| *** aojea has joined #openstack-kuryr | 00:45 | |
| *** aojea has quit IRC | 00:49 | |
| *** hongbin has joined #openstack-kuryr | 02:03 | |
| *** rh-jelabarre has quit IRC | 02:12 | |
| *** aojea has joined #openstack-kuryr | 06:08 | |
| *** aojea has quit IRC | 06:09 | |
| *** hongbin has quit IRC | 06:16 | |
| *** yboaron has joined #openstack-kuryr | 06:34 | |
| *** gcheresh_ has joined #openstack-kuryr | 06:39 | |
| *** ccamposr has joined #openstack-kuryr | 06:56 | |
| *** pcaruana has joined #openstack-kuryr | 07:12 | |
| openstackgerrit | Danil Golov proposed openstack/kuryr-kubernetes master: Use default subnet mapping if subnet is not specified https://review.openstack.org/629877 | 07:20 |
|---|---|---|
| *** maysams has joined #openstack-kuryr | 08:04 | |
| *** ccamposr has quit IRC | 08:27 | |
| *** ccamposr has joined #openstack-kuryr | 08:27 | |
| openstackgerrit | Genadi Chereshnya proposed openstack/kuryr-tempest-plugin master: Testing port_pool_min and port_pool_max https://review.openstack.org/620060 | 08:37 |
| openstackgerrit | Danil Golov proposed openstack/kuryr-kubernetes master: Use default subnet mapping if subnet is not specified https://review.openstack.org/629877 | 09:26 |
| *** yboaron_ has joined #openstack-kuryr | 09:32 | |
| *** yboaron has quit IRC | 09:33 | |
| *** yboaron_ has quit IRC | 09:52 | |
| *** yboaron_ has joined #openstack-kuryr | 09:57 | |
| *** yboaron_ has quit IRC | 10:15 | |
| *** rh-jelabarre has joined #openstack-kuryr | 12:07 | |
| *** gcheresh_ has quit IRC | 12:24 | |
| *** gcheresh_ has joined #openstack-kuryr | 12:51 | |
| *** aojea has joined #openstack-kuryr | 14:02 | |
| openstackgerrit | Maysa de Macedo Souza proposed openstack/kuryr-kubernetes master: Fix CRD update when NP has namespaceSelectors https://review.openstack.org/631230 | 14:28 |
| ltomasbo | ping dulek, I have a question for you | 14:34 |
| dulek | ltomasbo: pong! | 14:35 |
| ltomasbo | in one of the NP patch sets that got merged last week, I added a couple of fixes for a couple of problems I found, and I was expecting to move them out to a different commit before merging the patch | 14:35 |
| ltomasbo | dulek, ^^ however the patch got merged while I was away last week | 14:35 |
| ltomasbo | so, now we need to backport to rocky (and another one to queens) part of the code | 14:36 |
| ltomasbo | dulek, what is the best way of doing it? send it directly to rocky? | 14:36 |
| dulek | ltomasbo: I think it's best to merge follow up patch to master and then both squashed patches to rocky? | 14:37 |
| ltomasbo | dulek, let me explain myself a bit better | 14:38 |
| dulek | :) | 14:38 |
| ltomasbo | dulek, one patch was merged master which contains a couple of lines to fix one problem with the services | 14:38 |
| ltomasbo | dulek, now, I want to backport those couple of lines to rocky | 14:38 |
| ltomasbo | but they are part of an already merge patch | 14:38 |
| ltomasbo | dulek, and the NP patch cannot/shouldn't be backported to rocky | 14:39 |
| dulek | Oh, okay! | 14:39 |
| dulek | ltomasbo: Well, then just creating a new commit in stable/rocky sound totally fair. | 14:39 |
| ltomasbo | ok, I'll create a launchpad bug with the problem, and then manually add the information about it being fixed on the other merged patch for upstream | 14:40 |
| ltomasbo | thanks! | 14:40 |
| ltomasbo | dulek, actually, I found out the current master does not consider the NEUTRON_LBAAS_HAPROXY_PROVIDER case, which I guess it is ok as it is deprecated | 14:48 |
| dulek | ltomasbo: You mean for NP? | 14:48 |
| ltomasbo | dulek: yep | 14:49 |
| dulek | ltomasbo: I think it's fine that new feature doesn't work with deprecated stuff. | 14:49 |
| ltomasbo | well, it will not work for NP, but the changes to acommodate NP will leave the SG rules created for the lbaasv2 undeleted | 14:49 |
| ltomasbo | until the service is deleted | 14:49 |
| ltomasbo | dulek, ^^ | 14:50 |
| dulek | ltomasbo: I don't think I follow… Anyway nothing bad happens if you run with LBaaS v2 without NP, right? | 14:51 |
| ltomasbo | yes, it will on master | 14:51 |
| ltomasbo | if you modify the service definition, the security group rules added for the listeners will not be deleted | 14:51 |
| ltomasbo | dulek, ^^ | 14:53 |
| ltomasbo | dulek, so, I'm going to address that on master, and backport it to rocky adding the extra couple of lines that are already on master | 14:53 |
| ltomasbo | dulek, thougths? | 14:54 |
| dulek | ltomasbo: And that may happen with service_security_group_driver=default, and enabled_handlers not listing networkpolicy? | 14:54 |
| ltomasbo | dulek, yep | 14:54 |
| ltomasbo | to make it easier (master) https://github.com/openstack/kuryr-kubernetes/blob/master/kuryr_kubernetes/controller/drivers/lbaasv2.py#L390-L394 | 14:54 |
| dulek | ltomasbo: So I'd say - whatever easier of the two. Removing lbaasv2 or fixing this. :D | 14:55 |
| ltomasbo | rocky https://github.com/openstack/kuryr-kubernetes/blob/stable/rocky/kuryr_kubernetes/controller/drivers/lbaasv2.py#L248-L249 | 14:55 |
| ltomasbo | I need to add to the if at line 390 the next on master | 14:55 |
| ltomasbo | if loadbalancer.provider == const.NEUTRON_LBAAS_HAPROXY_PROVIDER: | 14:55 |
| ltomasbo | and them backport the 5-6 lines to rocky | 14:56 |
| *** dmellado has quit IRC | 14:56 | |
| *** aojea has quit IRC | 14:56 | |
| ltomasbo | removing lbaasv2 code sounds appealing... | 14:56 |
| ltomasbo | can I? | 14:57 |
| dulek | ltomasbo: Yeah, why not? Let me double check when we've deprecated it. | 15:39 |
| ltomasbo | dulek, I'm already removing it... | 15:39 |
| dulek | ltomasbo: Deprecation is on stable/rocky already, so I'd say - go for it! | 15:41 |
| ltomasbo | dulek, https://review.openstack.org/#/c/631261/ | 15:41 |
| dulek | ltomasbo: Was it that easy? I never knew how to start it. | 15:42 |
| ltomasbo | dulek, well, probably I've missed some parts... | 15:42 |
| dulek | ltomasbo: Oh, you need to remove lbaasv2 gate from .zuul.d. | 15:42 |
| ltomasbo | didn't we already do that? | 15:43 |
| ltomasbo | I'll update the patch with that | 15:43 |
| dulek | ltomasbo: And as next step we could move to use python-octaviaclient, but that can totally be done in a next patch. | 15:45 |
| openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Removing lbaasv2 related code https://review.openstack.org/631261 | 15:49 |
| dulek | ltomasbo: There's one more option to get removed, I've mentioned it in the comment. | 15:49 |
| dulek | ltomasbo: It's not used anywhere, so really easy. :D | 15:50 |
| ltomasbo | ohh, I just saw it... resubmitting... | 15:50 |
| ltomasbo | thanks! | 15:50 |
| openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Removing lbaasv2 related code https://review.openstack.org/631261 | 15:51 |
| *** gcheresh_ has quit IRC | 15:56 | |
| *** maysams has quit IRC | 16:03 | |
| *** pcaruana has quit IRC | 16:25 | |
| *** maysams has joined #openstack-kuryr | 16:30 | |
| *** ccamposr has quit IRC | 16:43 | |
| openstackgerrit | Michał Dulko proposed openstack/kuryr-kubernetes master: Remove way of running without kuryr-daemon https://review.openstack.org/631278 | 16:55 |
| dulek | ltomasbo: I guess it's cleanup day. ^ | 16:55 |
| ltomasbo | :) | 16:55 |
| *** mrostecki has quit IRC | 17:05 | |
| dulek | Heh, I'm surprised, but it seems to work on my env. | 17:13 |
| *** aperevalov has quit IRC | 17:20 | |
| *** mrostecki has joined #openstack-kuryr | 17:31 | |
| *** aojea has joined #openstack-kuryr | 17:48 | |
| *** aojea has quit IRC | 18:01 | |
| *** maysams has quit IRC | 18:38 | |
| *** dmellado has joined #openstack-kuryr | 19:22 | |
| *** threestrands has joined #openstack-kuryr | 22:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!