| *** sapd1_x has joined #openstack-containers | 00:11 | |
| *** absubram has quit IRC | 00:16 | |
| *** sapd1_x has quit IRC | 00:30 | |
| *** ricolin has joined #openstack-containers | 00:56 | |
| jakeyip | I think it's ok I got kubeconfig to work. Never mind me I think I was just confused. | 00:58 |
|---|---|---|
| *** threestrands has joined #openstack-containers | 01:24 | |
| *** hongbin has joined #openstack-containers | 01:37 | |
| *** itlinux has joined #openstack-containers | 02:09 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [fedora_atomic] Support auto healing for k8s https://review.openstack.org/631378 | 02:47 |
| *** itlinux has quit IRC | 02:52 | |
| *** itlinux has joined #openstack-containers | 02:54 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [k8s] Master nodes auto healing https://review.openstack.org/653270 | 03:16 |
| *** ykarel|away has joined #openstack-containers | 03:19 | |
| *** chhagarw has joined #openstack-containers | 03:24 | |
| *** chhagarw has quit IRC | 03:28 | |
| *** hongbin has quit IRC | 03:38 | |
| *** itlinux has quit IRC | 03:43 | |
| *** chhagarw has joined #openstack-containers | 04:09 | |
| *** ykarel|away has quit IRC | 04:27 | |
| *** udesale has joined #openstack-containers | 04:37 | |
| *** ykarel|away has joined #openstack-containers | 04:43 | |
| *** sapd1 has quit IRC | 04:50 | |
| *** ramishra has joined #openstack-containers | 04:50 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [k8s] Master nodes auto healing https://review.openstack.org/653270 | 04:54 |
| *** absubram has joined #openstack-containers | 05:01 | |
| *** absubram has quit IRC | 05:06 | |
| *** absubram has joined #openstack-containers | 05:10 | |
| *** ykarel|away is now known as ykarel | 05:10 | |
| *** ivve has joined #openstack-containers | 05:13 | |
| *** janki has joined #openstack-containers | 05:49 | |
| *** absubram has quit IRC | 06:01 | |
| *** pcaruana has joined #openstack-containers | 06:11 | |
| *** chhagarw has quit IRC | 06:22 | |
| *** chhagarw has joined #openstack-containers | 06:22 | |
| *** udesale has quit IRC | 06:37 | |
| *** udesale has joined #openstack-containers | 06:38 | |
| *** udesale has quit IRC | 06:39 | |
| *** udesale has joined #openstack-containers | 06:44 | |
| *** udesale has quit IRC | 06:46 | |
| *** udesale has joined #openstack-containers | 06:56 | |
| *** yolanda_ has joined #openstack-containers | 07:07 | |
| *** flwang1 has joined #openstack-containers | 07:34 | |
| *** gbialas has joined #openstack-containers | 07:36 | |
| * flwang1 is so excited that strigazi finally +2ed his autoscaling patch | 07:38 | |
| *** gbialas has quit IRC | 07:40 | |
| *** ykarel is now known as ykarel|lunch | 07:43 | |
| *** alisanhaji has joined #openstack-containers | 07:54 | |
| *** ttsiouts has joined #openstack-containers | 08:06 | |
| *** rcernin has quit IRC | 08:13 | |
| *** ttsiouts has quit IRC | 08:17 | |
| *** ttsiouts has joined #openstack-containers | 08:17 | |
| *** ttsiouts has quit IRC | 08:22 | |
| *** ttsiouts has joined #openstack-containers | 08:28 | |
| *** rcernin has joined #openstack-containers | 08:28 | |
| *** ykarel|lunch is now known as ykarel | 08:31 | |
| openstackgerrit | Merged openstack/magnum master: [fedora_atomic] Support auto healing for k8s https://review.openstack.org/631378 | 08:36 |
| openstackgerrit | Merged openstack/magnum stable/stein: Support multi DNS server https://review.openstack.org/652784 | 08:48 |
| openstackgerrit | Merged openstack/magnum master: Fix registry on k8s_fedora_atomic https://review.openstack.org/652096 | 08:48 |
| *** udesale has quit IRC | 08:51 | |
| flwang1 | strigazi: around for a chat? | 09:02 |
| flwang1 | just ping me when you're available | 09:03 |
| strigazi | flwang1: ping | 09:04 |
| flwang1 | strigazi: pong | 09:04 |
| flwang1 | strigazi: thank you for approving my patch :) | 09:04 |
| strigazi | flwang1 is so excited that strigazi finally +2ed his autoscaling patch, works great!!! | 09:04 |
| strigazi | flwang1: I'll override the v1.0 image in dockerhub with Thomas' fix | 09:05 |
| strigazi | makes sense? | 09:05 |
| flwang1 | i need to resolve the conflicts before backporting to stable/stein | 09:05 |
| strigazi | or v2.0 | 09:05 |
| flwang1 | strigazi: better override | 09:05 |
| strigazi | I prefer tags to be immutable, but I don't mind for something that is not merged | 09:05 |
| strigazi | I mean, not merged in magnum side | 09:06 |
| flwang1 | strigazi: it's happening now | 09:06 |
| flwang1 | so just do it | 09:06 |
| flwang1 | as for auto healing, i have another idea i'd like to get your comments | 09:06 |
| strigazi | before getting to that. I think I commented in all your patches, did I miss something? | 09:07 |
| openstackgerrit | Merged openstack/magnum master: Dropping the py35 testing https://review.openstack.org/652396 | 09:07 |
| flwang1 | strigazi: i'd like to get a general option for this https://review.openstack.org/#/c/643225/ | 09:08 |
| flwang1 | and this https://review.openstack.org/#/c/648590/ if you can | 09:08 |
| strigazi | pushed openstackmagnum/cluster-autoscaler:v1.0 | 09:11 |
| flwang1 | strigazi: cool | 09:11 |
| strigazi | not it has the fix https://github.com/kubernetes/autoscaler/pull/1873 | 09:11 |
| flwang1 | why? | 09:11 |
| flwang1 | shouldn't we release the latest version otherwise the scale down doesn't work | 09:12 |
| strigazi | s/not it has the fix https://github.com/kubernetes/autoscaler/pull/1873/NOW it has the fix https://github.com/kubernetes/autoscaler/pull/1873/ | 09:13 |
| strigazi | all good, right? | 09:14 |
| flwang1 | i'm happy now :) | 09:14 |
| strigazi | https://review.openstack.org/#/c/648590 +2 | 09:16 |
| *** ttsiouts has quit IRC | 09:17 | |
| flwang1 | thanks | 09:18 |
| *** ttsiouts has joined #openstack-containers | 09:18 | |
| strigazi | flwang1 for https://review.openstack.org/#/c/643225/ | 09:18 |
| flwang1 | https://review.openstack.org/#/c/643225/ thoughts? | 09:18 |
| strigazi | hehe, i was faster :) | 09:18 |
| flwang1 | well, you win | 09:19 |
| strigazi | I left a comment, After rebasing, works fine. Thanks! we just need the docs change | 09:19 |
| strigazi | patch looks fine | 09:19 |
| *** ttsiouts_ has joined #openstack-containers | 09:19 | |
| flwang1 | strigazi: the background is, we(catalyst cloud) would like a out-of-box keystone auth support just like GKE | 09:19 |
| *** ttsiouts has quit IRC | 09:19 | |
| flwang1 | user don't have to deal with the policy, just use it | 09:19 |
| strigazi | I just have a related question. Do you have a role (not the admin one) that can do role assignments | 09:20 |
| *** udesale has joined #openstack-containers | 09:20 | |
| flwang1 | strigazi: at project level, we do | 09:20 |
| flwang1 | we're using Adjutant to do that | 09:21 |
| flwang1 | Adjutant helps us doing those work need admin role | 09:21 |
| flwang1 | for example, sign up | 09:21 |
| flwang1 | and invite new user and assign roles for them | 09:21 |
| flwang1 | for this case, we will create those new roles in keystone, and adding them into Adjutant, then project admin can assign those roles for project members | 09:22 |
| flwang1 | strigazi: does that make sense? | 09:22 |
| strigazi | project admin will you openstack cli? | 09:23 |
| strigazi | yes it will | 09:24 |
| flwang1 | strigazi: at this moment, i'm not sure too sure if Adjutant cant support cli for that | 09:24 |
| flwang1 | but for dashboard, we do | 09:25 |
| strigazi | makes sense, we have something in house built long time ago. anyway | 09:25 |
| strigazi | did I miss any other patch | 09:25 |
| flwang1 | i will rebase the patch and update the user guide doc | 09:25 |
| strigazi | ? | 09:25 |
| flwang1 | no, you make my day | 09:25 |
| flwang1 | this is the most happy day for me in the last 3 months | 09:26 |
| strigazi | I hope you are exaggerating :) | 09:26 |
| strigazi | but in context, makes some sense | 09:26 |
| flwang1 | :D | 09:27 |
| flwang1 | before talking about rolling upgrade, give me 5 mins about the master node auto healing? | 09:27 |
| strigazi | +1 | 09:28 |
| strigazi | I mean, go on | 09:28 |
| flwang1 | thanks | 09:29 |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [k8s] Master nodes auto healing https://review.openstack.org/653270 | 09:29 |
| flwang1 | ^ this is my general idea | 09:30 |
| flwang1 | mostly are here https://review.openstack.org/#/c/653270/3/magnum/service/periodic.py@126 | 09:30 |
| strigazi | flwang1: it makes some sense | 09:32 |
| strigazi | I think with the template changes in upgrade, it *may* work | 09:32 |
| flwang1 | strigazi: exactly, the work may need your work in the upgrade patch | 09:33 |
| strigazi | I'm not sure I agree with the octavia status check | 09:33 |
| strigazi | it should ask kubernetes | 09:33 |
| flwang1 | strigazi: it's master, my friend | 09:33 |
| flwang1 | we can't get health status for each master | 09:33 |
| strigazi | correct, if the request times out | 09:33 |
| strigazi | but not through octavia, not everyone has it :) | 09:34 |
| flwang1 | we can only get the overall api status or the etcd api status | 09:34 |
| flwang1 | strigazi: i understand that | 09:34 |
| flwang1 | is there any other way to get the status for each master? so that i can make it plugable? | 09:34 |
| *** threestrands has quit IRC | 09:36 | |
| strigazi | I mean, we can ask directly the api_address. but we need cases, fip enabled and so on | 09:36 |
| flwang1 | strigazi: yep, i thought that as well | 09:36 |
| flwang1 | can we use the ca certs in magnum db to talk to etcd api? | 09:37 |
| strigazi | flwang1 unfortunately yes, we must change that | 09:37 |
| flwang1 | and will it still work if we generate different ca certs for api, etcd and front-proxy? | 09:38 |
| strigazi | I think I mentioned it in the past, we need a new ca for it | 09:38 |
| flwang1 | i know | 09:38 |
| flwang1 | there is a security hole | 09:38 |
| flwang1 | it's on my list | 09:38 |
| strigazi | ok | 09:38 |
| flwang1 | if i can make a base function like _give_me_the_broken_master(), are you happy with the overall desgin? | 09:39 |
| strigazi | +1 | 09:39 |
| flwang1 | cool | 09:39 |
| flwang1 | deal | 09:39 |
| *** lpetrut has joined #openstack-containers | 09:40 | |
| flwang1 | following that | 09:40 |
| flwang1 | actually, based on our current resize api, the worker node auto healing can be done by the similar way, i'd like to call it auto_healing_mode=hard, because user can always keep the same node　count | 09:41 |
| flwang1 | by checking the health_status_reason, we can know which worker is down, and then we can call resize api to replace the node | 09:41 |
| strigazi | what is soft? | 09:41 |
| flwang1 | soft is our current way with CA | 09:42 |
| strigazi | ok | 09:42 |
| openstackgerrit | Merged openstack/magnum master: [fedora atomic] Allow traffic between k8s workers https://review.openstack.org/652931 | 09:42 |
| flwang1 | how do you think? | 09:42 |
| strigazi | hard > soft | 09:42 |
| flwang1 | you like the 'hard'(rock) way? | 09:43 |
| strigazi | yes, I don't know if it will conflict with the CA | 09:44 |
| flwang1 | strigazi: it won't | 09:44 |
| flwang1 | the new label 'auto_healing_mode' can manage that | 09:45 |
| strigazi | if a node is NotReady, the CA won't do anything? | 09:45 |
| strigazi | I'll be back in a few mins <5 | 09:46 |
| flwang1 | assume auto_healing_enabled=True, if auto_healing_mode=hard, mangum-conductor will take over, otherwise CA+Draino will take over | 09:46 |
| strigazi | back | 09:48 |
| strigazi | flwang1: ok, but | 09:48 |
| strigazi | if auto_scaling_enabled=True and auto_healing_mode=hard ? | 09:48 |
| flwang1 | if auto_scaling_enabled=True and auto_healing_mode=hard, that's the magnum-conductor way | 09:49 |
| flwang1 | if auto_healing_enabled=False, no auto_healing for either master or node | 09:49 |
| flwang1 | as long as we have containerized master, we won't have to care the master | 09:50 |
| flwang1 | at that moment, the auto_healing_mode may make more sense | 09:50 |
| strigazi | if auto_scaling_enabled=True and auto_healing_mode=hard, that's the magnum-conductor way. OK. but what the CA does when it sees a node as not ready? | 09:51 |
| flwang1 | strigazi: AFAICT, CA won't automatically remove bad node | 09:54 |
| flwang1 | that's the work need to be down by Draino | 09:54 |
| flwang1 | but at this case, Draino won't be installed | 09:54 |
| strigazi | draino, just dains right? | 09:54 |
| flwang1 | cordon and drain | 09:54 |
| strigazi | And then CA scales down because it is empty right? | 09:55 |
| flwang1 | after drain, the node will be empty, and CA will remove the empty node | 09:55 |
| flwang1 | yes | 09:55 |
| strigazi | ok | 09:55 |
| strigazi | actually soft and hard can "co-exist" in a way. It is whoever comes first, conductor or CA | 09:56 |
| flwang1 | strigazi: it could be, but that will be messed and will be hard to debug | 09:56 |
| flwang1 | but technically, yes | 09:56 |
| *** gbialas has joined #openstack-containers | 09:58 | |
| flwang1 | so far, i think generally you like the idea? | 09:58 |
| strigazi | sounds good | 09:59 |
| flwang1 | cool, i will polish it | 09:59 |
| flwang1 | and that does need your upgrade patch | 09:59 |
| flwang1 | what's the status of the last part? | 09:59 |
| strigazi | I finish with the Software deployment of maste and we are good. The templates are a bit messed but it almost works. I don't have anything else to do today | 10:01 |
| strigazi | I have only a question | 10:02 |
| strigazi | Lixian tried to change somehting similar, shall we change the template version to newton? | 10:03 |
| strigazi | to have conditions | 10:03 |
| flwang1 | strigazi: personally i think we should | 10:03 |
| strigazi | now? | 10:03 |
| flwang1 | i don't think anyone can use magnum without upgrade heat to queens | 10:03 |
| strigazi | raelly? | 10:04 |
| strigazi | really? | 10:04 |
| flwang1 | fine | 10:04 |
| strigazi | due to the agent? | 10:04 |
| flwang1 | agent issue is a big problem | 10:04 |
| strigazi | fyi, we upgraded heat to rocky yesterday | 10:04 |
| flwang1 | but if they don't care region, then could be | 10:04 |
| flwang1 | but | 10:04 |
| flwang1 | i still think stick on very old heat version means we're losing more than we got | 10:05 |
| flwang1 | we can send an email to the community to survey if there is any company is using Magnum with an old heat ( older than newton) | 10:06 |
| flwang1 | if there is nobody, we're safe to go | 10:06 |
| strigazi | or queens | 10:07 |
| strigazi | ok, will you send an email? | 10:07 |
| flwang1 | strigazi: i can do that | 10:08 |
| flwang1 | and you can just focus on your work | 10:08 |
| strigazi | ok | 10:08 |
| strigazi | have you tested my patches for 1.14? | 10:08 |
| strigazi | conformance passes for calico and flannel | 10:08 |
| flwang1 | you mean the calico one? or the upgrade one? | 10:08 |
| strigazi | coredns/calico/1.14 | 10:09 |
| *** strigazi has quit IRC | 10:09 | |
| flwang1 | i reviewed them separatedly, but didn't test togehter | 10:10 |
| flwang1 | i do have some questions | 10:10 |
| *** strigazi has joined #openstack-containers | 10:10 | |
| flwang1 | 1. for the coreDNS, https://review.openstack.org/#/c/652058/2/magnum/drivers/common/templates/kubernetes/fragments/core-dns-service.sh@74 | 10:10 |
| flwang1 | so with the new coreDNS versioin, we don't have to provide the network cidr to coreDNS? | 10:11 |
| strigazi | flwang1 ok, easy, other than that, all good? | 10:11 |
| strigazi | flwang1: V | 10:12 |
| strigazi | flwang1: https://github.com/coredns/deployment/blob/master/kubernetes/deploy.sh#L7 | 10:12 |
| strigazi | I think both work | 10:13 |
| flwang1 | ok, i will take a look | 10:13 |
| strigazi | I can try passing the cidr like before | 10:13 |
| *** livelace has joined #openstack-containers | 10:13 | |
| flwang1 | 2. for calico v3.3, how calico know the etcd server to save data? | 10:13 |
| flwang1 | i didn't see how the etcd address being passed in as before | 10:14 |
| *** jchhatbar has joined #openstack-containers | 10:14 | |
| strigazi | it uses custom resource definitions, not more etcd. It uses the kubernetes api directly | 10:14 |
| flwang1 | fantastic | 10:15 |
| strigazi | flannel does the same, one of the reason I wanted to make it a cni | 10:15 |
| flwang1 | then we can remove those info https://review.openstack.org/#/c/652059/4/magnum/drivers/common/templates/kubernetes/fragments/calico-service.sh@10 | 10:16 |
| *** janki has quit IRC | 10:16 | |
| strigazi | I forgot about that | 10:16 |
| strigazi | sure | 10:16 |
| flwang1 | i you can propose new patchset, i'm happy to test it tomorrow and +2, | 10:17 |
| strigazi | ok | 10:17 |
| flwang1 | btw, i will take leave from this Firday 19th Apr to 28th | 10:17 |
| strigazi | just a question, | 10:17 |
| flwang1 | for camping | 10:17 |
| flwang1 | no laptop | 10:17 |
| flwang1 | sure | 10:17 |
| strigazi | flwang1 proper holidats | 10:17 |
| strigazi | flwang1 proper holidays | 10:18 |
| flwang1 | yep, Easter holidays | 10:18 |
| flwang1 | i do need a break, otherwise i will be broken soon | 10:18 |
| strigazi | CALICO_IPV4POOL_IPIP upstream is on, in our case it is off | 10:19 |
| strigazi | we use CALICO_IPV4POOL_NAT_OUTGOING, upstream example config it is not using it https://docs.projectcalico.org/v3.3/getting-started/kubernetes/installation/hosted/kubernetes-datastore/calico-networking/1.7/calico.yaml | 10:20 |
| strigazi | any reason you did it like this for v2? | 10:20 |
| flwang1 | strigazi: i can't remember the details, TBH | 10:21 |
| flwang1 | i'm ok to follow the community way | 10:21 |
| flwang1 | i mean the default yaml from calico | 10:22 |
| strigazi | ok, in any case, conformance passes. I'll check if the upstream config works too. | 10:22 |
| strigazi | let's check tomorrow same time, or you are off already? | 10:22 |
| strigazi | let's check tomorrow same time, or you will be off already? | 10:22 |
| *** janki has joined #openstack-containers | 10:23 | |
| flwang1 | are you all k8s nodes in a same l2 network? | 10:24 |
| flwang1 | strigazi: i can meet with you tomorrow at the same time | 10:24 |
| flwang1 | like 9:00 UTC | 10:25 |
| flwang1 | but won't stay long | 10:25 |
| strigazi | short and efficient is better | 10:25 |
| flwang1 | since we will leave very early on Friday | 10:25 |
| flwang1 | strigazi: cool | 10:25 |
| *** jchhatbar has quit IRC | 10:25 | |
| flwang1 | i will dig the CALICO_IPV4POOL_IPIP and leave comments on your calico patch | 10:25 |
| strigazi | I'm going home for holidays at the same time but not real | 10:25 |
| flwang1 | haha | 10:25 |
| flwang1 | now, tough question | 10:26 |
| strigazi | I have to do four presentations for the summit the week after | 10:26 |
| flwang1 | after i back from holiday, can i see a new patch set for the upgrade patch? | 10:26 |
| flwang1 | strigazi: ah, i see | 10:26 |
| strigazi | yes | 10:26 |
| flwang1 | good luck and i believe you will be fine | 10:26 |
| flwang1 | strigazi: cool, then i give you 10-NO-flwang-days | 10:27 |
| flwang1 | i will start to bug you since 29th | 10:27 |
| strigazi | Well see you tomorrow | 10:27 |
| flwang1 | see you, thanks for making my day | 10:27 |
| strigazi | cheers | 10:28 |
| openstackgerrit | Feilong Wang proposed openstack/magnum stable/stein: Fix registry on k8s_fedora_atomic https://review.openstack.org/653379 | 10:32 |
| *** gbialas has quit IRC | 10:32 | |
| *** livelace has quit IRC | 10:33 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum stable/stein: [fedora_atomic] Support auto healing for k8s https://review.openstack.org/653380 | 10:37 |
| *** ttsiouts_ has quit IRC | 10:37 | |
| *** ttsiouts has joined #openstack-containers | 10:38 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum stable/stein: [fedora_atomic] Support auto healing for k8s https://review.openstack.org/653380 | 10:39 |
| *** ttsiouts has quit IRC | 10:42 | |
| *** gbialas has joined #openstack-containers | 11:06 | |
| *** ttsiouts has joined #openstack-containers | 11:08 | |
| *** gbialas has quit IRC | 11:09 | |
| openstackgerrit | Merged openstack/magnum master: Allow admin update cluster/template in any project https://review.openstack.org/648590 | 11:17 |
| *** jchhatbar has joined #openstack-containers | 11:20 | |
| *** ttsiouts has quit IRC | 11:20 | |
| *** ttsiouts has joined #openstack-containers | 11:21 | |
| *** janki has quit IRC | 11:22 | |
| *** ttsiouts has quit IRC | 11:25 | |
| *** strigazi has quit IRC | 11:25 | |
| *** strigazi has joined #openstack-containers | 11:25 | |
| *** ttsiouts has joined #openstack-containers | 11:27 | |
| *** ttsiouts has quit IRC | 11:43 | |
| *** ttsiouts has joined #openstack-containers | 11:44 | |
| *** ttsiouts_ has joined #openstack-containers | 11:46 | |
| *** ttsiouts has quit IRC | 11:47 | |
| *** strigazi has quit IRC | 11:50 | |
| *** ttsiouts_ has quit IRC | 11:51 | |
| *** strigazi has joined #openstack-containers | 11:51 | |
| *** strigazi has quit IRC | 11:52 | |
| *** strigazi has joined #openstack-containers | 11:52 | |
| *** strigazi has quit IRC | 11:55 | |
| *** strigazi has joined #openstack-containers | 11:55 | |
| *** pcaruana has quit IRC | 12:30 | |
| *** pcaruana has joined #openstack-containers | 12:53 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: k8s: set functional jobs to voting https://review.openstack.org/627756 | 13:00 |
| *** ykarel is now known as ykarel|afk | 13:36 | |
| *** ykarel|afk has quit IRC | 13:41 | |
| openstackgerrit | Diogo Guerra proposed openstack/magnum master: [k8s] Set traefik to stable version v1.7.10 https://review.openstack.org/645632 | 14:16 |
| *** itlinux has joined #openstack-containers | 14:19 | |
| *** itlinux has quit IRC | 14:22 | |
| *** ykarel|afk has joined #openstack-containers | 14:24 | |
| *** ykarel|afk is now known as ykarel | 14:25 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: Revert "ci: Disable functional tests" https://review.openstack.org/642873 | 14:26 |
| *** lpetrut has quit IRC | 14:28 | |
| strigazi | mnaser: I think I figured out the issue you had with flannel, security groups see: https://review.openstack.org/#/c/652931/3 | 14:52 |
| mnaser | strigazi: that might make sense then | 14:53 |
| mnaser | merging that will fix it though, I think right? | 14:53 |
| *** jchhatbar has quit IRC | 14:53 | |
| *** jchhatbar has joined #openstack-containers | 14:54 | |
| strigazi | We merged this, will cherry-pick to stein | 14:54 |
| mnaser | strigazi: great, I can redeploy and let you know | 14:56 |
| strigazi | mnaser: use vxlan as backedn | 14:57 |
| strigazi | mnaser: use vxlan as backend | 14:57 |
| *** udesale has quit IRC | 14:57 | |
| strigazi | --labels flannel_backend=vxlan | 14:57 |
| mnaser | I guess it defaults to dup | 14:58 |
| mnaser | udp | 14:58 |
| strigazi | yes | 14:58 |
| openstackgerrit | Mohammed Naser proposed openstack/magnum stable/stein: [fedora atomic] Allow traffic between k8s workers https://review.openstack.org/653459 | 14:58 |
| mnaser | strigazi: to make your life easier ^ | 14:59 |
| *** udesale has joined #openstack-containers | 14:59 | |
| *** udesale has quit IRC | 14:59 | |
| *** udesale has joined #openstack-containers | 14:59 | |
| strigazi | mnaser: when it was failing for you it was udp or vxlan? | 14:59 |
| mnaser | strigazi: it was default, the cni0 interface never went up | 15:00 |
| mnaser | apparently though sometimes it never goes up unless the node has interfaces attached | 15:00 |
| strigazi | udp then | 15:00 |
| *** lpetrut has joined #openstack-containers | 15:01 | |
| *** udesale has quit IRC | 15:06 | |
| *** absubram has joined #openstack-containers | 15:17 | |
| *** absubram has quit IRC | 15:21 | |
| *** absubram has joined #openstack-containers | 15:29 | |
| *** jchhatbar has quit IRC | 15:33 | |
| *** jchhatbar has joined #openstack-containers | 15:33 | |
| *** itlinux has joined #openstack-containers | 15:35 | |
| *** jchhatbar has quit IRC | 15:36 | |
| *** jchhatbar has joined #openstack-containers | 15:36 | |
| *** jchhatbar has quit IRC | 15:40 | |
| *** lpetrut has quit IRC | 15:42 | |
| *** ykarel is now known as ykarel|away | 16:05 | |
| *** dims has quit IRC | 16:07 | |
| *** lpetrut has joined #openstack-containers | 16:09 | |
| *** ramishra has quit IRC | 16:26 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: Update coredns from upstream manifest and to 1.3.1 https://review.openstack.org/652058 | 16:51 |
| *** dims has joined #openstack-containers | 16:54 | |
| *** dims has quit IRC | 16:59 | |
| *** dims has joined #openstack-containers | 17:01 | |
| *** ricolin has quit IRC | 17:04 | |
| *** chhagarw has quit IRC | 17:43 | |
| *** chhagarw has joined #openstack-containers | 17:48 | |
| *** ivve has quit IRC | 17:52 | |
| *** chhagarw has quit IRC | 18:23 | |
| *** chhagarw has joined #openstack-containers | 18:23 | |
| *** alisanhaji has quit IRC | 18:24 | |
| *** lpetrut has quit IRC | 18:41 | |
| *** lpetrut has joined #openstack-containers | 18:42 | |
| *** chhagarw has quit IRC | 18:55 | |
| *** ykarel|away has quit IRC | 19:02 | |
| openstackgerrit | Merged openstack/magnum stable/stein: [fedora atomic] Allow traffic between k8s workers https://review.openstack.org/653459 | 19:23 |
| openstackgerrit | Merged openstack/magnum master: [k8s] Set traefik to stable version v1.7.10 https://review.openstack.org/645632 | 19:23 |
| *** lpetrut has quit IRC | 19:40 | |
| *** absubram has quit IRC | 20:35 | |
| *** pcaruana has quit IRC | 20:43 | |
| *** logan- has quit IRC | 21:34 | |
| *** logan- has joined #openstack-containers | 21:37 | |
| *** absubram has joined #openstack-containers | 21:46 | |
| *** hongbin has joined #openstack-containers | 21:52 | |
| *** itlinux has quit IRC | 23:26 | |
| *** hongbin has quit IRC | 23:38 | |
| *** itlinux has joined #openstack-containers | 23:44 | |
| *** itlinux has quit IRC | 23:44 | |
| *** absubram has quit IRC | 23:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!