| *** hongbin has joined #openstack-kuryr | 04:25 | |
| *** janki has joined #openstack-kuryr | 04:25 | |
| *** hongbin has quit IRC | 04:53 | |
| *** gcheresh has joined #openstack-kuryr | 06:12 | |
| *** gcheresh has quit IRC | 06:31 | |
| *** gcheresh has joined #openstack-kuryr | 07:20 | |
| *** yboaron has quit IRC | 07:35 | |
| *** maysams has joined #openstack-kuryr | 07:35 | |
| *** gkadam has joined #openstack-kuryr | 08:01 | |
| *** maysams has quit IRC | 08:34 | |
| *** maysams has joined #openstack-kuryr | 08:37 | |
| *** yboaron has joined #openstack-kuryr | 08:48 | |
| *** yboaron_ has joined #openstack-kuryr | 08:53 | |
| *** yboaron has quit IRC | 08:55 | |
| *** ccamposr has joined #openstack-kuryr | 09:06 | |
| *** aperevalov has joined #openstack-kuryr | 09:28 | |
| *** mrostecki has joined #openstack-kuryr | 09:30 | |
| aperevalov | Hello dulek, yboaron_. I'm regarding issues mentioned in HA doc. Did you create a bug/task for it? | 09:43 |
|---|---|---|
| dulek | aperevalov: I doubt it's there, feel free to create it. | 09:44 |
| aperevalov | dulek, ok, thank you! | 09:45 |
| *** garyloug has joined #openstack-kuryr | 09:51 | |
| *** aperevalov has quit IRC | 10:26 | |
| *** garyloug_ has joined #openstack-kuryr | 11:21 | |
| *** garyloug has quit IRC | 11:25 | |
| *** garyloug_ has quit IRC | 11:41 | |
| *** garyloug has joined #openstack-kuryr | 11:43 | |
| *** rh-jelabarre has joined #openstack-kuryr | 12:47 | |
| *** janki has quit IRC | 12:50 | |
| *** jchhatbar has joined #openstack-kuryr | 12:50 | |
| *** gcheresh has quit IRC | 13:01 | |
| *** gcheresh_ has joined #openstack-kuryr | 13:01 | |
| *** garyloug_ has joined #openstack-kuryr | 13:03 | |
| *** garyloug__ has joined #openstack-kuryr | 13:04 | |
| *** garyloug has quit IRC | 13:07 | |
| *** garyloug_ has quit IRC | 13:08 | |
| *** gcheresh has joined #openstack-kuryr | 13:43 | |
| *** gcheresh_ has quit IRC | 13:44 | |
| *** yboaron_ has quit IRC | 14:03 | |
| *** yboaron_ has joined #openstack-kuryr | 14:03 | |
| *** aperevalov has joined #openstack-kuryr | 14:08 | |
| *** aperevalov has quit IRC | 14:16 | |
| *** aperevalov1 has joined #openstack-kuryr | 14:16 | |
| *** aperevalov1 has quit IRC | 14:18 | |
| *** aperevalov has joined #openstack-kuryr | 14:20 | |
| aperevalov | I didn't see any mails regarding meeting cancellation. | 14:24 |
| *** aperevalov has quit IRC | 14:45 | |
| *** jchhatbar has quit IRC | 14:49 | |
| *** aperevalov has joined #openstack-kuryr | 14:54 | |
| dulek | aperevalov: I've started it right now, not sure what will be the turnout. | 15:01 |
| *** garyloug_ has joined #openstack-kuryr | 15:01 | |
| *** garyloug__ has quit IRC | 15:05 | |
| *** snapiri has quit IRC | 15:08 | |
| *** yboaron_ has quit IRC | 15:21 | |
| *** yboaron_ has joined #openstack-kuryr | 15:21 | |
| maysams | dulek ping | 15:30 |
| dulek | maysams: What's up? | 15:30 |
| maysams | dulek: I've been trying to "upgrade" the LBaaSPortSpec to make it compatible with 1.1 version by using obj_make_compatible | 15:32 |
| dulek | What happened? ;) | 15:32 |
| maysams | However, it seems that this function is never executed and so the LBaaSPortSpec object cannot be loaded since it's missing one field | 15:33 |
| dulek | maysams: Oh yes, it should explicitly be called somewhere. | 15:33 |
| maysams | dulek: I wonder if it's related to the fact that the LBaaSPortSpec is a field of LBaaSServiceSpec | 15:33 |
| maysams | dulek: really? I thought it would be handled by oslo | 15:34 |
| dulek | maysams: https://github.com/openstack/cinder/blob/master/cinder/objects/base.py#L570-L579 | 15:35 |
| dulek | E.g. here it was backporting objects before serialization. | 15:35 |
| dulek | maysams: Let's maybe start with implementing explicit check on the methods that use portSpec field. With the obj_is_set(). Tomorrow I'll check if and how it can be implemented through OVO. | 15:39 |
| dulek | This is implementation detail at this point. :P | 15:39 |
| maysams | dulek, I don't think we should only handle with the obj_is_set() | 15:41 |
| maysams | dulek: the first thing we do on the LBaaS side is to check for port changes | 15:41 |
| maysams | dulek: and in this line https://review.openstack.org/#/c/635039/8/kuryr_kubernetes/controller/handlers/lbaas.py@136 | 15:41 |
| *** yboaron_ has quit IRC | 15:42 | |
| maysams | dulek: we load the object with the respectives fields regsitered | 15:42 |
| *** yboaron_ has joined #openstack-kuryr | 15:42 | |
| maysams | so, even though the targetPort field is registered is not present in the port spec | 15:43 |
| maysams | dulek: we could overcome this by not checking for the targetPort attribute... but I don't think it's ideal | 15:44 |
| maysams | dulek, what do you think? | 15:45 |
| *** gkadam has quit IRC | 15:49 | |
| dulek | maysams: Ah, interesting. | 15:51 |
| dulek | maysams: So we can just have an utility function that will do obj_from_primitive() and obj_make_compatible() and use that when deserializing? | 15:52 |
| maysams | dulek, yup, I think this could work | 15:54 |
| maysams | ;-) | 15:54 |
| maysams | dulek, thanks for the tip | 15:55 |
| dulek | maysams: I'm not 100% sure if obj_make_compatible() works fine with "forwardporting" | 15:55 |
| dulek | It was initially designed with backporting in mind. | 15:55 |
| maysams | dulek, yup.. and I did not find any examples that do some "upgrade" as I aim to | 15:58 |
| *** garyloug__ has joined #openstack-kuryr | 16:08 | |
| *** garyloug_ has quit IRC | 16:12 | |
| *** gcheresh has quit IRC | 16:20 | |
| *** yboaron_ has quit IRC | 16:28 | |
| *** ccamposr has quit IRC | 16:57 | |
| *** maysams has quit IRC | 17:08 | |
| *** garyloug__ has quit IRC | 18:37 | |
| *** openstackgerrit has quit IRC | 18:51 | |
| *** dmellado has quit IRC | 20:39 | |
| *** dmellado has joined #openstack-kuryr | 21:43 | |
| *** jistr_ has joined #openstack-kuryr | 23:39 | |
| *** jistr has quit IRC | 23:46 | |
| *** openstackgerrit has joined #openstack-kuryr | 23:48 | |
| openstackgerrit | Maysa de Macedo Souza proposed openstack/kuryr-kubernetes master: Fix SG rules on targetPort update https://review.openstack.org/635039 | 23:48 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!