*** fragatina has joined #openstack-kolla | 00:00 | |
*** fragatina has quit IRC | 00:00 | |
*** fragatina has joined #openstack-kolla | 00:01 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 00:04 |
---|---|---|
Pavo | guess I need to write myself a ansible script so I don't have to do parted /dev/sdb -s -- mklabel gpt mkpart KOLLA_CEPH_OSD_BOOTSTRAP 1 -1 everytime I destroy and redeploy | 00:06 |
*** sdake has quit IRC | 00:08 | |
*** tonanhngo has joined #openstack-kolla | 00:11 | |
*** tonanhngo has quit IRC | 00:13 | |
Pavo | woot thanks sdake that worked | 00:16 |
*** yuanying_ has joined #openstack-kolla | 00:19 | |
*** yuanying has quit IRC | 00:19 | |
*** yuanying_ has quit IRC | 00:19 | |
*** yuanying has joined #openstack-kolla | 00:19 | |
*** yuanying_ has joined #openstack-kolla | 00:20 | |
*** yuanying has quit IRC | 00:20 | |
*** yuanying has joined #openstack-kolla | 00:21 | |
*** yuanying_ has quit IRC | 00:21 | |
*** yuanying_ has joined #openstack-kolla | 00:21 | |
*** yuanying has quit IRC | 00:21 | |
*** yuanying_ has quit IRC | 00:21 | |
*** yuanying has joined #openstack-kolla | 00:22 | |
*** yuanying has quit IRC | 00:22 | |
*** yuanying has joined #openstack-kolla | 00:22 | |
*** kong_ has joined #openstack-kolla | 00:26 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 00:28 |
*** Jeffrey4l has quit IRC | 00:29 | |
*** Jeffrey4l has joined #openstack-kolla | 00:30 | |
*** negronjl has joined #openstack-kolla | 00:30 | |
*** yingjun has joined #openstack-kolla | 00:30 | |
*** negronjl has quit IRC | 00:31 | |
*** negronjl has joined #openstack-kolla | 00:37 | |
SamYaple | Pavo: yep. if you are _redeploying_ as in a new ceph cluster, then youll need to wipe the disks | 00:38 |
SamYaple | thats more of a ceph thing and less of a kolla thing | 00:38 |
sbezverk | evening, anybody sent any PS for last few minutes? gates seem to be broke.. | 00:42 |
*** yingjun_ has joined #openstack-kolla | 00:42 | |
*** yingjun has quit IRC | 00:42 | |
*** tonanhngo has joined #openstack-kolla | 00:43 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 00:46 |
*** tonanhngo has quit IRC | 00:46 | |
*** JRobinson__ is now known as JRobinson__afk | 00:53 | |
*** yingjun has joined #openstack-kolla | 00:55 | |
*** yingjun_ has quit IRC | 00:57 | |
*** yingjun has quit IRC | 00:57 | |
*** hfu has joined #openstack-kolla | 01:00 | |
*** sayantan_ has joined #openstack-kolla | 01:01 | |
*** hieulq has joined #openstack-kolla | 01:02 | |
*** sayantan_ has quit IRC | 01:05 | |
*** hfu has quit IRC | 01:06 | |
*** newmember has joined #openstack-kolla | 01:07 | |
*** tonanhngo has joined #openstack-kolla | 01:11 | |
*** tonanhngo has quit IRC | 01:13 | |
*** JRobinson__afk is now known as JRobinson__ | 01:19 | |
*** yingjun has joined #openstack-kolla | 01:22 | |
*** tonanhngo has joined #openstack-kolla | 01:23 | |
*** jascott1 has joined #openstack-kolla | 01:24 | |
*** tonanhngo has quit IRC | 01:24 | |
*** jascott1 has quit IRC | 01:29 | |
*** jascott1 has joined #openstack-kolla | 01:36 | |
Pavo | yeah SamYaple its not a big deal, wonder if I can just add it as a task in the current kolla ansible files though | 01:40 |
Pavo | shouldn't to hard to do | 01:41 |
*** jascott1 has quit IRC | 01:41 | |
*** williamcaban has quit IRC | 01:47 | |
*** elynn has joined #openstack-kolla | 01:48 | |
*** williamcaban has joined #openstack-kolla | 01:48 | |
SamYaple | Pavo: something may have changed, but there was strong opposition to ever implementing a delete/cleanup task in ansible of any kind, much less for data | 01:49 |
SamYaple | best write your own for that | 01:49 |
Pavo | yeah thats true | 01:49 |
*** elynn_ has joined #openstack-kolla | 01:52 | |
*** williamcaban has quit IRC | 01:52 | |
*** bachp has quit IRC | 01:52 | |
*** elynn has quit IRC | 01:54 | |
Pavo | SamYaple think this would work? http://paste.openstack.org/show/591373/ | 01:59 |
*** hfu has joined #openstack-kolla | 02:00 | |
SamYaple | Pavo: yea | 02:02 |
SamYaple | but http://paste.openstack.org/show/591374/ is better | 02:02 |
Pavo | ok cool thanks | 02:02 |
SamYaple | you could even get rid of the controller/compute part | 02:02 |
Pavo | I am just now learning ansible | 02:02 |
SamYaple | if you had a list per group/host | 02:02 |
*** tonanhngo has joined #openstack-kolla | 02:02 | |
SamYaple | so it would be one task thats safe to run everyhwere if you wanted | 02:02 |
Pavo | nice I like the items thing | 02:02 |
Pavo | didn't think about that | 02:03 |
Pavo | yeah haven't learned how to do that inventory thing yet | 02:03 |
*** tonanhngo has quit IRC | 02:04 | |
*** kristian__ has joined #openstack-kolla | 02:09 | |
*** unicell1 has quit IRC | 02:12 | |
*** Pavo has quit IRC | 02:12 | |
*** Pavo has joined #openstack-kolla | 02:13 | |
*** kristian__ has quit IRC | 02:13 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install zaqar ui into horizon image https://review.openstack.org/406606 | 02:18 |
*** unicell has joined #openstack-kolla | 02:18 | |
*** williamcaban has joined #openstack-kolla | 02:18 | |
*** bachp has joined #openstack-kolla | 02:18 | |
*** tonanhngo has joined #openstack-kolla | 02:21 | |
*** tonanhngo has quit IRC | 02:23 | |
openstackgerrit | Merged openstack/kolla: Fix aodh role https://review.openstack.org/406203 | 02:24 |
Pavo | SamYaple wouldn't this be better? http://paste.openstack.org/show/591375/ | 02:25 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Read EXTRA_OPTS from env in kolla-ansible script https://review.openstack.org/406671 | 02:25 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Fix octavia images https://review.openstack.org/404645 | 02:33 |
SamYaple | Pavo: no. the paramater you are modifying isnt the label. its the starting MB of the partition | 02:33 |
SamYaple | Kolla handles the labeling | 02:33 |
*** tonanhngo has joined #openstack-kolla | 02:35 | |
*** tonanhngo has quit IRC | 02:37 | |
*** sdake_ has joined #openstack-kolla | 02:39 | |
sdake_ | evening folks | 02:39 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 02:44 |
*** hfu has quit IRC | 02:45 | |
*** hfu has joined #openstack-kolla | 02:46 | |
*** tonanhngo has joined #openstack-kolla | 02:48 | |
*** hfu has quit IRC | 02:49 | |
*** sayantan_ has joined #openstack-kolla | 02:49 | |
*** tonanhngo has quit IRC | 02:49 | |
*** sayantan_ has quit IRC | 02:50 | |
*** hfu has joined #openstack-kolla | 02:50 | |
*** sayantan_ has joined #openstack-kolla | 02:50 | |
*** tovin07 has joined #openstack-kolla | 02:50 | |
*** sdake_ has quit IRC | 02:51 | |
*** fragatina has quit IRC | 02:52 | |
*** yuanying has quit IRC | 02:53 | |
*** fragatina has joined #openstack-kolla | 02:53 | |
*** severion has quit IRC | 02:58 | |
*** yuanying has joined #openstack-kolla | 03:00 | |
*** tonanhngo has joined #openstack-kolla | 03:08 | |
*** tonanhngo has quit IRC | 03:08 | |
*** duonghq has joined #openstack-kolla | 03:11 | |
*** jtriley has joined #openstack-kolla | 03:12 | |
*** hfu has quit IRC | 03:13 | |
openstackgerrit | sayantani proposed openstack/kolla: Clean up kolla-ansible related files from kolla https://review.openstack.org/406429 | 03:13 |
*** sp_ has joined #openstack-kolla | 03:14 | |
*** hfu has joined #openstack-kolla | 03:16 | |
*** tonanhngo has joined #openstack-kolla | 03:21 | |
*** tonanhngo has quit IRC | 03:22 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 03:25 |
*** rstarmer has quit IRC | 03:26 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: Change HELM version to 2.0.1 https://review.openstack.org/406681 | 03:28 |
*** tonanhngo has joined #openstack-kolla | 03:34 | |
*** vks has joined #openstack-kolla | 03:34 | |
*** tonanhngo has quit IRC | 03:36 | |
*** adrian_otto has joined #openstack-kolla | 03:40 | |
Pavo | SamYaple ok so I just need to change the name of the task to MB instead of label? | 03:40 |
*** hfu has quit IRC | 03:42 | |
*** adrian_otto has quit IRC | 03:43 | |
*** hfu has joined #openstack-kolla | 03:44 | |
*** adrian_otto has joined #openstack-kolla | 03:45 | |
*** yuanying has quit IRC | 03:48 | |
vks | Hi , Kolla 3.0.1 deploy gets stuck/timesout at "TASK [haproxy : Waiting for virtual IP to appear]" | 03:52 |
*** tonanhngo has joined #openstack-kolla | 03:52 | |
vks | Any help would be much appreciated. Thanks | 03:53 |
*** hfu has quit IRC | 03:53 | |
*** tonanhngo has quit IRC | 03:53 | |
vks | fatal: [mtv5-at-os11]: FAILED! => {"changed": false, "elapsed": 301, "failed": true, "msg": "Timeout when waiting for xx.xx.xx.xx:3306"} | 03:54 |
*** elynn_ has quit IRC | 03:55 | |
*** hfu has joined #openstack-kolla | 03:56 | |
*** hfu has quit IRC | 03:58 | |
sbezverk | sdake: ping | 04:00 |
*** jtriley has quit IRC | 04:02 | |
*** vks has quit IRC | 04:03 | |
*** sayantan_ has quit IRC | 04:04 | |
*** fragatin_ has joined #openstack-kolla | 04:04 | |
*** fragatina has quit IRC | 04:05 | |
*** tonanhngo has joined #openstack-kolla | 04:05 | |
*** tonanhngo has quit IRC | 04:07 | |
*** jtriley has joined #openstack-kolla | 04:08 | |
*** vks has joined #openstack-kolla | 04:08 | |
vks | Hi , Kolla 3.0.1 deploy gets stuck/timesout at "TASK [haproxy : Waiting for virtual IP to appear]" | 04:09 |
*** vks_ has joined #openstack-kolla | 04:10 | |
vks | fatal: [mtv5-at-os11]: FAILED! => {"changed": false, "elapsed": 301, "failed": true, "msg": "Timeout when waiting for xx.xx.xx.xx:3306"} | 04:11 |
*** adrian_otto has quit IRC | 04:11 | |
*** Pavo has quit IRC | 04:12 | |
*** Pavo has joined #openstack-kolla | 04:14 | |
duonghq | anybody got error kolla-toolbox cannot create endpoint,... for Glance? | 04:14 |
*** vks_ has quit IRC | 04:15 | |
*** v1k0d3n has joined #openstack-kolla | 04:16 | |
*** tonanhngo has joined #openstack-kolla | 04:18 | |
*** tonanhngo has quit IRC | 04:20 | |
*** mdnadeem has joined #openstack-kolla | 04:26 | |
*** coolsvap has joined #openstack-kolla | 04:30 | |
*** tonanhngo has joined #openstack-kolla | 04:31 | |
duonghq | anybody got error kolla-toolbox cannot create endpoint,... for Glance? | 04:33 |
*** tonanhngo has quit IRC | 04:33 | |
*** tonanhngo has joined #openstack-kolla | 04:47 | |
*** tonanhngo has quit IRC | 04:48 | |
*** elynn has joined #openstack-kolla | 04:48 | |
*** adrian_otto has joined #openstack-kolla | 04:50 | |
*** sdake has joined #openstack-kolla | 04:50 | |
openstackgerrit | Merged openstack/kolla: Install searchlight-ui to horizon https://review.openstack.org/405956 | 04:50 |
*** elynn_ has joined #openstack-kolla | 04:53 | |
*** yuanying has joined #openstack-kolla | 04:54 | |
*** elynn has quit IRC | 04:54 | |
*** tonanhngo has joined #openstack-kolla | 05:00 | |
*** jmccarthy has quit IRC | 05:00 | |
*** pbourke has quit IRC | 05:00 | |
*** tonanhngo has quit IRC | 05:01 | |
*** pbourke has joined #openstack-kolla | 05:01 | |
*** jmccarthy has joined #openstack-kolla | 05:01 | |
*** adrian_otto has quit IRC | 05:02 | |
*** adrian_otto has joined #openstack-kolla | 05:04 | |
*** jtriley has quit IRC | 05:07 | |
*** JRobinson__ has quit IRC | 05:07 | |
*** JRobinson__ has joined #openstack-kolla | 05:08 | |
*** rstarmer has joined #openstack-kolla | 05:08 | |
openstackgerrit | Merged openstack/kolla: add os_gnocchi_apache_log.lua in heka https://review.openstack.org/384599 | 05:11 |
*** krtaylor has quit IRC | 05:11 | |
*** adrian_otto has quit IRC | 05:12 | |
*** tonanhngo has joined #openstack-kolla | 05:13 | |
*** tonanhngo has quit IRC | 05:15 | |
SamYaple | Pavo: that doesnt need to be configurable at all | 05:17 |
Pavo | not understanding, I tried without doing those commands and ceph containers restart over and over | 05:18 |
Pavo | what do you mean that doesn't need to be configured SamYaple? | 05:20 |
*** bjolo_ has joined #openstack-kolla | 05:21 | |
SamYaple | Pavo: you do not need to make what you called "label" variable. its not doing what you think | 05:21 |
*** bjolo has quit IRC | 05:22 | |
Pavo | SamYaple the docs http://docs.openstack.org/developer/kolla/ceph-guide.html say that in order for ceph to work it has to find a partition label that is set by using that command | 05:23 |
Pavo | is this wrong? | 05:23 |
openstackgerrit | zhangshuai proposed openstack/kolla-ansible: Add karbor ansible role https://review.openstack.org/402229 | 05:25 |
SamYaple | Pavo: parted {{ item.disk }} -s -- mklabel gpt mkpart KOLLA_CEPH_OSD_BOOTSTRAP {{ item.label }} -1 | 05:25 |
SamYaple | item.label is wrong | 05:25 |
SamYaple | it should always be 1 | 05:25 |
Pavo | for all disk? | 05:26 |
SamYaple | it should not be configurable | 05:26 |
SamYaple | yea | 05:26 |
Pavo | hmmm | 05:26 |
*** saneax-_-|AFK is now known as saneax | 05:26 | |
SamYaple | thats not doing what you think its doing | 05:26 |
*** sdake has quit IRC | 05:26 | |
*** tonanhngo has joined #openstack-kolla | 05:26 | |
Pavo | then what exactly is the 1 doing? picking the 1st table in the partition? | 05:26 |
Pavo | I thought it was labeling it as how many disk you have in the system | 05:27 |
SamYaple | that parted command is recrating a gpt partition label on {{item.disk}}. with a single partition called KOLLA_CEPH_OSD_BOOTSTRAP at START_MB of 1 and END_MB of -1 | 05:27 |
*** tonanhngo has quit IRC | 05:27 | |
SamYaple | you dont do the labeling | 05:27 |
SamYaple | i wrote the automatical labeling piece so the admin never has to worry about numbers like that | 05:27 |
Pavo | ah ok | 05:29 |
Pavo | well thanks for telling me this, someone might wanna explain that in the docs | 05:29 |
SamYaple | its all explained in the docs | 05:30 |
*** tonanhngo has joined #openstack-kolla | 05:39 | |
*** tonanhngo has quit IRC | 05:40 | |
Pavo | nice it worked, and I used the inventory file from kolla with it | 05:44 |
*** Jeffrey4l has quit IRC | 05:51 | |
*** Jeffrey4l has joined #openstack-kolla | 05:51 | |
*** zhangshuai has quit IRC | 05:52 | |
*** zhangshuai has joined #openstack-kolla | 05:52 | |
*** tonanhngo has joined #openstack-kolla | 05:52 | |
*** tonanhngo has quit IRC | 05:53 | |
*** ram_ has joined #openstack-kolla | 05:57 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 05:59 |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 06:00 |
*** jascott1 has joined #openstack-kolla | 06:02 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install zaqar ui into horizon image https://review.openstack.org/406606 | 06:04 |
*** hfu has joined #openstack-kolla | 06:05 | |
*** tonanhngo has joined #openstack-kolla | 06:06 | |
*** jascott1 has quit IRC | 06:06 | |
*** tonanhngo has quit IRC | 06:07 | |
*** Pavo has quit IRC | 06:12 | |
*** hfu has quit IRC | 06:15 | |
*** adrian_otto has joined #openstack-kolla | 06:17 | |
*** Pavo has joined #openstack-kolla | 06:17 | |
*** tonanhngo has joined #openstack-kolla | 06:19 | |
ram_ | Hi. I deployed Kolla on VM. And I configured kolla to deploy magnum. It deployed the magnum service. How can I look at the status of magnum. | 06:19 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 06:19 |
ram_ | When i ran $magnum service-list I was getting info as mentioned. http://paste.openstack.org/show/591384/ | 06:20 |
*** tonanhngo has quit IRC | 06:20 | |
*** adrian_otto has quit IRC | 06:21 | |
ram_ | Can any one please help me in this. | 06:21 |
openstackgerrit | caoyuan proposed openstack/kolla: Install designate dashboard into horizon image https://review.openstack.org/406726 | 06:22 |
*** Jeffrey4l has quit IRC | 06:24 | |
*** bachp has quit IRC | 06:24 | |
*** tovin07_ has joined #openstack-kolla | 06:25 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install mistral dashboard into horizon image https://review.openstack.org/406726 | 06:25 |
openstackgerrit | caoyuan proposed openstack/kolla-ansible: Enable mistral dashboard when enable_mistral is yes https://review.openstack.org/406728 | 06:27 |
*** hfu has joined #openstack-kolla | 06:31 | |
*** tonanhngo has joined #openstack-kolla | 06:32 | |
*** tonanhngo has quit IRC | 06:34 | |
*** Jeffrey4l has joined #openstack-kolla | 06:36 | |
*** Satya_ has joined #openstack-kolla | 06:36 | |
Satya_ | Hi jaffrey4l | 06:36 |
Satya_ | Hi SamYaple | 06:38 |
*** kristian__ has joined #openstack-kolla | 06:40 | |
*** elynn__ has joined #openstack-kolla | 06:42 | |
duonghq | sup Satya_ | 06:43 |
*** elynn_ has quit IRC | 06:43 | |
Satya_ | Hi duonghq | 06:44 |
Satya_ | how are you | 06:44 |
*** kristian__ has quit IRC | 06:44 | |
*** kristian__ has joined #openstack-kolla | 06:45 | |
*** tonanhngo has joined #openstack-kolla | 06:45 | |
duonghq | Finding my way in k8s and helm :) | 06:45 |
*** tonanhngo has quit IRC | 06:46 | |
*** sgordon has quit IRC | 06:47 | |
*** pomac_ has joined #openstack-kolla | 06:48 | |
*** qbjolof2 has joined #openstack-kolla | 06:48 | |
*** kristian__ has quit IRC | 06:49 | |
*** kong_ has quit IRC | 06:50 | |
*** sgordon has joined #openstack-kolla | 06:52 | |
*** yuanying_ has joined #openstack-kolla | 06:52 | |
*** pomac has quit IRC | 06:53 | |
*** qbjolof1 has quit IRC | 06:53 | |
*** Marx314 has quit IRC | 06:53 | |
*** unicell has quit IRC | 06:54 | |
Satya_ | duonghq are you aware what is the checksum method used by glance? | 06:55 |
*** yuanying has quit IRC | 06:55 | |
*** unicell has joined #openstack-kolla | 06:55 | |
*** unicell has joined #openstack-kolla | 06:55 | |
duonghq | Satya_, if I understand correctly, you mention image checksum? according to http://docs.openstack.org/developer/glance/glanceapi.html -> it's MD5. | 06:56 |
*** tonanhngo has joined #openstack-kolla | 06:58 | |
*** tonanhngo has quit IRC | 06:59 | |
Satya_ | dounghq it seems as per the bug "https://bugs.launchpad.net/glance/+bug/1516031" md5 is removed | 07:00 |
openstack | Launchpad bug 1516031 in OpenStack Security Notes "Use of MD5 in OpenStack Glance image signature (CVE-2015-8234)" [High,Fix released] - Assigned to Robert Clark (robert-clark) | 07:00 |
Satya_ | as i am also facing issue while using md5 checksum | 07:00 |
openstackgerrit | Merged openstack/kolla: Don't process/queue children not matched https://review.openstack.org/406389 | 07:02 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Implement octavia ansible role https://review.openstack.org/404648 | 07:09 |
*** tonanhngo has joined #openstack-kolla | 07:11 | |
*** tonanhngo has quit IRC | 07:13 | |
*** mdnadeem has quit IRC | 07:13 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 07:14 |
*** hfu has quit IRC | 07:18 | |
*** fragatin_ has quit IRC | 07:18 | |
*** hfu has joined #openstack-kolla | 07:22 | |
*** mdnadeem has joined #openstack-kolla | 07:28 | |
*** kristian__ has joined #openstack-kolla | 07:40 | |
ram_ | duonghq: Hi. How are you? Actually, for my Kolla AIO deployment I didn't have taken inventory file. But what was main use of that. When we will use that exactly? | 07:43 |
duonghq | ram_, it's used to deploy multinode | 07:43 |
*** rstarmer has quit IRC | 07:45 | |
*** kristian__ has quit IRC | 07:45 | |
*** JRobinson__ has quit IRC | 07:47 | |
*** rstarmer has joined #openstack-kolla | 07:47 | |
duonghq | Satya_, what do you mean? | 07:47 |
ram_ | duonghq: OK. Thanks. Before deploying kolla, I edited globals.yml , I enabled cinder and enabled_cinder_backend : iscsi (or) enabled_cinder_backend : lvm It was giving iscsid contaner issue : Internal server error : 500. http://paste.openstack.org/show/591385/ | 07:49 |
*** v1k0d3n has quit IRC | 07:59 | |
*** openstackgerrit has quit IRC | 08:03 | |
*** msimonin has joined #openstack-kolla | 08:10 | |
sp_ | Jeffrey4l: hi | 08:12 |
Jeffrey4l | sup sp_ | 08:12 |
*** Pavo has quit IRC | 08:12 | |
sp_ | Jeffrey4l: do you have some idea about the upgrading nova with kolla ?? | 08:13 |
Jeffrey4l | sp_, what kinda of idea? | 08:13 |
Jeffrey4l | did u see any issue? or something? sp_ | 08:13 |
sp_ | Jeffrey4l: I need the spec of nova upgrade | 08:14 |
sp_ | or swift upgrade | 08:14 |
Jeffrey4l | in kolla? there is no specially info about nova upgrade. just stop the container, remove old container then start a new one. | 08:14 |
sp_ | Jeffrey4l: so that i can get some guidance for creating the spec for ks-rolling-upgrade | 08:15 |
*** bachp has joined #openstack-kolla | 08:15 | |
Jeffrey4l | are u mean 0 down time upgrade? | 08:15 |
Jeffrey4l | if so, just remove it and start new one should work. | 08:16 |
Jeffrey4l | if not** | 08:16 |
sp_ | Jeffrey4l: thanks | 08:16 |
sp_ | Jeffrey4l: Be right back | 08:16 |
Jeffrey4l | ok | 08:16 |
*** Pavo has joined #openstack-kolla | 08:18 | |
*** mnasiadka has joined #openstack-kolla | 08:19 | |
*** yuanying_ has quit IRC | 08:19 | |
*** bachp has quit IRC | 08:19 | |
*** zhubingbing has joined #openstack-kolla | 08:24 | |
sp_ | Jeffrey4l: what about if some new operation performed in between, will we guarantee of our DB consistency ?? | 08:26 |
Jeffrey4l | sp_, nova-manage db will upgrade the DB. so only one command should be ran during upgrade. | 08:27 |
Jeffrey4l | and i have no idea if the operators run some thing during `nova-manage db upgrade`. | 08:27 |
*** kristian__ has joined #openstack-kolla | 08:29 | |
*** Satya_ has quit IRC | 08:32 | |
*** kristia__ has joined #openstack-kolla | 08:32 | |
*** kristian__ has quit IRC | 08:33 | |
*** mariso has joined #openstack-kolla | 08:35 | |
*** jistr is now known as jistr|mtgs | 08:36 | |
*** kristia__ has quit IRC | 08:37 | |
mariso | command 'kollakube resource create ..' returns 'hostvars' is undefined. What I'm missing? | 08:38 |
*** gfidente has joined #openstack-kolla | 08:41 | |
*** gfidente has joined #openstack-kolla | 08:41 | |
*** kristian__ has joined #openstack-kolla | 08:43 | |
*** openstackgerrit has joined #openstack-kolla | 08:46 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 08:46 |
*** zhangshuai has quit IRC | 08:46 | |
*** zhangshuai has joined #openstack-kolla | 08:47 | |
*** wwriverrat_ has joined #openstack-kolla | 08:47 | |
*** kristia__ has joined #openstack-kolla | 08:47 | |
*** wwriverrat has quit IRC | 08:48 | |
*** wwriverrat_ is now known as wwriverrat | 08:48 | |
*** slagle has quit IRC | 08:50 | |
*** palexster has quit IRC | 08:50 | |
*** kristian__ has quit IRC | 08:51 | |
*** slagle has joined #openstack-kolla | 08:51 | |
*** palexster has joined #openstack-kolla | 08:55 | |
*** zhangshuai has quit IRC | 08:55 | |
*** zhangshuai has joined #openstack-kolla | 08:55 | |
*** kristia__ has quit IRC | 08:58 | |
*** kristian__ has joined #openstack-kolla | 09:00 | |
*** zhangshuai has quit IRC | 09:05 | |
*** zhangshuai has joined #openstack-kolla | 09:05 | |
*** zhangshuai has quit IRC | 09:06 | |
*** zhangshuai has joined #openstack-kolla | 09:06 | |
*** yuanying has joined #openstack-kolla | 09:08 | |
*** Jeffrey4l has quit IRC | 09:11 | |
*** kristian__ has quit IRC | 09:14 | |
*** jascott1 has joined #openstack-kolla | 09:17 | |
*** athomas has joined #openstack-kolla | 09:19 | |
*** sp_ has quit IRC | 09:22 | |
*** kristian__ has joined #openstack-kolla | 09:27 | |
*** kong_ has joined #openstack-kolla | 09:31 | |
*** sp_ has joined #openstack-kolla | 09:34 | |
ram_ | HI. I did the basic kolla deployment. So if I want to restart a particular service. How ca I do that? | 09:36 |
SamYaple | ram_: `docker restart nova-api` | 09:37 |
SamYaple | there is no restart playbook to restart across the cluster though | 09:37 |
*** fragatina has joined #openstack-kolla | 09:38 | |
SamYaple | something like `ansible -i <inventory> -m command -a 'docker restart nova-api'` would work for that | 09:38 |
*** bachp has joined #openstack-kolla | 09:39 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla: Updated from global requirements https://review.openstack.org/406852 | 09:39 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-ansible: Updated from global requirements https://review.openstack.org/406853 | 09:39 |
ram_ | SamYaple: Hi. I didn't used inventory file. I did all in one Kolla deployment. Then What command I need to use exactly? | 09:41 |
zhubingbing | hi sam | 09:42 |
zhubingbing | we are aware of this problem | 09:42 |
*** flaper87 has quit IRC | 09:43 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 09:47 |
openstackgerrit | jianyi proposed openstack/kolla: Install senlin dashboard into horizon image https://review.openstack.org/405070 | 09:47 |
SamYaple | ram_: if its all in one, then just use `docker restart <service-name>` | 09:49 |
openstackgerrit | jianyi proposed openstack/kolla: Install senlin dashboard into horizon image https://review.openstack.org/405070 | 09:50 |
ram_ | SamYaple: Thanks. I will try that. | 09:50 |
zhubingbing | hi SamYaple, we can add restart role | 09:50 |
*** yingjun has quit IRC | 09:50 | |
*** flaper87 has joined #openstack-kolla | 09:50 | |
*** hfu has quit IRC | 09:51 | |
zhubingbing | kolla-ansible restart service | 09:51 |
SamYaple | zhubingbing: i mean that seems like a good thing. but really, its not a big deal in my opinion. im sure people will use it | 09:51 |
*** kjelly has quit IRC | 09:52 | |
ram_ | SamYaple/zhubingbing: Thanks. Have you ever tested configuring kolla to deploy magnum? | 09:52 |
zhubingbing | agree | 09:52 |
zhubingbing | i have to test for magnum | 09:52 |
openstackgerrit | jianyi proposed openstack/kolla: Install senlin dashboard into horizon image https://review.openstack.org/405070 | 09:53 |
ram_ | SamYaple/zhubingbing: I tried. Magnum service deployed. But when I tried to run $magnum service-list getting issue http://paste.openstack.org/show/591384/ | 09:54 |
zhubingbing | you can yum install python-magnumclient | 09:55 |
zhubingbing | you should install python-magnumclient packages | 09:56 |
ram_ | zhubingbing: Manually I need to do right? ansible playbook will not do that by default? | 09:59 |
*** sp_ has quit IRC | 09:59 | |
*** elynn__ has quit IRC | 10:00 | |
*** sp_ has joined #openstack-kolla | 10:01 | |
*** tonanhngo has joined #openstack-kolla | 10:01 | |
*** tonanhngo has quit IRC | 10:03 | |
*** sp_ has quit IRC | 10:06 | |
*** Pavo has quit IRC | 10:12 | |
*** matrohon has joined #openstack-kolla | 10:16 | |
*** rmart04 has joined #openstack-kolla | 10:16 | |
*** sp_ has joined #openstack-kolla | 10:17 | |
*** Pavo has joined #openstack-kolla | 10:17 | |
openstackgerrit | Sam Yaple proposed openstack/kolla: [WIP] Add monasca build files https://review.openstack.org/406422 | 10:18 |
openstackgerrit | Sam Yaple proposed openstack/kolla: [WIP] Add kafka image https://review.openstack.org/406874 | 10:18 |
*** fragatina has quit IRC | 10:20 | |
openstackgerrit | Sam Yaple proposed openstack/kolla: [WIP] Add kafka image https://review.openstack.org/406874 | 10:21 |
openstackgerrit | Sam Yaple proposed openstack/kolla: [WIP] Add monasca build files https://review.openstack.org/406422 | 10:21 |
*** zhangshuai has quit IRC | 10:23 | |
*** zhangshuai has joined #openstack-kolla | 10:23 | |
openstackgerrit | Vladislav Belogrudov proposed openstack/kolla-ansible: Sometimes merging of passwords.yml gives json https://review.openstack.org/406877 | 10:24 |
openstackgerrit | Sam Yaple proposed openstack/kolla: Add zookeeper container https://review.openstack.org/406878 | 10:25 |
*** jascott1 has quit IRC | 10:28 | |
*** slagle has quit IRC | 10:28 | |
ram_ | Hi. I have a basic kolla All in one deployment. So where we can see the service logs? | 10:29 |
*** slagle has joined #openstack-kolla | 10:29 | |
duonghq | ram_, it's mentioned in document | 10:29 |
duonghq | ram_, http://docs.openstack.org/developer/kolla/quickstart.html#debugging-kolla | 10:30 |
ram_ | duonghq: OH. sry. I will look at that. Thanks. | 10:30 |
*** Serlex has joined #openstack-kolla | 10:31 | |
duonghq | ram_, np | 10:32 |
duonghq | bye guys | 10:33 |
*** duonghq has quit IRC | 10:33 | |
*** ccesario has joined #openstack-kolla | 10:35 | |
*** kristia__ has joined #openstack-kolla | 10:39 | |
*** kristian__ has quit IRC | 10:40 | |
SamYaple | /win/win 22 | 10:41 |
*** DTadrzak has joined #openstack-kolla | 10:41 | |
*** zhangshuai has quit IRC | 10:42 | |
*** zhangshuai has joined #openstack-kolla | 10:43 | |
*** kristia__ has quit IRC | 10:44 | |
*** tovin07 has quit IRC | 10:44 | |
*** tovin07_ has quit IRC | 10:48 | |
*** portdirect__ has joined #openstack-kolla | 11:06 | |
*** portdirect__ has quit IRC | 11:06 | |
*** portdirect__ has joined #openstack-kolla | 11:07 | |
*** sp_ has quit IRC | 11:10 | |
*** yingjun has joined #openstack-kolla | 11:17 | |
*** sp_ has joined #openstack-kolla | 11:23 | |
*** kristian__ has joined #openstack-kolla | 11:26 | |
*** portdirect_ has quit IRC | 11:26 | |
*** sp_ has quit IRC | 11:28 | |
*** vks has quit IRC | 11:32 | |
*** kristian__ has quit IRC | 11:34 | |
*** jrich523 has quit IRC | 11:34 | |
*** mliima has joined #openstack-kolla | 11:35 | |
*** sayantan_ has joined #openstack-kolla | 11:37 | |
*** sayantan_ has quit IRC | 11:41 | |
*** sp_ has joined #openstack-kolla | 11:42 | |
*** Daviey has quit IRC | 11:43 | |
*** Daviey has joined #openstack-kolla | 11:43 | |
mliima | hello guys | 11:49 |
mliima | morning | 11:49 |
*** portdirect___ has joined #openstack-kolla | 11:57 | |
zhubingbing | hello | 11:59 |
zhubingbing | morning | 11:59 |
*** portdirect__ has quit IRC | 12:00 | |
*** prameswar has joined #openstack-kolla | 12:00 | |
*** portdirect__ has joined #openstack-kolla | 12:02 | |
*** hfu has joined #openstack-kolla | 12:02 | |
*** portdirect___ has quit IRC | 12:04 | |
*** prameswar has quit IRC | 12:08 | |
*** kristian__ has joined #openstack-kolla | 12:09 | |
*** kristia__ has joined #openstack-kolla | 12:09 | |
*** yee379 has joined #openstack-kolla | 12:10 | |
*** yee3799 has quit IRC | 12:11 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 12:11 |
openstackgerrit | Merged openstack/kolla-ansible: Add custom policies in service.json https://review.openstack.org/399712 | 12:12 |
openstackgerrit | Merged openstack/kolla-ansible: Enable zaqar ui when zaqar enabled https://review.openstack.org/406594 | 12:12 |
*** Pavo has quit IRC | 12:12 | |
*** kristian__ has quit IRC | 12:13 | |
*** portdirect__ has quit IRC | 12:14 | |
*** portdirect__ has joined #openstack-kolla | 12:14 | |
*** jascott1 has joined #openstack-kolla | 12:15 | |
*** kristia__ has quit IRC | 12:17 | |
*** Pavo has joined #openstack-kolla | 12:17 | |
*** mliima_ has joined #openstack-kolla | 12:18 | |
*** kristian__ has joined #openstack-kolla | 12:18 | |
*** portdirect__ has quit IRC | 12:18 | |
*** jascott1 has quit IRC | 12:19 | |
*** mliima has quit IRC | 12:21 | |
*** rhallisey has joined #openstack-kolla | 12:23 | |
*** kristian__ has quit IRC | 12:23 | |
openstackgerrit | Merged openstack/kolla-ansible: Updated from global requirements https://review.openstack.org/406853 | 12:23 |
*** prameswar has joined #openstack-kolla | 12:32 | |
*** mdnadeem has quit IRC | 12:35 | |
*** sdake has joined #openstack-kolla | 12:41 | |
openstackgerrit | Merged openstack/kolla: Install magnum dashboard into the horizon image https://review.openstack.org/405350 | 12:44 |
sdake | morning peeps | 12:47 |
*** srwilkers_ has quit IRC | 12:51 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Fix the set config error https://review.openstack.org/371669 | 12:53 |
*** prameswar has quit IRC | 12:56 | |
*** v1k0d3n has joined #openstack-kolla | 12:59 | |
*** kristian__ has joined #openstack-kolla | 13:00 | |
*** portdirect has joined #openstack-kolla | 13:00 | |
*** kristia__ has joined #openstack-kolla | 13:00 | |
portdirect | morning :) | 13:00 |
*** msimonin has quit IRC | 13:01 | |
*** kristi___ has joined #openstack-kolla | 13:02 | |
openstackgerrit | zhubingbing proposed openstack/kolla: Install murano dashboard into horizon image https://review.openstack.org/406474 | 13:02 |
*** krist____ has joined #openstack-kolla | 13:03 | |
*** kristia__ has quit IRC | 13:03 | |
*** kristian__ has quit IRC | 13:04 | |
*** duonghq has joined #openstack-kolla | 13:06 | |
*** kristi___ has quit IRC | 13:07 | |
*** msimonin has joined #openstack-kolla | 13:08 | |
*** kong_ has quit IRC | 13:10 | |
*** sp_ has quit IRC | 13:10 | |
*** prameswar has joined #openstack-kolla | 13:11 | |
*** portdirect has quit IRC | 13:13 | |
sbezverk | sdake: morning could you pease ack this one: https://review.openstack.org/406681 | 13:13 |
*** portdirect has joined #openstack-kolla | 13:14 | |
openstackgerrit | caoyuan proposed openstack/kolla-ansible: Enable watcher dashboard when enable_watcher is yes https://review.openstack.org/406608 | 13:14 |
*** portdirect has quit IRC | 13:15 | |
ram_ | duonghq: Hi. How do I login to the service container once after having the basic kolla deployment? | 13:16 |
*** portdirect has joined #openstack-kolla | 13:16 | |
duonghq | ram_, sorry but what do you mean? | 13:16 |
sbezverk | ram_: docker exec -tu root -ti {container} bash | 13:18 |
ram_ | duonghq: Kolla deploy services inside the docker containers right. How can I login to the container which is having cinder service? I want to restart the service. I tried like "kolla-ansible restart cinder-volume" (or) "docker restart "cinder-volume". But these two are not working . | 13:19 |
duonghq | ram_, you should restart the container | 13:19 |
duonghq | sbezverk, around? | 13:23 |
duonghq | about: https://review.openstack.org/#/c/405422/7/helm/all_values.yaml@27 | 13:24 |
duonghq | you mean 1 storage micro-service for all service or one for each service? | 13:25 |
ram_ | sbezverk/duonghq: thanks. But I am confusing . How do I confirm the status is running/stopped | 13:26 |
sbezverk | duonghq: yes, what is up? | 13:27 |
duonghq | ram_, docker status <container-name> | 13:27 |
duonghq | sbezverk, can you clarify "storage micro-service"? | 13:28 |
*** yingjun has quit IRC | 13:28 | |
*** Bico_Fino has joined #openstack-kolla | 13:28 | |
*** yingjun has joined #openstack-kolla | 13:28 | |
sbezverk | ram_: unless I missing hidden meaning in the question: if you run duocker ps | grep {container name} | 13:28 |
sbezverk | duonghq: we need to come up with a flexible solution for kolla-kube storage needs | 13:29 |
*** yingjun_ has joined #openstack-kolla | 13:29 | |
*** yingjun has quit IRC | 13:29 | |
sbezverk | duonghq: I suggested to kfox1111 to have a kind of storage microservice, but it was just the idea | 13:30 |
duonghq | sbezverk, but you mean this microservice is shared for all services or just locally in each service? | 13:30 |
duonghq | I mean the storage | 13:30 |
sbezverk | duonghq: I would like it to be shared, but | 13:31 |
sbezverk | again since it is just an idead I do not know how it will be at the end ;-) | 13:31 |
duonghq | sbezverk, I understood, so I want to catch w/ your idea | 13:31 |
*** sp_ has joined #openstack-kolla | 13:31 | |
sdake | sbezverk moment let me review | 13:32 |
sdake | sbezverk acked | 13:33 |
sbezverk | sdake: thank you | 13:33 |
sdake | just wasting money on perksatwork | 13:33 |
sdake | so - bought washer and dryer - all with my points from my amex :) | 13:33 |
sdake | actually I had to cough up a few hundred bucks in cash | 13:34 |
sdake | but more or less - points ftw ;) | 13:34 |
sbezverk | man, I need to check my points, maybe I have enough for a pen ;-) | 13:34 |
sdake | we just printed out $1100 in gift cards from the amex | 13:34 |
sdake | we drill all our expenses through amex gold preferred | 13:35 |
sdake | so get 2x points | 13:35 |
openstackgerrit | Merged openstack/kolla-kubernetes: Change HELM version to 2.0.1 https://review.openstack.org/406681 | 13:37 |
ram_ | sbezverk/duonghq: Thanks. | 13:39 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: WIP: packing and uploading registry https://review.openstack.org/404435 | 13:39 |
openstackgerrit | zhubingbing proposed openstack/kolla: Install murano dashboard into horizon image https://review.openstack.org/406474 | 13:40 |
sbezverk | sdake: got it, good approach.. | 13:40 |
sdake | cool thing about amex is - can't go into debt | 13:40 |
sdake | have to pay it off in full each month ;) | 13:40 |
*** jheroux has joined #openstack-kolla | 13:40 | |
sdake | or they shut off your card (I assume, I don't know I've never not paid in full) | 13:40 |
*** dave-mccowan has joined #openstack-kolla | 13:40 | |
sdake | the only downside of amex is the FR | 13:40 |
sdake | dave-mccowan morning fine sir | 13:40 |
dave-mccowan | good morning sdake! | 13:41 |
*** lbeliveau has joined #openstack-kolla | 13:41 | |
duonghq | sbezverk, how to you test the Chart? | 13:41 |
sbezverk | duonghq: at the gate | 13:41 |
duonghq | I see some ps don't have test | 13:42 |
sbezverk | duonghq: we modify ceph_workflow and replace old way deployment of a specific component with helm chart | 13:42 |
*** msimonin has quit IRC | 13:42 | |
sbezverk | duonghq: you can check my rabbit, now rabbit svc/job and pod are implemented and tested as helm chart.. | 13:43 |
duonghq | sbezverk, thank you | 13:43 |
duonghq | I'm confusing between helm-packaging and helm-microservices bp purpose | 13:43 |
*** msimonin has joined #openstack-kolla | 13:43 | |
sbezverk | duonghq: I think helm-microservices replaced helm-packaging | 13:44 |
duonghq | iirc this ps: https://review.openstack.org/#/c/396296/ is pushed beforce helm-microservices is written? | 13:44 |
duonghq | sdake, ping, can you give me some ideas? | 13:44 |
*** msimonin has quit IRC | 13:44 | |
duonghq | sbezverk, iirc, sdake said something about 2 layers in layer 3 of Ryan's specs | 13:45 |
duonghq | *2 sub-layers | 13:45 |
*** kristian__ has joined #openstack-kolla | 13:45 | |
sbezverk | duonghq: that was PS to include HELM support, so it was a start of helm in kolla | 13:45 |
sdake | duonghq moment reading scrollback | 13:45 |
*** sp_ has quit IRC | 13:45 | |
duonghq | sbezverk, sure, | 13:45 |
duonghq | so I think it belongs to helm packaging | 13:45 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Fix the set config error https://review.openstack.org/371669 | 13:46 |
duonghq | I'll be back in about 5mins, sorry guys | 13:46 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Use check_mode no instead of always_run https://review.openstack.org/406967 | 13:46 |
*** msimonin has joined #openstack-kolla | 13:47 | |
*** hfu has quit IRC | 13:48 | |
*** krist____ has quit IRC | 13:48 | |
*** lbeliveau_ has joined #openstack-kolla | 13:48 | |
*** lbeliveau has quit IRC | 13:49 | |
*** Bico_Fino has quit IRC | 13:49 | |
*** Jeffrey4l has joined #openstack-kolla | 13:50 | |
* duonghq back | 13:50 | |
*** ayoung has quit IRC | 13:53 | |
*** yingjun has joined #openstack-kolla | 13:53 | |
sdake | duonghq okback | 13:55 |
sdake | duonghq what do you need to know about that patchset you linked? | 13:55 |
sdake | duonghq basically we are making helm-native microservices | 13:55 |
*** Bico_Fino has joined #openstack-kolla | 13:55 | |
*** yingjun_ has quit IRC | 13:56 | |
duonghq | sdake, so what is the helm-packaging for? | 13:56 |
sdake | for packaging and exposing an API | 13:57 |
sdake | at some point we plan to get into making it do orchestration - if it can do that reliably | 13:57 |
sdake | sorry helm-native packaging | 13:57 |
sdake | not helm-native microservices :) | 13:57 |
sdake | we are making allt he microservices helm-native packaged :) | 13:58 |
*** srwilkers_ has joined #openstack-kolla | 13:58 | |
duonghq | so, guess that microservice is building block for helm-packaging? | 13:59 |
openstackgerrit | caoyuan proposed openstack/kolla-ansible: Enable watcher dashboard when enable_watcher is yes https://review.openstack.org/406608 | 14:00 |
sdake | right, and our microservices come from the kolla repo | 14:00 |
duonghq | roger, | 14:00 |
duonghq | thank sdake | 14:00 |
openstackgerrit | junbo proposed openstack/kolla-ansible: Integrate ceph-rgw with keystone https://review.openstack.org/406971 | 14:00 |
*** schwicht has joined #openstack-kolla | 14:01 | |
*** lamt has joined #openstack-kolla | 14:01 | |
*** mgiles has joined #openstack-kolla | 14:01 | |
sdake | so basically duonghq we are building up a stack | 14:03 |
*** yingjun has quit IRC | 14:03 | |
sdake | top layer = helm install openstack | 14:03 |
*** tonanhngo has joined #openstack-kolla | 14:03 | |
sdake | middle layer = helm install nova | 14:03 |
*** yingjun has joined #openstack-kolla | 14:03 | |
sdake | bottom layer = helm install nova-novncproxy, nova-conductror, etc | 14:04 |
*** tonanhngo has quit IRC | 14:04 | |
duonghq | clear :) | 14:04 |
sdake | we are implementing the bottom layer atm | 14:04 |
sdake | that is one of the "lyaers" in ryans arch document | 14:04 |
duonghq | got it | 14:05 |
duonghq | so we should build up from microservice :) | 14:06 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Optimized reconfigure action for glance https://review.openstack.org/406978 | 14:06 |
sdake | duonghq right but we need to build the microservice helm packaging layer first | 14:07 |
sdake | to have something to build up from :) | 14:07 |
srwilkers_ | good morning | 14:08 |
duonghq | morning srwilkers_ | 14:08 |
*** goldyfruit has joined #openstack-kolla | 14:08 | |
sdake | sup srwilkers_ | 14:08 |
*** schwicht has quit IRC | 14:09 | |
mgiles | good morning all | 14:10 |
srwilkers_ | morning mgiles | 14:10 |
mgiles | finally got my env set up end of last week so am ready to get started on the k8s work now | 14:10 |
duonghq | many guys is in morning :) | 14:10 |
mgiles | took way longer than I expected to get past all the proxy hurdles :( | 14:10 |
*** schwicht has joined #openstack-kolla | 14:10 | |
srwilkers_ | yeah, dealing with a proxy was terrible. thankfully though, have the ability to use public wifi in the office now | 14:11 |
mgiles | must be nice. :-) | 14:11 |
srwilkers_ | it was an enabler for sure | 14:11 |
mgiles | I'm a bit behidn on stuff due to all that setup mess. can anybody help point me to the best place at this point to jump in and help out? | 14:12 |
mgiles | I've taken a look at the helm microservice stuff in progress | 14:12 |
mgiles | is there a bp somewhere with pieces still to do there? | 14:12 |
*** Pavo has quit IRC | 14:12 | |
duonghq | mgiles, this one: https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-packaging | 14:13 |
duonghq | and this one: https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-packaging | 14:13 |
mgiles | thanks duonghq | 14:13 |
sbezverk | sdake: you were going to update task list at helm-microservices, right? | 14:13 |
duonghq | mgiles, also the specs/kolla-kubernetes-arch.rst from Ryan, thank to Ryan for amazing work | 14:14 |
*** jtriley has joined #openstack-kolla | 14:14 | |
mgiles | duonghq was that supposed to be two different bp links? | 14:14 |
duonghq | I'm working on Glance | 14:14 |
mgiles | yep - up to date on the arch spec. good stuff | 14:14 |
duonghq | mgiles, moment | 14:14 |
duonghq | mgiles, you can read from: http://eavesdrop.openstack.org/irclogs/%23openstack-kolla/%23openstack-kolla.2016-12-05.log.html#t2016-12-05T13:24:39 | 14:15 |
duonghq | sorry I cannot summarize this conversion | 14:15 |
mgiles | s'ok. hard to do sometimes. I've been trying to keep up on stuff but hard to read everything. :) | 14:16 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Optimized reconfigure action for glance https://review.openstack.org/406978 | 14:16 |
*** jascott1 has joined #openstack-kolla | 14:16 | |
sdake | sbezverk i was | 14:17 |
sdake | sbezverk i had personal business to deal with on thur/fri | 14:17 |
*** Pavo has joined #openstack-kolla | 14:17 | |
sdake | sbezverk only human ;) | 14:17 |
sbezverk | sdake: :-) we can copy already existing task list from helm-packaging to helm-microservices | 14:18 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Optimized reconfigure action for glance https://review.openstack.org/406978 | 14:18 |
sdake | cool - although then it needs to be broken out | 14:18 |
*** jtriley has quit IRC | 14:18 | |
*** tovin07 has joined #openstack-kolla | 14:19 | |
*** jascott1 has quit IRC | 14:21 | |
*** fragatina has joined #openstack-kolla | 14:21 | |
*** kristia__ has joined #openstack-kolla | 14:21 | |
duonghq | sdake, sbezverk it can be copied to helm-microservices and be broken out later? | 14:22 |
*** kristia__ has quit IRC | 14:22 | |
*** kristia__ has joined #openstack-kolla | 14:22 | |
*** kristia__ has quit IRC | 14:22 | |
sdake | duonghq wfm ;) | 14:23 |
sdake | who shall do it | 14:23 |
*** kristian__ has quit IRC | 14:24 | |
duonghq | I'll | 14:24 |
sdake | thanks duonghq | 14:25 |
*** fguillot has joined #openstack-kolla | 14:25 | |
*** kristian__ has joined #openstack-kolla | 14:25 | |
duonghq | sdake, sbezverk I put the work item to the microservices one, who received which service should broken out this if they feel it is needed. | 14:26 |
duonghq | sdake, my pleasure | 14:26 |
*** coolsvap has quit IRC | 14:27 | |
*** krtaylor has joined #openstack-kolla | 14:35 | |
sbezverk | duonghq: thank you. I update mine with the latest progress update.. | 14:36 |
rhallisey | sbezverk, hey | 14:36 |
sbezverk | rhallisey: morning rhallisey! | 14:36 |
rhallisey | sbezverk, morning :) | 14:36 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Add run_gate.sh script for CI https://review.openstack.org/406995 | 14:36 |
rhallisey | I was looking at mariadb recovery workflow | 14:37 |
rhallisey | we'll need to use a PD (persistent disk) | 14:37 |
sdake | rhallisey Jeffrey4l or SamYaple are your go to on that :) | 14:37 |
SamYaple | im not anything | 14:37 |
rhallisey | since there are multiple readers | 14:37 |
SamYaple | so i just had the coolest thing happen | 14:38 |
sdake | sorry he isn't anything :) | 14:38 |
rhallisey | SamYaple, what's that? | 14:38 |
SamYaple | i had a power outage where my 3 node galera cluster went hard down | 14:38 |
SamYaple | and it restarted. through no intervention of my own | 14:38 |
*** schwicht has quit IRC | 14:38 | |
Jeffrey4l | what? | 14:38 |
rhallisey | nice | 14:38 |
SamYaple | the latest mariadb and percona stuff seems to have an automagically recovery thing | 14:38 |
SamYaple | im still trying to track it down | 14:38 |
SamYaple | there was an update 2 days ago | 14:38 |
sdake | is that with clusterer? | 14:39 |
SamYaple | no, mariadb | 14:39 |
SamYaple | clusterer is rabbitmq | 14:39 |
Jeffrey4l | SamYaple, data will loss when you force to use --new-cluster parameter to the node which is not the last stopped node. | 14:39 |
SamYaple | it already survives full power outage | 14:40 |
SamYaple | Jeffrey4l: not true always | 14:40 |
SamYaple | esspecially the way we use it | 14:40 |
SamYaple | as long as you start teh last node _written_ to, almost always teh first node, its perfectly fine | 14:40 |
SamYaple | but like i said, this was no intervention at all | 14:40 |
portdirect | rhallisey: I'm hoping to have ceph PVC in for review this evening (late afternoon your time) - if you need something up and running (short term) before then I can throw you some k8s templates | 14:41 |
rhallisey | portdirect, oh neat :). I'm good though. I was just pointing out a slight change in mariadb | 14:41 |
sbezverk | portdirect: I need pvc/pv for rabbitmq too. we need to decide on common implementation.. | 14:42 |
rhallisey | there will need to be multiple readers of grastate.dat | 14:42 |
SamYaple | what is PVC/PV? | 14:42 |
Jeffrey4l | SamYaple, re start the last node written to, yes, but current mariadb-recovery still have some buggy, which can not cover all case. | 14:42 |
SamYaple | k8s stuff? | 14:42 |
rhallisey | SamYaple, yes | 14:42 |
portdirect | rhallisey: Persistant-Volume (and Claim) | 14:42 |
portdirect | whoops SamYaple ^^ | 14:42 |
SamYaple | Jeffrey4l: no, you dont understand. This is outside of kolla. it recovered on its own (this isnt kolla realted at all) | 14:43 |
SamYaple | Jeffrey4l: so i think its a new feature or something. im playing with it now | 14:43 |
sdake | SamYaple thanks on teh clusterer thing - just waking up:) | 14:43 |
SamYaple | portdirect: im not sure what the claim part is, help? | 14:43 |
portdirect | sbezverk: np - been firefighting some other stuff over the last few days (thankfully not my fires) but set aside all today to work on it | 14:44 |
sbezverk | SamYaple: it the way k8s allows attaching persistant volumes to pods | 14:44 |
portdirect | SamYaple: yeah, it's a but confusing a at first but is pretty awseome | 14:44 |
SamYaple | color me confused | 14:45 |
portdirect | SamYaple: a PV is a unamespaced bit of stoarage | 14:45 |
portdirect | and the PVC is a namespaced claim for that storage | 14:45 |
sbezverk | SamYaple: PV persistent volume describes the actual volume with its parameters, but pvc allows pod select pv based on required parameters | 14:45 |
SamYaple | i think that makes sense | 14:46 |
portdirect | SamYaple: sbezverk got there first :) | 14:46 |
*** lamt has quit IRC | 14:46 | |
Jeffrey4l | did i miss something? are we talking some mariadb new feature? | 14:47 |
rhallisey | SamYaple, is you galera kolla? Curious how it came back | 14:47 |
SamYaple | Jeffrey4l: no. two conversations | 14:47 |
SamYaple | rhallisey: idk. im trying to figure it out | 14:47 |
*** srwilkers_ has quit IRC | 14:47 | |
SamYaple | rhallisey: but they all 100% went down from power out. just the cluster recovered | 14:47 |
portdirect | SamYaple: exaple of PV/PVC useage (from my old work) http://paste.openstack.org/show/591434/ | 14:48 |
*** jtriley has joined #openstack-kolla | 14:50 | |
SamYaple | ok so i understand the volume part, i dont understand the Claim part fully. though i admittedly dont understand k8s as well as i could | 14:50 |
SamYaple | i get the concept though (i think :) ) | 14:50 |
*** Jeffrey4l has quit IRC | 14:51 | |
*** Jeffrey4l has joined #openstack-kolla | 14:51 | |
SamYaple | so i threw together an etherpad of a deployment im building with kolla images and salt. anyone feel free to take a look https://etherpad.openstack.org/p/kolla-salt | 14:52 |
SamYaple | this is based off my my original idea when i _first_ started with openstack in containers | 14:52 |
*** schwicht has joined #openstack-kolla | 14:54 | |
*** kristian__ has quit IRC | 14:56 | |
*** kristian__ has joined #openstack-kolla | 14:57 | |
prameswar | Hi All, can you please review https://review.openstack.org/#/c/406177/ | 15:00 |
*** kristian__ has quit IRC | 15:01 | |
portdirect | SamYaple: nice - i've though about making 'drop-in' containers and systemd units as well - didnt realise thats where you started :) Be great to get rkt as a 1st class citizen as well | 15:04 |
SamYaple | portdirect: yea originally i was doing rkt+salt, but that quickly became docker+salt, and then finally docker+ansible. Then i came to kolla to bring ansible into an established community | 15:05 |
SamYaple | portdirect: but really i want rkt+salt | 15:06 |
sdake | prameswar why must /var/lib/nova/mnt be a shared mount? | 15:06 |
SamYaple | i dont particuallarly like docker or ansible | 15:06 |
*** fragatina has quit IRC | 15:08 | |
*** fragatina has joined #openstack-kolla | 15:08 | |
sdake | prameswar the bifrost review i'd approve now, the qsg not sure on | 15:09 |
sdake | prameswar hence i asked in review to split them up | 15:09 |
prameswar | sdake , please see error in launchpad https://bugs.launchpad.net/kolla/+bug/1644602 . - dockerfile have "/var/lib/nova/mnt/:/var/lib/nova/mnt/:shared" | 15:09 |
openstack | Launchpad bug 1644602 in kolla "Path /var/lib/nova/mnt is mounted on / but it is not a shared mount" [Undecided,Fix released] - Assigned to prameswar (prameswar) | 15:09 |
portdirect | srwilkers: ping | 15:09 |
prameswar | for that we have to make it shared | 15:09 |
*** yingjun has quit IRC | 15:10 | |
*** msimonin has quit IRC | 15:10 | |
*** zhangshuai has quit IRC | 15:10 | |
Jeffrey4l | sdake, is it used for cinder+nfs storage. | 15:10 |
*** zhangshuai has joined #openstack-kolla | 15:11 | |
sdake | Jeffrey4l ok thanks - looks like that bug above reports a crater | 15:11 |
Jeffrey4l | sdake, i remove this when enable_cinder_backend_nfs is no https://review.openstack.org/406168 | 15:11 |
*** mnasiadka has quit IRC | 15:11 | |
Jeffrey4l | sdake, i want to make a big change for ci script. could u review this https://review.openstack.org/406995 and related project-config https://review.openstack.org/406997 | 15:12 |
sdake | Jeffrey4l so no more crater then if that value is no? | 15:12 |
*** ayoung has joined #openstack-kolla | 15:12 | |
mgiles | ls | 15:12 |
mgiles | oops | 15:12 |
sdake | Jeffrey4l zug, let me check my calendar | 15:12 |
Jeffrey4l | sdake, yes. | 15:13 |
sdake | Jeffrey4l yup - give me about 30-45 mins :) | 15:13 |
Jeffrey4l | it just change few lines. more work will be continue after project-config is merged. | 15:13 |
prameswar | sdake , then i have to add two PS for each doc ? | 15:13 |
Jeffrey4l | but this should be backport to all branch. | 15:14 |
sdake | prameswar one ps per doc | 15:14 |
sdake | you said big change :) | 15:14 |
sdake | ok reviewing anyway moment | 15:14 |
sdake | Jeffrey4l no mor etox for gate wfm i guess | 15:14 |
sdake | although the gate typically uses tox for kicking stuff off so there is a venv | 15:14 |
prameswar | but in quickstart i added only one blank newline | 15:15 |
Jeffrey4l | big means 1. it need backport to all branches. 3. lots of following change will be pushed after project-config is changed. | 15:15 |
sdake | Jeffrey4l still reviewing | 15:15 |
sdake | Jeffrey4l can you check out other gating - and make sure the integration tests dont use tox | 15:15 |
sdake | i think its not wise to blaze a trail on ci :) | 15:16 |
Jeffrey4l | sdake, devstack do not use tox. | 15:16 |
sdake | Jeffrey4l if nova/etc are not using tox wfm | 15:16 |
Jeffrey4l | we are still use tox but not call it in gate. | 15:16 |
sdake | perhaps it was us who blazed the tox trail :) | 15:16 |
Jeffrey4l | it is moved to `run_gate.sh` script. | 15:16 |
sdake | i see review :) | 15:16 |
sdake | moment let me check out other one | 15:16 |
*** saneax is now known as saneax-_-|AFK | 15:17 | |
*** zhangshuai has quit IRC | 15:17 | |
Jeffrey4l | what i want is: we can do more after tox is ran. let collect all logs, pack the docker registry which is what inc0 does right now. | 15:17 |
*** zhangshuai has joined #openstack-kolla | 15:18 | |
sdake | Jeffrey4l ok lgtm | 15:18 |
sdake | Jeffrey4l i did -1 the review | 15:18 |
openstackgerrit | Merged openstack/kolla: Updated from global requirements https://review.openstack.org/406852 | 15:18 |
sdake | needs one minor change | 15:18 |
openstackgerrit | Merged openstack/kolla: Install zaqar ui into horizon image https://review.openstack.org/406606 | 15:18 |
openstackgerrit | Merged openstack/kolla: Install designate dashboard into horizon image https://review.openstack.org/405933 | 15:18 |
openstackgerrit | Merged openstack/kolla: Install senlin dashboard into horizon image https://review.openstack.org/405070 | 15:18 |
*** msimonin has joined #openstack-kolla | 15:19 | |
Jeffrey4l | now project-config change depends on that one. because kolla script change will not break the gate, but project-config change will break all kolla's gate for a while. | 15:19 |
sdake | hmm i didn't catch that - moment | 15:20 |
*** kristian__ has joined #openstack-kolla | 15:20 | |
sdake | Jeffrey4l this is what happens when people ask me to review at 8am :) | 15:20 |
Jeffrey4l | ;) thanks. | 15:21 |
openstackgerrit | prameswar proposed openstack/kolla: QSG and bifrost improved https://review.openstack.org/406177 | 15:23 |
*** kristian__ has quit IRC | 15:25 | |
sdake | prameswar apologies or all the extra work - sometimes that is part of the process in getting changes merged quickly :) | 15:27 |
prameswar | no problem i am doing | 15:28 |
openstackgerrit | prameswar proposed openstack/kolla: QSG and bifrost improved https://review.openstack.org/406177 | 15:31 |
duonghq | prameswar, can you update the topic? | 15:33 |
sdake | duonghq topic doesn'tneed to be updated :) | 15:33 |
sdake | prameswar already been thru ringer ;) | 15:33 |
sdake | duonghq or do you mean the commit message ? | 15:34 |
duonghq | no, I mean the topic, it's "(detached" now | 15:34 |
duonghq | I think fixing character case, typo,... it's fine for but all of fix in one ps | 15:35 |
sdake | duonghq i htink a detached topic will merge | 15:36 |
sdake | duonghq but not certain | 15:37 |
openstackgerrit | prameswar proposed openstack/kolla: QSG and bifrost improved https://review.openstack.org/406177 | 15:37 |
duonghq | sdake, sorry, but what do you mean? | 15:37 |
sdake | prameswar the main reason i'd like em split up is so sean can have a look over the bifrost changes | 15:38 |
*** fragatina has quit IRC | 15:38 | |
openstackgerrit | prameswar proposed openstack/kolla: QSG and bifrost improved https://review.openstack.org/406177 | 15:39 |
*** Bico_Fino has quit IRC | 15:40 | |
*** adrian_otto has joined #openstack-kolla | 15:40 | |
openstackgerrit | Christian Berendt proposed openstack/kolla-ansible: Give plays in the playbooks a name https://review.openstack.org/404822 | 15:40 |
*** mgiles_ has joined #openstack-kolla | 15:41 | |
*** fragatina has joined #openstack-kolla | 15:42 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 15:42 |
*** mgiles has quit IRC | 15:42 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 15:43 |
*** Bico_Fino has joined #openstack-kolla | 15:44 | |
prameswar | sdake i added one patch for bifrost . you can review it now . i will add new patch for quick start after some time bcoz i want to add some more info to it. | 15:45 |
sdake | prameswar cool | 15:45 |
openstackgerrit | prameswar proposed openstack/kolla: QSG and bifrost improved https://review.openstack.org/406177 | 15:49 |
v1k0d3n | morning folks | 15:52 |
v1k0d3n | have a question for anyone out there. is anyone creating custom image sources for build? or is anyone using jenkins to create these (<< preferred). | 15:53 |
*** kristian__ has joined #openstack-kolla | 15:53 | |
v1k0d3n | wondering if anyone can share some ideas/thoughts/jenkins scripts for doing such things :) | 15:54 |
*** harlowja_at_home has joined #openstack-kolla | 15:54 | |
duonghq | v1k0d3n, http://docs.openstack.org/developer/kolla/image-building.html#build-openstack-from-source <- this does not fit w/ your need? | 15:55 |
v1k0d3n | i'm asking for a bit more than just that... | 15:55 |
*** Bico_Fino has quit IRC | 15:55 | |
v1k0d3n | sdake: and i were talking about this over the weekend. | 15:56 |
duonghq | you mean you want to integrate the build process into your CI? | 15:56 |
*** Bico_Fino has joined #openstack-kolla | 15:56 | |
v1k0d3n | we want to build a jenkins workflow around this process. | 15:56 |
v1k0d3n | yup | 15:56 |
sdake | we does not include sdake btw :) | 15:56 |
sdake | i've got too much on my plate to tackle that particular problem ;) | 15:57 |
v1k0d3n | should have a pipeline this week for doing this. | 15:57 |
v1k0d3n | want this as part of kolla-kubernetes | 15:57 |
*** kristian__ has quit IRC | 15:57 | |
v1k0d3n | in fact! i'll put in the blueprint now. | 15:57 |
sbezverk | portdirect: ping | 15:58 |
portdirect | sbezverk: tcp? | 15:59 |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP Hel-ming Rabbitmq https://review.openstack.org/405422 | 15:59 |
*** sayantan_ has joined #openstack-kolla | 15:59 | |
*** sayantan_ has joined #openstack-kolla | 15:59 | |
sbezverk | portdirect: icmp ;-) can you check my latest rabbitmq common_pv and common_pvc, we need to decide how to provide helm ceph cluster config info so it would be used for rendering | 16:00 |
*** kristian__ has joined #openstack-kolla | 16:00 | |
*** kristian__ has quit IRC | 16:00 | |
portdirect | portdirect: touche :) 2 mins. | 16:01 |
v1k0d3n | https://blueprints.launchpad.net/kolla-kubernetes/+spec/kolla-kubernetes-ci-for-custom-images | 16:03 |
duonghq | v1k0d3n, does it needs breakdown to some workitems? | 16:04 |
v1k0d3n | duonghq: what do you mean? | 16:05 |
*** duonghq has left #openstack-kolla | 16:05 | |
*** duonghq has joined #openstack-kolla | 16:05 | |
duonghq | v1k0d3n, I think it requires not little work, does it should be broken into some workitems? | 16:06 |
duonghq | for tracking purpose? | 16:06 |
*** ayoung has quit IRC | 16:06 | |
*** ayoung has joined #openstack-kolla | 16:07 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 16:07 |
v1k0d3n | duonghq: actually it will be quite easy. | 16:07 |
v1k0d3n | working through this now. | 16:08 |
v1k0d3n | will come back to it when i make some progress. | 16:08 |
duonghq | v1k0d3n, thank you, interesting bp | 16:08 |
v1k0d3n | but check out stable deployment of jenkins using helm. | 16:08 |
v1k0d3n | it includes all of the kubernetes bits already. | 16:08 |
duonghq | ya | 16:09 |
v1k0d3n | but some of the other stuff i can provide in code really. | 16:10 |
duonghq | sure, | 16:10 |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 16:10 |
v1k0d3n | a jenkins script, a vagrantfile or even better yet reuse kubernetes... | 16:10 |
duonghq | nice | 16:10 |
sdake | v1k0d3n placed into discussion | 16:11 |
*** gfidente has quit IRC | 16:12 | |
*** Pavo has quit IRC | 16:12 | |
*** fragatina has quit IRC | 16:14 | |
sdake | v1k0d3n note I added your workitems you put in the blueprint subject into he work item section | 16:14 |
sdake | v1k0d3n you can thank me later ;) | 16:14 |
v1k0d3n | i can thank you now sdake. thanks! :) | 16:15 |
sdake | np :) | 16:15 |
duonghq | you guys are so polite :P | 16:16 |
*** athomas has quit IRC | 16:16 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 16:16 |
*** zhubingbing has quit IRC | 16:17 | |
*** Pavo has joined #openstack-kolla | 16:17 | |
openstackgerrit | caoyuan proposed openstack/kolla: Install watcher dashboard into horizon image https://review.openstack.org/406607 | 16:19 |
duonghq | bye guys | 16:25 |
*** duonghq has quit IRC | 16:26 | |
sdake | night duonghq | 16:26 |
*** sayantan_ has quit IRC | 16:26 | |
*** rmart04 has quit IRC | 16:29 | |
*** yuanying has quit IRC | 16:29 | |
openstackgerrit | prameswar proposed openstack/kolla: QSG updated for shared volumes https://review.openstack.org/407061 | 16:33 |
kfox1111 | morning. | 16:36 |
sbezverk | kfox1111: good morning | 16:36 |
openstackgerrit | prameswar proposed openstack/kolla: QSG updated for shared volumes https://review.openstack.org/407061 | 16:36 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla: Tacker Docker configuration https://review.openstack.org/396391 | 16:37 |
*** lamt has joined #openstack-kolla | 16:38 | |
*** srwilkers_ has joined #openstack-kolla | 16:38 | |
sbezverk | kfox1111: I got rabbitmq svc/job and pod going, need to decide what to do with pv and pvc, please check the latest version of ps | 16:38 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Horizon custom policies https://review.openstack.org/399707 | 16:38 |
*** harlowja_at_home has quit IRC | 16:42 | |
kfox1111 | sbezverk: yeah, looking through it right now. | 16:42 |
openstackgerrit | Mauricio Lima proposed openstack/kolla: Fix zaqar ui https://review.openstack.org/407065 | 16:42 |
openstackgerrit | Mauricio Lima proposed openstack/kolla: Fix zaqar ui https://review.openstack.org/407065 | 16:43 |
*** sayantan_ has joined #openstack-kolla | 16:43 | |
*** sayantan_ has quit IRC | 16:45 | |
*** sayantan_ has joined #openstack-kolla | 16:45 | |
*** vhosakot has joined #openstack-kolla | 16:46 | |
openstackgerrit | prameswar proposed openstack/kolla: QSG updated for shared volumes https://review.openstack.org/407061 | 16:46 |
*** srwilkers_ has quit IRC | 16:46 | |
*** tovin07 has left #openstack-kolla | 16:46 | |
*** srwilkers_ has joined #openstack-kolla | 16:47 | |
*** jrich523 has joined #openstack-kolla | 16:47 | |
*** kristian__ has joined #openstack-kolla | 16:48 | |
*** kristia__ has joined #openstack-kolla | 16:50 | |
mattmceuen | kolla-ansible question for y'all. I'm following the Quickstart + Multinode Deployment docs and have a registry concern. | 16:50 |
mattmceuen | Quickstart says to "kolla-ansible pull" to verify that images are available. This works if I build my own images, but fails if I don't (The required image does not exist) | 16:52 |
*** adrian_otto has quit IRC | 16:52 | |
mattmceuen | If I create my registry as a proxy for docker hub, then it kolla-ansible pull seems to work as expected, but the Multinode docs don't say to create it this way. | 16:52 |
*** kristi___ has joined #openstack-kolla | 16:53 | |
*** vhosakot has quit IRC | 16:53 | |
*** kristian__ has quit IRC | 16:53 | |
mattmceuen | Should e.g. "-e REGISTRY_PROXY_REMOTEURL=https://registry-1.docker.io" be added to the registry creation script, or am I approaching this wrong? | 16:53 |
*** unicell has quit IRC | 16:54 | |
*** lrensing has joined #openstack-kolla | 16:55 | |
*** kristia__ has quit IRC | 16:56 | |
*** unicell has joined #openstack-kolla | 16:56 | |
*** kristian__ has joined #openstack-kolla | 16:58 | |
*** adrian_otto has joined #openstack-kolla | 16:59 | |
*** kristi___ has quit IRC | 17:01 | |
kfox1111 | heh. helm delete bug.... that sounds serious... | 17:07 |
kfox1111 | (and why I'm worried about relying on it for orchestration too quickly. :) | 17:08 |
kfox1111 | can we brainstorm some term other then pod for non pod objects? | 17:09 |
kfox1111 | I'd rather call them their service type at minimum, at the microservices layer, as then it would be obvious what they are. | 17:10 |
kfox1111 | but if we want to come up with a term for "generic k8s object that launches 1 or more pods" I'd be ok with that. | 17:11 |
sdake | naming - one of the hardest problems in computer science ;) | 17:11 |
kfox1111 | its confusing though to mix "launch, single no scalable non fault tollerent thing called a pod" with the other use case | 17:11 |
kfox1111 | sdake: indeed. | 17:11 |
sdake | perhaps we can have a 10 minute brainstorm session on wednesday if jeffrey4l is able to fit that in the agenda | 17:12 |
sdake | and just merge what we got now | 17:12 |
kfox1111 | hmm... k. | 17:12 |
*** jascott1 has joined #openstack-kolla | 17:12 | |
kfox1111 | I just see pod in the title and gut reastion is, "why would I ever want just a pod for that?" | 17:13 |
*** jascott1 has quit IRC | 17:13 | |
sdake | ya bit of a misnomer | 17:13 |
kfox1111 | and then its not obvious if I kubectl get pods that its really a daemonset or something. | 17:13 |
kfox1111 | we shouldh't overload k8s's own terminology. | 17:13 |
sdake | sounds like perfectly good use of a team meeting time :) | 17:14 |
sdake | in the meantime no sense blocking on naming conventions | 17:14 |
kfox1111 | if I can't make the meeting, will you raise the question for me please? | 17:14 |
sdake | if you put on agenda i'll get er done | 17:15 |
sdake | jeffrey4l wants to operate off the agenda rather then freeform | 17:15 |
sdake | an etherpad with what we are brainstorming would help :) | 17:16 |
*** portdirect is now known as portdirect_meeti | 17:16 | |
kfox1111 | sdake: done | 17:18 |
sdake | cool | 17:18 |
*** matrohon has quit IRC | 17:19 | |
*** jascott1 has joined #openstack-kolla | 17:21 | |
*** msimonin has quit IRC | 17:24 | |
*** adrian_otto1 has joined #openstack-kolla | 17:26 | |
*** portdirect_meeti is now known as portdirect | 17:26 | |
*** adrian_otto has quit IRC | 17:28 | |
portdirect | kfox1111: the official term is 'controller' AFAIK - which makes sense in the context of 'pod controller' (and the orginal constructs were named; replication controller, job controller etc) but sounds a bit weird on its own | 17:30 |
sdake | controller is a peice of software that controls a pod or daemonset or whatever ;) portdirect | 17:30 |
kfox1111 | portdirect: yeah, I think sdake's right. | 17:32 |
kfox1111 | its the piece of software the yaml is loaded into that then launches the pods. | 17:32 |
kfox1111 | I'm not aware of a generic term other then 'object' | 17:32 |
sdake | that isn't what i was thinking but thats possibly a controller as well ;) | 17:32 |
kfox1111 | but 'object' is more inclusive then just things that end up as pods. it includes secrets and configmaps and other stuff too. | 17:33 |
sdake | a vessel for example (operator renamed) is a container with a controller that runs the launching of pods/etc | 17:33 |
kfox1111 | sdake: yeah. | 17:33 |
kfox1111 | a controller is either provided by k8s, or is a 3rd party one, usually loaded into a container itself (operator/vessel/whatever :) | 17:34 |
kfox1111 | if we can't come up with a good term, I'd rather fall back to just calling them after their actual object. | 17:35 |
kfox1111 | neutron-server-deployment, heat-engine-statefulset etc. | 17:36 |
*** fragatina has joined #openstack-kolla | 17:36 | |
*** tummy has quit IRC | 17:37 | |
portdirect | Oh, sorry I was refering to: https://github.com/kubernetes/kubernetes.github.io/blob/master/docs/user-guide/pods/index.md#durability-of-pods-or-lack-thereof | 17:38 |
*** yuanying has joined #openstack-kolla | 17:38 | |
kfox1111 | portdirect: yeah. that is saying you should use a controller to watchdog them. | 17:38 |
kfox1111 | but the input to the controller generically is called what? | 17:39 |
*** tummy has joined #openstack-kolla | 17:39 | |
kfox1111 | "daemonset|statefulset|replicaset|deployment" ? | 17:39 |
kfox1111 | (I'm not aware of any term actually being defined today) | 17:39 |
kfox1111 | well, here's a candidate | 17:40 |
kfox1111 | "The use of collective APIs as the primary user-facing primitive is relatively common among cluster scheduling systems, including Borg, Marathon, Aurora, and Tupperware." | 17:40 |
kfox1111 | "collection" ? | 17:40 |
v1k0d3n | hey guys...i think i'm missing a step here. when pushing to a public registry like quay or docker hub....am i using the correct format here (getting errors trying to push). | 17:41 |
v1k0d3n | kolla-build keystone --base ubuntu --type source --push --registry quay.io/attcomdev/clcp-keystone --namespace attcomdev --tag v4.0.1 | 17:41 |
v1k0d3n | i'm assuming that this is because the individual repo is not created? | 17:42 |
*** fragatina has quit IRC | 17:45 | |
v1k0d3n | (it is btw...still failing) | 17:46 |
sdake | v1k0d3n did you login to the registry with docker login? | 17:46 |
v1k0d3n | sorry...yes i did log into the repo prior to attempting to push | 17:47 |
sdake | hrm that should work although i have never tried out quay - only dockerhub | 17:47 |
sdake | you have to login with the docker server tho, because it puts creds in ~?.docker so that docker-py may read them | 17:48 |
sdake | not directly into quay.io | 17:48 |
sdake | got an error output paste? | 17:48 |
sdake | ~/.docker that is | 17:48 |
*** sayantan_ has quit IRC | 17:49 | |
*** harlowja has joined #openstack-kolla | 17:50 | |
*** Serlex has quit IRC | 17:51 | |
v1k0d3n | well, and in this case...quay uses token...but should be the same | 17:56 |
*** kristia__ has joined #openstack-kolla | 17:58 | |
*** kristia__ has quit IRC | 17:59 | |
*** rstarmer has quit IRC | 18:00 | |
*** bmace has quit IRC | 18:00 | |
*** bmace has joined #openstack-kolla | 18:00 | |
*** jascott1 has quit IRC | 18:01 | |
*** kristian__ has quit IRC | 18:02 | |
openstackgerrit | prameswar proposed openstack/kolla: QSG updated for shared volumes https://review.openstack.org/407061 | 18:03 |
*** DTadrzak has quit IRC | 18:04 | |
*** adrian_otto1 has quit IRC | 18:06 | |
*** adrian_otto has joined #openstack-kolla | 18:07 | |
harlowja | who was mentioning something about jenkins? | 18:09 |
harlowja | i thought i overheard something | 18:10 |
harlowja | (or saw something) | 18:10 |
*** mdnadeem has joined #openstack-kolla | 18:10 | |
*** srwilkers_ is now known as srwilkers_away | 18:11 | |
*** rstarmer has joined #openstack-kolla | 18:11 | |
*** Pavo has quit IRC | 18:12 | |
v1k0d3n | harlowja: hey | 18:13 |
*** newmember has quit IRC | 18:13 | |
v1k0d3n | that was me | 18:13 |
v1k0d3n | yup | 18:13 |
harlowja | jenkins with kolla-k8s or just jenkins with kolla? | 18:13 |
v1k0d3n | i want to work on jenkins image build pipeline. | 18:14 |
harlowja | ok, well u are in luck | 18:14 |
v1k0d3n | well kolla...for image builds. | 18:14 |
harlowja | cause i've got one, lol | 18:14 |
v1k0d3n | SWEET!!! | 18:14 |
v1k0d3n | are you able to share, so i can review it? | 18:14 |
v1k0d3n | this would be a huge help. | 18:14 |
*** fragatina has joined #openstack-kolla | 18:15 | |
harlowja | v1k0d3n possibly :-P | 18:15 |
harlowja | http://imgur.com/a/LzRee is what we have so far | 18:15 |
harlowja | we have that pipeline + testing | 18:15 |
harlowja | all via jenkins-job builder templates | 18:16 |
harlowja | and groovy code to do all of it | 18:16 |
harlowja | and slack integration to a channel that tells the team what is happening | 18:16 |
harlowja | http://imgur.com/a/eKAzn (slack integration) | 18:17 |
*** Pavo has joined #openstack-kolla | 18:17 | |
harlowja | my guess is yes, i think we could open it, i might have to remove some stuff from it but seems doable | 18:17 |
v1k0d3n | harlowja: this would be great! | 18:18 |
v1k0d3n | what i want to do is basically build this into kolla-kubernetes | 18:18 |
v1k0d3n | we're using part of this for a PoC that we're doing...in which we're working back into kolla-kubernetes | 18:18 |
*** jascott1 has joined #openstack-kolla | 18:20 | |
harlowja | why the k8s stuff? | 18:21 |
*** lrensing has quit IRC | 18:23 | |
*** lrensing has joined #openstack-kolla | 18:23 | |
harlowja | why not keep the building seperate from kolla-k8s ? | 18:23 |
harlowja | that'd be my desire | 18:24 |
*** prameswar has quit IRC | 18:35 | |
*** mdnadeem has quit IRC | 18:38 | |
*** mariso has quit IRC | 18:40 | |
*** rstarmer has quit IRC | 18:46 | |
sdake | harlowja interesting idea - you mean keep the ci pipeline work in the kolla repo itself rather then kolla-ansible or kolla-k8s? | 18:48 |
sdake | that way both deliverables can benefit? | 18:49 |
sdake | ci pipeline seems more related to images and building rather then deployment | 18:50 |
harlowja | sdake yes | 18:51 |
*** fragatina has quit IRC | 18:52 | |
*** fragatina has joined #openstack-kolla | 18:52 | |
*** kristian__ has joined #openstack-kolla | 18:53 | |
harlowja | sdake i'll start shifting some stuff into https://github.com/harlowja/kolla-jenkins from our internal repo | 18:53 |
harlowja | and then go from here, ha | 18:53 |
harlowja | may take a little while, have to extract certain things | 18:53 |
sdake | harlowja ya - you know the drill ;) | 18:54 |
*** openstackstatus has joined #openstack-kolla | 18:54 | |
*** ChanServ sets mode: +v openstackstatus | 18:54 | |
harlowja | :-p | 18:55 |
*** jascott1 has quit IRC | 18:55 | |
harlowja | yup yup | 18:55 |
*** vhosakot has joined #openstack-kolla | 18:55 | |
*** awiddersheim has quit IRC | 18:58 | |
*** awiddersheim has joined #openstack-kolla | 18:58 | |
*** Bico_Fino has quit IRC | 19:00 | |
*** Bico_Fino has joined #openstack-kolla | 19:00 | |
*** msimonin has joined #openstack-kolla | 19:06 | |
*** jrich523 has quit IRC | 19:10 | |
*** adrian_otto has quit IRC | 19:14 | |
*** diogogmt has joined #openstack-kolla | 19:15 | |
*** jrich523 has joined #openstack-kolla | 19:16 | |
*** tonanhngo has joined #openstack-kolla | 19:17 | |
*** ayoung is now known as ayoung-afk | 19:17 | |
*** williamc_ has joined #openstack-kolla | 19:18 | |
sdake | v1k0d3n would you be ok with moving tht discusison blueprint from kolla-kubernetes to kolla? | 19:18 |
sdake | v1k0d3n since it sounds like you and harlowja may be collaborating there | 19:19 |
harlowja | yup | 19:19 |
harlowja | keep it seperate imho from k8s | 19:19 |
*** adrian_otto has joined #openstack-kolla | 19:19 | |
harlowja | where is that blueprint sdake ? | 19:20 |
harlowja | lol | 19:20 |
sdake | moment | 19:20 |
harlowja | faser | 19:20 |
harlowja | fasterrrrr | 19:20 |
harlowja | lol | 19:20 |
*** tonanhngo_ has joined #openstack-kolla | 19:20 | |
*** sayantani01__ has joined #openstack-kolla | 19:20 | |
sdake | https://blueprints.launchpad.net/kolla-kubernetes/+spec/kolla-kubernetes-ci-for-custom-images | 19:20 |
*** NachoDuck_ has joined #openstack-kolla | 19:20 | |
sdake | i am faster ;) | 19:21 |
harlowja | ya, i don't see a need to connect it into k8s at this point | 19:21 |
sdake | harlowja before we move it lets make sure v1k0d3n is in agreement | 19:21 |
*** tonanhngo has quit IRC | 19:21 | |
harlowja | its just image building via jenkins (which would be triggered by something to build more images) | 19:22 |
sdake | harlowja we use the whiteboard for discussion - i think like most projects in openstack | 19:22 |
harlowja | then u do what u want with those things | 19:22 |
harlowja | eck, whiteboard for discussion | 19:22 |
*** madgin has joined #openstack-kolla | 19:22 | |
harlowja | more like text box | 19:22 |
sdake | ya its the best we got ;) | 19:22 |
v1k0d3n | yup fine with moving it. | 19:22 |
harlowja | i always preferred etherpad :-P | 19:22 |
sdake | harlowja link an ehterpad then ;) | 19:22 |
sdake | we do that too | 19:22 |
dmsimard | The kolla-ansible bootstrap-servers is pretty cool. I only have one itch with it, this bit: https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/baremetal/tasks/install.yml#L67-L69 | 19:22 |
harlowja | :) | 19:22 |
v1k0d3n | just realize that we'd need jenkins to be a part of the containerized deployment. | 19:22 |
sdake | harlowja i am a big fan of linking etherpads in blueprints :) | 19:23 |
dmsimard | 1) It doesn't use a constraint on that version of docker-py ? 2) What if I want to use a binary/packaged version of docker-py instead ? | 19:23 |
v1k0d3n | so the image that helm uses has an image that would be helpful for kolla and kolla-kubernetes (pipeline, docker, kubernetes modules pre-baked in). | 19:23 |
sdake | dmsimard scratch it ;) | 19:23 |
sdake | dmsimard alternately file a bug and maybe somenoe else will :) | 19:24 |
dmsimard | sdake: I can scratch it, but I don't know what you guys want | 19:24 |
*** adrian_otto has quit IRC | 19:24 | |
sdake | dmsimard i think what you just mentioned sounds good to me and would be agreeable to kolla core reviewer team | 19:24 |
sdake | we definately want an upper constraint | 19:24 |
sdake | or I do atleast - sounds good to me :) | 19:25 |
sdake | from binary is good too | 19:25 |
*** tonanhngo_ has quit IRC | 19:25 | |
dmsimard | sdake: like, docker-py is in RDO and probably in Ubuntu/UCA but I'm not sure if the version is sufficient -- and it probably wouldn't work because this bootstrap phase predates the addition of RDO or UCA repositories (and would anyway be incompatible with source installs) | 19:25 |
sdake | docker-py that ships with distros should be a thing we should use | 19:25 |
dmsimard | the version of docker-py in base CentOS is probably ancient if it even exists | 19:25 |
* dmsimard looks | 19:25 | |
sdake | our requirements are in the globals | 19:25 |
*** pprokop_ has joined #openstack-kolla | 19:25 | |
sdake | i am pretty sure everything from 1.6.0 onwards works for binary | 19:26 |
dmsimard | looks like 1.7.2 is available from CentOS Extras | 19:26 |
sdake | that should work I think :) | 19:26 |
dmsimard | https://pkgs.org/search/python-docker-py .. not sure what would be the package name on Ubuntu-like | 19:26 |
sdake | soemone else changed global requirements to 1.8.0 min iirc | 19:27 |
sdake | i think magnum | 19:27 |
sdake | for kolla min requirement is 1.6.0 pretty sure - moment i'll tell you what i test on | 19:27 |
*** schwicht has quit IRC | 19:28 | |
*** rhallisey has quit IRC | 19:28 | |
*** jmccarthy has quit IRC | 19:28 | |
*** williamcaban has quit IRC | 19:28 | |
*** mattmceuen has quit IRC | 19:28 | |
*** sayantani01_ has quit IRC | 19:28 | |
*** NachoDuck has quit IRC | 19:28 | |
*** pprokop has quit IRC | 19:28 | |
*** jroll has quit IRC | 19:28 | |
dmsimard | looks like python-docker in Ubuntu | 19:28 |
*** sayantani01__ is now known as sayantani01_ | 19:28 | |
dmsimard | 1.8.0 in xenial | 19:29 |
dmsimard | hm, kolla actually doesn't want none of that 1.8.0 https://github.com/openstack/kolla/blob/master/requirements.txt#L5 | 19:30 |
dmsimard | https://github.com/openstack/kolla/commit/31af52fb77c7f74c0d651d09859be1f46ad5cd97 | 19:30 |
*** sdake_ has joined #openstack-kolla | 19:31 | |
*** tonanhngo has joined #openstack-kolla | 19:31 | |
sdake_ | 12:28:48] <sdake>dmsimard [sdake@minime-03 ~]$ pip show docker-py | 19:31 |
sdake_ | [12:28:48] <sdake>--- | 19:31 |
sdake_ | [12:28:48] <sdake>Metadata-Version: 1.1 | 19:31 |
sdake_ | [12:28:48] <sdake>Name: docker-py | 19:31 |
sdake_ | [12:28:48] <sdake>Version: 1.7.2 | 19:31 |
sdake_ | [12:29:08] <sdake>dmsimard no ide afor ubuntu - i run centos | 19:31 |
sdake_ | [12:29:20] <sdake>dmsimard if you submit a review i'm sure someone will tell you that runs ubuntu | 19:31 |
sdake_ | [12:29:33] <sdake>inc0 - our ptl - is moving atm | 19:31 |
sdake_ | [12:29:40] <sdake>he runs ubuntu | 19:31 |
sdake_ | [12:29:46] <sdake>also SamYaple runs ubuntu | 19:31 |
sdake_ | [12:29:56] <sdake>and there are a few more who could give you that information | 19:31 |
sdake_ | sorry got dced because of vpn madness | 19:31 |
*** NachoDuck_ is now known as NachoDuck | 19:31 | |
*** sdake has quit IRC | 19:31 | |
v1k0d3n | ok...so have questions on building images. | 19:32 |
dmsimard | sdake_: Ubuntu 16.04 ships 1.8.0 from base OS but Kolla reqs.txt states it does not want 1.8.0: https://github.com/openstack/kolla/blob/master/requirements.txt#L5 | 19:32 |
v1k0d3n | sorry this is my next focus...hence all the recently interest. | 19:32 |
v1k0d3n | if i want to build and push only keystone.... | 19:33 |
v1k0d3n | using the example docs, i think i just state keystone. but i'm building a few others along with, right? | 19:33 |
*** jmccarthy has joined #openstack-kolla | 19:33 | |
sdake_ | dmsimard nobody from kolla mde that global requirements change | 19:33 |
sdake_ | dmsimard so 1.8.0 is safe for us | 19:33 |
v1k0d3n | is there any way to only push ONE container to a registry? | 19:33 |
sdake_ | dmsimard let me check the global requirements blame list and see why tht change was made - moment | 19:33 |
dmsimard | sdake_: https://review.openstack.org/#/c/400770/6 | 19:33 |
dmsimard | +2'd by Jeffrey | 19:34 |
*** srwilkers_away is now known as srwilkers_ | 19:34 | |
sdake_ | ya we rubber stamp globla requirements changes | 19:34 |
dmsimard | and Paul :p | 19:34 |
sdake_ | that comes from a bot | 19:34 |
sdake_ | from global-requirements.txt | 19:34 |
sdake_ | in the requirements repo | 19:34 |
*** fragatina has quit IRC | 19:35 | |
*** adrian_otto has joined #openstack-kolla | 19:35 | |
sdake_ | https://github.com/openstack/requirements/blame/master/global-requirements.txt#L323 | 19:35 |
*** rhallisey has joined #openstack-kolla | 19:35 | |
*** jrich523 has quit IRC | 19:35 | |
sdake_ | hongbin made that change | 19:35 |
sdake_ | its a magnum specific change | 19:35 |
*** tonanhngo has quit IRC | 19:36 | |
*** msimonin1 has joined #openstack-kolla | 19:36 | |
*** jroll has joined #openstack-kolla | 19:36 | |
sdake_ | https://github.com/openstack/requirements/commit/4fc29359129efb66691cc8749c64cd355caf4258 | 19:36 |
sdake_ | dmsimard looks like something busted with docker-py 1.8.0 for magnum | 19:37 |
sdake_ | works for kolla tho ;) | 19:37 |
*** jrich523 has joined #openstack-kolla | 19:37 | |
sdake_ | sorry v1k0d3n harbie where were we :) | 19:38 |
sdake_ | harlowja rather | 19:38 |
harlowja | docker-py busted again | 19:38 |
harlowja | damn | 19:38 |
sdake_ | this is why we wrote our own docker module for ansible ;-) | 19:39 |
sdake_ | to insulate us from docker upstream speed daemon problems ;) | 19:39 |
*** msimonin has quit IRC | 19:39 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helm-ize memcached https://review.openstack.org/405677 | 19:40 |
*** msimonin1 has quit IRC | 19:41 | |
*** jascott1 has joined #openstack-kolla | 19:41 | |
v1k0d3n | so is this related to docker-py? | 19:42 |
sdake_ | v1k0d3n i thought we wer etalking about the ci pipeline work | 19:42 |
sdake_ | docker-py is something dmsimard needed | 19:42 |
*** fragatina has joined #openstack-kolla | 19:42 | |
v1k0d3n | yeah, sorry...just getting confused here. | 19:42 |
sdake_ | rather that discussion was for dmsimard | 19:42 |
v1k0d3n | so my command is: kolla-build keystone --base ubuntu --type source --push --registry quay.io/attcomdev --namespace attcomdev --tag v4.0.1 | 19:43 |
sdake_ | v1k0d3n ya maintaining context is hard - i've got 5 different chat clients and email rolling | 19:43 |
sdake_ | its super annoying | 19:43 |
v1k0d3n | and i'm trying to push to quay using a token (configured in ~/.docker/config.json. | 19:43 |
sdake_ | v1k0d3n can you ls .docker in your home dir | 19:43 |
v1k0d3n | what am i missing? | 19:43 |
v1k0d3n | sdake_: my config.json. | 19:44 |
sdake_ | cool | 19:44 |
sdake_ | can you try to push to the docker hub? | 19:44 |
v1k0d3n | which is a cut paste for bot | 19:44 |
v1k0d3n | lol | 19:44 |
sdake_ | i know you dont wnt to push to docker hub | 19:44 |
sdake_ | i just want to see if that part works | 19:45 |
v1k0d3n | i....suppose i could push to docker hub...although that's not what i use. | 19:45 |
sdake_ | yup - nobody has really tried pushing to quay although clearly it has been done ;) | 19:45 |
v1k0d3n | i would need to create a lot for that test...as in several repos etc. | 19:45 |
v1k0d3n | oh goodness. | 19:45 |
v1k0d3n | :-/ | 19:46 |
sdake_ | just build keystone independently and push | 19:46 |
v1k0d3n | ok | 19:46 |
sdake_ | the command you used above | 19:46 |
v1k0d3n | back to my question | 19:46 |
sdake_ | dont have to put the whole thing there | 19:46 |
v1k0d3n | keystone doesn't just push keystone | 19:46 |
v1k0d3n | it's pushing several | 19:46 |
sdake_ | oh right | 19:46 |
sdake_ | security thing? | 19:46 |
v1k0d3n | can i limit? | 19:46 |
v1k0d3n | that would help my test a lot... | 19:46 |
sdake_ | hrm, you can push manually | 19:47 |
sdake_ | to limit what is pushed | 19:47 |
sdake_ | there is no selection on what is pushed in kolla-build | 19:47 |
sdake_ | it pushes all deps in the chain | 19:47 |
sdake_ | would make a good feature to only push the leafs | 19:47 |
v1k0d3n | no, when i try to push keystone (command above) it pushes a bunch | 19:48 |
sdake_ | keystone, base, and opentack-base correct? | 19:48 |
sdake_ | watched a vid last night | 19:49 |
sdake_ | the guy said all the stuff he got in life for freee was most important to him | 19:49 |
sdake_ | and all the stuff he paid for was replaceable | 19:49 |
*** kristian__ has quit IRC | 19:49 | |
sdake_ | moving locations bb-5 | 19:49 |
*** adrian_otto has quit IRC | 19:50 | |
*** portdirect has quit IRC | 19:51 | |
*** portdirect has joined #openstack-kolla | 19:51 | |
*** sdake has joined #openstack-kolla | 19:53 | |
*** adrian_otto has joined #openstack-kolla | 19:54 | |
* portdirect getting spammed bigtime :/ brb | 19:55 | |
v1k0d3n | so sdake_ the formatting of the command above is correct? | 19:55 |
*** portdirect has quit IRC | 19:55 | |
openstackgerrit | David Moreau Simard proposed openstack/kolla-ansible: Provide a way to install docker-py from packaging in bootstrap https://review.openstack.org/407165 | 19:55 |
dmsimard | sdake: ^ | 19:56 |
*** sdake_ has quit IRC | 19:56 | |
*** vilobhmm11 has joined #openstack-kolla | 19:58 | |
sdake | v1k0d3n think so v1k0d3n | 19:58 |
sdake | dmsimard missed your message | 19:58 |
sdake | oh your review | 19:59 |
sdake | ;) | 19:59 |
dmsimard | sdake: yeah. | 19:59 |
sdake | lucnh has arrived | 20:00 |
sdake | bbiaf | 20:00 |
*** goldyfruit has quit IRC | 20:00 | |
*** tonanhngo has joined #openstack-kolla | 20:01 | |
*** tonanhngo has quit IRC | 20:01 | |
vilobhmm11 | sdake : was trying to do a AIO deployment of kolla using http://docs.openstack.org/developer/kolla/quickstart.html...the service come up; able to run nova, keystone, neutron, glance cli, able to boot instances (vms) but not able to ping them…looks like there is something wrong with the network interface settings..can you please help me ? | 20:01 |
*** v1k0d3n has quit IRC | 20:02 | |
*** tonanhngo has joined #openstack-kolla | 20:02 | |
*** fragatina has quit IRC | 20:03 | |
*** sayantan_ has joined #openstack-kolla | 20:04 | |
*** fragatina has joined #openstack-kolla | 20:05 | |
sayantan_ | Hi everone | 20:08 |
sayantan_ | everyone* | 20:08 |
sayantan_ | If I want to run the tests Jenkins runs locally before uploading a patch, how do I do that? | 20:09 |
*** portdirect has joined #openstack-kolla | 20:10 | |
*** portdirect is now known as portdirect__ | 20:10 | |
*** Pavo has quit IRC | 20:12 | |
*** Bico_Fino has quit IRC | 20:13 | |
vilobhmm11 | dmsimard : ^^ | 20:16 |
dmsimard | vilobhmm11: huh ? I have no clue - why ask me? :) | 20:17 |
*** goldyfruit has joined #openstack-kolla | 20:17 | |
vilobhmm11 | dmsimard : was trying to do a AIO deployment of kolla using http://docs.openstack.org/developer/kolla/quickstart.html...the service come up; able to run nova, keystone, neutron, glance cli, able to boot instances (vms) but not able to ping them…looks like there is something wrong with the network interface settings..can you please help me ? | 20:17 |
vilobhmm11 | 12:03 | 20:17 |
vilobhmm11 | v1k0d3n left the room (quit: Ping timeout: 260 secon | 20:17 |
*** Pavo has joined #openstack-kolla | 20:17 | |
vilobhmm11 | dmsimard : thought you might have an asnwer for this one | 20:17 |
dmsimard | why would I know ? I'm not very familiar with Kolla at all :P | 20:17 |
vilobhmm11 | :^^ | 20:18 |
vilobhmm11 | oh ok | 20:18 |
sdake | hey folks back from lunch | 20:18 |
vilobhmm11 | sdake : gr8 | 20:18 |
vilobhmm11 | sdake : was trying to do a AIO deployment of kolla using http://docs.openstack.org/developer/kolla/quickstart.html...the service come up; able to run nova, keystone, neutron, glance cli, able to boot instances (vms) but not able to ping them…looks like there is something wrong with the network interface settings..can you please help me ? | 20:18 |
sdake | vilobhmm11 i don't know much about networking - I do know that kolla neutron works fantastically | 20:18 |
sdake | vilobhmm11 unfortunately I can't guide you in network setup - setting up a network properly is difficult | 20:18 |
vilobhmm11 | is there a set of steps i can go with which will help me to do a AIO deployment of kolla | 20:19 |
vilobhmm11 | sdake : ^^ | 20:19 |
sdake | vilobhmm11 the key here is your neutron isn't cooperating with your network environment | 20:19 |
sdake | so I'd suggest sorting that part out | 20:19 |
sdake | vilobhmm11 i use init-runonce - it works for me - is tailored to my specific environment | 20:19 |
*** kong_ has joined #openstack-kolla | 20:19 | |
sdake | pavo may be able to help - he has been through the ringer on networking | 20:19 |
portdirect__ | vilobhmm11: Can you get VNC access to your booted instance? | 20:20 |
sdake | pavo finally got his networking rolling | 20:20 |
vilobhmm11 | sdake : i see :) | 20:20 |
vilobhmm11 | Since i am deploying kolla for the first time was expecting something of this | 20:20 |
sdake | ya networking always trips p eople up | 20:20 |
sdake | same story across all industries | 20:20 |
sdake | ;) | 20:21 |
vilobhmm11 | portdirect_ : will give vnc a try | 20:21 |
vilobhmm11 | sdake : right set of docs to get started http://docs.openstack.org/developer/kolla/quickstart.html right ? | 20:21 |
vilobhmm11 | or is there something more updated that i should be aware of | 20:21 |
sdake | thats most up to date | 20:21 |
sdake | that is published live from master | 20:21 |
vilobhmm11 | sdake : alrite..sure thing | 20:21 |
sdake | although our docs are in a state of disarray | 20:22 |
sdake | as a result of the repo split | 20:22 |
sdake | inc0 and Jeffrey4l are working on it | 20:22 |
vilobhmm11 | sdake : will focus on what you said | 20:22 |
vilobhmm11 | thanks for the help | 20:22 |
sdake | i'd recommmend reading the newton docs tho | 20:22 |
sdake | rather then master | 20:22 |
sdake | and deploying newton rather then master | 20:22 |
vilobhmm11 | ok | 20:22 |
sdake | those docs are not in a state of disarray | 20:22 |
vilobhmm11 | ok | 20:22 |
sdake | moment - i'll find a link | 20:23 |
vilobhmm11 | that will be gr8 | 20:23 |
sdake | http://docs.openstack.org/developer/kolla/newton/ | 20:23 |
portdirect__ | vilobhmm11: if you can it makes this sort of inital debugging much easier, as if your instance is getting an ip address and booting scussfully you are usually almost home and dry | 20:23 |
vilobhmm11 | portdirect__: I can see it through the console logs though | 20:24 |
*** fragatina has quit IRC | 20:24 | |
vilobhmm11 | sdake :thanks :) | 20:24 |
sdake | vilobhmm11 np - hope it helps :) | 20:24 |
*** fragatina has joined #openstack-kolla | 20:24 | |
vilobhmm11 | sdake : lets hope for the best | 20:24 |
*** fragatina has quit IRC | 20:25 | |
*** hyang has joined #openstack-kolla | 20:25 | |
sayantan_ | I have been working on the documentation. Plan to integrate it with the quickstart guide. Here is the link to the patch. https://review.openstack.org/#/c/404993/ | 20:26 |
portdirect__ | vilobhmm11: a simple debugging methodology I follow is to first (from the instance) ping the router, then if that scucceds try to ping the routher from your host, once you have these two ends of the problem solved your ususally eirther up and running or have a very clear idea what you need to do next. unfortuantly you cannot piing the router from the instance unless you you have vnc/spice access to it at this stage. | 20:26 |
*** adrian_otto has quit IRC | 20:29 | |
vilobhmm11 | portdirect__ : ok lemme try | 20:29 |
*** fragatina has joined #openstack-kolla | 20:29 | |
*** fragatina has quit IRC | 20:29 | |
*** adrian_otto has joined #openstack-kolla | 20:29 | |
sdake | sayantan_ we used to have a way to do that with tox -e gatename - Jeffrey4l took that out, not sure what he has in mind for a replacement in master | 20:30 |
portdirect__ | vilobhmm11: it's also worth checking that ICMP is open in your security group settings - It's caught me out more than a few times :) | 20:32 |
*** kristian__ has joined #openstack-kolla | 20:33 | |
sayantan_ | I am trying to delete the files passwords.yml and globals.yml from kolla/etc/kolla. However, I see that these files are still used in the setup_gate.sh | 20:34 |
*** hyang has quit IRC | 20:34 | |
*** kristian__ has quit IRC | 20:38 | |
*** adrian_otto has quit IRC | 20:39 | |
*** adrian_otto has joined #openstack-kolla | 20:39 | |
madgin | I have a multinode setup and am trying to get ceph working. 3 nodes with 2 disks each = 6 OSDs, which start correctly. | 20:40 |
*** madgin is now known as mattmceuen | 20:40 | |
*** fragatina has joined #openstack-kolla | 20:41 | |
mattmceuen | Does anyone know what would cause a HEALTH_ERR of "584 pgs are stuck inactive for more than 300 seconds"? | 20:41 |
mattmceuen | I don't have any custom config for ceph, beyond labelling my disks as KOLLA_CEPH_OSD_CACHE_BOOTSTRAP. They get bootstrapped correctly into 6 different OSDs during deployment. | 20:43 |
*** goldyfruit has quit IRC | 20:44 | |
mattmceuen | ^ using Newton; same behavior with 3.0.0 images and images built from latest newton source. | 20:45 |
*** kristian__ has joined #openstack-kolla | 20:45 | |
*** goldyfruit has joined #openstack-kolla | 20:46 | |
*** fragatina has quit IRC | 20:46 | |
*** schwicht has joined #openstack-kolla | 20:47 | |
*** rstarmer has joined #openstack-kolla | 20:52 | |
*** adrian_otto has quit IRC | 20:53 | |
*** sayantan_ has quit IRC | 20:58 | |
*** sayantan_ has joined #openstack-kolla | 20:59 | |
*** lamt has quit IRC | 21:03 | |
cliles | anyone on https://bugs.launchpad.net/kolla/+bug/1482196 ? I seem to be hitting this with a deployment from master, yet it's marked as fixed over a year ago?!? | 21:10 |
openstack | Launchpad bug 1482196 in kolla "Horizon dashboard not loading properly, can't click on anything" [Critical,Fix released] - Assigned to Sam Yaple (s8m) | 21:10 |
sdake | cliles recommend deploying from a stable branch | 21:11 |
sdake | cliles when deploying from master, you deploy master of openstack as well | 21:11 |
sdake | cliles which may not be what you want | 21:11 |
sdake | if it is what you want, I'd recommend filing a new bug as its a new issue | 21:12 |
cliles | sdake I want trove | 21:12 |
cliles | dodmn | 21:12 |
cliles | didn't see that in stable | 21:12 |
sdake | yup i believe trove recently landed | 21:13 |
cliles | yet trove dashboard doesn't seem to work correctly either :/ | 21:13 |
cliles | I'd love to open the bug, but I am still new and having a hard time following the links between all the projects. | 21:14 |
cliles | as in, is this a bug in the container, the deployment, or the rdo package | 21:14 |
sdake | cliles file bug in kolla-ansible's launchpad | 21:15 |
sdake | cliles it will be addressed | 21:15 |
sdake | cliles is all of your horizon dashboard not loading? | 21:16 |
cliles | so far I have identified 2 bugs, not sure if they are related | 21:16 |
cliles | symptoms of 1482196 | 21:16 |
sdake | cliles ok - well file em, and the team will look at em | 21:17 |
cliles | and that the trove-ui .pyc files are not in the location the contaner runtime expects them | 21:17 |
cliles | kk | 21:17 |
*** sdake has quit IRC | 21:22 | |
*** bachp has quit IRC | 21:22 | |
*** Marcellin__ has joined #openstack-kolla | 21:25 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: DO NOT MERGE!: Testing gate workflow (so I understand it) https://review.openstack.org/407207 | 21:27 |
harlowja | okie dokie, https://github.com/harlowja/kolla-jenkins mostly setup with a clone of stuff from our internal, probably still a bunch of GD specifics, but meh, its a start | 21:27 |
harlowja | https://github.com/harlowja/kolla-jenkins/blob/master/projects/nova.yaml is what creates the main jenkins job for say nova here | 21:27 |
harlowja | there is a repo it references internally that describes where to find versions of things for deployment and such | 21:28 |
vilobhmm11 | harlowja : will have a look ..thanks :) | 21:28 |
harlowja | ha | 21:28 |
harlowja | np | 21:28 |
srwilkers_ | awesome, thanks harlowja | 21:28 |
srwilkers_ | harlowja, curious to take a look | 21:28 |
harlowja | for a project say nova, u have to fill in https://github.com/harlowja/kolla-jenkins/blob/master/scripts/nova.groovy | 21:28 |
harlowja | the rest of the code is using jjb #include-raw mechanism (and a helper library that gets loaded at runtime) | 21:29 |
harlowja | https://github.com/harlowja/kolla-jenkins/blob/master/scripts/openstack.groovy (being the main piece of code shared among various openstack components) | 21:29 |
harlowja | and https://github.com/harlowja/kolla-jenkins/blob/master/lib/helpers.groovy being the helper lib | 21:29 |
*** sdake has joined #openstack-kolla | 21:29 | |
harlowja | with kolla building routine @ https://github.com/harlowja/kolla-jenkins/blob/master/lib/helpers.groovy#L340-L343 | 21:29 |
harlowja | knowledge required of jenkins, groovy, jjb | 21:31 |
harlowja | and probably kolla, ha | 21:31 |
*** fguillot has quit IRC | 21:32 | |
harlowja | questions welcome :-P | 21:34 |
harlowja | sdake after a little iteration prob will just make that under openstack, seem reasonable? | 21:34 |
harlowja | likely still to godaddy specific, ha | 21:35 |
*** Jeffrey4l has quit IRC | 21:35 | |
*** Jeffrey4l has joined #openstack-kolla | 21:36 | |
*** v1k0d3n has joined #openstack-kolla | 21:38 | |
harlowja | v1k0d3n ^ | 21:41 |
*** sayantan_ has quit IRC | 21:43 | |
*** sayantan_ has joined #openstack-kolla | 21:43 | |
sean-k-mooney | mliima_: hi, im just head home but did you get a chance to test manila with the conntrack firewall driver on friday? if not no worries just taught i would ask | 21:44 |
harlowja | and also https://jenkins.io/doc/book/pipeline/ | 21:47 |
harlowja | all this is what is being used | 21:47 |
*** lamt has joined #openstack-kolla | 21:49 | |
harlowja | and yes its probably very python like groovy code | 21:50 |
harlowja | to bad, ha | 21:50 |
sdake | harlowja i dont grok make that under openstack part ? | 21:54 |
harlowja | openstack/kolla-jenkins | 21:54 |
harlowja | (if people want to) | 21:54 |
sdake | harlowja thats a decision for inc0 to make | 21:54 |
harlowja | k | 21:54 |
sdake | harlowja and probably the core review team to vote on | 21:55 |
harlowja | right right | 21:55 |
harlowja | yes sir | 21:55 |
sdake | just guessing :) | 21:55 |
harlowja | :-p | 21:55 |
sdake | every time we add a repo - we add 2x pain multiplier ;) | 21:55 |
harlowja | lol | 21:55 |
harlowja | ya ya | 21:55 |
sdake | well time to go program myself in how to do agile - bbl | 21:56 |
harlowja | programming yourself is the hardest kind of programming | 21:56 |
*** sdake_ has joined #openstack-kolla | 21:57 | |
srwilkers_ | yes it is | 21:57 |
*** srwilkers_ has quit IRC | 21:58 | |
*** bachp has joined #openstack-kolla | 21:59 | |
*** ayoung-afk has quit IRC | 22:00 | |
openstackgerrit | Kris Lindgren proposed openstack/kolla: Spec to enable consistent use of user id's and group id's https://review.openstack.org/405647 | 22:00 |
*** awiddersheim has quit IRC | 22:00 | |
*** sdake has quit IRC | 22:01 | |
*** adrian_otto has joined #openstack-kolla | 22:02 | |
*** lrensing has quit IRC | 22:03 | |
*** awiddersheim has joined #openstack-kolla | 22:04 | |
*** harbie has quit IRC | 22:07 | |
sdake_ | core reviewers - need an ack on this before people change their minds :) | 22:07 |
sdake_ | https://review.openstack.org/#/c/372133/1 | 22:07 |
*** harbie has joined #openstack-kolla | 22:07 | |
sdake_ | sup srwilkers | 22:07 |
*** portdirect__ has left #openstack-kolla | 22:08 | |
*** portdirect__ has joined #openstack-kolla | 22:08 | |
*** harbie has quit IRC | 22:10 | |
*** harbie has joined #openstack-kolla | 22:11 | |
*** Pavo has quit IRC | 22:12 | |
*** portdirect__ has quit IRC | 22:13 | |
*** kristian__ has quit IRC | 22:13 | |
*** fragatina has joined #openstack-kolla | 22:14 | |
*** inc0 has joined #openstack-kolla | 22:17 | |
inc0 | hey | 22:17 |
inc0 | how are you guys doing? | 22:17 |
*** Pavo has joined #openstack-kolla | 22:17 | |
jascott1 | hey man! are you IRC'in whilst driving? ;) | 22:22 |
sean-k-mooney | sdake_: do we have similar disclamer for anything else? | 22:23 |
*** fragatina has quit IRC | 22:23 | |
openstackgerrit | Kris Lindgren proposed openstack/kolla: Spec to enable consistent use of user id's and group id's https://review.openstack.org/405647 | 22:23 |
*** jtriley has quit IRC | 22:24 | |
sean-k-mooney | sdake_: Open vSwitch and Ceph are both tradmarks but i dont think we have similar disclamers for those | 22:24 |
sean-k-mooney | sdake_: that said i dont really care either way just curios what promted adding that | 22:25 |
sdake_ | inc0 please ack https://review.openstack.org/#/c/372133/1 | 22:27 |
sdake_ | ok back to agile programming | 22:29 |
*** fragatina has joined #openstack-kolla | 22:29 | |
*** schwicht has quit IRC | 22:32 | |
*** sayantan_ has quit IRC | 22:34 | |
*** sayantan_ has joined #openstack-kolla | 22:35 | |
*** rstarmer has quit IRC | 22:35 | |
*** rstarmer has joined #openstack-kolla | 22:36 | |
*** rstarmer has quit IRC | 22:40 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helm-ize memcached https://review.openstack.org/405677 | 22:41 |
inc0 | done sdake_ | 22:44 |
sdake_ | thanks inc0 | 22:44 |
inc0 | jascott1, we depart tomorrow | 22:44 |
v1k0d3n | sdake_: hey man | 22:44 |
v1k0d3n | where is that jenkins BP again? | 22:44 |
v1k0d3n | can't find it | 22:44 |
sdake_ | inc0 note i asked jeffrey to ack a change re the gate since i thought you were moving | 22:45 |
jascott1 | this one? https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-packaging | 22:45 |
sdake_ | inc0 hope your cool with that | 22:45 |
sdake_ | jascott1 that is not the one | 22:45 |
inc0 | sdake_, yeah, I'll be away from tomorrow on | 22:45 |
sdake_ | v1k0d3n moment | 22:45 |
jascott1 | just pasting blindly | 22:45 |
v1k0d3n | this was addition of jenkins to create images for operators | 22:45 |
inc0 | even now I'm in literally almost empty apartment | 22:45 |
v1k0d3n | k | 22:45 |
jascott1 | as is my people's way | 22:45 |
sdake_ | inc0 oh well it merged :) | 22:45 |
inc0 | just router and pillow to sit on | 22:45 |
jascott1 | good thing you have VR! | 22:46 |
sdake_ | inc0 hopefully a toilet left in there too - the necessities of life :) | 22:46 |
v1k0d3n | sdake_: k | 22:46 |
jascott1 | inc0 hope you have a great trip. we got some snow today in anticipation of your arrival ;) | 22:47 |
sdake_ | v1k0d3n here are all blueprints for kolla-kubernetes atm | 22:47 |
sdake_ | https://blueprints.launchpad.net/kolla-kubernetes | 22:47 |
inc0 | yay | 22:47 |
v1k0d3n | sdake_: i think you moved it though ;) | 22:47 |
inc0 | jascott1, did you get contacted by car shipping company | 22:48 |
v1k0d3n | it was in kolla-kube...but i think you moved it somehwere. i checked all three...couldn't find it. | 22:48 |
inc0 | they should arrive tomorrow | 22:48 |
sdake_ | v1k0d3n its right in there - i haven't moved it | 22:48 |
jascott1 | inc0 no i have not. or at least not that I know of, will be on the lookout and will be WFH tomorrow so should smooth sailing | 22:48 |
sdake_ | v1k0d3n search for this on that page with CTRL-F | 22:48 |
sdake_ | kolla-kubernetes-ci-for-custom-i... | 22:48 |
v1k0d3n | im blind! | 22:49 |
v1k0d3n | sorry | 22:49 |
sdake_ | its ok - easy for stuff to get lost in the sea of blueprints outstanding :) | 22:49 |
inc0 | yeah, cool idea, harlowja is already working on something similar | 22:49 |
inc0 | just not k8s-y | 22:49 |
harlowja | shouldn't be k8s-y imho | 22:49 |
harlowja | k8s is a later stage imho | 22:50 |
harlowja | after u validate your images and crap | 22:50 |
harlowja | and put them through some tests and ... | 22:50 |
harlowja | then u activate the deployment solution | 22:50 |
harlowja | (ansible, k8s, blah blah) | 22:50 |
inc0 | yeah | 22:50 |
harlowja | building is way earlier | 22:50 |
inc0 | it's pretty pluggable | 22:50 |
inc0 | anyway, that would be one cool summit demo | 22:50 |
inc0 | "here, I submitted a patch and look at it go" | 22:51 |
harlowja | whats a summit | 22:51 |
harlowja | lol | 22:51 |
sdake_ | full ci pipeline? | 22:51 |
inc0 | yeah | 22:51 |
harlowja | CI pipeline not so hard, the CD part is imho harder | 22:51 |
harlowja | :-P | 22:51 |
sdake_ | 4 month cycle ... :) | 22:51 |
inc0 | patch->jenkins->container build->helm->deploy staging->perform tests->deploy prod | 22:51 |
*** jheroux has quit IRC | 22:51 | |
sdake_ | ya infrastructure as a service | 22:51 |
harlowja | inc0 ya | 22:51 |
harlowja | though imho a few more stages in there | 22:52 |
harlowja | a perf validation would be nice | 22:52 |
inc0 | yeah | 22:52 |
inc0 | ofc it's over simplification | 22:52 |
harlowja | :) | 22:52 |
inc0 | and ops would *hate* it for first months | 22:52 |
sdake_ | how long you been working on it harlowja | 22:52 |
harlowja | a few months after i got here | 22:52 |
harlowja | cause its the major gap :-P | 22:52 |
harlowja | so lets say 4 months? among other things, ha | 22:52 |
sdake_ | how many other godaddy cats working on it | 22:52 |
harlowja | most of the team, ha, but its complicated :-P | 22:53 |
inc0 | anyway, I'm going to finish up my apt packing and run all the errants needed for smooth move across country | 22:53 |
harlowja | ie u have to build out those other stages also | 22:53 |
harlowja | and u have to get your deployment data out of puppet (for general usage) | 22:53 |
harlowja | klindgren_ is doing some of that | 22:53 |
inc0 | be good when I'm away:) | 22:53 |
harlowja | klindgren_ is also trying to figure out how to glue and deploy it all in a way that doesn't blow up the current clouds | 22:53 |
harlowja | lol | 22:53 |
harlowja | (looks possible, just order dependent) | 22:54 |
sdake_ | ya brownfield would be a silver bullet ;) | 22:54 |
harlowja | ya, we may leave off say nova-libvirt though | 22:54 |
sbezverk | kfox1111: ping | 22:54 |
harlowja | (for a little wihle) | 22:54 |
*** schwicht has joined #openstack-kolla | 22:54 | |
sdake_ | understandable | 22:54 |
klindgren_ | yes, assessing kolla-ansible for ability to just handle pieces that we want it to. | 22:54 |
sdake_ | the upgrde model is complex from brown to brown | 22:54 |
harlowja | sdake_ we also aren't doing upgrade yet | 22:55 |
harlowja | just the same liberty with the same patches | 22:55 |
sdake_ | i mean upgrading from non-kolla to kolla brown :) | 22:55 |
harlowja | kk | 22:55 |
jascott1 | inc0 we have the best behavior, really just the best. Bigly even. | 22:55 |
harlowja | it will be done sdake_ do not fear | 22:55 |
harlowja | ha | 22:55 |
sdake_ | maybe upgrade is the wrong word | 22:55 |
harlowja | conversion? | 22:55 |
sdake_ | transition better word :) | 22:55 |
harlowja | ya, one of those :-P | 22:55 |
klindgren_ | we got part of the glance stuff running | 22:55 |
klindgren_ | but we really need this consistent uid/gid stuff | 22:55 |
inc0 | cyas all!:) | 22:56 |
harlowja | ya, didn't seem like it was crazy hard right klindgren_ | 22:56 |
*** inc0 has quit IRC | 22:56 | |
klindgren_ | because we have some stuff using nfs shares | 22:56 |
harlowja | except for uid problems | 22:56 |
sdake_ | klindgren_ did you talk to Jeffrey4l about it? | 22:56 |
klindgren_ | yes but still have questions which are covered in the spec that i just submitted | 22:56 |
harlowja | what are those questions :) | 22:56 |
sdake_ | ok - we probably didn't need a spec | 22:56 |
sdake_ | but whatever works for you :) | 22:56 |
klindgren_ | where/ how to place the service -> uid/gid mappings | 22:57 |
klindgren_ | I was told by you to do a spec! | 22:57 |
sdake_ | no i am anti-spec | 22:57 |
* klindgren_ sighs | 22:57 | |
harlowja | lol | 22:57 |
harlowja | lol | 22:57 |
sdake_ | if i said that, i must have been miscommunicating | 22:57 |
sdake_ | no fear | 22:57 |
sdake_ | sometimes we has hstuff out in a spec | 22:57 |
sdake_ | and then abandon it once consensus is reached | 22:57 |
klindgren_ | yea, just like what to use for the uid | 22:58 |
klindgren_ | where to do the sotring of that | 22:58 |
klindgren_ | where/how to override | 22:58 |
kfox1111 | sbezverk: hi | 22:58 |
kfox1111 | whats up? | 22:58 |
klindgren_ | how can we get that same mapping -> ansible | 22:58 |
sdake_ | cool - well add jeffrey and paul as reviewers | 22:58 |
sdake_ | if you wouldn't mind | 22:59 |
klindgren_ | whats their full names? jeff is not auto complete well | 23:00 |
harlowja | mr.jeff | 23:01 |
* harlowja runs away | 23:01 | |
v1k0d3n | harlowja: actually just reading above where you say later stage.... | 23:02 |
v1k0d3n | was actually going to use kubernetes as the cloud provider for doing the testing. | 23:02 |
harlowja | sure, u can | 23:02 |
v1k0d3n | however...by not doing that it makes it consumable by more. | 23:03 |
harlowja | well u can imho | 23:03 |
v1k0d3n | just use the docker that's already there. | 23:03 |
sbezverk | kfox1111: couple of questions. one Release name does not see to work, I checked memcached Jason is working on, has the same issue.. Possibly we are not using it correctly, but I do not see big big reason to use it. Second I still need to know where we store ceph variables.. | 23:03 |
*** adrian_otto has quit IRC | 23:03 | |
harlowja | v1k0d3n i think its ok to use any cloud u want for testing of a deployment | 23:03 |
harlowja | (pre-testing of a deployment) | 23:03 |
v1k0d3n | ^^ true | 23:03 |
harlowja | k8s or other | 23:03 |
jascott1 | sbezverk we need spec.selector.matchLabels | 23:03 |
harlowja | we have a lab that i'd like to target our stuff to (automatically) | 23:03 |
jascott1 | we have some code I copied that had just spec.selector | 23:04 |
v1k0d3n | harlowja: are you guys using operator driven or developer driven workflow | 23:04 |
v1k0d3n | ? | 23:04 |
sdake_ | klindgren_ jeffrey4l | 23:04 |
sdake_ | klindgren_ paul d bourke | 23:04 |
harlowja | v1k0d3n unsure what u mean :) | 23:04 |
jascott1 | but .Release.Name does indeed work | 23:04 |
sdake_ | ok - must focus - bbiaf :) | 23:04 |
*** srwilkers_ has joined #openstack-kolla | 23:04 | |
kfox1111 | sbezverk: let me test with relase name real quick. just need to figure out the right incantation. | 23:04 |
klindgren_ | jeffrey4l does not identify a registered user or group | 23:04 |
sbezverk | jascott1: why we need matchlabel? we have not used it before, just curious what was the need. | 23:04 |
*** schwicht has quit IRC | 23:05 | |
sdake_ | klindgren_ i'll ping him when i see him and ask him to take a look | 23:05 |
kfox1111 | as for ceph, yeah. not sure the best answer to that one. I think dynamic pvc's have a different object, something like a storagegroup config, where that stuff goes under. which is why I was asking about the 1.5 stuff. | 23:05 |
harlowja | v1k0d3n whats the difference :-P | 23:05 |
v1k0d3n | ^ good answer :) | 23:05 |
*** fragatin_ has joined #openstack-kolla | 23:05 | |
jascott1 | idk i couldnt get it to work wthout it, even tho I was seeing spec.selector = service: memcached-whatever, it was failing w 'spec.selector' cannot be empty | 23:05 |
v1k0d3n | i stopped typing when i saw that response. | 23:05 |
v1k0d3n | well...there is a difference, but not sure there should be. | 23:05 |
sbezverk | kfox1111: we need to decide them. from what I heard fro m sdake he wanted us to stay with 1.4.X for now.. | 23:06 |
*** rstarmer has joined #openstack-kolla | 23:06 | |
harlowja | v1k0d3n i hope not, with the right pipeline working the right way, it shouldn't matter | 23:06 |
sbezverk | I think so too, as then we need to convert petsets on the fly | 23:06 |
v1k0d3n | if you're using CI on images you're already developer focused lol | 23:06 |
v1k0d3n | sbezverk: remember...petsets aren't around very much longer. | 23:06 |
kfox1111 | sbezverk: yeah. we should talk with him then. a fair amount of difference there I think. | 23:06 |
v1k0d3n | statefulsets | 23:07 |
v1k0d3n | coming oh so soon! (as in now in our case) | 23:07 |
kfox1111 | but, I guess there may be a reason you might want to do either.... | 23:07 |
v1k0d3n | harlowja: :thumbsup | 23:07 |
*** JRobinson__ has joined #openstack-kolla | 23:07 | |
kfox1111 | so we could do it the old way, and add a dynamic version too. | 23:07 |
jascott1 | going to trudge across the tundra (mile after mile) bbiab | 23:07 |
*** jascott1 has quit IRC | 23:07 | |
sbezverk | kfox1111: I think doing both versions is perfectly fine. | 23:08 |
*** jascott1 has joined #openstack-kolla | 23:08 | |
*** fragatina has quit IRC | 23:08 | |
harlowja | v1k0d3n the part that gets harder is that u need a description of your envs to know what to do with these built images, where to target them (so u need environment files that describe that) | 23:08 |
harlowja | if u let k8s do most of that, probably makes it easier | 23:08 |
sbezverk | kfox1111: let's for now park ceph setting in common variables too, so we could move forward or into the statefull_service what ever it means ;-) | 23:09 |
*** jascott1 has quit IRC | 23:09 | |
kfox1111 | yeah, lets go with stateful_service and then put it all under there then? | 23:10 |
sbezverk | kfox1111: then prebuild will copy stateful_service to each microservice, right? | 23:11 |
kfox1111 | sbezverk: just tried it. this worked: | 23:12 |
kfox1111 | FOO: {{ .Release.Name }} | 23:12 |
kfox1111 | and it did set it to a random string if not specified on the commandline | 23:12 |
kfox1111 | helm install --dry-run --debug /tmp/neutron-server-3.0.0-1.tgz | grep FOO | 23:12 |
kfox1111 | FOO: esteemed-yak | 23:12 |
kfox1111 | helm install --dry-run --debug /tmp/neutron-server-3.0.0-1.tgz --name foo | grep FOO | 23:13 |
kfox1111 | FOO: foo | 23:13 |
sbezverk | kfox1111: strange, I will retest it | 23:13 |
kfox1111 | sbezverk: also: | 23:16 |
kfox1111 | [kfox@mantis ~]$ helm install /tmp/neutron-server-3.0.0-1.tgz ^C--name foo | 23:16 |
kfox1111 | [kfox@mantis ~]$ helm install /tmp/neutron-server-3.0.0-1.tgz --name foo | 23:16 |
kfox1111 | release name, is unique. | 23:16 |
kfox1111 | Error: a release named "foo" already exists | 23:16 |
*** ayoung-afk has joined #openstack-kolla | 23:20 | |
*** schwicht has joined #openstack-kolla | 23:21 | |
sbezverk | kfox1111: we use the same name for different kubernetes objects, we have pod rabbitmq, pvc/pv rabbitmq if we start using release.name instead of names, then we cannot refer to names since they will be unique | 23:22 |
*** hyang has joined #openstack-kolla | 23:22 | |
sbezverk | I guess that is why release.name would require labels | 23:24 |
*** vilobhmm111 has joined #openstack-kolla | 23:25 | |
*** jascott1 has joined #openstack-kolla | 23:25 | |
kfox1111 | we'd have to do helm install kolla/neutron-server --name neutron-server; helm install kolla/neutron-server-svc --name neutron-server-svc | 23:25 |
kfox1111 | as we can't double up on release name. and just hving it pick one really is unpleasent. the default's unusable. | 23:25 |
*** vilobhmm112 has joined #openstack-kolla | 23:26 | |
kfox1111 | or helm install kolla/neutron-server-daemonset --name neutron-server-daemonset or something. | 23:27 |
*** vilobhmm11 has quit IRC | 23:28 | |
*** vilobhmm111 has quit IRC | 23:29 | |
sbezverk | kfox1111: can you make ps to add copying of statefull_service: to all microservices? | 23:30 |
*** rstarmer has quit IRC | 23:30 | |
kfox1111 | sbezverk: sure | 23:31 |
kfox1111 | well, just the stateful services? | 23:31 |
kfox1111 | otherwise, we might as well put it in common. | 23:31 |
sbezverk | kfox1111: here is what I have now http://paste.openstack.org/show/591476/ | 23:33 |
kfox1111 | yeah. that looks good. | 23:34 |
srwilkers_ | just a nit, not sure if it matters, but i think they'll be stateful with one 'l' | 23:35 |
*** rstarmer has joined #openstack-kolla | 23:36 | |
sbezverk | srwilkers_: :-) stateful-service: | 23:37 |
srwilkers_ | :) | 23:38 |
sbezverk | srwilkers_: please bare with me, english is not my first language | 23:38 |
srwilkers_ | sbezverk, of course :) | 23:38 |
kfox1111 | sbezverk: do helm vars allow nesting lilke that? | 23:41 |
kfox1111 | how do you override them on the cli? | 23:41 |
*** schwicht has quit IRC | 23:42 | |
kfox1111 | sbezverk: oh.... and here's where it might fall apart.... | 23:42 |
kfox1111 | monitors shoudl be a list. | 23:42 |
kfox1111 | I wonder how helm handles that. | 23:42 |
jascott1 | hey guys memcache is failing the gate with "Error: transport is closing" I saw that before when I have helm 2.0.0 against k8s 1.5beta1. does that explain it or should I keep looking? | 23:43 |
kfox1111 | jascott1: we've seen that very speratically since we introduced helm. | 23:43 |
kfox1111 | :/ | 23:44 |
kfox1111 | seems like a bug in helm to me. | 23:44 |
sdake_ | sbezverk for not being your first language - you speak english really well :) | 23:44 |
sbezverk | yep, recheck is your friend | 23:44 |
kfox1111 | seen it against k8s 1.4 too. | 23:44 |
kfox1111 | sbezverk: I agree with sdake_ :) | 23:44 |
jascott1 | k thx good to know | 23:44 |
jascott1 | kfox1111 "monitors shoudl be a list" ? | 23:44 |
*** schwicht has joined #openstack-kolla | 23:44 | |
*** lamt has quit IRC | 23:44 | |
kfox1111 | jascott1: for ceph. | 23:45 |
*** schwicht has quit IRC | 23:45 | |
jascott1 | oh ok | 23:45 |
kfox1111 | there are usually 3. | 23:45 |
sdake_ | sbezverk i dont care if you stay with 1.4 for now or not - do whatever makes the most technical sense | 23:45 |
kfox1111 | but sometimes 1 or 5+ | 23:45 |
jascott1 | right right, thought you were talking probes sry | 23:45 |
kfox1111 | jascott1: ah. no worries. :) | 23:46 |
*** vhosakot has quit IRC | 23:46 | |
sdake_ | i think using a beta version of anything in the gate is not ideal though :) | 23:46 |
srwilkers_ | sbezverk, i agree with sdake_ and kfox1111 too :) | 23:46 |
kfox1111 | sdake_: yeah. though it looks like they are planning on dec 8 for the 1.5 release. so we can do beta, or we can even just wait. not that far out. | 23:47 |
kfox1111 | pleanty of work to do before then anyway. | 23:47 |
sdake_ | yup few days away | 23:47 |
sdake_ | assuming they are on schedule | 23:47 |
* kfox1111 nods | 23:47 | |
sdake_ | openstack for the most part runs on schedule, not sure about kubernetes - if they slip or not? | 23:48 |
kfox1111 | they have been pretty good about sticking to a schedule, | 23:51 |
kfox1111 | and they are much quicker releasing then openstack so far. | 23:51 |
kfox1111 | (and openstack releases very quickly) | 23:51 |
sbezverk | my ratonal is simple, we know 1.4.6 well, we had kolla-openstack working on it, so I do not expect any hidden obstacles or at least less, than with 1.5 | 23:53 |
kfox1111 | sbezverk: yeah. but we have a gate. so if we enable 1.5 and it just works, then great. if it fights, then we don't go there yet. :) | 23:53 |
*** Bico_Fino has joined #openstack-kolla | 23:54 | |
sbezverk | kfox1111: if it were that simple I would agree, but that means we need to maintain two versions of templates one with petset and one with stateful deployments.. | 23:54 |
sbezverk | since 1.5 is not backward compatible | 23:55 |
kfox1111 | sbezverk: we're going to run into that issue anyway, as people start running 1.5. | 23:57 |
kfox1111 | so either we support both versions, or we just start depending on 1.5 asap. | 23:58 |
kfox1111 | (I think the difference between statefulset and petset were just the name. so we could template it out by var if we reallly wanted to support it) | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!