| *** hongbin has joined #openstack-kuryr | 00:37 | |
| *** dulek has quit IRC | 01:21 | |
| *** dulek has joined #openstack-kuryr | 01:34 | |
| *** gkadam has joined #openstack-kuryr | 02:48 | |
| *** gkadam has quit IRC | 02:50 | |
| *** hongbin has quit IRC | 02:57 | |
| *** hongbin has joined #openstack-kuryr | 02:58 | |
| *** hongbin has quit IRC | 03:12 | |
| *** hongbin has joined #openstack-kuryr | 03:14 | |
| *** premsankar has quit IRC | 03:30 | |
| *** hongbin has quit IRC | 03:31 | |
| *** spsurya has joined #openstack-kuryr | 03:31 | |
| *** premsankar has joined #openstack-kuryr | 05:32 | |
| *** pcaruana has joined #openstack-kuryr | 05:35 | |
| *** openstackgerrit has joined #openstack-kuryr | 05:53 | |
| openstackgerrit | zhufl proposed openstack/kuryr-kubernetes master: Add missing ws separator between words https://review.openstack.org/649224 | 05:53 |
|---|---|---|
| *** ccamposr has joined #openstack-kuryr | 06:09 | |
| *** openstackgerrit has quit IRC | 06:09 | |
| *** aperevalov has joined #openstack-kuryr | 06:50 | |
| *** kmadac4 is now known as kmadac | 06:51 | |
| *** aperevalov has quit IRC | 06:58 | |
| *** gcheresh has joined #openstack-kuryr | 07:02 | |
| *** maysams has joined #openstack-kuryr | 07:27 | |
| *** aperevalov has joined #openstack-kuryr | 08:08 | |
| *** gkadam has joined #openstack-kuryr | 08:10 | |
| aperevalov | hi, Zuul gave me -1 here https://review.openstack.org/#/c/642491/, not so clear why. Is it due to problem on the gates, you discussed yesterday? | 08:53 |
| aperevalov | also danil submitted blueprint: https://blueprints.launchpad.net/kuryr-kubernetes/+spec/dpdk-applications-in-containers | 08:54 |
| aperevalov | Please take a look, but from my point of view it's just a part of the problem. | 08:54 |
| dulek | aperevalov: Yes, gates are still not fixed, I just successfully reproduced the issue locally and starting to debug it. | 09:52 |
| *** zigo has joined #openstack-kuryr | 09:59 | |
| dulek | dmellado: Hm, interestingly the gate issue is because amphora cannot reach 10.0.0.67. | 10:58 |
| dulek | ltomasbo: ^ | 10:58 |
| dmellado | how strange that octavia is the issue | 10:59 |
| dmellado | I'm fighting against another bug as well, in ooo deployment and it's octavia as well | 10:59 |
| dmellado | dulek: I wasn't able to reproduce that locally, how about you? | 10:59 |
| dulek | dmellado: Me neither, only on completely fresh installation today. | 11:00 |
| dulek | dmellado: At this point I'm 60% sure that it's because we had old amphora images cached. | 11:00 |
| dmellado | hmm probably | 11:00 |
| dmellado | what network is that 10.0.0.67 ? k8s api? | 11:01 |
| dulek | dmellado: Yep. I think the root cause is that amphora has the service network attached but it doesn't show up inside… | 11:01 |
| dmellado | how awesome... | 11:02 |
| dmellado | now let's check what could've changed from within octavia to modify this... | 11:02 |
| dulek | dmellado: I suspect it's not really Octavia. It might be a bug in Nova that prevents second network from being attached. | 11:02 |
| dulek | dmellado: Or it might be something in amphora. | 11:03 |
| dulek | dmellado: How often is the amphora image built? | 11:03 |
| dulek | dmellado: Because if it's rare, then the window is much bigger. | 11:03 |
| dmellado | hmm I assume that there'd be some kind of post in zuul to do the tarball | 11:05 |
| dmellado | let's see the date from the current ones | 11:05 |
| dulek | dmellado: Okay, screw this, now I think it's amphora - check out 23a411413fbad6e96ed6b3c3ec3876cba3d8f4fd. | 11:06 |
| *** janki has joined #openstack-kuryr | 11:14 | |
| ltomasbo | dulek, thanks for the info! | 11:32 |
| ltomasbo | dulek, do they octavia folks know about it? I assume they may have tempest test for that, right? | 11:33 |
| ltomasbo | dulek, nevermind, I see you are talking with Carlos already... | 11:34 |
| dulek | ltomasbo: Yup, we're looking at this. | 11:35 |
| ltomasbo | thanks! | 11:35 |
| *** openstackgerrit has joined #openstack-kuryr | 11:38 | |
| openstackgerrit | Michał Dulko proposed openstack/kuryr-kubernetes master: DNM: Testing the gates https://review.openstack.org/649285 | 11:38 |
| dulek | Let's see if revert of that commit helps. ^ | 11:38 |
| *** celebdor has quit IRC | 11:45 | |
| *** celebdor has joined #openstack-kuryr | 12:23 | |
| *** rh-jelabarre has joined #openstack-kuryr | 13:03 | |
| dulek | Eh, it does not help. :( | 13:06 |
| openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Add support for policyTypes at Network Policies https://review.openstack.org/648699 | 13:11 |
| dmellado | dulek: oh, amphora! | 14:28 |
| dmellado | damn | 14:28 |
| dmellado | dulek: I've been fighting another wars as well... | 14:28 |
| dulek | dmellado: I tested without that commit and it haven't helped, so a this point I don't know nothing again. :P | 14:29 |
| dmellado | lol | 14:29 |
| dmellado | xD | 14:30 |
| dmellado | I was reading your log with carlos and so far besides commuting it doesn't seem like it's useful at all also xD | 14:30 |
| dmellado | dulek: the sad thing is that I thought that commit could be culprit as well.... | 14:31 |
| dmellado | did you check out with any 'old' amphora? | 14:31 |
| dmellado | I'm not sure if there could be any older tar.gz. around | 14:31 |
| dulek | dmellado: I had an old one downloaded on my local env. | 14:31 |
| dmellado | how about cheking using that one in a newer env? | 14:32 |
| dulek | dmellado: No way, scp'ing it would take 1.5 hour. :P | 14:32 |
| dmellado | xD | 14:32 |
| dmellado | just use a vm xD | 14:32 |
| dmellado | oh, I see, local local | 14:32 |
| dmellado | like behind your isp xD | 14:32 |
| dulek | Yup. | 14:32 |
| dmellado | hmmm, I guess we'll have to freeze a vm and check up there | 14:36 |
| dmellado | let's grab carlos as well | 14:36 |
| dmellado | celebdor: won't you like to play octavia again? XD | 14:36 |
| dulek | dmellado: I have a freezed VM but I can't access it due to that IPv6 issues. Anyway I have that reproduced on another cloud. | 14:40 |
| dulek | dmellado: I can Carlos access, I guess. | 14:40 |
| dmellado | dulek: so you couldn't access it due to ipv6? | 14:41 |
| dmellado | yes, let carlos has access there and see | 14:41 |
| dulek | dmellado: Nah, but whatever, I see the same on my own VM, I don't need to check the gate one now. | 14:41 |
| *** gcheresh has quit IRC | 14:42 | |
| *** janki has quit IRC | 14:56 | |
| *** rh-jelabarre has quit IRC | 15:08 | |
| *** ccamposr has quit IRC | 15:33 | |
| *** maysams has quit IRC | 15:46 | |
| *** maysams has joined #openstack-kuryr | 16:17 | |
| *** rh-jelabarre has joined #openstack-kuryr | 16:20 | |
| *** celebdor has quit IRC | 16:31 | |
| *** gkadam has quit IRC | 17:02 | |
| *** rh-jelabarre has quit IRC | 17:25 | |
| *** rh-jelabarre has joined #openstack-kuryr | 17:29 | |
| openstackgerrit | Michał Dulko proposed openstack/kuryr-kubernetes master: DevStack: Fix Amphora->kubelet port connectivity https://review.openstack.org/649413 | 17:59 |
| dulek | Okay, hopefully this solves the issue. ^ | 17:59 |
| *** spsurya has quit IRC | 18:32 | |
| *** gmann is now known as gmann_afk | 19:04 | |
| *** rh-jelabarre has quit IRC | 19:26 | |
| *** rh-jelabarre has joined #openstack-kuryr | 19:30 | |
| *** celebdor has joined #openstack-kuryr | 19:30 | |
| *** pcaruana has quit IRC | 20:30 | |
| openstackgerrit | Michał Dulko proposed openstack/kuryr-kubernetes master: DevStack: Fix Amphora->kubelet port connectivity https://review.openstack.org/649413 | 20:40 |
| *** maysams has quit IRC | 21:27 | |
| openstackgerrit | Michał Dulko proposed openstack/kuryr-kubernetes master: DevStack: Fix Amphora->kubelet port connectivity https://review.openstack.org/649413 | 21:59 |
| *** gmann_afk is now known as gmann | 23:11 | |
| *** celebdor has quit IRC | 23:22 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!