| *** JRobinson__ has joined #openstack-kolla | 00:00 | |
| *** Bico_Fino has joined #openstack-kolla | 00:11 | |
| *** sayanta__ has quit IRC | 00:26 | |
| *** sayantan_ has joined #openstack-kolla | 00:26 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 00:27 |
|---|---|---|
| *** l4yerffeJ has quit IRC | 00:28 | |
| *** l4yerffeJ has joined #openstack-kolla | 00:28 | |
| *** yingjun has joined #openstack-kolla | 00:30 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 00:33 |
| *** sayantan_ has quit IRC | 00:34 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 00:35 |
| *** sayantan_ has joined #openstack-kolla | 00:35 | |
| kfox1111 | nice... ok. got a generic function working with docker_registry. | 00:36 |
| kfox1111 | going to rewrite it a little bit more, then will post the ps. | 00:36 |
| *** sayantan_ has quit IRC | 00:38 | |
| *** sayanta__ has joined #openstack-kolla | 00:38 | |
| *** fragatina has joined #openstack-kolla | 00:40 | |
| *** fragatina has quit IRC | 00:41 | |
| *** fragatina has joined #openstack-kolla | 00:41 | |
| *** sayantan_ has joined #openstack-kolla | 00:42 | |
| *** jmccarthy1 has quit IRC | 00:42 | |
| *** sayanta__ has quit IRC | 00:43 | |
| *** sayantan_ has quit IRC | 00:43 | |
| *** jmccarthy has joined #openstack-kolla | 00:43 | |
| *** awiddersheim has quit IRC | 00:46 | |
| kfox1111 | nice.. that really worked. :) | 00:48 |
| *** awiddersheim has joined #openstack-kolla | 00:51 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: global api prototype https://review.openstack.org/414350 | 00:51 |
| kfox1111 | sbezverk: have a look at this one. | 00:51 |
| kfox1111 | particularly the _common_defines. | 00:51 |
| kfox1111 | you can do this with it: | 00:52 |
| kfox1111 | http://pastebin.com/J7XXCbPB | 00:52 |
| kfox1111 | overwriting any of the settings all the way down. | 00:53 |
| *** tonanhngo has joined #openstack-kolla | 00:58 | |
| *** gfidente has quit IRC | 00:59 | |
| *** _chas has joined #openstack-kolla | 01:00 | |
| srwilkers | evening | 01:02 |
| *** tonanhngo has quit IRC | 01:02 | |
| *** sayantan_ has joined #openstack-kolla | 01:04 | |
| *** zhubingbing has joined #openstack-kolla | 01:05 | |
| *** _chas has quit IRC | 01:05 | |
| kfox1111 | srwilkers: evening. :) | 01:10 |
| kfox1111 | srwilkers: check out that review. | 01:11 |
| *** jtriley has joined #openstack-kolla | 01:11 | |
| kfox1111 | incorperates the new helm 2.1.3 stuff to support globals. | 01:11 |
| kfox1111 | if everyone likes the api, then we can try and get the concept merged into all microservices. | 01:11 |
| *** ssurana has joined #openstack-kolla | 01:12 | |
| *** ssurana has quit IRC | 01:12 | |
| *** jtriley has quit IRC | 01:16 | |
| *** JRobinson__ has quit IRC | 01:20 | |
| *** hfu has joined #openstack-kolla | 01:28 | |
| openstackgerrit | dupengfei proposed openstack/kolla: Not to pull the base image when building. https://review.openstack.org/408940 | 01:29 |
| openstackgerrit | dupengfei proposed openstack/kolla-kubernetes: Helm charts for keepalived pod https://review.openstack.org/413985 | 01:41 |
| *** fragatina has quit IRC | 01:42 | |
| *** unicell has quit IRC | 01:43 | |
| *** unicell has joined #openstack-kolla | 01:46 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: use static uid/gid in images https://review.openstack.org/412231 | 01:49 |
| *** unicell has quit IRC | 01:50 | |
| srwilkers | kfox1111, looking at it now. sorry, getting back to you late | 01:57 |
| srwilkers | been a busy few days | 01:57 |
| kfox1111 | srwilkers: no worries. got a new version coming soon. | 01:58 |
| kfox1111 | well, I guess I can post it now. | 01:58 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: global api prototype https://review.openstack.org/414350 | 01:59 |
| kfox1111 | there. | 01:59 |
| kfox1111 | still needs a tweak to prebuild to build values out of the box properly. | 01:59 |
| kfox1111 | but the concept is all there now. | 02:00 |
| openstackgerrit | caoyuan proposed openstack/kolla: Move Kafka down to the "Infrastructure components" section https://review.openstack.org/413909 | 02:01 |
| openstackgerrit | rahul sharma proposed openstack/kolla: Steps to add vnic interface https://review.openstack.org/413404 | 02:01 |
| openstackgerrit | caoyuan proposed openstack/kolla: Move Kafka down to the "Infrastructure components" section https://review.openstack.org/413909 | 02:04 |
| *** sdake has joined #openstack-kolla | 02:06 | |
| Jeffrey4l | morning | 02:22 |
| *** sdake has quit IRC | 02:24 | |
| portdirect | Hey Jeffrey4l o/ | 02:25 |
| *** Pavo has joined #openstack-kolla | 02:25 | |
| Pavo | anyone here? | 02:25 |
| Jeffrey4l | hey | 02:25 |
| portdirect | Pavo, I just work up in the middle of the night... seems like kolla-k8s has been busy as usual :) | 02:26 |
| Pavo | ok I have tried and tried over and over and still getting the same issue with horizon, I have also tried the official docker hub images also | 02:27 |
| Pavo | this is what I am getting | 02:27 |
| Pavo | http://paste.openstack.org/show/593214/ | 02:27 |
| Pavo | only happens when I try to upload a image | 02:27 |
| Pavo | using horizon | 02:27 |
| portdirect | but you can upload images via cli? | 02:27 |
| Pavo | I was told it was a qemu issue with centos 7.3 but I'm using centos 7.2 | 02:27 |
| Pavo | let me try | 02:28 |
| Pavo | this is really really getting on my nerves though | 02:29 |
| Pavo | even the init_runonce script that comes with kolla has issues running | 02:30 |
| Pavo | nope doesn't look like glance cli is working either | 02:32 |
| *** jianyi has joined #openstack-kolla | 02:32 | |
| *** sdake_ has joined #openstack-kolla | 02:32 | |
| portdirect | could you try with openstack cli with --debug and the post output? | 02:32 |
| Pavo | whats the openstack cmd? | 02:33 |
| portdirect | 'openstack' | 02:33 |
| Pavo | just openstack? | 02:33 |
| Pavo | nvm I know why I think | 02:33 |
| Pavo | lol | 02:33 |
| Pavo | opps | 02:34 |
| Pavo | forgot to remake adminrc file | 02:34 |
| *** Bico_Fino has quit IRC | 02:35 | |
| portdirect | oh - running with glance cmd is fine but i'll need the debug output from that to have a chance to help if it does'nt work | 02:35 |
| Pavo | ok hang on | 02:35 |
| Pavo | gonna try it again | 02:35 |
| *** sdake has joined #openstack-kolla | 02:36 | |
| Pavo | ok glance cli is working | 02:36 |
| Pavo | but why doesn't horizon work? | 02:36 |
| portdirect | looks ike a horizon bug to me | 02:37 |
| *** sdake has quit IRC | 02:37 | |
| Pavo | I have tried a ton of different things | 02:37 |
| Pavo | offical image, from source, binary and they all do the same thing | 02:37 |
| openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: DO NOT MERGE Helm plugin to manipulate chart versions https://review.openstack.org/414383 | 02:37 |
| portdirect | it's complaining that the visibily option cannot be found | 02:37 |
| Pavo | yeah I even tried selecting both public and private | 02:38 |
| portdirect | I think asking the horizon folks may be the best bet? | 02:38 |
| Pavo | also if I wanted to add the murano plugin into the horizon container, would I be able to do that? | 02:38 |
| Pavo | portdirect I wouldn't know what to ask them for? since its a container that is built by kolla | 02:39 |
| *** sdake_ has quit IRC | 02:39 | |
| portdirect | I'd ask them for help re your bug - tell themthat uploadging via glance cli works, but you are getting the errro that you showd me in the past when trying via horizon. | 02:40 |
| portdirect | i'd give them that paste - it should make sense to them | 02:40 |
| Pavo | ok | 02:40 |
| Pavo | will try | 02:40 |
| dupengfei | kfox1111: My work is almost done, please review ,thanks https://review.openstack.org/#/c/413985/ | 02:49 |
| Pavo | portdirect guess I'll be waiting to see if anyone answers | 02:50 |
| portdirect | dupengfei: what is keepalived used for in k8s context? | 02:52 |
| portdirect | dupengfei: i see we had it in original jinja2 templates but not sure why we have it? | 02:52 |
| kfox1111 | dupengfei: will do. eating now, then will have a look. | 02:52 |
| dupengfei | portdirect: to suply HA | 02:53 |
| dupengfei | kfox1111: Fine | 02:53 |
| dupengfei | portdirect: Consider we have 3 controllers, we need a unique VIP address to aceess the dashboard. | 02:54 |
| portdirect | ok - thats the bit thats confusing me - we should be doing that via k8s itself? | 02:55 |
| portdirect | either via a service and l4 routing, or via an ingress controller? | 02:56 |
| dupengfei | portdirect: in k8s module, the common way to implement keepalive+haproxy on the edge nodes. | 02:57 |
| dupengfei | portdirect: in k8s module, the common way is to implement keepalive+haproxy on the edge nodes. | 02:57 |
| dupengfei | portdirect: keepalived is a must service, while haproxy is a optional service. | 02:58 |
| dupengfei | portdirect: k8s use the VIP as a external IP | 02:58 |
| dupengfei | portdirect: While external IP is setted by keepalived. | 02:58 |
| portdirect | dupengfei: that makes sense prior to ingress controllers - but that architecture is a backward step with k8s, I'll not object to the ps, but think we should use native k8s solutions to this. | 03:01 |
| kfox1111 | dupengfei: looks really good. one little thing to fix. | 03:01 |
| dupengfei | kfox1111: I will modify it . | 03:02 |
| kfox1111 | portdirect: k8s leaves how an external vip gets mapped to a host as an excersize to the reader. | 03:02 |
| kfox1111 | the keepalived helm package is a batteries included thing. don't have to use it, but if you need it, its there to help. | 03:03 |
| dupengfei | Yes. | 03:03 |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move cloudkitty precheck into its own role https://review.openstack.org/414389 | 03:04 |
| portdirect | kfox1111: i think thats where my issue lies, im a bit uncomfortable with us using external vips in the absence of a real load balancer, but if some people may want to use it then that up to them :) | 03:05 |
| kfox1111 | portdirect: what would you suggest in absense of a real load balancer? | 03:06 |
| portdirect | kfox1111: ingress controller daemonset on edge nodes | 03:06 |
| kfox1111 | it does reuse k8s's built in lb. | 03:06 |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move cloudkitty precheck into its own role https://review.openstack.org/414389 | 03:06 |
| kfox1111 | portdirect: that has the same issue. | 03:07 |
| kfox1111 | you still need a lb at the end of the day for ingress controllers. | 03:07 |
| kfox1111 | lb's seem to be the recursive blackhole of ha. :) | 03:07 |
| kfox1111 | there's always another layer you can add to them to get more ha. :) | 03:08 |
| kfox1111 | as your lb has to be ha. | 03:08 |
| openstackgerrit | dupengfei proposed openstack/kolla-kubernetes: Helm charts for keepalived pod https://review.openstack.org/413985 | 03:08 |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move cloudkitty precheck into its own role https://review.openstack.org/414389 | 03:09 |
| portdirect | not quite sure i follow, but wfm if it wfy :) | 03:09 |
| kfox1111 | dns isn't a very good tool for lb/ha. if an ip goes away the client doesn't stop tryign to talk to it. | 03:10 |
| kfox1111 | so you need the service on the ip to be ha. | 03:10 |
| portdirect | thats why you attach a service to the ingress controller, and use headless services only on the inside of your cluster | 03:10 |
| kfox1111 | so just having a pool of ip's in dns with an ingress controller or lb like haproxy on it doesn't make it more ha. | 03:11 |
| kfox1111 | if the lb dies, users notice. | 03:11 |
| kfox1111 | vips on the other hand, don't have that issue. | 03:11 |
| kfox1111 | as soon as the nodne that is hosting the vip dies, another node can pick it up. | 03:11 |
| kfox1111 | so from the users perspective, it never goes down, or goes down for a very very small time. | 03:11 |
| kfox1111 | ideally you would have #vips = #lb's. and each fails over to the other nodes. | 03:12 |
| kfox1111 | then in the normal case, there's a 1:1 relationship and it performs well. | 03:12 |
| kfox1111 | but still can failover really quickly when things go wrong. | 03:13 |
| kfox1111 | there are lots of layers to higher and higher levels of ha. | 03:14 |
| portdirect | but this is why you cant do this from within your cluster | 03:14 |
| kfox1111 | dns lb, vip's, l2 lb's, l3 lb's, l7 lb's, etc. | 03:14 |
| *** mdnadeem has joined #openstack-kolla | 03:14 | |
| kfox1111 | ingress controllers are a form of l7 lb's. | 03:15 |
| kfox1111 | yeah, you can solve it within the same cluster. | 03:15 |
| *** jrich523 has quit IRC | 03:15 | |
| portdirect | within your cluste k8s is awesome at this for exuernal facing services you need an external solution to map to multiple edge nodes, unless you only have a sinlge ingress point? | 03:15 |
| kfox1111 | true, sort of. | 03:16 |
| kfox1111 | there's nothing in k8s natively to do that, right. | 03:16 |
| kfox1111 | but you can use k8s to manage containers that are net=host, | 03:16 |
| kfox1111 | and the net=host containers can setup your underlying stuff k8s doesn't support. | 03:16 |
| kfox1111 | like vips, l2/l3 lb's, etc. | 03:16 |
| portdirect | but that still cant cope with weather or not you are directing external traffic toward a node that is doen | 03:17 |
| kfox1111 | yeah you can. tahts what vip's are for. | 03:17 |
| kfox1111 | and keepalived manages vips. | 03:17 |
| kfox1111 | I do it. | 03:17 |
| kfox1111 | so for example. | 03:17 |
| kfox1111 | say you have 3 ndoes. | 03:17 |
| kfox1111 | 10.0.0.1, 10.0.0.2, and 10.0.0.3. | 03:17 |
| kfox1111 | you then create 3 vips. 10.0.0.4, 10.0.0.5, 10.0.0.6 | 03:18 |
| *** qiliang27 has quit IRC | 03:18 | |
| kfox1111 | you install keepavlied on each, and give priority for 4 to 1, 5 to 2, and 6 to 3. | 03:18 |
| *** qiliang27 has joined #openstack-kolla | 03:19 | |
| kfox1111 | you register 4 5 and 6 to your dns entry users talk to. | 03:19 |
| kfox1111 | now, all 3 addresses will exist when a node goes down cause another node will pick it up. | 03:19 |
| kfox1111 | then you bind your services to the vips, not the node's physical ip's. | 03:20 |
| portdirect | right but i thought k8s only allowed a single external ip? | 03:21 |
| kfox1111 | I'm not sure if it suports a list or not. if not, it should get that feature. | 03:21 |
| kfox1111 | but its still useful as an option for a small cluster of 2 nodes with failover in that case at very least. | 03:21 |
| kfox1111 | there's lots of ways to archetect ha systems. so we're supporting having the options to allow the end user decide what's best for their environment. | 03:22 |
| kfox1111 | even if they have 1 vip, that may be simpler, and simpler's cheper/less likely to break. | 03:23 |
| kfox1111 | if their api load isn't high enough, that may be the preffered solution. | 03:23 |
| portdirect | k - i wont be recommending it but as i said if it works for you then that's great :) | 03:23 |
| kfox1111 | yeah. there are lots of other ways to do it. so, whatever works for you. :) | 03:24 |
| dupengfei | Hello, I will goes into helm "cinder-create-keystone-endpoint-publicv2", Is anyone already working on it ? | 03:32 |
| *** fragatina has joined #openstack-kolla | 03:33 | |
| srwilkers | dupengfei, ive got a patch set im tidying up for it. have most of the work done, just a few bits to finish up | 03:34 |
| dupengfei | srwilkers: ok, you do it . | 03:34 |
| kfox1111 | dupengfei: how about the createdb/managedb/endpoint stuff for heat? I don't think anyone's taken those yet. | 03:35 |
| *** mdnadeem has quit IRC | 03:35 | |
| srwilkers | dupengfei, kfox1111: i think that'd be good to tackle | 03:35 |
| *** tonanhngo has joined #openstack-kolla | 03:36 | |
| dupengfei | OK, I will do about createdb/managedb/endpoint of heat. | 03:36 |
| *** sp__ has joined #openstack-kolla | 03:36 | |
| kfox1111 | awesome. thanks. :) | 03:37 |
| dupengfei | Pleasure, about 2 days, I am newer about heat. | 03:38 |
| kfox1111 | ok. cool. shoudl be pretty much the same as the other services for those I think. | 03:38 |
| *** fragatin_ has joined #openstack-kolla | 03:41 | |
| *** tonanhngo has quit IRC | 03:42 | |
| *** tonanhngo has joined #openstack-kolla | 03:42 | |
| openstackgerrit | Ravindra Yadav proposed openstack/kolla: update documentation to correct diirectory https://review.openstack.org/414407 | 03:42 |
| *** fragatina has quit IRC | 03:43 | |
| *** fragatin_ has quit IRC | 03:43 | |
| *** fragatina has joined #openstack-kolla | 03:44 | |
| kfox1111 | nice... really liking this 2.1.3 thing. :) | 03:45 |
| dupengfei | 2.1.3 thing. ?? | 03:46 |
| kfox1111 | helm. | 03:46 |
| *** tonanhngo has quit IRC | 03:46 | |
| kfox1111 | sry. | 03:46 |
| openstackgerrit | Ravindra Yadav proposed openstack/kolla: update documentation to correct diirectory https://review.openstack.org/414407 | 03:47 |
| *** l4yerffeJ has quit IRC | 03:57 | |
| *** sdake has joined #openstack-kolla | 04:00 | |
| kfox1111 | oh, so nice... | 04:00 |
| kfox1111 | loving this ability to edit dicts. :) | 04:00 |
| *** sp__ has quit IRC | 04:06 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: global api prototype https://review.openstack.org/414350 | 04:06 |
| *** sdake has quit IRC | 04:06 | |
| *** Pavo has quit IRC | 04:06 | |
| *** yingjun has quit IRC | 04:17 | |
| *** prameswar has joined #openstack-kolla | 04:18 | |
| *** mdnadeem has joined #openstack-kolla | 04:21 | |
| *** _chas has joined #openstack-kolla | 04:29 | |
| *** gomarivera has joined #openstack-kolla | 04:31 | |
| *** tonanhngo has joined #openstack-kolla | 04:32 | |
| *** _chas has quit IRC | 04:34 | |
| *** gomarivera has quit IRC | 04:36 | |
| *** yingjun has joined #openstack-kolla | 04:37 | |
| *** srwilkers has quit IRC | 04:40 | |
| Jeffrey4l | sbezverk, ping | 04:46 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: global api prototype https://review.openstack.org/414350 | 04:47 |
| *** v1k0d3n has quit IRC | 04:55 | |
| *** v1k0d3n has joined #openstack-kolla | 04:55 | |
| *** jianyi has quit IRC | 05:02 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 05:04 |
| openstackgerrit | Merged openstack/kolla-ansible: Do not mount /var/lib/nova/mnt unless nfs cinder backend is enabled https://review.openstack.org/406168 | 05:04 |
| openstackgerrit | Merged openstack/kolla: Fix typo for Octavia link https://review.openstack.org/414309 | 05:07 |
| *** v1k0d3n has quit IRC | 05:11 | |
| *** v1k0d3n has joined #openstack-kolla | 05:11 | |
| *** senk has joined #openstack-kolla | 05:16 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible: Romove repeat memcache configuration https://review.openstack.org/414417 | 05:18 |
| *** unicell has joined #openstack-kolla | 05:22 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Remove -H fd:// for docker.service in the baremetal role https://review.openstack.org/413911 | 05:38 |
| *** senk has quit IRC | 05:39 | |
| openstackgerrit | dupengfei proposed openstack/kolla: Not to pull the base images when building. https://review.openstack.org/408940 | 05:44 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: [wip] try pull docker images from tarballs.o.o https://review.openstack.org/413720 | 05:51 |
| *** sayantan_ has quit IRC | 06:09 | |
| *** sayantan_ has joined #openstack-kolla | 06:10 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible: WIP add fluentd role https://review.openstack.org/407392 | 06:10 |
| *** sayantan_ has quit IRC | 06:15 | |
| *** hfu has quit IRC | 06:17 | |
| *** senk has joined #openstack-kolla | 06:27 | |
| sp_ | portdirect: hi... | 06:31 |
| *** _chas has joined #openstack-kolla | 06:31 | |
| *** awiddersheim has quit IRC | 06:31 | |
| *** zhangshuai has joined #openstack-kolla | 06:32 | |
| *** _chas has quit IRC | 06:35 | |
| dupengfei | Hello， who can help me to analysis the jenkins failure： https://review.openstack.org/#/c/408940/ thanks | 06:39 |
| dupengfei | ??? | 06:43 |
| *** salv-orlando has joined #openstack-kolla | 06:44 | |
| *** gomarivera has joined #openstack-kolla | 06:52 | |
| *** gomarivera has quit IRC | 06:57 | |
| *** _chas has joined #openstack-kolla | 07:05 | |
| *** mnasiadka has joined #openstack-kolla | 07:10 | |
| openstackgerrit | dupengfei proposed openstack/kolla-kubernetes: Delete configMap volume in cinder_create_db.yaml https://review.openstack.org/414457 | 07:18 |
| openstackgerrit | jimmygc proposed openstack/kolla: Remove unneeded oslo-rootwrap in swift-base https://review.openstack.org/413392 | 07:18 |
| *** salv-orlando has quit IRC | 07:40 | |
| *** sayantan_ has joined #openstack-kolla | 07:48 | |
| *** sayantan_ has quit IRC | 07:53 | |
| *** pcaruana has quit IRC | 07:54 | |
| *** pcaruana has joined #openstack-kolla | 07:55 | |
| *** sdake has joined #openstack-kolla | 07:56 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Not to pull the base images when building. https://review.openstack.org/408940 | 07:59 |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible: WIP add fluentd role https://review.openstack.org/407392 | 08:00 |
| openstackgerrit | shaofeng cheng proposed openstack/kolla: kolla fix nova config error in quickstart.rst https://review.openstack.org/414476 | 08:22 |
| *** yingjun has quit IRC | 08:24 | |
| *** flaper87 has quit IRC | 08:24 | |
| *** yingjun has joined #openstack-kolla | 08:25 | |
| *** flaper87 has joined #openstack-kolla | 08:26 | |
| sdake | Jeffrey4l ping | 08:27 |
| *** sdake_ has joined #openstack-kolla | 08:30 | |
| *** sdake has quit IRC | 08:34 | |
| openstackgerrit | Zeyu Zhu proposed openstack/kolla-ansible: Modify the openstack_release in globals.yml https://review.openstack.org/414478 | 08:40 |
| *** jimmygc has joined #openstack-kolla | 08:46 | |
| jimmygc | The Jenkins is broken I think, it failed before building ubuntu image. | 08:50 |
| *** prameswar has quit IRC | 08:56 | |
| openstackgerrit | dupengfei proposed openstack/kolla-kubernetes: Helm charts for heat-create-db and heat-manage-db https://review.openstack.org/414496 | 09:00 |
| openstackgerrit | Kuo-tung Kao proposed openstack/kolla-ansible: provide a universal collectd template https://review.openstack.org/414499 | 09:05 |
| *** tonanhngo has quit IRC | 09:17 | |
| *** yingjun_1 has joined #openstack-kolla | 09:17 | |
| *** yingjun has quit IRC | 09:18 | |
| *** shardy has joined #openstack-kolla | 09:22 | |
| *** sayantan_ has joined #openstack-kolla | 09:36 | |
| *** sayantan_ has quit IRC | 09:41 | |
| *** salv-orlando has joined #openstack-kolla | 09:41 | |
| *** yingjun_1 has quit IRC | 09:44 | |
| portdirect | sp_: hey | 09:44 |
| *** salv-orlando has quit IRC | 09:46 | |
| *** gfidente has joined #openstack-kolla | 09:48 | |
| *** oredondo has joined #openstack-kolla | 09:49 | |
| *** oredondo has quit IRC | 09:50 | |
| portdirect | Jeffrey4l: ping | 09:54 |
| *** jimmygc has quit IRC | 10:04 | |
| *** salv-orlando has joined #openstack-kolla | 10:05 | |
| *** mannamne has joined #openstack-kolla | 10:07 | |
| Jeffrey4l | sdake_, portdirect pong | 10:08 |
| Jeffrey4l | sdake_, could u review this https://review.openstack.org/414235 which unblock ubuntu gate. | 10:09 |
| sp_ | portdirect: hi.. | 10:10 |
| sp_ | portdirect: there was some query related to kolla-k8s | 10:10 |
| sp_ | portdirect: thanks | 10:10 |
| *** Serlex has joined #openstack-kolla | 10:12 | |
| *** prameswar has joined #openstack-kolla | 10:12 | |
| *** pprokop_ has quit IRC | 10:16 | |
| *** msimonin has joined #openstack-kolla | 10:24 | |
| *** pprokop has joined #openstack-kolla | 10:25 | |
| portdirect | hey Jeffrey4l & sp_ back again | 10:28 |
| portdirect | sp_: about for the next hour | 10:28 |
| Jeffrey4l | sup portdirect | 10:29 |
| portdirect | Jeffrey4l: if you have time to look at my comments here it would be great: https://review.openstack.org/#/c/412205/ | 10:29 |
| portdirect | Jeffrey4l: totally happy to make adjustment as you see fit | 10:29 |
| portdirect | as I know it is ugly in its current form | 10:29 |
| portdirect | there is no plans to offer a non 64bit build - but naming the images this way is in line with how google release them | 10:30 |
| Jeffrey4l | portdirect, i am OK if other kolla-k8s team member to accept this. | 10:30 |
| *** msimonin has quit IRC | 10:30 | |
| portdirect | np - I'll get kfox1111 or sbezverk to review, thanks :) | 10:31 |
| Jeffrey4l | but personally, i prefer not to use such pattern. | 10:31 |
| Jeffrey4l | because i do not see such solution in hub.docker.com | 10:31 |
| Jeffrey4l | roger. | 10:31 |
| portdirect | totally accept that - and I could understand if you did not accept it as it does break the standard for kolla images | 10:32 |
| Jeffrey4l | cool. | 10:32 |
| *** rmart04 has joined #openstack-kolla | 10:57 | |
| sdake_ | portdirect - i -1'ed that review as well | 11:05 |
| sdake_ | portdirect i agree with jeffrey the directory name of the dockerfile should remove amd64 | 11:05 |
| portdirect | sdake_: a-ok with me :) | 11:06 |
| portdirect | will get new ps up soon | 11:06 |
| sdake_ | however the dockerfile contents should use the /amd64 - as this is how upstream releases it | 11:06 |
| sdake_ | morning btw portdirect :) | 11:07 |
| sdake_ | boy went to bed at 7pm | 11:07 |
| sdake_ | so glorious :) | 11:07 |
| sdake_ | my wife is still sleeping from 7pm atm - and its 4am | 11:07 |
| sdake_ | got about 10 hrs of sleep or so - guess I needed it :) | 11:07 |
| portdirect | oh yeah! morning sdake :D | 11:08 |
| sdake_ | Jeffrey4l - have you tested centos ansible lately? | 11:08 |
| sdake_ | I have new wifi and my boxes are not working properly - pavo has reported x86_64 is not working properly | 11:08 |
| sdake_ | Jeffrey4l as a result of my wifi I am unable to test his test scenarios | 11:09 |
| sdake_ | Jeffrey4l i am calling support in the morning to hopefully get it sorted out | 11:09 |
| sdake_ | or back to the store it goes :) | 11:09 |
| sdake_ | its called eero wireless | 11:09 |
| sdake_ | looks pretty good to me :) | 11:09 |
| sdake_ | its super simple | 11:09 |
| sdake_ | Jeffrey4l in the meantime, minime-03 will be out of service if you still use it | 11:10 |
| openstackgerrit | Merged openstack/kolla: Run sudo with -H parameter to set the proper home dir https://review.openstack.org/414235 | 11:10 |
| sdake_ | got new tv - samsung 8000 2016 model | 11:11 |
| sdake_ | super blowout sale - 1499 | 11:11 |
| sdake_ | pretty excited to get it mounted | 11:11 |
| sdake_ | wife has banned mounting until after christmas ;) | 11:11 |
| sdake_ | since it clocks in at 50 lbs | 11:12 |
| sdake_ | got a roku ultra too - much snappier then my roku 4 | 11:12 |
| openstackgerrit | Pete Birley proposed openstack/kolla: Add Kubernetes Images for Kolla-Kubernetes https://review.openstack.org/412205 | 11:12 |
| sdake_ | i don't know how roku stays in business | 11:13 |
| sdake_ | portdirect possibly dumb question - why containerizing kubernetes? | 11:13 |
| sdake_ | wouldn't distros do that? | 11:13 |
| portdirect | not dumb | 11:13 |
| sdake_ | thats why I qualified with possibly :) | 11:14 |
| portdirect | so that we can provide ceph rbd support - google upstream images dont do that | 11:14 |
| sdake_ | i see thanks | 11:14 |
| portdirect | i'm trying to get them to add it... | 11:14 |
| portdirect | so these are simply to enable someone who wants to use kubeadm or another k8s deployment tool that uses containers for the controll pane to have a set without building their own | 11:15 |
| portdirect | the primary user being our gate :) | 11:16 |
| portdirect | (at least at present) | 11:16 |
| sp_ | portdirect: hi.. I had some query atm | 11:16 |
| portdirect | hey sp_ | 11:16 |
| sp_ | portdirect: but thanks | 11:16 |
| sp_ | portdirect: i got my answer. np | 11:16 |
| sdake_ | portdirect cool got it | 11:17 |
| *** sdake_ is now known as sdake | 11:17 | |
| sdake | btw portdirect kfox1111 srwilker- jascott1 sbezverk - and anyone else interested, please respond on my thread re the naming conventions in the repo for microservices - if we decide to make a change here we can do it in 0.5.0 | 11:18 |
| portdirect | sp_: sweet :) I'll be away from this afternoon but will be back on 28th - if you have any issues before then you *may* be able to get me via email - but no guarantees | 11:18 |
| sdake | also please use reno release notes | 11:18 |
| sdake | i'm sort of in and out - at this time of the morning I can't do anything else in my hosuehold as it wakes up the family - so review helps kill time | 11:18 |
| Jeffrey4l | sdake, i am back. | 11:19 |
| Jeffrey4l | are u means install ansible from centos rpm package? sdake | 11:19 |
| sdake | so when i'm awake at 4-5 am - i may be doing some reviews here and there - but otherwise dont' expect me to get around | 11:19 |
| portdirect | sdake: i've reached out to the kube deveoplers to see if they have a preffered term still - but still awaiting reply - I'll respond regardless before i head off though :) | 11:19 |
| sdake | Jeffrey4l no - i mean the kolla-ansible deliverable doesn't work for pavo when he pip installs 3.0.1 | 11:19 |
| sdake | its related to centos7.3 and the virt sig fallout | 11:19 |
| sdake | (I believe) | 11:20 |
| Jeffrey4l | 3.0.1 should use kolla rather then kolla-ansible. | 11:20 |
| sdake | Jeffrey4l right - i miscommunicated, I mean he is using 3.0.1 of kolla (the ansible parts) and its not working | 11:20 |
| Jeffrey4l | any error message? | 11:21 |
| sdake | Jeffrey4l if you could sync up with him and sort out how to get a bug filed so we can track it, that would be helpful | 11:21 |
| Jeffrey4l | gate is OK. | 11:21 |
| sdake | gate has poor coverage | 11:21 |
| sp_ | portdirect: thanks a lot FYI. is there any to catch you on 26th and 27th ? | 11:21 |
| sdake | gate = dead chicken test | 11:21 |
| Jeffrey4l | yep. I will check with him | 11:21 |
| sp_ | portdirect: certainly after 9AM (GMT). | 11:21 |
| portdirect | sp_: on those dates I'll most likley be in the middle of the countryside with 2g intenet access only | 11:22 |
| sdake | Jeffrey4l to save you time on urban dictionary - dead chicken test = wave a dead chicken over the gear - if the chicken comes alive you know you have a problem ;) | 11:22 |
| portdirect | sdake: thats a new one on me! | 11:22 |
| Jeffrey4l | lol funny | 11:22 |
| sdake | ya - dead chicken test is old school | 11:22 |
| sdake | I learned it from soemone who was 60 when i was 30 :) | 11:22 |
| sdake | he is now 70ish | 11:22 |
| sdake | portdirect it may not even be in urban dictionary :) | 11:22 |
| sdake | thats along the lines of fat dumb and happy | 11:23 |
| sdake | which is harder to define : | 11:23 |
| sdake | ;) | 11:23 |
| portdirect | nice - I'm gonna use it for sure though. | 11:23 |
| sp_ | sdake: "dead chicken test is old school" please elaborate, seems quite interesting | 11:24 |
| sdake | old school = what people used in the old days | 11:24 |
| sdake | it basically means very *basic* testing - nothing extensive | 11:24 |
| sp_ | sdake: got it | 11:25 |
| sdake | we ought to bring that back into our lexicon :) | 11:25 |
| sdake | (our being the tech industry) :) | 11:25 |
| *** gomarivera has joined #openstack-kolla | 11:32 | |
| *** l4yerffeJ has joined #openstack-kolla | 11:33 | |
| *** gomarivera has quit IRC | 11:37 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO_NOT_MERGE: TEST MASTER BRANCH https://review.openstack.org/399897 | 11:38 |
| *** pprokop has quit IRC | 11:39 | |
| sdake | Jeffrey4l mastter branch passes gate | 11:39 |
| sdake | i believe it is more complex - pavo is doing something specific which is exposing a problem | 11:39 |
| sdake | i am concerned it is affecting many people | 11:39 |
| Jeffrey4l | yep. it should be. just double check ;) | 11:39 |
| Jeffrey4l | roger. will sync with him | 11:40 |
| sdake | re chinese new year - when was that again | 11:40 |
| sdake | so I can put it on my calendar | 11:40 |
| sdake | i was attempting to do so - but got hammered by 20 distractions at once :) | 11:40 |
| sdake | boy rht got hammered after their earnings report | 11:41 |
| sdake | missed their #'s by 2.7 million | 11:41 |
| sdake | probably one sales dude got hit by a bus | 11:42 |
| Jeffrey4l | we have 3 days holiday for new year 1.1 - 1.3 and have a long holiday for Chinese new year 1.27 - 2.2 | 11:42 |
| sdake | Jeffrey4l cool everyone will be out for new year | 11:42 |
| sdake | i'll put 27-2.2 on my calendar | 11:42 |
| sdake | Jeffrey4l is it lunar based, or consistently the same? | 11:42 |
| Jeffrey4l | lunar based. | 11:42 |
| sdake | ok - well i may ask you again next year then :) | 11:43 |
| sdake | that whole lunar thing is hard for me :) | 11:43 |
| Jeffrey4l | ;) yep. | 11:43 |
| Jeffrey4l | me too. i never use it. | 11:43 |
| portdirect | Jeffrey4l: chinese new year used to strike fear in manufacturing co's in the UK - we base pretty much our entire year about the shutdown :) | 11:43 |
| Jeffrey4l | hrm. " we base pretty much our entire year about the shutdown :)" means? | 11:45 |
| *** _chas_ has joined #openstack-kolla | 11:46 | |
| Jeffrey4l | manufacturing is almost close at chinese new year? or? portdirect | 11:46 |
| sdake | i bet chinese new strikes fear in the global economy :) | 11:46 |
| sdake | globalizaation ftl :) | 11:47 |
| portdirect | Jeffrey4l: sorry bad typing - for several comanies over here we were so reliant on orders coming from china, that if we forgot to take the new year holiday into account it caused major supply issues - as we tried to keep as low as possible an inventory in stock it meant that sometimes we would miss the cutoff for things shipping and not be able to run | 11:47 |
| portdirect | production lines for a month | 11:47 |
| sdake | i have a chinese manufactured tube amp | 11:47 |
| sdake | it totally kicks ass | 11:48 |
| sdake | hand made | 11:48 |
| sdake | point to point wiring | 11:48 |
| sdake | line magnetic 219ia :) | 11:48 |
| Jeffrey4l | portdirect, understood. thanks. yep, almost the whole china is stopped at chinese new year. | 11:48 |
| *** _chas has quit IRC | 11:49 | |
| sdake | portdirect - this is what I get for reviewing when i wake up | 11:51 |
| sdake | portdirect can you add a reno release note to the kubernetes docker image review? | 11:51 |
| portdirect | lol | 11:51 |
| sdake | the good and bad thing about china is there is no chatoic election of presidents every 4 years :) | 11:52 |
| portdirect | no - beause i dont know how - can you point me in the right direction and I'll try and work it out? | 11:52 |
| sdake | reno -n blurb | 11:52 |
| sdake | sudo pip install reno | 11:52 |
| sdake | reverse those steps :) | 11:52 |
| sdake | edit the reno file | 11:53 |
| sdake | also you need a blueprint opened which i can ack | 11:53 |
| sdake | (commit message should contain a blueprint) | 11:53 |
| portdirect | roger. cheers :) | 11:53 |
| sdake | link blueprint in channel and i'll ack it | 11:53 |
| sdake | i have a torture session at the dentist in about 4 hours - so sooner the better :) | 11:54 |
| sdake | seems like folks have fallen off the reno wagon as of late | 11:55 |
| sdake | we need those to track changes for our downstream consumers | 11:55 |
| sdake | WTB RAX WORKING GATE | 11:57 |
| sdake | Jeffrey4l any ideas about rax | 11:57 |
| Jeffrey4l | sdake, no ;( what did u got? | 11:58 |
| sdake | rechecking a million times to avoid rax is irritating at best :) | 11:58 |
| Jeffrey4l | sdake, logs? | 11:58 |
| sdake | https://review.openstack.org/#/c/412205/11 | 11:58 |
| Jeffrey4l | btw, nova_libvirt failed now and then on master. no idea why. | 11:58 |
| sdake | right - is this rax specific or does osic exhibit this problem too? | 11:59 |
| sdake | osic gear seems most reliable | 11:59 |
| Jeffrey4l | agree. | 11:59 |
| sdake | i think there is something wrong with our gate scripts as it reltes to rax | 11:59 |
| sdake | perhaps we could add a custom hack to deal with that cloud provider | 11:59 |
| Jeffrey4l | one issue in rax is: there is no ip on eth1 in some case http://logs.openstack.org/05/412205/11/check/gate-kolla-dsvm-build-centos-source-centos-7/ad4e560/console.html#_2016-12-23_11_13_23_341513 | 12:00 |
| sdake | ok - that should be easy to workaround - last I changed the gate scripts, I made it use eth1 iirc | 12:00 |
| Jeffrey4l | but i have no idea why nova_libvirt failed now and then. | 12:00 |
| sdake | it was some time ago | 12:00 |
| sdake | is the now and then new as of 2-3 weeks gao, or now and then new as of 2-3 months ago? | 12:01 |
| sdake | gao/aago | 12:01 |
| sdake | gah | 12:01 |
| Jeffrey4l | there should be a ip in eth1. | 12:01 |
| sdake | Jeffrey4l right - but do we need that ip? | 12:01 |
| Jeffrey4l | base on the /etc/nodepool | 12:01 |
| *** msimonin has joined #openstack-kolla | 12:01 | |
| sdake | i guess we need two interfaces to operate correctly | 12:02 |
| sdake | Jeffrey4l btw, i think dmsimard was wanting the tox -e build-xxx | 12:02 |
| Jeffrey4l | we depends on /etc/nodepool/private_address | 12:02 |
| sdake | to enable gating of their work | 12:02 |
| Jeffrey4l | if this is on a nic which do not get a ip, gate will be failed. | 12:02 |
| sdake | Jeffrey4l cool can you file a bug on that and make it essential and bring it into the release | 12:02 |
| sdake | i'd like it to be tracked at minimum | 12:03 |
| Jeffrey4l | ok. | 12:03 |
| sdake | or critical rather | 12:03 |
| sdake | if people can figure out how to fix it, winning :) | 12:03 |
| Jeffrey4l | i will file a bug. | 12:03 |
| sdake | we have been dealing with this for over a year | 12:03 |
| sdake | and its really disruptive | 12:03 |
| Jeffrey4l | another thing is nova_libvirt failed now and then. | 12:03 |
| sdake | only on rax? | 12:03 |
| Jeffrey4l | could u check the audit.log here http://logs.openstack.org/05/412205/11/check/gate-kolla-dsvm-deploy-centos-binary-centos-7-nv/74fe447/logs/system_logs/audit/audit.txt.gz | 12:03 |
| sdake | or on all providers | 12:03 |
| Jeffrey4l | is there any special? | 12:03 |
| Jeffrey4l | not sure. but seems so. | 12:04 |
| sdake | type=AVC msg=audit(1482491219.592:16): avc: denied { read } for pid=496 comm="ip6tables.init" name="modprobe.d" dev="xvda1" ino=3655937 scontext=system_u:system_r:iptables_t:s0 tcontext=system_u:object_r:modules_conf_t:s0 tclass=dir | 12:04 |
| sdake | i search for denied | 12:04 |
| sdake | rax does not use KVM under the hood | 12:04 |
| sdake | instead they use that other virt environment - forget which its called | 12:04 |
| openstackgerrit | Pete Birley proposed openstack/kolla: Add Kubernetes Images for Kolla-Kubernetes https://review.openstack.org/412205 | 12:04 |
| sdake | it may be the cause of some problems | 12:05 |
| portdirect | now onto blueprint - 2 mins... | 12:05 |
| Jeffrey4l | yep. i got the same error in another gate job, which is green. so it should not be a issue. | 12:05 |
| Jeffrey4l | sdake, check this http://logs.openstack.org/05/412205/11/check/gate-kolla-dsvm-deploy-centos-binary-centos-7-nv/74fe447/console.html#_2016-12-23_11_45_21_041991 | 12:06 |
| Jeffrey4l | nova_libvirt failed to start. but i can not get the detail error message. | 12:06 |
| sdake | Jeffrey4l why can't get detail error message from libvirt logs? | 12:09 |
| Jeffrey4l | do not know. ;( | 12:09 |
| sdake | is it not present? | 12:10 |
| sdake | oh - the rax virt layer is xen | 12:10 |
| sdake | SamYaple mentioned that the other day | 12:10 |
| Jeffrey4l | how can we make sure of that? libvirt can not work on xen? | 12:10 |
| sdake | i think our problems are related to ip addresses tho, not necesssarily xen | 12:10 |
| sdake | i don't know if libvirt works on xen or not | 12:11 |
| sdake | rather, i don't know if kvm on xen is a thing that works | 12:11 |
| sdake | mailing list msg on openstack-infra would help here | 12:11 |
| portdirect | sdake: https://blueprints.launchpad.net/kolla/+spec/add-kubernetes-images | 12:11 |
| sdake | (with link to the bug and this irc log:) | 12:11 |
| Jeffrey4l | we just run libvirt serivce, without boot any instance. | 12:12 |
| Jeffrey4l | and even if we boot instance, it is qemu ( without kvm ) | 12:12 |
| *** msimonin has quit IRC | 12:12 | |
| openstackgerrit | Pete Birley proposed openstack/kolla: Add Kubernetes Images for Kolla-Kubernetes https://review.openstack.org/412205 | 12:13 |
| openstackgerrit | Merged openstack/kolla-ansible: Move cloudkitty precheck into its own role https://review.openstack.org/414389 | 12:13 |
| portdirect | sdake: will make another ps hang on 2 sec before reviewing | 12:17 |
| sdake | portdirect sorry for the hassle ;) | 12:17 |
| sdake | portdirect this is all stuff that should be happening in the kolla-kubernetes repo too (reno for example) | 12:18 |
| sdake | i knwo i've been a bit absent from reviews on kolla-kubernetes | 12:18 |
| portdirect | lol - no probs man - serves me right for not smell checking | 12:18 |
| sdake | been drinking from firehose internally at work in december | 12:18 |
| *** msimonin has joined #openstack-kolla | 12:18 | |
| sdake | when you drink from a firehose you explode ;) | 12:18 |
| portdirect | dude - that waters got dead pigeons in it :/ | 12:19 |
| sdake | well i dont actually drink from a firehose :) | 12:19 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: TEST MASTER BRANCH https://review.openstack.org/398095 | 12:19 |
| sdake | in the us most hydrant water is on the main water distribution system - as in its drinkable | 12:19 |
| portdirect | sorry i have no concept of metaphors, they go right over my head. | 12:19 |
| sdake | potable :) | 12:19 |
| sdake | portdirect lol - i'm pretty sure you do :) | 12:20 |
| portdirect | :) | 12:20 |
| sdake | 31 hours of media in a 24 hour day :) | 12:20 |
| portdirect | exactly | 12:21 |
| sdake | there is a really good documentary on "Curiosity!" stream (11.99 a month for 4k, or 2.99 for regular sd) - forget what its called | 12:21 |
| sdake | but it is pretty good :) | 12:21 |
| sdake | about information overload | 12:21 |
| sdake | and how our brains are not designed to handle it | 12:21 |
| sdake | i love that channel | 12:22 |
| sdake | i watch it often when i have spare brainpower | 12:22 |
| sdake | if i dont i turn on the tv while working | 12:22 |
| sdake | or listen to my stereo | 12:22 |
| *** msimonin has quit IRC | 12:24 | |
| openstackgerrit | Pete Birley proposed openstack/kolla: Add Kubernetes Images for Kolla-Kubernetes https://review.openstack.org/412205 | 12:24 |
| *** mdnadeem has quit IRC | 12:32 | |
| *** Serlex has quit IRC | 12:32 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 12:34 |
| *** senk has quit IRC | 12:35 | |
| *** caowei has joined #openstack-kolla | 12:37 | |
| portdirect | sbezverk: ping | 12:40 |
| sbezverk | portdirect: pong | 12:41 |
| portdirect | sbezverk: morning o/ :) | 12:41 |
| portdirect | where are you getting the ip addrs from in the all_values.yaml? | 12:41 |
| sbezverk | portdirect: morning, if it is about lvm, then ignore them for now | 12:41 |
| portdirect | ok cool :) | 12:42 |
| sbezverk | I am just testing rendering different lvm backstorres | 12:42 |
| portdirect | np - figured that was the case - but thought I might have missed somthing | 12:42 |
| *** salv-orlando has quit IRC | 12:43 | |
| sbezverk | portdirect: the gate does not generate configmap for any iscsi components.. need to figure out how to change globals so not to break ceph at the same time.. | 12:43 |
| sbezverk | portdirect: btw I filed an issue I hit with kube yesterday. https://github.com/kubernetes/kubernetes/issues/39159 | 12:44 |
| sbezverk | not sure if you ever use iscsi, but just fyi | 12:44 |
| *** mgiles has joined #openstack-kolla | 12:47 | |
| *** msimonin has joined #openstack-kolla | 12:47 | |
| portdirect | roger, not used it directly - I wrote a cinder flexvolume driver for that | 12:48 |
| portdirect | but will track the issue - did you get ceph working for you in the end? | 12:49 |
| sbezverk | portdirect: yeah, ceph signle node works fine.. | 12:51 |
| sbezverk | with dynamic pvc | 12:52 |
| openstackgerrit | Merged openstack/kolla-kubernetes: Delete configMap volume in cinder_create_db.yaml https://review.openstack.org/414457 | 12:54 |
| sdake | morning sbezverk | 12:54 |
| *** salv-orlando has joined #openstack-kolla | 12:55 | |
| sdake | can someone explain to me how ceph works now | 12:55 |
| sdake | as in the structure | 12:55 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 12:56 |
| sdake | there is a ceph on baremetal where kubernetets is running | 12:56 |
| sdake | that provides service for storage for PVCs | 12:57 |
| *** Bico_Fino has joined #openstack-kolla | 12:57 | |
| sdake | what about the control palnes for things like nova/glance/etc | 12:57 |
| portdirect | can use same ceph | 12:58 |
| sdake | does att's dev env deploy that now? | 12:58 |
| sdake | or is it manual deploy of ceph? | 12:58 |
| portdirect | it uses k8s to deply ceph | 12:59 |
| sdake | so ceph is internal in the att dev env | 13:00 |
| sdake | in a real env, ceph would be external to k8s? | 13:00 |
| portdirect | ah - sorry crossed wires | 13:00 |
| sdake | shorting ftw :) | 13:00 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 13:00 |
| portdirect | the halcyon-kubernetes does not deploy ceph - but sets up k8s to be ready for it | 13:01 |
| sbezverk | sdake: morning, enjoying pto I see? | 13:01 |
| sdake | sbezverk ya - trying to keep my mind off the impending torture session at the dentist | 13:01 |
| portdirect | lol - sdake s as bad as the rest of us - his comment on my ps earlier today made me think: kolla, not even once. | 13:02 |
| sbezverk | sdake: sorry to hear that.. | 13:02 |
| *** senk has joined #openstack-kolla | 13:02 | |
| sdake | sbezverk its ok - this one should be relatively easy | 13:02 |
| sdake | installing a crown | 13:02 |
| sdake | vs making one | 13:02 |
| sdake | making one sucks | 13:02 |
| sdake | root canals are worse :( | 13:02 |
| sdake | my dentist told me my genetics are 1 in 500 for my teeth | 13:02 |
| sbezverk | sdake: they all use this excuse to compensate lack of skills or bigger bill ;-) | 13:03 |
| sdake | nah - they really are that bad unfortunately | 13:04 |
| kfox1111 | arg... woke up at 2 am and have spent the last 3 hours trying to get back down. | 13:06 |
| kfox1111 | I give up. :/ | 13:07 |
| sdake | sweet we are all on crazy sleep schedules :) | 13:07 |
| sdake | of course i went to bed at 7pm :) | 13:07 |
| kfox1111 | I wanted to go to sleep at 7, but was kept up until 11. :/ | 13:08 |
| sdake | bummer - i just hit the rack and that was that :) | 13:08 |
| kfox1111 | anyway, kolla's benifit I guess. | 13:08 |
| sdake | morning kfox1111 | 13:08 |
| kfox1111 | morning sdake | 13:08 |
| kfox1111 | :) | 13:08 |
| kfox1111 | portdirect: sbezverk what do you think of the globals values thing? | 13:09 |
| *** awiddersheim has joined #openstack-kolla | 13:11 | |
| *** ocaner has joined #openstack-kolla | 13:11 | |
| portdirect | kfox1111: the -ve 1 was for memecahed, though your point is fair - i kinda want it left so we are forced to fix it | 13:12 |
| portdirect | kfox1111: general approch seems ok though :) | 13:12 |
| kfox1111 | portdirect: I don't think memcached can be "fixed" though. its inherent in its archetecture. | 13:12 |
| kfox1111 | upstream would have to come up with a cache coherent memcache cluste version. | 13:13 |
| kfox1111 | I think thats highly unlkely to happen in the next year (and probably unlikely ever. :/ ) | 13:13 |
| kfox1111 | if you feel strongly about it, I can revert it. but it just feels like a foot gun. | 13:14 |
| portdirect | if you think its unfixable, suppose we better remove the gun : | 13:14 |
| portdirect | so how can we make it better? ds on nodes and bind it to docker ip? | 13:15 |
| kfox1111 | yeah. I'm sure we can't fix it. memcached would have to. :/ | 13:15 |
| portdirect | or put at a memcached container in each pod? | 13:15 |
| kfox1111 | it should fail nicely. if it dies, the cache is wiped, and the new one comes in clean. | 13:16 |
| kfox1111 | for coherency, each api service at least needs to talk to the same memcached. | 13:16 |
| kfox1111 | so we can share it per region, or per service. | 13:16 |
| portdirect | i think per service would make sense, though i just realised that actually running it as a statefulset may solve our issue? | 13:17 |
| sdake | shame memcache d oesn't sycnronize | 13:17 |
| portdirect | (eitherway just kill the var, you are right) | 13:18 |
| sdake | qq, what is memcache used for precisely? | 13:18 |
| SamYaple | morning | 13:18 |
| sdake | i never understood the details | 13:18 |
| sdake | morning SamYaple | 13:18 |
| *** jheroux has joined #openstack-kolla | 13:19 | |
| kfox1111 | statefulset would let you scale out the number of insntances, but would still require you to put every instance in the config of every thing using it. | 13:19 |
| kfox1111 | and dynamic scaling wouldn't really be possible. | 13:19 |
| SamYaple | memcached _shouldnt_ mirror caches, it fails gracefully, mostly | 13:19 |
| ocaner | morning SamYaple, thanks for the welcome and +1 :) | 13:19 |
| kfox1111 | so, yeah, kind of might help | 13:19 |
| sdake | say SamYaple - re adding salt - i was thinking about our wednesday meeting, and i believe it is necessary to hold a core reviewer vote on adding a deliverable | 13:19 |
| sdake | SamYaple we have done this for every other deliverable we have added or intended to add | 13:20 |
| SamYaple | sdake: talk to inc0, not me | 13:20 |
| sdake | i will - just wanted to let you know | 13:20 |
| sdake | i'm in favor | 13:20 |
| sdake | i just don't want you to feel ambushed :) | 13:20 |
| sdake | it has my +1 | 13:20 |
| portdirect | sdake: keystone tokens afaik so needs real care | 13:20 |
| SamYaple | portdirect: actually no! | 13:20 |
| SamYaple | yes its used with that, but not required | 13:20 |
| SamYaple | the only _required_ piece that uses memcached is nova-consoleauth | 13:21 |
| SamYaple | that is the _only_ reason its implemented | 13:21 |
| SamYaple | i can assure you | 13:21 |
| SamYaple | i tried so hard to never use memcached | 13:21 |
| kfox1111 | SamYaple: I've had to use it with horizon at times too. cookies got too big. | 13:22 |
| portdirect | ah ok :) though the in memory cahe for keystonemiddleware is depericated i think? | 13:22 |
| kfox1111 | SamYaple: but did switch to sql after a while. | 13:22 |
| SamYaple | portdirect: sure, it is, but that doesnt make memcached required | 13:22 |
| kfox1111 | I've tried to avoid memcached too for similar reasons I'm sure. | 13:22 |
| SamYaple | kfox1111: yea, but thats also not *required* | 13:22 |
| SamYaple | dont get me wrong, its useful | 13:22 |
| SamYaple | and we take advantage of it | 13:22 |
| SamYaple | but its only required to make multinode nova-consoleauth work | 13:22 |
| kfox1111 | yeah. +1 for optional memcached. | 13:22 |
| portdirect | yeah - with horzion without it and the db backend (required for murano) performance drops off a cliff | 13:23 |
| SamYaple | agreed | 13:23 |
| kfox1111 | I kind of wondered about nova-consoelauth. I couldn't figure it out, so just made it active-passive on the research cloud. | 13:23 |
| SamYaple | has anyone noticed the kolla gate is bad? | 13:23 |
| portdirect | SamYaple: what do you think we should do then? | 13:23 |
| SamYaple | portdirect: im not actually sure the issue, i just was 'what is memcached?' and went with talking | 13:24 |
| portdirect | how we should handle it in kolla-k8s | 13:24 |
| portdirect | we currently deploy 1 container for a whole deplyment and eveything that want to use it uses that | 13:24 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: use static uid/gid in images https://review.openstack.org/412231 | 13:24 |
| kfox1111 | sbezverk: you around? | 13:25 |
| kfox1111 | srwilker-: and you? | 13:25 |
| SamYaple | if you use fernet tokens, and dont plan on using revocations.... then you dont need to use it one for an entire environment | 13:25 |
| SamYaple | basically all the keystoneauth layers have to talk to the same memcached | 13:25 |
| portdirect | we could put a single container in each pod, or create a service that uses a k8s component called a stefullset to run multiple containers | 13:25 |
| SamYaple | outside of that, each service could have its own memcached layer | 13:25 |
| sbezverk | kfox1111: yep | 13:26 |
| kfox1111 | portdirect: per pod won't work, as it wont take invalidations properly? | 13:26 |
| portdirect | cool - sold on that :) | 13:26 |
| kfox1111 | sbezverk: what do you think about the global values aproach? | 13:26 |
| sbezverk | kfox1111: I could not find examples on how to read hashkey and set syntax, so it looks chineese to me, no ofence to chinees folks.. | 13:28 |
| kfox1111 | what about the heirarchy? seem ok? | 13:29 |
| kfox1111 | I'm hoping they can make the black magic in that one macro easier when helm implements a new function that does that logic for us. | 13:29 |
| sbezverk | kfox1111: I do not like so many levels, I think kolla level is really artificial | 13:29 |
| SamYaple | what do you mean sbezverk? | 13:30 |
| kfox1111 | sbezverk: sap said they were doing openstack as a component of a greatter application. | 13:30 |
| kfox1111 | I know others will want to do the same. | 13:30 |
| kfox1111 | putting it under globals.kolla allows that use case. | 13:30 |
| portdirect | kfox1111: what happens if there is a conflict eg port for memecaced? or will that be sorted in a later ps | 13:31 |
| sbezverk | kfox1111: Well our focus was kolla, getting in globals possible "mission to mars" maybe useful in future but should we suffer added complexity now? | 13:31 |
| kfox1111 | portdirect: you can override it at any level. | 13:31 |
| kfox1111 | so, say, you want a neutron-memcached element. | 13:31 |
| kfox1111 | helm install kolla/memcached --set element_name=nova-memcached | 13:32 |
| sbezverk | SamYaple: it is nothing offending to kolla. it is just about structure kfox1111 suggested | 13:32 |
| kfox1111 | and all its values can come from globals.kolla.nova-memcached.* | 13:32 |
| portdirect | what i was meaning was how they are combined -eg the values on 198 and 206 | 13:32 |
| kfox1111 | if you want to override the port in both the deployment and the svc, | 13:32 |
| kfox1111 | globals.kolla.nova-memcached.all.port = 72 | 13:33 |
| SamYaple | sdake: can i get an ack on this https://review.openstack.org/#/c/413119/ | 13:33 |
| SamYaple | already merged on teh kolla side a while bac | 13:33 |
| sbezverk | kfox1111: another thing the structure really hardcoded with ifs so adding level in data requires adding code, instead of just walking the tree | 13:34 |
| kfox1111 | sbezverk: I think kolla should fix it now if they efver want to support the use case, rather then ask users to rewrite config later. adding kolla into globals gives it its own namespace right away and isnt that onerious. | 13:35 |
| kfox1111 | sbezverk: true. but I don't think we can just walk the tree without ifs? | 13:35 |
| sdake | SamYaple looking moment | 13:35 |
| kfox1111 | sbezverk: its also pretty self contained, so if we feel there needs to be other levels, we can change it in one place. | 13:35 |
| sdake | SamYaple looks good | 13:36 |
| SamYaple | sdake: danke | 13:36 |
| sbezverk | kfox1111: you need if to check but probably going with range would make it more flexible | 13:36 |
| SamYaple | sdake: i did check into overlay on centos, but centos kernel is lagging as always | 13:36 |
| kfox1111 | sbezverk: hmm.. where does the range param come from? | 13:37 |
| kfox1111 | I think I see what your getting at... | 13:37 |
| kfox1111 | not sure how to walk the tree like that though. | 13:37 |
| kfox1111 | can an index take a dotted string? | 13:37 |
| kfox1111 | ala, index .Value ".globals.foo" ? | 13:37 |
| sbezverk | kfox1111: you can use string in quote | 13:38 |
| *** Bico_Fino has quit IRC | 13:38 | |
| sdake | SamYaple ok - spoke with inc0 on the ml | 13:38 |
| kfox1111 | if so, yeah, we could replace the iffs with a range. | 13:38 |
| kfox1111 | sbezverk: yeah, but will it let you do dots to do subelements? | 13:38 |
| kfox1111 | cause the tree isn't flat. | 13:38 |
| sbezverk | kfox1111: did not try that | 13:39 |
| openstackgerrit | Merged openstack/kolla-ansible: Switch Ubuntu to overlay2 docker driver https://review.openstack.org/413119 | 13:39 |
| kfox1111 | if it can, then I think we can implement your idea. :) | 13:39 |
| *** Bico_Fino has joined #openstack-kolla | 13:39 | |
| portdirect | SamYaple: is overlay ok on centos now? I had whiteout issues last time i used it anger | 13:40 |
| sbezverk | kfox1111: http://paste.openstack.org/show/593266/ | 13:41 |
| sdake | portdirect not really | 13:41 |
| SamYaple | portdirect: pretty sure its not. there was bugs with it | 13:41 |
| SamYaple | portdirect: it works *sometimes* with *somethings* | 13:41 |
| sdake | portdirect it works without RPM | 13:41 |
| sdake | portdirect the ovl plugin to rpm is busted imo | 13:41 |
| SamYaple | portdirect: until centos gets into 4.0 kernel, its going to be hard to backport all the needed things | 13:41 |
| kfox1111 | sbezverk: need a dotted value though. | 13:41 |
| sdake | portdirect personally i'd roll with btrfs on centos | 13:42 |
| kfox1111 | cause would have to lookup "globals.all.foo" | 13:42 |
| kfox1111 | the ifs all currently look at different places in the tree, | 13:42 |
| kfox1111 | so would have to be changed to look in the same place, and walk down the tree in the string to work with a range. | 13:42 |
| portdirect | SamYaple/sdake: yeah I'm using kernel-lt on my boxes with overlay2 to get round that, so suppose not really centos any more... | 13:42 |
| sdake | what is kernel-lt? | 13:42 |
| sdake | is that a centos package? | 13:43 |
| portdirect | elrepo kernel | 13:43 |
| sdake | cool - i cna't access my dev boxes atm because i got an eero wireless device | 13:43 |
| sdake | so can't check :) | 13:43 |
| sdake | wireless router that is | 13:43 |
| portdirect | so 3rd party - used to build my own but got lazy | 13:43 |
| sdake | sseems to work great | 13:43 |
| sdake | although my machines are not operational - waiting to call support once they are up | 13:44 |
| sbezverk | kfox1111: if you could point the place where you get syntax for hashkey and new set, that would be good. it will help to understand better | 13:44 |
| kfox1111 | k. sec | 13:45 |
| *** senk has quit IRC | 13:45 | |
| kfox1111 | https://godoc.org/github.com/Masterminds/sprig | 13:45 |
| kfox1111 | hasKey: | 13:45 |
| kfox1111 | - hasKey: Takes a dict and a key, and returns boolean true if the key is in the dict. | 13:45 |
| kfox1111 | eqiv to 'key in mydict' in python. | 13:45 |
| kfox1111 | set: Takes a dict, a key, and a value, and sets that key/value pair in | 13:46 |
| kfox1111 | the dict. `set $dict $key $value`. For convenience, it returns the dict, | 13:46 |
| kfox1111 | even though the dict was modified in place. | 13:46 |
| *** mdnadeem has joined #openstack-kolla | 13:46 | |
| kfox1111 | go doesn't like hanging returns, so they use a pattern of $_ to define away values. | 13:46 |
| kfox1111 | so {{ $_ := set mydict "foo" "bar" }} | 13:46 |
| kfox1111 | adds foo=bar to mydict. | 13:47 |
| sbezverk | kfox1111: ok $_ is just a placeholder as in go | 13:48 |
| kfox1111 | yeah. its a black hole var. | 13:48 |
| kfox1111 | the side affect of the set is the thing that makes it useful. | 13:49 |
| kfox1111 | being able to actually change the stuff in the dict. | 13:49 |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 13:49 |
| *** prameswar has quit IRC | 13:53 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 13:55 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 13:56 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 14:01 |
| kfox1111 | sbezverk: so, does the macro look ok as a concept and how its called? | 14:01 |
| kfox1111 | we could merge a dummy macro that all it does is call .Values.$foo for now, and then we can work on the details of the heirarchy in parallel with fixing up the microservice packages to call the macro. | 14:02 |
| sbezverk | kfox1111: the macro idea is great, for sure as it hides all these wierdness | 14:03 |
| sbezverk | kfox1111: I do not see why we should not merge it and play with it to see how it behaves in practice | 14:04 |
| kfox1111 | ok. cool. | 14:04 |
| kfox1111 | to answer the dotted question, it seems no: | 14:04 |
| kfox1111 | index Returns the result of indexing its first argument by the following arguments. Thus "index x 1 2 3" is, in Go syntax, x[1][2][3]. Each indexed item must be a map, slice, or array. | 14:04 |
| *** Serlex has joined #openstack-kolla | 14:05 | |
| dmsimard | sdake, Jeffrey4l: so I'm trying to understand the thing with tox -- are you looking to move the gate jobs away to something other than "tox -e build-something" ? | 14:05 |
| kfox1111 | so, it still might be doable, if we could make a new array that is the merging of two other arrays. | 14:05 |
| kfox1111 | not sure if gotl's flexible enough to do that though. :/ | 14:06 |
| *** ipsecguy_ has quit IRC | 14:06 | |
| *** prameswar has joined #openstack-kolla | 14:06 | |
| sbezverk | kfox1111: if you could clean up memcached ps, I think we could merge it.. | 14:07 |
| kfox1111 | k. I'll work on doing that. | 14:07 |
| *** ipsecguy has joined #openstack-kolla | 14:12 | |
| openstackgerrit | Merged openstack/kolla-kubernetes: Python 3.5 support for Kolla https://review.openstack.org/414270 | 14:12 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Global api https://review.openstack.org/414350 | 14:16 |
| SamYaple | w00t a sub 13m ubuntu gate run! | 14:26 |
| SamYaple | https://review.openstack.org/#/c/413394/ | 14:26 |
| SamYaple | thats the lowest yet | 14:26 |
| kfox1111 | nice. | 14:26 |
| SamYaple | overlay2 driver allows me to increase the thread count | 14:27 |
| SamYaple | for building | 14:27 |
| SamYaple | more performance | 14:27 |
| kfox1111 | thats just build? | 14:28 |
| *** salv-orl_ has joined #openstack-kolla | 14:31 | |
| *** salv-orlando has quit IRC | 14:34 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 14:35 |
| kfox1111 | thats really weird... | 14:38 |
| kfox1111 | did zuul get stuck? | 14:38 |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move aodh precheck into its own role https://review.openstack.org/404389 | 14:38 |
| kfox1111 | all jobs compled, but still in the gate and no gerrit review posted yet... | 14:38 |
| SamYaple | kfox1111: nah it just takes a second | 14:39 |
| *** jtriley has joined #openstack-kolla | 14:39 | |
| kfox1111 | was more like 2 minutes that time. | 14:40 |
| *** zhubingbing has quit IRC | 14:40 | |
| SamYaple | kfox1111: nah it just takes a seconds | 14:40 |
| SamYaple | :) | 14:40 |
| kfox1111 | never seen it take that long before. | 14:41 |
| *** Bico_Fino has quit IRC | 14:42 | |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move congress precheck into its own role https://review.openstack.org/413971 | 14:43 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Adds cinder volume lvm microservice https://review.openstack.org/414348 | 14:46 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Global api https://review.openstack.org/414350 | 14:47 |
| *** mdnadeem has quit IRC | 14:48 | |
| *** gomarivera has joined #openstack-kolla | 14:48 | |
| openstackgerrit | caoyuan proposed openstack/kolla-ansible: Move barbican precheck into its own role https://review.openstack.org/404465 | 14:48 |
| *** tonanhngo has joined #openstack-kolla | 15:02 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 15:02 |
| *** gomarivera has quit IRC | 15:03 | |
| *** senk has joined #openstack-kolla | 15:04 | |
| *** tonanhngo has quit IRC | 15:06 | |
| *** salv-orl_ has quit IRC | 15:12 | |
| openstackgerrit | Merged openstack/kolla: Readability improvement to base image https://review.openstack.org/411154 | 15:14 |
| *** goldyfruit has joined #openstack-kolla | 15:14 | |
| *** gomarivera has joined #openstack-kolla | 15:22 | |
| kfox1111 | sbezverk: https://review.openstack.org/#/c/414350/ | 15:23 |
| kfox1111 | portdirect: can you have a look too please? | 15:26 |
| *** salv-orlando has joined #openstack-kolla | 15:27 | |
| SamYaple | can this patch get some love? its been sitting for days https://review.openstack.org/#/c/412452/ | 15:27 |
| sbezverk | kfox1111: why do we still need common section? is it just for backward compatibility? | 15:29 |
| sbezverk | global is supposed to give visibility to variables to all subcharts | 15:30 |
| kfox1111 | the defaults need to go somewhere. | 15:30 |
| kfox1111 | if you run the subcharts on their own, without setting anything, it needs to get its defautls from its values file. | 15:30 |
| kfox1111 | I think once the api settles, we can consider flatting stuff back out to just regular values.yaml files in each microservice. | 15:31 |
| *** salv-orlando has quit IRC | 15:31 | |
| kfox1111 | but for now, the common stuff makes it so much easier to make a change and hit all the packages. | 15:32 |
| *** Pavo has joined #openstack-kolla | 15:32 | |
| SamYaple | DRY | 15:33 |
| kfox1111 | SamYaple: yeah. that too. :) | 15:33 |
| *** caowei has quit IRC | 15:33 | |
| kfox1111 | I'm a little curious if subchart's global defaults ripple back to the parent. | 15:35 |
| kfox1111 | if so, once we get to the global api everywhere, we might be able to just move the defaults to the kolla-common chart | 15:36 |
| sbezverk | kfox1111: I still not convinced that so many nesting is really necessary, but I guess I am tired of discussing it.. | 15:37 |
| kfox1111 | sbezverk: k. thanks. once we get a little further in, we can reevaluate if so many are useful or if we can remove some. Shoudl be pretty easy to remove later. | 15:39 |
| *** zhubingbing has joined #openstack-kolla | 15:40 | |
| kfox1111 | portdirect: can you please wf if your still ok with it? | 15:42 |
| *** mnasiadka has quit IRC | 15:42 | |
| *** rmart04 has quit IRC | 15:44 | |
| sdake | kfox1111 which patch | 15:45 |
| kfox1111 | https://review.openstack.org/#/c/414350/ | 15:45 |
| sdake | kfox1111 we should be using 2 crs now that we have beefed up the team :) | 15:45 |
| sdake | moment i'll review | 15:45 |
| kfox1111 | sdake: yup. thats why I'm asking for a second. :) | 15:45 |
| kfox1111 | I'm glad we've gotten to the point where we can get back to 2 x +2's. | 15:46 |
| sdake | is idea to move 15-183 into globals? | 15:46 |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 15:46 |
| kfox1111 | sdake: which file? | 15:47 |
| sdake | https://review.openstack.org/#/c/414350/8/helm/all_values.yaml | 15:47 |
| *** _chas_ has quit IRC | 15:48 | |
| kfox1111 | sdake: yeah. the ps allows for globals to be used in a heirchy so you can set a setting generically and hit all the microservices with one config file, | 15:48 |
| kfox1111 | or you can set a global and then override it on a specific instance of microservice. | 15:49 |
| kfox1111 | so in the end, it works similar to how /etc/kolla/globals.yaml works. | 15:49 |
| kfox1111 | you can write one file, and load it with any service/microservice you want to launch, like | 15:49 |
| kfox1111 | helm install kolla/foo --values myglobals.yaml | 15:49 |
| sdake | cool so that could handle subdeploy? | 15:50 |
| kfox1111 | yeah. it will work with that too. | 15:50 |
| sdake | hot | 15:51 |
| sdake | when did that come into helm? | 15:51 |
| kfox1111 | sdake: 2.1.3. :) | 15:51 |
| kfox1111 | the thing sbezverk and I was pushing for. :) | 15:51 |
| sdake | to me that was the big blocker of the hierarchrial approach | 15:51 |
| kfox1111 | like, yesterday. :) | 15:51 |
| kfox1111 | yeah. | 15:51 |
| sdake | now we can actually do that | 15:51 |
| kfox1111 | helm really stepped up for us. :) | 15:51 |
| sdake | nice - v1k0d3n had a hand in that :) | 15:52 |
| kfox1111 | +1 | 15:52 |
| sdake | i'd like to say i had, but i've gien up on upstreams doing precisely what a downstream needs | 15:52 |
| sdake | ;) | 15:52 |
| kfox1111 | v1k0d3n: thanks again for your help building bridges. :) | 15:52 |
| sdake | i wonder what other "smalish" fixes we can do to get to a more helm-native approach | 15:53 |
| sdake | (to helm that is) | 15:53 |
| v1k0d3n | of course. happy to help. | 15:53 |
| kfox1111 | I posted a wishlist. | 15:53 |
| v1k0d3n | :) | 15:53 |
| sdake | kfox1111 nice may I see? | 15:53 |
| kfox1111 | I put it in the etherpad inc0 posted too. | 15:53 |
| sdake | i guess i missed that etherpad | 15:54 |
| openstackgerrit | Merged openstack/kolla-kubernetes: Global api https://review.openstack.org/414350 | 15:54 |
| sdake | was that on the mailing list? | 15:54 |
| *** senk has quit IRC | 15:54 | |
| sdake | have a link? | 15:55 |
| sdake | kfox1111 link plz :) | 15:57 |
| kfox1111 | sry. my son just walked over with no diaper on.... | 15:59 |
| sdake | oops | 15:59 |
| sdake | that can be a problem :) | 15:59 |
| kfox1111 | I don't have a link handy. it was just after the helm meeting from the other day. | 15:59 |
| kfox1111 | he posted the link to this channel. so should be in the log. | 16:00 |
| Pavo | is it possible to add the murano plugin into horizon container manually? | 16:00 |
| sdake | link should be posted to ml - i'll hut it down in log after i return from torture session | 16:00 |
| kfox1111 | k | 16:00 |
| sdake | unless portdirect or sbezverk have it handy | 16:00 |
| Pavo | I am reading the guide on how to do it but there isn't a openstack_dashboard/local folder in the horizon container | 16:00 |
| *** sayantan_ has joined #openstack-kolla | 16:02 | |
| kfox1111 | darn. yeah, globals from subcharts don't ripple back to parent charts. | 16:04 |
| kfox1111 | so prebuilding values still is probably the best thing for now. | 16:04 |
| sdake | if we could get the ripple going - would that allow us to be more helm native? | 16:05 |
| kfox1111 | yeah. | 16:06 |
| sdake | sounds like a good thing to me | 16:06 |
| kfox1111 | I'm not entirely sure, but that might eliminate the need for prebuild entirely. | 16:06 |
| sdake | that would be pretty sweet - tbh i was never a fan of prebuild but it seemed necessary at the time | 16:07 |
| sdake | and still does now | 16:07 |
| kfox1111 | yeah. | 16:07 |
| kfox1111 | I'd like to get rid of it too, as soon as helm has all the needed bits. | 16:07 |
| sdake | v1k0d3n do you have a link to the etherpad? | 16:08 |
| sdake | we need etherpads for our etherpads :) | 16:08 |
| *** srwilkers has joined #openstack-kolla | 16:08 | |
| srwilkers | morning | 16:09 |
| sdake | srwilkers do you have a link to the etherpad inc0 put together of our collaboration with upstream | 16:09 |
| kfox1111 | sdake: https://etherpad.openstack.org/p/openstack-helm | 16:09 |
| sdake | thanks :) | 16:09 |
| sdake | obvious enoguh - should have tried that :) | 16:10 |
| kfox1111 | hindsight... :) | 16:11 |
| srwilkers | :) | 16:11 |
| *** zhubingbing has quit IRC | 16:11 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 16:11 |
| kfox1111 | sdake: the one that will make it possible to do more helm native is the "library values" thing. | 16:12 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Ensure all microservices pass Release https://review.openstack.org/413847 | 16:16 |
| *** ssurana has joined #openstack-kolla | 16:17 | |
| *** sdake has quit IRC | 16:23 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 16:24 |
| *** salv-orlando has joined #openstack-kolla | 16:27 | |
| Pavo | why do I keep getting Failed to get D-Bus connection: Operation not permitted when I try to restart the httpd service in horizon container? | 16:28 |
| sbezverk | kfox1111: which version of tool box jq is included? | 16:28 |
| kfox1111 | 3.0.1+ | 16:28 |
| Pavo | ah nvm systemctl doesn't work in containers | 16:29 |
| sbezverk | kfox1111: thanks | 16:29 |
| Pavo | how do you restart a service in a container? | 16:29 |
| kfox1111 | Pavo: depends I guess. cloud phylosophy is to recreate the container rather then restart. | 16:30 |
| Pavo | I added the murano dashboard files to the horizon container and need to restart the httpd service within horizon container | 16:31 |
| Pavo | I tried restarting the container but that didn't work | 16:31 |
| kfox1111 | did you add it to the container image, or the running container? | 16:31 |
| Pavo | running container | 16:31 |
| kfox1111 | then it gets destroyed as soon as the container is rebuilt. | 16:31 |
| *** salv-orlando has quit IRC | 16:32 | |
| Pavo | its still there when I just did a docker restart | 16:32 |
| Pavo | but trying to figure out how to restart the httpd service within the horizon container | 16:33 |
| Pavo | systemctl restart doesn't work, service httpd restart doesn't work | 16:33 |
| kfox1111 | try and kill the http process. | 16:33 |
| kfox1111 | that might cause the container to restart? | 16:34 |
| kfox1111 | not sure how kolla-ansible sets it up though. | 16:34 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 16:34 |
| *** inc0 has joined #openstack-kolla | 16:34 | |
| kfox1111 | hmm... we're pretty inconsistent with our variable definitions right now.... | 16:35 |
| kfox1111 | we need to clean that up. | 16:36 |
| inc0 | good morning | 16:36 |
| Pavo | hmmm | 16:36 |
| Pavo | killing process didn't work | 16:37 |
| Pavo | operation not permitted | 16:37 |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 16:37 |
| SamYaple | morning inc0 | 16:37 |
| kfox1111 | Pavo: can kill it from the host? | 16:38 |
| Pavo | yeah and now the container is gone lol | 16:38 |
| Pavo | lol oh well time to redeploy | 16:39 |
| kfox1111 | show up with a docker ps -a? | 16:39 |
| kfox1111 | redeply may recreate the container, loosing your changes. | 16:40 |
| Pavo | already did a destroy | 16:40 |
| kfox1111 | (you should edit the container image anyway, to prevent that issue) | 16:40 |
| Pavo | give me 10 | 16:40 |
| Pavo | how to edit container image? | 16:40 |
| kfox1111 | the sofware installs really should be in the image. seperates build from the deploy. | 16:40 |
| kfox1111 | not sure. never done it before. | 16:40 |
| kfox1111 | someone else here probably knows though. | 16:40 |
| Pavo | hmmm seems like to start the container in interactive mode | 16:41 |
| Pavo | then after edit, you comment the changes | 16:42 |
| Pavo | interesting | 16:42 |
| kfox1111 | thats one way, butI don't think thats the recommended kolla way? | 16:42 |
| Pavo | true | 16:42 |
| kfox1111 | that procedure produces a pet image. hard to rebuild when software gets updated. | 16:42 |
| Pavo | but again I don't really know the kolla way lol | 16:42 |
| kfox1111 | someone here's bound to know. | 16:43 |
| kfox1111 | or it may be documented. have you looked in the kolla docs? | 16:43 |
| Pavo | well once the murano changes get sent up to pip package it will work with just having murano enabled in globals I hope | 16:43 |
| Pavo | yeah | 16:43 |
| *** unicell has quit IRC | 16:46 | |
| *** Bico_Fino has joined #openstack-kolla | 16:48 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: DONOTMERGE https://review.openstack.org/413786 | 16:48 |
| SamYaple | black hole var? that sounds sexy. what does it do? kill the universe? | 16:49 |
| *** ssurana has quit IRC | 16:49 | |
| SamYaple | sorry just reading scrollback there kfox1111 | 16:49 |
| *** tonanhngo has joined #openstack-kolla | 16:52 | |
| *** Bico_Fino has quit IRC | 16:53 | |
| *** Mech422 has joined #openstack-kolla | 16:54 | |
| *** tonanhngo has quit IRC | 16:57 | |
| kfox1111 | SamYaple: values go in, and don't come back out. :) | 16:58 |
| kfox1111 | basically in go, all function return values need to be assigned to a variable. | 16:58 |
| kfox1111 | "so you know you didn't ignore something important" | 16:58 |
| kfox1111 | but, there is a second rule, | 16:59 |
| kfox1111 | you can't not use a variable. | 16:59 |
| kfox1111 | they enforce that code cleanliness issue as part of the language. | 16:59 |
| *** gfidente has quit IRC | 16:59 | |
| kfox1111 | so, they came up with a stopgap/hack to work around it. | 16:59 |
| kfox1111 | $_ is a magic variable that isn't bound by that second rule. | 17:00 |
| kfox1111 | you can assign anything you want into it, and then never access it. | 17:00 |
| kfox1111 | so, an "I don't care about this thing" value. | 17:01 |
| kfox1111 | like, /dev/null :) | 17:01 |
| *** jrich523 has joined #openstack-kolla | 17:01 | |
| portdirect | SamYaple: ping | 17:06 |
| srwilkers | $_, taking me back to my perl days :/ | 17:08 |
| kfox1111 | yeah. but a little bit different. | 17:09 |
| srwilkers | yep, different, but still giving me flashbacks nonetheless | 17:09 |
| kfox1111 | I kind of like perl's $_ | 17:09 |
| kfox1111 | most of the rest of the launguage... | 17:09 |
| kfox1111 | :) | 17:09 |
| srwilkers | its great for what it's meant to be used for, but not in my top favorites for sure | 17:10 |
| kfox1111 | got real mixed feeling about it. | 17:10 |
| srwilkers | liked the work i did using it though | 17:10 |
| kfox1111 | its best feature is its worst. | 17:10 |
| SamYaple | portdirect: pong | 17:10 |
| kfox1111 | its so flexible you can basically do anythign you want in whatever way you want. | 17:10 |
| kfox1111 | which is great for a single developer. | 17:10 |
| kfox1111 | its hell for a multidev team. | 17:10 |
| srwilkers | terrible for a team | 17:10 |
| kfox1111 | yeah. :) | 17:11 |
| kfox1111 | and since its rare a code won't pass hands at some point, its hell to be the next guy to pick up the code. :/ | 17:11 |
| *** Bico_Fino has joined #openstack-kolla | 17:12 | |
| portdirect | SamYaple: thought you might know without needing to trawl through a load of docs :) what are the requiremnts for a ceph key in the keyrings? | 17:12 |
| srwilkers | yep. i worked at a lab for awhile that used it, and getting up to speed was a challenge for sure | 17:12 |
| portdirect | SamYaple: it would be awesome if i could easily create them using sprig | 17:13 |
| kfox1111 | srwilkers: yeah. and then that knowldge doesn't nessisarily apply when you go to a different code. :/ | 17:13 |
| kfox1111 | portdirect: uh, they come from ceph. | 17:13 |
| kfox1111 | you ask ceph to create you one, then you store the output. | 17:14 |
| *** pcaruana has quit IRC | 17:14 | |
| portdirect | kfox1111: I'm looking to create them outside of ceph | 17:15 |
| sbezverk | kfox1111: do you define which packages are installed on VM for the gate? | 17:15 |
| kfox1111 | portdirect: they are created by ceph and returned to ceph as an auth token. you can't just build one yourself. | 17:18 |
| kfox1111 | sbezverk: yeah. sec. | 17:18 |
| SamYaple | portdirect: hmm define requirement? | 17:18 |
| kfox1111 | sbezverk: see setup_gate line 51 | 17:19 |
| SamYaple | portdirect: im just a bit confused here. so i never make the keys, ceph does | 17:20 |
| SamYaple | portdirect: i just _ask_ ceph for the keys | 17:20 |
| SamYaple | portdirect: i ask in a way that says if you dont have the key, create the key | 17:20 |
| kfox1111 | SamYaple: I think he's trying to create it like a pw and push it to multiple places, | 17:20 |
| SamYaple | kfox1111: nope. cant to that. | 17:20 |
| kfox1111 | instead of creating it in ceph, pulling it back and pushing it to the rest of the places. | 17:20 |
| kfox1111 | yeah. | 17:20 |
| SamYaple | actually thats wrong, you _can_ but yo uabsultely shouldnt | 17:20 |
| SamYaple | you can manually build a crush map | 17:20 |
| SamYaple | but you shouldnt | 17:20 |
| SamYaple | thats a mistake | 17:21 |
| SamYaple | anyway, look at what i do for ceph in kolla-ansible. i query ceph from within the container | 17:21 |
| SamYaple | and then distribute that key | 17:21 |
| sbezverk | kfox1111: I need to make sure that iscsi tools are NOT installed on VM | 17:21 |
| kfox1111 | SamYaple: yeah, but he's trying to create a auth credential. I don't think you can do that at all. | 17:21 |
| sbezverk | I suspect you do not really care for ceph right? | 17:21 |
| *** shardy has quit IRC | 17:21 | |
| kfox1111 | sbezverk: I dont know what the gate itself installs. | 17:22 |
| portdirect | SamYaple: was looking fo clarification behind this: https://github.com/ceph/ceph-docker/blob/master/examples/kubernetes/generator/ceph-key.py | 17:22 |
| portdirect | will hit up leseb when hes about | 17:22 |
| kfox1111 | sbezverk: ceph doesn't use the iscsi bits at all, so should be pretty safe to always uninstall it if you find it? | 17:22 |
| SamYaple | portdirect: oh. yea you could do that | 17:23 |
| kfox1111 | oh, thats knarly. :/ | 17:23 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 17:23 |
| portdirect | yeah thats why i was wanting clarification :) | 17:23 |
| kfox1111 | I don't think that random pool is cryptographically secure. | 17:23 |
| sbezverk | kfox1111: yeah found it and tested in my test bed.. | 17:24 |
| SamYaple | kfox1111: its not | 17:24 |
| SamYaple | portdirect: you know that you still have to inject that key into ceph right? | 17:24 |
| portdirect | yes | 17:24 |
| SamYaple | i dont know about kube to much, but if you are in a possition to inject it, why can you not _read_ it too? | 17:24 |
| kfox1111 | portdirect: yeah, I'd recommend against that aproach. it is insecure from a randomness perspective, and deals with the internals of a non public datastructure | 17:25 |
| SamYaple | kfox1111: actually that part is public | 17:25 |
| SamYaple | thats just generating the initial keys | 17:25 |
| kfox1111 | SamYaple: just a little harder to order if you have to pull. | 17:25 |
| SamYaple | you can define a key from a file, no problem | 17:25 |
| SamYaple | i didnt realize he was ok with _pushing_ said key into ceph | 17:25 |
| portdirect | yup, sorry not wanting to open th rabbit warren - I'll leave you guys alone | 17:25 |
| kfox1111 | SamYaple: what do you mean inital keys? the dont change I thought. | 17:26 |
| SamYaple | portdirect: is the main concern you want to have the key available outside of ceph (you can ask ceph and distribute the response)? | 17:26 |
| SamYaple | kfox1111: they dont, but you can also generate and push keys into ceph for pools and stuff (the thing configs need) | 17:26 |
| SamYaple | portdirect: id like to change my answer to 'yes' | 17:27 |
| portdirect | ^^ as sam says above | 17:27 |
| SamYaple | if you have the ability to push into ceph, but you just cant easily read, then you can generate keys, sure | 17:27 |
| portdirect | yeah - was wanting to get the spec of the key i need to generate, but suspect it'll be a case of going to the source | 17:28 |
| SamYaple | portdirect: nah its documented somewhere | 17:28 |
| SamYaple | portdirect: kfox1111 https://github.com/ceph/ceph-deploy/blob/master/ceph_deploy/new.py#L21 | 17:30 |
| SamYaple | yesh. thats in the main ceph tool | 17:31 |
| kfox1111 | SamYaple: heh. well, not any worse then that then. :/ | 17:31 |
| portdirect | np thanks guys | 17:34 |
| *** mannamne has quit IRC | 17:36 | |
| SamYaple | portdirect: now lets talk about cryptography! | 17:36 |
| *** senk has joined #openstack-kolla | 17:36 | |
| * portdirect double facepalm | 17:36 | |
| kfox1111 | heh. | 17:37 |
| portdirect | "It is well-known that five-way secret sharing has been illegal since the Protestant Reformation [Luther1517]. However, using recent advances in polynomial-time Bojangle projections, we demonstrate how a set of peers who are frenemies can exchange up to five snide remarks that are robust to Bojangle-chosen plaintext attacks." | 17:37 |
| SamYaple | https://turncoffeeintocode.files.wordpress.com/2015/07/2009-02-02-emacs-vs-vim.png | 17:38 |
| *** ssurana has joined #openstack-kolla | 17:39 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: tDONOTMERGE https://review.openstack.org/413786 | 17:41 |
| *** msimonin has quit IRC | 17:48 | |
| *** ipsecguy has quit IRC | 17:48 | |
| *** cliles has quit IRC | 17:50 | |
| *** sdake has joined #openstack-kolla | 17:50 | |
| *** cliles has joined #openstack-kolla | 17:50 | |
| *** ipsecguy has joined #openstack-kolla | 17:54 | |
| sdake | kfox1111 portdirect i think one thing that would hep with - atlast with kfox's wishlist as it relates to helm native, is to order them | 17:55 |
| sdake | a bullet list is unordered | 17:55 |
| sdake | can we get them into an ordered list | 17:55 |
| sdake | and by we, I mean you :) | 17:55 |
| kfox1111 | sure. | 17:56 |
| kfox1111 | but in a bit. | 17:56 |
| sdake | portdirect i cut and pasted your name to make it alphabetical, unfortunately it lost your color profile | 17:57 |
| sdake | can you fix that? :) | 17:57 |
| sdake | tx :) | 17:57 |
| kfox1111 | sbezverk: sdake: portdirect: https://review.openstack.org/#/c/413847/ is ready | 17:59 |
| *** sdake_ has joined #openstack-kolla | 18:00 | |
| *** sdake has quit IRC | 18:02 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 18:03 |
| *** jtriley has quit IRC | 18:04 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: tDONOTMERGE https://review.openstack.org/413786 | 18:06 |
| kfox1111 | sdake_: https://review.openstack.org/#/c/413847/ ? | 18:06 |
| sdake_ | gah wall of review :) | 18:07 |
| sdake_ | moment | 18:07 |
| kfox1111 | sdake_: yeah. now that helm 2.1.3 is in, going to try and crank through a bunch of changes to get microservices in good shape for 0.4.0 | 18:07 |
| kfox1111 | will have to touch pretty much all the files. :/ | 18:08 |
| sdake_ | kfox1111 looks good | 18:10 |
| sdake_ | although few small nitpicks - no blueprint id, no reno file | 18:10 |
| sdake_ | might as well start using reno sooner rather then later :) | 18:10 |
| sdake_ | it *proves* to people that work is being done | 18:10 |
| sdake_ | without looking at the git repo | 18:11 |
| sdake_ | reno content doesn't have to be extensive | 18:11 |
| sdake_ | it can be something simple like "Pass release values to all microservices." | 18:11 |
| sdake_ | it literally will take you 5 minutes | 18:12 |
| kfox1111 | sdake_: I think the 30 ps's coming in the next 24 hours will be proof enough there is work being done. ;) | 18:13 |
| kfox1111 | and the time to put useless to the user values like "Pass release values to all microservices." will slow that down. :/ | 18:13 |
| kfox1111 | really I think reno is for building notes for the users that they will care about. | 18:14 |
| kfox1111 | rather then commit logs which they dont. | 18:14 |
| kfox1111 | how about this, I promice to start using reno as soon as the globals refactor is done? | 18:14 |
| kfox1111 | arg... changing a commit log will trigger a whole gate rebuild. :/ | 18:15 |
| kfox1111 | and 30 more min at least. :/ | 18:15 |
| sdake_ | kfox1111 wfm | 18:18 |
| kfox1111 | k. thx. | 18:18 |
| sdake_ | ok dont sweat this one | 18:18 |
| inc0 | release notes are most important post 1.0 | 18:18 |
| sdake_ | try to do the blueprints in future and reno after refactor | 18:18 |
| sdake_ | also important prior | 18:18 |
| kfox1111 | k. thx. | 18:18 |
| inc0 | as form of documentation | 18:19 |
| inc0 | but they wont replace documentation | 18:19 |
| kfox1111 | yeah. I'm planning on trying to use 0.4.0 as a user, so would benifit from reno post 0.4.0 | 18:19 |
| openstackgerrit | Edmund Rhudy proposed openstack/kolla: Enable building without Internet access for most Ubuntu binary containers https://review.openstack.org/414639 | 18:19 |
| sdake_ | inc0 ya not a replacement for docs for sure | 18:19 |
| sdake_ | alot of people look at the release notes not the git repo however | 18:20 |
| *** sdake_ is now known as sdake | 18:20 | |
| kfox1111 | yeah. right before cutting 0.4.0 we should do a reno entry for what was genenrally done. | 18:21 |
| kfox1111 | then we can start tracking changes. | 18:21 |
| sdake | kfox1111 yup we did that for liberty - that works | 18:21 |
| jrich523 | you guys are working mighty hard for christmas eve :) | 18:21 |
| sdake | but it was a hack :) | 18:21 |
| sdake | christmas eve is tomorrow jrich523 :) | 18:22 |
| jrich523 | .... | 18:22 |
| jrich523 | you sir, are correct | 18:22 |
| jrich523 | let me try again | 18:22 |
| jrich523 | you guys are working might hard for the friday before xmas :) | 18:22 |
| sdake | ;-) | 18:22 |
| kfox1111 | sdake: yeah, I think tis a common thing for reno not to be in place on day one of a new project. | 18:22 |
| sdake | ya it is friday :) | 18:22 |
| jrich523 | yeah it sure is :) | 18:22 |
| openstackgerrit | Merged openstack/kolla-kubernetes: Ensure all microservices pass Release https://review.openstack.org/413847 | 18:22 |
| sdake | kfox1111 we introduced reno early on | 18:22 |
| jrich523 | hope you guys are taking some time off | 18:22 |
| sdake | on purpose so people could track their changes | 18:23 |
| sdake | jrich i am taking time off - starting 22nd - 4th :) | 18:23 |
| sdake | my torture session wasn't as bad as I thought :) | 18:23 |
| *** prameswar has quit IRC | 18:23 | |
| jrich523 | oh wow, nice, going places? | 18:23 |
| sdake | my house is orderly | 18:23 |
| sdake | staying at home | 18:23 |
| jrich523 | lol always nice | 18:23 |
| sdake | my presents are wrapped | 18:23 |
| sdake | my house is bordering in even clean :) | 18:24 |
| sdake | in/on | 18:24 |
| sdake | so i'm basically bored atm :) | 18:24 |
| sdake | and throwing peanuts from the peanut gallery :) | 18:24 |
| jrich523 | lol | 18:24 |
| sdake | jrich523 this is what defines passion :) | 18:24 |
| jrich523 | i wish i could say the same | 18:24 |
| jrich523 | indeed | 18:24 |
| jrich523 | i have a small hint of passion some place in side me | 18:24 |
| jrich523 | i wish i had the same level that you do | 18:25 |
| jrich523 | but alas, i like video games too much | 18:25 |
| jrich523 | lol | 18:25 |
| sdake | nah you probably dont :) | 18:25 |
| sdake | i liked wow alot when i was younger | 18:25 |
| sdake | had an unhealthy obsession with it | 18:25 |
| jrich523 | meh, it would be nice to accomplish something | 18:25 |
| sdake | my kids are at school or sleeping | 18:25 |
| jrich523 | lol wow ate a chunk of my life | 18:25 |
| sdake | i did accomplish something with wow | 18:25 |
| sdake | I learned how to work with 20 year olds as a 30 year old :) | 18:25 |
| jrich523 | lol nothing you should be too proud of tho | 18:25 |
| jrich523 | lol | 18:25 |
| inc0 | I still have unhealthy obsession about games | 18:25 |
| sdake | jrich523 i was in a top 50 us guild in wow | 18:26 |
| sdake | now I don't have time for games :) | 18:26 |
| jrich523 | lol im not surprised, its your personality | 18:26 |
| sdake | atleast those without a pause button | 18:26 |
| inc0 | yeah I don't play much mmo's any more | 18:26 |
| sdake | jrich523 yo ugot that right, i dont do anything half assed ;) | 18:26 |
| jrich523 | lol yeah my taste in games has changed, if it doesnt have a 15-20 min round, i dont play | 18:26 |
| sdake | i play diablo a bit here and there | 18:26 |
| jrich523 | im the king of half ass lol.. you need a project 80% done, im your guy | 18:27 |
| sdake | 15 mins and then i can go do something else | 18:27 |
| jrich523 | thats a great game, i was playing path of exile, free/similar and good | 18:27 |
| sdake | haven't played | 18:27 |
| inc0 | I like progression idea in PoE | 18:27 |
| sdake | my son's pc is broken and he has taken over my office | 18:27 |
| inc0 | innovative | 18:27 |
| jrich523 | had to introduce the wife to 'click to move games', start there and move to D3 if she lasts :) | 18:27 |
| jrich523 | inc0: its infuriating lol | 18:27 |
| jrich523 | sdake: make him fix it | 18:27 |
| jrich523 | its how i learned, and got good with PCs | 18:28 |
| sdake | jrich523 i can't even fix it | 18:28 |
| jrich523 | "oh crap i broke it, dads going to kill me" | 18:28 |
| sdake | its water cooled and whatnot | 18:28 |
| inc0 | sdake, just act like you do | 18:28 |
| inc0 | and have him do it | 18:28 |
| jrich523 | lol well, he's younger, and if motivated (no pc till you fix it) he'll get it done :) | 18:28 |
| sdake | i can change out the ram - that may fix it | 18:28 |
| jrich523 | lol good plan inc0 :) | 18:28 |
| sdake | going to try that today actually | 18:28 |
| inc0 | I've setup my water cooled pc because I never had one;) | 18:29 |
| inc0 | now it's nice and quiet | 18:29 |
| sdake | i bought one from digital storm for both kids | 18:29 |
| sdake | triple way sli | 18:29 |
| sdake | its pretty sweet - wish I ahd one :) | 18:29 |
| inc0 | to play diablo3?;) | 18:29 |
| sdake | they play all kinds of games | 18:29 |
| sdake | my 1080 does everything i need on diablo 3 | 18:29 |
| sdake | they unfortunately ahve 980s | 18:30 |
| inc0 | 1080 will run all the games I know of on max detail without SLI | 18:30 |
| inc0 | 980 too | 18:30 |
| sdake | not with hi res screens | 18:30 |
| inc0 | I have radeon r9 390 and it handles VR very well | 18:30 |
| Pavo | I run a RX 480 in my macpro runs very well | 18:30 |
| inc0 | so 4k+90hz refresh rate | 18:30 |
| sdake | my mac pro with integrated video is sluggish on d3 | 18:30 |
| sdake | kids aren't doing vr until they get older | 18:31 |
| sdake | too immersive | 18:31 |
| sdake | i debated even buying them a microphone | 18:31 |
| Pavo | what gfx card you have in your macpro sdake | 18:31 |
| sdake | but bought one anyway | 18:31 |
| sdake | pavo i don't recall i think r280 | 18:31 |
| inc0 | yeah vr is crazy | 18:31 |
| inc0 | and I mean it | 18:31 |
| Pavo | should run fine in D3 | 18:31 |
| portdirect | I've got 12 7970's, but no games :/ | 18:31 |
| sdake | portdirect i don't even want ot knwo ;) | 18:31 |
| jrich523 | do you think water cooled is worth it? and do you OC? | 18:32 |
| inc0 | I need to move to nvidia from radeon to play arond cuda | 18:32 |
| sdake | jrich523 yup memory is oced, cpu oced, video oced | 18:32 |
| inc0 | jrich523, I don't OC, but it's super quiert | 18:32 |
| inc0 | and not that expensive any more | 18:32 |
| jrich523 | thats a solid point inc0, but im typically with headphones on, so i dont care :) | 18:32 |
| jrich523 | mostly i dont see the reason (i dont oc/noise meh) | 18:32 |
| jrich523 | seems, dangerous too | 18:33 |
| inc0 | I use headphones mostly too, but for times I don't it makes sense | 18:33 |
| jrich523 | one little leaks and boom gone | 18:33 |
| sdake | my workstation is a hp z820 | 18:33 |
| sdake | its pretty solid | 18:33 |
| sdake | 128gb ram | 18:33 |
| jrich523 | oh i know you do inc0 ;) | 18:33 |
| inc0 | also if I want to keep PC running during night | 18:33 |
| sdake | 1.2gb intel 750 ssd | 18:33 |
| jrich523 | btw hows that treating you? | 18:33 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 18:33 |
| sdake | and 10gig nics | 18:33 |
| inc0 | :D | 18:33 |
| jrich523 | 10gb in the home? | 18:33 |
| jrich523 | oh right, you work for cisco, nvm | 18:33 |
| inc0 | btw ATH are great, thanks for pointers | 18:33 |
| sdake | jrich523 my network for work is netgear actually | 18:33 |
| jrich523 | glad you're enjoying them... im still loving them :) | 18:34 |
| sdake | jrich523 paid out of pocket for my home lab | 18:34 |
| jrich523 | sdake: lol one of the few cisco hasnt bought? :) | 18:34 |
| portdirect | lol - that much ram would be a gamechanger | 18:34 |
| jrich523 | pssh you should make them gear you up | 18:34 |
| sbezverk | portdirect: ping | 18:34 |
| jrich523 | its the least they could do | 18:34 |
| inc0 | cable jack could be more stable (it disconnects with little encouragement) | 18:34 |
| sdake | ya 800 bucks for a 8 port 10gig switch | 18:34 |
| portdirect | sbezverk: pong | 18:34 |
| portdirect | sup? | 18:34 |
| sdake | easy purchase for me | 18:34 |
| inc0 | but on the other hand I'd rather have this than broken jack | 18:34 |
| inc0 | again | 18:34 |
| jrich523 | i took the meraki class just for the freee gear | 18:34 |
| sdake | inc0 did you buy a house or renting? | 18:34 |
| jrich523 | i suspect there must be some internal class you can take to get gear | 18:34 |
| inc0 | rent apartament | 18:34 |
| sdake | jrich523 i dont get get from cisco except in my cisco labs | 18:35 |
| jrich523 | i have a house, and an office, so this pc runs all the time and it could be as loud as a server and i wouldnt care :) | 18:35 |
| sdake | jrich523 i just ask - and if the money gods think its approiate they deliver :) | 18:35 |
| sdake | ya i leave my pc on 24/7 | 18:35 |
| sdake | keeps the computer from wearing out | 18:35 |
| jrich523 | you would think, with the buy back stuff, that they'd give you guys stuff to play with | 18:35 |
| sdake | like a lightbulb - if you leave it on forever, it doesn't usually break | 18:35 |
| jrich523 | lol money gods... they rip people off so bad, you'd think it would be no big deal to toss gear at you guys | 18:36 |
| sbezverk | portdirect: can you do me a favor. When I run iscsi/tgtd ps in my baremetal test bed, it runs fine but at the gate iscsd failes. COuld you please try to run it in your test bed to see? | 18:36 |
| sdake | jrich523 we get small deals on some stuff but not much | 18:36 |
| sdake | jrich523 nah its a complex thing to get gear - has to be racked/etc | 18:36 |
| jrich523 | cisco is one of my most disliked IT companies | 18:36 |
| sdake | jrich523 and it must be in a dc | 18:36 |
| sdake | jrich523 sorry to hear it - I enjoy working at cisco :) | 18:36 |
| jrich523 | thats good to hear | 18:36 |
| sdake | although I haven't totally drank the koolaid :) | 18:36 |
| jrich523 | a lot of the folks i work with are great | 18:36 |
| sbezverk | portdirect: you do not need any special configs | 18:36 |
| portdirect | sbezverk: can you give me link - should be able to if no ceph requirement | 18:37 |
| sdake | i'm mostly an upstream guy | 18:37 |
| inc0 | I need ps4 pro tho | 18:37 |
| sdake | so no koolaid necessary | 18:37 |
| jrich523 | i jsut dont really agree with the direction of the company, or the general "must buy cisco" view of 99% of companies | 18:37 |
| inc0 | just for final fantasy if anything | 18:37 |
| sbezverk | portdirect: no ceph required | 18:37 |
| jrich523 | lol | 18:37 |
| sdake | jrich523 nobody ever went wrong buying cisco :) | 18:37 |
| jrich523 | so they say | 18:37 |
| sdake | yup | 18:37 |
| jrich523 | but i've seen PLENTY of people go wrong lol | 18:37 |
| sdake | lol :) | 18:37 |
| sbezverk | https://review.openstack.org/#/c/414254/ | 18:37 |
| inc0 | depends how much you'd like to spend | 18:38 |
| jrich523 | lol | 18:38 |
| inc0 | if I'd build cloud, no cisco would be involved | 18:38 |
| inc0 | open hardware all the way | 18:38 |
| jrich523 | ok got to run for a bit, i'll bbl for more interesting conversation! :) | 18:38 |
| sdake | sbezverk gate failures | 18:38 |
| inc0 | cyas | 18:38 |
| sdake | sbezverk people don't look at reviews with gate failures, although I'd like to change that :) | 18:38 |
| sdake | later inc0 | 18:38 |
| sdake | enjoy the holidays | 18:38 |
| inc0 | I'm here | 18:38 |
| inc0 | saying bye to jrich523 ;) | 18:39 |
| sdake | inc0 if you could get the video and notes published from wed, that would be helpful (in the etherpad) | 18:39 |
| sbezverk | sdake: not sure the resaon for this comment, it does not either useful or helpful | 18:39 |
| inc0 | dont have it | 18:39 |
| sdake | sbezverk huh? | 18:39 |
| inc0 | I've sent mail to Lachlan yesterday | 18:39 |
| sbezverk | portdirect: you might need to use these two configmaps | 18:39 |
| inc0 | and I'm not sure if anyone took notes | 18:39 |
| sbezverk | http://paste.openstack.org/show/593291/ | 18:40 |
| sdake | inc0 cool - can you put a note in the etherpad waiting on notes :) | 18:40 |
| sdake | inc0 i got impression lachian was going to send out notes | 18:40 |
| sbezverk | portdirect: ^^ | 18:40 |
| sdake | inc0 if he does, don't make a liar out of me and publish them via google docs :) | 18:41 |
| sdake | late rjr | 18:41 |
| inc0 | sdake, I created etherpad myself remember? | 18:41 |
| sdake | jrich523 that is | 18:41 |
| sdake | inc0 roger | 18:41 |
| inc0 | ;) | 18:41 |
| sdake | inc0 i wasn't here when you created etherpad, was getting presents sorted out for all the peeps | 18:41 |
| inc0 | I wanted to send out email you've sent when I get video tbh | 18:42 |
| inc0 | thats why I was waiting with it | 18:42 |
| sdake | inc0 but i did get a link and posted to ml | 18:42 |
| sdake | oh i see | 18:42 |
| sdake | well sorry for jumping the gun there | 18:42 |
| inc0 | no problem | 18:42 |
| *** bmace has quit IRC | 18:42 | |
| sdake | i just don't want us operating in secret :) | 18:42 |
| inc0 | well, no secrets involved | 18:42 |
| sdake | right | 18:42 |
| *** bmace has joined #openstack-kolla | 18:42 | |
| inc0 | I've sent mail to Lachlan yesterday and was waiting for reply before making announcement | 18:43 |
| sdake | people might infer that, and after that flameout on the ml, we need to avoid that in the future ;) | 18:43 |
| inc0 | well, I didn't start flameout;) | 18:43 |
| sdake | yes doug ended it however | 18:43 |
| inc0 | yeah, I think we were agreeing for a while | 18:43 |
| inc0 | but certain people kept it going artificially | 18:44 |
| sdake | welcome to flamewars 10101 :) | 18:44 |
| *** mdnadeem has joined #openstack-kolla | 18:46 | |
| *** tonanhngo has joined #openstack-kolla | 18:47 | |
| portdirect | sbezverk: would it be ok if i ran that test later? having trouble building the services on this pc - and cant ssh into my home :/ | 18:51 |
| *** Bico_Fino has quit IRC | 18:51 | |
| *** tonanhngo has quit IRC | 18:52 | |
| *** sdake has quit IRC | 18:52 | |
| sbezverk | portdirec: sure thing, appreciate your help here as I am puzzeled here.. I suspect it is something at the gate that forces it to break.. | 18:53 |
| portdirect | also kfox1111, sdake, sbezverk, srwilkers, v1k0d3n, Pavo and the rest of the kolla chan, hope you guys have an awesome holiday (and try to take one) | 18:54 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 18:54 |
| kfox1111 | portdirect: you too. :) | 18:54 |
| Pavo | you too portdirect merry christmas and happy new year | 18:54 |
| portdirect | http://www.swaindestinations.com/blog/wp-content/uploads/2013/12/Santa_with_koala_2-638.jpg laters peeps! | 18:55 |
| *** mdnadeem has quit IRC | 18:55 | |
| portdirect | ^^ thats for you inc0 | 18:56 |
| *** portdirect has left #openstack-kolla | 18:56 | |
| *** inc0 has quit IRC | 18:59 | |
| *** sdake has joined #openstack-kolla | 19:00 | |
| *** inc0 has joined #openstack-kolla | 19:00 | |
| *** MasterOfBugs has quit IRC | 19:03 | |
| kfox1111 | sbezverk: sdake: portdirect: https://review.openstack.org/#/c/413849/ is ready | 19:11 |
| sdake | kfox1111 let me finish my lunch - 5 mins - and i'll review | 19:13 |
| kfox1111 | sdake: k. thanks. | 19:13 |
| *** inc0 has quit IRC | 19:17 | |
| sdake | kfox1111 is https://review.openstack.org/#/c/413849/9/helm/all_values.yaml templating tags on your wishlist? | 19:18 |
| srwilkers | looking now kfox1111 | 19:18 |
| srwilkers | sorry, been yo-yoing like crazy the past few days :/ | 19:19 |
| sdake | what is yoyoing :) | 19:19 |
| sdake | is that like hohoing? :) | 19:19 |
| *** coolsvap has quit IRC | 19:19 | |
| srwilkers | just working on too many things at once | 19:19 |
| sdake | srwilkers oh ya - gotta get good at doing multipel things at once - takes a long time to learn that one :) | 19:20 |
| srwilkers | yeah, im learning real quick how terrible i am at it | 19:20 |
| sdake | there is that old idiom do one thing do it well | 19:20 |
| sdake | i was terrible too 5-10 years ago | 19:20 |
| sdake | now the new idiom is jack of all trades, master of many :) | 19:20 |
| kfox1111 | sdake: yeah. | 19:21 |
| sdake | in other words, hybrid employees are highly valued | 19:21 |
| kfox1111 | srwilkers: thanks for looking. | 19:21 |
| *** inc0 has joined #openstack-kolla | 19:21 | |
| kfox1111 | srwilkers: sorry but its going to be a bit noisy ps wise.doing a bunch of cleanup to facilitate the global api stuff. | 19:21 |
| srwilkers | kfox1111, no need to thank me. sorry ive been slacking on reviewing things the past few days | 19:21 |
| srwilkers | kfox1111, no worries | 19:21 |
| kfox1111 | srwilkers: no worries. its the holidays. :) | 19:22 |
| kfox1111 | sbezverk: should we try and force it on always? maybe its ok to let the user decide to force it off? | 19:23 |
| sdake | srwilkers ya no need to apologize for slacking on reviewing | 19:23 |
| kfox1111 | I think maybe we should just default that one package to on? | 19:24 |
| sdake | srwilkers its a volunteer project | 19:24 |
| sdake | its not like we get paid for this stuff :) | 19:24 |
| sdake | kfox1111 i think that particular one sbezverk pointed out is mandatory all the time so should be forced on all the time | 19:24 |
| *** MasterOfBugs has joined #openstack-kolla | 19:24 | |
| srwilkers | dont get paid for it, but i enjoy doing it | 19:24 |
| sbezverk | kfox1111: another thing but it was a miss in openvswitch, that mount in dependecy container is not needed at all | 19:26 |
| *** MasterOfBugs has quit IRC | 19:26 | |
| *** MasterOfBugs has joined #openstack-kolla | 19:26 | |
| sbezverk | kfox1111: it is only needed in main init container | 19:26 |
| kfox1111 | sbezverk: I thought the socket needed to be read by the entrypoint container? | 19:27 |
| kfox1111 | so, what if we switch it back to the way I had it initially. | 19:27 |
| kfox1111 | we have both a specific dependency container for the socket, so it can't be turned off. | 19:28 |
| kfox1111 | then include the macro version. | 19:28 |
| kfox1111 | two seperate init containers. | 19:28 |
| inc0 | how fun, vagrant 1.9.1 doesn't work with openstack provider | 19:28 |
| sbezverk | kfox1111: vswitchd needs two init containers one main init, second dependency on a socket which always needs to be on. | 19:29 |
| kfox1111 | yeah, I guess I'm proposing 3. | 19:30 |
| kfox1111 | main init for config, dependency for the socket, | 19:30 |
| sbezverk | in PS I see you call for common_dependency_container_multi | 19:30 |
| kfox1111 | and a 3rd optional one for users to orchestrate with. | 19:30 |
| sdake | victory: [sdake@minime-03 ~]$ | 19:30 |
| sdake | now to setup my other wireless meshes | 19:31 |
| sdake | bbiaf | 19:31 |
| sbezverk | which is subject to condition | 19:31 |
| sbezverk | if you want third (not sure why it is needed here) if fine by socket init must be there and it is gone now | 19:31 |
| *** Bico_Fino has joined #openstack-kolla | 19:32 | |
| sbezverk | s/by/but/ | 19:32 |
| kfox1111 | sbezverk: so users can override dependencies with extra stuff if they need. | 19:32 |
| *** inc0 has quit IRC | 19:32 | |
| sbezverk | kfox1111: socket is NOT overridable | 19:32 |
| kfox1111 | sbezverk: I don't think your hearing me. | 19:33 |
| kfox1111 | there is a case where users might want to add extra deps. | 19:33 |
| kfox1111 | not removing socket. | 19:33 |
| kfox1111 | let me recode it so that the socket is not optional. | 19:34 |
| kfox1111 | but user can still add optional deps... | 19:34 |
| sbezverk | kfox1111: ok if you want to add third init container go ahead, but just return explicit depenency on socket as it was before.. | 19:34 |
| kfox1111 | thats what I'm doing now. | 19:34 |
| *** unicell1 has joined #openstack-kolla | 19:34 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 19:36 |
| kfox1111 | sbezverk: hows that look? | 19:36 |
| sbezverk | almost. extra volume thing is not needed that was a left over which I forgot to remove.. | 19:38 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 19:38 |
| jrich523 | one of these days, id like to figure out how to connect to the openstack api with powershell..... its tough getting help since no openstack guy really knows windows (dev) | 19:38 |
| v1k0d3n | portdirect you're a cool dude man! happy holidays to you and everyone else on the kolla team :) | 19:39 |
| kfox1111 | sbezverk: which extra volumes mount? | 19:39 |
| srwilkers | oh man, its that v1k0d3n guy | 19:40 |
| kfox1111 | jrich523: they have an equiv of curl and jq? if so, should have all you need. | 19:40 |
| sbezverk | kfox1111: I replied as a comment | 19:40 |
| jrich523 | yeah, it appears to be a simple thing, but i've had issues | 19:40 |
| kfox1111 | sbezverk: thanks. | 19:40 |
| jrich523 | partly because of how it handles the auth token with .net | 19:41 |
| kfox1111 | sbezverk: k. I'll pull that. | 19:41 |
| jrich523 | .net hides it, and i think its screwing me up | 19:41 |
| v1k0d3n | v1k0d3n = troublemaker | 19:41 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 19:41 |
| srwilkers | can confirm | 19:41 |
| kfox1111 | jrich523: ah. bummer. | 19:43 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 19:43 |
| jrich523 | yeah.. its really a stupid problem to have lol | 19:43 |
| jrich523 | once i figure it out tho, i should be able to get a lot of stuff done | 19:43 |
| jrich523 | mapping to powershell will be easy work i mean | 19:43 |
| kfox1111 | doh. typo... | 19:44 |
| jrich523 | im going to try and make a powershell module for it.. since... thats what i do... make powershell modules :) | 19:44 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 19:44 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 19:45 |
| *** srwilkers is now known as srwilkers_brb | 19:45 | |
| sbezverk | kfox1111: I think it looks good now. thanks | 19:46 |
| kfox1111 | cool. thanks for reviewing. | 19:47 |
| *** salv-orlando has joined #openstack-kolla | 19:47 | |
| sbezverk | I need to run for some errands, I guess sdake or swinklers will ack after the gate check.. | 19:48 |
| sbezverk | bbl | 19:48 |
| kfox1111 | ok. ttl | 19:48 |
| *** senk has quit IRC | 19:54 | |
| *** Pavo has quit IRC | 19:54 | |
| *** senk has joined #openstack-kolla | 19:55 | |
| *** senk has quit IRC | 19:55 | |
| *** senk has joined #openstack-kolla | 19:55 | |
| *** senk has quit IRC | 19:56 | |
| *** Bico_Fino has quit IRC | 19:56 | |
| *** senk has joined #openstack-kolla | 19:57 | |
| *** sdake has quit IRC | 19:57 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 19:57 |
| *** senk has quit IRC | 19:57 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: tDONOTMERGE https://review.openstack.org/413786 | 19:58 |
| *** senk has joined #openstack-kolla | 19:58 | |
| *** senk has quit IRC | 19:59 | |
| *** senk has joined #openstack-kolla | 20:00 | |
| *** senk has quit IRC | 20:00 | |
| *** senk has joined #openstack-kolla | 20:00 | |
| *** senk has quit IRC | 20:01 | |
| *** senk has joined #openstack-kolla | 20:01 | |
| *** senk has quit IRC | 20:02 | |
| *** senk has joined #openstack-kolla | 20:03 | |
| *** senk has quit IRC | 20:03 | |
| *** senk has joined #openstack-kolla | 20:04 | |
| *** senk has quit IRC | 20:05 | |
| *** senk has joined #openstack-kolla | 20:06 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: tDONOTMERGE https://review.openstack.org/413786 | 20:06 |
| *** senk has quit IRC | 20:06 | |
| *** senk has joined #openstack-kolla | 20:07 | |
| *** senk has quit IRC | 20:08 | |
| *** ssurana has quit IRC | 20:09 | |
| *** senk has joined #openstack-kolla | 20:09 | |
| *** senk has quit IRC | 20:09 | |
| *** senk has joined #openstack-kolla | 20:10 | |
| *** ssurana has joined #openstack-kolla | 20:10 | |
| *** senk has quit IRC | 20:10 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla-ansible: tDONOTMERGE https://review.openstack.org/413786 | 20:11 |
| *** senk has joined #openstack-kolla | 20:11 | |
| *** senk has quit IRC | 20:12 | |
| *** senk has joined #openstack-kolla | 20:13 | |
| *** senk has quit IRC | 20:13 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize vals for keystone user creation https://review.openstack.org/414643 | 20:13 |
| *** senk has joined #openstack-kolla | 20:14 | |
| *** sdake has joined #openstack-kolla | 20:14 | |
| *** portdirect__ has joined #openstack-kolla | 20:14 | |
| *** senk has quit IRC | 20:15 | |
| *** senk has joined #openstack-kolla | 20:15 | |
| *** senk has quit IRC | 20:15 | |
| *** portdirect has joined #openstack-kolla | 20:15 | |
| *** senk has joined #openstack-kolla | 20:16 | |
| openstackgerrit | Merged openstack/kolla-kubernetes: Add init containers for deps to missing things https://review.openstack.org/413849 | 20:16 |
| *** senk has quit IRC | 20:17 | |
| *** senk has joined #openstack-kolla | 20:18 | |
| *** senk has quit IRC | 20:18 | |
| *** sdake_ has joined #openstack-kolla | 20:18 | |
| *** senk has joined #openstack-kolla | 20:19 | |
| openstackgerrit | Edmund Rhudy proposed openstack/kolla: Enable building without Internet access for most Ubuntu binary containers https://review.openstack.org/414639 | 20:19 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Standardize some values https://review.openstack.org/414637 | 20:19 |
| *** senk has quit IRC | 20:19 | |
| *** senk has joined #openstack-kolla | 20:19 | |
| *** senk has quit IRC | 20:20 | |
| *** sdake has quit IRC | 20:20 | |
| *** senk has joined #openstack-kolla | 20:21 | |
| *** senk has quit IRC | 20:21 | |
| *** senk has joined #openstack-kolla | 20:22 | |
| *** senk has quit IRC | 20:22 | |
| *** senk has joined #openstack-kolla | 20:22 | |
| *** senk has quit IRC | 20:23 | |
| *** senk has joined #openstack-kolla | 20:24 | |
| *** senk has quit IRC | 20:24 | |
| *** senk has joined #openstack-kolla | 20:25 | |
| *** senk has quit IRC | 20:25 | |
| *** senk has joined #openstack-kolla | 20:26 | |
| *** senk has quit IRC | 20:27 | |
| *** senk has joined #openstack-kolla | 20:28 | |
| *** senk has quit IRC | 20:28 | |
| *** senk has joined #openstack-kolla | 20:29 | |
| *** inc0 has joined #openstack-kolla | 20:29 | |
| *** senk has quit IRC | 20:30 | |
| *** senk has joined #openstack-kolla | 20:31 | |
| *** salv-orl_ has joined #openstack-kolla | 20:31 | |
| *** senk has quit IRC | 20:31 | |
| *** senk has joined #openstack-kolla | 20:32 | |
| *** senk has quit IRC | 20:33 | |
| *** sdake_ has quit IRC | 20:33 | |
| *** senk has joined #openstack-kolla | 20:34 | |
| *** senk has quit IRC | 20:34 | |
| *** salv-orlando has quit IRC | 20:34 | |
| *** senk has joined #openstack-kolla | 20:35 | |
| *** senk has quit IRC | 20:36 | |
| *** senk has joined #openstack-kolla | 20:37 | |
| *** senk has quit IRC | 20:37 | |
| *** senk has joined #openstack-kolla | 20:38 | |
| *** senk has quit IRC | 20:39 | |
| *** senk has joined #openstack-kolla | 20:40 | |
| *** senk has quit IRC | 20:40 | |
| *** senk has joined #openstack-kolla | 20:40 | |
| *** senk has quit IRC | 20:41 | |
| *** senk has joined #openstack-kolla | 20:42 | |
| *** senk has quit IRC | 20:42 | |
| *** senk has joined #openstack-kolla | 20:43 | |
| *** senk has quit IRC | 20:43 | |
| *** senk has joined #openstack-kolla | 20:44 | |
| *** senk has quit IRC | 20:45 | |
| *** senk has joined #openstack-kolla | 20:46 | |
| *** senk has quit IRC | 20:46 | |
| *** senk has joined #openstack-kolla | 20:47 | |
| *** senk has quit IRC | 20:48 | |
| *** senk has joined #openstack-kolla | 20:49 | |
| *** senk has quit IRC | 20:49 | |
| *** senk has joined #openstack-kolla | 20:50 | |
| *** senk has quit IRC | 20:51 | |
| *** senk has joined #openstack-kolla | 20:52 | |
| *** senk has quit IRC | 20:52 | |
| *** sdake has joined #openstack-kolla | 20:52 | |
| *** senk has joined #openstack-kolla | 20:53 | |
| *** senk has quit IRC | 20:54 | |
| *** senk has joined #openstack-kolla | 20:55 | |
| *** senk has quit IRC | 20:55 | |
| *** senk has joined #openstack-kolla | 20:56 | |
| *** senk has quit IRC | 20:57 | |
| *** senk has joined #openstack-kolla | 20:57 | |
| *** senk has quit IRC | 20:57 | |
| *** senk has joined #openstack-kolla | 20:58 | |
| *** senk has quit IRC | 20:59 | |
| *** senk has joined #openstack-kolla | 20:59 | |
| *** senk has quit IRC | 21:00 | |
| *** senk has joined #openstack-kolla | 21:01 | |
| *** senk has quit IRC | 21:01 | |
| *** senk has joined #openstack-kolla | 21:02 | |
| *** senk has quit IRC | 21:03 | |
| *** senk has joined #openstack-kolla | 21:04 | |
| *** sdake_ has joined #openstack-kolla | 21:04 | |
| *** senk has quit IRC | 21:04 | |
| *** senk has joined #openstack-kolla | 21:04 | |
| *** senk has quit IRC | 21:05 | |
| *** senk has joined #openstack-kolla | 21:06 | |
| *** sdake has quit IRC | 21:06 | |
| *** senk has quit IRC | 21:06 | |
| *** senk has joined #openstack-kolla | 21:07 | |
| *** senk has quit IRC | 21:08 | |
| *** senk has joined #openstack-kolla | 21:09 | |
| *** senk has quit IRC | 21:09 | |
| *** jemcevoy has quit IRC | 21:10 | |
| *** senk has joined #openstack-kolla | 21:10 | |
| *** senk has quit IRC | 21:11 | |
| *** senk has joined #openstack-kolla | 21:12 | |
| sdake_ | general veers, prepare your men! | 21:12 |
| *** senk has quit IRC | 21:12 | |
| kfox1111 | https://review.openstack.org/#/c/414643/ looks ready. | 21:13 |
| *** senk has joined #openstack-kolla | 21:13 | |
| *** senk has quit IRC | 21:14 | |
| *** senk has joined #openstack-kolla | 21:15 | |
| *** senk has quit IRC | 21:15 | |
| *** senk has joined #openstack-kolla | 21:16 | |
| *** senk has quit IRC | 21:17 | |
| sdake_ | kfox1111 your client is broken | 21:17 |
| *** senk has joined #openstack-kolla | 21:17 | |
| sdake_ | senk | 21:17 |
| sdake_ | can you please stop | 21:17 |
| *** senk has quit IRC | 21:18 | |
| *** senk has joined #openstack-kolla | 21:18 | |
| *** senk has quit IRC | 21:18 | |
| *** senk has joined #openstack-kolla | 21:19 | |
| openstackgerrit | Edmund Rhudy proposed openstack/kolla: Enable building without Internet access for most Ubuntu binary containers https://review.openstack.org/414639 | 21:19 |
| *** senk has quit IRC | 21:20 | |
| *** senk has joined #openstack-kolla | 21:21 | |
| *** senk has quit IRC | 21:21 | |
| *** senk has joined #openstack-kolla | 21:22 | |
| *** senk has quit IRC | 21:22 | |
| mnaser | +b *@5.147.141.151 for a bit [y/n]? | 21:23 |
| mnaser | :< | 21:23 |
| *** senk has joined #openstack-kolla | 21:23 | |
| *** senk has quit IRC | 21:24 | |
| *** jheroux has quit IRC | 21:24 | |
| *** senk has joined #openstack-kolla | 21:24 | |
| *** senk has quit IRC | 21:24 | |
| *** tonanhngo has joined #openstack-kolla | 21:25 | |
| *** senk has joined #openstack-kolla | 21:25 | |
| *** senk has quit IRC | 21:26 | |
| *** senk has joined #openstack-kolla | 21:27 | |
| *** senk has quit IRC | 21:27 | |
| *** senk has joined #openstack-kolla | 21:28 | |
| *** tonanhngo_ has joined #openstack-kolla | 21:29 | |
| *** senk has quit IRC | 21:29 | |
| *** senk has joined #openstack-kolla | 21:30 | |
| *** senk has quit IRC | 21:30 | |
| *** senk has joined #openstack-kolla | 21:31 | |
| *** senk has quit IRC | 21:32 | |
| *** l4yerffeJ_ has joined #openstack-kolla | 21:32 | |
| *** l4yerffeJ_ has quit IRC | 21:32 | |
| *** senk has joined #openstack-kolla | 21:32 | |
| *** tonanhngo has quit IRC | 21:32 | |
| *** senk has quit IRC | 21:33 | |
| *** senk has joined #openstack-kolla | 21:33 | |
| *** l4yerffeJ_ has joined #openstack-kolla | 21:33 | |
| *** senk has quit IRC | 21:33 | |
| *** senk has joined #openstack-kolla | 21:34 | |
| *** Jeffrey4l has quit IRC | 21:35 | |
| *** senk has quit IRC | 21:35 | |
| *** l4yerffeJ has quit IRC | 21:35 | |
| *** senk has joined #openstack-kolla | 21:36 | |
| *** senk has quit IRC | 21:36 | |
| *** senk has joined #openstack-kolla | 21:37 | |
| *** senk has quit IRC | 21:37 | |
| *** senk has joined #openstack-kolla | 21:38 | |
| *** senk has quit IRC | 21:39 | |
| *** sdake_ has quit IRC | 21:39 | |
| *** senk has joined #openstack-kolla | 21:40 | |
| *** senk has quit IRC | 21:40 | |
| *** senk has joined #openstack-kolla | 21:41 | |
| *** senk has quit IRC | 21:42 | |
| *** senk has joined #openstack-kolla | 21:43 | |
| *** senk has quit IRC | 21:43 | |
| *** senk has joined #openstack-kolla | 21:44 | |
| *** senk has quit IRC | 21:45 | |
| *** senk has joined #openstack-kolla | 21:46 | |
| *** senk has quit IRC | 21:46 | |
| *** senk has joined #openstack-kolla | 21:47 | |
| *** senk has quit IRC | 21:48 | |
| *** gomarivera has quit IRC | 21:48 | |
| *** Jeffrey4l has joined #openstack-kolla | 21:48 | |
| *** senk has joined #openstack-kolla | 21:49 | |
| *** senk has quit IRC | 21:49 | |
| *** senk has joined #openstack-kolla | 21:50 | |
| *** senk has quit IRC | 21:51 | |
| jrich523 | mnaser: seems like it would be worth while.. | 21:51 |
| *** senk has joined #openstack-kolla | 21:52 | |
| *** senk has quit IRC | 21:52 | |
| *** sdake has joined #openstack-kolla | 21:52 | |
| mgiles | kfox1111 You around? | 21:52 |
| jrich523 | sdake: is back | 21:53 |
| *** senk has joined #openstack-kolla | 21:53 | |
| jrich523 | sdake: ^ constant join/quit msgs | 21:53 |
| jrich523 | been going on for like the last 30 min | 21:54 |
| *** senk has quit IRC | 21:54 | |
| *** senk has joined #openstack-kolla | 21:55 | |
| *** senk has quit IRC | 21:55 | |
| mgiles | Anybody else here that might know the answer? After a ps by kfox1111 was merged today I get _common_defines.yaml:10: function "hasKey" not defined | 21:56 |
| *** senk has joined #openstack-kolla | 21:56 | |
| mgiles | Anybody know how to resolve this? | 21:56 |
| *** senk has quit IRC | 21:57 | |
| *** senk has joined #openstack-kolla | 21:58 | |
| *** senk has quit IRC | 21:58 | |
| *** senk has joined #openstack-kolla | 21:58 | |
| *** senk has quit IRC | 21:59 | |
| *** senk has joined #openstack-kolla | 22:00 | |
| *** senk has quit IRC | 22:00 | |
| *** senk has joined #openstack-kolla | 22:01 | |
| *** senk has quit IRC | 22:02 | |
| *** senk has joined #openstack-kolla | 22:03 | |
| *** senk has quit IRC | 22:03 | |
| *** senk has joined #openstack-kolla | 22:04 | |
| *** senk has quit IRC | 22:05 | |
| *** senk has joined #openstack-kolla | 22:06 | |
| *** senk has quit IRC | 22:06 | |
| *** senk has joined #openstack-kolla | 22:07 | |
| *** senk has quit IRC | 22:07 | |
| *** senk has joined #openstack-kolla | 22:08 | |
| *** senk has quit IRC | 22:09 | |
| *** senk has joined #openstack-kolla | 22:09 | |
| *** senk has quit IRC | 22:09 | |
| *** senk has joined #openstack-kolla | 22:10 | |
| *** senk has quit IRC | 22:11 | |
| *** senk has joined #openstack-kolla | 22:12 | |
| *** sdake has quit IRC | 22:12 | |
| *** senk has quit IRC | 22:12 | |
| *** senk has joined #openstack-kolla | 22:13 | |
| *** senk has quit IRC | 22:14 | |
| *** senk has joined #openstack-kolla | 22:15 | |
| *** senk has quit IRC | 22:15 | |
| *** senk has joined #openstack-kolla | 22:16 | |
| *** senk has quit IRC | 22:17 | |
| *** senk has joined #openstack-kolla | 22:18 | |
| *** senk has quit IRC | 22:18 | |
| *** senk has joined #openstack-kolla | 22:19 | |
| *** senk has quit IRC | 22:19 | |
| *** senk has joined #openstack-kolla | 22:20 | |
| *** senk has quit IRC | 22:21 | |
| *** senk has joined #openstack-kolla | 22:22 | |
| *** tonanhngo_ has quit IRC | 22:22 | |
| *** senk has quit IRC | 22:22 | |
| *** sdake has joined #openstack-kolla | 22:22 | |
| *** senk has joined #openstack-kolla | 22:23 | |
| *** tonanhngo has joined #openstack-kolla | 22:24 | |
| *** senk has quit IRC | 22:24 | |
| *** senk has joined #openstack-kolla | 22:25 | |
| *** senk has quit IRC | 22:25 | |
| *** senk has joined #openstack-kolla | 22:25 | |
| *** ChanServ sets mode: +o sdake | 22:26 | |
| *** senk has quit IRC | 22:26 | |
| *** sdake sets mode: +b *!senk*@* | 22:26 | |
| *** senk has joined #openstack-kolla | 22:27 | |
| *** sdake sets mode: +b *senk*!*@* | 22:27 | |
| *** senk has quit IRC | 22:28 | |
| sdake | jrich523 enjoy | 22:29 |
| jrich523 | thanks :) | 22:29 |
| *** Serlex has left #openstack-kolla | 22:29 | |
| *** sdake sets mode: -b *!senk*@* | 22:30 | |
| srwilkers_brb | wow | 22:31 |
| srwilkers_brb | thats a lot of channel spam | 22:32 |
| *** srwilkers_brb is now known as srwilkers | 22:32 | |
| sdake | i like to think that is unintentional | 22:34 |
| *** sdake sets mode: -o sdake | 22:34 | |
| srwilkers | i would like to think the smae | 22:34 |
| srwilkers | same | 22:34 |
| *** jtriley has joined #openstack-kolla | 22:39 | |
| sdake | sup mgiles | 22:39 |
| openstackgerrit | Merged openstack/kolla-kubernetes: Standardize vals for keystone user creation https://review.openstack.org/414643 | 22:43 |
| sdake | ok peeps - i'm out for a bit :) | 22:49 |
| sdake | enjoy the holidays :) | 22:49 |
| srwilkers | take care sdake :) | 22:53 |
| sbezverk | mgiles: have you rebased after kfox1111 ps? | 22:53 |
| *** tonanhngo has quit IRC | 22:54 | |
| sbezverk | mgiles: another thing to try is to re-run helm_prebuild_microservice | 22:54 |
| *** Pavo has joined #openstack-kolla | 22:54 | |
| *** ocaner has quit IRC | 22:58 | |
| sdake | pavo - before i depart for christmas - wanted to mention jeffrey4l is aware of your issue and wants to speak to you about filing a bug with reelevant data so he can reproduce | 23:01 |
| sdake | now i'm out ;-) | 23:01 |
| jrich523 | ah still here | 23:01 |
| Pavo | which issue? the horizon issue? | 23:01 |
| *** sdake has quit IRC | 23:01 | |
| jrich523 | lol | 23:01 |
| jrich523 | boom gone | 23:01 |
| *** mannamne has joined #openstack-kolla | 23:11 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 23:11 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 23:12 |
| *** l4yerffeJ_ has quit IRC | 23:12 | |
| *** jtriley has quit IRC | 23:13 | |
| sbezverk | kfox1111: ping | 23:13 |
| *** fragatina has quit IRC | 23:15 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP PS adds iscsid and tgtd microservices https://review.openstack.org/414254 | 23:16 |
| *** ssurana has quit IRC | 23:17 | |
| *** huzhengchuan has joined #openstack-kolla | 23:22 | |
| *** huzhengchuan_ has joined #openstack-kolla | 23:22 | |
| kfox1111 | sbezverk: ping | 23:25 |
| kfox1111 | whats up? | 23:25 |
| mgiles | sbezverk Yep - rebased and reran the prebuild steps. | 23:27 |
| *** tonanhngo has joined #openstack-kolla | 23:27 | |
| mgiles | kfxox1111 since you're online now - any idea why i'd get _common_defines.yaml:10: function "hasKey" not defined after your ps was committed? | 23:27 |
| mgiles | kfox1111 mistyped your name here ^^^ | 23:27 |
| kfox1111 | hmm.... | 23:28 |
| kfox1111 | do you have a ps your working on or is it one of the packages in trunk? | 23:30 |
| kfox1111 | oh. | 23:30 |
| kfox1111 | thats probably because we're requiring helm 2.1.3 now. | 23:30 |
| kfox1111 | hasKey was introducted in that version. | 23:30 |
| mgiles | I think my client is at 2.1.3 - not sure about my server, but I can try upgrading that. | 23:31 |
| mgiles | thanks | 23:31 |
| kfox1111 | yeah. I think the server is the one that processes the macro's. | 23:31 |
| kfox1111 | eastiest way I've found is kubectl delete deployment tiller-deploy --namespace=kube-system; helm init | 23:31 |
| mgiles | I'll give that a try. | 23:32 |
| mgiles | kfox1111 That did it. Thanks | 23:33 |
| kfox1111 | np. :) | 23:33 |
| *** salv-orl_ has quit IRC | 23:37 | |
| *** dave-mccowan has joined #openstack-kolla | 23:38 | |
| *** mgiles has quit IRC | 23:42 | |
| *** inc0 has quit IRC | 23:42 | |
| *** tonanhngo has quit IRC | 23:43 | |
| *** tonanhngo has joined #openstack-kolla | 23:44 | |
| *** dave-mccowan has quit IRC | 23:46 | |
| *** Pavo has quit IRC | 23:48 | |
| *** tonanhngo has quit IRC | 23:49 | |
| *** sbezverk_ has joined #openstack-kolla | 23:49 | |
| *** dave-mccowan has joined #openstack-kolla | 23:49 | |
| *** sbezverk has quit IRC | 23:49 | |
| *** goldyfruit has quit IRC | 23:50 | |
| *** dave-mccowan has quit IRC | 23:51 | |
| *** huzhengchuan_ has quit IRC | 23:54 | |
| *** huzhengchuan has quit IRC | 23:54 | |
| *** jtriley has joined #openstack-kolla | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
