Tuesday, 2018-04-17

« Monday, 2018-04-16 Index Wednesday, 2018-04-18 »
*** salv-orlando has joined #openstack-kuryr00:00
*** salv-orlando has quit IRC00:05
*** yamamoto has joined #openstack-kuryr00:06
*** deepika has quit IRC00:09
*** celebdor has quit IRC00:09
*** mrostecki[m] has quit IRC00:09
*** yamamoto has quit IRC00:11
*** kiennt26 has joined #openstack-kuryr00:35
*** yamamoto has joined #openstack-kuryr01:07
*** hongbin_ has joined #openstack-kuryr01:12
*** yamamoto has quit IRC01:13
*** janki has joined #openstack-kuryr01:42
*** pliu has quit IRC02:04
*** pliu has joined #openstack-kuryr02:04
*** yamamoto has joined #openstack-kuryr02:09
*** yamamoto has quit IRC02:15
*** kiennt26 has quit IRC02:18
*** maysamacedos has quit IRC02:35
*** maysamacedos has joined #openstack-kuryr02:35
*** kiennt26 has joined #openstack-kuryr02:47
*** kiennt26 has quit IRC02:54
*** jchhatbar has joined #openstack-kuryr03:00
*** kiennt26 has joined #openstack-kuryr03:02
*** janki has quit IRC03:03
*** jchhatbar is now known as janki03:07
*** kiennt26 has quit IRC03:08
*** yamamoto has joined #openstack-kuryr03:11
*** kiennt26 has joined #openstack-kuryr03:11
*** kiennt26 has quit IRC03:13
*** kiennt26 has joined #openstack-kuryr03:16
*** yamamoto has quit IRC03:16
*** maysamacedos has quit IRC03:17
*** kiennt26 has quit IRC03:21
*** kiennt26 has joined #openstack-kuryr03:36
*** yamamoto has joined #openstack-kuryr04:13
*** kiennt26 has quit IRC04:15
*** jchhatbar has joined #openstack-kuryr04:18
*** yamamoto has quit IRC04:18
*** janki has quit IRC04:20
*** hongbin_ has quit IRC04:21
*** gcheresh_ has joined #openstack-kuryr04:43
*** salv-orlando has joined #openstack-kuryr04:55
*** mrostecki[m] has joined #openstack-kuryr05:33
*** celebdor has joined #openstack-kuryr05:42
*** deepika has joined #openstack-kuryr05:46
*** jchhatbar is now known as janki06:12
*** yamamoto has joined #openstack-kuryr06:16
*** pcaruana has joined #openstack-kuryr06:16
*** yamamoto has quit IRC06:22
*** jchhatbar has joined #openstack-kuryr06:50
*** yboaron has joined #openstack-kuryr06:50
*** janki has quit IRC06:53
*** yamamoto has joined #openstack-kuryr07:10
*** yamamoto has quit IRC07:14
*** yamamoto has joined #openstack-kuryr07:14
*** salv-orlando has quit IRC07:28
*** salv-orlando has joined #openstack-kuryr07:28
*** salv-orlando has quit IRC07:32
*** yamamoto_ has joined #openstack-kuryr07:39
*** yamamoto has quit IRC07:42
*** salv-orlando has joined #openstack-kuryr07:45
*** celebdor1 has joined #openstack-kuryr07:46
*** vkuramshin has joined #openstack-kuryr07:54
*** garyloug has joined #openstack-kuryr07:55
*** yboaron_ has joined #openstack-kuryr08:26
*** openstackgerrit has joined #openstack-kuryr08:27
openstackgerritMerged openstack/kuryr-kubernetes master: Dockerfiles: COPY code after installing packages  https://review.openstack.org/54505908:27
*** yboaron has quit IRC08:29
*** yamamoto_ has quit IRC08:36
*** yamamoto has joined #openstack-kuryr08:37
*** yboaron_ has quit IRC08:52
openstackgerritMerged openstack/kuryr-kubernetes master: Modify maintainers of Dockerfiles  https://review.openstack.org/56120308:52
*** maysamacedos has joined #openstack-kuryr08:55
openstackgerritMerged openstack/kuryr-kubernetes master: Vagrant: Document the increased Octavia reqs  https://review.openstack.org/55502309:06
*** yboaron_ has joined #openstack-kuryr10:59
*** atoth has joined #openstack-kuryr11:27
*** maysamacedos has quit IRC11:57
yboaron_dulek, ping12:21
*** yamamoto has quit IRC12:25
*** yamamoto has joined #openstack-kuryr12:25
*** yamamoto has quit IRC12:26
*** yamamoto has joined #openstack-kuryr12:26
dulekyboaron_: hi!12:31
yboaron_dulek, Hi! question regarding release notes12:31
dulekyup?12:31
yboaron_dulek, of  https://review.openstack.org/#/c/555735/ , 'features' section should be sufficient, right?12:31
dulekyboaron_: Works for me, can't find a better fit.12:32
yboaron_dulek, OK,10x12:33
*** maysamacedos has joined #openstack-kuryr12:35
openstackgerritYossi Boaron proposed openstack/kuryr-kubernetes master: Kuryr-Controller: make handlers pluggable  https://review.openstack.org/55573512:45
*** atoth has quit IRC12:47
dulekltomasbo, dmellado: BTW about container building using upper constraints - that cherry-pick needs to point to stable/queens UCs.12:47
dmelladodulek: sure, I assumed that when cherry-picking12:48
dulekdmellado: (moving here) What do you mean by healthchecks on DevStack?12:51
dmelladodulek: i.e to check when devstack finishes that i.e kuryr containers are ready12:51
dmelladoand otherwise spit out an error12:52
dmelladowe put up containerized versions but aren't checking that all is fine when stacking12:52
dulekdmellado: Ah, this one. Probably a good idea, though it'll need a bit of waiting with timeout as containers are spawned just in the end.12:52
dmelladoI guess timeout'd be fine12:52
dmelladohmm12:53
*** salv-orlando has quit IRC12:53
dmelladooddly I'm seeing some issues on my BM environment12:53
dmelladodifferent from what I thought'd be xD12:53
*** salv-orlando has joined #openstack-kuryr12:54
dmelladokube-system   kuryr-cni-ds-swbdr                  0/1       CrashLoopBackOff    2          4m12:55
dmelladokuryr-cni goes to hell xD12:55
*** salv-orlando has quit IRC12:58
dulekdmellado: Logs?13:01
dulekdmellado: kubectl -n kube-system logs  kuryr-cni-ds-swbdr13:01
dmelladonvm, beaker issues13:02
dmelladowith iptables issues by default13:02
*** atoth has joined #openstack-kuryr13:03
* dmellado sighs, why would they block stuff by default when not asking13:06
*** yamamoto has quit IRC13:39
*** yamamoto has joined #openstack-kuryr13:39
*** yamamoto has quit IRC13:44
openstackgerritYossi Boaron proposed openstack/kuryr-kubernetes master: Devstack deployment - handle LoadBalancer creation failure  https://review.openstack.org/55861613:49
*** yamamoto has joined #openstack-kuryr14:01
openstackgerritYossi Boaron proposed openstack/kuryr-kubernetes master: Kuryr-Controller: make handlers pluggable  https://review.openstack.org/55573514:02
*** hongbin_ has joined #openstack-kuryr14:10
*** fkautz has quit IRC14:11
*** fkautz has joined #openstack-kuryr14:12
dulekcelebdor1, dmellado: I've tested https://review.openstack.org/#/c/561244/. Seems to be working fine.14:30
*** celebdor1 is now known as apuimedo14:30
dulekI knew celebdor1 is apuimedo!14:30
apuimedodulek: you need irenab to merge https://review.openstack.org/#/c/561244/214:32
apuimedoonly the ancients of kuryr can merge to stable14:32
apuimedoIIRC14:33
apuimedos/merge/approve/14:33
dulekapuimedo: Hm. For kuryr-kubernetes you've gave me stable rights as well.14:33
apuimedodulek: cool14:34
apuimedoone day I have to cleanup the group hierarchy14:34
*** kiennt26_ has joined #openstack-kuryr14:34
dmelladoheh14:36
dmelladodulek: sso you tested locally the containerized versions?14:37
dulekdmellado: Yep, we don't have a gate testing containerized in stable.14:37
dmelladodulek: ack, Ill trust you, I just hope it won't be just like in Indiana Jones... Follow me! I know the way! xD14:38
dulekdmellado: Don't worry, I don't wear a hat. I don't even look good in red fedora.14:41
dmelladoxD14:41
openstackgerritYossi Boaron proposed openstack/kuryr-kubernetes master: Kuryr controller - pluggable handlers documentation  https://review.openstack.org/56144714:43
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Create CNI bin dir in OpenShift DevStack plugin  https://review.openstack.org/55677714:45
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Add `privileged` SCC to SA in OpenShift DevStack  https://review.openstack.org/55695914:45
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Add HTTPS support to K8s API healthchecks  https://review.openstack.org/55696014:45
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: WiP: Gate with containerized deployment and OpenShift  https://review.openstack.org/55731314:45
dulekLet's see how this chain will do. If anyone doesn't remember this is supposed to unblock DevStack with OpenShift and containerized Kuryr.14:45
dulekProbably useful for openshfit-ansible as well if we ever want that. ;)14:46
apuimedodmellado: dulek: I brought it up yesterday14:47
apuimedoabout backporting the zuul improvements dmellado did to stable/queens14:47
apuimedowhat's your take on it?14:47
dulekapuimedo: I think there was a lot of them. Which improvements are we talking about?14:48
apuimedodulek: I was talking about all the localrc instead of relying on the gaterc files14:48
openstackgerritMichał Dulko proposed openstack/kuryr-kubernetes master: Gate with containerized deployment and OpenShift  https://review.openstack.org/55731314:49
*** yboaron_ has quit IRC14:49
dulekOkay, so the move to the *true* Zuul v3.14:49
apuimedodulek: why does the kuryr controller need privileges?14:49
apuimedohttps://review.openstack.org/#/c/556959/3/devstack/plugin.sh14:49
dulekapuimedo: Uh, just a sec…14:50
dulekapuimedo: https://github.com/openstack/kuryr-kubernetes/blob/1ab7e66a82e667cb439404f94cd36b2f328d20c6/devstack/lib/kuryr_kubernetes#L40914:51
dulekapuimedo: Because we've named our Kuryr ServiceAccount like that.14:51
dulekapuimedo: But with Daemon CNI container needs it as well.14:51
dulekAnd the name meaning got a bit wrong now.14:52
apuimedook14:52
dulekWe could create two different SA TBH.14:52
apuimedodulek: https://review.openstack.org/#/c/556960/3 will need backporting, won't it14:52
dulekBut I don't think that should be our priotiyu.14:52
dulekSo I'll file a bug for that now.14:52
apuimedodulek: it is not a priority14:52
dulekapuimedo: Yes, that patch is a nice thing to have. We haven't noticed that until tls-proxy wasn't forced on our gate.14:53
apuimedodulek: so please backport it once it is merged14:54
dulekapuimedo: Ah, but that one will not be trivial to backport.14:54
dulekAs CNI healthchecks got in only in Rocky.14:54
*** salv-orlando has joined #openstack-kuryr14:55
dulekAnd this sparks a good discussion actually.14:55
*** gcheresh_ has quit IRC14:55
*** gcheresh_ has joined #openstack-kuryr14:56
dulekAbout what do we do with CNI healthchecks for downstream Rocky.14:56
dulekapuimedo: ^14:56
apuimedodulek: ah, right14:56
apuimedoso then nothing14:56
apuimedoxD14:56
apuimedodulek: you mean downstream queens14:56
apuimedoright?14:56
dulekapuimedo: I mean I can backport controller part easily.14:56
dulekapuimedo: Yup.14:56
dmelladoapuimedo: about backporting the zuulv3 I'm ok *but* the multinode ones14:56
dmelladowhose depen-on patches aren't ready yet14:57
apuimedodmellado: do you think you could do that?14:57
apuimedodulek: I would like to backport the cni health checks, obviously14:57
dmelladoyeah, I'll need to check some stuff with andreaf first, though14:57
apuimedobut I think we should rather do an intermediate release14:57
apuimedoand add them for origin 3.1114:57
dmelladoapuimedo: yeah, didn't we agree on that?14:57
dulekI'm a little bit concerned about backporting Zuul V3. As probably both of you are as well.14:58
* dmellado has too many things as of lately so he might mix up things14:58
dmelladodulek: that's why I need to check stuff14:58
dulekdmellado: Don't we all have too much on our plates?14:58
dmelladodulek: heh, I guess totally xD14:58
apuimedodmellado: ok. If you are concerned, I don't want it14:58
apuimedo:-)14:58
apuimedoI'll backport without14:58
dulekdmellado: If you can double check that we won't drop any testing that's in stable/queens now, I'll approve the backport.14:58
apuimedo(backport the octavia fix)14:58
dmelladoapuimedo: dulek: I'll create a patch and chack14:59
dmelladowill update you after it14:59
apuimedodmellado: what else is on your plate?14:59
dulekapuimedo: Ah, come one, this is trivial to backport that Octavia fix without dmellado's changes. :P14:59
duleks/one/on :P14:59
dmelladoapuimedo: multinode gates, openshift on openstack integration, network policy support15:00
*** salv-orlando has quit IRC15:00
dmellado+ the dreaded final report, which is worse than cifuentes master's degree15:00
dmelladoxD15:00
apuimedodulek: is that you volunteering?15:01
apuimedoxD15:01
apuimedowhat about openshift on openstack?15:01
apuimedodmellado: which masters degree?15:01
apuimedoThe one she never had or the one she "returned"15:01
apuimedoxD15:01
dmelladoxDDDD15:02
dmelladoapuimedo: I'm not really sure about which one is worse, that one or casado's15:02
dulekapuimedo: Ah why not, give me 5 minutes. xD15:02
dulekWhat's the deal with cifuentes master's degree?15:03
*** salv-orlando has joined #openstack-kuryr15:05
dmelladodulek: that's a local politician who claimed to have a master's degree but wasn't able to show her work on that at all15:05
dmelladoshe didn't go to classes nor did a final job15:05
dmelladoall that involved quite a big corruption network going from a public university to several politicians15:05
dmelladothere was another one who claimed to have a postgraduate degree from harvard and it was a 4-day course from a local university in a town called Aravaca15:06
dulekdmellado: Oh, yes. I've even used that fact in a conversation lately to prove it's not only Poland's system that broken on many levels. ;)15:06
dmelladohttps://twitter.com/XarnegoSedicios/status/984685688951050241/photo/115:07
dulekdmellado: Oh, that's niiiice. :D15:08
apuimedodulek: I think https://review.openstack.org/561447 is ready15:10
dmelladodulek: also, the funny thing is that if she resigns, she'll have to go into court due to another corruption case, which is currently not being processed due to her current legal state as a governor... something at berlusconi's levels15:10
dmelladoso she'd stick like a tick15:11
apuimedodmellado: I see another round of catalan arrests if that happens15:11
dulekapuimedo: I have it opened, but got engaged into backporting your patch (almost there!).15:11
apuimedoto keep the press occupied15:11
dmelladoapuimedo: ttotally, that's the smokescreen15:11
apuimedofucking hell, I can't get stacking15:12
dmelladoapuimedo: grab my system15:12
dmelladoI'm outta here to home anyway15:12
dmelladoso I'll be commuting15:12
dmelladogimme your ssh key again15:12
apuimedohttps://gist.github.com/celebdor/77f1130eb8763078a2c997a2ebf9149415:12
apuimedoI'll reprovision mine as well15:12
dulekdmellado: We were talking about Indiana Jones, so grab a less politicized meme from the east of Europe: https://i.redd.it/zq5qyvn93vr01.jpg (it's former Prime Minister on the left). ;)15:13
dmelladolol15:14
apuimedoLao Che!15:14
*** yamamoto has quit IRC15:17
*** yamamoto has joined #openstack-kuryr15:17
dulekapuimedo: https://review.openstack.org/#/c/561974/ - this should be fine, let's wait for CI.15:19
*** pcaruana has quit IRC15:28
celebdorVery well! Thanks!15:31
openstackgerritMerged openstack/kuryr-kubernetes master: Devstack deployment - handle LoadBalancer creation failure  https://review.openstack.org/55861615:34
*** kiennt26_ has quit IRC15:43
openstackgerritMerged openstack/kuryr-kubernetes master: Kuryr-Controller: make handlers pluggable  https://review.openstack.org/55573515:50
dulekapuimedo: "V" should be lowercase? Like really? :D15:51
dulekI won't interrupt testing now, but will fix in next patchset.15:51
*** maysamacedos has quit IRC16:16
*** pcaruana has joined #openstack-kuryr16:18
*** gcheresh_ has quit IRC16:27
*** gcheresh_ has joined #openstack-kuryr16:28
*** yamamoto_ has joined #openstack-kuryr16:30
*** gcheresh_ has quit IRC16:32
*** yamamoto has quit IRC16:33
*** maysamacedos has joined #openstack-kuryr16:57
*** vkuramshin has quit IRC17:07
celebdorXD17:10
celebdordulek: what makes you think the Ubuntu minimal is needed and that it will be backported?17:11
dulekcelebdor: Quick talk with Octavia folks: http://eavesdrop.openstack.org/irclogs/%23openstack-lbaas/%23openstack-lbaas.2018-04-17.log.html#t2018-04-17T16:37:0017:12
dulekcelebdor: I'm not happy about it, it'll definitely take time.17:12
celebdorWe don't have the time17:12
dulekcelebdor: Anyway we're broken because DIB doesn't work with pip 10. And as workaround Octavia has this minimal Ubuntu.17:13
dulekcelebdor: So we need to wait for either DIB fix or Octavia's workaround.17:13
celebdorDamn17:13
dulekcelebdor: We could backport dmellado's amphorae tarball thing.17:14
dulekcelebdor: The problem is - I'm not sure if that's not the cause of Octavia errors we see in gate.17:14
dulekcelebdor: If we're in a big hurry, then we're more or less screwed here with force pushing as last resort.17:15
celebdorI was thinking the same17:15
*** yamamoto_ has quit IRC17:15
*** yamamoto has joined #openstack-kuryr17:16
celebdorI'll check later when I put the kids to sleep17:16
celebdorBut master Gates work now, right?17:17
dulekcelebdor: No. Due to this second Octavia issue.17:21
dulekcelebdor: The one I've asked cgoncalves in the morning. :(17:21
dulekcelebdor: I was suspecting it's because we use tarballed amphorae, but I don't have a proof.17:22
*** jchhatbar has quit IRC17:27
*** hongbin_ has quit IRC18:41
*** hongbin_ has joined #openstack-kuryr18:41
*** maysamacedos has quit IRC20:04
*** yamamoto has quit IRC20:29
*** yamamoto has joined #openstack-kuryr20:30
*** yamamoto has quit IRC20:34
*** pcaruana has quit IRC20:42
*** garyloug has quit IRC21:15
*** yamamoto has joined #openstack-kuryr21:52
*** rh-jelabarre has quit IRC21:58
*** yamamoto has quit IRC21:58
*** maysamacedos has joined #openstack-kuryr22:03
*** yamamoto has joined #openstack-kuryr22:15
*** yamamoto has quit IRC22:20
openstackgerritAntoni Segura Puimedon proposed openstack/kuryr-kubernetes master: devstack: fix octavia allocation ranges  https://review.openstack.org/56206722:24
*** apuimedo has quit IRC22:31
*** salv-orl_ has joined #openstack-kuryr22:33
*** yamamoto has joined #openstack-kuryr22:35
*** salv-orlando has quit IRC22:36
*** yamamoto has quit IRC22:42
*** hongbin_ has quit IRC22:58
*** yamamoto has joined #openstack-kuryr22:58
*** yamamoto has quit IRC23:03
*** yamamoto has joined #openstack-kuryr23:19
*** yamamoto has quit IRC23:24
*** atoth has quit IRC23:25
*** yamamoto has joined #openstack-kuryr23:40
*** yamamoto has quit IRC23:46
« Monday, 2018-04-16 Index Wednesday, 2018-04-18 »

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