Monday, 2019-03-18

*** gkadam_ has joined #openstack-kuryr02:11
*** gkadam_ has quit IRC02:17
*** irclogbot_1 has quit IRC02:23
*** hongbin has joined #openstack-kuryr03:12
*** hongbin has quit IRC03:21
*** janki has joined #openstack-kuryr04:05
openstackgerritMerged openstack/kuryr-kubernetes master: DevStack: Bump SG quotas for "policy" SG driver  https://review.openstack.org/64359206:52
*** pcaruana has joined #openstack-kuryr07:19
*** gcheresh has joined #openstack-kuryr07:58
*** maysams has joined #openstack-kuryr08:03
openstackgerritMerged openstack/kuryr-kubernetes master: Add option to deploy coredns  https://review.openstack.org/64207209:12
*** maysams has quit IRC10:04
*** alisanhaji has joined #openstack-kuryr10:06
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Add gate for Octavia provider OVN  https://review.openstack.org/60403610:13
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: WiP: Run K8s e2e network policy tests  https://review.openstack.org/64332710:13
*** maysams has joined #openstack-kuryr10:22
openstackgerritMerged openstack/kuryr-kubernetes master: Switch Octavia API calls to openstacksdk  https://review.openstack.org/63825811:25
*** kaiokmo has quit IRC11:37
*** rh-jelabarre has joined #openstack-kuryr11:57
*** pcaruana|afk| has joined #openstack-kuryr12:54
*** pcaruana has quit IRC12:56
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: WiP: Run K8s e2e network policy tests  https://review.openstack.org/64332713:25
aperevalov /join #openstack-meeting-413:43
*** pcaruana|afk| has quit IRC13:48
*** alisanhaji has quit IRC14:02
*** pcaruana has joined #openstack-kuryr14:33
*** alisanhaji has joined #openstack-kuryr14:35
dulekI guess this means we should run the meeting today. ;)14:58
*** janki has quit IRC15:26
*** gcheresh has quit IRC16:02
dulekmaysams: Seems like your patch isn't working: http://paste.openstack.org/show/747978/16:16
dulekmaysams: I'm unable to debug this deeper at the moment, but wanted to give feedback as soon as I see the result. ;)16:17
maysamsdulek: good to know.. I will look into it later today16:17
maysamsdulek: thanks16:17
maysamsdulek: Was this test passing before?16:18
dulekmaysams: No, no, it's about named ports.16:18
maysamsokay16:19
dulekmaysams: I'm trying to test it with https://review.openstack.org/#/c/643080/ applied.16:19
maysamsdulek, yup yup16:20
openstackgerritMerged openstack/kuryr-libnetwork master: Migrate legacy jobs to Ubuntu Bionic  https://review.openstack.org/64232116:22
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Add option to tag Octavia resources created by us  https://review.openstack.org/63848316:29
dulekmaysams: Hey, do we support such policies: https://github.com/kubernetes/kubernetes/blob/master/test/e2e/network/network_policy.go#L87-L111 ?16:33
dulekBecause this one failed with a pod that shouldn't be able to access the service, accessed it.]16:33
maysamsdulek, checking16:34
dulekmaysams: Hm, kuryr-controller was restarted 7 times…16:35
maysamsdulek: we do support this kind of policy16:36
maysamsdulek: so, Just to confirm it failed because a pod that was not supposed to be selected was being able to access the svc?16:39
*** ccamposr has joined #openstack-kuryr16:39
*** ccamposr__ has joined #openstack-kuryr16:39
maysamsdulek: Do you have the controller log?16:39
dulekmaysams: Yup. But no log, seems like pod got restarted. :(16:39
dulekmaysams: It's pretty possible that some etcd events got lost.16:39
dulekmaysams: I'll need to dig more here.16:40
maysamsdulek: right, later I will try to double check the matching selector code and see if we're missing some case16:41
maysamsdulek: let me know if you find some log16:41
maysamsdulek: thanks!16:41
dulekltomasbo: Sure!16:41
ltomasbo??16:42
ltomasbodulek, ^16:42
dulekltomasbo: Hm? I'm running those K8s e2e tests on my env.16:43
ltomasbodulek, I saw your ping: 'ltomasbo: Sure!'16:43
ltomasbodulek, not sure you were asking anything... or I asked something time ago... xD16:44
dulekltomasbo: Oh, I've tagged wrong person, sorry. :P16:44
ltomasbo:)16:44
dulekmaysams: One more data point - "should allow ingress access on one named port" passed while "should allow egress access on one named port" failed.16:55
maysamsdulek: and do you recall if this was passing before?16:56
dulekmaysams: Both were failing previously.16:56
maysamsdulek: okay16:56
dulekSo either your commit almost works, or it's my env failure. Doesn't hurt to double check if there's some significant difference in how ingress and egress are handled. :)16:57
maysamsdulek: it's possible that we're missing something else besides the text port16:57
maysamsdulek: sure sure. I will check16:57
dulekmaysams: I doubt we're missing anything besides text port, but obviously it's possible. :)16:58
*** ccamposr has quit IRC17:06
*** ccamposr__ has quit IRC17:06
maysamsdulek: yep, me too.. But lets see17:07
*** snapiri has quit IRC17:19
*** maysams has quit IRC17:31
*** maysams has joined #openstack-kuryr18:04
*** maysams has quit IRC18:27
*** gmann is now known as gmann_afk18:34
*** gmann_afk is now known as gmann18:56
*** mrostecki has quit IRC19:25
*** mrostecki has joined #openstack-kuryr19:30
*** irclogbot_1 has joined #openstack-kuryr20:12
*** irclogbot_1 has quit IRC20:25
*** irclogbot_1 has joined #openstack-kuryr20:27
*** alisanhaji has quit IRC20:38
*** pcaruana has quit IRC20:54
*** rh-jelabarre has quit IRC21:07
*** irclogbot_1 has quit IRC21:26
*** irclogbot_1 has joined #openstack-kuryr21:28

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!