| *** caoyuan has joined #openstack-kolla | 00:06 | |
| *** adrian_otto has quit IRC | 00:16 | |
| *** yangyapeng has joined #openstack-kolla | 00:19 | |
| *** yangyapeng has quit IRC | 00:20 | |
| kolla-slack | <kfox1111> @fooliouno setting to different nodes is tested in the gate. works good. :slightly_smiling_face: | 00:27 |
|---|---|---|
| *** caoyuan has quit IRC | 00:29 | |
| *** dave-mccowan has quit IRC | 00:29 | |
| openstackgerrit | Mohammed Naser proposed openstack/kolla master: Merged multiple ENV into a single layer https://review.openstack.org/449371 | 00:30 |
| *** jrobinson has quit IRC | 00:31 | |
| *** duonghq has joined #openstack-kolla | 00:34 | |
| duonghq | morning guys | 00:34 |
| *** sayantani01 has quit IRC | 00:43 | |
| *** dave-mccowan has joined #openstack-kolla | 00:49 | |
| spsurya | duonghq: morning | 00:54 |
| duonghq | morning spsurya | 00:54 |
| *** dave-mcc_ has joined #openstack-kolla | 00:55 | |
| duonghq | sdake, ping | 00:57 |
| *** dave-mccowan has quit IRC | 00:57 | |
| *** tovin07 has joined #openstack-kolla | 00:58 | |
| *** qiliang27 has joined #openstack-kolla | 01:05 | |
| *** jrobinson has joined #openstack-kolla | 01:08 | |
| *** dave-mccowan has joined #openstack-kolla | 01:09 | |
| *** dave-mcc_ has quit IRC | 01:11 | |
| *** qiliang28 has joined #openstack-kolla | 01:12 | |
| *** sayantani01 has joined #openstack-kolla | 01:13 | |
| *** qiliang27 has quit IRC | 01:14 | |
| *** qiliang28 is now known as qiliang27 | 01:14 | |
| *** egonzalez has quit IRC | 01:15 | |
| *** MasterOfBugs has joined #openstack-kolla | 01:17 | |
| duonghq | sbezverk_, are you there? | 01:22 |
| *** zhurong has joined #openstack-kolla | 01:26 | |
| openstackgerrit | yunfeng zhou proposed openstack/kolla-ansible master: change log-* to flog-*. https://review.openstack.org/449384 | 01:26 |
| openstackgerrit | yunfeng zhou proposed openstack/kolla-ansible master: change log-* to flog-*. https://review.openstack.org/449384 | 01:27 |
| *** cuongnv has joined #openstack-kolla | 01:28 | |
| *** zhubingbing has joined #openstack-kolla | 01:28 | |
| kolla-slack | <portdirect> Oh this is cool o/ hey peeps :D | 01:35 |
| *** japestinho has quit IRC | 01:46 | |
| *** japestinho has joined #openstack-kolla | 01:47 | |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 02:02 |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 02:10 |
| *** eaguilar has joined #openstack-kolla | 02:13 | |
| duonghq | Jeffrey4l, are you there? | 02:15 |
| Jeffrey4l | duonghq, yep. sup | 02:17 |
| duonghq | https://review.openstack.org/#/c/445573/ about this ps, where do you get the cherry pick commit id? | 02:18 |
| Jeffrey4l | inc0, please review the release patch https://review.openstack.org/448711 ? | 02:18 |
| duonghq | the commit id I get from cherry-pick feature of gerrit is differ from your | 02:18 |
| Jeffrey4l | duonghq, check https://review.openstack.org/#/q/I6b77773238cd65a3cc009d8d9a38bdbea0035212,n,z | 02:18 |
| Jeffrey4l | duonghq, the commit sha will be changed, for sure. | 02:19 |
| Jeffrey4l | it is how git works. | 02:19 |
| kolla-slack | <inc0> Done Jeffrey | 02:20 |
| duonghq | hmm, given that gerrit can search for this | 02:20 |
| duonghq | I think it's ok | 02:20 |
| Jeffrey4l | inc0, cool. whats this? slack channel? | 02:20 |
| duonghq | although I prefer merged id | 02:21 |
| kolla-slack | <inc0> Yeah sdake connected slack with irc | 02:21 |
| Jeffrey4l | duonghq, yep. as long as the change id is the same. | 02:21 |
| kolla-slack | <inc0> Writing from phone | 02:21 |
| Jeffrey4l | inc0, how can i join it? | 02:21 |
| kolla-slack | <inc0> Sure, Kubernetes slack team, kolla | 02:22 |
| kolla-slack | <inc0> It's open channel | 02:22 |
| duonghq | hmm, we have this channel long ago | 02:22 |
| duonghq | but kolla-slack is new? | 02:22 |
| kolla-slack | <inc0> No, it's same channel | 02:22 |
| kolla-slack | <inc0> Just now it's mirrored | 02:22 |
| *** caoyuan has joined #openstack-kolla | 02:23 | |
| *** caowei has quit IRC | 02:23 | |
| duonghq | I mean the kolla-slack bot is new? | 02:23 |
| duonghq | have not seen it before | 02:24 |
| *** yangyapeng has joined #openstack-kolla | 02:25 | |
| *** sayantani01 has quit IRC | 02:27 | |
| *** schwicht has quit IRC | 02:27 | |
| kolla-slack | <jeffrey4l> i am in. interesting | 02:27 |
| duonghq | Jeffrey4l, have fun ;) | 02:28 |
| *** caoyuan has quit IRC | 02:29 | |
| *** caoyuan has joined #openstack-kolla | 02:31 | |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 02:33 |
| sdake | sup peeps | 02:44 |
| sdake | duonghq kolla-aslack is a bot i setup | 02:44 |
| sdake | duonghq it is new indeed | 02:45 |
| duonghq | nice | 02:45 |
| *** unicell1 has quit IRC | 02:45 | |
| sdake | kolla-slack isa a bot that runs on a server and connects to slack and irc and mirrors the conversation | 02:46 |
| *** shashank_t_ has joined #openstack-kolla | 02:46 | |
| sdake | it took about 30 minutes to tsetup up | 02:46 |
| duonghq | does it consume much resource? | 02:47 |
| sdake | we couldn't set it up prior | 02:47 |
| sdake | the resource it consumes the msot of is the scrollback (in the free lack) | 02:47 |
| sdake | I think someone in kubernetes land is paying for paid slack which is 7 or 8 bucks a month per user | 02:47 |
| sdake | weh ave between 800-1k messages a day | 02:48 |
| Jeffrey4l | sdake, set up the bot from zero? or leverage some existing service? | 02:48 |
| sdake | free slack has a 10k msg scrollback before messages drop off in free | 02:48 |
| sdake | Jeffrey4l i used this : https://github.com/ekmartin/slack-irc | 02:48 |
| sdake | npm install slack-irc | 02:49 |
| sdake | and sepnd 3 minutes wirting a cnofig file | 02:49 |
| sdake | GTG | 02:49 |
| Jeffrey4l | roger. thanks. | 02:49 |
| sdake | thatis good to go | 02:49 |
| sdake | there is a paid service that does the same thing which is 490$/yr | 02:49 |
| * sdake starts a business hosting slack-irc instances :) | 02:50 | |
| Jeffrey4l | lol | 02:50 |
| Jeffrey4l | btw, slack is hard to use in China. very slow in sometime. ;( | 02:50 |
| sdake | Jeffrey4l right - plan is to still continue to use irc | 02:50 |
| sdake | hwoever there are alot of people that either don't want or can't run irc for whatever eason | 02:51 |
| Jeffrey4l | sync bot is a good idea. | 02:51 |
| sdake | but can run slack - so its essentiallly free for others to participate in kolla dev | 02:51 |
| spsurya | sdake: +1 | 02:53 |
| spsurya | sdake: my team is also using slack | 02:53 |
| spsurya | good to use | 02:53 |
| sdake | the slack side integration is pretty good, the irc side integration isa little clunky | 02:57 |
| sdake | i think we mostly want to focus our work here - but provide an opportunity to build a bridge to the kubernetes community - and part of that happens via slack :) | 02:58 |
| sdake | here being irc | 02:58 |
| *** yuanying has quit IRC | 02:58 | |
| *** eaguilar has quit IRC | 03:04 | |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 03:08 |
| *** dave-mccowan has quit IRC | 03:08 | |
| spsurya | Jeffrey4l: sdake btw what is the problem with IRC ? | 03:26 |
| Jeffrey4l | i do not see the issue of irc. but slack is easy to integrate with other service. and k8s team use slack and openstack communit use irc | 03:26 |
| sdake | right the problem we are fixing is not that irc is broken but that the communities don't have a common way to communicate that doesn't involve two clients | 03:28 |
| sdake | now they do :) | 03:28 |
| sdake | they/we | 03:28 |
| spsurya | Jeffrey4l: sdake roger | 03:30 |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 03:32 |
| Jeffrey4l | btw, i found http://wiki.znc.in/ZNC . it is a IRC bouncer | 03:33 |
| *** Margin has joined #openstack-kolla | 03:34 | |
| spsurya | Jeffrey4l: if we will integrate slack then we will have the same problem for kolla developer and other service developer in communication | 03:35 |
| spsurya | sdake | 03:35 |
| Jeffrey4l | what's the "same problem"? | 03:36 |
| sdake | must have missed a message | 03:37 |
| sdake | "same problem?" | 03:37 |
| spsurya | Jeffrey4l: as we have in communication with kubernetes member as i understand :( | 03:38 |
| sdake | spsurya if you want to talk directly with helm community or kubernetes community - the correct answer is to use the tools they are accustomed to | 03:39 |
| sdake | (which means slack) | 03:39 |
| sdake | we can't necessarily solve that problem | 03:39 |
| Jeffrey4l | i do not think we have better solution expect the current sync bot. | 03:39 |
| sdake | the problem we can solve is linking hte kubernetes slack #kolla to #openstack-kolla on irc | 03:40 |
| sdake | spsurya the kubernetes slack is open to anyone | 03:40 |
| sdake | spsurya not sure what the exact issue you are concerned with | 03:40 |
| sdake | spsurya mind expanding | 03:40 |
| spsurya | Jeffrey4l: sdake roger | 03:42 |
| spsurya | sdake: as i understand the problem is linking the kuberenes slack and #openstack-kolla | 03:46 |
| spsurya | right ? | 03:46 |
| spsurya | i.e running common channel of slack and IRC | 03:47 |
| spsurya | for kolla-kuberenetes | 03:47 |
| spsurya | and kubernetes | 03:47 |
| spsurya | sdake: i have no issue in communication with helm community or kubernetes community :) | 03:50 |
| spsurya | just wanted to understand the yours and Jeffrey4l conversation | 03:50 |
| spsurya | Jeffrey4l: sdake BRB | 03:52 |
| *** unicell has joined #openstack-kolla | 03:53 | |
| spsurya | sorry for too much mind expansion :( | 03:53 |
| *** cliles has quit IRC | 04:01 | |
| *** zhurong has quit IRC | 04:07 | |
| *** n0isyn0ise has quit IRC | 04:12 | |
| *** n0isyn0ise has joined #openstack-kolla | 04:14 | |
| *** unicell1 has joined #openstack-kolla | 04:18 | |
| *** unicell has quit IRC | 04:18 | |
| *** jrobinson has quit IRC | 04:19 | |
| *** yuanying has joined #openstack-kolla | 04:20 | |
| *** unicell has joined #openstack-kolla | 04:22 | |
| *** unicell1 has quit IRC | 04:23 | |
| *** jrobinson has joined #openstack-kolla | 04:24 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla master: DO NOT MERGE: PoC of CI friendly share nothing img https://review.openstack.org/449426 | 04:33 |
| *** caoyuan has quit IRC | 04:35 | |
| *** yangyapeng has quit IRC | 04:40 | |
| sdake | nothing like the "please stop" passive aggressive response in a review | 04:43 |
| * sdake cackles insanely | 04:43 | |
| *** skramaja has joined #openstack-kolla | 04:44 | |
| duonghq | sdake, sup? | 04:47 |
| sdake | duonghq nothing much | 04:48 |
| sdake | duonghq just personal attacks in reviews followed by "please stop" | 04:48 |
| duonghq | ya | 04:48 |
| duonghq | not good | 04:48 |
| sdake | i dont understand why people can't just accept there is a system in place to handle these things and work through the normal processes and channels to make things work well | 04:48 |
| sdake | instead attack me for stating facts | 04:49 |
| sdake | its not like I wrote linuxfoundation's trademark policy | 04:49 |
| sdake | I simply pointed it out | 04:49 |
| duonghq | hmm, | 04:50 |
| sdake | https://review.openstack.org/449426 | 04:50 |
| sdake | for those interested - bring some salt ! | 04:50 |
| duonghq | I don't get idea behind it | 04:51 |
| sdake | who knows who cares | 04:51 |
| duonghq | ah, got the idea | 04:51 |
| sdake | if someone wants to do something wfm | 04:51 |
| duonghq | one of the PTG topic | 04:51 |
| sdake | yup that may be it | 04:52 |
| duonghq | hmm, don't understand why keystone do not get resolve.conf from configmap | 04:53 |
| sdake | sbezverk_ speaking of resolv.conf | 04:53 |
| duonghq | it was ok with kargo | 04:53 |
| sdake | sbezverk_ that workaround you ahve in place is not needed if docker-engine is setup properly | 04:53 |
| duonghq | but when I setup cluster with kubeadm, it doesn't like me anymore | 04:53 |
| *** unicell1 has joined #openstack-kolla | 04:53 | |
| sdake | kubeadm uses a resolv.conf with the name "kube-dns" | 04:54 |
| sdake | kargo uses "kubedns" | 04:54 |
| sdake | this is a delta | 04:54 |
| duonghq | ya, I fix that | 04:54 |
| duonghq | fixed | 04:54 |
| *** unicell has quit IRC | 04:54 | |
| duonghq | the generated configmap is ok | 04:55 |
| *** haplo37 has quit IRC | 04:59 | |
| *** haplo37 has joined #openstack-kolla | 05:08 | |
| Jeffrey4l | spsurya, around? wanna talk about https://review.openstack.org/432300 this patch with u. | 05:09 |
| spsurya | Jeffrey4l: yes | 05:10 |
| Jeffrey4l | spsurya, still can not know why we need the old version. | 05:10 |
| Jeffrey4l | what the new and old is used for the same purpose. and there are some issue in the old implementation. | 05:11 |
| Jeffrey4l | can not parse what u said " Supose if any operator wants to find the latest version of kolla rather than version of the latest version from openstack release" | 05:12 |
| Jeffrey4l | spsurya, ^^ | 05:12 |
| spsurya | Jeffrey4l: new implementation is good for getting the latest version from openstack/release | 05:12 |
| spsurya | Jeffrey4l: i want to say that that suppose if we want to get the version for specific release | 05:13 |
| Jeffrey4l | for example? | 05:13 |
| Jeffrey4l | openstack release? or nova release? | 05:13 |
| spsurya | Jeffrey4l: I mean for mitaka, Newton or ocata | 05:14 |
| spsurya | not the latest i.e ocata | 05:14 |
| spsurya | Jeffrey4l: new implemenattion is good | 05:15 |
| Jeffrey4l | spsurya, the old implement can not handle this but the new implement can handle this. | 05:15 |
| Jeffrey4l | this is what i said: old have some bug. | 05:15 |
| Jeffrey4l | old implement always get the latest version. | 05:15 |
| Jeffrey4l | he do not know which version is the latest ocata version. | 05:15 |
| Jeffrey4l | then it resulted in we are using newton swift in mitaka branch, now. | 05:16 |
| *** zhurong has joined #openstack-kolla | 05:16 | |
| Jeffrey4l | spsurya, check the --openstack-release parameter in the new implementation. | 05:16 |
| spsurya | i Jeffrey4l sorry but if we see from commit message it look like what i am saying | 05:16 |
| Jeffrey4l | and it use .gitreview branch as default. | 05:17 |
| spsurya | Jeffrey4l: roger | 05:17 |
| Jeffrey4l | version-check gets latest versions from | 05:17 |
| Jeffrey4l | openstack/releases repo, which is faster and get also the latest | 05:17 |
| Jeffrey4l | versions for certain openstack release. | 05:17 |
| spsurya | Jeffrey4l: checking | 05:17 |
| Jeffrey4l | i.e.: it support master and certain openstack release. | 05:17 |
| spsurya | Jeffrey4l: I also commented *Please correct if I am on wrong track. Thanks !!* | 05:18 |
| spsurya | Jeffrey4l: i think we need to correct the commit message | 05:19 |
| Jeffrey4l | yep. so we are talking this ;) | 05:19 |
| Jeffrey4l | yep. the commit message is a little confused . | 05:19 |
| spsurya | Jeffrey4l: that is why i got confused | 05:19 |
| Jeffrey4l | as a non-native speaker, this is always hard for me ;) | 05:19 |
| spsurya | Jeffrey4l: updation of commit message will make me green for you patch :) | 05:20 |
| Jeffrey4l | i will add some usage in the commit message for clarify | 05:20 |
| Jeffrey4l | cool. | 05:20 |
| spsurya | Jeffrey4l: roger , thanks | 05:20 |
| Jeffrey4l | thanks. | 05:20 |
| spsurya | Jeffrey4l: any way ! what about the your travellng to summit ? | 05:21 |
| spsurya | Jeffrey4l: Are we going to be on same stage ? | 05:21 |
| spsurya | :) | 05:21 |
| Jeffrey4l | still preparing the visa. others are find. if it will be accepted. | 05:22 |
| Jeffrey4l | yep. hope that so. | 05:22 |
| Jeffrey4l | what's your progress? | 05:22 |
| spsurya | same in VISA process | 05:22 |
| spsurya | :( | 05:22 |
| Jeffrey4l | aha | 05:22 |
| spsurya | Jeffrey4l: hope we will get that | 05:24 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: Get OpenStack service versions from openstack/releases repo https://review.openstack.org/432300 | 05:26 |
| Jeffrey4l | spsurya, updated. please check again. ^^ | 05:27 |
| *** caoyuan has joined #openstack-kolla | 05:27 | |
| spsurya | Jeffrey4l: sure | 05:27 |
| Jeffrey4l | thanks. | 05:29 |
| spsurya | Jeffrey4l: LGTM | 05:38 |
| Jeffrey4l | thanks. | 05:38 |
| spsurya | waiting for jenkin jobs | 05:39 |
| *** sayantani01 has joined #openstack-kolla | 05:45 | |
| *** bmace has quit IRC | 05:48 | |
| *** bmace has joined #openstack-kolla | 05:48 | |
| *** shashank_t_ has quit IRC | 05:53 | |
| *** mdnadeem has joined #openstack-kolla | 05:54 | |
| *** shashank_t_ has joined #openstack-kolla | 05:59 | |
| duonghq | anybody get TSP result? | 06:02 |
| *** shashank_t_ has quit IRC | 06:03 | |
| duonghq | spsurya, did you get TSP result? | 06:05 |
| spsurya | duonghq: didn't apply, hoping from employer :) | 06:06 |
| duonghq | ok | 06:06 |
| *** yee37914 has quit IRC | 06:06 | |
| *** yee379 has joined #openstack-kolla | 06:08 | |
| *** yee379 has quit IRC | 06:12 | |
| *** yee379 has joined #openstack-kolla | 06:12 | |
| *** yee379 has quit IRC | 06:14 | |
| *** yee379 has joined #openstack-kolla | 06:15 | |
| *** tovin07_ has joined #openstack-kolla | 06:18 | |
| *** sayantani01 has quit IRC | 06:25 | |
| *** sayantani01 has joined #openstack-kolla | 06:26 | |
| openstackgerrit | jimmygc proposed openstack/kolla-ansible master: Fix glance Upload large image haproxy timeout https://review.openstack.org/449440 | 06:30 |
| *** MasterOfBugs has quit IRC | 06:32 | |
| *** yee379 has quit IRC | 06:44 | |
| *** yee379 has joined #openstack-kolla | 06:44 | |
| *** jrobinson has quit IRC | 06:47 | |
| *** targon has joined #openstack-kolla | 06:52 | |
| *** unicell has joined #openstack-kolla | 06:57 | |
| sdake | zhubingbing around? | 06:57 |
| *** unicell1 has quit IRC | 06:57 | |
| zhubingbing | hear | 06:57 |
| zhubingbing | ；） | 06:58 |
| *** jascott1_ has quit IRC | 07:01 | |
| *** jascott1_ has joined #openstack-kolla | 07:01 | |
| spsurya | Jeffrey4l: ping.... | 07:05 |
| *** jascott1_ has quit IRC | 07:06 | |
| sdake | zhubingbing any progress on https://review.openstack.org/#/c/439740/ | 07:07 |
| spsurya | sdake there many BPs which are still in ocata/newton series | 07:10 |
| spsurya | should not we move these to pike ? | 07:10 |
| sdake | spsurya not sure - i dont work on kolla-ansible all that much anymore | 07:10 |
| sdake | spsurya i dont do triage for the blueprints/bugs in that deliverable | 07:11 |
| spsurya | sdake: there are many in kolla deliverable too | 07:11 |
| sdake | spsurya i dont do triage at all for bugs in kolla or kolla-ansible deliverable unless someone specifically asks me to do so | 07:12 |
| spsurya | sdake: roger | 07:12 |
| sdake | spsurya i barely do triage for blueprints | 07:12 |
| sdake | spsurya i have alot on my plate at present, and there are others in the various drivers team that can do that triage | 07:12 |
| spsurya | sdake: i was just getting aware of that :( | 07:13 |
| sdake | spsurya how do you mean | 07:13 |
| spsurya | sdake: i wanted know ? Is there any specific reason for the BPs which are in the ocata series | 07:14 |
| spsurya | sdake: my apology. If i asked something wrong :( | 07:16 |
| sdake | spsurya no reason to apologize | 07:17 |
| sdake | spsurya i dont know why people put blueprints in ocata - people self organize in kolla | 07:17 |
| sdake | typically in the past ocata blueprints that didn't finish should be removed from ocata and retargeted towards pike | 07:18 |
| duonghq | hmm, bare kubeadm is way different to kargo | 07:18 |
| spsurya | sdake : roger, i was also thinking the same | 07:19 |
| spsurya | thanks | 07:19 |
| sdake | spsurya if your not in the drivers team i can ad dyou and you can do that retargeting | 07:20 |
| sdake | anyone in the drivers team shoudl be doing that work | 07:20 |
| sdake | we shouldn't have SPOFs in our community around any specific individual | 07:20 |
| spsurya | sdake: i am in already in driver team | 07:21 |
| spsurya | i will move them to pike-1 | 07:21 |
| sdake | spsurya well then you can retarget blueprints :) | 07:21 |
| *** manheim has joined #openstack-kolla | 07:22 | |
| *** manheim has quit IRC | 07:22 | |
| *** manheim has joined #openstack-kolla | 07:23 | |
| spsurya | roger ......)))))) | 07:23 |
| duonghq | I have meeting now, BRB | 07:24 |
| *** sayantani01 has quit IRC | 07:24 | |
| *** jascott1_ has joined #openstack-kolla | 07:25 | |
| *** jmccarthy has joined #openstack-kolla | 07:26 | |
| sdake | bedtime | 07:29 |
| sdake | later | 07:29 |
| *** caoyuan_ has joined #openstack-kolla | 07:31 | |
| *** Margin is now known as MarginHu | 07:32 | |
| *** caoyuan has quit IRC | 07:33 | |
| spsurya | Jeffrey4l: done for https://review.openstack.org/#/c/432300/13 | 07:33 |
| *** satyar has joined #openstack-kolla | 07:33 | |
| *** zhuzeyu has joined #openstack-kolla | 07:41 | |
| *** jascott1_ has quit IRC | 07:44 | |
| *** jascott1_ has joined #openstack-kolla | 07:45 | |
| *** jascott1_ has quit IRC | 07:50 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ironic-pxe: handle non-x86 architectures https://review.openstack.org/434817 | 07:54 |
| hrw | morning | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: enable all images enabled for Ubuntu https://review.openstack.org/432787 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: Add support for non-x86 architectures (aarch64, ppc64le) https://review.openstack.org/430940 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: openstack-base/kolla-toolbox: use mariadb-devel for CentOS and !x86-64 https://review.openstack.org/444832 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-libvirt: handle ppc64le https://review.openstack.org/434810 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ceph: move ceph-fuse package to cephfs-fuse image https://review.openstack.org/440634 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: kubernetes: disable for architectures other than x86-64 https://review.openstack.org/445638 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-compute: handle rtslib(-fb) for debian-binary builds too https://review.openstack.org/435941 | 07:54 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: move to stretch https://review.openstack.org/434453 | 07:54 |
| openstackgerrit | pangliye proposed openstack/kolla-ansible master: Update configuration of Telegraf about mysql https://review.openstack.org/449465 | 07:54 |
| *** matrohon has joined #openstack-kolla | 07:54 | |
| *** jaosorior has joined #openstack-kolla | 07:55 | |
| hrw | can someone review/+2 https://review.openstack.org/449135 and then https://review.openstack.org/449086 one? properly working yum is important | 07:55 |
| *** yangyapeng has joined #openstack-kolla | 07:56 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/kolla master: RHEL: Add mod_ssl for services running over httpd https://review.openstack.org/446911 | 08:02 |
| *** shardy_afk is now known as shardy | 08:09 | |
| duonghq | hrw, done for the 1st | 08:11 |
| hrw | duonghq: thx | 08:12 |
| duonghq | hrw, can you recheck the 2nd one | 08:12 |
| *** mgoddard has joined #openstack-kolla | 08:12 | |
| hrw | duonghq: it requires 1st one to be merged to pass | 08:12 |
| duonghq | ah, yeah, need waiting for it | 08:13 |
| hrw | duonghq: I made 2nd because 1st failed | 08:14 |
| duonghq | hrw, roger | 08:14 |
| duonghq | hrw, need leaving now, cya | 08:18 |
| hrw | cu duonghq | 08:21 |
| *** duonghq has quit IRC | 08:21 | |
| *** mnasiadka has joined #openstack-kolla | 08:23 | |
| *** caowei has joined #openstack-kolla | 08:30 | |
| *** manheim has quit IRC | 08:33 | |
| *** matrohon has quit IRC | 08:37 | |
| *** mgoddard has quit IRC | 08:40 | |
| *** matrohon has joined #openstack-kolla | 08:41 | |
| *** zhubingbing has quit IRC | 08:44 | |
| *** zhubingbing has joined #openstack-kolla | 08:45 | |
| *** magicboiz has quit IRC | 08:45 | |
| *** sean-k-mooney has quit IRC | 08:48 | |
| *** athomas has joined #openstack-kolla | 08:57 | |
| *** zhubingbing has quit IRC | 09:02 | |
| *** manheim has joined #openstack-kolla | 09:03 | |
| *** MarginHu has left #openstack-kolla | 09:06 | |
| *** zhubingbing_ has joined #openstack-kolla | 09:09 | |
| zhubingbing_ | sdake | 09:09 |
| zhubingbing_ | i am not begin, i will start in this nighhe | 09:10 |
| zhubingbing_ | night | 09:10 |
| *** yuanying has quit IRC | 09:12 | |
| *** magicboiz has joined #openstack-kolla | 09:16 | |
| *** mgoddard has joined #openstack-kolla | 09:18 | |
| openstackgerrit | Merged openstack/kolla master: ironic-conductor: add missing comma for centos/source build https://review.openstack.org/449135 | 09:23 |
| *** zhubingbing_ has quit IRC | 09:24 | |
| *** zhubingbing_ has joined #openstack-kolla | 09:25 | |
| *** pbandark has joined #openstack-kolla | 09:29 | |
| *** zhubingbing_ has quit IRC | 09:32 | |
| *** jrist has quit IRC | 09:34 | |
| openstackgerrit | jimmygc proposed openstack/kolla-ansible master: Add VMware vSphere support https://review.openstack.org/449529 | 09:35 |
| *** zhubingbing_ has joined #openstack-kolla | 09:37 | |
| *** caowei has quit IRC | 09:37 | |
| *** egonzalez has joined #openstack-kolla | 09:40 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla stable/newton: ironic-conductor: add missing comma for centos/source build https://review.openstack.org/449535 | 09:41 |
| *** zhuzeyu has quit IRC | 09:43 | |
| *** deadnull has joined #openstack-kolla | 09:43 | |
| *** bmace has quit IRC | 09:48 | |
| *** skramaja_ has joined #openstack-kolla | 09:51 | |
| *** rabel has joined #openstack-kolla | 09:52 | |
| openstackgerrit | jimmygc proposed openstack/kolla-ansible master: Add VMware vSphere support https://review.openstack.org/449529 | 09:54 |
| *** skramaja has quit IRC | 09:54 | |
| *** yuanying has joined #openstack-kolla | 09:54 | |
| *** sean-k-mooney has joined #openstack-kolla | 09:54 | |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 09:55 |
| *** Serlex has joined #openstack-kolla | 09:57 | |
| *** bmace has joined #openstack-kolla | 10:01 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-kubernetes master: Fix problem with kollakube res create configmap https://review.openstack.org/439740 | 10:04 |
| *** pcaruana has joined #openstack-kolla | 10:05 | |
| *** zhubingbing_ has quit IRC | 10:06 | |
| *** cuongnv has quit IRC | 10:07 | |
| *** dixiaoli has joined #openstack-kolla | 10:08 | |
| *** skramaja_ is now known as skramaja | 10:10 | |
| *** tovin07_ has quit IRC | 10:11 | |
| *** egonzalez has quit IRC | 10:15 | |
| *** pbourke has quit IRC | 10:17 | |
| *** pbourke has joined #openstack-kolla | 10:18 | |
| *** manheim has quit IRC | 10:24 | |
| *** manheim has joined #openstack-kolla | 10:26 | |
| *** matrohon has quit IRC | 10:30 | |
| *** pbandark has quit IRC | 10:31 | |
| *** manheim has quit IRC | 10:35 | |
| *** manheim has joined #openstack-kolla | 10:36 | |
| hawi | hi. i do have question - i deployed some machines and because quota limits it failed. now overview shows that memory is in use and cpus are in use, but actually i cant see failed instances. | 10:36 |
| hawi | so - how to free resources again? | 10:37 |
| *** Pavo has joined #openstack-kolla | 10:38 | |
| hawi | i guess something was updated in mariadb and i need to fix it manually? | 10:39 |
| *** zhurong has quit IRC | 10:40 | |
| *** matrohon has joined #openstack-kolla | 10:41 | |
| *** pbandark has joined #openstack-kolla | 10:44 | |
| *** athomas has quit IRC | 10:50 | |
| *** athomas has joined #openstack-kolla | 10:51 | |
| *** athomas has quit IRC | 10:52 | |
| *** zhubingbing_ has joined #openstack-kolla | 10:53 | |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 10:57 |
| *** athomas has joined #openstack-kolla | 10:58 | |
| *** dixiaoli has quit IRC | 11:03 | |
| *** Pavo has quit IRC | 11:04 | |
| *** magicboiz has quit IRC | 11:07 | |
| *** rwallner has joined #openstack-kolla | 11:10 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla master: Fix source install for vmtp https://review.openstack.org/448515 | 11:10 |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 11:11 |
| jaosorior | Hey guys, if you have time could you check these out https://review.openstack.org/#/q/status:open+project:openstack/kolla+branch:master+topic:bug/1675308 ? | 11:11 |
| *** tvignaud has quit IRC | 11:16 | |
| *** Pavo has joined #openstack-kolla | 11:18 | |
| *** magicboiz has joined #openstack-kolla | 11:18 | |
| *** tovin07 is now known as tovin07_at_work | 11:20 | |
| *** tovin07_at_work has quit IRC | 11:20 | |
| pbourke | jaosorior: done | 11:23 |
| jaosorior | thanks pbourke! | 11:23 |
| *** Pavo has quit IRC | 11:25 | |
| spsurya | sup guys | 11:27 |
| *** tvignaud has joined #openstack-kolla | 11:29 | |
| *** tvignaud has quit IRC | 11:37 | |
| *** blallau has joined #openstack-kolla | 11:41 | |
| *** caoyuan_ has quit IRC | 11:49 | |
| *** tvignaud has joined #openstack-kolla | 11:51 | |
| *** Serlex has quit IRC | 11:54 | |
| *** tvignaud has quit IRC | 11:58 | |
| jmccarthy | Hmm ok ansible question, in the keystone bootstrap.yml task, there is a conditional include at the end for bootstrap_service.yml when:database.changed. database is registered in the first task 'Creating keystone database' - but I dont see any 'changed_when' or - how does it know it's changed ? | 11:59 |
| openstackgerrit | jimmygc proposed openstack/kolla master: Add vSphere Support https://review.openstack.org/449395 | 12:08 |
| spsurya | Jeffrey4l: ping... | 12:10 |
| Jeffrey4l | spsurya, pong but sorry, busy. | 12:11 |
| spsurya | Jeffrey4l: np ;) | 12:11 |
| *** tvignaud has joined #openstack-kolla | 12:12 | |
| spsurya | whenever you got few moment please reply to this | 12:12 |
| spsurya | Jeffrey4l: | 12:12 |
| Jeffrey4l | spsurya, reply which ? | 12:13 |
| spsurya | I was re-targeting the all the kolla and kolla-ansible to pike | 12:13 |
| *** rabel has left #openstack-kolla | 12:13 | |
| spsurya | i.e all BPs | 12:14 |
| Jeffrey4l | spsurya, please... | 12:14 |
| Jeffrey4l | i can add u to kolla-drivers. then you can do this. | 12:14 |
| spsurya | Jeffrey4l: please check | 12:14 |
| spsurya | I think i am already there | 12:15 |
| Jeffrey4l | spsurya, yes. you are. | 12:16 |
| Jeffrey4l | then you have permission to do this. | 12:16 |
| *** fooliouno has quit IRC | 12:16 | |
| spsurya | Jeffrey4l: yes | 12:17 |
| spsurya | so moving all to pike-1 | 12:17 |
| Jeffrey4l | thansks ;) | 12:17 |
| *** tvignaud has quit IRC | 12:18 | |
| *** manheim has quit IRC | 12:20 | |
| *** manheim has joined #openstack-kolla | 12:21 | |
| *** mliima has joined #openstack-kolla | 12:23 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Remove Ubuntu 14.04 related code https://review.openstack.org/448792 | 12:23 |
| *** schwicht has joined #openstack-kolla | 12:24 | |
| mliima | morning guyas | 12:27 |
| *** srwilkers has joined #openstack-kolla | 12:29 | |
| *** schwicht has quit IRC | 12:30 | |
| *** yangyapeng has quit IRC | 12:31 | |
| *** tvignaud has joined #openstack-kolla | 12:31 | |
| *** shardy is now known as shardy_lunch | 12:31 | |
| *** caoyuan has joined #openstack-kolla | 12:32 | |
| *** schwicht has joined #openstack-kolla | 12:34 | |
| openstackgerrit | Merged openstack/kolla stable/mitaka: Updated from global requirements https://review.openstack.org/448723 | 12:35 |
| *** jascott1_ has joined #openstack-kolla | 12:36 | |
| *** caoyuan_ has joined #openstack-kolla | 12:36 | |
| *** caoyuan has quit IRC | 12:36 | |
| srwilkers | morning | 12:36 |
| portdirect | morning mliima o/ | 12:36 |
| spsurya | mliima: morning | 12:38 |
| *** jascott1_ has quit IRC | 12:41 | |
| *** schwicht has quit IRC | 12:47 | |
| *** Serlex has joined #openstack-kolla | 12:52 | |
| *** rwallner has quit IRC | 12:55 | |
| *** david-lyle has quit IRC | 12:55 | |
| kolla-slack | <v1k0d3n> morning everyone | 12:55 |
| portdirect | 'morning v1k0d3n. sup dude? | 12:56 |
| *** lamt has joined #openstack-kolla | 12:56 | |
| *** goldyfruit has joined #openstack-kolla | 12:56 | |
| v1k0d3n | nothing portdirect...just reading over things, catching up. TGIF. | 12:57 |
| portdirect | lol - just had to look up TGIF. Indeed :D | 12:57 |
| *** rwallner has joined #openstack-kolla | 12:57 | |
| v1k0d3n | haha welcome to 'merica buddy! :-) | 12:58 |
| *** magicboiz has quit IRC | 13:01 | |
| *** tvignaud has quit IRC | 13:03 | |
| sdake | morning peeps | 13:07 |
| *** schwicht has joined #openstack-kolla | 13:08 | |
| sdake | sbezverk_ ping | 13:09 |
| sbezverk_ | sdake: pong | 13:14 |
| sdake | sbezverk_ there was a report of ironic not working properly (cash loop backoff) related to dependenciesnot setup in the services chart properly | 13:15 |
| sdake | ring any bells? | 13:15 |
| sdake | sbezverk_ heard it third hand so ymmv | 13:15 |
| *** mbruzek has joined #openstack-kolla | 13:15 | |
| spsurya | sdake: morning | 13:16 |
| *** tvignaud has joined #openstack-kolla | 13:16 | |
| sdake | hey spsurya | 13:16 |
| sbezverk_ | sdake: at the gate? | 13:16 |
| sdake | sbezverk_ i think it was not at the gate | 13:16 |
| sdake | sbezverk_ I looked at the pod ith describe | 13:17 |
| sdake | and it didnt'have any depednencies defined at all | 13:17 |
| *** shardy_lunch is now known as shardy | 13:17 | |
| sbezverk_ | sdake: then it must be config issue, my cluster and the gate shows green.. | 13:17 |
| sdake | sbezverk_ if you can login to my boxes for 3 mins and look that might be h elpful | 13:17 |
| sbezverk_ | sdake: sorry, I have few things I need to finish | 13:18 |
| sbezverk_ | sdake: I can try in the afternoon the earliest | 13:19 |
| sdake | sbezverk_ ok thanks | 13:19 |
| sdake | sbezverk_ note this was with 4.0.0 images | 13:19 |
| sdake | which are not gated at all | 13:19 |
| spsurya | Jeffrey4l: sdake: re-targeted all the BPs of kolla and kolla-ansible to pike-1 which were approved and Essential or High or medium priority and the one which was got assignee | 13:19 |
| sdake | I bought a slew of 96khz/24bit msuci last night for 20% off | 13:21 |
| *** yangyapeng has joined #openstack-kolla | 13:21 | |
| *** manheim has quit IRC | 13:21 | |
| spsurya | sdake: nice | 13:21 |
| *** jrist has joined #openstack-kolla | 13:22 | |
| sbezverk_ | sdake: gate jobs run 4.0.0 images | 13:23 |
| sbezverk_ | I mean for ironic | 13:23 |
| sdake | spsurya yup i load it up on my phone for when i have to get on a plane and decode it with: https://www.amazon.com/OPPO-HA-2SE-Portable-Headphone-Amplifier/dp/B01LG8FWSY/ref=sr_1_4?s=audio-video-accessories&ie=UTF8&qid=1490361822&sr=1-4 | 13:24 |
| sdake | spsurya hooks up to my phone via OTG USB | 13:24 |
| sdake | sbezverk_ thats good to hear - so possibly a problem in my environment | 13:24 |
| sdake | sbezverk_ what is wierd is the same exact thing works with 3.0.3-beta1 images | 13:24 |
| sdake | same exact everything - cloud.yaml etc | 13:25 |
| openstackgerrit | Merged openstack/kolla-ansible master: Delete the deprecated option from the group "DEFAULT" https://review.openstack.org/447368 | 13:29 |
| openstackgerrit | Merged openstack/kolla-ansible master: change log-* to flog-*. https://review.openstack.org/449384 | 13:29 |
| sbezverk_ | sdake: there were tons of changes on ironic side introduced by latest PS, I am not surprised. Check my cloud yaml I posted yesterday it has all setting you need to run ironic | 13:30 |
| *** jtriley has joined #openstack-kolla | 13:31 | |
| *** zhubingbing_ has quit IRC | 13:36 | |
| *** zhubingbing_ has joined #openstack-kolla | 13:39 | |
| hawi | i wonder if there is some trick to make central logging working on otaca? i enabled it in globals.yml, got containers on control nodes, but it seems that something wrong with elastic config (should i configure data dir there?) - at least kibana doesnt find log-* | 13:42 |
| hawi | also i tried to enable grafana - nothing happened at all. what i need to enable at same time, if i say "grafana=yes"? | 13:42 |
| *** manheim has joined #openstack-kolla | 13:42 | |
| *** skramaja has quit IRC | 13:48 | |
| *** blallau has quit IRC | 14:01 | |
| *** mdnadeem has quit IRC | 14:06 | |
| mnaser | morning | 14:07 |
| *** dave-mccowan has joined #openstack-kolla | 14:07 | |
| *** lamt has quit IRC | 14:09 | |
| openstackgerrit | Mohammed Naser proposed openstack/kolla master: Merged multiple ENV into a single layer https://review.openstack.org/449371 | 14:14 |
| *** satyar_ has joined #openstack-kolla | 14:17 | |
| *** satyar has quit IRC | 14:18 | |
| *** jbadiapa has quit IRC | 14:18 | |
| *** shashank_t_ has joined #openstack-kolla | 14:22 | |
| *** david-lyle has joined #openstack-kolla | 14:26 | |
| jaosorior | mnaser, sdake hey guys, if you have time could you check these out https://review.openstack.org/#/q/status:open+project:openstack/kolla+branch:master+topic:bug/1675308 ? | 14:30 |
| *** jbadiapa has joined #openstack-kolla | 14:33 | |
| *** matrohon has quit IRC | 14:33 | |
| mnaser | jaosorior i know you did a lot of work on this | 14:35 |
| mnaser | but im a bit curious | 14:35 |
| jaosorior | mnaser: what's up? | 14:35 |
| mnaser | if we can remove a lot of deduplication by adding like | 14:35 |
| mnaser | openstack-base-http | 14:35 |
| mnaser | i know you did a lot of work on those patches but maybe we can remove a lot of deduplication with that? (i'd hate to have to rework all that i know you put in a lot of effort into it) | 14:36 |
| mnaser | but again im not a core so its just my "personal thought" | 14:36 |
| jaosorior | MAYBE that could work? I honestly don't know the code well enough to figure out if the configurations all match up and if we could do that. | 14:37 |
| mnaser | jaosorior i can imagine of a few reason why it wouldnt right now | 14:37 |
| jaosorior | pretty new to working with kolla myself | 14:37 |
| mnaser | ex: nova-api and nova-placement-api need httpd | 14:37 |
| mnaser | nova-*api base off of nova-base | 14:37 |
| jaosorior | is nova-api running over httpd? | 14:38 |
| mnaser | https://review.openstack.org/#/c/446911/8/docker/nova/nova-api/Dockerfile.j2 i guess? | 14:38 |
| jaosorior | cause as far as I knew that was broken | 14:38 |
| mnaser | according to your patch :-P | 14:38 |
| jaosorior | or which nova component was it... | 14:38 |
| *** jascott1_ has joined #openstack-kolla | 14:38 | |
| mnaser | id hope its not | 14:39 |
| jaosorior | mnaser: well, you would have noticed. Since it was "supposed" to be broken for around a month now | 14:39 |
| jaosorior | mnaser: it might have been fixed, which means that I could start re-enabling running nova-api over httpd in TripleO | 14:40 |
| mnaser | jaosorior mm im not sure, in prod we've stuck to running nova-api for now | 14:41 |
| jaosorior | mnaser: so you don't run nova-api over containers? | 14:41 |
| mnaser | jaosorior we do, but stable/newton | 14:41 |
| mnaser | so i believe that uses nova-api | 14:41 |
| jaosorior | ah, that makes sense | 14:42 |
| jaosorior | so yeah, nova-api over httpd in pike is currently not working properly AFAIK. Or at least with the configuration and requirements we have in TripleO | 14:42 |
| *** jascott1_ has quit IRC | 14:43 | |
| *** targon has quit IRC | 14:47 | |
| *** erhudy has quit IRC | 14:50 | |
| jemcevoy | inc0: I have a new issue. kolla-build is failing on the customer site because docker build does not have the http/https man-in-the-middle proxy configured inside the centos7 containers so the rpm --import https://<blah> fails because it needs access to the ca.crt file for the proxy. Any ideas on how to get around this? | 14:51 |
| *** yangyapeng has quit IRC | 14:51 | |
| *** matrohon has joined #openstack-kolla | 14:51 | |
| *** yangyapeng has joined #openstack-kolla | 14:52 | |
| *** haplo37 has quit IRC | 14:55 | |
| *** jbadiapa has quit IRC | 14:55 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Designate: manage Api and Central services workers https://review.openstack.org/444929 | 14:56 |
| *** jbadiapa has joined #openstack-kolla | 14:56 | |
| *** esharao has joined #openstack-kolla | 14:58 | |
| esharao | Hello.. im trying to bring up kolla kubernetes... and i was succuessfully able to bring up all pods except nova-compute and nova-libvirt.. it keeps on crashing.. any advice as to how to proceed? | 14:59 |
| *** lucasxu has joined #openstack-kolla | 14:59 | |
| mandre | jaosorior, mnaser: I like the idea of an openstack-base-http image, that way we would also be able to save a lot of memory on the host if I'm not mistaken | 15:00 |
| mandre | one alternative (to reduce code duplication, but not bits duplication) would be to write an jinja2 macro that installs the required http related packages | 15:01 |
| esharao | when i look at the logs from k8s dashboard i see the last statement in the log as - Running command: 'nova-compute' | 15:03 |
| sdake | esharao which version of the contianers are you using | 15:04 |
| sdake | esharao i find the same result with 4.0.0 contianers | 15:04 |
| sdake | sbezverk_ has indicated this is because the cloud.yaml is incorrect | 15:05 |
| esharao | 4.0.0 | 15:05 |
| sdake | esharao how are you deploying? | 15:05 |
| *** haplo37 has joined #openstack-kolla | 15:05 | |
| esharao | i used the kolla-kubernetes to deploy | 15:05 |
| esharao | i also tried using HELM and it also failed for nova-compute | 15:06 |
| *** fooliouno has joined #openstack-kolla | 15:09 | |
| *** mnasiadka has quit IRC | 15:10 | |
| sdake | esharao what I mean is what set of tools did you use to deploy | 15:11 |
| sdake | esharao or doucmentation | 15:11 |
| *** blallau has joined #openstack-kolla | 15:12 | |
| esharao | i followed the docs (Quickstart guide) | 15:13 |
| esharao | with some changes to files (removal of petsets, etc) | 15:13 |
| sdake | esharao i see - those docs are really old - may I recommend: https://review.openstack.org/#/c/447356/ | 15:14 |
| sdake | esharao i am going to submit a review now to move those docs to a deprecated section | 15:14 |
| sdake | esharao note these docs are incomplete | 15:14 |
| sdake | esharao the one in the review | 15:14 |
| sdake | esharao testing wanted :) | 15:14 |
| sdake | esharao you probbably spent weeks trying to comb through the old docs to get helm operational | 15:15 |
| sdake | esharao atleast i did | 15:15 |
| *** david-lyle has quit IRC | 15:15 | |
| esharao | yep... | 15:15 |
| esharao | 3 weeks | 15:15 |
| esharao | and was able to get it up based on multiple docs.. and stuff.. | 15:15 |
| esharao | i can delete my whole stuff and recreate following this review doc | 15:15 |
| esharao | and will let you know | 15:15 |
| esharao | i only have a single machine.. will this work? | 15:16 |
| esharao | and can i use kube-deploy to get my kubernetes up or do i need to use kubeadm? | 15:17 |
| rwellum | sdake: that's a great qn from esharao - why is 'deployment' hinged on 2-machines? | 15:17 |
| mnaser | wiw has abyieb ever kiijed thriygh thus> | 15:22 |
| mnaser | https://github.com/grammarly/rocker | 15:22 |
| mnaser | oh boy | 15:22 |
| mnaser | i was index+1 on my keyboard | 15:22 |
| mnaser | wow has anyone ever looked through this? | 15:22 |
| sdake | 2 machines - one is master one is minion | 15:23 |
| sdake | I really believe we can taint or untaint the master so we can use one node | 15:23 |
| sdake | I'd like to sort out how to do that | 15:23 |
| *** gkadam has joined #openstack-kolla | 15:24 | |
| esharao | Ok.. and what about kube-deploy vs kubeadm? | 15:24 |
| sdake | kube-deploy is what tool? | 15:24 |
| sdake | is that kargo? | 15:24 |
| esharao | https://github.com/kubernetes/kube-deploy | 15:24 |
| sdake | esharao the only delta we have found with deployment tools is the dns pod has a different name | 15:25 |
| esharao | https://docs.openstack.org/developer/kolla-kubernetes/kubernetes-setup.html | 15:25 |
| esharao | ok..and ubuntu 16.04 for base os works .. i hope.. so will use that and follow this doc and see where it takes me.. thanks sdake | 15:26 |
| sdake | esharao there are some rough spots int he docs that need tuning | 15:27 |
| sdake | esharao specifically aroudn how to create a cloud.yaml file | 15:27 |
| sdake | Ibelieve rwellum is stuck there | 15:27 |
| sdake | also there is a new cloudl.yaml file that sbezverk_ recommends to start with to keep the 4.0.0 images working: http://paste.openstack.org/show/603915/ | 15:28 |
| *** eanylin has joined #openstack-kolla | 15:28 | |
| esharao | ok.. | 15:29 |
| openstackgerrit | Ben Swartzlander proposed openstack/kolla-ansible master: Add missing pip install to quickstart guide https://review.openstack.org/449685 | 15:29 |
| sdake | esharao ubuntu should work for a base os, however, there are small incompatibilities - if your doing an eval i think its fine if your doing a production deploy - i'd recommend against mixing | 15:29 |
| esharao | Ok.. that makes sense.. | 15:29 |
| *** jaosorior has quit IRC | 15:30 | |
| *** hrw has quit IRC | 15:31 | |
| esharao | bbiaf | 15:31 |
| *** athomas has quit IRC | 15:33 | |
| *** caoyuan_ has quit IRC | 15:35 | |
| eanylin | Hi sdake, I used the instructions and were able to bring up the same set of services as seen in your video other than that for nova-compute. You mentioned that there might be a problem with cloud.yaml? | 15:35 |
| sdake | eanylin so instructions worked for you? | 15:36 |
| eanylin | it did | 15:36 |
| sdake | eanylin did you have any modifications? | 15:36 |
| eanylin | I tested on Ubuntu 16.04 | 15:36 |
| *** hrw has joined #openstack-kolla | 15:36 | |
| eanylin | I made modifications to correct the indentation for cinder | 15:36 |
| sdake | eanylin how long did it take you to run through the instructions? | 15:36 |
| eanylin | as pointed out in the etherpad | 15:36 |
| eanylin | it took many nights, unfortunately :) | 15:36 |
| eanylin | but once I got the right config | 15:36 |
| eanylin | it went pretty fast | 15:37 |
| sdake | cool - right config - this is where we are blocked ;) | 15:37 |
| eanylin | maybe 1 to 2 hrs to get to the stage where nova-compute is crashing | 15:37 |
| sdake | 1 or 2 hours isn't super but its pretty good :) | 15:37 |
| eanylin | yeah | 15:37 |
| sdake | (i mean as a general community thing not in your specific case) | 15:37 |
| eanylin | from base CentOS | 15:37 |
| eanylin | with just docker | 15:37 |
| eanylin | I used the latest Docker CE | 15:37 |
| eanylin | since I dont have the older version and I been tearing down my environment multiple times | 15:38 |
| sdake | eanylin regarding the failing nova-compute | 15:38 |
| eanylin | had to skip initial flight test to get the kubeadm init working | 15:38 |
| eanylin | ok | 15:38 |
| sdake | I know the cloud.yaml works well for 3.0.3-beta.1 images | 15:39 |
| eanylin | ok | 15:39 |
| sdake | as in it boots nova properly | 15:39 |
| eanylin | so I should not use tag 4.0.0 in cloud.yaml | 15:39 |
| eanylin | but should use 3.0.3-beta.1? | 15:39 |
| sdake | eanylin 4.0.0 works in the gate | 15:39 |
| sdake | eanylin i'd recommend using 4.0.0 images | 15:39 |
| eanylin | I know your initial config was for 3.0.2 | 15:39 |
| eanylin | ok | 15:39 |
| eanylin | so 4.0.0 is still the right image | 15:39 |
| sdake | eanylin what we need to do is sort out why the 4.0.0 images aren't coming up properly | 15:40 |
| sdake | we know from the gate the 4.0.0 images are good | 15:40 |
| eanylin | I see | 15:40 |
| eanylin | so the images are ok | 15:40 |
| sdake | the 4.0.0 images are immutable | 15:40 |
| eanylin | but somehow the config has some problems | 15:40 |
| sdake | right | 15:40 |
| eanylin | and hence nova-compute is not coming up properly | 15:40 |
| eanylin | I see | 15:40 |
| eanylin | makes senes | 15:40 |
| sdake | and by config - I believe that is cloud.yaml :) | 15:40 |
| eanylin | sense* | 15:40 |
| eanylin | yeah, I think so | 15:40 |
| sdake | this is my current line of analysis | 15:40 |
| eanylin | thats the one that I fed to helm | 15:41 |
| eanylin | based on your sample :) | 15:41 |
| sdake | eanylin are you using 2 nodes or 1 | 15:41 |
| eanylin | I only have 1 laptop actually | 15:41 |
| eanylin | where I used the wifi card and eth0 | 15:41 |
| eanylin | as the 2 interface | 15:41 |
| eanylin | wasnt sure if this is good enough | 15:41 |
| sdake | that should work i think - but not certain | 15:41 |
| eanylin | I have another more powerful laptop that can create 2 VMs | 15:42 |
| eanylin | ok | 15:42 |
| sdake | one vm should be fine - i'm curious how you got controllers/computes on one vm | 15:42 |
| eanylin | will VM work or we definitely need bare metal ? | 15:42 |
| sdake | rwellum had asked above | 15:42 |
| eanylin | tainted it | 15:42 |
| eanylin | then it will allow running of all containers isnt it? | 15:42 |
| eanylin | I got libvirt-bin up | 15:42 |
| eanylin | as in that container | 15:42 |
| sdake | eanylin thanks - tainted it - could you add that to this review: https://review.openstack.org/#/c/447356/ | 15:43 |
| eanylin | sure | 15:43 |
| sdake | eanylin with exact steps you followed | 15:43 |
| sdake | to taint it | 15:43 |
| eanylin | its part of the kubeadm set up | 15:43 |
| eanylin | if you look at the webpage to set up kubeadm | 15:43 |
| sdake | i must have skipped over that part of the docs :) | 15:43 |
| eanylin | https://kubernetes.io/docs/getting-started-guides/kubeadm/#instructions | 15:43 |
| eanylin | Step 2 says that we can taint it if we only have 1 node for dev work | 15:44 |
| sdake | # MASTER_NODE="actual_master_node_name" | 15:44 |
| sdake | # TAINT_KEY=$(kubectl get no ${MASTER_NODE} --template="") | 15:44 |
| sdake | # kubectl taint nodes ${MASTER_NODE} ${TAINT_KEY}- | 15:44 |
| eanylin | I think I used the old command | 15:45 |
| eanylin | not this one | 15:45 |
| eanylin | they updated this instructions as well I think | 15:45 |
| sdake | used old command where - in that review? | 15:45 |
| sdake | eanylin kolla-kubernetes has moved most of the deploy guide in the etherpad to the review here: https://review.openstack.org/#/c/447356/ - it has undergone some change as well | 15:46 |
| eanylin | I am quite sure the old ones work | 15:47 |
| eanylin | I tried the new one and I think I had problems | 15:47 |
| eanylin | # kubectl taint nodes --all dedicated- | 15:47 |
| eanylin | node "test-01" tainted | 15:47 |
| eanylin | taint key="dedicated" and effect="" not found. | 15:47 |
| eanylin | taint key="dedicated" and effect="" not found. | 15:47 |
| eanylin | so this one will work | 15:47 |
| eanylin | yes, its in the review, it can be found in this link, http://deploy-preview-1321.kubernetes-io-vnext-staging.netlify.com/docs/getting-started-guides/kubeadm/ | 15:47 |
| eanylin | which is mentioned | 15:47 |
| *** vhosakot has joined #openstack-kolla | 15:48 | |
| sdake | so just: https://review.openstack.org/#/c/447356/ | 15:48 |
| eanylin | so I was able to get to the same kind of failure with weave net as well as flannel deploy using master.sh (kube-deploy) | 15:48 |
| eanylin | I mena with nova-compute | 15:48 |
| eanylin | mean* | 15:48 |
| sdake | rathe so just : # kubectl taint nodes --all dedicated- | 15:48 |
| eanylin | yeah, that should work | 15:48 |
| eanylin | for CentOS 7 | 15:48 |
| eanylin | and latest docker CE | 15:48 |
| eanylin | and kubeadm | 15:48 |
| eanylin | on single node | 15:48 |
| sdake | eanylin nice! | 15:49 |
| vhosakot | sending this message from my IRC client to test if this is seen in my slack client... testing 1 2 3 a b c $%^&*{}[] | 15:49 |
| kolla-slack | <vhosakot> sending this message from my slack client to test if this is seen in my IRC client... testing 1 2 3 a b c $%^&*{}[] | 15:49 |
| fooliouno | sdake: enaylin: FWIW, I brought up 4.0.0 images using the old workflow, and nova-compute is a crashloop there as well. | 15:49 |
| eanylin | ok | 15:49 |
| eanylin | yeha, same | 15:49 |
| eanylin | the video was very helpful, I can see it was working at that time | 15:49 |
| sdake | yup it does indeed work :) | 15:50 |
| sdake | its a config problem | 15:50 |
| eanylin | the console shows "nova-compute" and crash | 15:50 |
| eanylin | yeah | 15:50 |
| eanylin | brb | 15:50 |
| vhosakot | hmm... if I'm logged into both IRC client and slack client, messaged sent from slack client are seen with username kolla-slack in my IRC client | 15:50 |
| sdake | vhosakot yup thats how the integration works | 15:51 |
| rwellum | fyi weave never worked for me, had to use canal or calico as an SDN. That was an issue for me in the developer guide - that it recommends weave that seems to be unreliable for most. | 15:51 |
| vhosakot | sdake: so, to chat in the #opensatck-kolla, should we use slack or IRC? | 15:51 |
| sdake | vhosakot your choice | 15:51 |
| vhosakot | I prefer IRC :) | 15:52 |
| vhosakot | i use IRC to chat in other IRC rooms like neutron, nova, glance, etc | 15:52 |
| sdake | ok - well I've got 1.5 hrs, wouldl folks be interested in a webex session with screen sharing to sort out the config bits? | 15:52 |
| fooliouno | eanylin: would you mind pasting the cloud.yml that worked for you on Ubuntu16.04 (on pastebin). Did you use kubernetes 1.5.4 or 1.5.2? | 15:52 |
| vhosakot | also, the slack emojis are not seen in IRC client :) | 15:52 |
| rwellum | sdake: +1 | 15:53 |
| fooliouno | vhosakot: Did you resolve your dashboard issues? | 15:54 |
| *** Pavo has joined #openstack-kolla | 15:55 | |
| fooliouno | rwellum: Weave failed for me when ovs containers deployed | 15:56 |
| vhosakot | fooliouno: no, thanks for your help. so, I've found that kubeadm needs service account creation for both dashboard and "helm init" to work so the secrets/certs are setup correctly. I need to fix my service account in kubeadm before proceeding. I am making slow progress as I'm working on many things now :) | 15:56 |
| *** manheim has quit IRC | 15:56 | |
| fooliouno | Worked like a charm until then | 15:56 |
| *** manheim has joined #openstack-kolla | 15:56 | |
| vhosakot | fooliouno: https://kubernetes.io/docs/user-guide/secrets/#service-accounts-automatically-create-and-attach-secrets-with-api-credentials is what I need to do after deploying kubeadm. | 15:57 |
| vhosakot | fooliouno: do you have dashboard working on kubeadm? | 15:57 |
| fooliouno | vhosakot: cool. Yes. I have kubeadm and dashboard. | 15:57 |
| fooliouno | Looks like its scheduled on the kube master right now. So its up. Will try to recreate it on a minion and see if it works. | 15:58 |
| vhosakot | fooliouno: did dashboard work out of the box for you? you did "kubectl create -f https://raw.githubusercontent.com/kubernetes/dashboard/v1.6.0/src/deploy/kubernetes-dashboard.yaml", and boom the dashboard was up? | 15:58 |
| fooliouno | One other thing. If you are using Flannel and docker 1.13, you may be tripped up by IP forwarding rules. | 15:58 |
| vhosakot | fooliouno: I'm using canal which sbezverk_ and kfox1111 recommended. | 15:59 |
| fooliouno | Yes .. it did (when scheduled on master) | 15:59 |
| *** krtaylor has joined #openstack-kolla | 15:59 | |
| fooliouno | Canal may have the same issue since it uses Flannel underneath | 15:59 |
| sdake | i've got 1.5 hrs to go through the workflow on single node and sort out that review doc | 16:00 |
| sdake | if folks want to join in, http://cisco.webex.com/join/stdake | 16:00 |
| fooliouno | sdake: I will listen in | 16:00 |
| mnaser | oracle linux mirrors having any problems? | 16:00 |
| mnaser | oraclelinux jobs are timing out | 16:01 |
| vhosakot | fooliouno: how did you pin the dashboard pod to the master? edit the dashboard sepc YAML and add label+nodeSelector? | 16:01 |
| *** manheim has quit IRC | 16:01 | |
| sdake | mnaser oracle linux uses rdo mirrors i believe | 16:01 |
| mnaser | only for the openstack packages | 16:01 |
| mnaser | $os packages still use their mirros | 16:01 |
| sdake | mnaser also for infrastructure pakages | 16:01 |
| fooliouno | No .. it was scheduled that way by kubernetes. If it fails, delete and retry | 16:01 |
| vhosakot | fooliouno: ah, when I re-try, it ends on one of the masters... I thought no pod is scheduled on the master except the k8s control-plane pods (scheduler, controller-manager, etcd, api-server), right? | 16:03 |
| sdake | fooliouno note you can login with a web-browser only so no need to download a client | 16:03 |
| *** pcaruana has quit IRC | 16:03 | |
| vhosakot | fooliouno: one of the workers I meant (not one of the masters) | 16:03 |
| fooliouno | dashboard has atolerance to schedule on master | 16:03 |
| vhosakot | fooliouno: I read about it... where is the tolerence? let me read the dashboard spec.. | 16:04 |
| fooliouno | In the yaml | 16:04 |
| sdake | srujan can you click "conenct to audio" | 16:05 |
| sdake | anyone else interestedi n participating before i get started giving the instructions a dry run? | 16:05 |
| sdake | rwellum ? | 16:05 |
| *** Pavo has quit IRC | 16:06 | |
| vhosakot | fooliouno: is this the tolerence annotation --> https://github.com/kubernetes/dashboard/blob/master/src/deploy/kubernetes-dashboard.yaml#L38-L46 | 16:06 |
| vhosakot | fooliouno: did you have to change this in the dashboard YAML? --> https://github.com/kubernetes/dashboard/blob/master/src/deploy/kubernetes-dashboard.yaml#L59 | 16:07 |
| eanylin | http://pastebin.com/GUw5G2XJ | 16:08 |
| vhosakot | sdake: we're moving our lab now, I'll move me kubeadm cluster, fix my secrets, and ping when I get "helm init" working to continue | 16:08 |
| eanylin | its similar to what sdake did | 16:08 |
| eanylin | I used similar config | 16:08 |
| vhosakot | turns out "helm init" and dashboard both not working out-of-the-box with kubeadm is a pretty common problem I found. | 16:09 |
| eanylin | my k8s server version is 1.5.5 | 16:10 |
| *** Pavo has joined #openstack-kolla | 16:12 | |
| *** manheim has joined #openstack-kolla | 16:13 | |
| *** vhosakot_ has joined #openstack-kolla | 16:14 | |
| *** vhosakot has quit IRC | 16:14 | |
| fooliouno | vhosakot: sorry on sdake webex. will look and respond later | 16:16 |
| vhosakot_ | cool. np | 16:18 |
| *** skramaja has joined #openstack-kolla | 16:19 | |
| *** pbandark has quit IRC | 16:20 | |
| *** Pavo has quit IRC | 16:21 | |
| *** manheim has quit IRC | 16:22 | |
| *** manheim has joined #openstack-kolla | 16:22 | |
| *** lamt has joined #openstack-kolla | 16:23 | |
| *** mattmceuen has joined #openstack-kolla | 16:24 | |
| *** blallau has quit IRC | 16:24 | |
| *** manheim has quit IRC | 16:27 | |
| eanylin | @vhosakot, you should be able to install dashboard by | 16:28 |
| eanylin | kubectl create -f https://rawgit.com/kubernetes/dashboard/master/src/deploy/kubernetes-dashboard.yaml | 16:28 |
| eanylin | based on instructions from https://github.com/kubernetes/dashboard | 16:28 |
| *** kolla-slack has quit IRC | 16:32 | |
| *** david-lyle has joined #openstack-kolla | 16:33 | |
| *** shashank_t_ has quit IRC | 16:35 | |
| *** shashank_t_ has joined #openstack-kolla | 16:35 | |
| *** jascott1_ has joined #openstack-kolla | 16:36 | |
| *** jascott1_ has quit IRC | 16:36 | |
| *** jascott1_ has joined #openstack-kolla | 16:42 | |
| *** MasterOfBugs has joined #openstack-kolla | 16:46 | |
| *** matrohon has quit IRC | 16:52 | |
| *** Serlex has quit IRC | 16:56 | |
| *** rwsu has quit IRC | 16:57 | |
| *** rwsu has joined #openstack-kolla | 17:00 | |
| *** Serlex has joined #openstack-kolla | 17:03 | |
| *** lamt has quit IRC | 17:17 | |
| vhosakot_ | eanylin: I did that, and it does not work with kubeadm :) | 17:18 |
| *** unicell has quit IRC | 17:19 | |
| *** jrich523 has joined #openstack-kolla | 17:21 | |
| *** jmccarthy has left #openstack-kolla | 17:22 | |
| *** jrich523 has quit IRC | 17:22 | |
| eanylin | I think I tested yesterday night, it was ok with version 1.6.0 | 17:23 |
| eanylin | image: gcr.io/google_containers/kubernetes-dashboard-amd64:v1.6.0 | 17:23 |
| eanylin | you can try and update the yaml file? | 17:23 |
| *** schwicht has quit IRC | 17:26 | |
| *** Serlex has quit IRC | 17:27 | |
| *** sayantani01 has joined #openstack-kolla | 17:28 | |
| *** schwicht has joined #openstack-kolla | 17:28 | |
| *** jrich523 has joined #openstack-kolla | 17:28 | |
| *** lucasxu has quit IRC | 17:36 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla-kubernetes master: Move glance config to kolla-k8s https://review.openstack.org/449755 | 17:38 |
| *** unicell has joined #openstack-kolla | 17:45 | |
| *** manheim has joined #openstack-kolla | 17:49 | |
| *** manheim has quit IRC | 17:49 | |
| *** unicell has quit IRC | 17:49 | |
| *** unicell has joined #openstack-kolla | 17:50 | |
| *** manheim has joined #openstack-kolla | 17:50 | |
| *** mgoddard has quit IRC | 17:58 | |
| *** sayantani01 has quit IRC | 17:58 | |
| *** sayantani01 has joined #openstack-kolla | 17:59 | |
| *** zhubingbing_ has quit IRC | 18:00 | |
| *** skramaja has quit IRC | 18:00 | |
| *** Serlex has joined #openstack-kolla | 18:06 | |
| *** manheim has quit IRC | 18:14 | |
| *** manheim has joined #openstack-kolla | 18:15 | |
| *** gkadam has quit IRC | 18:15 | |
| *** manheim has quit IRC | 18:24 | |
| *** MasterOfBugs has quit IRC | 18:28 | |
| *** mgoddard has joined #openstack-kolla | 18:35 | |
| *** lamt has joined #openstack-kolla | 18:39 | |
| *** shashank_t_ has quit IRC | 18:40 | |
| *** MasterOfBugs has joined #openstack-kolla | 18:46 | |
| sbezverk_ | inc0: ping | 18:50 |
| *** lucasxu has joined #openstack-kolla | 18:50 | |
| inc0 | whats up Serguey? | 18:50 |
| *** shashank_t_ has joined #openstack-kolla | 18:50 | |
| sbezverk_ | inc0: do you know when this is going to be fixed? https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/nova/templates/qemu.conf.j2#L3 | 18:51 |
| sbezverk_ | inc0: it breaks ubuntu ironic deployment | 18:51 |
| sbezverk_ | inc0: as I am getting permission denied when I try to access from the host libvirtd socket | 18:52 |
| sbezverk_ | inc0: I see it in 4.0.0 images | 18:52 |
| inc0 | hmm hard to tell | 18:58 |
| sbezverk_ | inc0: :( cause you see, libvirt runs as nova in the container, but on the host nova user is not known | 19:00 |
| sbezverk_ | inc0: maybe I could define user nova with the same user id and try to run virsh not with sudo bu with user nova id? | 19:01 |
| jemcevoy | inc0: I have a new issue. kolla-build is failing on the customer site because docker build does not have the http/https man-in-the-middle proxy configured inside the centos7 containers so the rpm --import https://<blah> fails because it needs access to the ca.crt file for the proxy. Any ideas on how to get around this? | 19:02 |
| inc0 | jemcevoy: well, you can use overrides | 19:04 |
| inc0 | to inject files/configs to container buildtime | 19:04 |
| inc0 | hold on | 19:04 |
| inc0 | sbezverk_: I'll get back to you in a second, having few discussions at once and yours needs full attention;) | 19:05 |
| inc0 | https://docs.openstack.org/developer/kolla/image-building.html#generic-customisation | 19:05 |
| inc0 | jemcevoy: ^ | 19:05 |
| inc0 | jemcevoy: https://github.com/openstack/kolla/blob/master/docker/base/Dockerfile.j2#L16 you can extend this block to inject any piece of dockerfile-compatible code at the very beggining of base image | 19:07 |
| inc0 | or look for block better suited for your needs | 19:07 |
| Serlex | sdake | 19:08 |
| jemcevoy | That sounds like a perfect solution... Looking into it. | 19:08 |
| Serlex | did the second part of that webex happen? | 19:08 |
| sdake | Serlex give me 20 mins | 19:08 |
| sdake | Serlex and i'll get it going - need to feed myself | 19:08 |
| sdake | serlex possibly 30 | 19:08 |
| *** tonanhngo has joined #openstack-kolla | 19:09 | |
| *** tonanhngo has quit IRC | 19:09 | |
| sbezverk_ | inc0: sure sure no rush | 19:10 |
| Serlex | ok - but we should ask around if the others are interested as I was purely spectating | 19:12 |
| inc0 | sbezverk_: right | 19:12 |
| inc0 | so issue was vm files | 19:13 |
| inc0 | with wrong permissions | 19:13 |
| inc0 | my question is...why do you try to access libvirt from host? | 19:14 |
| inc0 | I mean...we run everything in container | 19:14 |
| *** mgoddard has quit IRC | 19:16 | |
| sbezverk_ | inc0: it is only at the gate when I need to define baremetal VM using virsh | 19:16 |
| inc0 | well yeah adding user with same uid as used by nova in container might work | 19:17 |
| sbezverk_ | inc0: since it is a very specific scenario I think the trick with the user should be good enough | 19:17 |
| sbezverk_ | inc0: do you remember from the top of your head where or how user ids are getting generated, I think at one point you were going to use a static ids, is that right? | 19:18 |
| inc0 | sbezverk_: https://github.com/openstack/kolla/blob/master/docker/base/Dockerfile.j2#L7 | 19:19 |
| *** tonanhngo has joined #openstack-kolla | 19:19 | |
| inc0 | https://github.com/openstack/kolla/blob/master/kolla/common/config.py#L543 | 19:19 |
| sbezverk_ | inc0: right but where user.uid comes from? | 19:19 |
| inc0 | second link | 19:19 |
| sbezverk_ | inc0: cool, missed it | 19:20 |
| sbezverk_ | thanks a lot! | 19:20 |
| kfox1111 | sbezverk_: can you exec into the container and use its virsh? | 19:20 |
| *** erhudy has joined #openstack-kolla | 19:20 | |
| inc0 | alias virsh = docker exec -it nova_libvirt virsh :D | 19:21 |
| inc0 | kfox1111: https://review.openstack.org/#/c/449254/ | 19:21 |
| kfox1111 | inc0: won't quite work.... isn't named that way. ;) | 19:22 |
| kfox1111 | reviewing. | 19:22 |
| *** yangyapeng has quit IRC | 19:24 | |
| sbezverk_ | kfox1111: hm, but then I need somehow dump xml into the container | 19:24 |
| *** yangyapeng has joined #openstack-kolla | 19:24 | |
| sbezverk_ | kfox1111: adding nova user on the host with the same uid/guid should do the trick I think | 19:25 |
| kfox1111 | sbezverk_: cat foo.xml | docker exec ... 'virsh <(cat)' ? | 19:26 |
| inc0 | my god | 19:26 |
| kfox1111 | hehe | 19:26 |
| kfox1111 | shellfu. :) | 19:26 |
| *** shashank_t_ has quit IRC | 19:27 | |
| *** shashank_t_ has joined #openstack-kolla | 19:27 | |
| inc0 | shell fu - bashing faces | 19:27 |
| kfox1111 | :) | 19:28 |
| inc0 | bash and boot | 19:28 |
| sdake | ok | 19:28 |
| sbezverk_ | kfox1111: why not :) | 19:28 |
| sdake | nova-compute and nova-compute-ironic don't come up in my env | 19:28 |
| sdake | any pointers? | 19:28 |
| sdake | with 4.0.0 images | 19:28 |
| sdake | with sbezverk_ 's cloud.yaml | 19:28 |
| sbezverk_ | sdake: what they are complaning about? | 19:28 |
| sdake | they enter running state then enter a crashlopbackoff | 19:28 |
| sdake | sbezverk_ not sure what to check | 19:29 |
| sbezverk_ | sdake: what is the crash reason? | 19:29 |
| sdake | sbezverk_ run docker logs on em ? | 19:29 |
| Serlex | did you create initramfs .kernel files sdake? | 19:29 |
| sdake | Serlex not sure | 19:29 |
| sbezverk_ | Serlex: initramfs is not needed for containers to run | 19:30 |
| sbezverk_ | only to deply baremetal | 19:30 |
| Serlex | actually if it deployed, yeah you are past that requirement, my bad | 19:30 |
| sdake | sbezverk_ how should I debug? | 19:30 |
| sdake | sbezverk_ everyone in the community stuck at same point with 4.0.0 images | 19:31 |
| sdake | sbezverk_ that has managed to get past the deploy kubernetes gauntlet | 19:31 |
| *** shashank_t_ has quit IRC | 19:31 | |
| sbezverk_ | sdake: why do you deploy ironic? have you tried just nova compute for fun and see if it is working? | 19:32 |
| sbezverk_ | have you deployed nva placement | 19:32 |
| sdake | nova placement... | 19:32 |
| sbezverk_ | have you deplyed nova cell0? | 19:32 |
| sdake | letm e look | 19:32 |
| kfox1111 | sdake: 4.0.0 images are.... lightly tested. | 19:32 |
| sdake | kfox1111 i understand | 19:32 |
| kfox1111 | we need some gates for them. | 19:32 |
| openstackgerrit | Rich Wellum proposed openstack/kolla-kubernetes master: setup_gate.sh echos non-quoted params https://review.openstack.org/449790 | 19:32 |
| sbezverk_ | sdake: the easiet is to take the latest deploy scripts from the gate and see what you missed | 19:33 |
| sbezverk_ | I can guarantee you, you missed something | 19:33 |
| sdake | the deploy scripts deploy 4.0.0? | 19:33 |
| sbezverk_ | sdake: please check ironic gate job | 19:34 |
| sdake | sbezverk_ thanks | 19:34 |
| sbezverk_ | you will see two main scripts one generic deployment with iscsi backend and second is ironic specific script | 19:34 |
| sbezverk_ | just follwo them | 19:34 |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla-kubernetes master: Move glance config to kolla-k8s https://review.openstack.org/449755 | 19:35 |
| sdake | sbezverk_ is the placement api and cell0 in the nova service chart? | 19:36 |
| sbezverk_ | nope | 19:36 |
| sbezverk_ | because they will fail on old images | 19:36 |
| sbezverk_ | I could not put them there | 19:36 |
| sbezverk_ | you need to deploy them manually after the service | 19:37 |
| jemcevoy | inc0: I had to fix a bug in base_yum_repo_keys in that same file: The GPG key should be https://www.percona.com/downloads/RPM-GPG-KEY-percona not a local file... https://github.com/openstack/kolla/blob/master/docker/base/Dockerfile.j2#L63 | 19:37 |
| sbezverk_ | sdake: we can only put them in services after all images are 4.0.0 and higher | 19:37 |
| sdake | how about we make a nova service chart for ocata | 19:38 |
| sdake | or alternately remove 2.0 and 3.0 images :) | 19:38 |
| kfox1111 | sbezverk_: we should put in a conditional dep. | 19:43 |
| sbezverk_ | what will you check? in this dependency?? ;) | 19:48 |
| inc0 | can we please remove 2 and 3 images? pretty pretty please? | 19:48 |
| sbezverk_ | kfox1111: needs to confirm that with 4.0.0 no issues with ceph | 19:49 |
| inc0 | why would it? | 19:49 |
| sbezverk_ | inc0: it seems kfox1111 has laready tried in the past and there was a problem with cinder volumes | 19:52 |
| sbezverk_ | kfox1111: unfortunately crazy bash thing will not work, as vbmc uses normal way to start/stop vm :( | 19:53 |
| sdake | kfox1111 how do you do a conditional dep | 20:06 |
| inc0 | kfox1111: https://review.openstack.org/#/c/449755/ these gates seems to fail for reasons other than patchset..? | 20:08 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes master: Fixing ironic deployment failure on Ubuntu https://review.openstack.org/449803 | 20:09 |
| sdake | kfox1111 need m0ar info on conditonal dep plz | 20:10 |
| sbezverk_ | sdake: making it conditional will not help much, because there is no way reliably identify what images you use. | 20:12 |
| sbezverk_ | they can be overridden at multiple levels. | 20:13 |
| sdake | we could set a "deploy_ocata" field | 20:13 |
| sbezverk_ | sdake: and in pike we change it for deploy_pike??? it does not sound as a good solution | 20:13 |
| sdake | a solution that works isbetter then no solution at all :) | 20:14 |
| sbezverk_ | why do not you deploy it just after that manually ?? it is not big deal.. | 20:14 |
| sbezverk_ | until we sort out images issue | 20:14 |
| sdake | we will always have this problem of images being added to services in new versions | 20:14 |
| sdake | why have a nova service chart if some of the services must be deployed manually? | 20:15 |
| sdake | deploying manually a bunch of extra services is not very intuitive (to answer your question) | 20:16 |
| sdake | and compute-kit won't deploy the whole schebang | 20:16 |
| sdake | helm install compute-kit | 20:16 |
| sdake | bam - 4.0.0 images should work with that use casse | 20:16 |
| sbezverk_ | sdake: we have not release kolla-kube | 20:16 |
| sbezverk_ | so now it is used mostly by developers | 20:17 |
| sbezverk_ | of course at the release time all these will be sorred out | 20:17 |
| sdake | how about a legacy_images flag | 20:17 |
| sdake | that points backwards instead of deploy_ocata which is a point in time | 20:18 |
| sbezverk_ | man :( 1.5.5 kube is broken all gates are red | 20:18 |
| sbezverk_ | we should stop pulling for the latest kube | 20:19 |
| sbezverk_ | kfox1111: what do you think? | 20:19 |
| sdake | that kind of thiking leads up down the road of keeping 2.0.0 images in the tree ;) | 20:19 |
| sbezverk_ | sdake: yeah of course it is much better to have all gate jobs red | 20:20 |
| sdake | i didn't say it was | 20:20 |
| sdake | a temporary pin is fine - a permanent pin not fine - temporary often turns into permanent in kolla-kubernetes... | 20:20 |
| *** mliima has quit IRC | 20:21 | |
| *** fooliouno has quit IRC | 20:31 | |
| sdake | inc0 when you get bored, you might have a look at updating this: https://review.openstack.org/#/c/346455/ | 20:32 |
| *** rwallner has quit IRC | 20:33 | |
| sdake | inc0 ttx looking for an update and your it :) | 20:35 |
| rwellum | sdake: I missed k8's meeting part 2 and it all works right? :p | 20:35 |
| sdake | rwellum i haven' theld a part 2 | 20:35 |
| sdake | i was thinking on the hour we can do that - had some things creep into my schedule | 20:36 |
| rwellum | sdake: It's Friday too... | 20:36 |
| sdake | rwellum I am pretty sure we have the info now to get past the problem everyone is suffering with wrt nova and 4.0.0 images | 20:36 |
| sdake | rwellum right - thats 5pm your time? | 20:36 |
| sdake | or 4pm? | 20:36 |
| rwellum | Depends are we rounding up or down? | 20:37 |
| rwellum | 4:37 | 20:37 |
| sdake | its 1;37 PM now where I am | 20:37 |
| inc0 | sdake: I'll reup it today, thanks | 20:37 |
| sdake | ok so on the hour would be 5pm your time | 20:37 |
| sdake | thans inc0 | 20:37 |
| inc0 | somehow I missed this review | 20:37 |
| sdake | rwellum is that too late to go - for 1 hour? | 20:37 |
| sdake | inc0 yup happens to me all the time | 20:38 |
| rwellum | sdake: unless someone @home complains it's good. | 20:38 |
| *** imcsk8 has quit IRC | 20:38 | |
| rwellum | sdake: I thought you were MST for some reason. But you're further West I guess. | 20:39 |
| sdake | rwellum I am MST year round | 20:40 |
| sdake | although right now most people in M?T are MDT | 20:40 |
| sdake | MDT is often abbreviated MT | 20:41 |
| sdake | its confusing for everyone that is on daylight savings | 20:41 |
| kfox1111 | what a day. :/ | 20:41 |
| sdake | for people not on daylight savings, its not confusing :) | 20:41 |
| kfox1111 | all the computers are rebelling. :/ | 20:41 |
| sdake | atleast its not the 13th | 20:41 |
| rwellum | Yeah I'd expect a 2 hr difference not 3 hr. | 20:41 |
| kfox1111 | sdake: pretty sure it is somehow the 13th today. :/ | 20:42 |
| sdake | kfox1111 and full moon :) | 20:42 |
| kfox1111 | yeah. that too. | 20:42 |
| kfox1111 | there are warewolfs out for sure. :/ | 20:42 |
| sdake | rwellum if you type "what time is it in arizona" it pops out the proper time :) | 20:42 |
| sdake | type into google that is | 20:42 |
| kfox1111 | sdake: https://github.com/kubernetes/helm/issues/1585 | 20:43 |
| rwellum | Oh right - that's not a cool feature of irc then... | 20:44 |
| *** srwilkers has quit IRC | 20:44 | |
| sdake | rwellum sorry for the duh moment :) | 20:45 |
| *** imcsk8 has joined #openstack-kolla | 20:45 | |
| sdake | rwellum timeanddate says MST: https://www.timeanddate.com/worldclock/usa/phoenix | 20:46 |
| rwellum | that's actually pretty interesting... Phoenix has chosen not to observe DST. | 20:48 |
| rwellum | It's a per-city choice? | 20:48 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: Helm Version File https://review.openstack.org/449814 | 20:51 |
| *** matrohon has joined #openstack-kolla | 20:52 | |
| *** imcsk8 has quit IRC | 20:54 | |
| *** imcsk8 has joined #openstack-kolla | 20:56 | |
| *** jtriley has quit IRC | 21:00 | |
| *** dave-mccowan has quit IRC | 21:01 | |
| sdake | rwellum all of az except parts of the reservation don't observe dst | 21:02 |
| sdake | setting up webex now | 21:02 |
| sdake | for folks that want to make a go at bringing up kubernetes + kolla_kubernetes in an AIO environment | 21:02 |
| sdake | continuation from this morning session | 21:02 |
| sdake | http://cisco.webex.com/join/stdake | 21:03 |
| *** eaguilar has joined #openstack-kolla | 21:04 | |
| *** eaguilar has quit IRC | 21:05 | |
| sdake | rwellum ^^ | 21:06 |
| *** pramodrj07 has joined #openstack-kolla | 21:16 | |
| *** mattmceuen has quit IRC | 21:18 | |
| esharao | https://github.com/kubernetes/kubernetes/issues/38118 | 21:19 |
| *** MasterOfBugs has quit IRC | 21:20 | |
| *** jascott1 has quit IRC | 21:21 | |
| *** shasha_t_ has quit IRC | 21:21 | |
| *** pbandark has joined #openstack-kolla | 21:24 | |
| *** pbandark has quit IRC | 21:24 | |
| *** eaguilar has joined #openstack-kolla | 21:26 | |
| *** satyar_ has quit IRC | 21:42 | |
| kfox1111 | the evermind seems mad at me today. | 21:42 |
| kfox1111 | all sorts of computers are rebelling. | 21:42 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: Helm Native Values. https://review.openstack.org/449826 | 21:44 |
| dims | kfox1111 : LOL | 21:45 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: Helm Native Values. https://review.openstack.org/449826 | 21:46 |
| *** Pavo has joined #openstack-kolla | 21:48 | |
| eanylin | https://review.openstack.org/#/c/439409/1/kolla_kubernetes/pathfinder.py | 21:49 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: Helm Version File https://review.openstack.org/449814 | 21:50 |
| eanylin | http://paste.openstack.org/raw/600976/ | 21:55 |
| *** Serlex has quit IRC | 21:59 | |
| *** sayantani01 has quit IRC | 22:01 | |
| *** pramodrj07 has quit IRC | 22:03 | |
| *** pramodrj07 has joined #openstack-kolla | 22:03 | |
| eanylin | $ curl https://raw.githubusercontent.com/kubernetes/helm/master/scripts/get > get_helm.sh | 22:06 |
| eanylin | $ chmod 700 get_helm.sh | 22:06 |
| eanylin | $ ./get_helm.sh | 22:06 |
| *** schwicht has quit IRC | 22:07 | |
| eanylin | https://github.com/kubernetes/helm/blob/master/docs/install.md | 22:07 |
| *** matrohon has quit IRC | 22:19 | |
| sbezverk_ | rwellum: ping | 22:21 |
| *** esharao has quit IRC | 22:21 | |
| *** lucasxu has quit IRC | 22:27 | |
| *** tonanhngo has quit IRC | 22:30 | |
| *** schwicht has joined #openstack-kolla | 22:30 | |
| sbezverk_ | kfox1111: should we pin 1.5.4 at the gate until 1.5.5 gets fixed? | 22:34 |
| kfox1111 | sure. | 22:38 |
| kfox1111 | do you have a log of a broken 1.5.5 run? | 22:38 |
| kfox1111 | nm. I've got a log. | 22:38 |
| sbezverk_ | kfox1111: all jobs are failing | 22:40 |
| sbezverk_ | I checked and I could not see any obvious reason other than api and other containers are restarting | 22:40 |
| kfox1111 | weird. the changelog says its just a security fix. | 22:41 |
| sbezverk_ | kfox1111: maybe bad build? | 22:41 |
| kfox1111 | did canal change? | 22:42 |
| sbezverk_ | do not think so, and it seems it was not restarting (surprisingly) | 22:42 |
| kfox1111 | wondering if they changed ip's again. | 22:43 |
| kfox1111 | nope. they are still the same... | 22:45 |
| kfox1111 | so, kube-dns is consistently failing. | 22:50 |
| sdake | kube-dns renamed possibly? | 22:50 |
| sdake | note i saw kube-dns consistently failing on canal earlier today | 22:51 |
| sdake | it was not renamed however it was failing consistently | 22:51 |
| sdake | and I can reproduce it locally | 22:51 |
| kfox1111 | here we go | 22:51 |
| kfox1111 | http://logs.openstack.org/14/449814/2/check/gate-kolla-kubernetes-deploy-centos-binary-2-iscsi-nv/f39afd8/logs/pods/kube-system-kube-dns-2924299975-91mp3-kube-dns.txt | 22:51 |
| sdake | it is possible the gate problem is not kubernetes 1.5.5 however canal | 22:52 |
| kfox1111 | its saying 1.6.0-alpha..... | 22:52 |
| kfox1111 | possibly bad kubeadm | 22:52 |
| sbezverk_ | kfox1111: you see kubedns might flap like a toilet seat, but it should not impact api-server | 22:52 |
| kfox1111 | leaking in parts of 1.6. | 22:52 |
| sbezverk_ | kfox1111: we have seen it many times before, but api-server was always rock solid | 22:53 |
| sbezverk_ | at least in my experience | 22:53 |
| kfox1111 | sbezverk_: where are you seeing api-server failing? | 22:54 |
| kfox1111 | http://logs.openstack.org/14/449814/2/check/gate-kolla-kubernetes-deploy-centos-binary-2-iscsi-nv/f39afd8/console.html says it failed to start kube-dns. | 22:54 |
| sbezverk_ | kfox1111: number of restarts for api-server is not 9 | 22:54 |
| *** goldyfruit has quit IRC | 22:54 | |
| sbezverk_ | not 0 sorry | 22:54 |
| kfox1111 | its running though. | 22:54 |
| sdake | i deployed this is what i am running: [sdake@kolla ~]$ rpm -qa | grep kubeadm | 22:54 |
| sdake | kubeadm-1.6.0-0.alpha.0.2074.a092d8e0f95f52.x86_64 | 22:54 |
| kfox1111 | and not restarting frequently. | 22:54 |
| sbezverk_ | kfox1111: http://logs.openstack.org/90/449790/1/check/gate-kolla-kubernetes-deploy-centos-binary-2-ceph-nv/2d106b3/console.html | 22:55 |
| sdake | and found that everything works fine with that versoin of kubeadm | 22:55 |
| sdake | (with weave) | 22:55 |
| *** mbruzek has quit IRC | 22:55 | |
| *** eaguilar has quit IRC | 22:55 | |
| kfox1111 | sbezverk_: 1 is not a problem. | 22:55 |
| sdake | calico, kube-dns fails | 22:55 |
| kfox1111 | its responding enough to respond to the get pods request. | 22:55 |
| kfox1111 | gate's failing because kubedns isn't fully up. | 22:55 |
| kfox1111 | and I'm guessing thats because its a 1.6 kube-dns against a 1.5 cluster? | 22:56 |
| kfox1111 | sdake: not sure about weave. never really deployed it. | 22:57 |
| sbezverk_ | kfox1111: would be very surprised | 22:57 |
| kfox1111 | sbezverk_: suprprised about what? | 22:57 |
| sdake | kfox1111 the point is kube-dns didnt start with calico, and kube-dns did start with weave | 22:57 |
| sdake | kube-dns + calico = crashloopbackoff; kube-dns + weave = golden | 22:58 |
| kfox1111 | sdake: seen the exact opposite more often. :/ | 22:58 |
| sbezverk_ | Mar 24 19:47:48 localhost journal: E0324 19:47:48.646703 1 reflector.go:199] pkg/dns/dns.go:148: Failed to list *api.Service: Get https://172.16.128.1:443/api/v1/services?resourceVersion=0: dial tcp 172.16.128.1:443: i/o timeout | 22:59 |
| kfox1111 | kubedns uses cni though, so if its not working, it can be network related. | 22:59 |
| kfox1111 | sbezverk_: whats that from? | 22:59 |
| kfox1111 | sbezverk_: and does it reconnect afterwards? | 22:59 |
| sbezverk_ | messages | 22:59 |
| kfox1111 | 172.16... is that an address we gave it? | 23:00 |
| *** erhudy has quit IRC | 23:00 | |
| kfox1111 | ah. yeah, it is. | 23:00 |
| kfox1111 | but... | 23:01 |
| kfox1111 | we do some sed'ing in setup_kubernetes.sh too for addresses. | 23:01 |
| kfox1111 | maybe the sed is failing... | 23:01 |
| kfox1111 | in tools/setup_kubernetes.sh | 23:01 |
| *** schwicht has quit IRC | 23:01 | |
| sbezverk_ | kfox1111: it looks like cni is failing, damn | 23:02 |
| sdake | i suspect its the cni+calico connection | 23:03 |
| sbezverk_ | kfox1111: scratch that at the very end of log it got net interface initialized | 23:04 |
| *** jascott1 has joined #openstack-kolla | 23:04 | |
| kfox1111 | I0324 19:46:54.514390 1 manager.go:149] Using interface with name eth0 and address 104.239.174.208 | 23:04 |
| kfox1111 | that seems wrong.... | 23:04 |
| kfox1111 | well, maybe thats the nodes address... | 23:05 |
| sbezverk_ | http://paste.openstack.org/show/604112/ | 23:06 |
| sbezverk_ | it might be ip address of physical box? | 23:07 |
| *** eaguilar has joined #openstack-kolla | 23:07 | |
| sbezverk_ | kfox1111: yep it is a host at rackspace | 23:08 |
| sbezverk_ | we should try to use cisco's contiv ;) | 23:09 |
| kfox1111 | any cni plugin should work in theory. | 23:11 |
| *** fooliouno has joined #openstack-kolla | 23:12 | |
| sbezverk_ | sure, the keyword is in theory ;) | 23:12 |
| kfox1111 | so.... | 23:12 |
| kfox1111 | 1.5.5 was released the 21st. | 23:13 |
| kfox1111 | so unlkely its was that. | 23:13 |
| kfox1111 | 1.6.0rc1 was released today. | 23:13 |
| kfox1111 | stuff broke today? | 23:13 |
| sbezverk_ | kfox1111: are you sure yesterday's jobs were using 1.5.5?? | 23:13 |
| kfox1111 | sbezverk_: unsure. but would guess so. | 23:14 |
| sbezverk_ | hm you are right, I check a job from 22nd | 23:15 |
| sdake | any idea on this: 1h10s264{kubelet kolla}WarningFailedSyncError syncing pod, skipping: failed to "StartContainer" for "tiller" with ImagePullBackOff: "Back-off pulling image \"gcr.io/kubernetes-helm/tiller:v2.2.3\"" | 23:15 |
| sdake | [sdake@kolla ~]$ docker pull gcr.io/kubernetes-helm/tiller:v2.2.3 | 23:15 |
| sbezverk_ | and it was using 1.5.5 | 23:15 |
| sbezverk_ | sdake: is there version 2.2.3???? | 23:15 |
| sdake | indeed | 23:16 |
| sbezverk_ | I thought 2.2.2 is the latest | 23:16 |
| sdake | also 2.2.2 fials to pull | 23:16 |
| kfox1111 | sdake: do a docker pull manually on it. | 23:16 |
| kfox1111 | sometimes tells you more about it. | 23:16 |
| sdake | i did kfox1111 : [sdake@kolla ~]$ docker pull gcr.io/kubernetes-helm/tiller:v2.2.3 | 23:16 |
| sdake | Error response from daemon: target is unknown | 23:16 |
| kfox1111 | weird. | 23:17 |
| kfox1111 | what docker version? | 23:17 |
| sdake | 1.3.0 | 23:17 |
| sdake | 1.13.0 | 23:17 |
| sbezverk_ | gcr.io/kubernetes-helm/tiller:v2.2.2 | 23:17 |
| sbezverk_ | works for me | 23:17 |
| kfox1111 | 1.13.0 is broken I think. | 23:17 |
| kfox1111 | [kfox@mantis ~]$ docker pull gcr.io/kubernetes-helm/tiller:v2.2.3 | 23:17 |
| kfox1111 | Trying to pull repository gcr.io/kubernetes-helm/tiller ... | 23:17 |
| kfox1111 | v2.2.3: Pulling from gcr.io/kubernetes-helm/tiller | 23:17 |
| kfox1111 | 53ebc9bfbcc0: Pull complete | 23:17 |
| sdake | [sdake@kolla ~]$ docker pull gcr.io/kubernetes-helm/tiller:v2.2.2 | 23:17 |
| sdake | Error response from daemon: target is unknown | 23:17 |
| kfox1111 | I think 1.13.0 broke the docker hub protocol. | 23:17 |
| * sdake groans | 23:18 | |
| kfox1111 | see if there is a 1.13.1 | 23:18 |
| kfox1111 | or revert to the distro provided docker. | 23:18 |
| sbezverk_ | could be I run 1.12.6 docker | 23:18 |
| kfox1111 | I'm on 1.10.3-44.el7.centos | 23:18 |
| kfox1111 | docker inc made changes in 1.13 that broke google's repo. | 23:19 |
| sbezverk_ | kfox1111: no wonder you like 2.0.2 images ;) | 23:19 |
| *** vhosakot_ has quit IRC | 23:19 | |
| kfox1111 | sbezverk_: no, in this case, I prefer redhat's lack of breaking stuff, over dockerinc's habbit of breaking thigns. ;) | 23:19 |
| kfox1111 | sdake's seeing why first hand. :/ | 23:20 |
| sbezverk_ | kfox1111: I get docker from epel I think and they keep it at 1.12.6 | 23:20 |
| kfox1111 | sbezverk_: ah. yeah, they may have a newer version now. this test box hasn't ben updated in a little while. | 23:20 |
| kfox1111 | oh. yeah, in fact, I was mid update on this box when it ate itself..... thats right... | 23:21 |
| kfox1111 | I probaby shouldn't restart the update today... the evermind's not being kind to me today... | 23:22 |
| sbezverk_ | kfox1111: yeah, good idea, no risky operations on friday afternoon ;) | 23:22 |
| kfox1111 | too late. docker = 1.12.6 :) | 23:23 |
| kfox1111 | and it pulls tiller:v2.2.3 ok. | 23:23 |
| kfox1111 | sdake: see if you can upgrade past 1.13.0. | 23:23 |
| kfox1111 | that one I think is a month+ old. maybe they have a fix by now. | 23:24 |
| sdake | this seems to work: Version: 17.03.0-ce | 23:26 |
| kfox1111 | ah. yeah. they seem to be going more freemium. :/ | 23:27 |
| *** tonanhngo has joined #openstack-kolla | 23:30 | |
| kfox1111 | sbezverk_: I'm seeing kubelet 1.5.4 from the repo when I pull it local. | 23:32 |
| kfox1111 | kubeadm is v1.6.0-alpha.0.2074+a092d8e0f95f52, but built on 2016-12-13T17:03:18Z | 23:32 |
| kfox1111 | so not sure anything changed there. | 23:32 |
| sbezverk_ | kfox1111: a couple of day ago new version of cni was released | 23:34 |
| *** tonanhngo has quit IRC | 23:34 | |
| sbezverk_ | maybe we could pull it into the gate job and test | 23:35 |
| sbezverk_ | if the issue is the same, then it is somethign else | 23:35 |
| sdake | sbezverk_ how does this look for a cloud.yaml: http://paste.openstack.org/show/604113/ | 23:36 |
| kfox1111 | just wondering what actuallyl changed... | 23:36 |
| sdake | kfox1111 ^^ | 23:36 |
| kfox1111 | sdake: looks ok. | 23:36 |
| kfox1111 | well, unquote the port_externals I think. | 23:37 |
| sbezverk_ | kfox1111: cni we install is very very old | 23:38 |
| sbezverk_ | 0.3.1 but the current stable is 0.5.1 | 23:38 |
| kfox1111 | sbezverk_: there's another possibility... | 23:39 |
| kfox1111 | ironic merged today? | 23:39 |
| sbezverk_ | kfox1111: nope | 23:39 |
| sbezverk_ | last night, after that inc0 patch merged | 23:39 |
| kfox1111 | ah. | 23:39 |
| kfox1111 | any successful gate runs after that? | 23:39 |
| *** eaguilar has quit IRC | 23:40 | |
| *** eaguilar has joined #openstack-kolla | 23:40 | |
| *** lamt has quit IRC | 23:41 | |
| sbezverk_ | kfox1111: I think the gate was running ok after that.. | 23:41 |
| sbezverk_ | last night 6:35 the gate was ok | 23:41 |
| sbezverk_ | https://review.openstack.org/449254 | 23:41 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: Test gate https://review.openstack.org/449843 | 23:42 |
| kfox1111 | sbezverk_: when you reviewed it, it was ok, | 23:42 |
| kfox1111 | but it probably checked at 2:47 + ~30 min | 23:42 |
| sbezverk_ | kfox1111: doubt filtering could be the cause as the gate was ok, but let's see | 23:43 |
| kfox1111 | yeah, maybe not. just throwing a bunch of things against the wall to see what sticks. :) | 23:44 |
| sbezverk_ | kfox1111: +1 | 23:44 |
| *** lucasxu has joined #openstack-kolla | 23:45 | |
| sdake | rwellum using the proper verson of docker gets those doc sto work | 23:47 |
| sbezverk_ | the moment of truth | 23:47 |
| sdake | rwellum i haen't sorted out how to start the api placement microchart or the cell0 stuff | 23:48 |
| *** Pavo has quit IRC | 23:48 | |
| *** Pavo has joined #openstack-kolla | 23:49 | |
| *** schwicht has joined #openstack-kolla | 23:51 | |
| kfox1111 | sbezverk_: same failure. | 23:53 |
| sbezverk_ | :( | 23:53 |
| kfox1111 | your off the hook. ;) | 23:54 |
| sbezverk_ | :) | 23:54 |
| sbezverk_ | let's compare 6:35 log with the one we have now | 23:55 |
| sbezverk_ | to see if we can spot any differences | 23:55 |
| kfox1111 | ah. good plan. | 23:55 |
| sbezverk_ | do you have any good tool to compare other than diff? | 23:56 |
| kfox1111 | not really. :/ | 23:57 |
| kfox1111 | got a diff.. reviewing.... | 23:59 |
| sbezverk_ | kfox1111: yeah but timestamp screwed things up | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!