| sdake | sup spsurya | 00:05 |
|---|---|---|
| sdake | spsurya there was a discussion earlier today that affects the work you are doing | 00:05 |
| sdake | spsurya moment | 00:05 |
| spsurya | sdake yes | 00:06 |
| spsurya | that was related to | 00:06 |
| sdake | http://eavesdrop.openstack.org/irclogs/%23openstack-kolla/%23openstack-kolla.2017-04-17.log.html#t2017-04-17T15:59:55 | 00:06 |
| spsurya | docker not starting | 00:06 |
| sdake | spsurya please read that log | 00:06 |
| spsurya | sure | 00:07 |
| *** guoguo has joined #openstack-kolla | 00:12 | |
| *** guoguo has quit IRC | 00:13 | |
| *** guoguo has joined #openstack-kolla | 00:14 | |
| sdake | kfox1111 sbezverk - this is ready for review plz: https://review.openstack.org/#/c/451495/8 | 00:22 |
| sdake | spsurya what did you parse out the log | 00:22 |
| *** jascott1 has joined #openstack-kolla | 00:24 | |
| sbezverk | sdake: let's wait for the gate | 00:25 |
| sdake | sbezverk thanks :) | 00:26 |
| sbezverk | sdake: https://review.openstack.org/456906 | 00:26 |
| spsurya | sdake: till now what i understood is we are in right direction . So we need to complete the things of gate and dev env i.e the current work | 00:26 |
| spsurya | please correct if i am wrong | 00:26 |
| sdake | spsurya the way i parsed that conversation is as follows | 00:27 |
| sdake | the work in my original devtools patch - some variant of that, or a refactor of the existing tools that setup canal, setup kubernetes etc stay put - without error detection | 00:28 |
| sdake | spsurya everything gate related goes to the /gate directory | 00:28 |
| sdake | the gate tools then call the simpler more generalized setup_* tools in tools dir | 00:28 |
| sdake | rather the gate scripts in /gate call them | 00:28 |
| sdake | then we can make forward progress from there | 00:29 |
| sdake | spsurya the end result of this for this milestone is that the devtools are gated | 00:29 |
| sdake | and we don't endelessly spin on rewriting gating and development tools nad documentation | 00:29 |
| sdake | spsurya that is probably a lot to digest, and you will have to become an expert in the gate | 00:30 |
| sdake | spsurya i'd recommend tackling one file at a time if your up for the mission | 00:30 |
| sdake | kfox1111 ca nyou validate plz | 00:31 |
| sdake | jascott1 ^^ | 00:31 |
| sdake | jascott1 spsurya the good news of this is you will become experts in the gate | 00:32 |
| sdake | and also be able to become expert level contributors | 00:32 |
| sbezverk | sdake: why touch gate??? why do not you create a separate directory for dev env and work in it?? | 00:32 |
| sdake | sbezverk that isn't what kfox1111 wanted | 00:33 |
| sdake | sbezverk if you read that log above he had a fit because i did as you suggested :) | 00:33 |
| sdake | http://eavesdrop.openstack.org/irclogs/%23openstack-kolla/%23openstack-kolla.2017-04-17.log.html#t2017-04-17T15:59:55 | 00:33 |
| *** huzhengchuan has joined #openstack-kolla | 00:33 | |
| sbezverk | sdake: kfox1111 is not going to work on this, if it were him then I think it would be doable, anybody else including myself would not be able to do it quickly and painlessly | 00:34 |
| sdake | i could do it quickly and painlessly | 00:34 |
| sbezverk | so this will just burn cycles and efforts of a one or possible more developers | 00:34 |
| sdake | however the goal isn't quick and painless | 00:34 |
| sdake | sbezverk these devs don't know how to contirbute to kolla-kubernetes | 00:35 |
| sbezverk | instead of actually contributing usefull code | 00:35 |
| sbezverk | sdake: I do not beleive it | 00:35 |
| sdake | the reaosn they dont know how to contribute is because the devtools don't exist and the gate is broken | 00:35 |
| openstackgerrit | Merged openstack/kolla-kubernetes master: Converting keepalived into helm charts https://review.openstack.org/456906 | 00:35 |
| sbezverk | it is so damn easy to setup, if somebody really wanted, he/she would have completed it in one day max | 00:35 |
| sdake | what is so damn easy to setup? | 00:36 |
| sbezverk | development env | 00:36 |
| sdake | sbezverk it took me 2 months with the help of 15 contributors to write the deployment guide | 00:36 |
| sdake | sbezverk that was all about learning for me how the gate worked | 00:36 |
| sdake | that is a long ramp up time | 00:36 |
| sbezverk | sdake: because you wanted a magic script | 00:36 |
| sbezverk | that is all | 00:36 |
| sdake | sbezverk incorrct - I wanted documentaiton people could understand | 00:36 |
| sdake | a magic script is what the gate is | 00:37 |
| sdake | black fucking magic | 00:37 |
| sbezverk | just doing copy and paste from gates with minor modification would bring env up in no time | 00:37 |
| sdake | sbezverk either you trust my judgement or you dont - this is the best path forward | 00:37 |
| sbezverk | sdake: I am sorry sdake but I see waste of valuable cycles of other developers on grand initiatives.. | 00:38 |
| sdake | learning is not wasting | 00:38 |
| sdake | i attempted to mrege a review that simplifies the devleopment experience | 00:39 |
| sdake | kfox1111 wants a grand initiative | 00:39 |
| sdake | here i tis | 00:39 |
| sdake | I also want spsurya and jascott1 to learn how the gate works if they are to be contributor | 00:40 |
| sdake | s | 00:40 |
| sbezverk | sdake: well, how many weeks you think it will take? do we really have this time?? | 00:40 |
| sdake | MY POV: it is impossible to contribute to this project if you don't understand how the gate operates | 00:40 |
| sdake | sbezverk it is scheduled for 1 month | 00:40 |
| sdake | the alternative is these devs spin cycles on doing nothing | 00:41 |
| sbezverk | sdake: -2 I contributed iscsi to kolla without any idea of how gates work | 00:41 |
| sdake | as in they are completely blocked now | 00:41 |
| sdake | iscsi didn't have a gate | 00:41 |
| sdake | still doesn't | 00:41 |
| sdake | kolla-ansible has much less stringent gating requirements | 00:41 |
| sdake | that is why you were able to make such fast progress | 00:41 |
| sbezverk | sdake: anything I contributed to kolla was done without understanding the gate | 00:41 |
| sdake | sbezverk every one of my patches includes gate chagnes | 00:42 |
| sdake | sbezverk or requires gate changes | 00:42 |
| sdake | sbezverk except docs of course | 00:42 |
| *** zhurong has joined #openstack-kolla | 00:42 | |
| sbezverk | sdake: hmm, even 10k of deleting PS? | 00:42 |
| sdake | sbezverk yup even those | 00:42 |
| jascott1 | sdake you said my name three times and I have been summoned | 00:42 |
| sdake | i had to understand the gate | 00:42 |
| sdake | jascott1 :) | 00:42 |
| sdake | jascott1 do you want to learn how the gate works? | 00:43 |
| sdake | jascott1 this is what leads to understanding how to contribute to kolla-kubernetes | 00:43 |
| spsurya | sdake: sbezverk following your conversation | 00:44 |
| sdake | sbezverk it is unfortunate that kolla-kubernetes requires a deep understanding of gating to contribute to - that is what has been built | 00:44 |
| jascott1 | ive been thru the gate a few months ago, im sure its changed a lot. I used it to understand what the hell was going on with the kolla-kubernetes install when the docs were so outta whack | 00:44 |
| jascott1 | but yes I do want to learn how the gate works | 00:44 |
| sdake | jascott1 here is your opportunity | 00:44 |
| sdake | ;) | 00:44 |
| sdake | jascott1 then you can actually contribute code | 00:44 |
| spsurya | same here sdake | 00:45 |
| sbezverk | that is only sdake POV, I think to contribute to kolla-kube is not required gate | 00:45 |
| sdake | since you will be able to make gate changes on code contributions | 00:45 |
| sbezverk | knowledge, it is nice to have but not required | 00:45 |
| sdake | sbezverk please find me one example of a patch that doesn't touch the gate that isn't some sort of cleanup (such as removal of dead code) | 00:45 |
| sdake | oen is all I ask | 00:45 |
| sbezverk | actually if you use gate scripts to setup your dev env, you will get that knwoledge | 00:45 |
| sdake | sbezverk you need to experience working on the gate to understand it | 00:46 |
| jascott1 | I did some of that and did learn but its not ideal | 00:46 |
| sdake | my understanding of the gate came through working on it | 00:46 |
| sdake | sbezverk find me one example :) | 00:47 |
| sbezverk | how do you exaplina kolla/kolla-ansible? I bet only 10-20% of contributors have knowledge about the gate | 00:47 |
| sbezverk | but they are still ACTIVELY contributing | 00:47 |
| *** tonanhngo has quit IRC | 00:47 | |
| sdake | sbezverk i explain it because the gating requirements are not stringent | 00:47 |
| sdake | the gate barely tests anything in kolla-ansible | 00:47 |
| sdake | the kolla gate is braindead easy -either it builds or doesn't | 00:47 |
| *** Margin has joined #openstack-kolla | 00:48 | |
| sdake | sbezverk we have a couple options - reduce gating requirements - or teach people how to work on the gate | 00:48 |
| sdake | sbezverk and kfox1111 are of the opinion that every patch should pas the gate - so lets enable contributors to do that | 00:48 |
| sdake | sbezverk if you have a third option I'm listening | 00:49 |
| sdake | and sitll waiting for that one patch | 00:49 |
| sdake | sbezverk think back to our ironic gating work - which was fantastic btw | 00:50 |
| sbezverk | sdake: please recheck I am worried about ironic job failure | 00:50 |
| sdake | why did you do that gating work? | 00:50 |
| openstackgerrit | jeck proposed openstack/kolla-ansible master: Removing swift references from fstab https://review.openstack.org/456805 | 00:50 |
| sdake | sbezverk will do | 00:50 |
| sdake | sbezverk recheck which | 00:50 |
| masber | good morning all, according to this doc https://github.com/openstack/kolla-ansible/blob/master/doc/ironic-guide.rst ironic is in tech preview. Is there a roadmap about fully support ironic? | 00:50 |
| sbezverk | sdake: https://review.openstack.org/#/c/451495/ | 00:51 |
| sbezverk | ironic failure after your modification, I want to make sure it is transient issue | 00:51 |
| sdake | looks like stuck in gate | 00:51 |
| sdake | rather stuck in build | 00:51 |
| sdake | http://logs.openstack.org/95/451495/8/check/gate-kolla-kubernetes-deploy-centos-source-4-ironic-nv/3d333b7/console.html#_2017-04-18_00_15_03_285473 | 00:51 |
| sdake | sbezverk ya the ansible change I ntroduced also suffered this transient fialure | 00:51 |
| *** newmember has quit IRC | 00:53 | |
| sdake | jascott1 spsurya is this work you are interested in - minus the fact that there is clearlly disagreemnet ont he topic :) | 00:53 |
| jascott1 | sure | 00:53 |
| sdake | jascott1 spsurya cool once you understand the gate, you can do anything with kolla-kubernetes | 00:53 |
| *** tonanhngo has joined #openstack-kolla | 00:53 | |
| jascott1 | maybe someone can give a deep dive or at least overview of the finer points | 00:54 |
| *** newmember has joined #openstack-kolla | 00:54 | |
| sdake | jascott1 when would you like? | 00:54 |
| sdake | jascott1 avialalbe now | 00:54 |
| sdake | spsurya ? | 00:54 |
| spsurya | here | 00:54 |
| spsurya | please start | 00:54 |
| sdake | jascott1 had asked for a deep dive on the gate, i think that would helpy ou too spsurya if your interested in contributing | 00:54 |
| jascott1 | ive got a hard stop in about 30 | 00:54 |
| jascott1 | but yes | 00:55 |
| sdake | webex? | 00:55 |
| sdake | http://cisco.webex.com/join/stdake | 00:55 |
| jascott1 | sure | 00:55 |
| jascott1 | one sec | 00:55 |
| sdake | 30 minutes wont be much of a deep dive | 00:55 |
| *** tonanhngo_ has joined #openstack-kolla | 00:56 | |
| jascott1 | idk part1 ? | 00:56 |
| sdake | yup part 1 | 00:56 |
| *** tonanhngo has quit IRC | 00:58 | |
| *** tonanhngo_ has quit IRC | 01:01 | |
| *** cuongnv has joined #openstack-kolla | 01:02 | |
| sbezverk | sdake: I need to learn from you your way of giving deep dives ;) | 01:02 |
| *** tovin07_ has joined #openstack-kolla | 01:02 | |
| sdake | damnit webex crashed | 01:08 |
| sdake | pile o garb | 01:08 |
| *** StephenWang1991 has joined #openstack-kolla | 01:09 | |
| *** StephenWang1991 has quit IRC | 01:13 | |
| *** tervancovan has joined #openstack-kolla | 01:13 | |
| *** MasterOfBugs has quit IRC | 01:16 | |
| *** zhubingbing has joined #openstack-kolla | 01:27 | |
| *** hrw has quit IRC | 01:30 | |
| *** hrw has joined #openstack-kolla | 01:32 | |
| sdake | sbezverk you are correct - doing what kfox1111 wants is difficult | 01:35 |
| sdake | sbezverk we do have the time to do it | 01:35 |
| sdake | eanylin spsurya jascott1 thanks for attending | 01:35 |
| *** tonanhngo has joined #openstack-kolla | 01:36 | |
| spsurya | sdake nice | 01:37 |
| *** MasterOfBugs has joined #openstack-kolla | 01:40 | |
| *** negronjl_ has joined #openstack-kolla | 01:42 | |
| *** al498u_ has quit IRC | 01:42 | |
| sdake | thanks for the ack sbezverk | 01:44 |
| *** tonanhngo has quit IRC | 01:45 | |
| *** StephenWang1991 has joined #openstack-kolla | 01:47 | |
| *** eanylin has quit IRC | 01:47 | |
| *** britthouser has quit IRC | 01:49 | |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module https://review.openstack.org/457163 | 01:49 |
| *** britthouser has joined #openstack-kolla | 01:49 | |
| sdake | sup britthouser | 01:50 |
| *** iceyao has joined #openstack-kolla | 01:51 | |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module https://review.openstack.org/457163 | 01:52 |
| *** ssurana has joined #openstack-kolla | 01:54 | |
| *** iceyao has quit IRC | 02:03 | |
| *** tonanhngo has joined #openstack-kolla | 02:04 | |
| *** iceyao has joined #openstack-kolla | 02:04 | |
| *** Margin has quit IRC | 02:05 | |
| *** unicell has quit IRC | 02:08 | |
| *** tonanhngo has quit IRC | 02:11 | |
| *** super has joined #openstack-kolla | 02:12 | |
| *** tonanhngo has joined #openstack-kolla | 02:16 | |
| *** eanylin has joined #openstack-kolla | 02:20 | |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module in glance_base image https://review.openstack.org/457163 | 02:23 |
| *** tonanhngo has quit IRC | 02:28 | |
| *** StephenW_ has joined #openstack-kolla | 02:30 | |
| *** strigazi has quit IRC | 02:30 | |
| *** strigazi has joined #openstack-kolla | 02:31 | |
| *** lujinluo has joined #openstack-kolla | 02:32 | |
| *** StephenWang1991 has quit IRC | 02:33 | |
| *** tonanhngo has joined #openstack-kolla | 02:35 | |
| *** mbruzek has joined #openstack-kolla | 02:36 | |
| *** japestinho has joined #openstack-kolla | 02:37 | |
| *** iceyao has quit IRC | 02:49 | |
| *** schwicht has quit IRC | 02:53 | |
| *** zhurong has quit IRC | 03:03 | |
| *** lamt has joined #openstack-kolla | 03:04 | |
| *** tonyb has quit IRC | 03:04 | |
| *** dave-mccowan has quit IRC | 03:05 | |
| *** iceyao has joined #openstack-kolla | 03:06 | |
| *** sudipto has joined #openstack-kolla | 03:08 | |
| *** sudipto_ has joined #openstack-kolla | 03:08 | |
| *** zhurong has joined #openstack-kolla | 03:12 | |
| *** tonanhngo has quit IRC | 03:15 | |
| *** gkadam has joined #openstack-kolla | 03:15 | |
| *** gkadam is now known as gkadam-brb | 03:15 | |
| *** lamt has quit IRC | 03:18 | |
| *** lamt has joined #openstack-kolla | 03:19 | |
| *** iceyao has quit IRC | 03:20 | |
| *** iceyao has joined #openstack-kolla | 03:21 | |
| *** mbruzek has quit IRC | 03:31 | |
| *** StephenWang1991 has joined #openstack-kolla | 03:35 | |
| *** jtriley has quit IRC | 03:35 | |
| *** iceyao has quit IRC | 03:35 | |
| *** iceyao has joined #openstack-kolla | 03:36 | |
| *** StephenW_ has quit IRC | 03:37 | |
| *** iceyao has quit IRC | 03:42 | |
| *** lamt has quit IRC | 03:45 | |
| *** gkadam-brb is now known as gkadam | 03:46 | |
| *** lamt has joined #openstack-kolla | 03:52 | |
| *** zhurong has quit IRC | 04:00 | |
| *** lamt has quit IRC | 04:00 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Fix 'node_config_directory' variable bad usage https://review.openstack.org/456588 | 04:02 |
| *** iceyao has joined #openstack-kolla | 04:03 | |
| *** lamt has joined #openstack-kolla | 04:07 | |
| *** iceyao has quit IRC | 04:08 | |
| duonghq | mnaser, ping | 04:08 |
| spsurya | duonghq: can you please set +w https://review.openstack.org/#/c/455866/ | 04:09 |
| *** lamt has quit IRC | 04:13 | |
| *** lamt has joined #openstack-kolla | 04:14 | |
| *** unicell has joined #openstack-kolla | 04:15 | |
| *** lamt has quit IRC | 04:15 | |
| *** unicell has quit IRC | 04:16 | |
| *** zhurong has joined #openstack-kolla | 04:18 | |
| *** unicell has joined #openstack-kolla | 04:19 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 04:20 |
| duonghq | spsurya, we cannot link to neighbor file? | 04:21 |
| *** zhurong has quit IRC | 04:25 | |
| spsurya | duonghq: you mean this neigbour file https://github.com/openstack/kolla-ansible/blob/master/doc/vagrant-dev-env.rst | 04:26 |
| openstackgerrit | Merged openstack/kolla master: Updated from global requirements https://review.openstack.org/455928 | 04:26 |
| spsurya | duonghq: ? | 04:26 |
| spsurya | btw the link i provided also fine | 04:28 |
| *** zhubingbing has quit IRC | 04:28 | |
| spsurya | IMO | 04:28 |
| duonghq | spsurya, mins, I'm reading rst manual | 04:29 |
| duonghq | :) | 04:29 |
| duonghq | sorry, need making sure about smt | 04:29 |
| *** unicell has quit IRC | 04:34 | |
| *** unicell has joined #openstack-kolla | 04:35 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Switch kolla_docker to rely on SHA256 for image changes https://review.openstack.org/454265 | 04:35 |
| *** chas has joined #openstack-kolla | 04:53 | |
| *** chas has quit IRC | 04:53 | |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module in glance_base image https://review.openstack.org/457163 | 04:53 |
| *** chas has joined #openstack-kolla | 04:54 | |
| openstackgerrit | Merged openstack/kolla master: Use ironic-inspector user for ironic-inspector https://review.openstack.org/441605 | 04:55 |
| *** chas has quit IRC | 04:58 | |
| *** zhurong has joined #openstack-kolla | 05:00 | |
| *** bmace has quit IRC | 05:12 | |
| *** bmace has joined #openstack-kolla | 05:13 | |
| *** sudipto_ has quit IRC | 05:18 | |
| *** sudipto has quit IRC | 05:18 | |
| *** sudipto_ has joined #openstack-kolla | 05:19 | |
| *** sudipto has joined #openstack-kolla | 05:19 | |
| *** jrobinson has quit IRC | 05:19 | |
| *** sudipto_ has quit IRC | 05:19 | |
| *** sudipto has quit IRC | 05:19 | |
| *** zhubingbing_ has joined #openstack-kolla | 05:21 | |
| *** skramaja has joined #openstack-kolla | 05:26 | |
| portdirect | jascott1: nice work re the 'helm init --net-host' cool stuff :) | 05:30 |
| jascott1 | hey! whats up? | 05:31 |
| jascott1 | i like that it was ~5 lines total iirc | 05:31 |
| portdirect | yeah - I played around with running outside the cluster to achive the same effect - but having it in is really nice | 05:32 |
| SamYaple | portdirect-- the man who never sleeps | 05:32 |
| portdirect | for installing proper networking backends is super useful | 05:33 |
| portdirect | SamYaple: we meet again. | 05:33 |
| jascott1 | what up SamYaple! | 05:33 |
| jascott1 | yeah i need to make a proper PR, going to add selectors and tolerations looks like | 05:33 |
| *** iceyao has joined #openstack-kolla | 05:47 | |
| *** zhugaoxiao has quit IRC | 05:56 | |
| SamYaple | yo jascott1. hows it going. still being awesome? | 06:00 |
| *** Serlex has joined #openstack-kolla | 06:01 | |
| *** cuongnv has quit IRC | 06:02 | |
| *** cuongnv has joined #openstack-kolla | 06:02 | |
| jascott1 | haha i try to for a few minutes everyday | 06:07 |
| *** tonanhngo has joined #openstack-kolla | 06:07 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 06:09 |
| *** dvx76 has joined #openstack-kolla | 06:14 | |
| *** matrohon has joined #openstack-kolla | 06:19 | |
| *** iceyao has quit IRC | 06:22 | |
| *** sayantan_ has quit IRC | 06:23 | |
| *** sudipto_ has joined #openstack-kolla | 06:26 | |
| *** sudipto has joined #openstack-kolla | 06:26 | |
| *** imcsk8 has quit IRC | 06:31 | |
| *** tonanhngo has quit IRC | 06:32 | |
| *** iceyao has joined #openstack-kolla | 06:35 | |
| openstackgerrit | jeck proposed openstack/kolla-ansible master: Removing swift references from fstab https://review.openstack.org/456805 | 06:37 |
| *** sayantan_ has joined #openstack-kolla | 06:43 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Grafana: automatically set InfuxDB datasource https://review.openstack.org/457094 | 06:45 |
| tervancovan | h all | 06:46 |
| tervancovan | hi all | 06:46 |
| tervancovan | ls | 06:48 |
| *** sayantan_ has quit IRC | 06:48 | |
| openstackgerrit | shaofeng cheng proposed openstack/kolla-ansible master: Add VMware DataStore support to glance https://review.openstack.org/452176 | 06:49 |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible stable/ocata: Fix 'node_config_directory' variable bad usage https://review.openstack.org/457477 | 06:50 |
| *** bjolo has joined #openstack-kolla | 06:50 | |
| bjolo | morning | 06:50 |
| tervancovan | I have configured kolla-kubernetes environment using kubeadm-aio | 06:52 |
| tervancovan | and I think that there is more settings for launching kolla-kubernetes in kubeadm-aio | 06:54 |
| *** manheim has joined #openstack-kolla | 06:54 | |
| tervancovan | Is it right? | 06:55 |
| *** bjolo has quit IRC | 06:56 | |
| *** bjolo has joined #openstack-kolla | 06:57 | |
| *** pcaruana has joined #openstack-kolla | 06:59 | |
| *** tonanhngo has joined #openstack-kolla | 07:01 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Simplify Openstack services configuration possibilities https://review.openstack.org/456688 | 07:02 |
| *** tonanhngo has quit IRC | 07:02 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Cleanup /run/netns host directory volume mount https://review.openstack.org/454179 | 07:07 |
| *** sayantan_ has joined #openstack-kolla | 07:09 | |
| *** shardy has joined #openstack-kolla | 07:09 | |
| *** Serlex has quit IRC | 07:13 | |
| openstackgerrit | Duong Ha-Quang proposed openstack/kolla-kubernetes master: Heat service deployment https://review.openstack.org/437252 | 07:15 |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Generalize api_interface_address variable usage https://review.openstack.org/457479 | 07:15 |
| *** chas has joined #openstack-kolla | 07:23 | |
| *** magicboiz has joined #openstack-kolla | 07:29 | |
| openstackgerrit | Javier Castillo proposed openstack/kolla-ansible master: Fix incorrect fluend destination address https://review.openstack.org/457262 | 07:31 |
| *** egonzalez has joined #openstack-kolla | 07:38 | |
| *** haplo37- has quit IRC | 07:39 | |
| *** haplo37_ has joined #openstack-kolla | 07:41 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla master: Fix zun images https://review.openstack.org/456251 | 07:44 |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible stable/ocata: Multi-regions Support https://review.openstack.org/457494 | 07:49 |
| egonzalez | zhubingbing_, thanks for zun logs fix ;) | 07:50 |
| zhubingbing_ | ；） | 07:50 |
| zhubingbing_ | https://review.openstack.org/#/c/457494/ <egonzalez | 07:51 |
| zhubingbing_ | please review it | 07:51 |
| *** jaosorior has joined #openstack-kolla | 08:03 | |
| *** ssurana has quit IRC | 08:05 | |
| *** ssurana has joined #openstack-kolla | 08:08 | |
| *** zhurong has quit IRC | 08:09 | |
| *** ssurana has quit IRC | 08:10 | |
| spsurya | zhubingbing_: http://paste.openstack.org/show/606938/ \ | 08:11 |
| spsurya | can you please check | 08:11 |
| spsurya | have faced this issue | 08:11 |
| spsurya | kk8s | 08:11 |
| zhubingbing_ | hmm | 08:12 |
| spsurya | so what is happening ? | 08:12 |
| spsurya | what could be the problem in this ? | 08:12 |
| zhubingbing_ | k8s 1.6.0 have some problems | 08:13 |
| spsurya | regarding what ? will please elaborate | 08:13 |
| *** iceyao has quit IRC | 08:13 | |
| spsurya | will you* | 08:13 |
| zhubingbing_ | i thinks you should install 1.5.4 | 08:13 |
| spsurya | zhubingbing_: but we are on the rack of 1.6.1 | 08:14 |
| openstackgerrit | zhangdongqi proposed openstack/kolla-ansible master: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is neutron_vpnaas_agent.volumes not neutron_vpnaas_agent.image https://review.openstack.org/457505 | 08:14 |
| spsurya | you already know, i think | 08:14 |
| *** iceyao has joined #openstack-kolla | 08:15 | |
| zhubingbing_ | wait | 08:15 |
| zhubingbing_ | i give you a link | 08:15 |
| *** iceyao has quit IRC | 08:17 | |
| spsurya | zhubingbing_: sure | 08:17 |
| *** iceyao has joined #openstack-kolla | 08:18 | |
| *** jascott1 has quit IRC | 08:19 | |
| *** MasterOfBugs has quit IRC | 08:19 | |
| *** jascott1 has joined #openstack-kolla | 08:19 | |
| hrw | morning | 08:20 |
| spsurya | hrw: morning | 08:21 |
| zhubingbing_ | https://github.com/kubernetes/kubernetes/issues/43815 | 08:21 |
| hrw | time to check patches and refresh what is needed | 08:21 |
| zhubingbing_ | spsurya | 08:21 |
| *** dmk0202 has joined #openstack-kolla | 08:22 | |
| spsurya | zhubingbing_: seems issue has been closed | 08:22 |
| spsurya | ;) | 08:23 |
| *** jascott1 has quit IRC | 08:24 | |
| *** iceyao has quit IRC | 08:26 | |
| *** zhurong has joined #openstack-kolla | 08:26 | |
| openstackgerrit | Chen proposed openstack/kolla master: Use LABEL instead of MAINTAINER (deprecated) in Dockerfile.j2 https://review.openstack.org/457513 | 08:27 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: Support non-x86 architectures (aarch64, ppc64le) https://review.openstack.org/430940 | 08:31 |
| *** openstackgerrit has quit IRC | 08:33 | |
| *** openstackgerrit has joined #openstack-kolla | 08:33 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Add trustee_domain_id in magnum.conf https://review.openstack.org/448438 | 08:33 |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Add trustee_domain_id in magnum.conf https://review.openstack.org/448438 | 08:35 |
| magicboiz | good morning | 08:36 |
| egonzalez | magicboiz, sup | 08:36 |
| magicboiz | good thx! | 08:36 |
| magicboiz | egonzalez, any experience with trove? | 08:36 |
| egonzalez | nope | 08:37 |
| magicboiz | :( | 08:37 |
| openstackgerrit | zhangdongqi proposed openstack/kolla-ansible master: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is wrong. https://review.openstack.org/457505 | 08:37 |
| magicboiz | would you happen to know someone here?? :) | 08:38 |
| egonzalez | magicboiz, post the question and maybe someone can answer | 08:40 |
| magicboiz | well, I have deployed trove in (ocata/ubuntu/source/tls) and I'm facing a trove strange error: I cannot create a trove instance, I'm getting these kind of messages: Unable to establish connection to http://0.0.0.0:5000/v2.0/tokens: HTTPConnectionPool(host='0.0.0.0', port=5000): Max retries exceeded with url: /v2.0/tokens (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection ob... | 08:41 |
| *** Serlex has joined #openstack-kolla | 08:42 | |
| magicboiz | digging into logs, I see that trove is asking Nova for an instance-id "None" which is wrong...."GET call to compute for http://192.168.12.254:8774/v2/b89517e0f2bb4561b6e2cd2fb03b66cf/servers/None" | 08:42 |
| magicboiz | ERROR trove.instance.models [-] Could not find nova server_id(None). | 08:43 |
| magicboiz | it's strange. I cannot even delete failed trove instances.....¿? | 08:43 |
| egonzalez | magicboiz, may be trove bug https://bugs.launchpad.net/trove/+bug/1671655 | 08:47 |
| openstack | Launchpad bug 1671655 in OpenStack DBaaS (Trove) "trove_auth_url is not populated properly everywhere" [Medium,In progress] - Assigned to Erik Olof Gunnar Andersson (eandersson) | 08:47 |
| egonzalez | OSA have the same issue and are pointing to trove bugs https://bugs.launchpad.net/openstack-ansible/+bug/1647681 | 08:47 |
| openstack | Launchpad bug 1647681 in openstack-ansible "trove error during publish_exists_event" [Undecided,In progress] - Assigned to Donovan Francesco (donovan-francesco) | 08:47 |
| magicboiz | egonzalez, like always, thanks! | 08:50 |
| magicboiz | https://review.openstack.org/#/c/455463/ | 08:50 |
| openstackgerrit | zhangdongqi proposed openstack/kolla-ansible master: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is wrong. https://review.openstack.org/457505 | 08:51 |
| *** newmember has quit IRC | 08:53 | |
| *** mgoddard has joined #openstack-kolla | 08:53 | |
| *** newmember has joined #openstack-kolla | 08:54 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: Support non-x86 architectures (aarch64, ppc64le) https://review.openstack.org/430940 | 08:56 |
| *** sayantan_ has quit IRC | 09:00 | |
| *** sayantan_ has joined #openstack-kolla | 09:00 | |
| *** gfidente has joined #openstack-kolla | 09:01 | |
| *** gfidente has quit IRC | 09:01 | |
| *** gfidente has joined #openstack-kolla | 09:01 | |
| openstackgerrit | Duong Ha-Quang proposed openstack/kolla-kubernetes master: Heat service deployment https://review.openstack.org/437252 | 09:03 |
| *** derekjhyang has quit IRC | 09:03 | |
| *** sayantan_ has quit IRC | 09:06 | |
| *** Margin has joined #openstack-kolla | 09:10 | |
| *** sambetts|afk is now known as sambetts | 09:19 | |
| *** guoguo has quit IRC | 09:26 | |
| sdake | tervancovan need some help? | 09:27 |
| openstackgerrit | Merged openstack/kolla-ansible master: Cleanup /run/netns host directory volume mount https://review.openstack.org/454179 | 09:32 |
| tervancovan | yes sdake | 09:34 |
| tervancovan | i need some help~! | 09:34 |
| sdake | tervancovan tryign to deploy kolla-kubernetes? | 09:35 |
| tervancovan | yes right | 09:35 |
| sdake | tervancovan ran through https://docs.openstack.org/developer/kolla-kubernetes/deployment-guide.html | 09:35 |
| tervancovan | So I have checked deployment-guide.rst in kolla-kuberenetes | 09:36 |
| sdake | ok - i guess itwould help if the steps were numbered | 09:36 |
| sdake | which step are you on - step 1 or 2 or 3? | 09:36 |
| *** chas has quit IRC | 09:37 | |
| tervancovan | I am in step 2 | 09:37 |
| sdake | validating kubernetes? | 09:37 |
| tervancovan | yes right | 09:37 |
| sdake | can you paste the output you see on your screen during that nslookup operation? | 09:37 |
| *** chas has joined #openstack-kolla | 09:37 | |
| tervancovan | https://thepasteb.in/p/Vmh03Q0vxqXh4 | 09:40 |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module in glance_base image https://review.openstack.org/457163 | 09:41 |
| *** chas has quit IRC | 09:42 | |
| spsurya | sdake: for DNS configuration with CIDR what should i put in /etc/systemd/system/kubelet.service.d/10-kubeadm.conf if my centos 7 VM ip is 192.168.121.45/24 ? | 09:42 |
| *** zhurong has quit IRC | 09:42 | |
| sdake | spsurya in that case, the pod CIDR and service CIDRs need not be modified | 09:43 |
| sdake | the DNS server comes from the service CIDR | 09:43 |
| openstackgerrit | shaofeng cheng proposed openstack/kolla master: Add oslo-vmware module in glance_base image https://review.openstack.org/457163 | 09:43 |
| sdake | Warning This documentation assumes a POD CIDR of 10.1.0.0/16 and a service CIDR of 10.3.3.0/24. Two rules must be followed when reading this guide. | 09:44 |
| sdake | The service and pod cidr cannot overlap | 09:44 |
| sdake | The address spaces cannot already be allocated by your organization | 09:44 |
| sdake | If the POD and CIDR addresses overlap in this documentation with your organizations’s IP address ranges, they may be changed. Simply subtitute anywhere these addresses are used with the custom cidrs you hae chosen. | 09:44 |
| sdake | in this case since your a vm, the pod and service cidrs never overlap and are not allocated by your org | 09:44 |
| sdake | your in a vm | 09:44 |
| spsurya | roger | 09:45 |
| sdake | tervancovan you did not follow the instructions :) | 09:51 |
| sdake | tervancovan the kubeadm defaults are 10.9616 fro the service cidr | 09:51 |
| sdake | 10.96/16 for th eservice cidr | 09:51 |
| sdake | tervancovan i suspect you didn't change the dns server | 09:51 |
| sdake | tervancovan please cat /etc/systemd/system/kubelet.service.d/10-kubeadm.conf | 09:52 |
| sdake | -> paste | 09:53 |
| *** skramaja has quit IRC | 09:56 | |
| *** sayantan_ has joined #openstack-kolla | 10:02 | |
| *** shardy has quit IRC | 10:06 | |
| *** sayantan_ has quit IRC | 10:07 | |
| *** Margin has quit IRC | 10:10 | |
| *** cuongnv has quit IRC | 10:12 | |
| spsurya | sdake: http://paste.openstack.org/show/606990/ | 10:12 |
| spsurya | cleaned many times | 10:12 |
| *** tovin07_ has quit IRC | 10:13 | |
| *** asmaa has joined #openstack-kolla | 10:13 | |
| sdake | spsurya -> [root@localhost vagrant]# CGROUP_DRIVER=$(sudo docker info | grep "Cgroup Driver" | awk '{print $3}') | 10:14 |
| sdake | you see this command is failing | 10:14 |
| sdake | spsurya prior to that operation | 10:14 |
| sdake | (the docs may be wrong) | 10:15 |
| sdake | spsurya can you clean up | 10:15 |
| sdake | and stop prior to that CGRUP_DRIVER operation plz | 10:15 |
| sdake | and wait for further instruction | 10:15 |
| spsurya | roger | 10:15 |
| sdake | Cannot connect to the Docker daemon. Is the docker daemon running on this host? | 10:15 |
| sdake | this is how i know there is an error | 10:15 |
| sdake | first error is what we need to debug | 10:15 |
| sdake | I know exacty what the problem is | 10:15 |
| spsurya | sdake once i installed docker with `curl -sSL https://get.docker.io | bash` | 10:18 |
| *** chas has joined #openstack-kolla | 10:18 | |
| sdake | spsurya - not correcct | 10:18 |
| sdake | spsurya please undo that | 10:18 |
| *** chas has quit IRC | 10:19 | |
| *** chas_ has joined #openstack-kolla | 10:19 | |
| sdake | upstream docker does not work with kubeadm | 10:19 |
| spsurya | yes | 10:19 |
| sdake | spsurya reset to clean environment, then stop prior to the CGRUOP_DRIVER operation | 10:20 |
| spsurya | sure | 10:21 |
| openstackgerrit | Vladislav Belogrudov proposed openstack/kolla-ansible master: Allow use of any backup backend for any volume backend https://review.openstack.org/457554 | 10:21 |
| sdake | spsurya when you get here -> sudo yum install -y docker ebtables kubeadm kubectl kubelet kubernetes-cni | 10:22 |
| sdake | stop | 10:22 |
| sdake | spsurya inbetween 63 and 69, i don't see that you have run one command as specified the deployment guide | 10:22 |
| sdake | actually two commands | 10:23 |
| *** sayantan_ has joined #openstack-kolla | 10:23 | |
| spsurya | moment | 10:24 |
| spsurya | destroying and vagrant up | 10:24 |
| *** yangyapeng has joined #openstack-kolla | 10:24 | |
| sdake | k | 10:24 |
| spsurya | now trying from fresh | 10:24 |
| sdake | spsurya would you be kindenough to share our vagrantfile | 10:24 |
| sdake | our/your | 10:24 |
| spsurya | sure | 10:25 |
| sdake | i'll try on my end | 10:25 |
| *** skramaja has joined #openstack-kolla | 10:25 | |
| sdake | tervancovan where u at dude ;) | 10:27 |
| *** sayantan_ has quit IRC | 10:27 | |
| spsurya | sdake: vagrant file http://paste.openstack.org/show/606991/ | 10:28 |
| *** shardy has joined #openstack-kolla | 10:29 | |
| spsurya | sdake: now at --->>> sudo cp -a kubernetes.repo /etc/yum.repos.d | 10:31 |
| spsurya | now at * sudo yum install -y docker ebtables kubeadm kubectl kubelet kubernetes-cni | 10:32 |
| tervancovan | Thanks sdake | 10:35 |
| *** rmart04 has joined #openstack-kolla | 10:35 | |
| sdake | tervancovan did you get running? | 10:35 |
| sdake | spsurya ok - now after that step - what is the enxt step :) | 10:35 |
| *** chas has joined #openstack-kolla | 10:36 | |
| tervancovan | Not yet I am just back from dinner | 10:36 |
| *** gfidente has quit IRC | 10:36 | |
| tervancovan | So The thing that I have to do is fix configuration in docker container of kubeadm-aio | 10:37 |
| sdake | tervancovan ok - well looking at your output, I notice your dns is not configured correctly | 10:37 |
| tervancovan | For kolla-kubernetes CIDR rule | 10:37 |
| sdake | tervancovan which would indicate a direction was skipped ;) | 10:37 |
| tervancovan | yes right | 10:37 |
| *** duonghq has quit IRC | 10:38 | |
| spsurya | sdake: after that this one --->>> CGROUP_DRIVER=$(sudo docker info | grep "Cgroup Driver" | awk '{print $3}') | 10:38 |
| *** chas_ has quit IRC | 10:38 | |
| sdake | spsurya no- prior to that step | 10:39 |
| sdake | spsurya click "RELOAD" on the web browser | 10:39 |
| sdake | spsurya it is possibel you ahve old docs cached | 10:39 |
| spsurya | sdake: you are right | 10:41 |
| spsurya | was at old one ;) | 10:41 |
| *** chas_ has joined #openstack-kolla | 10:41 | |
| sdake | spsurya would you submit a docs change and add a warning tag at the top that says "these docs are under rpaid change. Please use the reload button before running through these instructions" | 10:42 |
| sdake | or something similar | 10:42 |
| sdake | spsurya there is already a warning the docs dont'work | 10:42 |
| sdake | I think that canbe removed as the docs DO work | 10:42 |
| sdake | or altered | 10:42 |
| spsurya | sdake sure | 10:43 |
| spsurya | i will | 10:43 |
| sdake | tervancovan this may apply toy ou as well | 10:43 |
| *** chas_ has quit IRC | 10:43 | |
| *** chas has quit IRC | 10:44 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Cleanup /lib/modules volume mounts in containers https://review.openstack.org/454214 | 10:45 |
| tervancovan | sdake I think this command should be applyed --->"sudo kubeadm init --pod-network-cidr=10.1.0.0/16 --service-cidr=10.3.3.0/24" | 10:47 |
| spsurya | sdake: should we be ok with the warning Usage of loopback devices ? | 10:47 |
| sdake | spsurya i dont follow :) | 10:48 |
| sdake | tervancovan ithat is what the documents recommend | 10:49 |
| spsurya | sdake i mean http://paste.openstack.org/show/606992/ | 10:49 |
| sdake | spsurya ya - we can ignore that for now | 10:49 |
| sdake | spsurya that is mainly a warning not an error | 10:50 |
| *** jaosorior has quit IRC | 10:55 | |
| tervancovan | root@hosoup-ubuntu-64:~# kubeadm init --pod-network-cidr=10.1.0.0/16 --service-cidr=10.3.3.0/24 | 10:55 |
| tervancovan | [kubeadm] WARNING: kubeadm is in beta, please do not use it for production clusters. | 10:55 |
| tervancovan | [init] Using Kubernetes version: v1.6.0 | 10:55 |
| tervancovan | [init] Using Authorization mode: RBAC | 10:55 |
| tervancovan | [preflight] Running pre-flight checks | 10:55 |
| *** tervancovan has quit IRC | 10:55 | |
| spsurya | sdake worked :) | 10:55 |
| sdake | spsurya good news :) | 10:55 |
| sdake | spsurya which step, step 2? | 10:56 |
| spsurya | at least got of that waiting | 10:56 |
| sdake | or step 1 | 10:56 |
| openstackgerrit | jeck proposed openstack/kolla-ansible master: Remove /etc/fstab Swift references during cleanup-host https://review.openstack.org/456805 | 10:56 |
| *** tervancovan has joined #openstack-kolla | 10:56 | |
| spsurya | sdake: 1 | 10:56 |
| sdake | spsurya on my gige internet that step takes about 30 seconds | 10:57 |
| sdake | spsurya YMMV - would be interesteed to know what i ttakes for you | 10:57 |
| sdake | kubeadm is downloading a slew of containers | 10:57 |
| spsurya | sdake now onwards it will take 29secs for me ;) | 10:57 |
| sdake | spsurya cool | 10:58 |
| sdake | spsurya can you submit that warning about the "click refresh on browser" when you ahve an opportunity | 10:58 |
| spsurya | yes for sure | 10:58 |
| spsurya | sdake | 10:58 |
| sdake | spsurya thanks :) | 10:59 |
| tervancovan | This is output of this command in container -->" sudo kubeadm init --pod-network-cidr=10.1.0.0/16 --service-cidr=10.3.3.0/24 | 10:59 |
| tervancovan | " https://thepasteb.in/p/JZhpZMQZZMXhg | 10:59 |
| *** jaosorior has joined #openstack-kolla | 11:01 | |
| sdake | tervancovan run kubeadm reset | 11:05 |
| sdake | tervancovan you can't kubeadm followed by kubeadm without a reset inbetween | 11:05 |
| tervancovan | thank you sdake | 11:05 |
| sdake | tervancovan your DNS server was not set proeprly in kubeadm | 11:05 |
| sdake | tervancovan so - if you could please paste the output of : | 11:05 |
| openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-kubernetes master: Updated the warning message of deployment guide https://review.openstack.org/457571 | 11:05 |
| sdake | etc/systemd/system/kubelet.service.d/10-kubeadm.conf | 11:06 |
| *** shardy is now known as shardy_lunch | 11:06 | |
| tervancovan | this is outout of 10-kubeadm.conf https://thepasteb.in/p/98hRQ7YXBMEsk | 11:07 |
| sdake | tervancovan looks good | 11:07 |
| sdake | tervancovan did you restart kubeadm | 11:07 |
| sdake | sorry | 11:07 |
| sdake | did you restartkubelet | 11:07 |
| *** gfidente has joined #openstack-kolla | 11:08 | |
| *** gfidente has quit IRC | 11:08 | |
| *** gfidente has joined #openstack-kolla | 11:08 | |
| sdake | sudo systemctl stop kubelet | 11:08 |
| sdake | sudo systemctl enable kubelet | 11:08 |
| sdake | sudo systemctl start kubelet | 11:08 |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible master: Update simple_cell_setup to manual creation https://review.openstack.org/456920 | 11:09 |
| tervancovan | "kubeadm reset" -> https://thepasteb.in/p/RghnEZgYAV7cz | 11:09 |
| sdake | tervancovan run the 3 above sudo commands please | 11:10 |
| *** zhurong has joined #openstack-kolla | 11:14 | |
| tervancovan | I already did it | 11:16 |
| tervancovan | btw I use "--skip-preflight-checks" this options to kubeadm init | 11:16 |
| sdake | tervancovan don't use that optoin | 11:17 |
| sdake | tervancovan follow the guide exactly | 11:17 |
| sdake | tervancovan do not deviate plz | 11:17 |
| spsurya | sdake http://paste.openstack.org/show/606998/ | 11:19 |
| *** chas has joined #openstack-kolla | 11:20 | |
| *** jascott1 has joined #openstack-kolla | 11:20 | |
| *** jascott1 has quit IRC | 11:25 | |
| sdake | spsurya this is what it should look like: | 11:26 |
| sdake | [vagrant@localhost ~]$ kubectl run -i -t $(uuidgen) --image=busybox --restart=Never | 11:26 |
| sdake | If you don't see a command prompt, try pressing enter. | 11:26 |
| sdake | Server: 10.3.3.10 | 11:26 |
| sdake | Address 1: 10.3.3.10 kube-dns.kube-system.svc.cluster.local | 11:26 |
| sdake | Name: kubernetes | 11:26 |
| sdake | Address 1: 10.3.3.1 kubernetes.default.svc.cluster.local | 11:26 |
| sdake | note I just run with your vagrant setup | 11:26 |
| sdake | and it wfm | 11:26 |
| sdake | spsurya the error you ahve is indicative of not changing the dns server | 11:27 |
| sdake | spsurya run kubectl get nodes -> paste | 11:27 |
| sdake | after typing exit a few time to get out of that busybox image | 11:27 |
| spsurya | https://www.irccloud.com/pastebin/LnOD0u2I/ | 11:28 |
| spsurya | sdake | 11:29 |
| sdake | spsurya ok kubeadm get pods -n kube-system -> paste | 11:29 |
| sdake | rathe kubectl get pods -n kube-system -> paste | 11:29 |
| spsurya | https://www.irccloud.com/pastebin/Emvvvwsh/ | 11:30 |
| sdake | dns 3/3 | 11:30 |
| sdake | looks good | 11:30 |
| spsurya | seems fine too me | 11:30 |
| sdake | ok - cat /etc/systemd/system/kubelet.service.d/10-kubeadm.conf | 11:30 |
| sdake | Note Kubernetes uses x.x.x.10 as the DNS server. The Kolla developers don’t know precisely | 11:30 |
| spsurya | Environment="KUBELET_DNS_ARGS=--cluster-dns=10.3.3.10 --cluster-domain=cluster.local" | 11:32 |
| spsurya | correct | 11:32 |
| sdake | ok - that looks good | 11:32 |
| sdake | did you restart kubelet? | 11:32 |
| spsurya | not yet | 11:32 |
| sdake | ok | 11:32 |
| sdake | well - | 11:32 |
| sdake | these instructions need to be followed in the order specified in the docs: | 11:33 |
| sdake | Then reload the hand-modified service files: | 11:33 |
| sdake | sudo systemctl daemon-reload | 11:33 |
| sdake | Then stop kubelet if it is running: | 11:33 |
| sdake | sudo systemctl stop kubelet | 11:33 |
| sdake | Then enable and start docker and kubelet: | 11:33 |
| sdake | sudo systemctl enable kubelet | 11:33 |
| sdake | sudo systemctl start kubelet | 11:33 |
| spsurya | do we need to update in doc too | 11:35 |
| spsurya | ? | 11:35 |
| sdake | that is what is in the odc | 11:35 |
| sdake | straight cut and paste from it | 11:36 |
| sdake | here is cut and past efrom the docs: | 11:36 |
| sdake | Setup the DNS server with the service CIDR: | 11:36 |
| sdake | sudo sed -i 's/10.96.0.10/10.3.3.10/g' /etc/systemd/system/kubelet.service.d/10-kubeadm.conf | 11:36 |
| sdake | Note Kubernetes uses x.x.x.10 as the DNS server. The Kolla developers don’t know precisely why this is the case, however, current speculation is that that 1..9 are reserved for future expansion of Kubernetes infrastructure services. | 11:36 |
| sdake | Then reload the hand-modified service files: | 11:36 |
| sdake | sudo systemctl daemon-reload | 11:36 |
| sdake | Then stop kubelet if it is running: | 11:36 |
| sdake | sudo systemctl stop kubelet | 11:36 |
| sdake | Then enable and start docker and kubelet: | 11:36 |
| sdake | sudo systemctl enable kubelet | 11:36 |
| sdake | sudo systemctl start kubelet | 11:36 |
| sdake | you did the sed operation | 11:36 |
| sdake | i am not convinced you did the daemon reload operation or stop kubelet or enable kubelet or start kubelet | 11:36 |
| spsurya | i did | 11:37 |
| sdake | ok | 11:38 |
| sdake | it could be a timing issue | 11:38 |
| sdake | i just ran through steps 1 and step2 using your vagrantfile | 11:38 |
| sdake | and it WFM | 11:38 |
| spsurya | nice | 11:38 |
| sdake | so lets do this | 11:39 |
| sdake | don't reset a clean environment | 11:39 |
| sdake | instead, run cleanup-k8s.sh | 11:39 |
| spsurya | may be you can take a look on this http://paste.openstack.org/show/606999/ | 11:40 |
| sdake | spsurya run a new busybox and try again | 11:42 |
| sdake | with the nslookup | 11:42 |
| sdake | spsurya i think you are not waiting for kubernetes to come up before running busybox | 11:43 |
| sdake | spsurya although I am not certain | 11:43 |
| sdake | spsurya ok - new recommendation | 11:44 |
| sdake | destroy the vagrant env | 11:44 |
| sdake | and do vagrant up again | 11:44 |
| sdake | vagrant ssh | 11:44 |
| sdake | and start the steps | 11:44 |
| spsurya | moment | 11:44 |
| sdake | prior to validation, stop | 11:44 |
| sdake | (e.g. stop at step 2) | 11:44 |
| sdake | i.e. rather | 11:44 |
| spsurya | now doing destroy and up | 11:45 |
| *** bjolo has quit IRC | 11:46 | |
| openstackgerrit | Martin André proposed openstack/kolla master: Ensure tests run in isolation https://review.openstack.org/457214 | 11:48 |
| pbourke | hi sdake, looks like I missed some msgs from you over the weekend? | 11:51 |
| sdake | pbourke just wondering how lifes treating ya :) | 11:52 |
| pbourke | sdake: pretty good :) just back from long easter weekend | 11:52 |
| *** rwallner has joined #openstack-kolla | 11:53 | |
| *** rwallner has quit IRC | 11:55 | |
| tervancovan | Thanks for helping sdake | 11:58 |
| tervancovan | I have to go home right now , and I need more trying for it | 11:59 |
| tervancovan | tomorrow | 11:59 |
| *** lujinluo has quit IRC | 12:01 | |
| *** rwallner has joined #openstack-kolla | 12:01 | |
| sdake | tervancovan enjoy | 12:06 |
| *** dave-mccowan has joined #openstack-kolla | 12:08 | |
| spsurya | sdake again tainted | 12:12 |
| spsurya | i.e step 1 | 12:12 |
| spsurya | now holding | 12:12 |
| sdake | spsurya run command kubectl get pods -n kube-system | 12:13 |
| spsurya | ran | 12:13 |
| sdake | into a past e:) | 12:13 |
| spsurya | all are good | 12:13 |
| sdake | into past eplz | 12:13 |
| *** goldyfruit has quit IRC | 12:14 | |
| spsurya | https://www.irccloud.com/pastebin/YYYgwapc/ | 12:14 |
| sdake | those are not all good | 12:14 |
| sdake | you need to wait for dns to be 3/3 RUNNING | 12:14 |
| sdake | then you are all good :) | 12:14 |
| sdake | you notice dns is 0/3 pending | 12:14 |
| spsurya | yeah | 12:14 |
| spsurya | how long it takes ;) | 12:15 |
| sdake | downloading images | 12:15 |
| spsurya | its 9:15 PM here | 12:15 |
| sdake | on my gige network it takes 2 minutes to bring up kubernetes | 12:15 |
| sdake | how fast is your network? | 12:16 |
| *** shardy_lunch is now known as shardy | 12:16 | |
| spsurya | your ntw is fabulous | 12:16 |
| sdake | ya 99$/mo | 12:16 |
| sdake | not a bad deal :) | 12:16 |
| spsurya | its office one | 12:16 |
| spsurya | now got the dns 3/3 | 12:16 |
| *** Margin has joined #openstack-kolla | 12:17 | |
| sdake | now run busybox | 12:17 |
| spsurya | https://www.irccloud.com/pastebin/cS5Jf8E2/ | 12:19 |
| spsurya | gotch :) | 12:19 |
| *** salv-orlando has joined #openstack-kolla | 12:19 | |
| spsurya | sdake | 12:20 |
| sdake | spsurya ok two actions | 12:23 |
| sdake | which are two patches | 12:23 |
| *** tervancovan has quit IRC | 12:23 | |
| sdake | 1) add what you learned regarding the fact that you need to refresh the page to th edocument warning at the top | 12:23 |
| sdake | 2) add what you learned regarding the fact that you need to wait for kubernetes to come up before proceeding | 12:23 |
| sdake | spsurya I'll provide fast reviews on these and get them through the review queue quickly | 12:24 |
| sdake | thanks :) | 12:24 |
| *** sayantan_ has joined #openstack-kolla | 12:24 | |
| spsurya | for sure sdake | 12:24 |
| spsurya | 1 done | 12:24 |
| spsurya | 2 getting it done | 12:24 |
| *** sayantan_ has quit IRC | 12:28 | |
| *** srwilkers has joined #openstack-kolla | 12:30 | |
| *** lamt has joined #openstack-kolla | 12:44 | |
| *** gkadam has quit IRC | 12:44 | |
| openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-kubernetes master: Deployment guide updated with few more information https://review.openstack.org/457614 | 12:44 |
| spsurya | sdake https://review.openstack.org/457571 https://review.openstack.org/457614 | 12:45 |
| *** zhurong has quit IRC | 12:45 | |
| spsurya | sdake thanks for continuos support | 12:46 |
| sdake | spsurya thank you :) | 12:46 |
| *** lamt has quit IRC | 12:46 | |
| spsurya | will start from 3rd step once reach home | 12:47 |
| spsurya | hope eyes will be open otherwise tomorrow | 12:47 |
| *** mbruzek has joined #openstack-kolla | 12:52 | |
| *** StephenWang1991 has quit IRC | 12:53 | |
| *** zhubingbing_ has quit IRC | 12:58 | |
| *** rwallner has quit IRC | 13:02 | |
| *** tonanhngo has joined #openstack-kolla | 13:03 | |
| *** sayantan_ has joined #openstack-kolla | 13:04 | |
| *** sudipto has quit IRC | 13:04 | |
| *** sudipto_ has quit IRC | 13:04 | |
| *** huzhengchuan has quit IRC | 13:06 | |
| *** jaosorior has quit IRC | 13:06 | |
| *** rwallner has joined #openstack-kolla | 13:07 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes master: Pinning OS_VOLUME_API_VERSION=2 https://review.openstack.org/457626 | 13:08 |
| *** tonanhngo has quit IRC | 13:08 | |
| *** goldyfruit has joined #openstack-kolla | 13:11 | |
| *** Pavo has joined #openstack-kolla | 13:12 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes master: Adding kolla_kubernetes_external_subnet var https://review.openstack.org/457627 | 13:13 |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla-ansible master: Use ansible from within the kolla_toolbox virtualenv https://review.openstack.org/435024 | 13:13 |
| sdake | spsurya i left reviews on your patches | 13:14 |
| sdake | spsurya i think they could use a small bit of tidying | 13:14 |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla master: Reparent kolla-toolbox from openstack-base https://review.openstack.org/435023 | 13:14 |
| *** jaosorior has joined #openstack-kolla | 13:14 | |
| *** dmsimard|afk is now known as dmsimard | 13:15 | |
| *** esharao has joined #openstack-kolla | 13:17 | |
| *** mbruzek has quit IRC | 13:17 | |
| *** jtriley has joined #openstack-kolla | 13:17 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla master: Update oraclelinux tag to new slim base image https://review.openstack.org/437329 | 13:20 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add fwaas extension in l3 agent when neutron fwaas is enabled https://review.openstack.org/457631 | 13:20 |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla-ansible master: Only start one glance_api when using file backend https://review.openstack.org/448654 | 13:20 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible stable/ocata: Fix qos rules are not applied on ocata with ovs https://review.openstack.org/457633 | 13:24 |
| *** pcaruana has quit IRC | 13:27 | |
| openstackgerrit | zhangdongqi proposed openstack/kolla-ansible master: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is wrong https://review.openstack.org/457505 | 13:35 |
| *** eanylin has quit IRC | 13:39 | |
| *** mbruzek has joined #openstack-kolla | 13:41 | |
| *** iceyao has joined #openstack-kolla | 13:47 | |
| *** lamt has joined #openstack-kolla | 13:49 | |
| *** schwicht has joined #openstack-kolla | 13:50 | |
| *** MarginHu2017 has joined #openstack-kolla | 13:53 | |
| *** al498u has joined #openstack-kolla | 13:55 | |
| *** Margin has quit IRC | 13:56 | |
| *** al498u_ has joined #openstack-kolla | 13:57 | |
| sdake | 0.6.0 is done and in the process of being tagged: https://launchpad.net/kolla-kubernetes/+milestone/0.6.0 | 14:00 |
| *** al498u has quit IRC | 14:01 | |
| *** lamt has quit IRC | 14:02 | |
| sdake | 0.7.0 in progress - deadline may 15th: https://launchpad.net/kolla-kubernetes/+milestone/0.7.0 | 14:03 |
| *** lamt has joined #openstack-kolla | 14:05 | |
| *** sudipto_ has joined #openstack-kolla | 14:08 | |
| *** sudipto has joined #openstack-kolla | 14:08 | |
| *** cuongnv has joined #openstack-kolla | 14:08 | |
| *** pcaruana has joined #openstack-kolla | 14:10 | |
| *** StephenWang1991 has joined #openstack-kolla | 14:13 | |
| *** sayantan_ has quit IRC | 14:13 | |
| openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes master: Version 0.7.0 https://review.openstack.org/457657 | 14:18 |
| *** sayantan_ has joined #openstack-kolla | 14:20 | |
| *** sayantan_ has quit IRC | 14:21 | |
| openstackgerrit | zhangdongqi proposed openstack/kolla-ansible master: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is wrong https://review.openstack.org/457505 | 14:23 |
| *** dave-mccowan has quit IRC | 14:24 | |
| *** MarginHu2017 has quit IRC | 14:25 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Use ironic-inspector user for ironic inspector https://review.openstack.org/445587 | 14:30 |
| *** unicell1 has joined #openstack-kolla | 14:34 | |
| *** huzhengchuan has joined #openstack-kolla | 14:34 | |
| *** unicell has quit IRC | 14:34 | |
| *** sudipto_ has quit IRC | 14:38 | |
| *** sudipto has quit IRC | 14:38 | |
| *** tovin07 has joined #openstack-kolla | 14:44 | |
| *** dave-mccowan has joined #openstack-kolla | 14:44 | |
| *** tovin07 has left #openstack-kolla | 14:46 | |
| *** iceyao has quit IRC | 14:47 | |
| *** MarginHu2017 has joined #openstack-kolla | 14:49 | |
| *** MarginHu2017 has quit IRC | 14:50 | |
| spsurya | sdake: ok | 14:50 |
| *** zhubingbing_ has joined #openstack-kolla | 14:53 | |
| *** manheim has quit IRC | 14:56 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Set fluentd output to elasticsearch to use elasticsearch_address https://review.openstack.org/447654 | 14:56 |
| *** mbruzek has quit IRC | 14:58 | |
| *** StephenWang1991 has quit IRC | 15:03 | |
| *** srwilkers has quit IRC | 15:03 | |
| *** srwilkers has joined #openstack-kolla | 15:06 | |
| *** blallau has joined #openstack-kolla | 15:07 | |
| *** Serlex has quit IRC | 15:08 | |
| *** StephenWang1991 has joined #openstack-kolla | 15:13 | |
| *** iceyao has joined #openstack-kolla | 15:14 | |
| sdake | kfox1111 portdirect srwilkers if any of you ahve a moment - this is good for merge: https://review.openstack.org/#/c/457048/ | 15:15 |
| *** skramaja has quit IRC | 15:16 | |
| *** StephenWang1991 has quit IRC | 15:17 | |
| sbezverk | it is already merged?!?!? | 15:17 |
| sbezverk | do you want double merge now? | 15:17 |
| *** StephenWang1991 has joined #openstack-kolla | 15:17 | |
| sbezverk | only one merge counts against stackalitics | 15:17 |
| sdake | sbezverk moment | 15:17 |
| sdake | sbezverk oh wierd - it wasn't merged a bit ago | 15:17 |
| sdake | sbezverk perhaps gerrit returned me the wrong results - or perhaps i didn't look at the correct review | 15:18 |
| sdake | sbezverk ya - that is wrong link | 15:18 |
| sbezverk | kfox1111: please ping me when you have a chance | 15:19 |
| *** iceyao has quit IRC | 15:19 | |
| sdake | sbezverk ya - wrong review - my bad: the correct one is: https://review.openstack.org/#/c/451495/ | 15:19 |
| sdake | sbezverk also - stackalytics can DIAF :) | 15:19 |
| openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-kubernetes master: Updated the warning message of deployment guide https://review.openstack.org/457571 | 15:20 |
| *** jascott1 has joined #openstack-kolla | 15:21 | |
| *** rmart04 has quit IRC | 15:22 | |
| *** manheim has joined #openstack-kolla | 15:22 | |
| *** huzhengchuan has quit IRC | 15:23 | |
| *** jascott1 has quit IRC | 15:25 | |
| *** manheim has quit IRC | 15:27 | |
| openstackgerrit | Merged openstack/kolla-kubernetes master: Updated the warning message of deployment guide https://review.openstack.org/457571 | 15:27 |
| *** tonanhngo has joined #openstack-kolla | 15:29 | |
| sbezverk | kolla-kubernetes cores, please check a couple of PS, minor issues found and addressed based on mutlinode cluster deployment: https://review.openstack.org/457626 and https://review.openstack.org/457627 | 15:31 |
| openstackgerrit | Merged openstack/kolla-kubernetes master: Nova-control service chart dependencies https://review.openstack.org/451495 | 15:33 |
| kfox1111 | morning. | 15:33 |
| kfox1111 | interesting.... whats required to support cinder v3? | 15:34 |
| *** chas has quit IRC | 15:34 | |
| *** chas has joined #openstack-kolla | 15:35 | |
| sdake | super kfox1111 | 15:37 |
| *** huzhengchuan has joined #openstack-kolla | 15:38 | |
| *** chas has quit IRC | 15:39 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla master: Reparent kolla-toolbox from openstack-base https://review.openstack.org/435023 | 15:39 |
| sdake | portdirect thanks | 15:41 |
| *** pcaruana has quit IRC | 15:41 | |
| sbezverk | kfox1111: cinder client when not pinned to v2 tries to find end poit cinderv3 | 15:43 |
| sbezverk | and fails since we do not support it | 15:43 |
| kfox1111 | its just an entry in keystone pointing back to the regular cinder server? or is it something more? | 15:44 |
| sbezverk | kfox1111: it is keystone end point, remember we have cinder and cinderv2 | 15:45 |
| sbezverk | now it appears we need cinderv3 | 15:45 |
| kfox1111 | so, the real fix is cp -a cinderv2/cinderv3 | sed 's/v2/v3/' ? | 15:45 |
| kfox1111 | fun. | 15:45 |
| sbezverk | volumev2 | 15:45 |
| sbezverk | no, we need to add volumev2 and volumev3 | 15:46 |
| kfox1111 | ah. that too. | 15:46 |
| sbezverk | I mean add only volumev3 | 15:46 |
| openstackgerrit | Merged openstack/kolla-kubernetes master: Adding kolla_kubernetes_external_subnet var https://review.openstack.org/457627 | 15:46 |
| sbezverk | kfox1111: are you busy now? I came accross really wierd issue in mutlinode, need a second pair of eyes.. | 15:46 |
| kfox1111 | yeah, I've got a bit. whats up? | 15:47 |
| sbezverk | kfox1111: check this out: http://paste.openstack.org/show/607020/ | 15:47 |
| *** matrohon has quit IRC | 15:47 | |
| sbezverk | kfox1111: remember we added a check to openvswitch-agent to wait for opensvwitch to be operations | 15:48 |
| kfox1111 | hmm... | 15:48 |
| kfox1111 | is this in gate, or on your own box? | 15:48 |
| *** salv-orl_ has joined #openstack-kolla | 15:49 | |
| sbezverk | kfox1111: it is my 5 nodes cluster I am building summit demo | 15:49 |
| kfox1111 | ah. | 15:49 |
| kfox1111 | did you remember to turn off cri on all nodes? | 15:49 |
| kfox1111 | (we should document that somewhere... ) | 15:49 |
| sbezverk | kfox1111: DAMN!!!!!!! | 15:50 |
| *** magicboiz has quit IRC | 15:50 | |
| kfox1111 | yeah.... I'm disapointed they enabled the beta cri by default. :/ | 15:51 |
| sbezverk | kfox1111: :) thanks a lot for pointing it out | 15:51 |
| kfox1111 | though I totally get why. | 15:51 |
| kfox1111 | np. :) | 15:51 |
| *** salv-orlando has quit IRC | 15:51 | |
| kfox1111 | docker inc renaming docker has been a huge pain. | 15:52 |
| *** magicboiz has joined #openstack-kolla | 15:52 | |
| kfox1111 | I've started to have to educate people that when they say docker, they really mean containerd now. | 15:52 |
| kfox1111 | they don't want "docker" anymore. | 15:52 |
| *** mgoddard has quit IRC | 15:54 | |
| *** iceyao has joined #openstack-kolla | 15:54 | |
| *** jascott1 has joined #openstack-kolla | 15:55 | |
| *** rwallner has quit IRC | 15:55 | |
| sbezverk | kfox1111: :) | 15:55 |
| sdake | kfox1111 +100 | 15:56 |
| sbezverk | kfox1111: appears nothing want to make our life easier | 15:56 |
| *** magicboiz has quit IRC | 15:56 | |
| kfox1111 | sbezverk: thats why we make the big bucks. ;) | 15:56 |
| kfox1111 | or something like that.... | 15:56 |
| kfox1111 | :) | 15:56 |
| *** gfhellma has joined #openstack-kolla | 15:57 | |
| sbezverk | kfox1111: that was it.. now openvswitch-agent is very happy | 15:59 |
| *** iceyao has quit IRC | 15:59 | |
| kfox1111 | cool. | 15:59 |
| *** jascott1 has quit IRC | 16:00 | |
| *** rwallner has joined #openstack-kolla | 16:01 | |
| *** jtriley_ has joined #openstack-kolla | 16:02 | |
| *** inc0 has joined #openstack-kolla | 16:02 | |
| kfox1111 | wow... this order is coming by snail. may 5-26. :) | 16:04 |
| *** jtriley has quit IRC | 16:05 | |
| *** egonzalez has quit IRC | 16:05 | |
| *** blallau has quit IRC | 16:05 | |
| openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-kubernetes master: Deployment guide updated with few more information https://review.openstack.org/457614 | 16:08 |
| *** magicboiz has joined #openstack-kolla | 16:08 | |
| inc0 | good morning | 16:08 |
| *** rwallner has quit IRC | 16:12 | |
| *** salv-orl_ has quit IRC | 16:14 | |
| spsurya | inc0: morning | 16:15 |
| *** inc0 has quit IRC | 16:16 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 16:17 |
| *** cuongnv has quit IRC | 16:17 | |
| *** sudipto_ has joined #openstack-kolla | 16:18 | |
| *** sudipto has joined #openstack-kolla | 16:18 | |
| *** StephenWang1991 has quit IRC | 16:21 | |
| sbezverk | kfox1111: that pinning PS, are you ok to ack it? adding volumev3 atm does not seem to be an option due to time constraints | 16:22 |
| kfox1111 | sbezverk: yeah, but can you file a blueprint so we dont forget? | 16:23 |
| sdake | and link so I can approve it plz | 16:23 |
| sdake | morning inc0 | 16:23 |
| sbezverk | sdake: you cannot approve to times ;) | 16:24 |
| *** gfidente has quit IRC | 16:24 | |
| *** inc0 has joined #openstack-kolla | 16:24 | |
| sbezverk | kfox1111: techdebt or blue-print? | 16:24 |
| kfox1111 | I guess its a blue-print as its a new feature kind of? | 16:25 |
| kfox1111 | not really sure. | 16:26 |
| sbezverk | kfox1111: blueprint is done | 16:27 |
| kfox1111 | sbezverk: thanks. | 16:27 |
| *** jascott1 has joined #openstack-kolla | 16:29 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Add fwaas extension in l3 agent when neutron fwaas is enabled https://review.openstack.org/457631 | 16:30 |
| openstackgerrit | Merged openstack/kolla-ansible stable/ocata: Fix qos rules are not applied on ocata with ovs https://review.openstack.org/457633 | 16:31 |
| *** StephenWang1991 has joined #openstack-kolla | 16:31 | |
| inc0 | kfox1111, so reading your mail about tags, I agree with you but what you're asking for is overhaul of build system really | 16:33 |
| kfox1111 | possibly? | 16:34 |
| inc0 | 1. develop rock-solid version manifesto system | 16:34 |
| kfox1111 | I think initially, I can do it by just retagging before upload. | 16:34 |
| inc0 | 2. check existing version manifesto | 16:34 |
| *** mgoddard has joined #openstack-kolla | 16:34 | |
| inc0 | 3. determine differences | 16:35 |
| kfox1111 | yeah. | 16:35 |
| inc0 | 4. tag accordingly and upload | 16:35 |
| openstackgerrit | Merged openstack/kolla-kubernetes master: Pinning OS_VOLUME_API_VERSION=2 https://review.openstack.org/457626 | 16:35 |
| kfox1111 | so, I'm doing a bit of it now: | 16:35 |
| kfox1111 | http://tarballs.openstack.org/kolla-kubernetes/gate/containers/ | 16:35 |
| kfox1111 | the list of versioned containers is here: http://tarballs.openstack.org/kolla-kubernetes/gate/containers/centos-binary-2-ceph-containers.txt | 16:36 |
| *** StephenWang1991 has quit IRC | 16:36 | |
| kfox1111 | and here: http://tarballs.openstack.org/kolla-kubernetes/gate/containers/kubernetes-containers.txt | 16:36 |
| kfox1111 | if it ends up pulling differently versioned containers, it updates the tarballs. | 16:36 |
| *** StephenWang1991 has joined #openstack-kolla | 16:36 | |
| kfox1111 | we would need a refined version that had revisions and a container fingerprint. | 16:37 |
| kfox1111 | something like 'rpm -qa | sort | md5sum' in the container. | 16:37 |
| kfox1111 | if a freshly built one had a different fingerprint, its a canditate for updating. it should be tested at that point, and if passes, released with revision+1. | 16:38 |
| kfox1111 | bbiab | 16:39 |
| *** strigazi is now known as strigazi_AFK | 16:39 | |
| *** strigazi_AFK is now known as strigazi | 16:41 | |
| *** dave-mccowan has quit IRC | 16:41 | |
| *** StephenWang1991 has quit IRC | 16:41 | |
| *** zhubingbing_ has quit IRC | 16:42 | |
| inc0 | kfox1111, but that will be the case every time | 16:42 |
| inc0 | for source builds at least | 16:42 |
| inc0 | and binary builds aren't voting in gates | 16:43 |
| *** jaosorior has quit IRC | 16:43 | |
| *** mbruzek has joined #openstack-kolla | 16:45 | |
| *** dave-mccowan has joined #openstack-kolla | 16:45 | |
| *** ksumit has joined #openstack-kolla | 16:46 | |
| *** harlowja_ has joined #openstack-kolla | 16:50 | |
| *** harlowja has quit IRC | 16:52 | |
| *** dmk0202 has quit IRC | 16:52 | |
| *** ssurana has joined #openstack-kolla | 16:52 | |
| *** unicell1 has quit IRC | 16:55 | |
| *** rwallner has joined #openstack-kolla | 16:59 | |
| *** rwallner_ has joined #openstack-kolla | 17:01 | |
| *** rwallne__ has joined #openstack-kolla | 17:02 | |
| *** rwallne__ has quit IRC | 17:03 | |
| *** rwallne__ has joined #openstack-kolla | 17:03 | |
| *** rwallner has quit IRC | 17:04 | |
| *** rwallner_ has quit IRC | 17:05 | |
| *** mbruzek has quit IRC | 17:17 | |
| *** tonanhngo has quit IRC | 17:21 | |
| *** jascott1 has quit IRC | 17:21 | |
| *** gfhellma_ has joined #openstack-kolla | 17:26 | |
| *** gfhellma has quit IRC | 17:29 | |
| *** shardy has quit IRC | 17:30 | |
| openstackgerrit | Marcus Williams proposed openstack/kolla master: Add OpenDaylight Container to kolla https://review.openstack.org/416369 | 17:32 |
| *** chas has joined #openstack-kolla | 17:36 | |
| *** jtriley_ has quit IRC | 17:36 | |
| openstackgerrit | Marcus Williams proposed openstack/kolla master: Add OpenDaylight Container to kolla https://review.openstack.org/416369 | 17:36 |
| *** unicell has joined #openstack-kolla | 17:37 | |
| openstackgerrit | Marcus Williams proposed openstack/kolla master: Add OpenDaylight Container to kolla https://review.openstack.org/416369 | 17:39 |
| *** chas has quit IRC | 17:40 | |
| *** jascott1 has joined #openstack-kolla | 17:43 | |
| *** inc0 has quit IRC | 17:43 | |
| *** jascott1 has quit IRC | 17:44 | |
| *** StephenWang1991 has joined #openstack-kolla | 17:46 | |
| *** rwallne__ has quit IRC | 17:49 | |
| *** japestinho has quit IRC | 17:50 | |
| *** StephenWang1991 has quit IRC | 17:51 | |
| *** jascott1 has joined #openstack-kolla | 17:52 | |
| *** rwellum has joined #openstack-kolla | 17:59 | |
| *** srwilkers has quit IRC | 18:01 | |
| *** rwallner has joined #openstack-kolla | 18:06 | |
| *** sudipto has quit IRC | 18:07 | |
| *** sudipto_ has quit IRC | 18:07 | |
| *** gfhellma_ has quit IRC | 18:07 | |
| *** sambetts is now known as sambetts|afk | 18:08 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla stable/ocata: Add ironic-ui in horizon binary install https://review.openstack.org/457755 | 18:09 |
| *** eaguilar has joined #openstack-kolla | 18:17 | |
| *** jascott1 has quit IRC | 18:21 | |
| *** jascott1 has joined #openstack-kolla | 18:22 | |
| *** inc0 has joined #openstack-kolla | 18:23 | |
| inc0 | back | 18:23 |
| *** jascott1 has quit IRC | 18:25 | |
| sdake | inc0 since your back - mind hitting the review queue | 18:29 |
| sdake | inc0 we have a bunch of stuff that is pretty easy to review and needs just needs a w+ | 18:30 |
| inc0 | omw | 18:30 |
| sdake | tx :) | 18:30 |
| *** eaguilar has quit IRC | 18:32 | |
| *** eaguilar_ has joined #openstack-kolla | 18:32 | |
| sdake | kfox1111 sbezverk can we have that cells discussion now | 18:34 |
| sdake | tell me what you want and i'll make it happen :) | 18:35 |
| *** mgoddard has quit IRC | 18:37 | |
| *** srwilkers has joined #openstack-kolla | 18:39 | |
| *** rwallner has quit IRC | 18:42 | |
| *** tonanhngo has joined #openstack-kolla | 18:43 | |
| *** tonanhngo has quit IRC | 18:43 | |
| *** matrohon has joined #openstack-kolla | 18:57 | |
| *** rwallner has joined #openstack-kolla | 18:57 | |
| *** gfhellma has joined #openstack-kolla | 19:00 | |
| *** Manheim has joined #openstack-kolla | 19:01 | |
| Manheim | hello | 19:04 |
| *** dave-mccowan has quit IRC | 19:06 | |
| *** rwallner has quit IRC | 19:08 | |
| *** ksumit has quit IRC | 19:09 | |
| *** ksumit has joined #openstack-kolla | 19:09 | |
| *** rwallner has joined #openstack-kolla | 19:10 | |
| *** rwallner has quit IRC | 19:12 | |
| *** inc0_ has joined #openstack-kolla | 19:13 | |
| *** rwallner has joined #openstack-kolla | 19:15 | |
| *** rwallner_ has joined #openstack-kolla | 19:16 | |
| *** mbruzek has joined #openstack-kolla | 19:16 | |
| *** rwallner has quit IRC | 19:19 | |
| *** jascott1 has joined #openstack-kolla | 19:25 | |
| kfox1111 | sdake: I think the idea is, we don't put any ordering of when cell setup is performed. | 19:28 |
| kfox1111 | instead, we do a nova-api-cell-discover-host-job. | 19:29 |
| kfox1111 | maybe explicit outside the service chart? | 19:30 |
| kfox1111 | not sure there's a non racy way to do it inside the service chart. | 19:30 |
| kfox1111 | cause you never know if you got all the nova compute's up. | 19:30 |
| *** dave-mccowan has joined #openstack-kolla | 19:32 | |
| sdake | kfox1111 how would one do computekit in that model | 19:33 |
| sbezverk | kfox1111: I wanted to build service chart with to parts | 19:33 |
| sbezverk | cell0 prep/create and discover | 19:33 |
| kfox1111 | sdake: two part. helm install compute-kit.... make sure all computes are up/ready, helm install nova-api-cell-discover-host-job | 19:33 |
| kfox1111 | the other way to do it requires something like an operator... | 19:34 |
| sbezverk | in this case it looks like service and almost independent from other things | 19:34 |
| kfox1111 | you need to know what nodes are suposed to be running nova-compute, | 19:34 |
| kfox1111 | and to know if its showing in nova service list. | 19:34 |
| kfox1111 | when all are there, you can run the discover-host-job. | 19:34 |
| kfox1111 | maybe that functionality could be added to discover-host-job itself conditionally? | 19:35 |
| *** pbourke has quit IRC | 19:36 | |
| *** pbourke has joined #openstack-kolla | 19:36 | |
| *** chas has joined #openstack-kolla | 19:37 | |
| *** jascott1 has quit IRC | 19:37 | |
| kfox1111 | part of the problem is there are two sources of knowlege to deay running the discover-host-job. | 19:37 |
| kfox1111 | one's k8s, and the other is nova. | 19:37 |
| kfox1111 | its probably non trivial to get that logic right. | 19:39 |
| sdake | vt | 19:39 |
| kfox1111 | 1. kubectl get nodes, check for compute=x. count them. | 19:40 |
| kfox1111 | 2. run nova service-list | grep nova-compute | wc -l | 19:40 |
| sdake | 55rf | 19:40 |
| kfox1111 | if !=, wait a bit and go back to step 1. | 19:40 |
| *** chas has quit IRC | 19:41 | |
| *** magicboiz has quit IRC | 19:41 | |
| sbezverk | kfox1111: I do not get why we need this complexity? | 19:44 |
| sbezverk | discovery does not need specific nodes | 19:44 |
| sbezverk | it will discover themm.. | 19:44 |
| *** jascott1 has joined #openstack-kolla | 19:44 | |
| kfox1111 | sbezverk: it only can discover them, if they are up and running. | 19:45 |
| sbezverk | basicallly whatever gets registered | 19:45 |
| kfox1111 | right. but for helm install compute-kit, | 19:45 |
| kfox1111 | folks would assume they can laucnh a vm once its done. | 19:45 |
| kfox1111 | but they need to wait for computes to finish spawning, then run discover-host-job. | 19:46 |
| sbezverk | why not to run it as a separate step?? | 19:46 |
| sbezverk | frankly I do not think it is a problem, especially since nobody will run compute kit with 2.0.0 | 19:46 |
| sbezverk | image | 19:47 |
| sbezverk | so prod people will start using from version 1.0.0 | 19:47 |
| kfox1111 | thats ok too.. we could say, helm install compute kit.... then wait until nova service-list looks correct to you , then run helm nova-api-discover-host-job | 19:47 |
| sbezverk | and good chance it will be 4.0.0 images or higher | 19:47 |
| sbezverk | kfox1111: exactly | 19:47 |
| sbezverk | that if how I thought when proposed separate | 19:48 |
| sbezverk | service chart including discovery job | 19:48 |
| kfox1111 | I don't htink it needs a seperate service chart though. | 19:48 |
| kfox1111 | nova-controller should be enough? | 19:48 |
| sbezverk | hmm, it is closer to compute | 19:49 |
| kfox1111 | maybe we're discussing idfferent things. | 19:49 |
| sbezverk | since it is kind deoends | 19:49 |
| kfox1111 | I'm suggesting, | 19:49 |
| *** iceyao has joined #openstack-kolla | 19:50 | |
| kfox1111 | cell0-simple-setup belongs in nova-controller, and nova-compute and nova-api-cell-discover-host-job are seperate things? | 19:50 |
| kfox1111 | you launch nova-controller/nova-compute, then wait for thigns to settle, then run the third one. | 19:50 |
| kfox1111 | or if you do computekit, it does the first 2 together. | 19:50 |
| sbezverk | one way or another it will work | 19:51 |
| kfox1111 | I think sdake was hoping to get the discover-host-job into the compute-kit too thoguh. but like I was saying, its a fair amount of logic work. | 19:51 |
| sbezverk | we just need chose it ;) | 19:51 |
| kfox1111 | my vote would be to throw nova-cell0-create-db-job in nova-controler service package, and add no deps on nova-compute (don't think that works anyway) | 19:52 |
| kfox1111 | then add a documentation step to do discover-host-job | 19:53 |
| jascott1 | lol http://fabric8.io/presentations/jfall-2016-nov/images/say-microservices.jpg | 19:53 |
| *** david-lyle has joined #openstack-kolla | 19:53 | |
| *** ksumit has quit IRC | 19:53 | |
| kfox1111 | jascott1: hehe. | 19:54 |
| jascott1 | was looking for k8s diagrams and found that | 19:54 |
| *** iceyao has quit IRC | 19:54 | |
| kfox1111 | hehe | 19:54 |
| kfox1111 | meeting next. | 19:55 |
| kfox1111 | jascott1: almost got flannel deploying with helm. :) | 19:55 |
| jascott1 | nice | 19:55 |
| kfox1111 | one more minor thing I think. | 19:55 |
| kfox1111 | theres another change I think needed to helm, but I'll drop the info in the review. | 19:55 |
| kfox1111 | gotta head to next meeting. | 19:55 |
| jascott1 | ok | 19:55 |
| kfox1111 | wanted to make sure I got the whole thing working end to end. | 19:55 |
| kfox1111 | there was a feature regression in 1.6 as aposed to 1.5 where I tested the concept. | 19:56 |
| kfox1111 | I found a workaround though. | 19:56 |
| kfox1111 | bbiab | 19:56 |
| *** chas has joined #openstack-kolla | 19:57 | |
| openstackgerrit | Marcus Williams proposed openstack/kolla master: Add OpenDaylight Container to kolla https://review.openstack.org/416369 | 19:58 |
| *** chas has quit IRC | 20:02 | |
| *** ksumit has joined #openstack-kolla | 20:02 | |
| *** srwilkers has quit IRC | 20:04 | |
| *** ksumit has quit IRC | 20:04 | |
| *** Pavo has quit IRC | 20:09 | |
| *** inc0_ has quit IRC | 20:15 | |
| *** chas has joined #openstack-kolla | 20:18 | |
| *** chas has quit IRC | 20:23 | |
| *** rwallner_ has quit IRC | 20:26 | |
| inc0 | devenv is almost ready | 20:30 |
| inc0 | anything anyone wants me to look at before I move on? | 20:31 |
| *** iceyao has joined #openstack-kolla | 20:31 | |
| *** openstackgerrit has quit IRC | 20:33 | |
| *** gfhellma_ has joined #openstack-kolla | 20:35 | |
| *** iceyao has quit IRC | 20:36 | |
| *** gfhellma has quit IRC | 20:36 | |
| SamYaple | inc0: yea | 20:37 |
| SamYaple | check this out http://i.imgur.com/IdFtpWC.jpg | 20:37 |
| inc0 | SamYaple: saw it before | 20:38 |
| inc0 | too much internet. | 20:38 |
| SamYaple | you just asked if anyone had anything to look at | 20:39 |
| SamYaple | you didnt specify anything about "new" | 20:39 |
| *** chas has joined #openstack-kolla | 20:39 | |
| inc0 | I also didn't say that I won't criticize it after I looked at it | 20:39 |
| *** matrohon has quit IRC | 20:42 | |
| *** chas has quit IRC | 20:44 | |
| *** salv-orlando has joined #openstack-kolla | 20:45 | |
| *** sean-k-mooney has quit IRC | 20:46 | |
| *** gfhellma__ has joined #openstack-kolla | 20:46 | |
| *** gfhellma_ has quit IRC | 20:47 | |
| *** shasha___ has joined #openstack-kolla | 20:48 | |
| *** david-lyle has quit IRC | 20:49 | |
| *** chas has joined #openstack-kolla | 21:00 | |
| *** schwicht has quit IRC | 21:03 | |
| *** chas has quit IRC | 21:05 | |
| SamYaple | haha | 21:06 |
| *** david-lyle has joined #openstack-kolla | 21:10 | |
| *** dmk0202 has joined #openstack-kolla | 21:11 | |
| *** spsurya has quit IRC | 21:13 | |
| *** iceyao has joined #openstack-kolla | 21:14 | |
| *** iceyao has quit IRC | 21:19 | |
| *** tonanhngo has joined #openstack-kolla | 21:21 | |
| *** chas has joined #openstack-kolla | 21:21 | |
| *** chas has quit IRC | 21:25 | |
| *** newmember has quit IRC | 21:25 | |
| *** mbruzek has quit IRC | 21:27 | |
| *** ksumit has joined #openstack-kolla | 21:27 | |
| *** eanylin has joined #openstack-kolla | 21:28 | |
| *** shasha___ has quit IRC | 21:28 | |
| *** Manheim has quit IRC | 21:32 | |
| *** eaguilar_ has quit IRC | 21:37 | |
| *** chas has joined #openstack-kolla | 21:42 | |
| *** manheim has joined #openstack-kolla | 21:44 | |
| *** chas has quit IRC | 21:46 | |
| *** manheim has quit IRC | 21:48 | |
| *** salv-orl_ has joined #openstack-kolla | 21:48 | |
| *** manheim has joined #openstack-kolla | 21:49 | |
| *** salv-orlando has quit IRC | 21:50 | |
| *** gfhellma__ has quit IRC | 21:51 | |
| *** MasterOfBugs has joined #openstack-kolla | 21:55 | |
| *** david-lyle has quit IRC | 21:55 | |
| *** iceyao has joined #openstack-kolla | 21:56 | |
| *** gfhellma has joined #openstack-kolla | 21:57 | |
| *** dmk0202 has quit IRC | 21:59 | |
| *** esharao has quit IRC | 22:00 | |
| *** iceyao has quit IRC | 22:00 | |
| *** gfhellma has quit IRC | 22:01 | |
| *** chas has joined #openstack-kolla | 22:03 | |
| *** goldyfruit has quit IRC | 22:05 | |
| *** chas has quit IRC | 22:07 | |
| jascott1 | why does this work `kubectl taint nodes kube1 node-role.kubernetes.io/master=:NoSchedule` | 22:10 |
| jascott1 | but this does not kubectl taint nodes kube1 node-role.kubernetes.io/master=:NoSchedule- | 22:10 |
| *** Pavo has joined #openstack-kolla | 22:15 | |
| jascott1 | whered every koala go? | 22:18 |
| sbezverk | jascott1: because to remove label you just specify labe name | 22:18 |
| sbezverk | kubectl taint nodes kube1 node-role.kubernetes.io/master- | 22:18 |
| sbezverk | you remove label this way | 22:18 |
| jascott1 | sbezverk thanks! | 22:19 |
| jascott1 | that worked but curios in that would remove all taints | 22:19 |
| jascott1 | with that key | 22:19 |
| jascott1 | docs say can do just the one | 22:19 |
| jascott1 | but thanks, trudging on | 22:19 |
| jascott1 | sbezverk docs say http://paste.openstack.org/show/607062/ | 22:20 |
| sbezverk | jascott1: sorry I did not get your question? | 22:23 |
| *** Pavo has quit IRC | 22:23 | |
| sbezverk | do you mean taints from all nodes or something else? | 22:23 |
| *** chas has joined #openstack-kolla | 22:23 | |
| jascott1 | no the specific NoSchedule taint | 22:24 |
| jascott1 | in my pastebin see the difference | 22:24 |
| jascott1 | first is just removing NoSchedule | 22:24 |
| jascott1 | second removes all taints with that key | 22:24 |
| jascott1 | idk its not critical for me atm but confusing | 22:24 |
| sbezverk | jascott1: I see now what you meant.. frankly I have not played much with this other tham removing it | 22:26 |
| jascott1 | ah that removed the taint but I have different taint problem | 22:27 |
| jascott1 | No nodes are available that match all of the following predicates:: PodToleratesNodeTaints | 22:27 |
| *** Pavo has joined #openstack-kolla | 22:28 | |
| *** chas has quit IRC | 22:28 | |
| jascott1 | so I removed all taints from master, waiting on tiller to retry to schedule and it failed with that error. Maybe it doesnt recheck nodes? | 22:28 |
| sbezverk | jascott1: tiller goes on master without remving taints if I am not mistaken | 22:31 |
| jascott1 | im not sure, kfox1111 said something about it floating around the cluster | 22:32 |
| jascott1 | ive never payed attention and usually use a multinode setup even if its just vagrant based k8s | 22:33 |
| *** Pavo has quit IRC | 22:34 | |
| sbezverk | jascott1: tiller lands in kube-system namespace | 22:34 |
| jascott1 | true | 22:35 |
| sbezverk | not sure if it can actually float arround especially since multimaster model is not yet supported | 22:35 |
| *** Pavo has joined #openstack-kolla | 22:35 | |
| jascott1 | sbezverk https://github.com/kubernetes/helm/issues/2222#issuecomment-294567589 | 22:36 |
| *** Pavo has quit IRC | 22:36 | |
| jascott1 | idk if everything in kube-system runs on master by default? | 22:37 |
| *** schwicht has joined #openstack-kolla | 22:37 | |
| jascott1 | i think my untaint didnt work because of the = | 22:39 |
| *** manheim has quit IRC | 22:40 | |
| sbezverk | jascott1: right you do not need =, only taint's name | 22:42 |
| jascott1 | mystery solved and I was correct in assuming i was doing something dumb :) | 22:43 |
| *** chas has joined #openstack-kolla | 22:44 | |
| kfox1111 | back. | 22:46 |
| jascott1 | we were just discussing the finer points of taint removal :) | 22:47 |
| kfox1111 | tiller has nothing to pin it to any node, so it floats by default. but if theh master is tainted, it won't land there. | 22:47 |
| kfox1111 | hehe. yeah. | 22:47 |
| jascott1 | you are correct sir as I have just experienced | 22:47 |
| kfox1111 | hence the taint feature request. :) | 22:48 |
| jascott1 | i was thinking the existing net-host could go in by itself and I can add the rest in another PR. does that make sense? | 22:48 |
| *** salv-orl_ has quit IRC | 22:48 | |
| *** chas has quit IRC | 22:49 | |
| kfox1111 | yeah. though I think the ohter bit may need to go in. | 22:49 |
| kfox1111 | so, here's the ps I've been working on: | 22:49 |
| kfox1111 | https://review.openstack.org/#/c/456964/ | 22:49 |
| sbezverk | kfox1111: in multinode scenario if multiple controllers are exist in kolla, it seems I have to use distributed neutron l3 agent | 22:50 |
| kfox1111 | the helm init --net-host would have worked on 1.5, but on 1.6, they have a new behavior. | 22:50 |
| kfox1111 | the node is tainted if cni is not started yet. | 22:50 |
| kfox1111 | and the kube-scheduler skips scheduling deployments on the node. :/ | 22:51 |
| kfox1111 | even if net=host. | 22:51 |
| kfox1111 | (there is a bug in for that). | 22:51 |
| kfox1111 | but if its a DaemonSet and net=host, it works fine. | 22:51 |
| kfox1111 | see, https://review.openstack.org/#/c/456964/7/tools/setup_helm.sh | 22:51 |
| kfox1111 | lines 17-19. | 22:51 |
| kfox1111 | the issue is kind of two part. DaemonSet works, but tiller can only have one instance. | 22:52 |
| kfox1111 | so that can be worked around in a DaemonSet by supporting labels, and only labeling one node. | 22:52 |
| kfox1111 | sbezverk: you talking dvr or l3-ha? | 22:53 |
| jascott1 | but if we just add taints and labels to init we are good? | 22:53 |
| kfox1111 | (or both) | 22:53 |
| *** david-lyle has joined #openstack-kolla | 22:53 | |
| sbezverk | kfox1111: yep | 22:54 |
| kfox1111 | jascott1: ideally, yes. and thats what they would like to do, but 1.6 doesn't fully support taints/tolerations for cni yet, and the scheculer has some hard coded logic still. :/ | 22:54 |
| jascott1 | ooooh | 22:54 |
| jascott1 | bummer | 22:54 |
| kfox1111 | sbezverk: ideally, all new deployments would use dvr + l3ha. | 22:54 |
| kfox1111 | jascott1: yeah. :/ | 22:54 |
| sbezverk | even if I use legacy, it starts on both controllers and it does not work well | 22:54 |
| kfox1111 | jascott1: thats part of the reason the 1.6.0 release went so badly. :/ | 22:54 |
| kfox1111 | sbezverk: it should still work. it should register them with seperate agent names, and l3 traffic should be assigned to just one or the other. | 22:55 |
| kfox1111 | but failover should take a long time. | 22:55 |
| *** Reepicheep has quit IRC | 22:56 | |
| *** Reepicheep has joined #openstack-kolla | 22:56 | |
| *** david-lyle has quit IRC | 22:56 | |
| kfox1111 | jascott1: so, I think to make the net=host thing work as intended, for 1.6 support at least, we need a --daemonset flag and a --label key=value support. | 22:57 |
| kfox1111 | the former just sed 's/Deployment/DaemonSet/' and the --node-selector support I mean, not labels. | 22:57 |
| kfox1111 | key value set of node labels to add in. | 22:58 |
| jascott1 | hmmm im not sure they (helm) will dig the daemonset flag | 22:58 |
| jascott1 | since it seems like a workaround | 22:58 |
| kfox1111 | not sure. I'd kind of like to use it anyway for one of my production systems. | 22:58 |
| kfox1111 | I'm pinning a tiller to a particular host and net=host to make it run within the hosts firewall, | 22:59 |
| kfox1111 | preventing others from being able to access it and create/delete pods. | 22:59 |
| kfox1111 | will keep needing to do that until k8s/helm authentication is fully implemented. | 22:59 |
| *** huzhengchuan has quit IRC | 23:00 | |
| kfox1111 | the daemonset thing will be really nice too when master election is added. | 23:00 |
| kfox1111 | just run daemonset and a nodeselector of controller=true, and your ha setup is complete. :) | 23:01 |
| *** goldyfruit has joined #openstack-kolla | 23:01 | |
| *** StephenWang1991 has joined #openstack-kolla | 23:02 | |
| jascott1 | kfox1111 hmmm you might want to pitch that on the 2222 issue or open a new one, although I havent had much input from technosophos | 23:03 |
| kfox1111 | k. | 23:03 |
| kfox1111 | I wonder how much the buyout is sucking up his time. | 23:03 |
| jascott1 | shopping for yachts? | 23:03 |
| jascott1 | i hope they are all but doubt it | 23:03 |
| kfox1111 | anyone have a deb box handy? | 23:04 |
| jascott1 | just ubuntu | 23:04 |
| kfox1111 | nm... I cna just run it in docker... | 23:05 |
| kfox1111 | docker run -it --rm ubuntu /bin/bash :) | 23:05 |
| *** chas has joined #openstack-kolla | 23:05 | |
| *** StephenWang1991 has quit IRC | 23:07 | |
| jascott1 | since Intel and Rackspace are no longer together we dont have dev environments. halycon kinda broke on 1.6 so I created a playbook and at least have a dev env now | 23:07 |
| jascott1 | the kolla-kubernetes part needs work ;) | 23:07 |
| kfox1111 | hows this look for a container fingerprinter: | 23:08 |
| kfox1111 | which deb 2>/dev/null && echo pip && pip freeze | sort; which dpkg 2>/dev/null && echo dpkg && dpkg -l | awk '{print $1, $2, $3, $4}' | sort -k 2; which rpm 2>/dev/null && echo rpm && rpm -qa | sort | 23:08 |
| jascott1 | can you add a couple more lines so my eyes will fully bleed out | 23:08 |
| kfox1111 | hehe. | 23:08 |
| kfox1111 | which pip 2>/dev/null && echo pip && pip freeze | sort | 23:09 |
| kfox1111 | which dpkg 2>/dev/null && echo dpkg && dpkg -l | awk '{print $1, $2, $3, $4}' | sort -k 2 | 23:09 |
| kfox1111 | which rpm 2>/dev/null && echo rpm && rpm -qa | sort | 23:09 |
| kfox1111 | so if anything in that output changes between two containers, the container got updated. | 23:10 |
| *** chas has quit IRC | 23:10 | |
| kfox1111 | sbezverk: so... what problem are you seeing with multiple l3 agents? | 23:13 |
| kfox1111 | I'm running them on one of my clusters and its ok. so maybe there's a misconfiguration somewhere? | 23:14 |
| jascott1 | i wonder if they will fix the cni loading thing soon since it was such a problem | 23:17 |
| *** ksumit has quit IRC | 23:18 | |
| jascott1 | haha just noticed the github issue count just gives up at 5000+ | 23:25 |
| *** goldyfruit has quit IRC | 23:26 | |
| *** chas has joined #openstack-kolla | 23:26 | |
| *** tonanhngo has quit IRC | 23:29 | |
| *** tonanhngo has joined #openstack-kolla | 23:29 | |
| *** chas has quit IRC | 23:31 | |
| jascott1 | kfox1111 if you can link me to relevant kubernetes issues I would be grateful.. no rush | 23:33 |
| *** tonanhngo has quit IRC | 23:34 | |
| *** lamt has quit IRC | 23:40 | |
| *** tonanhngo has joined #openstack-kolla | 23:40 | |
| *** goldyfruit has joined #openstack-kolla | 23:42 | |
| *** manheim has joined #openstack-kolla | 23:42 | |
| *** tonanhngo has quit IRC | 23:45 | |
| *** iniazi has joined #openstack-kolla | 23:46 | |
| *** manheim has quit IRC | 23:46 | |
| *** chas has joined #openstack-kolla | 23:47 | |
| *** chas has quit IRC | 23:52 | |
| *** eaguilar has joined #openstack-kolla | 23:52 | |
| kfox1111 | jascott1: https://github.com/kubernetes/kubernetes/issues/43815 | 23:53 |
| *** spsurya has joined #openstack-kolla | 23:53 | |
| inc0 | kfox1111: will you join me on kolla project update session? | 23:54 |
| spsurya | morning all | 23:55 |
| inc0 | I could also use help during hands on workshops for kolla-ansible | 23:55 |
| *** yangyapeng has quit IRC | 23:55 | |
| *** eanylin has quit IRC | 23:56 | |
| *** al498u_ has quit IRC | 23:56 | |
| *** goldyfruit has quit IRC | 23:58 | |
| kfox1111 | spsurya: hi. :) | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
