| *** salv-orl_ has quit IRC | 00:03 | |
| *** salv-orlando has joined #openstack-kuryr | 00:09 | |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: IPAM: Add /Plugin.Activate, /IpamDriver.RequestPool and placeholders https://review.openstack.org/252796 | 00:28 |
|---|---|---|
| *** dingboopt has quit IRC | 00:32 | |
| *** banix has joined #openstack-kuryr | 00:37 | |
| *** vikasc__ has quit IRC | 00:37 | |
| *** vikasc__ has joined #openstack-kuryr | 00:41 | |
| *** salv-orlando has quit IRC | 01:25 | |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: IPAM: Implement /IpamDriver.RequestAddress https://review.openstack.org/252835 | 01:36 |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: IPAM: Implement /IpamDriver.RequestAddress https://review.openstack.org/252835 | 01:40 |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: IPAM: Implement /IpamDriver.ReleasePool https://review.openstack.org/252848 | 01:49 |
| *** dingboopt has joined #openstack-kuryr | 01:52 | |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: IPAM: Implement /IpamDriver.ReleaseAddress https://review.openstack.org/252860 | 01:56 |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: Create subnets in /NetworkDriver.CreateNetwork using IPv4Data and IPv6Data https://review.openstack.org/253241 | 01:58 |
| openstackgerrit | Taku Fukushima proposed openstack/kuryr: Add trailing commans to follow the coding standard https://review.openstack.org/254493 | 02:25 |
| *** tfukushima has joined #openstack-kuryr | 02:26 | |
| openstackgerrit | vikas choudhary proposed openstack/kuryr: /NetworkDriver.CreateEndpoint cleanup https://review.openstack.org/253323 | 02:54 |
| *** vikasc__ has quit IRC | 02:59 | |
| *** vikasc__ has joined #openstack-kuryr | 03:01 | |
| openstackgerrit | Baohua Yang proposed openstack/kuryr: Using direct kwargs instead of a dict creation/destroy https://review.openstack.org/254071 | 03:15 |
| *** vikasc__ has quit IRC | 03:17 | |
| *** salv-orlando has joined #openstack-kuryr | 03:29 | |
| banix | tfukushima: hi Taku | 03:31 |
| tfukushima | banix: Hi banix, | 03:31 |
| banix | had a question for you | 03:31 |
| banix | regarding pyroute2, do you need to be rrot to use the library? | 03:31 |
| tfukushima | banix: Oh, sorry I forgot to reply to your email. | 03:32 |
| banix | I can have it working only as root | 03:32 |
| tfukushima | banix: Yes, pyroute2 uses rtnetlink, which opens up the special socket against the kernel. So it requires the root privilege. | 03:33 |
| tfukushima | Same as iproute2. | 03:33 |
| *** salv-orlando has quit IRC | 03:34 | |
| tfukushima | I mean for creating resources. Listing and showing the details of the resources can run without root. | 03:34 |
| banix | so then 2 questions: so we need to run kutyt as root? | 03:34 |
| banix | kuryr that is | 03:34 |
| banix | the 2nd question a minor one: we dont need the sudo in the ovs binary for example | 03:36 |
| tfukushima | Kuryr needs to be run by the root user, especially the pyroute2 part. | 03:37 |
| tfukushima | Honestly I'm not familiar with the way to separate the user privilege in the same Python process. | 03:38 |
| tfukushima | We need the root privilege only for pyroute2. | 03:38 |
| tfukushima | For creating a veth pair. That's the only place we need the root. | 03:38 |
| banix | I recall similar discussion on neutron/nova side wrt using librarires such as this; not sure what that led to | 03:38 |
| banix | so the other approach would be using sudo, etc which is far from ideal…. | 03:39 |
| tfukushima | Yeah, if we use the ip command, we could just add the "sudo" prefix. | 03:40 |
| banix | ok so we can/should discuss this further; just wanted to make sure I am not doing something incorrectly | 03:41 |
| tfukushima | We could use the ip command but it would be the string based command construction rather than the pure Python program. | 03:42 |
| banix | yes, I understand. | 03:42 |
| banix | the other thing I wanted to discuss tfukushima is that for different bindings (plugs/unplugs) we may have the same vif type | 03:44 |
| banix | in particular for ovs and ovs_hybrid | 03:44 |
| banix | I have been thinking about how to deal with this | 03:44 |
| banix | one approach could be delegating both ovs and ovs_hybrid to the same binary and have the binary figure out which plugging it needs to do | 03:45 |
| tfukushima | You mean you might want to execute the different binding binary even if binding:vif_type is set to the single value, say "ovs"? | 03:46 |
| banix | yes or somehow a single binary has to distinguish between different types | 03:47 |
| banix | for ovs and ovs_hybrid the plugging operations are different but both have the vif type as ovs | 03:47 |
| tfukushima | That'd be a little bit tough. We would need the additional context to do that. | 03:48 |
| tfukushima | binding:vif_type can't be set to ovs_hybrid? | 03:49 |
| banix | yeah so in nova you have a config option for it as far as i recall. we may need to have an additional config option. | 03:49 |
| banix | not really. you see there is the binding and then plugging the port. i think we have been mixing these a bit. even though the binding is the same the plugging is different | 03:51 |
| banix | will work on a solution and propose and see if it makes sense to you and others | 03:51 |
| tfukushima | Hmm, that's a bit tricky. I guess we need to add the logic for the special case. | 03:52 |
| banix | actually there may be other plugins with ovs type but different plugging but the prominent case is ovs and ovs_hybrid | 03:52 |
| tfukushima | :-/ | 03:53 |
| tfukushima | The radical way is to overwrite "ovs" script by their own but I'm pretty much sure we don't want it. | 03:54 |
| tfukushima | So please submit a patch when you come to find the workaround. We'd appreciate it. | 03:54 |
| banix | yeah something less bad :) | 03:56 |
| banix | thanks. talk to you later. | 03:57 |
| tfukushima | banix: Have a good night. | 03:57 |
| openstackgerrit | Taku Fukushima proposed openstack/kuryr: [WIP] Add spec of blueprint existing-neutron-network https://review.openstack.org/254005 | 04:05 |
| *** tfukushima has quit IRC | 04:11 | |
| *** tfukushima has joined #openstack-kuryr | 04:14 | |
| *** tfukushima has quit IRC | 04:20 | |
| *** banix has quit IRC | 04:39 | |
| *** fawadkhaliq has joined #openstack-kuryr | 05:12 | |
| *** tfukushima has joined #openstack-kuryr | 05:20 | |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: Seperate unit tests from fullstack tests https://review.openstack.org/254565 | 06:22 |
| *** diga has joined #openstack-kuryr | 06:29 | |
| gsagie | Will be good if we can merge https://review.openstack.org/254565 fast, it might effect other patches as i moved the unit tests to a different location | 06:30 |
| *** tfukushima has quit IRC | 06:31 | |
| *** tfukushima has joined #openstack-kuryr | 06:31 | |
| irenab | gsagie: checking | 06:32 |
| gsagie | tfukushima: https://review.openstack.org/254565 , best to merge this fast as i moved unit tests location | 06:32 |
| gsagie | irenab: thanks | 06:34 |
| irenab | gsagie: done | 06:38 |
| tfukushima | gasagie: Let me take a look at it. | 06:39 |
| tfukushima | gsagie: So in this patch the unit tests are just moved, right? | 06:40 |
| gsagie | tfukushima: yes | 06:41 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: Seperate unit tests from fullstack tests https://review.openstack.org/254565 | 06:46 |
| gsagie | tfukushima: just want to seperate the tests that run in the python27 gate job (same as you will write 'tox') and the tests that require devstack with Kuryr deployed which we will run in the gate | 06:48 |
| gsagie | python27 gate is for the unit tests | 06:48 |
| tfukushima | Ok. | 06:48 |
| *** salv-orlando has joined #openstack-kuryr | 06:49 | |
| gsagie | irenab: once this is merged going to use these comments :) | 06:49 |
| gsagie | i can take them off if its that of a problem, but since i am multiplexing with too many patches i prefer not to forget stuff ;) but as you see fit | 06:50 |
| irenab | gsagie: I will try to be less pedantic. Will try hard :-) | 06:50 |
| openstackgerrit | Baohua Yang proposed openstack/kuryr: Using direct kwargs instead of a dict creation/destroy https://review.openstack.org/254071 | 06:50 |
| gsagie | irenab: np ;) be yourself! | 06:50 |
| irenab | gsagie: having some REVISIT/TODO will be great | 06:51 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: Seperate unit tests from fullstack tests https://review.openstack.org/254565 | 06:52 |
| gsagie | irenab: nvm, just removed it :) | 06:52 |
| *** salv-orlando has quit IRC | 06:54 | |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: Separate unit tests from fullstack tests https://review.openstack.org/254565 | 07:08 |
| tfukushima | gsagie: I need to attend a meeting, so I just left brief comments. | 07:13 |
| *** fawadkhaliq has quit IRC | 07:41 | |
| *** fawadkhaliq has joined #openstack-kuryr | 07:42 | |
| tfukushima | gsagie: So you need your patch merged now? | 08:40 |
| openstackgerrit | kexiaodong proposed openstack/kuryr: Replace full-width quotes with half-width quotes https://review.openstack.org/254614 | 08:43 |
| gsagie | tfukushima: yes, will be good to make it now | 08:47 |
| openstackgerrit | Merged openstack/kuryr: Using direct kwargs instead of a dict creation/destroy https://review.openstack.org/254071 | 08:47 |
| gsagie | tfukushima: will do the fixes after, as all the patches that work on the unit tests and any new one will need to rebase | 08:47 |
| tfukushima | gsagie: This might cause some conflicts, for instance Vikas' IPAM patches. So I'm a little bit hesitated. | 08:48 |
| gsagie | tfukushima: yes, but its better we do it now rather then wait | 08:48 |
| gsagie | it will get worst | 08:48 |
| tfukushima | Ok. Finger crossed. | 08:49 |
| gsagie | i will rebase the patches for him | 08:49 |
| gsagie | is he online? | 08:49 |
| tfukushima | I don't think so. | 08:50 |
| openstackgerrit | Merged openstack/kuryr: Separate unit tests from fullstack tests https://review.openstack.org/254565 | 08:52 |
| tfukushima | gsagie: Merged and almost all patches got -1 by the change on the test. :-p | 08:56 |
| gsagie | yes need to move the test dir | 08:57 |
| gsagie | will fix it now | 08:57 |
| tfukushima | Or we can just announce the change on the mailing list. | 08:59 |
| gsagie | tfukushima: sec, let me see how much effort it is, i might be able to resolve it quickly | 09:01 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Add /Plugin.Activate, /IpamDriver.RequestPool and placeholders https://review.openstack.org/252796 | 09:02 |
| tfukushima | It was resolved automatically against my patch when I did rebase on top of the current master. | 09:02 |
| openstackgerrit | Taku Fukushima proposed openstack/kuryr: Add trailing commans to follow the coding standard https://review.openstack.org/254493 | 09:02 |
| gsagie | yeah, but was hoping i only need to do the merge once, but seems i need to do it for every patch | 09:04 |
| gsagie | ohh well, will do it | 09:04 |
| openstackgerrit | Merged openstack/kuryr: Replace full-width quotes with half-width quotes https://review.openstack.org/254614 | 09:07 |
| gsagie | ohh god :| | 09:08 |
| tfukushima | kexiaodong: We'd appreciate if you could add some reviewers next time you submit the patch. Refer to https://review.openstack.org/#/c/254565/ for instance. | 09:08 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Implement /IpamDriver.RequestAddress https://review.openstack.org/252835 | 09:18 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Add /Plugin.Activate, /IpamDriver.RequestPool and placeholders https://review.openstack.org/252796 | 09:18 |
| kexiaodong | ok, I will add some reviewer next time | 09:19 |
| tfukushima | kexiaodong: Thanks! | 09:20 |
| gsagie | tfukushima: doing some rebases on Vikas patches, please re-check me if you can :) | 09:20 |
| gsagie | hope i am not doing any mistakes | 09:20 |
| tfukushima | gsagie: I'm too lazy and I'd just announce the change and let people rebase their patches if I were you. :-p | 09:21 |
| *** diga has quit IRC | 09:22 | |
| gsagie | tfukushima: okie | 09:23 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Add /Plugin.Activate, /IpamDriver.RequestPool and placeholders https://review.openstack.org/252796 | 09:29 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Implement /IpamDriver.RequestAddress https://review.openstack.org/252835 | 09:34 |
| gsagie | ok, nvm i just did it wrong the first time | 09:34 |
| gsagie | bah :) | 09:34 |
| gsagie | oops | 09:35 |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Implement /IpamDriver.RequestAddress https://review.openstack.org/252835 | 09:43 |
| openstackgerrit | Berezovsky Irena proposed openstack/kuryr: Add trailing commas to follow the coding standard https://review.openstack.org/254493 | 10:09 |
| *** tfukushima has quit IRC | 10:10 | |
| *** tfukushima has joined #openstack-kuryr | 10:11 | |
| *** tfukushima has quit IRC | 10:14 | |
| *** fawadkhaliq has quit IRC | 10:21 | |
| *** fawadkhaliq has joined #openstack-kuryr | 10:21 | |
| *** salv-orlando has joined #openstack-kuryr | 10:25 | |
| *** salv-orlando has quit IRC | 10:30 | |
| *** openstackgerrit has quit IRC | 11:02 | |
| *** openstackgerrit has joined #openstack-kuryr | 11:02 | |
| openstackgerrit | Gal Sagie proposed openstack/kuryr: IPAM: Add /Plugin.Activate, /IpamDriver.RequestPool and placeholders https://review.openstack.org/252796 | 11:05 |
| *** salv-orlando has joined #openstack-kuryr | 11:06 | |
| *** salv-orlando has quit IRC | 11:14 | |
| gsagie | we need vikas :( | 11:50 |
| gsagie | apuimedo: here? | 11:50 |
| *** fawadkhaliq has quit IRC | 12:14 | |
| *** fawadkhaliq has joined #openstack-kuryr | 12:37 | |
| *** fawadkhaliq has quit IRC | 12:40 | |
| *** fawadkhaliq has joined #openstack-kuryr | 12:40 | |
| *** diogogmt has quit IRC | 13:58 | |
| *** diogogmt has joined #openstack-kuryr | 13:58 | |
| *** gsagie_ has joined #openstack-kuryr | 14:01 | |
| *** gsagie_ has quit IRC | 14:12 | |
| *** banix has joined #openstack-kuryr | 14:16 | |
| *** diogogmt has quit IRC | 14:25 | |
| *** fawadkhaliq has quit IRC | 14:45 | |
| *** fawadkhaliq has joined #openstack-kuryr | 14:46 | |
| *** vikasc__ has joined #openstack-kuryr | 14:50 | |
| *** gsagie_ has joined #openstack-kuryr | 15:02 | |
| *** fawadkhaliq has quit IRC | 15:16 | |
| banix | gsagie: did you get your docker problem issue resolved? I think we saw the same issue at some point after change of kernel version; some mismatch between certain version of docker and the kernel | 15:18 |
| banix | gsagie_: ^^^ | 15:19 |
| gsagie_ | banix: haven't tried yet, will update you tomorrow, haven't updated kernel, just re installed docker | 15:21 |
| gsagie_ | from get.docker.com script | 15:21 |
| banix | so you may need to upgrade the kernel. probably. | 15:21 |
| gsagie_ | They exit with code 127 just after they are started? | 15:22 |
| gsagie_ | banix: which kernel version solved it? | 15:23 |
| banix | can look at docker log, or start docker in debug; had this issue a while back… | 15:23 |
| banix | this is a while back so not the latest docker but had to go to 3.16. where running trusty. may be different for you. | 15:23 |
| gsagie_ | well, it did work for me like 1-2 weeks ago | 15:23 |
| banix | but you updated the docker you said? | 15:24 |
| gsagie_ | banix: ok thanks will check tommorow | 15:24 |
| banix | sure | 15:24 |
| gsagie_ | banix: yeah | 15:24 |
| *** vikasc__ has quit IRC | 15:25 | |
| *** gsagie_ has quit IRC | 15:31 | |
| *** diogogmt has joined #openstack-kuryr | 15:36 | |
| *** salv-orlando has joined #openstack-kuryr | 16:03 | |
| *** salv-orlando has quit IRC | 16:03 | |
| *** salv-orlando has joined #openstack-kuryr | 16:03 | |
| *** fawadkhaliq has joined #openstack-kuryr | 16:12 | |
| *** salv-orl_ has joined #openstack-kuryr | 16:26 | |
| *** salv-orlando has quit IRC | 16:29 | |
| *** openstackgerrit has quit IRC | 16:47 | |
| *** openstackgerrit has joined #openstack-kuryr | 16:47 | |
| *** salv-orl_ has quit IRC | 17:16 | |
| *** yasui_ has quit IRC | 17:35 | |
| *** yasui_ has joined #openstack-kuryr | 17:41 | |
| *** salv-orlando has joined #openstack-kuryr | 18:17 | |
| *** salv-orlando has quit IRC | 18:21 | |
| *** openstackgerrit has quit IRC | 19:02 | |
| *** openstackgerrit has joined #openstack-kuryr | 19:02 | |
| *** salv-orlando has joined #openstack-kuryr | 19:04 | |
| *** salv-orlando has quit IRC | 19:11 | |
| *** mestery has quit IRC | 19:28 | |
| *** mestery has joined #openstack-kuryr | 19:28 | |
| *** irenab has quit IRC | 19:42 | |
| *** fawadkhaliq has quit IRC | 20:55 | |
| *** salv-orlando has joined #openstack-kuryr | 21:50 | |
| *** salv-orlando has quit IRC | 21:55 | |
| *** salv-orlando has joined #openstack-kuryr | 22:05 | |
| *** banix has quit IRC | 22:40 | |
| *** salv-orlando has quit IRC | 22:53 | |
| *** salv-orlando has joined #openstack-kuryr | 22:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!