remember when I first heard about quantum | 10:58 |
| apuimedo | I googled it | 10:58 |
| apuimedo | and a youtube video of you showed up :-) | 10:58 |
| *** fawadkhaliq has quit IRC | 11:30 | |
| *** vikasc has quit IRC | 12:16 | |
| *** asti has joined #openstack-kuryr | 12:20 | |
| *** asti has quit IRC | 12:34 | |
| *** oanson has quit IRC | 12:35 | |
| *** banix has joined #openstack-kuryr | 12:39 | |
| *** salv-orl_ has quit IRC | 12:50 | |
| *** diogogmt has quit IRC | 13:17 | |
| *** irenab has quit IRC | 13:26 | |
| *** salv-orlando has joined #openstack-kuryr | 14:07 | |
| *** salv-orlando has quit IRC | 14:11 | |
| *** yamamoto has joined #openstack-kuryr | 14:12 | |
| *** asti has joined #openstack-kuryr | 14:17 | |
| *** fawadkhaliq has joined #openstack-kuryr | 14:29 | |
| *** diogogmt has joined #openstack-kuryr | 14:33 | |
| *** fawadkhaliq has quit IRC | 14:33 | |
| *** banix has quit IRC | 14:37 | |
| *** irenab has joined #openstack-kuryr | 14:41 | |
| *** yamamoto has quit IRC | 14:45 | |
| *** salv-orlando has joined #openstack-kuryr | 14:49 | |
| *** fawadkhaliq has joined #openstack-kuryr | 15:24 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Remove app.debug as we do not use it any more. https://review.openstack.org/318747 | 15:40 |
| *** yamamoto has joined #openstack-kuryr | 15:46 | |
| *** yamamoto has quit IRC | 15:57 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Add proper space in error log message. https://review.openstack.org/318754 | 15:59 |
| *** gangil has joined #openstack-kuryr | 16:03 | |
| *** fawadkhaliq has quit IRC | 16:11 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Fix run docker error on the network with host route https://review.openstack.org/318774 | 16:43 |
| *** fawadkhaliq has joined #openstack-kuryr | 16:46 | |
| *** shashank_hegde has joined #openstack-kuryr | 16:48 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Remove log_level as kuryr is using oslo_log now https://review.openstack.org/318777 | 16:51 |
| *** shashank_hegde has quit IRC | 16:54 | |
| *** gangil has quit IRC | 17:13 | |
| *** gangil has joined #openstack-kuryr | 17:13 | |
| *** shashank_hegde has joined #openstack-kuryr | 17:16 | |
| *** gangil has quit IRC | 17:18 | |
| *** banix has joined #openstack-kuryr | 17:25 | |
| *** devvesa has quit IRC | 17:27 | |
| *** fawadkhaliq has quit IRC | 17:33 | |
| *** gangil has joined #openstack-kuryr | 17:58 | |
| *** abhiraut has joined #openstack-kuryr | 18:00 | |
| *** abhiraut has quit IRC | 18:05 | |
| *** abhiraut has joined #openstack-kuryr | 18:10 | |
| *** apuimedo has quit IRC | 18:26 | |
| *** apuimedo has joined #openstack-kuryr | 18:26 | |
| *** oanson has joined #openstack-kuryr | 18:28 | |
| *** asti has quit IRC | 18:36 | |
| *** asti has joined #openstack-kuryr | 18:36 | |
| *** abhiraut has quit IRC | 18:52 | |
| *** abhiraut has joined #openstack-kuryr | 18:54 | |
| *** diogogmt has quit IRC | 18:55 | |
| *** salv-orlando has quit IRC | 19:01 | |
| *** abhiraut has quit IRC | 19:05 | |
| *** salv-orlando has joined #openstack-kuryr | 19:10 | |
| *** diogogmt has joined #openstack-kuryr | 19:12 | |
| *** diogogmt has quit IRC | 19:15 | |
| *** banix has quit IRC | 19:33 | |
| *** diogogmt has joined #openstack-kuryr | 19:37 | |
| *** diogogmt has quit IRC | 19:55 | |
| *** diogogmt has joined #openstack-kuryr | 20:04 | |
| *** diogogmt has quit IRC | 20:08 | |
| *** diogogmt has joined #openstack-kuryr | 20:09 | |
| *** banix has joined #openstack-kuryr | 20:13 | |
| *** oanson has quit IRC | 20:15 | |
| *** diogogmt has quit IRC | 20:25 | |
| *** diogogmt has joined #openstack-kuryr | 20:29 | |
| *** abhiraut has joined #openstack-kuryr | 20:30 | |
| *** abhiraut has quit IRC | 20:33 | |
| *** abhiraut has joined #openstack-kuryr | 20:35 | |
| *** yamamoto has joined #openstack-kuryr | 20:44 | |
| *** salv-orl_ has joined #openstack-kuryr | 20:48 | |
| *** salv-orlando has quit IRC | 20:51 | |
| *** abhiraut has quit IRC | 21:05 | |
| *** abhiraut has joined #openstack-kuryr | 21:17 | |
| *** abhiraut has quit IRC | 21:18 | |
| *** abhiraut has joined #openstack-kuryr | 21:20 | |
| *** asti has quit IRC | 21:38 | |
| *** diogogmt has quit IRC | 21:50 | |
| *** diogogmt has joined #openstack-kuryr | 22:19 | |
| *** abhiraut has quit IRC | 22:29 | |
| *** abhiraut has joined #openstack-kuryr | 22:40 | |
| *** banix has quit IRC | 22:52 | |
| *** diogogmt has quit IRC | 22:56 | |
| *** salv-orl_ has quit IRC | 23:12 | |
| *** shashank_hegde has quit IRC | 23:30 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!