| cloudnull | thanks spotz! | 00:01 |
|---|---|---|
| *** hybridpollo has joined #openstack-ansible | 00:09 | |
| *** yifei has joined #openstack-ansible | 00:11 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Updated from OpenStack Ansible Tests https://review.openstack.org/525821 | 00:12 |
| *** vnogin has joined #openstack-ansible | 00:13 | |
| *** yifei has quit IRC | 00:15 | |
| openstackgerrit | Jerry Zhao proposed openstack/openstack-ansible master: make include tasks dynamic when there is rescue https://review.openstack.org/527546 | 00:16 |
| *** vnogin has quit IRC | 00:19 | |
| *** hamzy has joined #openstack-ansible | 00:21 | |
| *** yifei has joined #openstack-ansible | 00:23 | |
| *** openstackgerrit has quit IRC | 00:33 | |
| *** chyka_ has quit IRC | 00:34 | |
| *** openstackgerrit has joined #openstack-ansible | 00:45 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-repo_build stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/527551 | 00:45 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Further simplify the upgrade check https://review.openstack.org/524086 | 00:52 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Only build ARA report when test fails https://review.openstack.org/512217 | 00:57 |
| *** masuberu has quit IRC | 00:59 | |
| *** vnogin has joined #openstack-ansible | 01:08 | |
| *** liangqi has joined #openstack-ansible | 01:19 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Allow disabling container restarts https://review.openstack.org/527256 | 01:22 |
| *** woodard has joined #openstack-ansible | 01:23 | |
| *** hw_wutianwei has joined #openstack-ansible | 01:26 | |
| *** vnogin has quit IRC | 02:01 | |
| *** gouthamr has joined #openstack-ansible | 02:06 | |
| *** chyka has joined #openstack-ansible | 02:06 | |
| *** jafeha__ has joined #openstack-ansible | 02:07 | |
| *** jafeha has quit IRC | 02:11 | |
| *** chyka has quit IRC | 02:11 | |
| *** rodolof has quit IRC | 02:12 | |
| *** lbragstad has quit IRC | 02:25 | |
| *** masber has joined #openstack-ansible | 02:27 | |
| spotz | welcome cloudnull | 02:31 |
| *** poopcat has quit IRC | 03:10 | |
| *** pbandark has quit IRC | 03:19 | |
| *** poopcat has joined #openstack-ansible | 03:22 | |
| *** poopcat1 has joined #openstack-ansible | 03:27 | |
| *** poopcat has quit IRC | 03:27 | |
| *** cuongnv has joined #openstack-ansible | 03:40 | |
| *** jwitko_ has quit IRC | 03:40 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add pip config to lxc_hosts https://review.openstack.org/527308 | 03:42 |
| *** nshetty has joined #openstack-ansible | 03:48 | |
| *** m|y|k has joined #openstack-ansible | 04:06 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-plugins master: Remove the vars plugin as it's no longer used https://review.openstack.org/527415 | 04:09 |
| *** threestrands has joined #openstack-ansible | 04:14 | |
| *** threestrands has quit IRC | 04:14 | |
| *** threestrands has joined #openstack-ansible | 04:14 | |
| *** jappleii__ has quit IRC | 04:15 | |
| *** threestrands has quit IRC | 04:15 | |
| *** threestrands has joined #openstack-ansible | 04:16 | |
| *** threestrands has quit IRC | 04:16 | |
| *** threestrands has joined #openstack-ansible | 04:16 | |
| *** jwitko_ has joined #openstack-ansible | 04:17 | |
| *** udesale has joined #openstack-ansible | 04:17 | |
| *** openstackgerrit has quit IRC | 04:18 | |
| *** chhavi__ has joined #openstack-ansible | 04:22 | |
| *** openstackgerrit has joined #openstack-ansible | 04:22 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-pip_install master: Set pip to use the pypi-server when locked down https://review.openstack.org/526889 | 04:22 |
| *** dave-mcc_ has quit IRC | 04:25 | |
| *** pmannidi has quit IRC | 04:27 | |
| *** jwitko_ has quit IRC | 04:29 | |
| *** pmannidi has joined #openstack-ansible | 04:32 | |
| *** taseer1 is now known as Taseer | 04:32 | |
| *** threestrands has quit IRC | 04:36 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create master: Unify container network interfaces with networkd https://review.openstack.org/523218 | 04:37 |
| *** threestrands has joined #openstack-ansible | 04:43 | |
| *** threestrands has quit IRC | 04:44 | |
| *** threestrands has joined #openstack-ansible | 04:44 | |
| *** threestrands has quit IRC | 04:44 | |
| *** threestrands has joined #openstack-ansible | 04:44 | |
| *** threestrands has quit IRC | 04:45 | |
| *** threestrands has joined #openstack-ansible | 04:46 | |
| *** threestrands has quit IRC | 04:47 | |
| *** threestrands has joined #openstack-ansible | 04:47 | |
| *** m|y|k has quit IRC | 04:49 | |
| *** m|y|k has joined #openstack-ansible | 04:50 | |
| *** m|y|k has quit IRC | 04:59 | |
| *** jose-phillips has quit IRC | 05:13 | |
| *** poopcat1 has quit IRC | 05:16 | |
| *** m|y|k has joined #openstack-ansible | 05:20 | |
| *** jose-phillips has joined #openstack-ansible | 05:21 | |
| *** chyka has joined #openstack-ansible | 05:23 | |
| *** chyka has quit IRC | 05:27 | |
| *** nshetty_ has joined #openstack-ansible | 05:28 | |
| *** woodard has quit IRC | 05:30 | |
| *** nshetty has quit IRC | 05:31 | |
| *** woodard has joined #openstack-ansible | 05:31 | |
| *** Smeared_Beard has quit IRC | 05:42 | |
| *** m|y|k has quit IRC | 05:45 | |
| *** m|y|k has joined #openstack-ansible | 05:47 | |
| *** hybridpollo has quit IRC | 05:57 | |
| *** threestrands has quit IRC | 06:03 | |
| *** nshetty__ has joined #openstack-ansible | 06:03 | |
| *** nshetty_ has quit IRC | 06:07 | |
| *** armaan_ has joined #openstack-ansible | 06:11 | |
| *** armaan has quit IRC | 06:11 | |
| *** armaan_ has quit IRC | 06:16 | |
| *** SmearedBeard has joined #openstack-ansible | 06:17 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Set the default machinectl limit to unlimited https://review.openstack.org/527592 | 06:20 |
| *** SmearedBeard has quit IRC | 06:22 | |
| *** nshetty__ is now known as nshetty|BMGR | 06:34 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Show debug message when cert is not found https://review.openstack.org/527598 | 06:38 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Show debug message when cert is not found https://review.openstack.org/527598 | 06:39 |
| *** m|y|k has quit IRC | 06:47 | |
| *** m|y|k_ has joined #openstack-ansible | 06:47 | |
| *** gouthamr has quit IRC | 06:48 | |
| *** chhavi__ has quit IRC | 06:49 | |
| *** chhavi__ has joined #openstack-ansible | 06:49 | |
| *** udesale__ has joined #openstack-ansible | 07:01 | |
| *** arbrandes has joined #openstack-ansible | 07:01 | |
| *** udesale has quit IRC | 07:03 | |
| *** arbrandes1 has quit IRC | 07:03 | |
| *** udesale has joined #openstack-ansible | 07:04 | |
| *** chhavi__ has quit IRC | 07:05 | |
| *** chhavi__ has joined #openstack-ansible | 07:06 | |
| *** udesale__ has quit IRC | 07:06 | |
| *** armaan has joined #openstack-ansible | 07:12 | |
| *** chhavi has joined #openstack-ansible | 07:14 | |
| *** chhavi__ has quit IRC | 07:16 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Fix SSL cert distribution tasks https://review.openstack.org/527598 | 07:28 |
| *** pmannidi has quit IRC | 07:29 | |
| *** epalper has joined #openstack-ansible | 07:32 | |
| *** sxc731 has joined #openstack-ansible | 07:41 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Converge neutron agents onto Baremetal https://review.openstack.org/454450 | 07:42 |
| *** nshetty|BMGR is now known as nshetty|lunch | 07:44 | |
| *** shardy has joined #openstack-ansible | 07:46 | |
| *** udesale__ has joined #openstack-ansible | 07:46 | |
| *** udesale has quit IRC | 07:49 | |
| *** armaan has quit IRC | 07:53 | |
| *** armaan has joined #openstack-ansible | 07:53 | |
| *** mbuil has joined #openstack-ansible | 07:59 | |
| *** DanyC has joined #openstack-ansible | 08:00 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 08:01 |
| evrardjp | cloudnull: I will unpin indeed | 08:04 |
| evrardjp | cloudnull: now that the convergence into baremetal and more converged stuff exists, will you focus on the migration? | 08:05 |
| evrardjp | so that everyone upgrading can be in the state of master? | 08:05 |
| *** ThomasS has joined #openstack-ansible | 08:09 | |
| *** jvidal has joined #openstack-ansible | 08:09 | |
| ThomasS | hi | 08:09 |
| ThomasS | i wanted to reconfiigure keystone | 08:10 |
| ThomasS | and it looks like it "hangs" forever at | 08:10 |
| ThomasS | TASK [pip_install : Install UCA repository key] | 08:10 |
| *** jafeha__ has quit IRC | 08:10 | |
| ThomasS | no error no timeout | 08:10 |
| ThomasS | i am on 16.0.4 | 08:10 |
| ThomasS | after reruning it it says fatal: [infra1_keystone_container-3cbadb01]: FAILED! => {"changed": false, "failed": true, "msg": "Failed to lock apt for exclusive operation"} | 08:11 |
| evrardjp | you already have an apt running then | 08:12 |
| ThomasS | trying to restart the affected container | 08:12 |
| ThomasS | i guess so | 08:12 |
| ThomasS | looks like it is "hanging" now again at TASK [pip_install : Install UCA repository key] | 08:13 |
| ThomasS | what could this be? | 08:13 |
| evrardjp | may you uncleanly interrupted something, and you still have apt locks? | 08:17 |
| evrardjp | connect on the machine and try to do an apt update / apt upgrade to see the issue | 08:17 |
| evrardjp | don't do the upgrade | 08:17 |
| evrardjp | just do a fake one | 08:18 |
| evrardjp | just to see | 08:18 |
| ThomasS | restarted the container and the lock is gone | 08:19 |
| ThomasS | now the tasks "hangs" at the TASK [pip_install : Install UCA repository key] | 08:20 |
| ThomasS | it sits there forever | 08:20 |
| ThomasS | how can i test in the container it this works? | 08:20 |
| evrardjp | yeah | 08:22 |
| ThomasS | i am in the container | 08:22 |
| ThomasS | apt-get update hangs | 08:22 |
| ThomasS | restart repo cotainers? | 08:23 |
| ThomasS | internet works flawlessly from the container | 08:23 |
| evrardjp | no there is something wrong on your container that happened | 08:23 |
| evrardjp | you have to fix it | 08:23 |
| evrardjp | then you can restart it if you want | 08:23 |
| evrardjp | but it looks like all your containers got an issue | 08:23 |
| evrardjp | so I'd be worried | 08:23 |
| evrardjp | maybe try to restart then all just in case | 08:23 |
| evrardjp | but I'd check at the root cause | 08:24 |
| ThomasS | i am on a different host | 08:24 |
| ThomasS | same keystone container | 08:24 |
| ThomasS | apt-get also hanging | 08:24 |
| ThomasS | looks like all containers | 08:25 |
| evrardjp | I would be worried that your base template was broken due to something you did. | 08:25 |
| ThomasS | so restart the repo? | 08:25 |
| evrardjp | is that a new environment? | 08:25 |
| evrardjp | you're still in testing phase right? | 08:25 |
| ThomasS | a test one works fine so far | 08:25 |
| evrardjp | What did you do before this happened | 08:25 |
| ThomasS | yes but i would not want to destroy it again | 08:25 |
| evrardjp | ? | 08:25 |
| evrardjp | ThomasS: why? | 08:26 |
| ThomasS | reconfigured haproxy with a cutsom service | 08:26 |
| evrardjp | no I mean something that touches apt | 08:26 |
| ThomasS | and then added ldap to users var | 08:26 |
| ThomasS | nothing | 08:26 |
| evrardjp | Did you control C on something? | 08:26 |
| ThomasS | yes | 08:26 |
| ThomasS | but only after i noticed that the task was hanging | 08:26 |
| evrardjp | mmm | 08:27 |
| evrardjp | restart your containers see if it goes better. | 08:27 |
| evrardjp | but I am worried that there would be something wrong somewhere. | 08:27 |
| evrardjp | I don't have time to look at that in more details, I am sorry | 08:27 |
| evrardjp | but this is definitely something in your env | 08:28 |
| ThomasS | ok | 08:28 |
| evrardjp | finding the root cause is the key to know how to not have it happen again | 08:28 |
| ThomasS | i think i know what has happend.... | 08:32 |
| *** sxc731 has quit IRC | 08:35 | |
| *** yolanda__ has joined #openstack-ansible | 08:37 | |
| evrardjp | ThomasS: yes? | 08:37 |
| *** yolanda has quit IRC | 08:40 | |
| evrardjp | SamYaple: your advice with permissions of yesterday didn't change anything it's something else | 08:44 |
| evrardjp | if you have another idea | 08:45 |
| evrardjp | on keystone I meant | 08:45 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [DNM] Build everything on metal https://review.openstack.org/504224 | 08:46 |
| *** yolanda__ has quit IRC | 08:46 | |
| *** yolanda__ has joined #openstack-ansible | 08:46 | |
| *** yolanda__ is now known as yolanda | 08:47 | |
| *** nshetty|lunch is now known as nshetty | 08:48 | |
| *** vnogin has joined #openstack-ansible | 08:49 | |
| *** jafeha__ has joined #openstack-ansible | 08:51 | |
| *** pmannidi has joined #openstack-ansible | 08:52 | |
| *** vnogin has quit IRC | 08:53 | |
| ThomasS | just for the records | 08:56 |
| ThomasS | it was a problem in the front firewall :-) | 08:56 |
| ThomasS | thanks evrardjp for your help | 08:56 |
| evrardjp | I couldn't help I am sorry | 08:57 |
| evrardjp | but I am glad you found the root cause | 08:57 |
| evrardjp | front firewall gave wrong headers and broke apt? | 08:57 |
| evrardjp | or is there a more hidden issue like mtu? | 08:57 |
| *** jafeha has joined #openstack-ansible | 08:59 | |
| *** chyka has joined #openstack-ansible | 08:59 | |
| ThomasS | oh no | 08:59 |
| ThomasS | now i am getting the "getting modern pip" error again.... | 09:00 |
| *** jafeha__ has quit IRC | 09:00 | |
| ThomasS | TASK [pip_install : Install pip packages (fall back mode)] | 09:00 |
| ThomasS | this fails... | 09:00 |
| ThomasS | all containers can reach the internet | 09:00 |
| ThomasS | the first task which fails on os-keystone is this one | 09:01 |
| ThomasS | TASK [pip_install : Get Modern PIP] | 09:01 |
| *** chyka has quit IRC | 09:03 | |
| *** pmannidi has quit IRC | 09:04 | |
| ThomasS | ok | 09:04 |
| ThomasS | it wants to pull a file from the repo http://10.0.100.9:8181/os-releases/16.0.5/ubuntu-16.04-x86_64/get-pip.py | 09:04 |
| ThomasS | which is not there | 09:04 |
| ThomasS | how can this be fixed? | 09:04 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Fix automatic log copying in Zuul runs https://review.openstack.org/516448 | 09:05 |
| *** pmannidi has joined #openstack-ansible | 09:05 | |
| evrardjp | ThomasS: run repo build | 09:06 |
| ThomasS | ok | 09:06 |
| ThomasS | should i checkour 16.0.5 ? | 09:06 |
| ThomasS | i have the system on 16.0.4 | 09:06 |
| evrardjp | You probably already did | 09:06 |
| evrardjp | else it would not know | 09:06 |
| evrardjp | so when you checkout a new version you should always bootstrap ansible again and run all the playbooks | 09:07 |
| evrardjp | in your case just repo build would do the trick | 09:07 |
| ThomasS | i will follow the minor upgrade guide | 09:08 |
| *** masber has quit IRC | 09:09 | |
| ThomasS | i really know how to break things ;:-( | 09:09 |
| *** sxc731 has joined #openstack-ansible | 09:15 | |
| *** pradiprwt has joined #openstack-ansible | 09:19 | |
| pradiprwt | Hi Everyone, I am trying to configure networking in centos 7 but I don't have much experience with centos. Can any one please provide or refer the sample configuration for interface, vlan interface and bridge interface. | 09:22 |
| *** rpittau has quit IRC | 09:26 | |
| evrardjp | pradiprwt: we have a wishlist bug item for that. Until it's solved you'll have to sort it out yourself because I sadly can't help you :( | 09:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Revert "Revert "Use Ansible 2.4"" https://review.openstack.org/527200 | 09:29 |
| evrardjp | is there no redhat network configuration guide? | 09:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/pike: Allow additional parameters for ansible-lint https://review.openstack.org/527645 | 09:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Skip ansible-lint test against roles https://review.openstack.org/527647 | 09:31 |
| pradiprwt | evrardjp : There is only ubuntu and centos is available in documentation but there is example is available for network configuration in centos | 09:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Reduce console output in gate jobs https://review.openstack.org/527648 | 09:32 |
| *** nshetty is now known as nshetty|afk | 09:32 | |
| pradiprwt | evrardjp : No example for centos | 09:33 |
| hwoarang | good morning | 09:34 |
| odyssey4me | o/ | 09:36 |
| evrardjp | good morning hwoarang | 09:38 |
| *** rpittau has joined #openstack-ansible | 09:42 | |
| *** Taseer has quit IRC | 09:44 | |
| *** vnogin has joined #openstack-ansible | 09:44 | |
| *** nshetty|afk is now known as nshetty | 09:46 | |
| evrardjp | hwoarang: could you vote on https://review.openstack.org/#/c/527127/ ? Neutron gates are broken right now due to this: http://logs.openstack.org/07/526307/1/check/openstack-ansible-functional-ubuntu-xenial/3704b3b/job-output.txt.gz#_2017-12-13_09_10_36_339009 | 09:46 |
| hwoarang | on it | 09:47 |
| evrardjp | thanks | 09:47 |
| *** yifei has quit IRC | 09:49 | |
| odyssey4me | evrardjp needs adjustment | 09:52 |
| odyssey4me | you'll have to add that into each of the existing overrides files, otherwise it won't work | 09:52 |
| *** hw_wutianwei has quit IRC | 09:56 | |
| *** electrofelix has joined #openstack-ansible | 09:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/pike: Fix systemd init template for program override https://review.openstack.org/527080 | 10:01 |
| *** dun has quit IRC | 10:01 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron master: Include neutron plugins for neutron testing https://review.openstack.org/527127 | 10:04 |
| odyssey4me | evrardjp you've misunderstood the feedback in the review | 10:04 |
| evrardjp | wasn't clear enough then | 10:04 |
| evrardjp | you wanted to use os_neutron-overrides | 10:05 |
| evrardjp | right? | 10:05 |
| evrardjp | we are not using it by default so it seems | 10:05 |
| odyssey4me | that is the default | 10:05 |
| odyssey4me | and every functional test has its own override file | 10:06 |
| evrardjp | yes | 10:06 |
| evrardjp | I see that | 10:06 |
| odyssey4me | so don't change tox, just add the plugins to the existing override files | 10:06 |
| evrardjp | could you show me where it's used by default? | 10:06 |
| evrardjp | th os_neutron-overrides file | 10:06 |
| odyssey4me | the default is to ise <role_name>-overrides.yml: https://github.com/openstack/openstack-ansible-tests/blob/master/test-ansible-functional.sh#L38 | 10:06 |
| odyssey4me | *use | 10:06 |
| odyssey4me | so we only need to set the override if we want to change that, like for a scenario for example | 10:07 |
| evrardjp | maybe I am blind but I don't see it sourced from run tests | 10:08 |
| odyssey4me | every single functional test uses that script: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tox.ini#L104 | 10:08 |
| evrardjp | run_tests -> run tests common -> tox | 10:08 |
| odyssey4me | tox -> test-ansible-functional.sh | 10:08 |
| evrardjp | right. | 10:09 |
| evrardjp | ok | 10:09 |
| evrardjp | so why does it fail then? | 10:09 |
| evrardjp | mmm | 10:09 |
| odyssey4me | why does what fail? | 10:09 |
| odyssey4me | you're adding a new override file which is missing all the other bits to make that test work | 10:09 |
| evrardjp | the tests for neutron, because the plugins are installed | 10:09 |
| evrardjp | no I mean they have the same content | 10:09 |
| odyssey4me | oh right, I see: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tests/os_neutron-overrides.yml#L18-L21 | 10:09 |
| *** cuongnv has quit IRC | 10:10 | |
| odyssey4me | ok, where's a fail? | 10:10 |
| odyssey4me | are yu sure it's not meant to switch to https://github.com/openstack/neutron-tempest-plugin ? it's a community goal to have plugins in their own repositories IIRC | 10:10 |
| *** dun has joined #openstack-ansible | 10:16 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/newton: Fix tempest plugins installation https://review.openstack.org/527070 | 10:18 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible master: Add networking-sfc repo to repo_packages https://review.openstack.org/525264 | 10:19 |
| jafeha | good morning | 10:19 |
| *** Taseer has joined #openstack-ansible | 10:28 | |
| *** nshetty_ has joined #openstack-ansible | 10:32 | |
| *** nshetty__ has joined #openstack-ansible | 10:34 | |
| *** vnogin has quit IRC | 10:35 | |
| *** nshetty has quit IRC | 10:35 | |
| *** stuartgr has joined #openstack-ansible | 10:36 | |
| *** nshetty_ has quit IRC | 10:37 | |
| *** vnogin has joined #openstack-ansible | 10:41 | |
| *** nshetty__ has quit IRC | 10:51 | |
| *** nshetty__ has joined #openstack-ansible | 10:52 | |
| *** hamza21 has joined #openstack-ansible | 10:56 | |
| *** gun1x has joined #openstack-ansible | 11:02 | |
| gun1x | hey guys | 11:02 |
| gun1x | how do you upgrade everything (including galera cluster) ? | 11:02 |
| gun1x | and rabbit mq ... all from one command. is that possible? | 11:02 |
| *** SmearedBeard has joined #openstack-ansible | 11:03 | |
| *** pmannidi has quit IRC | 11:06 | |
| *** pmannidi has joined #openstack-ansible | 11:07 | |
| *** udesale__ has quit IRC | 11:08 | |
| evrardjp | openstack-ansible setup-infrastructure.yml | 11:10 |
| *** armaan has quit IRC | 11:11 | |
| *** m|y|k_ has quit IRC | 11:12 | |
| *** armaan has joined #openstack-ansible | 11:12 | |
| evrardjp | gun1x: that's what we do in our major upgrades: https://github.com/openstack/openstack-ansible/blob/master/scripts/run-upgrade.sh#L182-L189 | 11:12 |
| evrardjp | sorry | 11:12 |
| evrardjp | https://github.com/openstack/openstack-ansible/blob/master/scripts/run-upgrade.sh#L174-L197 | 11:12 |
| *** pmannidi has quit IRC | 11:13 | |
| gun1x | evrardjp: i just mean from pike to pike | 11:15 |
| *** SmearedBeard has quit IRC | 11:16 | |
| evrardjp | technically re-running the playbooks should do. That's what we said we should do at least... Maybe worth asking ppl on this chan here if that changed. | 11:16 |
| *** m|y|k has joined #openstack-ansible | 11:18 | |
| gloksinia | evrardjp: I just wanna know, What should i focus or concern about deploy openstack using OSA like scaling, upgrading, updating for production at scale. | 11:26 |
| *** pbandark has joined #openstack-ansible | 11:28 | |
| gloksinia | evrardjp: I'm newbie with this things but I need to build a production asap right now I'm concerned T_T | 11:31 |
| *** chhavi has quit IRC | 11:33 | |
| *** chhavi has joined #openstack-ansible | 11:33 | |
| ThomasS | gloksinia: this is definitively very hard even with OSA | 11:37 |
| *** armaan has quit IRC | 11:41 | |
| *** armaan has joined #openstack-ansible | 11:41 | |
| *** armaan has quit IRC | 11:41 | |
| odyssey4me | gloksinia a three node control plane will give you scale up to around 250 computes, depending on your usage profile | 11:41 |
| odyssey4me | if you're network heavy, then rather dedicate some hosts for the network control plane | 11:42 |
| *** armaan has joined #openstack-ansible | 11:42 | |
| odyssey4me | if you're using telemetry, then build a dedicated telemetry rabbitmq cluster | 11:42 |
| odyssey4me | OSA allows you to architect in any way you want, but you have to understand architecting openstack and your use-case's performance profile | 11:42 |
| odyssey4me | if you don't know, then just build something standard and have a really good set of monitoring systems so that you can observe where bottlenecks are and can adjust your arch accordingly | 11:43 |
| gloksinia | ThomasS: Do you experienced using OSA on production ? | 11:43 |
| *** m|y|k has quit IRC | 11:43 | |
| odyssey4me | the only things that matter initially are your choices of network infra (use linuxbridge or something else) and your choices of storage... see the deploy guide's architecture section for more info | 11:44 |
| *** Guest7416 has quit IRC | 11:44 | |
| *** yifei has joined #openstack-ansible | 11:44 | |
| gloksinia | odyssey4me: That much information let me digest a sec LOL | 11:45 |
| *** nshetty__ is now known as nshetty | 11:47 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Change the galera health check for better cluster health https://review.openstack.org/527099 | 11:47 |
| gloksinia | odyssey4me: Okay for now i have got a standard deployment using OVS as network and Ceph as backend for testing. | 11:48 |
| gloksinia | odyssey4me: But only one problem of mine is how can i monitoring it well ? | 11:50 |
| gloksinia | odyssey4me: Do you have any suggestion ? | 11:50 |
| odyssey4me | gloksinia that's going to be up to you - we don't have a convenient suggestion unfortunatey | 11:52 |
| odyssey4me | some use zabbix, some use nagios, some use prometheus, some use sensu | 11:52 |
| *** pradiprwt has quit IRC | 11:52 | |
| odyssey4me | others still use various proprietary bits | 11:52 |
| odyssey4me | the only stuff we have which could be useful is https://github.com/openstack/openstack-ansible-ops/tree/master/cluster_metrics but that's just metrics, not alerting afaik | 11:53 |
| gokhan | hi folks, I see a bug on magnum on openstack ansible pike . Because of keystone public endpoint we can not create kubernates and swarm cluster. keystone public endpoint uses keystone version 2. But magnum need v3 | 11:56 |
| odyssey4me | gokhan we've been using v3 for the public endpoint since kilo | 11:57 |
| gloksinia | odyssey4me: Oh yeah that's one cloudnull suggest me too so i'm implementing but it can be monitoring only resource usage right ? Does it enough ? Do i need to monitoring on specific for some service ? | 11:57 |
| odyssey4me | if you have a v2 endpoint, then something's not right | 11:57 |
| gokhan | it creates public endpoint https://example.org:5000 but magnum needs https://example.org:5000/v3 | 11:57 |
| odyssey4me | gokhan what we do find, though, sometimes is that the conf for the service is set to use v2 because of the service not understanding how to use v3... so usually it's a misconfig in the service conf, nothing to do with keystone | 11:58 |
| odyssey4me | gokhan that means that magnum is not able to use versionless endpoints, which is a bug in magnum... or it's a misconfig in the magnum.conf because the config tells it how to use it and the keystoneauth library does the version discovery and stuff | 11:59 |
| CloudToko | gloksinia, we use check_mk | 11:59 |
| odyssey4me | I'd suggest comparing the cinder/nova/neutron/glance keystone client config and service config bits to spot differences | 11:59 |
| CloudToko | custom scripts , rally to test the platform from time to time ... check_mk for the hosts | 11:59 |
| odyssey4me | gloksinia if you want to scale, you want to monitor everyting - you need all the data you can get to be informed | 11:59 |
| odyssey4me | a word of advice though, don't try to get it perfect first time - that is an impossible task | 12:00 |
| odyssey4me | prepare the environment to change, and change often | 12:01 |
| odyssey4me | make sure you have a good test env, one that you can build/rebuild over and over again to try different things to figure out how you can evolve and improve | 12:01 |
| gokhan | odyssey4me, magnum use keystone public endpoint for trustee_keystone_interface. it does not get it from keystone auth. You said we have been using v3 but when I run on any openstack instance curl -vk https://example.org:5000 it says it is using v2 | 12:04 |
| gloksinia | cloudtoko: Thank you so much for your advice. I will working on it. | 12:07 |
| *** yifei has quit IRC | 12:09 | |
| odyssey4me | gokhan ok, so https://github.com/openstack/openstack-ansible-os_magnum/blob/master/templates/magnum.conf.j2 contains a bunch of different keystone stuff | 12:10 |
| odyssey4me | there is the magnum service bit which is the keystone_authtoken section, which is not what your concern is | 12:11 |
| odyssey4me | there is the trust section, which sets some things but may not be setting everything, and it's plausible that magnum's defaults set it to use v2 instead of v3 | 12:11 |
| gloksinia | odyssey4me: Yep fighting, I will working hard on it. So last question what about 3 site openstack any suggestion ?? | 12:12 |
| odyssey4me | then there are all the client sections which, as you can see, we do minimal config for in the hope that magnum has good defaults | 12:12 |
| odyssey4me | not sure if this helps, but in neutron there is also a trust thing, which we configure like https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/neutron.conf.j2#L107-L116 | 12:13 |
| odyssey4me | but I think you'll need to look into what's being configured in your env, what the defaults are, and therefore what might need changing to make it do the right things | 12:13 |
| odyssey4me | gloksinia what about 3 site openstack? you can do it in many, many ways - it depends what you're trying to achieve | 12:13 |
| *** sxc731 has quit IRC | 12:14 | |
| gloksinia | odyssey4me: Okay, Thank you so much :) | 12:16 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Update static files https://review.openstack.org/526308 | 12:16 |
| *** chhavi__ has joined #openstack-ansible | 12:18 | |
| *** chhavi has quit IRC | 12:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/ocata: Implement a proper WSREP check for galera https://review.openstack.org/527679 | 12:22 |
| ThomasS | gloksinia: one suggestion from my side keep it as simple as possible, i would also go with linuxbrigdes instead of ovs | 12:23 |
| *** dave-mccowan has joined #openstack-ansible | 12:23 | |
| *** armaan has quit IRC | 12:31 | |
| *** armaan has joined #openstack-ansible | 12:32 | |
| *** masuberu has joined #openstack-ansible | 12:33 | |
| *** armaan has quit IRC | 12:37 | |
| *** armaan has joined #openstack-ansible | 12:38 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/ansible-hardening master: tasks: auth: Use standard Grub2 authentication mechanism https://review.openstack.org/527682 | 12:39 |
| *** yifei has joined #openstack-ansible | 12:42 | |
| *** jamesdenton has joined #openstack-ansible | 12:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/newton: Implement a proper WSREP check for galera https://review.openstack.org/527684 | 12:48 |
| *** smatzek has joined #openstack-ansible | 12:51 | |
| *** nshetty has quit IRC | 12:53 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/ansible-hardening master: tasks: Add missing tags for async tasks https://review.openstack.org/527685 | 12:55 |
| *** vnogin has quit IRC | 12:56 | |
| *** armaan has quit IRC | 12:56 | |
| *** armaan has joined #openstack-ansible | 12:57 | |
| gokhan | odyssey4me, magnum use keystone v3, magnum doesn't work with v2. because of that it doesn't work clearly. is it possible to change keystone public endpoint from https://example.com:5000 to https://example.com:5000/v3 ? is it be problem on other services ? | 12:57 |
| odyssey4me | gokhan the way things are moving, all endpoints are going versionless | 12:57 |
| odyssey4me | the libraries that magnum uses understand how to discover the version | 12:58 |
| odyssey4me | if magnum is not able to consume a versionless endpoint, then find out from them how to configure it so that it either overrides the endpoint or uses the library to derive it from the versionless endpoint | 12:58 |
| *** udesale has joined #openstack-ansible | 12:58 | |
| odyssey4me | so yes, the issue is with magnum's code, or magnum's configuration | 12:59 |
| gokhan | odyssey4me, ok thanks for your help I guess issue is with magnum's code | 13:02 |
| *** m|y|k has joined #openstack-ansible | 13:11 | |
| *** vnogin has joined #openstack-ansible | 13:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/newton: Implement a proper WSREP check for galera https://review.openstack.org/527684 | 13:16 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron master: Fix tempest tests https://review.openstack.org/527686 | 13:17 |
| odyssey4me | evrardjp good find and fix, just one question in there | 13:20 |
| mhayden | good morning, folks | 13:28 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Create and use a local LXC RPM repo https://review.openstack.org/526739 | 13:31 |
| *** woodard has quit IRC | 13:37 | |
| *** woodard has joined #openstack-ansible | 13:38 | |
| mhayden | hwoarang: thanks for tackling that grub2 issue | 13:41 |
| mhayden | but i had a question or two > https://review.openstack.org/527682 | 13:41 |
| hwoarang | yeah i needed that for suse today and i noticed it didn't work there too | 13:42 |
| hwoarang | let me check | 13:42 |
| hwoarang | mhayden: i moved grub_custom_file to vars/main.yml because it's common across all distros | 13:43 |
| hwoarang | so i dropped it from all the distro files | 13:43 |
| hwoarang | is that what you mean? | 13:43 |
| mhayden | oh, i think i see what's going on now | 13:43 |
| mhayden | LGTM -- let's give this a try | 13:44 |
| *** hw_wutianwei_ has joined #openstack-ansible | 13:47 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_nova master: Handle Pike changes for Nova https://review.openstack.org/525515 | 13:49 |
| gun1x | any ideea why i always get this wenn trying to add an image: "403 Forbidden: It's not allowed to add locations if locations are invisible. (HTTP 403)" ? | 13:54 |
| *** hw_wutianwei has joined #openstack-ansible | 13:55 | |
| bndzor | check glance log gun1x | 14:04 |
| bndzor | possibly https://bugs.launchpad.net/glance/+bug/1595335 ? | 14:04 |
| openstack | Launchpad bug 1595335 in Glance "Add image location fails when show_multiple_locations = false" [Undecided,Confirmed] | 14:04 |
| *** sxc731 has joined #openstack-ansible | 14:08 | |
| openstackgerrit | Merged openstack/openstack-ansible-plugins master: Remove the vars plugin as it's no longer used https://review.openstack.org/527415 | 14:13 |
| *** smatzek has quit IRC | 14:19 | |
| mhayden | cloudnull: interesting LWN post about process tagging and how it can help identify processes in containers -> https://lwn.net/Articles/741261/ | 14:22 |
| mhayden | thought that might be interesting to ya since you did the slice work in systemd | 14:22 |
| evrardjp | mhayden: subscription required! | 14:30 |
| mgariepy | morning everyone | 14:31 |
| evrardjp | morning | 14:32 |
| evrardjp | how is the leap? | 14:32 |
| evrardjp | of faith | 14:32 |
| evrardjp | ;p | 14:32 |
| mgariepy | it works | 14:32 |
| evrardjp | ohoh I just received a notification that my snes mini is waiting for me at a shop! wooot. I hope I will get the chance to go to the shop this week :) | 14:33 |
| mgariepy | now need to upgrade to xenial :) haha | 14:33 |
| evrardjp | hahaha | 14:33 |
| evrardjp | do you have a blog where you can post your war stories? | 14:33 |
| mgariepy | almost updating to an EOL if fine | 14:34 |
| mgariepy | nop no blog. | 14:34 |
| mgariepy | haha | 14:34 |
| mgariepy | went pretty easy. | 14:34 |
| mgariepy | just had to do 2-3 patches to fix stuff ;) | 14:34 |
| mgariepy | haha | 14:34 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Update glance NFS for systemd https://review.openstack.org/526930 | 14:34 |
| *** openstackstatus has quit IRC | 14:36 | |
| *** openstackstatus_ has joined #openstack-ansible | 14:36 | |
| *** openstackstatus has joined #openstack-ansible | 14:37 | |
| *** ChanServ sets mode: +v openstackstatus | 14:37 | |
| *** lbragstad has joined #openstack-ansible | 14:37 | |
| *** openstackstatus_ has quit IRC | 14:37 | |
| *** openstackstatus has quit IRC | 14:38 | |
| *** openstackstatus_ has joined #openstack-ansible | 14:38 | |
| *** openstack has quit IRC | 14:39 | |
| *** openstack has joined #openstack-ansible | 14:42 | |
| *** ChanServ sets mode: +o openstack | 14:42 | |
| *** openstack has quit IRC | 14:43 | |
| *** openstack has joined #openstack-ansible | 14:47 | |
| *** ChanServ sets mode: +o openstack | 14:47 | |
| *** strattao has left #openstack-ansible | 14:49 | |
| *** strattao has joined #openstack-ansible | 14:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Fix automatic log copying in Zuul runs https://review.openstack.org/527701 | 14:50 |
| *** strattao has left #openstack-ansible | 14:51 | |
| *** marst has joined #openstack-ansible | 14:56 | |
| mgariepy | hwoarang, i responded to your comment on : https://review.openstack.org/#/c/525654/ | 14:59 |
| gun1x | can i run apt dist-upgrade on systems running osa ? :D | 14:59 |
| mgariepy | if you want i can remove the file created on that task. | 14:59 |
| hwoarang | mgariepy: what I mean was that if you run this script twice, the file will hold the IPs from the first invocation | 15:00 |
| hwoarang | as well as those from the second run | 15:00 |
| *** hw_wutianwei has quit IRC | 15:00 | |
| hwoarang | not sure if that's a problem really, i just thought to point that out :) but if it mixes ips from different runs then it would be hard to extract info from it | 15:01 |
| *** armaan has quit IRC | 15:01 | |
| *** armaan has joined #openstack-ansible | 15:02 | |
| *** dun has quit IRC | 15:02 | |
| *** sxc731 has quit IRC | 15:03 | |
| *** ianychoi has quit IRC | 15:04 | |
| mgariepy | well i'm not sure it would cause an issue. | 15:04 |
| mgariepy | and this script will only be run once anyway, there are several files created not to repeat steps along the run. | 15:05 |
| gun1x | bndzor: glance log: https://bpaste.net/show/9768a14a5660 | 15:05 |
| *** gouthamr has joined #openstack-ansible | 15:06 | |
| *** udesale has quit IRC | 15:07 | |
| mgariepy | hwoarang, https://github.com/openstack/openstack-ansible-ops/blob/master/leap-upgrades/re-deploy.sh#L37 | 15:07 |
| hwoarang | ok i see what you mean | 15:09 |
| mgariepy | but still the file is not used anywhere so i can just remove it. | 15:09 |
| mgariepy | and it's more for the community because my leap upgrade is done and i will not need it in the future (I really hope) haha | 15:10 |
| hwoarang | :p | 15:12 |
| mgariepy | lol | 15:12 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone master: [DNM] debug why on metal fails on gates and not on my machine. https://review.openstack.org/524954 | 15:15 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [DNM] Build everything on metal https://review.openstack.org/504224 | 15:15 |
| *** dun has joined #openstack-ansible | 15:18 | |
| *** sxc731 has joined #openstack-ansible | 15:19 | |
| openstackgerrit | Kaio Kassiano Moura Oliveira proposed openstack/openstack-ansible-os_monasca-agent master: Initial OSA zuul v3 role jobs https://review.openstack.org/525645 | 15:19 |
| *** smatzek has joined #openstack-ansible | 15:20 | |
| *** SerenaFeng has joined #openstack-ansible | 15:22 | |
| *** SerenaFeng has quit IRC | 15:22 | |
| *** SerenaFeng has joined #openstack-ansible | 15:23 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron master: Update static files https://review.openstack.org/526307 | 15:23 |
| *** hamza21 has quit IRC | 15:24 | |
| gun1x | i found this on the neutron container: /etc/neutron/plugins/ml2/linuxbridge_agent.ini:physical_interface_mappings = flat:eth12,vlan:eth11 | 15:25 |
| gun1x | i can't figure the difference between eth12 and eth11 | 15:25 |
| dmsimard | odyssey4me: so fwiw our not-awesome workaround for the time being was to get a private network and then provide a jump box | 15:29 |
| dmsimard | we'll have to think this one through for the next time | 15:29 |
| mgariepy | gun1x, it's defined in your openstack_user_config.yml | 15:32 |
| mgariepy | gun1x, https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L76-L93 | 15:33 |
| *** wspthr has quit IRC | 15:35 | |
| gun1x | mgariepy: yea but the difference between the vlan and the flat is not clear. however, that is basic openstack knowledge and has nothing to do with OSA. | 15:36 |
| mgariepy | gun1x, vlan interface will be able to have tagged vlan for tenant network | 15:36 |
| mgariepy | for network**9 | 15:36 |
| gun1x | mgariepy: yea, i read something similar online. but you usually have only one subnet on the physical network, so i fail to see how vlan will help if they all end up on the same layer 3 segment :D | 15:36 |
| *** wspthr has joined #openstack-ansible | 15:37 | |
| *** yifei has quit IRC | 15:38 | |
| gun1x | mgariepy: oh, the system is designed so that you can supply the vlan segregation that was already configured on the switches. | 15:38 |
| *** SerenaFeng has quit IRC | 15:38 | |
| mgariepy | yep. | 15:38 |
| gun1x | mgariepy: "disable gateway" means you don't have a L3 gateway on the physical router, right? and if you add the gateway, that will be supplied via DHCP openstack will NOT create an gateway there, right? | 15:39 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/newton: Implement a proper WSREP check for galera https://review.openstack.org/527684 | 15:39 |
| mgariepy | like you can have your external network on vlan XXX and then create the external network in openstack with that vlan for the external traffic. | 15:40 |
| logan- | hwoarang: hey, on the ceph-ansible stuff you were asking about.. here is how they're doing the role breakout: https://github.com/ceph/ceph-ansible/blob/master/contrib/push-roles-to-ansible-galaxy.sh .. I think their CI is broken or the script they use is broken but somehow it is not fully reliable. that push-roles-to-ansible-galaxy.sh is a modified version of a script I wrote for https://git.lstn.net/mirror which is published here: https://gis | 15:40 |
| logan- | t.github.com/Logan2211/528a7fd661e759181875d6edf4dd7c3f ... i have a CI job that just triggers on git changes to upstream ceph-ansible and runs this and have never had any issue. not sure why ceph's thing is so unreliable :/ | 15:40 |
| hwoarang | logan-: so what can we do about it ? :( | 15:41 |
| hwoarang | woud switching to ceph-ansible solve our issues? | 15:41 |
| *** phalmos has joined #openstack-ansible | 15:42 | |
| logan- | it has pros and cons but yes it would workaround this particular issue | 15:42 |
| mgariepy | gun1x, where are you reading that ? | 15:42 |
| logan- | should be possible without too much difficulty | 15:43 |
| logan- | you'll have to modify the osa roles_path I think? | 15:43 |
| gun1x | mgariepy: this is not specified anywhere, that's why i am asking. i tried the following websites: redhat, openstack, ubuntu ... | 15:43 |
| gun1x | mgariepy: all the info is "well if you disable gateway, you don't have a gateway." wow omg that helped, what if i already have an external gateway? :D | 15:44 |
| jamesdenton | gun1x are you talking --no-gateway on subnet creation? | 15:46 |
| gun1x | jamesdenton: i am talking about the "disable gateway" option in horizon, when creating a subnet on a flat network | 15:46 |
| jamesdenton | Ok. By default, when you create a subnet, openstack assigns the first IP of that CIDR as the gateway IP for that subnet | 15:47 |
| gun1x | jamesdenton: it's unclear for me if "disable gateway" means openstack does not create a router, or if it means that it doesn't tell the VMs they already have a gateway on that network | 15:47 |
| jamesdenton | and that IP will be used when pushing DHCP info to clients, or when attaching networks/subnets to routers | 15:47 |
| gun1x | jamesdenton: ok, so it doesn't create a router. | 15:48 |
| logan- | hwoarang: maybe we could advocate IRR upstream there :D | 15:48 |
| jamesdenton | no | 15:49 |
| jamesdenton | err. correct, it does not create a router | 15:49 |
| *** sxc731 has quit IRC | 15:50 | |
| hwoarang | logan-: :) well i think it's worth switching to the monolithinc role. seems upstream has no expectations regarding the split roles | 15:50 |
| hwoarang | i tried to bug them to the IRC channel but nope... | 15:50 |
| *** udesale has joined #openstack-ansible | 15:51 | |
| *** udesale has quit IRC | 15:51 | |
| *** udesale has joined #openstack-ansible | 15:58 | |
| *** Neptu_ has joined #openstack-ansible | 16:00 | |
| evrardjp | odyssey4me: where do I know that a project has subscribed to be follow the requirements process or not? | 16:01 |
| evrardjp | maybe in governance? | 16:01 |
| *** CloudToko has quit IRC | 16:01 | |
| *** Yvan has quit IRC | 16:01 | |
| evrardjp | prometheanfire: ^ | 16:01 |
| *** Yvan has joined #openstack-ansible | 16:01 | |
| *** udesale has quit IRC | 16:01 | |
| odyssey4me | evrardjp https://github.com/openstack/requirements/blob/master/projects.txt | 16:03 |
| evrardjp | thanks | 16:04 |
| odyssey4me | logan- hwoarang yeah, andymccr mentioned that all he does is add the path of /etc/ansible/roles/ceph-ansible to the roles path env var, and tada - everything's awesome | 16:04 |
| evrardjp | odyssey4me: could you remind me why we are in there, while we don't produce anything python related, except maybe our inventory? | 16:05 |
| evrardjp | consuming and producing seems different to me | 16:05 |
| odyssey4me | evrardjp to keep our requirements/test-requirements in sync with all of openstack | 16:05 |
| *** ianychoi has joined #openstack-ansible | 16:06 | |
| odyssey4me | we could switch to no longer using tox, and therefore no longer needing either of those files - and therefore not be part of that process | 16:06 |
| hwoarang | odyssey4me: i will have a look thanks! | 16:07 |
| odyssey4me | the question I have is, why does it pain you? | 16:07 |
| evrardjp | I need to weight in the pros and cons | 16:07 |
| evrardjp | well the tox should indeed not exist for us | 16:07 |
| odyssey4me | if the docs requirements split into their own file, which I think is the plan, then we very likely won't need to use anything else from there any more | 16:07 |
| evrardjp | but that's not really a problem | 16:08 |
| evrardjp | could you tell me more about that? | 16:08 |
| evrardjp | there is a plan to manage requirements inside repos directories like releasenotes/ and docs/ ? | 16:09 |
| odyssey4me | yes, I thought I saw a notification about that being part of the movements for docs | 16:09 |
| evrardjp | so proposal bot will change then | 16:09 |
| odyssey4me | basically eventually releasenotes and docs will all share a common conf file, and a common requirements file | 16:09 |
| evrardjp | oh I see. | 16:10 |
| odyssey4me | and those requirements will split from the python requirements file | 16:10 |
| evrardjp | I understand | 16:10 |
| evrardjp | I like this | 16:10 |
| odyssey4me | yeah, me too | 16:11 |
| odyssey4me | but we have to be patient - it will take a few cycles to get done | 16:11 |
| *** dave-mccowan has quit IRC | 16:11 | |
| odyssey4me | see in https://specs.openstack.org/openstack/docs-specs/specs/pike/os-manuals-migration.html#proposed-change the 'during a later phase' | 16:11 |
| *** vnogin has quit IRC | 16:13 | |
| odyssey4me | hwoarang could you review https://review.openstack.org/527679 (ocata version) given that you approved the newton version? :) | 16:15 |
| evrardjp | oh great more features backport | 16:16 |
| prometheanfire | evrardjp: yep, projects.txt, we have a gate job that checks when people update that they actually subscribe to it | 16:18 |
| evrardjp | thanks | 16:18 |
| evrardjp | prometheanfire: what will be the impact of this future change of plans with the docs? | 16:18 |
| evrardjp | do you expect ppl leaving the process? | 16:19 |
| prometheanfire | I missed something, what? | 16:19 |
| hwoarang | evrardjp: they are not features. They are extended fixes :D | 16:19 |
| evrardjp | hwoarang: haha I heard that before :) | 16:19 |
| prometheanfire | you only want requirements fixes for docs related stuff? | 16:20 |
| evrardjp | prometheanfire: no, I was just curious how all of this will happen in the future | 16:20 |
| odyssey4me | evrardjp no differently | 16:20 |
| evrardjp | I know nothing will change, but there are maybe projects that will say "we don't need this anymore" | 16:21 |
| odyssey4me | currently two files are maintained, in the future there may be more | 16:21 |
| evrardjp | like we could. | 16:21 |
| odyssey4me | why would we do that? | 16:21 |
| openstackgerrit | Merged openstack/openstack-ansible master: Update test tooling for manual testing https://review.openstack.org/527137 | 16:21 |
| prometheanfire | no one has unsubscribed from reqs | 16:21 |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix release-yaml-file-prep.py to include ansible-hardening https://review.openstack.org/503083 | 16:21 |
| odyssey4me | let's not go in circles about the click library again, please | 16:22 |
| prometheanfire | lol | 16:22 |
| odyssey4me | if you would like to subscribe to some changes by the reqs repo, but not others and have a good reason for it, then that should be discussed as an option at the PTG with the teams involved | 16:22 |
| evrardjp | I just noticed we install click as part of our process in gate-check-commit . | 16:22 |
| evrardjp | because we install ara | 16:22 |
| odyssey4me | it may or may not be acceptable, depending on the proposal's presentation | 16:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Fix tempest tests https://review.openstack.org/527686 | 16:22 |
| evrardjp | but let's not talk about that. | 16:23 |
| evrardjp | I am just wondering why are we bothering having things if they are more constraints than solutions. Right now it's more solutions than constraints so we are good. | 16:23 |
| odyssey4me | I'd be happy to actually raise the discussion actually, because both us and infra are using python things outside of g-r and it would be nice if we could not have to do it with hacks | 16:24 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Allow additional parameters for ansible-lint https://review.openstack.org/527645 | 16:24 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Fix release-yaml-file-prep.py to include ansible-hardening https://review.openstack.org/527734 | 16:24 |
| prometheanfire | for osa and infra it's almost as if you have your own external global-constraints, because you don't really do openstack, you manage it | 16:24 |
| evrardjp | odyssey4me: if you like to add that to your to do list, go ahead, I won't prevent you. | 16:24 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Fix release-yaml-file-prep.py to include ansible-hardening https://review.openstack.org/527735 | 16:24 |
| evrardjp | prometheanfire: exactly. | 16:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Fix release-yaml-file-prep.py to include ansible-hardening https://review.openstack.org/527736 | 16:25 |
| evrardjp | and we manage also a proposal bot | 16:25 |
| prometheanfire | ya | 16:25 |
| evrardjp | odyssey4me: don't bother changing release-yaml-file-prep, I am not using it. | 16:26 |
| odyssey4me | evrardjp maybe you aren't, but the next PTL might - or someone else might | 16:26 |
| evrardjp | if there is someone else doing releases, maybe worth doing | 16:26 |
| evrardjp | I expect that I'd give a more recent version of things that work | 16:27 |
| odyssey4me | it's a bit sad that you aren't using the tooling we have - if it doesn't work for you, why aren't you making it work for you? | 16:27 |
| odyssey4me | our git repositories are supposed to be the collective knowledge over time - when we start working in isolation and not sharing, the collective wisdom breaks down | 16:27 |
| evrardjp | odyssey4me: I tried, but when I started this, both you and andymccr were not available, so after a few hours I decided to do my own thing that works better at the end. | 16:27 |
| odyssey4me | and we are doomed to repeat the same mistakes again and again | 16:28 |
| odyssey4me | it works better? | 16:28 |
| evrardjp | for me | 16:28 |
| *** Yvan has quit IRC | 16:28 | |
| odyssey4me | is that why we haven't had static file updates for ages, until you did the patches manually recently? | 16:28 |
| evrardjp | do you really want to go that way on the critics? | 16:29 |
| odyssey4me | and why we've also had a few releases with mismatching versions? | 16:29 |
| odyssey4me | these are all problems which were solved | 16:29 |
| odyssey4me | they're not perfect, but they are waiting to be improved | 16:29 |
| evrardjp | the mismatch of version happened when I used the old tooling | 16:29 |
| odyssey4me | so, I'm asking, please improve what's there and use it | 16:29 |
| *** Yvan has joined #openstack-ansible | 16:29 | |
| odyssey4me | that way we all benefit, and we get consistent results | 16:29 |
| evrardjp | because of a messup I did. That happens. Humar error. | 16:29 |
| evrardjp | the files are broken for me, and you've seen it when we tried a release together. At the end you said, you were NEXT TO ME, to drop it down. | 16:30 |
| evrardjp | doing it again now has the same issues. | 16:30 |
| evrardjp | anyway | 16:30 |
| odyssey4me | uh, dude - no need to shout, no animosity here | 16:30 |
| evrardjp | I am not here to be criticized | 16:30 |
| evrardjp | I am here to help you all | 16:31 |
| odyssey4me | I'm just trying to respond to "don't bother updating that tool" with, well, you should - or delete it. | 16:31 |
| evrardjp | I want to delete it indeed | 16:31 |
| evrardjp | or clean it up | 16:31 |
| evrardjp | keep the parts that matter | 16:31 |
| odyssey4me | don't delete it without replacing it please | 16:31 |
| cloudnull | mornings | 16:32 |
| *** sxc731 has joined #openstack-ansible | 16:32 | |
| prometheanfire | moin | 16:33 |
| Bico_Fino | Hello! | 16:33 |
| evrardjp | hello! | 16:35 |
| evrardjp | cloudnull: Bico_Fino how are things? | 16:35 |
| odyssey4me | well, now that's quite a thing! http://lists.openstack.org/pipermail/openstack-dev/2017-December/125473.html | 16:35 |
| Bico_Fino | not bad and you evrardjp ? | 16:36 |
| *** openstacking_123 has joined #openstack-ansible | 16:36 | |
| evrardjp | okay I guess, is there anything I can do to help you Bico_Fino ? | 16:37 |
| Bico_Fino | to change external_lb_vip_address, just run haproxy playbook again right? | 16:37 |
| *** vnogin has joined #openstack-ansible | 16:38 | |
| *** jvidal has quit IRC | 16:39 | |
| evrardjp | odyssey4me: yay! | 16:39 |
| odyssey4me | evrardjp yeah, I think it's good too actually | 16:40 |
| evrardjp | Bico_Fino: no, because it will be used for the endpoints | 16:40 |
| Bico_Fino | hum | 16:40 |
| Bico_Fino | what else I need to run evrardjp ? | 16:40 |
| evrardjp | You can either update your endpoints manually, or run the playbooks, I think we now update the endpoints | 16:40 |
| cloudnull | evrardjp: good, and you ? | 16:41 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Fix SSL cert distribution tasks https://review.openstack.org/527598 | 16:41 |
| evrardjp | cloudnull: depends on who asks I guess :) | 16:41 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Fix SSL cert distribution tasks https://review.openstack.org/527598 | 16:41 |
| Bico_Fino | thanks for the info evrardjp | 16:42 |
| Bico_Fino | On another subject, it's possible to remove a controller node and add another(Like the compute nodes)? | 16:42 |
| evrardjp | Bico_Fino: I think you can manage the endpoints in the openstack cli | 16:42 |
| evrardjp | if you want, directly on the utility container | 16:42 |
| evrardjp | :p | 16:42 |
| evrardjp | Bico_Fino: that is more tricky | 16:43 |
| evrardjp | you want to reduce galera cluster size then increase it again? | 16:43 |
| evrardjp | Bico_Fino: the alternative would be to recover from a failure | 16:43 |
| cloudnull | evrardjp: https://review.openstack.org/#/c/477017/ what would you like to see split out ? | 16:43 |
| cloudnull | I can break out the contianer tech changes and then add nspawn after ? | 16:44 |
| evrardjp | cloudnull: I was just scared, and I saw a list of "notable changes", so I thought each of them could be a commit :p | 16:44 |
| Bico_Fino | evrardjp want to remove the full hardware and add another as controller, and use the old hardware to be only a compute node | 16:44 |
| Bico_Fino | maybe I should just deploy it again? :P | 16:44 |
| evrardjp | Bico_Fino: can you switch IPs? | 16:44 |
| Bico_Fino | evrardjp yes | 16:45 |
| evrardjp | else you could say your node infra is dead, you remove its mgmt ip, and reconfigure it with a new ip as compute node, cleanup and reinstall as compute | 16:45 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Provide support for SFC deployments https://review.openstack.org/510909 | 16:45 |
| evrardjp | the new node will take the infra node IP, and will get reinstalled to be part of the cluster | 16:45 |
| Bico_Fino | evrardjp bingo. Good idea! | 16:46 |
| evrardjp | "will take" -> "need to be configured with" | 16:46 |
| openstackgerrit | Merged openstack/openstack-ansible master: Remove symlinked git cache functionality https://review.openstack.org/521749 | 16:46 |
| evrardjp | this way you separate the issue in two | 16:46 |
| evrardjp | and I would start with removing the infra and readding it with a new node, then only use the old node as compute | 16:47 |
| evrardjp | but that's just me | 16:47 |
| *** chhavi__ has quit IRC | 16:47 | |
| odyssey4me | wow, touch wood but the gates appear to be much more reliable today than they have been for ages | 16:48 |
| evrardjp | cloudnull: I trust you can do the right thing to make it more digestable :) But if it's not possible, we'll deal with it later, like we usually do :) | 16:48 |
| evrardjp | the last series of changes made the builds faster. | 16:49 |
| odyssey4me | it does appear to be helping, yes | 16:49 |
| evrardjp | a ceph build under 1h20 is good | 16:49 |
| odyssey4me | we're no longer hitting timeouts except on *some( providers | 16:49 |
| evrardjp | I am waiting for all of this to become bare metal | 16:49 |
| evrardjp | to shave 40 minutes. | 16:50 |
| evrardjp | well | 16:50 |
| evrardjp | not ALL | 16:50 |
| odyssey4me | but that will only be one test, surely? | 16:50 |
| evrardjp | but you understand the idea, faster to test | 16:50 |
| odyssey4me | we'll still have other tests using containers, so it will still take time | 16:50 |
| evrardjp | odyssey4me: for now yes, but what if ... ? | 16:50 |
| odyssey4me | of course, it opens new options | 16:51 |
| evrardjp | I agree, but I like the proposition of cloudnull to have more smaller jobs | 16:51 |
| evrardjp | so we could have smaller jobs with containers, larger jobs with bare metal | 16:51 |
| odyssey4me | I'd still prefer to sprinkle those across many repositories, instead of all in one repo | 16:51 |
| *** pbandark1 has joined #openstack-ansible | 16:52 | |
| odyssey4me | if we have a 5% chance of failure, and 20 tests, then we're not far from having at least one of those fail in the check/gate cycle and be in recheck hell | 16:52 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_server master: Update default paths of SSL cert files https://review.openstack.org/527740 | 16:54 |
| *** pbandark has quit IRC | 16:54 | |
| *** pbandark1 is now known as pbandark | 16:54 | |
| cloudnull | I have to agree. the more I work on the gate refactor the more I think having the nova tests for the nova scenarios in the nova role makes the most sense. | 16:55 |
| hwoarang | yeah makes sense | 16:55 |
| evrardjp | Yeah | 16:55 |
| evrardjp | We are just talking about integration here | 16:55 |
| cloudnull | if that means we need to build the jobs out in the integrated repo so the roles can consume them, wonderful | 16:55 |
| evrardjp | I don't think the importance is where we write the jobs | 16:56 |
| cloudnull | all that said, the AIO needs to get smaller. | 16:56 |
| evrardjp | I think what's good is increasing the scenarios coverage while reducing integrated job lengths | 16:56 |
| *** vnogin has quit IRC | 16:56 | |
| evrardjp | more responibilities to roles and more reuse is good | 16:57 |
| evrardjp | as a side product | 16:57 |
| cloudnull | as seen with that PR to refactor the jobs, the suse tests mostly pass in time. | 16:57 |
| *** armaan has quit IRC | 16:57 | |
| cloudnull | but with the current AIO they're timing out due to the 2 hour limit | 16:57 |
| evrardjp | cloudnull: suse is not a timeout offender, suse is generally a quick fail when there is a temporarily issue, which doesn't happen that often anymore | 16:58 |
| *** armaan has joined #openstack-ansible | 16:58 | |
| evrardjp | cloudnull: there are still slower than ubuntu though :) | 16:58 |
| *** vnogin has joined #openstack-ansible | 17:00 | |
| *** chyka has joined #openstack-ansible | 17:01 | |
| hwoarang | suse timesout normally on tempest | 17:02 |
| hwoarang | which is better because it used to timeout way before that in the past | 17:02 |
| cloudnull | have a look at the results for PS 39 | 17:03 |
| cloudnull | https://review.openstack.org/#/c/516002/39 | 17:03 |
| cloudnull | suse is almost all passing | 17:03 |
| hwoarang | as i told mhayden in the past, centos and suse spend double the time buidling the wheel artifacts | 17:03 |
| hwoarang | and i have no clue why we need so much time for centos and suse | 17:03 |
| hwoarang | the rest of the tasks are pretty similar | 17:03 |
| hwoarang | but that wheel building is killing us | 17:03 |
| *** m|y|k has quit IRC | 17:03 | |
| *** Prateek has quit IRC | 17:04 | |
| mhayden | i haven't figured that out either | 17:08 |
| mhayden | it's not a network thing as far as i can tell | 17:08 |
| *** vnogin has quit IRC | 17:08 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_server master: Update default paths of SSL cert files https://review.openstack.org/527740 | 17:12 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Implement a proper WSREP check for galera https://review.openstack.org/527679 | 17:14 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server master: Remove sylinked git cache functionality https://review.openstack.org/521750 | 17:14 |
| evrardjp | hwoarang: what would be nice is to know in which part of the process of the repo build, is that when building the wheel themselves, or when doing the venvs? | 17:17 |
| evrardjp | because else the change that odyssey4me is introducing will help | 17:17 |
| odyssey4me | hwoarang evrardjp well, regardless of the double-time - my next step is to kill most of whe wheel building by default - only build the hweels we *need* and then build the venvs | 17:19 |
| evrardjp | that's hwat I meant. | 17:19 |
| evrardjp | what* | 17:19 |
| evrardjp | thanks for clarifying | 17:19 |
| odyssey4me | hwoarang mhayden evrardjp hmm, one thing that may be worth checking is whether there are wheels pre-build in infra for the other platforms | 17:19 |
| odyssey4me | if not, that might be ok - but the question is then whether we're enabling the use of the pre-built wheels done on ubuntu, for suse/centos | 17:20 |
| odyssey4me | lemme finish what I'm doing here and I'll take a peek | 17:20 |
| evrardjp | thanks odyssey4me | 17:20 |
| mhayden | so if i look at the repo_build wheel results | 17:25 |
| mhayden | centos builds wheels 20 seconds slower | 17:25 |
| mhayden | but i don't think we have many wheels there | 17:25 |
| mhayden | in that role | 17:25 |
| *** dun has quit IRC | 17:27 | |
| hwoarang | no the role itself is fine | 17:29 |
| hwoarang | it's in the AIO builds that takes forever | 17:29 |
| evrardjp | hwoarang: IO related? | 17:29 |
| hwoarang | in the repo_build role all distros are +/- about the same. but in AIO there is a huge difference | 17:29 |
| hwoarang | evrardjp: i dont know | 17:29 |
| mhayden | we do the builds in a container in the repo_build role, too | 17:30 |
| evrardjp | there is one good way to find out, bare metal! | 17:30 |
| mhayden | so i doubt it's lxc being silly | 17:30 |
| evrardjp | oh | 17:30 |
| evrardjp | well it still has its toll | 17:30 |
| hwoarang | odyssey4me: i couldn't figure out whether we use the infra wheel stuff at all but it's been a while since i checked | 17:30 |
| evrardjp | is there anyone willing to help on my bare metal? It's been two weeks I am stuck now. | 17:30 |
| evrardjp | maybe 3. | 17:31 |
| evrardjp | I have all the logs, the best logs. | 17:31 |
| evrardjp | wow ceph in 1h06 today even | 17:32 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 17:33 |
| evrardjp | ping me anytime if you feel up to the challenge! | 17:33 |
| cloudnull | evrardjp: ^ | 17:33 |
| evrardjp | cloudnull: great! | 17:33 |
| cloudnull | first part of the break down | 17:33 |
| cloudnull | opps, missing things. | 17:34 |
| cloudnull | but im pulling the commit in two | 17:34 |
| mhayden | so if i examine a couple of gate jobs, i notice that the wheel building is almost doubled, but building venvs on centos is only 5-10% longer | 17:35 |
| mhayden | so i'm not sure if this is I/O | 17:35 |
| *** cshen has joined #openstack-ansible | 17:35 | |
| mhayden | i wish we had container logs when openstack-ansible runs | 17:36 |
| evrardjp | cloudnull: when splitting it allows a more thorough review | 17:37 |
| evrardjp | thanks! | 17:38 |
| evrardjp | maybe we should push odyssey4me 's changes in the future as a priority then | 17:39 |
| *** dun has joined #openstack-ansible | 17:39 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 17:40 |
| evrardjp | cloudnull: haha | 17:40 |
| cloudnull | ? | 17:41 |
| odyssey4me | mhayden maybe you should do a version of https://github.com/openstack/openstack-ansible-tests/commit/b2a1f4e8792337d6de668f77b1d21da4f4a7cb24 for theintegrated build? | 17:41 |
| odyssey4me | or, perhaps even better, we should try to use the same log collection script for the integrated build and the roles | 17:41 |
| mhayden | true | 17:42 |
| evrardjp | you were almost good on the first commit, you'd have kept the aio name and that commit was good. | 17:42 |
| * mhayden is building a centos vm to see if i can figure out why it can't start LXC containers now :/ | 17:42 | |
| cloudnull | evrardjp: yea, but i needed to include the actual variable changes in the playbooks | 17:43 |
| cloudnull | I'll reset the upgrade to aio | 17:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 17:44 |
| cloudnull | ^ that sets the upgrade jobs back to aio | 17:44 |
| *** germs has joined #openstack-ansible | 17:45 | |
| *** cshen has quit IRC | 17:46 | |
| *** vnogin has joined #openstack-ansible | 17:46 | |
| *** cshen has joined #openstack-ansible | 17:46 | |
| spotz | cloudnull: multiple container techs aka not just LXC anymore? | 17:46 |
| gun1x | i get this after a fresh install: "No valid host was found. There are not enough hosts available." ... both kvm hosts are up and running | 17:47 |
| gun1x | afaik other people encountered this, is there a quick fix? | 17:47 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 17:47 |
| cloudnull | spotz: yes. I'm adding nspawn | 17:47 |
| spotz | cool | 17:47 |
| cloudnull | I added all these bits in this PR https://review.openstack.org/#/c/477017/ | 17:48 |
| cloudnull | but it was asked I break that change up, So the first commit will add the required scafolding and I'll rebase the original to be just the inclusion of nspawn | 17:49 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 17:52 |
| cloudnull | sorry evrardjp, last edit i think | 17:52 |
| *** germs has quit IRC | 17:52 | |
| cloudnull | i believe that's all the bits i can break out from the larger nspawn addition | 17:52 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 17:54 |
| gun1x | nevermind, i am on a good path, i can solve this. hosts don't have kvm capabilities | 17:56 |
| *** mbuil has quit IRC | 17:57 | |
| *** octo-fs has joined #openstack-ansible | 18:04 | |
| *** vnogin has quit IRC | 18:05 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible stable/pike: Avoid using gather_facts as a variable https://review.openstack.org/527753 | 18:09 |
| *** vnogin has joined #openstack-ansible | 18:09 | |
| evrardjp | I am off, see you next year guys! | 18:10 |
| evrardjp | well I will probably be around | 18:10 |
| mhayden | evrardjp: enjoy! | 18:10 |
| evrardjp | but have nice winter holidays and enjoy xmas and all that | 18:11 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 18:11 |
| cloudnull | take care evrardjp | 18:11 |
| cloudnull | see you in a couple weeks | 18:11 |
| evrardjp | if someone got the chance to look at why my bare metal doesn't work I'd be super happy | 18:11 |
| spotz | seeya evrardjp | 18:11 |
| evrardjp | or ping me when (s)he have cycles for anything for the future :) | 18:12 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Create and use a local LXC RPM repo https://review.openstack.org/526739 | 18:16 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add scafolding for multiple container techs https://review.openstack.org/527749 | 18:17 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 18:18 |
| *** cshen has quit IRC | 18:18 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 18:19 |
| cloudnull | ok. sorry for the spam | 18:19 |
| cloudnull | but i believe they're all broken out | 18:19 |
| cloudnull | if folks have time to review those patches it'd be greatly appreciated. | 18:21 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Always template a localhost config when server https://review.openstack.org/527756 | 18:22 |
| *** vnogin has quit IRC | 18:23 | |
| *** Yvan has quit IRC | 18:24 | |
| *** vnogin has joined #openstack-ansible | 18:25 | |
| *** octo-fs has quit IRC | 18:26 | |
| jrosser | cloudnull: have you tried your nspawn stuff with zfs? | 18:26 |
| *** vnogin has quit IRC | 18:26 | |
| mhayden | mgariepy: ah, i ran into that lxc container stop hang today for the first time on master | 18:34 |
| mgariepy | mhayden, yep i noticed that it wasnt fixed.. | 18:35 |
| mgariepy | i wonder what's happening there.. | 18:36 |
| mhayden | so the container is doing something | 18:37 |
| mhayden | because i get booted out over ssh | 18:37 |
| mgariepy | yeah, | 18:37 |
| *** sxc731 has quit IRC | 18:37 | |
| mhayden | so journald is the only process left running in the container | 18:38 |
| mgariepy | I guess there is only one process left in the container | 18:38 |
| mhayden | http://paste.openstack.org/show/628890/ | 18:38 |
| mgariepy | is systemd-sysv installed ? | 18:39 |
| mhayden | yeah | 18:39 |
| mhayden | i'm stracing journald while stopping the container | 18:39 |
| mhayden | all of the cgroups get cleared up, then it goes into epoll_wait | 18:40 |
| mhayden | finally gets a SIGKILL | 18:40 |
| mgariepy | lxc try to stop the container then kill every process after 60sec. | 18:40 |
| mhayden | so it umounts the root fs | 18:41 |
| *** poopcat has joined #openstack-ansible | 18:41 | |
| mhayden | but doesn't get to that "Reached target shutdown" part | 18:41 |
| mgariepy | I wonder when it changed. | 18:42 |
| mgariepy | and why master but not pike or ocata | 18:42 |
| mgariepy | something we changed ? | 18:42 |
| mhayden | no idea | 18:43 |
| mhayden | this is funky deluxe | 18:43 |
| *** electrofelix has quit IRC | 18:43 | |
| mhayden | if i issue the poweroff inside the vm, all is well | 18:43 |
| mhayden | err inside the container | 18:44 |
| *** poopcat1 has joined #openstack-ansible | 18:44 | |
| *** dave-mccowan has joined #openstack-ansible | 18:44 | |
| mhayden | and it's not specific to a certain container | 18:45 |
| mhayden | does it to repo + rabbitmq | 18:45 |
| mgariepy | from what i've seen it all the containers. | 18:46 |
| *** poopcat has quit IRC | 18:46 | |
| *** hamza21 has joined #openstack-ansible | 18:48 | |
| *** germs has joined #openstack-ansible | 18:53 | |
| cloudnull | jrosser: I've not | 18:53 |
| cloudnull | hopefully soon though :) | 18:54 |
| *** stuartgr has quit IRC | 18:54 | |
| *** smatzek has quit IRC | 18:54 | |
| jrosser | cloudnull: hopefully :) our recent boxes have (almost) nothing else | 18:56 |
| cloudnull | The backend should support it, it's just a matter of getting the right testing in | 18:58 |
| cloudnull | once I can get the initial commit through, I'll work on that. | 18:58 |
| jrosser | that would be cool. these servers are small and only have two disks, so for a mirrored setup its somewhat all or nothing one the zfs front | 18:59 |
| cloudnull | the lxc implementation w/ ZFS is working wonderfully though :) | 19:04 |
| odyssey4me | anyone happy to give this backport another review? https://review.openstack.org/#/c/527648/1 | 19:06 |
| cloudnull | I have machinectl, zfs, and lvm all running within the same deployment, the machinectl and zfs backends have been great so far. | 19:06 |
| jrosser | cloudnull: https://github.com/jimsalterjrs/sanoid/blob/master/README.mdõ | 19:06 |
| cloudnull | odyssey4me: looking | 19:06 |
| cloudnull | jrosser: that's cool! | 19:07 |
| cloudnull | are you running that / | 19:08 |
| cloudnull | **? | 19:08 |
| jrosser | im half way through writing a role for it | 19:08 |
| jrosser | been doing that today | 19:08 |
| cloudnull | nice | 19:09 |
| gun1x | ok, so, i fixed virtualization for the hosts but i still get "No valid host was found. There are not enough hosts available." | 19:10 |
| cloudnull | odyssey4me: trade you https://review.openstack.org/#/c/527513/ :) | 19:10 |
| cloudnull | gun1x: is this on an instance build? | 19:10 |
| gun1x | cloudnull: multiple VMs on my home desktop. 2 kvm hosts, 1 controller, 1 cinder, 1 swift | 19:11 |
| gun1x | i can't start the first instance | 19:11 |
| cloudnull | the logs from the nova scheduler should indicate why filters it hit | 19:12 |
| cloudnull | and the compute node logs should should the check-ins w/ the available resources. | 19:12 |
| gun1x | i can grep after requiest ID | 19:12 |
| cloudnull | ...node logs should show the... | 19:12 |
| cloudnull | if the flavor is larger than the reported resources and the overcommit hasn't been set to allow for whatever is hitting the over provisioning you'll get that error | 19:13 |
| mhayden | mgariepy: okay, i'm getting somewhere | 19:14 |
| mhayden | if i install the @Core package group, it shuts off in 2-3 seconds | 19:14 |
| *** sxc731 has joined #openstack-ansible | 19:14 | |
| mhayden | so there's something in there that pleases lxc | 19:14 |
| mgariepy | ok. the other day only systemd-sysv was ok for me.. | 19:14 |
| gun1x | cloudnull: https://bpaste.net/show/5e3e4a82392e | 19:15 |
| gun1x | cloudnull: flavor has 1 cpu, 256 mb ram ... i have 3 cpus per host and 3 GB per host :D | 19:15 |
| *** vnogin has joined #openstack-ansible | 19:16 | |
| mhayden | mgariepy: got meetings now -- will take a look later on | 19:17 |
| mgariepy | hmm | 19:18 |
| mgariepy | digging.. | 19:18 |
| *** smatzek has joined #openstack-ansible | 19:21 | |
| cloudnull | gun1x: Binding failed for port 0cb9dc22-6891-4b0e-92ea-36f1b6b61f45, please check neutron logs for more information. | 19:23 |
| cloudnull | is there an issue with neutron | 19:24 |
| cloudnull | ? | 19:24 |
| gun1x | oh i didn't see that | 19:25 |
| *** openstackstatus has quit IRC | 19:27 | |
| gun1x | cloudnull: there seems to be an error for the falt network i was talking about on the channel before. i will switch to vxlan | 19:27 |
| cloudnull | oh, did you remove the flat network type in the openstack_user_config.yml ? | 19:28 |
| cloudnull | then rerun the os-neutron-playbooks to cleanup the flat network references. | 19:29 |
| gun1x | it failed on vxlan too | 19:30 |
| gun1x | osa-controller_nova_conductor_container-47c69f85/nova/nova-conductor.log:2017-12-13 21:30:02.571 195 ERROR nova.scheduler.utils [req-e4a9c39a-baa6-445b-86f8-d8484b27ec53 470153175f344be7a62d0b855abc32f7 a607ae87d705424f94f275888ac34258 - default default] [instance: 0de1b90e-3b83-4da6-b4f8-eee6a8391585] Error from last host: osa-lxd-1 (node osa-lxd-1): [u'Traceback (most recent call last):\n', u' | 19:31 |
| gun1x | File "/openstack/venvs/nova-16.0.3/lib/python2.7/site-packages/nova/compute/manager.py", line 1829, in _do_build_and_run_instance\n filter_properties)\n', u' File "/openstack/venvs/nova-16.0.3/lib/python2.7/site-packages/nova/compute/manager.py", line 2059, in _build_and_run_instance\n instance_uuid=instance.uuid, reason=six.text_type(e))\n', u'RescheduledException: Build of instance | 19:31 |
| gun1x | 0de1b90e-3b83-4da6-b4f8-eee6a8391585 was re-scheduled: Binding failed for port 0388e530-1832-490b-bbb9-e2e1d3f01407, please check neutron logs for more information.\n'] | 19:31 |
| gun1x | same error on VXLAN | 19:31 |
| cloudnull | what does the neutron logs say? | 19:32 |
| cloudnull | if it's complaining about access to a flat network you'll have to correct the config things before it'll work | 19:32 |
| openstackgerrit | David Wilde proposed openstack/openstack-ansible-specs master: Add spec for ELK stack in openstack-ansible https://review.openstack.org/527769 | 19:34 |
| gun1x | cloudnull: i will look into it | 19:34 |
| cloudnull | d34dh0r53: nice! | 19:36 |
| d34dh0r53 | cloudnull: tys | 19:37 |
| *** vnogin has quit IRC | 19:39 | |
| *** pbandark1 has joined #openstack-ansible | 19:44 | |
| *** pbandark has quit IRC | 19:45 | |
| *** pbandark1 is now known as pbandark | 19:45 | |
| *** Trident has joined #openstack-ansible | 19:47 | |
| *** Trident has quit IRC | 19:47 | |
| gun1x | cloudnull: the error i keep getting is https://bpaste.net/show/dead21c36121 | 19:49 |
| gun1x | for neutron | 19:49 |
| *** Trident has joined #openstack-ansible | 19:55 | |
| *** Trident has quit IRC | 19:55 | |
| *** cshen has joined #openstack-ansible | 19:59 | |
| *** Trident has joined #openstack-ansible | 20:00 | |
| mgariepy | mhayden, diff between master and pike: https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L177 https://github.com/openstack/openstack-ansible/blob/stable/pike/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L182 | 20:02 |
| *** poopcat1 is now known as poopcat | 20:03 | |
| openstackgerrit | tonytan4ever proposed openstack/openstack-ansible-tests master: Force gzip compression when file has a lot of links https://review.openstack.org/527478 | 20:05 |
| *** schwicht has joined #openstack-ansible | 20:06 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Always template a localhost config when server https://review.openstack.org/527756 | 20:07 |
| *** hybridpollo has joined #openstack-ansible | 20:09 | |
| *** chyka has quit IRC | 20:21 | |
| *** chyka has joined #openstack-ansible | 20:21 | |
| *** epalper has quit IRC | 20:22 | |
| mgariepy | mhayden, it's very weird, some are ok others aren't | 20:26 |
| *** openstackstatus has joined #openstack-ansible | 20:26 | |
| *** ChanServ sets mode: +v openstackstatus | 20:26 | |
| mgariepy | on my install I have memcach and utility container that are ""ok"" but others aren't | 20:27 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Update default paths of SSL cert files https://review.openstack.org/527740 | 20:31 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: Remove galera_address from playbook vars https://review.openstack.org/527778 | 20:34 |
| mgariepy | mhayden, start: stop fast, works correctly, wait for the network to be up and then it fails. | 20:37 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_server master: Remove the galera_ssl_address variable https://review.openstack.org/527779 | 20:37 |
| *** mattt has quit IRC | 20:43 | |
| *** neillc has quit IRC | 20:43 | |
| *** charz has quit IRC | 20:43 | |
| *** prometheanfire has quit IRC | 20:43 | |
| *** logan- has quit IRC | 20:43 | |
| *** tobberydberg has quit IRC | 20:43 | |
| *** logan- has joined #openstack-ansible | 20:43 | |
| *** mattt has joined #openstack-ansible | 20:43 | |
| *** charz has joined #openstack-ansible | 20:43 | |
| *** neillc has joined #openstack-ansible | 20:44 | |
| *** gouthamr has quit IRC | 20:45 | |
| *** tobberydberg has joined #openstack-ansible | 20:45 | |
| cloudnull | gun1x: mind sharing your openstack_user_config | 20:52 |
| gun1x | cloudnull: i found that both my br-storage and my vxlan were on 172.29.244.0/22 ... so i reverted everything to snapshot, i am going over the network config again and starting the deployment another time | 20:53 |
| cloudnull | mgariepy: are you finding that the machinectl bits are slowing things down ? | 20:54 |
| cloudnull | gun1x: ok | 20:55 |
| mgariepy | cloudnull, not sure. | 20:55 |
| cloudnull | is this related to the wheel build slowness ? | 20:55 |
| mgariepy | just pointing in every direction to make mhayden work.. lol | 20:55 |
| cloudnull | hahaha | 20:55 |
| mgariepy | cloudnull, nop | 20:55 |
| *** sxc731 has quit IRC | 20:56 | |
| mgariepy | cloudnull, http://logs.openstack.org/33/527333/1/check/openstack-ansible-deploy-aio-centos-7/e63e213/job-output.txt.gz#_2017-12-13_15_54_17_020088 | 20:56 |
| mhayden | if i disable systemd-networkd, i'm good to go | 20:57 |
| mhayden | systemd-networkd and the old sysv network scripts are both coming up at boot | 20:57 |
| mhayden | yeah, if systemd-networkd + network are both enabled, it hangs | 20:58 |
| mgariepy | hmm we might want to keep systemd-networkd and trash old crufty stuff. | 20:58 |
| mhayden | if only one of them are enabled, all is well | 20:58 |
| cloudnull | mhayden: i added that package to the lxc-hosts role a while back, https://github.com/openstack/openstack-ansible-lxc_hosts/commit/6c953712108415f9ba4bbae4a2ebd8dcea568b32 | 20:58 |
| mhayden | but systemd-networkd isn't bringing up the net devices properly | 20:58 |
| cloudnull | maybe that service should just be disabled | 20:58 |
| cloudnull | the hope was to get this in https://review.openstack.org/#/c/523218/ | 20:59 |
| mhayden | i think we've gotta pick one or the other | 20:59 |
| cloudnull | but until i can work on that maybe the package should be removed / service disabled ? | 20:59 |
| mgariepy | new one ! | 20:59 |
| mhayden | either we go full systemd, or we use the old sysv network stuff | 20:59 |
| mhayden | mgariepy: well, if we use networkd, we need to change our configs | 20:59 |
| mgariepy | going systemd would unify network config. | 20:59 |
| cloudnull | ++ | 21:00 |
| cloudnull | I have to put the suse network scripts back to make that pr pas | 21:00 |
| cloudnull | **pass | 21:00 |
| cloudnull | sadly there's no networkd in leap | 21:00 |
| mhayden | cloudnull: perhaps we need to revert the installation of networkd until we can hammer this out | 21:01 |
| cloudnull | ++ | 21:01 |
| mgariepy | ++ | 21:01 |
| mhayden | i'll go push the button | 21:01 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:02 |
| cloudnull | mhayden: mgariepy: reverted | 21:02 |
| *** japestinho has quit IRC | 21:02 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:03 |
| mgariepy | revert my systemd-sysv | 21:03 |
| mhayden | just bumped the commit msg | 21:03 |
| mgariepy | also | 21:03 |
| mhayden | merge conflict ;) | 21:03 |
| mhayden | cloudnull: i'm in a mtg -- can you get networkd + systemd-sysv removed | 21:03 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:04 |
| cloudnull | oh both packages need to be removed ? | 21:04 |
| mgariepy | i though the systemd-sysv was needed. | 21:05 |
| mgariepy | hmm. | 21:05 |
| mgariepy | wait a sec i'll test quickly | 21:05 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:06 |
| cloudnull | done. | 21:06 |
| cloudnull | oh. | 21:06 |
| cloudnull | hahaha | 21:06 |
| cloudnull | welp we can change the pr if needed | 21:06 |
| mgariepy | it's fine without it. | 21:07 |
| *** bndzor has quit IRC | 21:08 | |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:09 |
| *** bndzor has joined #openstack-ansible | 21:09 | |
| mgariepy | fixed it;p | 21:09 |
| *** Pingue_ has quit IRC | 21:09 | |
| mgariepy | have a nice evening everyone | 21:09 |
| *** Pingue has joined #openstack-ansible | 21:10 | |
| gun1x | cloudnull: this is the build i am starting: | 21:10 |
| gun1x | https://bpaste.net/show/f9afa8c2bd0b | 21:10 |
| cloudnull | did we figure anything out regarding the wheel building slowness ? | 21:10 |
| gun1x | last time i removed lxd flag so that i get kvm. but i want to try it out with lxd now | 21:10 |
| cloudnull | gun1x: I'd get rid of line 43:51 | 21:12 |
| cloudnull | the section for flat networks | 21:12 |
| cloudnull | and just make sure br-{mgmt,storage,vlan,vxlan} are on the required hosts. | 21:12 |
| gun1x | cloudnull: why? i want vlan ... | 21:13 |
| cloudnull | vlan is good flat is not | 21:13 |
| gun1x | cloudnull: i have a network with ip 192.168.100.0/24, dedicated for provider network ... so that i test that out too | 21:13 |
| cloudnull | ok. | 21:13 |
| gun1x | cloudnull: why is flat bad? | 21:13 |
| cloudnull | it's not bad | 21:14 |
| cloudnull | it's just causes a lot of confusion | 21:14 |
| gun1x | why is it not good? :D | 21:14 |
| cloudnull | if you have an upstream dhcp service running you'll have conflicts | 21:15 |
| cloudnull | in that setup you'll need eth12 as a veth pair plugged into your br-vlan bridge | 21:15 |
| gun1x | cloudnull: https://bpaste.net/show/ffd4bb101650 | 21:15 |
| gun1x | i do have dhcp on that network | 21:15 |
| gun1x | should i remove dhcp and leave only gateway? | 21:16 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts stable/pike: Create and use a local LXC RPM repo https://review.openstack.org/527786 | 21:16 |
| cloudnull | like this https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/aio_interfaces.cfg#L44-L59 | 21:16 |
| cloudnull | yes for the flat network to work you will need to disable dhcp and only use it for floating ip space. | 21:16 |
| gun1x | cloudnull: the veph pair is there on every host, for the br-vlan network | 21:16 |
| cloudnull | cool! | 21:17 |
| gun1x | cloudnull: so does it make sense to disable vlan and leave flat in this case? | 21:17 |
| cloudnull | vlan won't conflict | 21:17 |
| *** hamza21 has quit IRC | 21:17 | |
| cloudnull | neutron will tag all of the traffic passing through the br-vlan bridge | 21:17 |
| cloudnull | so all of that will be fine. | 21:18 |
| *** threestrands has joined #openstack-ansible | 21:18 | |
| *** threestrands has quit IRC | 21:18 | |
| *** threestrands has joined #openstack-ansible | 21:18 | |
| *** smatzek has quit IRC | 21:20 | |
| *** smatzek has joined #openstack-ansible | 21:20 | |
| *** smatzek has quit IRC | 21:20 | |
| gun1x | cloudnull: ok, than i leave vlan and disable flat, as you set | 21:22 |
| gun1x | so when creating a network from horizon, i have to select vlan, not flat? | 21:22 |
| gun1x | btw, here is how network looks on hosts: https://bpaste.net/show/b5d8d747d460 | 21:22 |
| cloudnull | don't get me wrong, they can co-exist. | 21:22 |
| cloudnull | i've just seen folks run into issues with flat | 21:22 |
| cloudnull | nice! | 21:23 |
| cloudnull | that looks like it'll work great | 21:23 |
| *** pramodrj07 has joined #openstack-ansible | 21:24 | |
| gun1x | ok. are there differences in behaviour between lxd and kvm? | 21:25 |
| *** cshen has quit IRC | 21:27 | |
| *** prometheanfire has joined #openstack-ansible | 21:31 | |
| mhayden | gun1x: network behavior differences? or generally differences between the two? | 21:33 |
| gun1x | mhayden: differences from the point of view of the openstack user ... like, does everything work? like networks, backups, storage and so on ... | 21:34 |
| mhayden | i haven't used lxd quite yet -- i'm trying to remember who put that code in... | 21:34 |
| *** sep_ has joined #openstack-ansible | 21:36 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_monasca-agent master: Initial OSA zuul v3 role jobs https://review.openstack.org/525645 | 21:37 |
| *** sep has quit IRC | 21:38 | |
| gun1x | mhayden: you just convinced me to use kvm instead of lxd | 21:40 |
| mhayden | kvm is heavily tested ;) | 21:40 |
| *** openstacking_123 has quit IRC | 21:43 | |
| gun1x | mhayden: yea i know | 21:44 |
| gun1x | mhayden: but people usually want the new shiny | 21:44 |
| *** openstacking_123 has joined #openstack-ansible | 21:44 | |
| *** openstacking_123 has quit IRC | 21:45 | |
| cloudnull | the only think i can think of, in terms of setup difference, is lxd will need raw images. | 21:46 |
| cloudnull | https://docs.openstack.org/openstack-ansible-os_nova/latest/#compute-driver-compatibility | 21:47 |
| cloudnull | and that's about all that's required to get it going | 21:47 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Revert "Add the systemd-networkd to centos 7 installations" https://review.openstack.org/527783 | 21:48 |
| mhayden | master openstack-ansible should be happy with centos + lxc now | 21:50 |
| mhayden | thanks mgariepy / cloudnull | 21:50 |
| prometheanfire | you mean people use centos? | 21:51 |
| mhayden | :| | 21:52 |
| prometheanfire | :D | 21:52 |
| mhayden | i'm consdering putitng in some patches to set state=present for epel-release/yum-priorities | 21:56 |
| mhayden | because it's not worth checking to see if there's a new one there | 21:56 |
| cloudnull | mhayden: i think that'd be useful | 22:01 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 22:23 |
| *** marst has quit IRC | 22:24 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 22:28 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 22:32 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 22:38 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Reduce console output in gate jobs https://review.openstack.org/527648 | 22:38 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Fix automatic log copying in Zuul runs https://review.openstack.org/527701 | 22:38 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Allows for the VMs preseed mirror to be overridden https://review.openstack.org/527802 | 22:39 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 22:41 |
| *** schwicht has quit IRC | 22:44 | |
| *** pmannidi has joined #openstack-ansible | 22:54 | |
| *** vnogin has joined #openstack-ansible | 22:57 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add networking-sfc repo to repo_packages https://review.openstack.org/525264 | 22:57 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: Move inventory files to folder in root of repo https://review.openstack.org/516032 | 23:04 |
| *** Neptu_ has quit IRC | 23:11 | |
| *** CloudToko has joined #openstack-ansible | 23:12 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_client master: Fix SSL cert distribution tasks https://review.openstack.org/527598 | 23:12 |
| *** dxiri has joined #openstack-ansible | 23:15 | |
| *** Neptu_ has joined #openstack-ansible | 23:20 | |
| *** dave-mccowan has quit IRC | 23:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Remove the galera_ssl_address variable https://review.openstack.org/527779 | 23:23 |
| *** esberglu has quit IRC | 23:30 | |
| gun1x | guys, any reason why the keyston checksum is missing from venvs ? | 23:34 |
| gun1x | TASK [repo_build : Execute the venv build scripts asynchonously] | 23:40 |
| gun1x | this one must be messed up. it doesn't build checksum or .tgz for keystone | 23:40 |
| *** esberglu has joined #openstack-ansible | 23:43 | |
| *** CloudToko has quit IRC | 23:45 | |
| *** CloudToko has joined #openstack-ansible | 23:46 | |
| *** CloudToko has quit IRC | 23:46 | |
| *** esberglu has quit IRC | 23:48 | |
| gun1x | ok,i am stuck | 23:50 |
| gun1x | it seems the list of venvs gets populated here: roles/repo_build/vars/main.yml | 23:51 |
| gun1x | i THINK this points to roles/repo_build/tests/test-vars.yml | 23:51 |
| gun1x | so somewhere here keystone is missing | 23:51 |
| gun1x | i guess | 23:51 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone master: Add MySQL connection SSL support https://review.openstack.org/527798 | 23:54 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!