| *** kstev has joined #openstack-ansible | 00:03 | |
| *** kstev has quit IRC | 00:10 | |
| *** csmart has quit IRC | 00:13 | |
| *** evrardjp[m] has quit IRC | 00:13 | |
| *** acormier has joined #openstack-ansible | 00:24 | |
| *** brad[] has joined #openstack-ansible | 00:25 | |
| *** arijr_ has joined #openstack-ansible | 00:28 | |
| *** acormier has quit IRC | 00:28 | |
| *** arijr has quit IRC | 00:32 | |
| *** foutatoro has quit IRC | 00:32 | |
| *** esberglu has quit IRC | 00:38 | |
| *** csmart has joined #openstack-ansible | 00:44 | |
| xdfil | anyone use magnum? my containers are spamming up the daemon.log with " [Errno 32] Broken pipe " | 00:46 |
|---|---|---|
| xdfil | its also in syslog but to a lesser dgree | 00:47 |
| xdfil | /openstack/venvs/magnum-16.0.7/lib/python2.7/site-packages/eventlet/greenio/base.py | 00:48 |
| *** hamzy_ has quit IRC | 00:50 | |
| *** evrardjp[m] has joined #openstack-ansible | 00:59 | |
| *** acormier has joined #openstack-ansible | 01:03 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Ensure neutron agents & rabbitmq do not restart when upgrading https://review.openstack.org/541320 | 01:05 |
| *** hamzy_ has joined #openstack-ansible | 01:13 | |
| *** weezS has joined #openstack-ansible | 01:24 | |
| *** tasker has joined #openstack-ansible | 01:26 | |
| tasker | how can I instruct OSA to reclone git repositories for `repo-build` playbook? | 01:27 |
| *** acormier has quit IRC | 01:27 | |
| tasker | I've tried deleting facts and the /var/www/repo/openstack/project and /var/www/repo/openstack directories to no avail. | 01:27 |
| *** acormier has joined #openstack-ansible | 01:27 | |
| tasker | when the /var/www/repo/openstack/project folder is gone, there is something else that is instructing OSA to think that it doesn't need to download. | 01:28 |
| *** tux_ has joined #openstack-ansible | 01:28 | |
| *** acormier has quit IRC | 01:32 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Change stable/newton branch to newton-eol https://review.openstack.org/541407 | 01:39 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/newton: Update git_branch for nova/ironic https://review.openstack.org/541406 | 01:39 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/newton: Collect logs to a subfolder properly https://review.openstack.org/541428 | 01:39 |
| *** weezS has quit IRC | 01:40 | |
| *** woodard_ has joined #openstack-ansible | 01:43 | |
| *** woodard has quit IRC | 01:43 | |
| *** chyka has quit IRC | 01:46 | |
| *** acormier has joined #openstack-ansible | 01:55 | |
| *** acormier has quit IRC | 01:58 | |
| *** acormier has joined #openstack-ansible | 01:58 | |
| *** masber has quit IRC | 01:59 | |
| tasker | is there anyway to force OSA to reclone git repositories that doesn't involve destorying the repo containers? | 02:06 |
| *** arijr_ has quit IRC | 02:14 | |
| tasker | great. I've destroyed the containers and now I have other problems. | 02:21 |
| tasker | time to rip ansible open. | 02:21 |
| tasker | s/ansible/OSA | 02:21 |
| cloudnull | tasker: yes. | 02:28 |
| cloudnull | you can pass in an option | 02:28 |
| cloudnull | one sec | 02:28 |
| cloudnull | from here. https://docs.openstack.org/openstack-ansible-repo_build/latest/ | 02:29 |
| cloudnull | see "repo_build_git_reclone" | 02:29 |
| cloudnull | the same can be set for "repo_build_venv_rebuild" and "repo_build_wheel_rebuild" | 02:30 |
| tasker | hehe. . ) | 02:30 |
| tasker | thanks for the info. | 02:30 |
| cloudnull | for the next time :) | 02:30 |
| cloudnull | the nuke from orbit option always works too | 02:30 |
| tasker | not this time. . / | 02:31 |
| cloudnull | oh? | 02:32 |
| tasker | https://dpaste.de/mV1c | 02:32 |
| cloudnull | did you lxc-destroy or run the `openstack-ansible lxc-container-destroy.yml --limit repo_all` | 02:32 |
| tasker | the latter. | 02:32 |
| tasker | the files in `/etc/ansible/roles/lxc_container/vars` are the only ones in /etc I can see that match the critera of //etc/ansible/roles/lxc_container_create/tasks/main.yml#L32 | 02:33 |
| cloudnull | then `openstack-ansible lxc-container-create.yml --limit hosts:repo_all` | 02:33 |
| tasker | not that exact incantation. | 02:33 |
| tasker | openstack-ansible lxc-containers-create.yml --limit "caiah-controller-*_repo_container*" | 02:34 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 02:34 |
| tasker | yours is working just fine. | 02:34 |
| cloudnull | I don't think * works in limits / tags | 02:34 |
| cloudnull | i'm not sure though | 02:34 |
| *** guhcampos_ has quit IRC | 02:35 | |
| *** guhcampos has quit IRC | 02:35 | |
| tasker | it worked for the destroy. . ) | 02:35 |
| tasker | and in other places. as best I understand, it's a minimal reged. | 02:35 |
| tasker | *regex | 02:35 |
| cloudnull | ah. makes sense | 02:36 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Add transaction cleanup to cent prep https://review.openstack.org/541522 | 02:45 |
| *** evrardjp[m] has quit IRC | 02:54 | |
| *** csmart has quit IRC | 02:55 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Fix ansible pipelining https://review.openstack.org/541524 | 02:55 |
| cloudnull | odyssey4me: evrardjp: mhayden: ^ that should help a lot with performance. seems pipelining has been disabled since we switched to 2.4 | 03:00 |
| *** gkadam has joined #openstack-ansible | 03:02 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Clean-up old systemd prep and allow machinctl to grow https://review.openstack.org/527592 | 03:04 |
| tux_ | Anyone aware of this issue? http://paste.openstack.org/show/664421/ | 03:05 |
| *** woodard_ has quit IRC | 03:09 | |
| *** tinwood has quit IRC | 03:20 | |
| *** tinwood has joined #openstack-ansible | 03:21 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/newton: Zuul: Remove project name https://review.openstack.org/538509 | 03:24 |
| tux_ | The error was: 'dict object' has no attribute 'ansible_pkg_mgr'\n\nThe error appears to have been in '/opt/openstack-ansible/playbooks/common-tasks/os-lxc-container-setup.yml': | 03:29 |
| tux_ | my package manager is yum not apt/zypper | 03:38 |
| *** gkadam has quit IRC | 03:40 | |
| *** woodard has joined #openstack-ansible | 03:44 | |
| tux_ | Just reported bug https://bugs.launchpad.net/openstack-ansible/+bug/1747813 | 03:45 |
| openstack | Launchpad bug 1747813 in openstack-ansible "Lxc container restart has no attribute 'ansible_pkg_mgr'" [Undecided,New] | 03:45 |
| *** woodard has quit IRC | 03:48 | |
| *** ianychoi_ has quit IRC | 03:49 | |
| *** ianychoi_ has joined #openstack-ansible | 03:50 | |
| cloudnull | tux_: that normally happens when there are no valid facts for the instance. | 03:51 |
| cloudnull | what playbook was being run ? | 03:51 |
| tux_ | openstack-ansible setup-openstack.yml | 03:52 |
| tux_ | fatal: [aio1_cinder_scheduler_container-0299e6e8]: FAILED | 03:52 |
| tux_ | when you said no valid facts does that means system facter? | 03:54 |
| tux_ | which specific facts it is looking for? | 03:54 |
| *** udesale has joined #openstack-ansible | 03:56 | |
| tux_ | I have comment 103 line for now and running build | 03:56 |
| *** abelur_ has quit IRC | 03:58 | |
| *** abelur has joined #openstack-ansible | 03:58 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add experimental tests of new ansible versions https://review.openstack.org/537302 | 04:03 |
| *** openstackgerrit has quit IRC | 04:04 | |
| *** acormier has quit IRC | 04:05 | |
| *** exodusftw has quit IRC | 04:08 | |
| *** exodusftw has joined #openstack-ansible | 04:15 | |
| cloudnull | tux_: ansible will gather facts before a run | 04:38 |
| cloudnull | however there are times when the facts are stale | 04:38 |
| cloudnull | you can force gathering facts from everywhere using `anisble -m setup hosts` | 04:38 |
| cloudnull | and `ansible -m setup all_containers` | 04:39 |
| tux_ | hmm! | 04:39 |
| tux_ | will try that in next build and see if that fix issue | 04:40 |
| *** openstackgerrit has joined #openstack-ansible | 04:43 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 04:43 |
| tasker | the problem with -e "repo_build_git_reclone=yes" is that it reclones _all_ of the repos. and I just need it to pull in two. | 04:47 |
| tasker | it fails to pull in in the first place and subsequent runs without the above force flag skip right over the repo as if it's already been downloaded. | 04:48 |
| tasker | is there another way to be more selective with what upstream repositories need to be cloned? | 04:48 |
| tasker | what is it in the host that suggest that the clone was successful in the first place? | 04:49 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Change default for sshd to only all keybased auth https://review.openstack.org/539020 | 04:49 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add defaults for lxc container_tech https://review.openstack.org/541315 | 04:50 |
| *** woodard has joined #openstack-ansible | 04:51 | |
| *** woodard has quit IRC | 04:55 | |
| *** dariko has joined #openstack-ansible | 04:57 | |
| tasker | additionally, the "clone git repositories asynchornously" isn't all that asynchronous. even with "-f 22" | 04:58 |
| tasker | this combined with having to reclone all of the repos makes some very long test cycles. | 04:58 |
| *** eumel8 has joined #openstack-ansible | 04:59 | |
| -tux_- TASK [ceph_client : Gather variables for each operating system] ********************************************************************************************************************* | 05:06 | |
| -tux_- Tuesday 06 February 2018 08:41:35 -0500 (0:00:00.008) 1:26:30.930 ****** | 05:06 | |
| -tux_- fatal: [aio1_gnocchi_container-08a39dcf]: FAILED! => {"failed": true, "msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"} | 05:06 | |
| tasker | "Clone git repositories asynchronously -------------- 740.68s" | 05:11 |
| tasker | tux_: I saw that same message earlier today. It went away when I adjusted what my --limit filter was. are you using a --limit when you run your playboom? | 05:14 |
| tux_ | no | 05:14 |
| tux_ | I am not running limit | 05:15 |
| *** poopcat has quit IRC | 05:16 | |
| tasker | hmm. and I don't know why changing the limit on my run had any effect. sorry, I can't help. | 05:16 |
| tux_ | what causing this issue? | 05:17 |
| *** dariko has quit IRC | 05:20 | |
| *** dariko has joined #openstack-ansible | 05:21 | |
| *** gkadam has joined #openstack-ansible | 05:25 | |
| *** taseer1 has joined #openstack-ansible | 05:26 | |
| *** Taseer has quit IRC | 05:26 | |
| *** taseer2 has joined #openstack-ansible | 05:26 | |
| *** taseer1 has quit IRC | 05:30 | |
| *** taseer2 is now known as Taseer | 05:30 | |
| *** dariko has quit IRC | 05:36 | |
| *** olivierbourdon38 has joined #openstack-ansible | 05:41 | |
| *** jbadiapa has joined #openstack-ansible | 05:42 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 05:42 |
| *** tux_ has quit IRC | 05:44 | |
| *** gkadam_ has joined #openstack-ansible | 05:49 | |
| *** dariko has joined #openstack-ansible | 05:50 | |
| *** gkadam has quit IRC | 05:51 | |
| *** csmart has joined #openstack-ansible | 06:13 | |
| *** olivierbourdon38 has quit IRC | 06:23 | |
| *** evrardjp[m] has joined #openstack-ansible | 06:26 | |
| *** woodard has joined #openstack-ansible | 06:47 | |
| *** olivierbourdon38 has joined #openstack-ansible | 06:50 | |
| *** woodard has quit IRC | 06:51 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-galera_client master: RedHat: Ensure that conflicting MariaDB packages are removed https://review.openstack.org/541581 | 06:59 |
| *** udesale has quit IRC | 07:02 | |
| *** efaf has joined #openstack-ansible | 07:05 | |
| *** udesale has joined #openstack-ansible | 07:05 | |
| *** udesale has quit IRC | 07:06 | |
| *** udesale has joined #openstack-ansible | 07:06 | |
| *** acormier has joined #openstack-ansible | 07:06 | |
| jafeha__ | good morning | 07:06 |
| *** jafeha__ is now known as jafeha | 07:06 | |
| *** acormier has quit IRC | 07:10 | |
| *** jbadiapa has quit IRC | 07:15 | |
| tasker | hello. | 07:15 |
| *** efaf has quit IRC | 07:15 | |
| *** sar has joined #openstack-ansible | 07:15 | |
| *** olivierbourdon38 has quit IRC | 07:16 | |
| *** SmearedBeard has joined #openstack-ansible | 07:20 | |
| *** sar has quit IRC | 07:22 | |
| *** taseer1 has joined #openstack-ansible | 07:28 | |
| *** Taseer has quit IRC | 07:29 | |
| *** taseer2 has joined #openstack-ansible | 07:29 | |
| *** sar has joined #openstack-ansible | 07:30 | |
| *** taseer2 is now known as Taseer | 07:30 | |
| *** taseer1 has quit IRC | 07:33 | |
| *** armaan has joined #openstack-ansible | 07:36 | |
| *** SmearedBeard has quit IRC | 07:40 | |
| *** olivierbourdon38 has joined #openstack-ansible | 07:46 | |
| *** woodard has joined #openstack-ansible | 07:47 | |
| *** SmearedBeard has joined #openstack-ansible | 07:47 | |
| *** udesale_ has joined #openstack-ansible | 07:48 | |
| *** portante has quit IRC | 07:50 | |
| *** udesale has quit IRC | 07:51 | |
| *** lhinds has quit IRC | 07:51 | |
| *** woodard has quit IRC | 07:51 | |
| evrardjp | good morning | 07:53 |
| jafeha | good morning evrardjp | 07:56 |
| prometheanfire | moin | 07:56 |
| tasker | what generates "/var/www/repo/os-releases/14.2.8/ubuntu-14.04-x86_64/requirements_absolute_requirements.txt"? it's pulling in a version of "python-ldap" that needs a conflicting version of "pyasn1". is there a way to modify the requirements file? | 07:59 |
| *** portante has joined #openstack-ansible | 08:02 | |
| *** lhinds has joined #openstack-ansible | 08:04 | |
| tasker | http://paste.openstack.org/show/664644/ | 08:07 |
| *** DanyC has joined #openstack-ansible | 08:08 | |
| openstackgerrit | sun cheng proposed openstack/openstack-ansible-os_neutron master: Update the url links in doc file of openstack-ansible-os_neutron https://review.openstack.org/541603 | 08:10 |
| *** DanyC_ has joined #openstack-ansible | 08:10 | |
| tasker | is there a way to see what set the requirement of python-ldap to something that isn't officially releaesed yet? https://www.python-ldap.org/en/latest/installing.html#installing | 08:10 |
| *** peri has joined #openstack-ansible | 08:11 | |
| *** DanyC has quit IRC | 08:13 | |
| *** nurdie has quit IRC | 08:13 | |
| *** mbuil has joined #openstack-ansible | 08:14 | |
| *** ianychoi_ has quit IRC | 08:14 | |
| *** nurdie has joined #openstack-ansible | 08:14 | |
| *** SmearedBeard has quit IRC | 08:15 | |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: [WIP] Ceph RadosGW integration https://review.openstack.org/517856 | 08:17 |
| *** santacloud has joined #openstack-ansible | 08:17 | |
| *** SmearedBeard has joined #openstack-ansible | 08:17 | |
| *** ianychoi_ has joined #openstack-ansible | 08:19 | |
| hwoarang | good morning | 08:20 |
| *** threestrands has quit IRC | 08:22 | |
| *** ianychoi_ has quit IRC | 08:28 | |
| *** sar has quit IRC | 08:28 | |
| *** sar has joined #openstack-ansible | 08:29 | |
| evrardjp | tasker: repo build | 08:29 |
| evrardjp | please update your repo | 08:29 |
| *** ianychoi_ has joined #openstack-ansible | 08:30 | |
| evrardjp | tasker: please update your sources too | 08:30 |
| evrardjp | this has changed a few months ago, it's outside our control | 08:30 |
| evrardjp | we adapted our code to make it work | 08:30 |
| evrardjp | hwoarang jafeha how are things? | 08:30 |
| hwoarang | so far so good. how about you? :) | 08:32 |
| evrardjp | okayish | 08:33 |
| hwoarang | i hope they transition to 'ok' soon :) | 08:33 |
| jafeha | same here. deadlines incoming :) | 08:34 |
| tasker | evrardjp: i will do so. though, i just built the repo this afternoon. | 08:35 |
| tasker | *repo containers. | 08:35 |
| *** SmearedBeard has quit IRC | 08:35 | |
| tasker | which repo are you talking about? OSA repo container, openstack-ansible git repository, ubuntu package repo, or something else? | 08:37 |
| *** pcaruana has joined #openstack-ansible | 08:38 | |
| evrardjp | tasker: the repo has to be rebuilt, but your sources are not updated | 08:39 |
| evrardjp | so update your openstack-ansible sources, rebootstrap ansible, and re-run all the playbooks | 08:39 |
| evrardjp | that should fix your problem | 08:40 |
| evrardjp | I mean that should fix the pyldap thing | 08:40 |
| evrardjp | hwoarang: ahah. I am just a little sick, so yeah I hope too! | 08:41 |
| *** gillesMo has joined #openstack-ansible | 08:41 | |
| hwoarang | ah get well soon then :) | 08:42 |
| tasker | and by sources, are you meaning the ubuntu package /etc/apt/sources.list or the OSA git repo not being at 14.2.13? | 08:43 |
| tasker | I'm doing the latter now | 08:43 |
| tasker | just trying to fully grok what you're saying. | 08:43 |
| *** woodard has joined #openstack-ansible | 08:44 | |
| *** ianychoi_ has quit IRC | 08:45 | |
| *** hamza21 has joined #openstack-ansible | 08:46 | |
| *** ianychoi_ has joined #openstack-ansible | 08:46 | |
| *** SmearedBeard has joined #openstack-ansible | 08:48 | |
| *** woodard has quit IRC | 08:49 | |
| *** santacloud has quit IRC | 08:52 | |
| *** SmearedBeard has quit IRC | 08:54 | |
| evrardjp | tasker: the osa git repo is not up to date, and so are the roles | 08:57 |
| evrardjp | haha sorry for being obscure | 08:57 |
| evrardjp | if you are updating your osa git repo to a little later in your branch, or to your last tag, you'd then have to bootstrap-ansible again (to get the latest roles), and then you can redeploy | 08:58 |
| *** SmearedBeard has joined #openstack-ansible | 08:58 | |
| evrardjp | in your redeploy, the repo build will get rebuilt, and everything would work again | 08:59 |
| tasker | no worries. thanks for the help and explanation. | 08:59 |
| *** rpittau has quit IRC | 09:01 | |
| *** gameon has joined #openstack-ansible | 09:15 | |
| *** sar has quit IRC | 09:22 | |
| *** olivierbourdon38 has quit IRC | 09:28 | |
| *** ivveh has quit IRC | 09:29 | |
| *** ivveh_ has joined #openstack-ansible | 09:29 | |
| *** SmearedBeard has quit IRC | 09:29 | |
| odyssey4me | tasker that python-ldap requirement was fixed quite some time ago - are you using an old tag? | 09:39 |
| odyssey4me | tasker if you switch to a new tag for the OSA repo, you must ensure that you re-bootstrap ansible to update the roles | 09:39 |
| odyssey4me | tasker see https://docs.openstack.org/openstack-ansible/latest/user/minor-upgrade.html | 09:40 |
| tasker | odyssey4me: evrardjp identified the same situation. . ) I was using 14.2.8 and have updated OSA to 14.2.13, deleted the old "ansible-runtime", re-ran "scripts/bootstrap-ansible.sh", and am running "repo-build' right now. | 09:42 |
| tasker | thanks for the confirmation. . ) | 09:42 |
| odyssey4me | evrardjp I see centos is killing us due to that yum repo url change - your thoughts on whether we should let https://review.openstack.org/541419 through as-is, or whether the URL should be left using the same repo provider? | 09:42 |
| hwoarang | i am curious where the new url is coming from | 09:45 |
| *** woodard has joined #openstack-ansible | 09:46 | |
| odyssey4me | yeah, it would have been nice for them to implement some backwards compatibility! | 09:49 |
| *** woodard has quit IRC | 09:51 | |
| hwoarang | it's interesting that it's only centos that changed | 09:52 |
| hwoarang | sles, rhel etc are the same | 09:53 |
| hwoarang | maybe they made a mistake | 09:53 |
| *** hw_wutianwei has quit IRC | 09:54 | |
| odyssey4me | there are several patches to improve tests in stable branches I'd like to get through asap, hwoarang evrardjp if you could look through https://review.openstack.org/#/q/owner:jesse-pretorius+project:%22%255Eopenstack/(monitorstack%257Copenstack-ansible).*%22+status:open+label:Verified%253E%253D1%252Czuul+NOT+label:Workflow%253C%253D-1+NOT+label:Code-Review%253C%253D-1 I'd appreciate it | 09:56 |
| *** hw_wutianwei has joined #openstack-ansible | 09:56 | |
| *** pcaruana has quit IRC | 09:57 | |
| hwoarang | on it | 09:58 |
| odyssey4me | thanks hwoarang | 09:58 |
| odyssey4me | I have many more, but the centos builds are all failing due to mariadb :/ | 09:59 |
| odyssey4me | perhaps you could peek at https://review.openstack.org/541443 too, even though it's failing centos - with another vote, I can port it back to where I need it (ocata) to debug the upgrade test better | 10:00 |
| hwoarang | well maybe we could remove centos voting quickly until major is up and sort out the url thing? | 10:00 |
| *** sar has joined #openstack-ansible | 10:00 | |
| odyssey4me | well, if we all agree using the same provider as last time is better, then we could just change up his patches | 10:01 |
| odyssey4me | he's changed it from yum.mariadb.org to use logan-'s company mirror | 10:01 |
| odyssey4me | which is ok, except that there doesn't appear to be any reason for that and logan- has also not been able to respond to whether it's ok... so I'd rather we stick to the previous provider which is apparently OVH | 10:02 |
| hwoarang | personally i would prefer the mariadb one since using another mirror doubles the amount of checks we need to do. which one is broken at $random time? upstream? limestone? | 10:02 |
| odyssey4me | overall I think I'd much rather get an infra mirror in place though | 10:02 |
| hwoarang | odyssey4me: re https://review.openstack.org/#/c/541263/ did you use a script to get the releasenote? i am just curious | 10:02 |
| odyssey4me | nope, I made the releasenote myself | 10:03 |
| hwoarang | ah ok | 10:03 |
| hwoarang | should that release note be part of the role git repo itself? | 10:04 |
| odyssey4me | it could, and probably should - happy to adjust that if you feel that's better | 10:05 |
| hwoarang | i was just curious | 10:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Update role test log collection https://review.openstack.org/541457 | 10:07 |
| hwoarang | because if i understand correctly release notes for roles go to the respective git repo and then copied to the integration repo | 10:07 |
| hwoarang | anyway doesn't matter much | 10:07 |
| odyssey4me | yeah, I can also add the reno to the role retrospectively after it's merged into the integrated repo | 10:08 |
| hwoarang | it's just my curiosity :) | 10:08 |
| odyssey4me | generally a reno should go with the related code | 10:08 |
| odyssey4me | sometimes a reno belongs in the integrated repo | 10:08 |
| odyssey4me | lemme push a patch to get that into the os_nova role too, seeing as it should actually be there - the integrated buildpatch doesn't need to change for that to happen | 10:09 |
| *** sar has quit IRC | 10:10 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/pike: Add reno for novnc/spice known issue https://review.openstack.org/541648 | 10:12 |
| *** pcaruana has joined #openstack-ansible | 10:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Add reno for novnc/spice known issue https://review.openstack.org/541651 | 10:15 |
| *** hw_wutianwei has quit IRC | 10:17 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Improve console install and restart services appropriately https://review.openstack.org/540329 | 10:18 |
| *** hw_wutianwei has joined #openstack-ansible | 10:18 | |
| odyssey4me | added the reno to that one seeing as it hasn't merged yet - made it assume that it only gets pulled in on the next sha bump, given that the newton->ocata upgrade is failing right now, which I'm working on | 10:19 |
| *** ianychoi_ has quit IRC | 10:21 | |
| *** ianychoi_ has joined #openstack-ansible | 10:24 | |
| tasker | do I need to pass --forks in order to take advantage of some of the asynchronous tasks? I tried that earlier but OSA still looked like it was doing things synchronously. | 10:25 |
| odyssey4me | tasker nope, async tasks in ansible aren't very asynchronous in the way they execute | 10:26 |
| odyssey4me | the only way to actually do them asynchronously is this pattern: https://github.com/openstack/openstack-ansible-repo_build/blob/master/tasks/repo_clone_git.yml#L34-L64 | 10:27 |
| odyssey4me | but unfortunately that pattern also means that if the task fails, there's no retry on the task... so async tasks aren't very failure tolerant | 10:27 |
| *** nikeshm has joined #openstack-ansible | 10:27 | |
| odyssey4me | hwoarang could your please review the reno's :) https://review.openstack.org/#/q/Ie48ce3fe62f91298c2beb03f5fa09a4b3a8e12e4 | 10:28 |
| tasker | what's a reno? | 10:29 |
| odyssey4me | tasker 'release note' = reno, due to the tool name: https://docs.openstack.org/reno/latest/ | 10:29 |
| tasker | ah! | 10:29 |
| tasker | ok, next one. is "oslo" short for anything? | 10:30 |
| odyssey4me | oslo is the name space for all common libraries in openstack-land | 10:30 |
| odyssey4me | https://docs.openstack.org/project-team-guide/oslo.html | 10:30 |
| nikeshm | hi | 10:31 |
| nikeshm | i am referring https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/deploymenthost.html | 10:31 |
| tasker | ah, neat! | 10:31 |
| tasker | ok. I'll let you get back to whatever more important things you were doing. . ) | 10:31 |
| nikeshm | under "Configure the operating system" | 10:31 |
| nikeshm | "There is future work planned to create proper firewall rules for OpenStack services in OpenStack-Ansible deployments. Until that work is complete, deployers must maintain their own firewall rulesets or disable the firewall entirely." | 10:32 |
| nikeshm | what are firewall rulesets | 10:32 |
| nikeshm | which we can maintain | 10:32 |
| nikeshm | with keeping firewall enable | 10:32 |
| evrardjp | odyssey4me: https://review.openstack.org/#/c/537015/ has a merge conflict, fyi | 10:32 |
| odyssey4me | nikeshm if you want to implement any host firewalling, that's on you to implement and maintain | 10:33 |
| odyssey4me | evrardjp yeah, I'm not certain that patch is worth progressing in this cycle... I'm thinking of rather switching it all up as per the spec next cycle instead of doing more churn now | 10:33 |
| odyssey4me | but, maybe I'll change my mind ;) | 10:34 |
| evrardjp | odyssey4me: I am not sure what you meant about the url on https://review.openstack.org/#/c/541419/1 | 10:34 |
| evrardjp | are you afraid of a security issue? | 10:34 |
| odyssey4me | evrardjp no, we're changing SP's | 10:34 |
| nikeshm | odyssey4me : is there any link which suggests firewall rules need to be added | 10:35 |
| evrardjp | you mean we are relying on logan's infra instead of upstream? | 10:35 |
| odyssey4me | from yum.mariadb.org -> mirror.lstn.net with no justification made | 10:35 |
| evrardjp | ok | 10:35 |
| evrardjp | yeah | 10:35 |
| evrardjp | I don't like it too, I'd rather we stay close to an upstream source, and on top of that, we won't take logan- 's BW :D | 10:35 |
| odyssey4me | nikeshm not to my knowledge - there may be something in general openstack documentation, and I think there is some in ours too now that I think of it - hang on | 10:35 |
| evrardjp | odyssey4me: commented | 10:36 |
| odyssey4me | nikeshm there is an entire security guide https://docs.openstack.org/security-guide/ | 10:37 |
| evrardjp | odyssey4me: and fyi changing sp is what I could consider as a possible security issue :) | 10:37 |
| odyssey4me | evrardjp my concern there was on two fronts - one, the change without justification as to why, then second the use of logan-'s mirror without him giving the OK that we're going to send a ton of traffic his way! | 10:38 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Add reno for novnc/spice known issue https://review.openstack.org/541651 | 10:39 |
| evrardjp | ok, I just added the security concerns on top of your two then! :p | 10:39 |
| evrardjp | I see | 10:39 |
| evrardjp | well, we are in sync, good! | 10:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client master: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541419 | 10:40 |
| *** udesale_ is now known as udesale | 10:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541417 | 10:41 |
| nikeshm | odyssey4me : any input on SELINUX rule | 10:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: Add reno for novnc/spice known issue https://review.openstack.org/541648 | 10:42 |
| odyssey4me | nikeshm no, I have no input on host firewall rules at all - within OSA there is some implementations for selinux, but mgariepy and mhayden would be your experts in that space | 10:42 |
| odyssey4me | evrardjp I modified the patches as mentioned in my comment - we can't wait and need a fix in ASAP | 10:43 |
| nikeshm | odyssey4me : ok thansk | 10:45 |
| nikeshm | thanks | 10:45 |
| hwoarang | odyssey4me: evrardjp: the galera url thing prob needs backport all the way back to newton :( | 10:46 |
| odyssey4me | hwoarang yep | 10:47 |
| odyssey4me | well, to ocata - newton has no centos gates I don't think | 10:47 |
| *** SmearedBeard has joined #openstack-ansible | 10:47 | |
| hwoarang | true | 10:47 |
| odyssey4me | I'd honestly prefer to remove the centos gates from ocata too, given that the only two contributors brave enough to have ever used it are now running pike+ | 10:48 |
| odyssey4me | and also given that the ocata implementation was *very* experimental, and doesn't appear to work very well in tests | 10:48 |
| evrardjp | odyssey4me: we should probably remove the old urls though | 10:49 |
| evrardjp | because upgrade's gonna fail | 10:49 |
| evrardjp | or be slow | 10:49 |
| evrardjp | more likely slow than fail | 10:49 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Add transaction cleanup to cent prep https://review.openstack.org/541522 | 10:49 |
| *** woodard has joined #openstack-ansible | 10:50 | |
| odyssey4me | evrardjp hwoarang yeah, no worries - I'll port them back to cata given that I'm already doing work on ocata anyway to resolve the os_nova upgrade failure | 10:50 |
| odyssey4me | *ocata | 10:50 |
| evrardjp | we don't gate centos on N | 10:50 |
| *** sar has joined #openstack-ansible | 10:51 | |
| evrardjp | odyssey4me: mmm | 10:52 |
| evrardjp | the thing is | 10:52 |
| evrardjp | I am pretty sure we should clean up after ourselves, and remove the old url | 10:52 |
| evrardjp | else only greenfields would be okay | 10:52 |
| evrardjp | so I'd rather we not rush this | 10:52 |
| odyssey4me | evrardjp I don't think that the yum config works the same as the apt config | 10:53 |
| evrardjp | getting the pattern right avoid works | 10:53 |
| evrardjp | broken mirrors on yum is even more annoying | 10:53 |
| odyssey4me | I think the yum config replaces whatever's there, not like the apt config which leaves the old thing there. | 10:53 |
| evrardjp | interesting | 10:53 |
| evrardjp | I'll double check | 10:53 |
| *** woodard has quit IRC | 10:54 | |
| *** SmearedBeard has quit IRC | 11:02 | |
| evrardjp | odyssey4me: double checked, ok. backport ready | 11:02 |
| evrardjp | at least according to master conditions | 11:02 |
| *** ianychoi_ has quit IRC | 11:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/ocata: tests: test-keystone-upgrades.sh: Remove the exit trap https://review.openstack.org/541469 | 11:05 |
| *** ianychoi_ has joined #openstack-ansible | 11:07 | |
| *** pcaruana has quit IRC | 11:09 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/ocata: tests: test-neutron-upgrades.sh: Remove the exit trap https://review.openstack.org/541472 | 11:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541417 | 11:11 |
| odyssey4me | hwoarang evrardjp ^ updated to include the dependency on the client patch | 11:11 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/ocata: tests: test-cinder-upgrades.sh: Remove the exit trap https://review.openstack.org/541465 | 11:13 |
| *** nikeshm has quit IRC | 11:15 | |
| tasker | tah! you all have a wonderful day. | 11:21 |
| *** tasker has quit IRC | 11:21 | |
| *** ivveh has joined #openstack-ansible | 11:24 | |
| *** ivveh_ has quit IRC | 11:25 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_hosts master: Update the UCA repo configured to queens https://review.openstack.org/541680 | 11:28 |
| *** udesale has quit IRC | 11:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/pike: tests: test-swift-upgrades.sh: Remove the exit trap https://review.openstack.org/541466 | 11:32 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/ocata: tests: test-swift-upgrades.sh: Remove the exit trap https://review.openstack.org/541467 | 11:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Remove placement_database related options/code https://review.openstack.org/541685 | 11:39 |
| odyssey4me | hwoarang evrardjp andymccr ^ just for funsies, that needs to go all the way back to ocata :/ | 11:39 |
| *** stuartgr has joined #openstack-ansible | 11:41 | |
| *** ivveh has quit IRC | 11:42 | |
| evrardjp | meh? | 11:45 |
| evrardjp | I thought this was the placement api | 11:46 |
| odyssey4me | nope | 11:46 |
| evrardjp | we have duplicate things then? | 11:46 |
| evrardjp | interesting | 11:46 |
| odyssey4me | there is only actually the 'nova' and 'nova_api' database that's used | 11:46 |
| *** woodard has joined #openstack-ansible | 11:48 | |
| odyssey4me | evrardjp the question is - do we want to clean up that database when upgrading? or should we just leave it alone? | 11:51 |
| odyssey4me | I think I'll just add a reno informing deployers that it has no function and can be removed. | 11:52 |
| *** woodard has quit IRC | 11:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Remove placement_database related options/code https://review.openstack.org/541685 | 11:57 |
| *** SmearedBeard has joined #openstack-ansible | 11:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Remove placement_database related options/code https://review.openstack.org/541691 | 11:58 |
| *** ianychoi_ has quit IRC | 12:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Remove placement_database related options/code https://review.openstack.org/541691 | 12:01 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-galera_client master: RedHat: Ensure that conflicting MariaDB packages are removed https://review.openstack.org/541581 | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Remove placement_database related options/code https://review.openstack.org/541692 | 12:02 |
| evrardjp | odyssey4me: yeah that's fine | 12:06 |
| *** Smeared_Beard has joined #openstack-ansible | 12:06 | |
| *** SmearedBeard has quit IRC | 12:07 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Update all SHAs for 16.0.8 https://review.openstack.org/540605 | 12:12 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Update all SHAs for 14.2.15 https://review.openstack.org/540789 | 12:12 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Update all SHAs for 15.1.16 https://review.openstack.org/540777 | 12:12 |
| openstackgerrit | Merged openstack/openstack-ansible master: Add defaults for lxc container_tech https://review.openstack.org/541315 | 12:12 |
| *** pcaruana has joined #openstack-ansible | 12:13 | |
| *** dave-mcc_ has joined #openstack-ansible | 12:15 | |
| *** Smeared_Beard has quit IRC | 12:17 | |
| *** mbuil has quit IRC | 12:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client master: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541419 | 12:29 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone stable/pike: Use the good configuration directive for memcache servers https://review.openstack.org/540931 | 12:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client stable/pike: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541698 | 12:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client stable/ocata: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541699 | 12:30 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone stable/pike: Use the good configuration directive for memcache servers https://review.openstack.org/540931 | 12:31 |
| evrardjp | I hate this series of jinja patches for keystone | 12:34 |
| evrardjp | I have the impression it's the best for forward backwards compatibility though, and have the right thing done in Q onwards. | 12:34 |
| *** mbuil has joined #openstack-ansible | 12:35 | |
| evrardjp | odyssey4me: does newton need a change too? | 12:36 |
| odyssey4me | evrardjp yeah, it's complicated - but ideally we want a working cache mechanism for anything we're delivering | 12:36 |
| evrardjp | oh no it's just cento | 12:36 |
| odyssey4me | and yes, those fixes to get caching working again will need to go all the way back to newton | 12:36 |
| evrardjp | no that's not what I meant | 12:37 |
| odyssey4me | oh - you mean centos... well, I couldn't do a clean pick and we have no tests for centos in newton so I don't think it's important | 12:37 |
| odyssey4me | for newton we only had the infra done | 12:37 |
| evrardjp | I meant for the galera things | 12:37 |
| evrardjp | http://mirror.rackspace.com/mariadb/repo/10.0/ubuntu/dists/xenial/main/ | 12:37 |
| evrardjp | I was afraid that kind of urls was replicated for all those mirrors | 12:37 |
| odyssey4me | ubuntu's still fine as far as I know? | 12:38 |
| evrardjp | but yeah, because that part of the folder is handled by apt | 12:38 |
| evrardjp | woot | 12:38 |
| evrardjp | I was afraid we're gonna get this kind of issues for ubuntu in a few days. | 12:39 |
| evrardjp | and break newton | 12:39 |
| evrardjp | I wonder why I focused on newton though, it would have broken everything. | 12:39 |
| evrardjp | brain brain brain! | 12:39 |
| evrardjp | I should get some sleep | 12:39 |
| odyssey4me | haha, yeah - I ad too much spinning in my mind last night - couldn't sleep | 12:40 |
| odyssey4me | I think I only fell asleep at around 4am :/ | 12:40 |
| *** acormier has joined #openstack-ansible | 12:40 | |
| evrardjp | wow that's bad. | 12:40 |
| evrardjp | I am sick though, so I slept like a baby yesterday, but still aren't fine today :( | 12:40 |
| evrardjp | will check on the placement thing | 12:41 |
| evrardjp | needs to know how it works then | 12:41 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Update os_nova role SHA https://review.openstack.org/541263 | 12:42 |
| openstackgerrit | Merged openstack/openstack-ansible master: Update ansible to the latest stable https://review.openstack.org/541381 | 12:42 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Update os_nova and repo_build role SHA's https://review.openstack.org/541259 | 12:42 |
| odyssey4me | the placement api sits on the nova_api database | 12:42 |
| evrardjp | ok | 12:43 |
| *** hw_wutianwei has quit IRC | 12:43 | |
| evrardjp | and? | 12:43 |
| evrardjp | I thought it was the plan to have it a separate thing | 12:44 |
| Miouge | I am having some problems with openstack-ansible-deploy-ceph-ubuntu-xenial in https://review.openstack.org/#/c/517856/19/playbooks/ceph-rgw-install.yml "aio1_ceph-rgw_container-45628537 FAILED! => python-keystoneclient is missing", I thought that the delegate_to: "{{ groups['cinder_all'][0] }}" would execute it on a cinder_all container which has that pip module installed from what I have tested. Any pointers or advice? | 12:44 |
| *** hw_wutianwei has joined #openstack-ansible | 12:44 | |
| evrardjp | oh I read your reno | 12:45 |
| odyssey4me | evrardjp it is a separate thing, just not in the way initially planned it would seem | 12:45 |
| evrardjp | funsies | 12:45 |
| evrardjp | yeah it was planned to be really really separate and all iirc. but yeah, I see. | 12:45 |
| *** woodard has joined #openstack-ansible | 12:46 | |
| odyssey4me | I'm not terribly fussed about the detailed of why and all that. But I don't have an interest in reducing tech debt and confusion. ;) | 12:46 |
| *** acormier has quit IRC | 12:46 | |
| *** acormier has joined #openstack-ansible | 12:48 | |
| *** woodard has quit IRC | 12:50 | |
| evrardjp | Yes I read the whole thing now. Damn for the backport back. | 12:51 |
| evrardjp | Miouge: delegating to utility container or deploy node seems more logical to me | 12:52 |
| odyssey4me | Miouge yeah, if you delegate to the cinder container you'd have to ensure that the cinder venv is used to execute from because the clients are not installed on the host as far as I know | 12:53 |
| odyssey4me | the utility container, however, has all the clients installed on the system | 12:54 |
| *** acormier has quit IRC | 12:54 | |
| *** acormier has joined #openstack-ansible | 12:54 | |
| evrardjp | It's probably fair to run it from your target node directly? | 12:56 |
| Miouge | It was delegated to utility_all in patchset 16, it generated the same error: http://logs.openstack.org/56/517856/16/check/openstack-ansible-deploy-ceph-ubuntu-xenial/401225d/job-output.txt.gz#_2018-02-05_17_31_43_484978 | 12:56 |
| evrardjp | opening the logs now | 12:56 |
| evrardjp | Miouge: I'd not delegate at all | 12:56 |
| evrardjp | oh wait | 12:57 |
| evrardjp | I checked the playbook -- yeah delegating seems kinda necessary, I don't want the end node to have all the keystone stuff! | 12:57 |
| Miouge | rgw is a C or C++ blob, so adding the venv just to add the service, user, and endpoints is a bit overkill | 12:57 |
| *** acormier has quit IRC | 12:58 | |
| *** udesale has joined #openstack-ansible | 12:58 | |
| *** acormier has joined #openstack-ansible | 12:58 | |
| evrardjp | Miouge: agreed. Could you check if upstream os_ modules can provide everything you need? | 12:59 |
| evrardjp | Before thinking of large changes I want to push :p | 12:59 |
| *** gillesMo has quit IRC | 13:03 | |
| evrardjp | Miouge: the reason why it failed, is that the utility container is not configured yet at that point | 13:04 |
| Miouge | evrardjp: At a glance, it looks like os_keystone_service os_user and os_user_role could do the same type of thing. | 13:05 |
| Miouge | Oh, right, “- include: ceph-install.yml” then it does “- include: utility-install.yml” | 13:06 |
| evrardjp | yes so, even if you have pip already, you don't have all those clients | 13:07 |
| evrardjp | maybe we can move the utility install on top, after the repo use | 13:08 |
| evrardjp | which is after the repo build | 13:08 |
| evrardjp | let me try if it's gonna break everything | 13:08 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Move setup-utility before other plays https://review.openstack.org/541709 | 13:10 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Move setup-utility before other plays https://review.openstack.org/541709 | 13:10 |
| Miouge | I could re-order things like so: ceph-install, utility-install, ceph-rgw-install. Or another option is to split the ceph-rgw-install into one part for setup-infra and another part for setup-openstack. | 13:10 |
| *** acormier has quit IRC | 13:12 | |
| evrardjp | I'd say it's better to put it into setup-openstack, because it needs keystone right? :p | 13:12 |
| evrardjp | I think moving utility earlier in the builds is good for many reasons in the future, so I have published a separate patch | 13:12 |
| evrardjp | Miouge: I think you should put the rgw install after the keystone play. | 13:13 |
| Miouge | Yes, ceph-rgw is a swift so it is openstack-ish. It uses keystone for auth | 13:13 |
| Miouge | *swift compatible API | 13:13 |
| evrardjp | so yes, you added it too soon in the process I think | 13:13 |
| odyssey4me | Miouge if RGW requires keystone then it would make sense to have it execute after keystone is running | 13:13 |
| odyssey4me | you may wish to have it execute around the same place as swift as it performs a similar function? | 13:14 |
| odyssey4me | that'll also be helpful for things like glance/cinder which will want to make use of it | 13:14 |
| *** acormier has joined #openstack-ansible | 13:14 | |
| *** bhujay has joined #openstack-ansible | 13:14 | |
| *** olivierbourdon38 has joined #openstack-ansible | 13:16 | |
| Miouge | You are right. Actually I see that Logan already fixed the setup-openstack.yml part but I added it in ceph-install.yml which is executed in setup-infra.yml, so my bad :( | 13:18 |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: [WIP] Ceph RadosGW integration https://review.openstack.org/517856 | 13:19 |
| Miouge | Thank you evrardjp and odyssey4me I wouldn’t have found the problem with your help | 13:22 |
| mgariepy | good morning everyone | 13:27 |
| *** tux_ has joined #openstack-ansible | 13:34 | |
| *** woodard has joined #openstack-ansible | 13:46 | |
| *** woodard has quit IRC | 13:50 | |
| *** gameon has quit IRC | 13:52 | |
| *** markvoelker has joined #openstack-ansible | 13:57 | |
| tux_ | Good morning! | 13:58 |
| tux_ | Guys any clue? | 13:58 |
| tux_ | fatal: [aio1_gnocchi_container-08a39dcf]: FAILED! => {"failed": true, "msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"} | 13:58 |
| evrardjp | tux_: you don't have facts for your container? | 13:58 |
| *** acormier has quit IRC | 13:59 | |
| *** acormier has joined #openstack-ansible | 13:59 | |
| evrardjp | that, or an unsupported os. But I don't know which tasks fails, so it's kinda psychic skills | 13:59 |
| tux_ | I am using CentOS7 | 14:00 |
| evrardjp | which branch? | 14:00 |
| tux_ | stable/pike | 14:00 |
| *** kstev has joined #openstack-ansible | 14:00 | |
| tux_ | How do i check facts for my container? | 14:00 |
| evrardjp | yeah centos 7 is supported on branch | 14:01 |
| *** SerenaFeng has joined #openstack-ansible | 14:01 | |
| evrardjp | you have a fact cache on your deploy node | 14:01 |
| tux_ | show i just run "ansible -m setup all_containers" ? | 14:01 |
| evrardjp | https://docs.openstack.org/openstack-ansible/latest/admin/troubleshooting.html#cached-ansible-facts-issues | 14:01 |
| tux_ | and re-run task? | 14:01 |
| evrardjp | yeah | 14:01 |
| evrardjp | should do the trick | 14:01 |
| tux_ | let me try | 14:01 |
| tux_ | problem is build take very longtime to reach that task :( | 14:02 |
| evrardjp | tux_: don't use the higher level playbooks then, and use the ones that are included directly | 14:03 |
| *** acormier has quit IRC | 14:03 | |
| tux_ | can i just run "openstack-ansible os-gnocchi-install.yml" ? | 14:04 |
| *** kstev1 has joined #openstack-ansible | 14:04 | |
| evrardjp | yeha, if everything ran fine before that | 14:04 |
| tux_ | i thought there must be dependency which may cause some other issue. | 14:04 |
| tux_ | let me give it a try | 14:04 |
| *** kstev has quit IRC | 14:05 | |
| evrardjp | yeah | 14:05 |
| evrardjp | gnocchi doesn't have much love tbh, and under centos, ... that's even less love | 14:05 |
| tux_ | :) | 14:05 |
| tux_ | if i want to disable some component in openstack-ansible where i should do that? | 14:06 |
| tux_ | i don't want cinder/swift etc.. | 14:06 |
| *** hachi has joined #openstack-ansible | 14:06 | |
| evrardjp | I think it must be in our docs of the deploy guide | 14:07 |
| evrardjp | basically don't say your host x or y will have cinder | 14:07 |
| evrardjp | with an empty inventory group for cinder_all, nothing would be deployed cinder related | 14:07 |
| *** hamza21 has quit IRC | 14:08 | |
| evrardjp | tux_: but if you have a very specific use case for only one service, maybe you can use the role directly and bypass our integrated system altogether | 14:08 |
| tux_ | oh! that make sense | 14:08 |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible-repo_build master: Use pike in UCA instead of ocata https://review.openstack.org/541723 | 14:08 |
| evrardjp | tux_: great to hear that :) | 14:09 |
| tux_ | EXIT NOTICE [Playbook execution success] ************************************** | 14:09 |
| tux_ | : | 14:09 |
| tux_ | :) | 14:09 |
| evrardjp | We try to make it simple for ppl to help maintain it | 14:09 |
| evrardjp | ahah good news. | 14:09 |
| tux_ | Thanks! | 14:10 |
| evrardjp | I think there are quite a few ppl on the channel that can help with gnocchi | 14:10 |
| evrardjp | I am, sadly, not one of them. | 14:10 |
| tux_ | But it was helpful | 14:10 |
| evrardjp | and welcome :) | 14:11 |
| tux_ | in my case all i need nova/neutron/horizon/glance, because i am building very basic cloud service | 14:11 |
| tux_ | for testing i am installing everything to understand relation | 14:11 |
| evrardjp | no cinder is a little weird, but I get that | 14:11 |
| tux_ | We need CPU/memory only | 14:11 |
| evrardjp | make sense | 14:12 |
| mgariepy | evrardjp, https://github.com/evrardjp/ansible-keepalived/blob/master/vars/ubuntu-16.04.yml#L29 | 14:12 |
| tux_ | we store no data on disk, everything is on flight, data crunching | 14:12 |
| evrardjp | yeah there are things to configure in the conf.d or openstack_user_config to remove cinder | 14:12 |
| mgariepy | does that need to be set to ocata by the keepalived role ? | 14:12 |
| odyssey4me | tux_ take a peep at https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/configure.html#initial-environment-configuration for the descroption of how to add services | 14:13 |
| evrardjp | mgariepy: we can use Pike now indeed | 14:13 |
| evrardjp | I still advise using the ppa | 14:13 |
| *** mattt has joined #openstack-ansible | 14:13 | |
| *** acormier has joined #openstack-ansible | 14:14 | |
| tux_ | Thanks @Odyssey4me | 14:14 |
| *** acormier has quit IRC | 14:14 | |
| mgariepy | evrardjp, ppa or uca by default ? | 14:14 |
| *** acormier has joined #openstack-ansible | 14:14 | |
| evrardjp | ppa by default on the role, let me check for osa | 14:15 |
| mgariepy | seems like uca | 14:15 |
| *** esberglu has joined #openstack-ansible | 14:15 | |
| mgariepy | should we set that var in the osa playbook to match the current release ? | 14:16 |
| evrardjp | https://github.com/openstack/openstack-ansible/blob/7c9dfcc8bd4caa8a9689b0d2e511873563787525/inventory/group_vars/haproxy_all/keepalived.yml#L19-L20 | 14:16 |
| evrardjp | I think it's fair to update uca to latest pike | 14:16 |
| mgariepy | if we set the var in the osa playbook we could use the latest on in osa all the time ? | 14:17 |
| odyssey4me | mgariepy oh dear - https://review.openstack.org/#/c/541723/1/vars/ubuntu-16.04.yml - I'm surprised that we're even implementing the repo there... we shouldn't be | 14:17 |
| odyssey4me | that must be some sort of leftover | 14:17 |
| mgariepy | odyssey4me, yeah probalby. | 14:17 |
| mgariepy | haha | 14:17 |
| *** ianychoi has joined #openstack-ansible | 14:17 | |
| evrardjp | odyssey4me: yeah I completely missed it too. | 14:18 |
| evrardjp | We should probably remove it. | 14:18 |
| evrardjp | for keepalived it's a little different, so I'll fix that. | 14:18 |
| mgariepy | for keepalived maybe we can set keepalived_uca_openstack_release in inventory/group_vars/haproxy_all/keepalived.yml | 14:19 |
| odyssey4me | mgariepy left a comment - those vars aren't used - let's remove them | 14:20 |
| odyssey4me | I think for Pike it might be overriding the defaults in the pip_install role, so we should remove them there too! | 14:20 |
| odyssey4me | lemme check quick | 14:20 |
| mgariepy | we should implement some cleaning also :) | 14:21 |
| *** esberglu has quit IRC | 14:21 | |
| evrardjp | mgariepy: focus on next cycle | 14:21 |
| odyssey4me | mgariepy yeah, those vars aren't used in the repo_build role at all... | 14:21 |
| evrardjp | I mean that's the focus | 14:21 |
| mgariepy | I guess it would be fair to assume the file : ubuntu_cloud_archive_canonical_com_ubuntu.list only contain 1 entry | 14:22 |
| odyssey4me | mgariepy for queens we do some cleanup there :) | 14:22 |
| evrardjp | odyssey4me: could you vote on https://review.openstack.org/#/c/504224/ ? | 14:22 |
| odyssey4me | I just pushed https://review.openstack.org/541680 up this morning. | 14:22 |
| evrardjp | mgariepy: too | 14:22 |
| *** gkadam_ has quit IRC | 14:23 | |
| evrardjp | mgariepy: ok I checked why I haven't updated ansible-keepalived | 14:25 |
| evrardjp | it doesn't matter :) | 14:25 |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible-repo_build master: Cleanup UCA vars https://review.openstack.org/541723 | 14:25 |
| *** tescra has quit IRC | 14:25 | |
| evrardjp | https://packages.ubuntu.com/search?keywords=keepalived&searchon=names&suite=xenial-updates§ion=all | 14:26 |
| evrardjp | http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/ both for xenial don't have keepalived . | 14:26 |
| evrardjp | as you can see here: http://ubuntu-cloud.archive.canonical.com/ubuntu/pool/main/k/keepalived/ | 14:26 |
| *** DanyC_ has quit IRC | 14:26 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Stable keepalived priorities https://review.openstack.org/532233 | 14:26 |
| evrardjp | but, for the sake of the exercise, I can bump it. | 14:28 |
| evrardjp | odyssey4me: you just bumped to queens? | 14:28 |
| evrardjp | this way I bump it to queens for consistency | 14:28 |
| mgariepy | evrardjp, seems like it's not comming from UCA :P | 14:29 |
| *** vnogin has joined #openstack-ansible | 14:29 | |
| evrardjp | ? | 14:29 |
| odyssey4me | mgariepy mind if I add a little bit of extra info to your commit message? | 14:30 |
| mgariepy | http://paste.openstack.org/show/664761/ | 14:30 |
| mgariepy | it's on xenial-updates not from cloud archive. | 14:30 |
| mgariepy | but from archive.u.c | 14:30 |
| mgariepy | odyssey4me, nop i don't mind :) | 14:31 |
| *** esberglu has joined #openstack-ansible | 14:32 | |
| *** ivve has quit IRC | 14:33 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: Remove leftover UCA variables https://review.openstack.org/541723 | 14:33 |
| odyssey4me | there we go - I think that about covers it :) | 14:33 |
| evrardjp | mgariepy: exactly | 14:33 |
| mgariepy | thanks odyssey4me | 14:33 |
| mgariepy | so you should remove the uca part of keepalived | 14:34 |
| mgariepy | :) | 14:34 |
| mgariepy | at least for xenial. not too sure about trusty. | 14:34 |
| evrardjp | mgariepy: no | 14:36 |
| evrardjp | well | 14:37 |
| mgariepy | why ? | 14:37 |
| evrardjp | it's used outside osa | 14:37 |
| evrardjp | but as default, we are using it in ppa. | 14:37 |
| evrardjp | So I guess you have a point | 14:37 |
| mgariepy | yes but keepalived isn't in uca it's in the archive.ubuntu.com | 14:37 |
| evrardjp | why keeping it around if uca is enough | 14:37 |
| evrardjp | if system* | 14:37 |
| evrardjp | that's exactly what I said above | 14:38 |
| mgariepy | lol | 14:38 |
| mgariepy | now i'm confused ahha | 14:38 |
| evrardjp | me too | 14:39 |
| evrardjp | :) | 14:39 |
| mgariepy | keepalived doesn't need uca. | 14:39 |
| evrardjp | depends | 14:39 |
| evrardjp | :D | 14:39 |
| evrardjp | in xenial with updates, no need for uca | 14:39 |
| mgariepy | so you install it from ppa or archive.ubuntu.com | 14:39 |
| mgariepy | http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/xenial-updates/pike/main/binary-amd64/Packages | 14:40 |
| evrardjp | what I mean is that it's not ansible-keepalived to change | 14:41 |
| evrardjp | it's probably osa to not ask ansible-keepalived to install uca | 14:41 |
| mgariepy | http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/trusty-updates/mitaka/main/binary-amd64/Packages | 14:41 |
| mgariepy | so you need uca if running trusty | 14:41 |
| evrardjp | exactly | 14:42 |
| evrardjp | or ppa | 14:42 |
| mgariepy | trusty > ppa or uca , xenial > ppa or archive | 14:42 |
| evrardjp | yes | 14:42 |
| evrardjp | so | 14:42 |
| evrardjp | ansible-keepalived does take the assumption of ppa | 14:43 |
| evrardjp | let me fix what I think should be fixed | 14:43 |
| evrardjp | you'll understand :) | 14:43 |
| mgariepy | for xenial the options should be ppa or system (not uca) | 14:43 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Use native packages to deploy keepalived https://review.openstack.org/541733 | 14:44 |
| mgariepy | if we unset uca var for xenial in OSA we will install it from ppa (not what I want), and we don't need keepalived to install uca. | 14:44 |
| evrardjp | mgariepy: ^ is that clearer? | 14:45 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible master: Make Opendaylight as the BGP speaker using Quagga https://review.openstack.org/523907 | 14:45 |
| *** bhujay has quit IRC | 14:45 | |
| *** ivve has joined #openstack-ansible | 14:45 | |
| mgariepy | https://github.com/evrardjp/ansible-keepalived/blob/master/defaults/main.yml#L27 | 14:46 |
| mgariepy | https://github.com/evrardjp/ansible-keepalived/blob/master/defaults/main.yml#L23 | 14:46 |
| evrardjp | yeah that was compatibility | 14:46 |
| evrardjp | mgariepy: have a look at https://review.openstack.org/#/c/541733/ | 14:47 |
| mgariepy | so the role install from either ppa or uca ? | 14:47 |
| *** vnogin has quit IRC | 14:47 | |
| mgariepy | ok. | 14:47 |
| mgariepy | perfect. | 14:47 |
| mgariepy | but i just didn't saw it in default.. | 14:47 |
| evrardjp | it's in the docs. | 14:47 |
| evrardjp | the default is ppa | 14:48 |
| evrardjp | and will always be | 14:48 |
| mgariepy | need more cofee | 14:48 |
| *** hachi has quit IRC | 14:48 | |
| mgariepy | it's ok now i understand .. haha | 14:48 |
| evrardjp | ansible-keepalived is slowly phasing out things, and is not dependant of openstack branching strategies | 14:49 |
| evrardjp | uca was introduced for convenience | 14:50 |
| mgariepy | yeah it's ok, i now understand it. the default var was just confusing to me. | 14:51 |
| evrardjp | it's all about compatibility | 14:52 |
| evrardjp | I am sorry if I wasn't clear | 14:52 |
| evrardjp | I try to be though. | 14:52 |
| evrardjp | :D | 14:52 |
| mgariepy | haha | 14:52 |
| mgariepy | communication is not easy. | 14:53 |
| mgariepy | i was just slow to understand haha | 14:53 |
| mgariepy | odyssey4me, can you take a look a my comment please ? https://review.openstack.org/#/c/541680/ | 14:57 |
| evrardjp | odyssey4me: the thing you noticed with nova, is it blocking gates right now on master ? | 15:00 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Add a status line for SELinux status https://review.openstack.org/541371 | 15:01 |
| odyssey4me | mgariepy I had suggested that to evrardjp , but he didn't like it :p | 15:01 |
| odyssey4me | evrardjp the placement api thing? no, that has no effect on a running environment except to have an extra database | 15:02 |
| *** rodolof has joined #openstack-ansible | 15:03 | |
| odyssey4me | evrardjp tests are only failing because we need https://review.openstack.org/541417 merged :/ | 15:03 |
| odyssey4me | it'd be useful if y'all voted https://review.openstack.org/#/q/I213b66cddcad8328d7606385354a0a7194a5c569 through while we wait :) | 15:04 |
| evrardjp | mmm it was a tempest issue, so I thought it was something different. | 15:05 |
| *** SerenaFeng has quit IRC | 15:05 | |
| *** rodolof has quit IRC | 15:08 | |
| *** rodolof has joined #openstack-ansible | 15:08 | |
| *** shardy has joined #openstack-ansible | 15:08 | |
| idlemind | block_device_allocate_retries and block_device_allocate_retries_interval are not present variables in osa for nova.conf right? (they need to be added) | 15:09 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone master: Test keystone with a memcached pool https://review.openstack.org/540001 | 15:14 |
| odyssey4me | evrardjp os_nova is blocked on the ocata branch with tempest issues, yes - but it's not tempest that's got a problem, it's nova and I'm figuring that stuff out | 15:17 |
| evrardjp | I am talking about master, and something that failed on tempest, so it's not related. | 15:19 |
| evrardjp | let's just merge the other patches in between. | 15:20 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Update the UCA repo configured to queens https://review.openstack.org/541680 | 15:25 |
| mnaser | does the openstack-ansible ci checkout modules directly from git or does it use the repos cloned by zuul? | 15:28 |
| mnaser | it looks like it's manually checking them out? | 15:28 |
| cloudnull | mornings | 15:29 |
| *** jbadiapa has joined #openstack-ansible | 15:29 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build master: Remove leftover UCA variables https://review.openstack.org/541723 | 15:30 |
| mnaser | it seems like it does but for some reason with a depends-on, it doesn't seem to be working. | 15:30 |
| *** tannazn has quit IRC | 15:30 | |
| mnaser | i wonder if it's because im using the old change-id way of doing things, that hasnt been dropped? | 15:30 |
| odyssey4me | mnaser the integrated build pulls from git - the role tests use the zuul git cache and pull anything that's missing | 15:31 |
| mnaser | weird | 15:31 |
| mnaser | https://review.openstack.org/#/c/541272/ | 15:31 |
| odyssey4me | if you use a depends-on commit, then it should work | 15:31 |
| mnaser | if you look at the centos-7 failure there, it failed with the galera repo | 15:31 |
| mnaser | yet depends-on is there | 15:31 |
| cloudnull | anyone around mind giving https://review.openstack.org/#/q/starredby:cloudnull+status:open,n,z a nudge ? | 15:32 |
| mgariepy | odyssey4me, evrardjp are you ok with patches in stables release to remove previous release uca repo ? | 15:32 |
| idlemind | mnaser is that w/the updated or overriden galera repo that was found yesterday | 15:32 |
| mnaser | idlemind: yeah, if you click the change-id, you should see it redirect to the galera repo fix | 15:32 |
| odyssey4me | mnaser did the test run perhaps on an older patch set - the galera_Server patch has a depends-on to the galera_client patch - both need to be there for it to work | 15:32 |
| *** woodard has joined #openstack-ansible | 15:32 | |
| mnaser | maybe, zuul was pretty messed up yesterday | 15:33 |
| evrardjp | mgariepy: yeah | 15:33 |
| mnaser | maybe i'll just wait for things to clear out | 15:33 |
| *** woodard has quit IRC | 15:33 | |
| odyssey4me | mnaser that dependency was only added this morning | 15:33 |
| mnaser | ah gotcha | 15:33 |
| *** acormier has quit IRC | 15:34 | |
| *** woodard has joined #openstack-ansible | 15:34 | |
| mnaser | so ceilometer and magnum gate has fixes in ready to be merged | 15:34 |
| mnaser | and trove (but it's -1 because of centos) | 15:34 |
| mnaser | aodh needs some work so i'll iterate more on that today | 15:34 |
| odyssey4me | cloudnull done | 15:35 |
| cloudnull | tyvm ! | 15:35 |
| odyssey4me | mnaser gokhan fantastic, thanks for pushing to get those all fixed up :) | 15:35 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone stable/ocata: Add security headers to web accessable services. https://review.openstack.org/513974 | 15:36 |
| *** SmearedBeard has joined #openstack-ansible | 15:36 | |
| cloudnull | evrardjp: are you still wanting to block https://review.openstack.org/#/c/513974/ ? | 15:36 |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: [WIP] Ceph RadosGW integration https://review.openstack.org/517856 | 15:37 |
| evrardjp | cloudnull: no. I just didn't remove it | 15:37 |
| cloudnull | ok. | 15:38 |
| evrardjp | mnaser: thank you very much. | 15:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build stable/pike: Remove leftover UCA variables https://review.openstack.org/541753 | 15:43 |
| *** SmearedBeard has quit IRC | 15:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build stable/ocata: Remove leftover UCA variables https://review.openstack.org/541754 | 15:43 |
| *** jbadiapa has quit IRC | 15:45 | |
| evrardjp | cloudnull: https://review.openstack.org/#/q/I213b66cddcad8328d7606385354a0a7194a5c569 ? | 15:45 |
| *** idlemind has quit IRC | 15:46 | |
| cloudnull | approved | 15:46 |
| *** idlemind has joined #openstack-ansible | 15:47 | |
| evrardjp | cloudnull: https://review.openstack.org/#/c/541733/ | 15:48 |
| evrardjp | :D | 15:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build stable/newton: Remove leftover UCA variables https://review.openstack.org/541757 | 15:48 |
| *** acormier has joined #openstack-ansible | 15:49 | |
| *** acormier has quit IRC | 15:52 | |
| *** weezS has joined #openstack-ansible | 15:52 | |
| *** acormier has joined #openstack-ansible | 15:53 | |
| *** SerenaFeng has joined #openstack-ansible | 15:56 | |
| *** acormier_ has joined #openstack-ansible | 15:57 | |
| odyssey4me | it looks like galera_server is failing all tests because it can't start the cluster: http://logs.openstack.org/17/541417/4/check/openstack-ansible-functional-ubuntu-xenial/63532db/job-output.txt.gz#_2018-02-07_15_40_16_555695 | 15:57 |
| *** woodard has quit IRC | 15:57 | |
| odyssey4me | anyone got some time to figure out what's going on there? without galera_server working we're dead in the water | 15:58 |
| *** woodard has joined #openstack-ansible | 16:00 | |
| *** mxevgenis has joined #openstack-ansible | 16:00 | |
| mxevgenis | hi everyone. I have deployed an OpenStack cloud following the OSAD guide and now i would like to integrate with Ceph. All the OpenStack services are running on containers. Do you have any idea which guide i should follow? | 16:00 |
| *** SerenaFeng has quit IRC | 16:01 | |
| *** acormier has quit IRC | 16:01 | |
| openstackgerrit | Merged openstack/openstack-ansible-plugins master: Fix nsenter commands on centos https://review.openstack.org/540956 | 16:01 |
| mgariepy | http://logs.openstack.org/17/541417/4/check/openstack-ansible-functional-ubuntu-xenial/63532db/logs/openstack/container2/mysql_logs/galera_server_error.log.txt.gz#_2018-02-07_15_30_05 | 16:02 |
| *** pcaruana has quit IRC | 16:03 | |
| evrardjp | odyssey4me: reliabily failing? | 16:05 |
| odyssey4me | evrardjp both rechecks this morning on the yum patch produced failures for *all* tests | 16:06 |
| evrardjp | on it. | 16:07 |
| evrardjp | master I guess? | 16:07 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/ocata: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541699 | 16:07 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/pike: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541698 | 16:07 |
| odyssey4me | evrardjp yep | 16:07 |
| *** SerenaFeng has joined #openstack-ansible | 16:07 | |
| odyssey4me | master timed out for all functional tests, the timeout was due to the cluster start hang by the looks of the ubuntu result | 16:07 |
| evrardjp | odyssey4me: I got a successful run this morning | 16:08 |
| evrardjp | just fyi | 16:08 |
| evrardjp | will have a look in more details anyway, spinning up a new machine | 16:09 |
| odyssey4me | yeah, I'm wondering if we suddenly have a new upstream package release which is broken or incompatible with one of our other pins or something like that | 16:09 |
| *** bhujay has joined #openstack-ansible | 16:09 | |
| evrardjp | or auto bump to 11! | 16:10 |
| evrardjp | 10.2 | 16:10 |
| evrardjp | but eleven is funnier | 16:10 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: Remove hard-coded host group value https://review.openstack.org/529876 | 16:12 |
| *** acormier has joined #openstack-ansible | 16:14 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Build everything on metal https://review.openstack.org/504224 | 16:15 |
| *** acormier_ has quit IRC | 16:16 | |
| *** acormier has quit IRC | 16:16 | |
| mgariepy | 10.1.31+maria-1~xenial was 10.1.30+maria-1~xenial on feb 5 | 16:17 |
| *** acormier has joined #openstack-ansible | 16:17 | |
| *** acormier_ has joined #openstack-ansible | 16:18 | |
| *** acormier_ has quit IRC | 16:18 | |
| idlemind | rando thing ... letsencrypt might be letting wildcard certs in? that could be fun ... maybe an application in osa | 16:18 |
| *** acormier_ has joined #openstack-ansible | 16:18 | |
| odyssey4me | idlemind sure it'd be nice - the only trouble would be that we'd have to keep the current implementations for self-signed certs in-place as the defaults have no names to use for certs | 16:20 |
| evrardjp | idlemind: not sure if you have wildcard, but osa only need one certificate, you can definitely have it | 16:20 |
| idlemind | ya you'd have to name everything and reference it by name | 16:20 |
| evrardjp | if you want you just have the external lb vip address set to a name | 16:20 |
| evrardjp | then all the public interface is there. | 16:20 |
| idlemind | speaking of cerst you guys see the bug i put about the default of checking certs being true when it should be false for all self-signed deploys? | 16:21 |
| idlemind | boom master segway | 16:21 |
| *** acormier has quit IRC | 16:22 | |
| odyssey4me | could do with a final core vote for this newton backport: https://review.openstack.org/541757 | 16:22 |
| idlemind | this ugly guy https://bugs.launchpad.net/openstack-ansible/+bug/1747684 | 16:23 |
| openstack | Launchpad bug 1747684 in openstack-ansible "Default Values Do Not Allow Image Uploads to Glance from Horizon" [Undecided,New] | 16:23 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Drop MongoDB installation and API file configurations https://review.openstack.org/541275 | 16:23 |
| *** dariko has quit IRC | 16:25 | |
| *** dariko has joined #openstack-ansible | 16:26 | |
| tux_ | finally AIO has been installed successfully on my VM | 16:28 |
| tux_ | very big question show do i access horizon :) | 16:28 |
| tux_ | very big question how do i access horizon :) | 16:28 |
| *** jwitko_ has joined #openstack-ansible | 16:28 | |
| tux_ | I think i need to inject route for 172 net on my host machine | 16:28 |
| idlemind | tux_ congrats | 16:29 |
| *** mxevgenis has left #openstack-ansible | 16:29 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Skip creation of Swift user if Ceilometer support is disabled https://review.openstack.org/541374 | 16:30 |
| *** SmearedBeard has joined #openstack-ansible | 16:30 | |
| idlemind | tux_ can't be 100% sure, the host machine might need routes and you might need routes | 16:32 |
| tux_ | my horizon vip will be haproxy right on host | 16:33 |
| *** mardim has quit IRC | 16:33 | |
| mgariepy | tux_, should be the external vip in the config | 16:33 |
| tux_ | you are right it was external IP of my host machine, no route required, sweet! | 16:37 |
| tux_ | I am in | 16:37 |
| idlemind | tux_ victory | 16:37 |
| idlemind | rando question about that too though, can internal and external lb vip address be the same or does external_lb have to be different or at least a hostname? | 16:38 |
| tux_ | one week of epic journey! it was fun...! Thanks you folks whoever help me directly or indirectly :) | 16:38 |
| evrardjp | tux_: \o/ | 16:38 |
| odyssey4me | idlemind internal and external vip can be the same only *if* https is disabled for the public endpoints... nothing can serve two protocols on the same port ;) | 16:39 |
| idlemind | odyssey4me so you need an extra "external" interface to tie the vips to or does a hostname that gets nat'd work just fine too? | 16:40 |
| idlemind | for now i'm just using an extra interface | 16:40 |
| odyssey4me | generally speaking it's better to just use two different IP's, or perhaps use a resolvable name for the external address | 16:40 |
| odyssey4me | see the example configs in the deploy guide appendices | 16:40 |
| idlemind | cool i'll reread that i'm pretty sure i hit that early on cuz i tried to use the same initially but i couldn't remember | 16:41 |
| *** eumel8 has quit IRC | 16:46 | |
| *** sar has quit IRC | 16:46 | |
| *** markvoelker_ has joined #openstack-ansible | 16:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build stable/pike: Remove leftover UCA variables https://review.openstack.org/541753 | 16:48 |
| *** markvoelker has quit IRC | 16:49 | |
| *** tux_ has quit IRC | 16:51 | |
| *** tux_ has joined #openstack-ansible | 16:52 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build stable/ocata: Remove leftover UCA variables https://review.openstack.org/541754 | 16:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541798 | 16:54 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Add a status line for SELinux status https://review.openstack.org/541371 | 16:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/ocata: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541799 | 16:54 |
| evrardjp | odyssey4me: second time I have no issue passing the galera_server. I am letting it run through the end now. | 16:55 |
| odyssey4me | ^ I cherry-pick the master patch as it's been approved -while the issue in master is being worked out, hopefully we can unblock pike/ocata | 16:55 |
| idlemind | did you do the galera client ones too odyssey4me? | 16:56 |
| *** tux_ has quit IRC | 16:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541798 | 16:56 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/ocata: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541799 | 16:57 |
| odyssey4me | idlemind the client ones are already merged | 16:57 |
| idlemind | ah sorry | 16:57 |
| idlemind | love me dem stable/pike backports | 16:57 |
| evrardjp | :) | 16:57 |
| odyssey4me | unfortunatelu pike/ocata will also need integrated build role SHA updates as well | 16:58 |
| odyssey4me | *unfortunately ... it would appear I've been at this desk too long | 16:58 |
| *** shardy has quit IRC | 17:00 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build stable/newton: Remove leftover UCA variables https://review.openstack.org/541757 | 17:01 |
| *** SerenaFeng has quit IRC | 17:01 | |
| *** SerenaFeng has joined #openstack-ansible | 17:04 | |
| *** chyka has joined #openstack-ansible | 17:07 | |
| *** rromans has quit IRC | 17:09 | |
| *** SerenaFeng has quit IRC | 17:09 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix ansible pipelining https://review.openstack.org/541524 | 17:10 |
| *** udesale has quit IRC | 17:13 | |
| *** markvoelker_ has quit IRC | 17:14 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client master: RedHat: Ensure that conflicting MariaDB packages are removed https://review.openstack.org/541581 | 17:14 |
| *** markvoelker has joined #openstack-ansible | 17:14 | |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-lxc_container_create stable/newton: Zuul: Remove project name https://review.openstack.org/541808 | 17:17 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 17:20 |
| *** rromans has joined #openstack-ansible | 17:23 | |
| *** peri has quit IRC | 17:23 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Build everything on metal https://review.openstack.org/504224 | 17:27 |
| *** Leo_m has quit IRC | 17:31 | |
| *** Leo_m has joined #openstack-ansible | 17:32 | |
| *** bhujay has quit IRC | 17:33 | |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: [WIP] Ceph RadosGW integration https://review.openstack.org/517856 | 17:43 |
| evrardjp | odyssey4me: no issue in master, it was temporary or at least I can't reproduce it in my machines | 17:48 |
| evrardjp | retrying once again | 17:48 |
| odyssey4me | evrardjp it looks like opensuse passed in the gate this time - waiting for the others | 17:48 |
| odyssey4me | oh, maybe I'm looking at the backports | 17:49 |
| evrardjp | ubuntu is generally the most reliable. Sometimes I see a tempest failure, but that's very seldom. Then suse, then centos. | 17:49 |
| evrardjp | odyssey4me: mmm | 17:49 |
| evrardjp | I don't know but I'm retrying. 3rd time now. | 17:49 |
| odyssey4me | yeah, but in this case it's just the galera_server role tests | 17:49 |
| odyssey4me | maybe we just hit a bad batch of test nodes | 17:49 |
| evrardjp | OH | 17:49 |
| evrardjp | You told me integrated! | 17:50 |
| evrardjp | well | 17:50 |
| evrardjp | ok | 17:50 |
| odyssey4me | oh no, I said galera_server | 17:50 |
| evrardjp | lol | 17:50 |
| odyssey4me | it's specifically the cluster start that timed out | 17:50 |
| evrardjp | ok | 17:50 |
| evrardjp | hahah | 17:50 |
| odyssey4me | which only happens if there's actually a cluster ;) | 17:50 |
| evrardjp | that can have lasted long. | 17:50 |
| evrardjp | ofc | 17:50 |
| evrardjp | well my role test is busy on keystone, finishing real quick, then working on it | 17:51 |
| *** mbuil has quit IRC | 17:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/newton: Zuul: Remove project name https://review.openstack.org/541808 | 17:52 |
| *** gkadam has joined #openstack-ansible | 17:55 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Replace pxelinux with iPXE support in MNAIO https://review.openstack.org/539327 | 17:55 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Add a status line for SELinux status https://review.openstack.org/541371 | 17:57 |
| odyssey4me | evrardjp failed again - very strange: https://review.openstack.org/#/c/541417/ | 17:58 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Use package_state=present for CentOS 7 testing https://review.openstack.org/541017 | 17:58 |
| *** olivierbourdon38 has quit IRC | 17:59 | |
| odyssey4me | the pike one has passed opensuse - the rest are still running | 17:59 |
| odyssey4me | so maybe something merged in master which is breaking this? | 17:59 |
| odyssey4me | that could be in galera_client, or the tests repo | 18:00 |
| *** tux_ has joined #openstack-ansible | 18:01 | |
| odyssey4me | ot the galera_server repo | 18:01 |
| odyssey4me | *or | 18:01 |
| *** vnogin has joined #openstack-ansible | 18:05 | |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 18:05 |
| *** acormier has joined #openstack-ansible | 18:10 | |
| *** dim_ has joined #openstack-ansible | 18:13 | |
| dim_ | Hi Gurus! | 18:14 |
| *** acormier_ has quit IRC | 18:14 | |
| dim_ | I have a tiny problem here when firing up the instance | 18:14 |
| dim_ | ```No valid host was found. There are not enough hosts available.``` | 18:14 |
| dim_ | I have deployed my environment using the following template https://www.irccloud.com/pastebin/r6VrbGNK/mytemplate | 18:15 |
| idlemind | dim_ in horizon under admin -> hypervisors do you see your hypervisor(s) listed there? | 18:18 |
| dim_ | sure | 18:19 |
| dim_ | ``` | 18:20 |
| dim_ | Hostname Type VCPUs (used) VCPUs (total) RAM (used) RAM (total) Local Storage (used) Local Storage (total) Instances | 18:20 |
| dim_ | node02.lab.---.---QEMU 0 24 2GB 62.8GB 0Bytes 28GB 0 | 18:20 |
| dim_ | ``` | 18:20 |
| *** vnogin has quit IRC | 18:20 | |
| dim_ | https://www.irccloud.com/pastebin/VUYAGfue/ | 18:21 |
| *** acormier has quit IRC | 18:21 | |
| idlemind | dim_ if you login (or attach) to your syslog container do you see anything useful in the nova or horizon logs? | 18:23 |
| idlemind | maybe nova-scheduler | 18:24 |
| dim_ | https://www.irccloud.com/pastebin/qYnGDvuh/ | 18:25 |
| dim_ | I was thinking of basic DNS problem, however I found the following in logs on nova-sheduler | 18:25 |
| *** TxGirlGeek has joined #openstack-ansible | 18:25 | |
| dim_ | https://www.irccloud.com/pastebin/s15Og2LH/ | 18:26 |
| *** eumel8 has joined #openstack-ansible | 18:27 | |
| dim_ | https://www.irccloud.com/pastebin/scunNvwZ/ | 18:28 |
| idlemind | dim_ seems like the same log repeated over and over again in pastes. what does the log for node02 nova say? | 18:29 |
| idlemind | (plain nova not one of the services like api or scheduler) | 18:30 |
| *** jwitko_ has quit IRC | 18:30 | |
| dim_ | https://www.irccloud.com/pastebin/yKiaoZ2g/ | 18:31 |
| openstackgerrit | Merged openstack/openstack-ansible master: Use native packages to deploy keepalived https://review.openstack.org/541733 | 18:36 |
| openstackgerrit | Merged openstack/openstack-ansible master: Remove impossible tests https://review.openstack.org/541220 | 18:36 |
| *** hamza21 has joined #openstack-ansible | 18:37 | |
| dim_ | it looks like this person had the same problem https://openstack.nimeyo.com/108318/openstack-operators-openstack-unable-to-launch-an-instance | 18:40 |
| idlemind | dim_ what about neutron log on the physical host | 18:44 |
| idlemind | dim_ sheduling will fail and look like nova's fault when it (nova) cannot connect to the network you've setup on the underlying infra | 18:45 |
| dim_ | I will have a look | 18:47 |
| dim_ | thanks | 18:47 |
| dim_ | well I have setup my network the way it's described in the article | 18:47 |
| dim_ | all hosts are reachable within br-mgmt | 18:48 |
| idlemind | dim_ take a look at neutron log to rule it out | 18:48 |
| dim_ | will do | 18:49 |
| dim_ | rebooting the infra just in case | 18:49 |
| *** acormier has joined #openstack-ansible | 18:52 | |
| *** acormier has quit IRC | 18:52 | |
| *** acormier has joined #openstack-ansible | 18:52 | |
| odyssey4me | @cloudnull I'm curious - why bother with the symlinks in https://github.com/openstack/openstack-ansible/commit/d5860004652368b1f31aa8599a98545879c0174c - the dynamic inclusion could just be '{{ container_tech }}-hosts-setup.yml', etc... ? | 19:01 |
| odyssey4me | And by the way - forward looking, that dynamic inclusion won't work with include_playbook/import_playbook | 19:02 |
| cloudnull | I added the links so that we didn't change the deployer interactions | 19:03 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 19:03 |
| *** SmearedBeard has quit IRC | 19:03 | |
| cloudnull | though we could remove them | 19:03 |
| odyssey4me | cloudnull yeah, but that's my point - no renames of plays were necessary | 19:03 |
| odyssey4me | if your inclusion was changed to adopt the existing names, those symlinks wouldn't be necessary | 19:04 |
| cloudnull | they were renamed for ease of understanding and consistency. | 19:04 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Update MariaDB CentOS yum mirror URL https://review.openstack.org/541799 | 19:04 |
| odyssey4me | it's a 1:1 mapping, and the new naming structure is not all that different | 19:04 |
| odyssey4me | lxc-hosts-setup vs containers-lxc-host ? | 19:04 |
| cloudnull | yea that's the point. | 19:05 |
| cloudnull | as we add new drivers it would be containers-nspawn-host ... | 19:05 |
| odyssey4me | Personally I'm not seeing the value... all the existing files use lxc- as a prefix, so they're pretty clear what they're for | 19:05 |
| cloudnull | and baremetal would skip containers-* | 19:05 |
| odyssey4me | if new files have nspawn as a prefix, they have the same | 19:05 |
| cloudnull | sure. it couldd | 19:05 |
| odyssey4me | the baremetal already skips as lxc_hosts or nspawn_hosts is empty | 19:06 |
| cloudnull | sure. but if we look at the plays we can see containers-* are for containers. | 19:06 |
| cloudnull | similar to os-* is for openstack | 19:06 |
| cloudnull | is there an issue you're seeing with that stuff ? | 19:06 |
| odyssey4me | with all the other infra littered in-between ;) | 19:06 |
| cloudnull | yea, :) | 19:07 |
| odyssey4me | nah - looks functional to me - posted some questions in the reviews | 19:07 |
| cloudnull | we could make that consistently named too. | 19:07 |
| odyssey4me | I'm not too bothered with the change of play names, just questioning the value. | 19:07 |
| odyssey4me | If we're doing an across the board naming standard, them I'm game | 19:07 |
| cloudnull | I think it'd be good from a ux point of view | 19:08 |
| cloudnull | but not the most pressing issue | 19:08 |
| odyssey4me | The only issue I'm seeing thus far is that ansible 2.4 throws monstrous warnings about using include: because it's deprecated... and the dynamic inclusion of plays won't work with include_playbook or import_playbook | 19:08 |
| *** stuartgr has quit IRC | 19:08 | |
| cloudnull | so how do we tackle that ? | 19:08 |
| mnaser | we have to deploy gnocchi to get the aodh alarming tempest tests to pass as it needs it :< | 19:08 |
| odyssey4me | It would be more ideal for us to figure out a better way of doing it - maybe through some sort of plugin that picks up a config or env var somewhere | 19:09 |
| odyssey4me | although, that said, I don't think we really need the dynamic inclusion - as long as lxc_hosts is empty, the play will just skip | 19:10 |
| cloudnull | I think that's a great idea. | 19:10 |
| mnaser | is there a way to add it to common/ ? or what would be a good way without duplicating the code that deploys it in the aodh role? | 19:10 |
| odyssey4me | mnaser the tests repo only needs things which are common across more than one repo | 19:10 |
| odyssey4me | like the keystone playbook | 19:10 |
| odyssey4me | the roles used for a role test are specified in each repo's tests/ansible-role-requirements.yml file and the tests/ folder has that role's playbooks with tests.yml as the entrypoint | 19:11 |
| mnaser | so if aodh needs gnocchi (and maybe at some point ceilometer and panko probably), should we consider adding it to common? | 19:11 |
| odyssey4me | mnaser yeah, sounds about right - 'common' is actually a clone of openstack-ansible-tests | 19:12 |
| evrardjp | odyssey4me: ok I started to have a look at it. So I reproduced it on my machine | 19:12 |
| cloudnull | odyssey4me: maybe I can make the change get picked up in inventory so that it populates the correct container tech groups. | 19:12 |
| cloudnull | I'll look into that | 19:12 |
| evrardjp | and because there was no change in the role, I expect it's a packaging issue, so I will check if it's .31 instead of .30 | 19:12 |
| mnaser | odyssey4me: ok i guess ill look into getting it added to openstack-ansible-tests | 19:13 |
| odyssey4me | cloudnull that might be nice to improve usability a bit, but the code as-is will work without the dynamic inclusion I think... I'll verify in a bit once the metal patch is done testing | 19:13 |
| *** SmearedBeard has joined #openstack-ansible | 19:15 | |
| *** qp4 has joined #openstack-ansible | 19:17 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 19:17 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541270 | 19:18 |
| *** qp4 has quit IRC | 19:18 | |
| mnaser | apparently there was already a playbook for gnocchi so yay | 19:18 |
| *** dim__ has joined #openstack-ansible | 19:18 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Stop any running VMs when re-deploying https://review.openstack.org/541416 | 19:20 |
| odyssey4me | cloudnull I see now why it's utually exclusive currently - commented in the review | 19:22 |
| odyssey4me | *mutually exclusive | 19:22 |
| odyssey4me | mnaser the tests repo is a bit of a hodgepodge - my condolences in advance :/ | 19:23 |
| mnaser | odyssey4me: didnt even have to touch it, gnocchi playbook was already there so np | 19:23 |
| *** SmearedBeard has quit IRC | 19:27 | |
| idlemind | well looky there the cloudbase folks got back to me about the ipv6 issue i had in their win2k12r2 image :) maybe we can get a new master out for download | 19:29 |
| odyssey4me | righto - I'm out for the night folks, have a great day/eve! | 19:29 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Zuul: Remove project name https://review.openstack.org/541093 | 19:32 |
| *** gkadam has quit IRC | 19:32 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: fix misspelling of 'configuration' and 'list' https://review.openstack.org/534195 | 19:32 |
| *** markvoelker has quit IRC | 19:33 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Updated from global requirements https://review.openstack.org/535051 | 19:33 |
| *** markvoelker_ has joined #openstack-ansible | 19:33 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Updated from OpenStack Ansible Tests https://review.openstack.org/518909 | 19:34 |
| mnaser | (rebasing stuff) | 19:34 |
| mnaser | (sorry for spam) | 19:34 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Replace virtualenv-tools by a script https://review.openstack.org/533648 | 19:34 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Adds ceilometer-upgrade task https://review.openstack.org/526655 | 19:35 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Allow the experimental trigger of the integrated repo https://review.openstack.org/525950 | 19:35 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-rsyslog_client stable/pike: Fix pep8 doc config lint failures https://review.openstack.org/541847 | 19:35 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Remove pip_install dependency https://review.openstack.org/525546 | 19:35 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_ceilometer master: Deprecate os_endpoint_type option Option "os_endpoint_type" from groups "service_credentials" and "keystone_authtoken" is deprecated. Use option "interface" from groups "service_credentials" and "keystone_authtoken". https://review.openstack.org/521857 | 19:35 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-rsyslog_client stable/pike: Correct task tag https://review.openstack.org/541024 | 19:37 |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-tests stable/pike: Zuul: Remove project name https://review.openstack.org/541848 | 19:38 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_magnum master: Zuul: Remove project name https://review.openstack.org/538627 | 19:38 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_magnum master: Replace virtualenv-tools by a script https://review.openstack.org/533657 | 19:38 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_magnum master: Updated from global requirements https://review.openstack.org/535061 | 19:39 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_trove master: Remove Trove's support for Heat https://review.openstack.org/534677 | 19:41 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_trove master: Zuul: Remove project name https://review.openstack.org/540308 | 19:41 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_trove master: Replace virtualenv-tools by a script https://review.openstack.org/533668 | 19:42 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_trove master: Updated from global requirements https://review.openstack.org/535074 | 19:42 |
| mnaser | ok im done i promise | 19:42 |
| mnaser | any cores mind giving this a +W again to kick it off to the gate? https://review.openstack.org/#/c/536441/ | 19:42 |
| mnaser | better than recheck and having it go through check then gate for no reason | 19:43 |
| evrardjp | odyssey4me: I have passed the day of caring for bare metal. If ppl don't want to get it in, too bad. More regressions like the one you found will appear, and I am done fixing other ppls stuff for now. | 19:48 |
| dim__ | @idlemind ``` | 19:49 |
| idlemind | dim_ ya? | 19:49 |
| dim__ | one sec | 19:49 |
| dim__ | ``` | 19:51 |
| *** pcaruana has joined #openstack-ansible | 19:52 | |
| *** dim__ has quit IRC | 19:55 | |
| *** dim__ has joined #openstack-ansible | 19:55 | |
| dim__ | ``` | 19:56 |
| dim__ | ``` | 19:56 |
| idlemind | uh oh | 19:56 |
| dim__ | @idlemind | 19:56 |
| dim__ | ``` | 19:56 |
| idlemind | i just see backticks | 19:56 |
| dim__ | can't sent multilne | 19:56 |
| idlemind | put in pastebin and send link | 19:57 |
| dim_ | https://www.irccloud.com/pastebin/6s4EFmQW/ | 19:58 |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_monasca stable/pike: Zuul: Remove project name https://review.openstack.org/541854 | 19:58 |
| idlemind | dim_ that looks a nova log, what's in neutron log | 19:59 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541270 | 19:59 |
| dim_ | https://www.irccloud.com/pastebin/pehm86Ex/ | 20:00 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server master: [WIP] Prevent latest mariadb version https://review.openstack.org/541855 | 20:00 |
| dim_ | https://www.irccloud.com/pastebin/0RN5qQef/ | 20:01 |
| idlemind | linux bridge agent | 20:01 |
| dim_ | it's pretty clean | 20:03 |
| dim_ | https://www.irccloud.com/pastebin/P0eUAHEl/ | 20:04 |
| evrardjp | spotz: could you ping me when you have time to work on docs with me? | 20:05 |
| evrardjp | or we plan something | 20:06 |
| spotz | evrardjp: Yeah, trying to switch stuff from one mac to another right now | 20:06 |
| evrardjp | so it means okay or not? Not so sure... | 20:06 |
| spotz | Um not right this second?:) How long you gonna be around for tonight? | 20:07 |
| *** pcaruana has quit IRC | 20:07 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Remove jinja from failed_when https://review.openstack.org/541391 | 20:12 |
| dim__ | @idlemind do you think it's the link between Controller and the Compute node? | 20:18 |
| dim__ | node01 is a controller node in my case and node02 is a Compute node, node03 is a Storage node | 20:18 |
| idlemind | dim__ a lot of times that error has to do w/the networking configuration and neutron not running correctly | 20:18 |
| dim__ | I see | 20:18 |
| idlemind | dim__ are you on ubuntu or centos? | 20:19 |
| dim__ | but this setup is common, and config file from it should look good? | 20:19 |
| dim__ | https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-test.html#test-environment-config | 20:19 |
| dim__ | it's ubuntu 16.04 | 20:19 |
| dim__ | and Pike release 16.0.6 | 20:19 |
| dim__ | all I changed in networking configuration was the interface name convention from eth0 to enp1s0f0 | 20:20 |
| idlemind | dim__ is the neutron service running on your compute node (it appears to be) | 20:21 |
| dim__ | http://pastebin.ca/3969565 | 20:23 |
| dim__ | i don't see "neutron" in ps aux at node02 (Compute) | 20:23 |
| dim__ | http://pastebin.ca/3969567 | 20:24 |
| idlemind | dim__ on node2: systemctl status neutron-linuxbridge-agent | 20:26 |
| evrardjp | spotz: not sure, it's already late and I don't feel real good ... Let's postpone | 20:26 |
| *** SmearedBeard has joined #openstack-ansible | 20:27 | |
| cloudnull | feel better evrardjp | 20:27 |
| spotz | evrardjp: okie, ping in the morning. I need to get a few things done tonight (CFP and UC nom) | 20:27 |
| spotz | evrardjp: Feel better | 20:27 |
| evrardjp | oh cloudnull while you're here | 20:27 |
| evrardjp | I tried to do something but it doesn't work: setting a pin | 20:28 |
| evrardjp | on an apt package | 20:28 |
| evrardjp | https://review.openstack.org/#/c/541855/ | 20:28 |
| evrardjp | you know why this doesn't work? | 20:28 |
| cloudnull | no idea. is it not generating the config? | 20:29 |
| evrardjp | I see the pin on the container, but it's still that same version that gets installed | 20:29 |
| dim__ | idlemind_ http://pastebin.ca/3969572 | 20:29 |
| evrardjp | nah it's generating that pin, it's just also installing that same version. While I suppose previous version is still available in repos | 20:29 |
| idlemind | dim_ _ active from 1 second ago, did you just start or restart it? | 20:30 |
| *** hw_wutianwei has quit IRC | 20:31 | |
| dim__ | looks like it's restarting all the time | 20:31 |
| dim__ | http://pastebin.ca/3969576 | 20:31 |
| evrardjp | well it's probably a xtradb cluster thing | 20:31 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Replace pxelinux with iPXE support in MNAIO https://review.openstack.org/539327 | 20:32 |
| idlemind | dim__ k that's probably where our trouble is | 20:32 |
| *** hw_wutianwei has joined #openstack-ansible | 20:32 | |
| dim__ | idlemind__ I hope we will be able to fix it now | 20:32 |
| evrardjp | cloudnull: interesting enough: http://repo.percona.com/apt/dists/xenial/main/ there are two folders: binary-x86_64/ and binary-amd64 | 20:32 |
| idlemind | dim__ can you see anything with journalctl -xe | 20:33 |
| evrardjp | let's hope they are symlinked. | 20:33 |
| evrardjp | not that it really matters for this issue | 20:33 |
| dim__ | idlemind http://pastebin.ca/3969577 | 20:35 |
| evrardjp | cloudnull: ok my bad, mariadb-server-10.1 (.30) isn't on the repo Packages file anymore | 20:35 |
| idlemind | dim__ can you post your neutron.conf | 20:36 |
| idlemind | also from node02's ssh (console) grep ERROR /var/log/neutron/neutron-linuxbridge-agent.log | 20:37 |
| dim__ | sure idlemind | 20:38 |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-ceph_client stable/ocata: Zuul: Remove project name https://review.openstack.org/541869 | 20:38 |
| dim__ | idlemind http://pastebin.ca/3969578 | 20:39 |
| idlemind | cat /etc/neutron/neutron.conf | 20:40 |
| dim__ | http://pastebin.ca/3969579 | 20:40 |
| dim__ | http://pastebin.ca/3969579 | 20:40 |
| dim__ | http://pastebin.ca/3969579 | 20:40 |
| dim__ | http://pastebin.ca/3969579 | 20:40 |
| dim__ | https://pastebin.ca/3969579 | 20:40 |
| idlemind | yar | 20:41 |
| idlemind | i know the problem | 20:41 |
| dim__ | sorry it pasted several times I don't know why | 20:41 |
| idlemind | talked about this yesterday w/the crew. the docs break the flat network out to a physical interface because we're not allowed to use br-vlan for flat and vlan | 20:41 |
| dim__ | idlemind it's the first time I get help in this chat and it feels soooo good | 20:41 |
| idlemind | that's why there's that bit about host-override | 20:41 |
| dim__ | thank you | 20:41 |
| idlemind | lol np sorry you got stuck w/me i'm a newbie like you the other chaps probably would've had you squared away a bit ago | 20:42 |
| idlemind | https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-test.html#test-environment-config | 20:42 |
| idlemind | so if you look there, in the "Environment layout" for /etc/openstack_deploy/openstack_user_config.yml | 20:43 |
| idlemind | it's defining your provider networks | 20:43 |
| idlemind | one of them is the "flat" network (name and type) | 20:43 |
| idlemind | you'll see it has "host_bind_override: eth12" | 20:43 |
| dim__ | brvlan is flat | 20:43 |
| *** sar has joined #openstack-ansible | 20:43 | |
| idlemind | this is saying, set the flat network on linuxbridge agents to use eth12 not br-vlan for this flat network | 20:43 |
| idlemind | sad part is, eth12 doesn't exist | 20:43 |
| idlemind | so neutron bombs when it is started | 20:44 |
| dim__ | wow | 20:44 |
| dim__ | ok, should I change the config to use br-vlan instead or should I create eth12 interface on node02? | 20:44 |
| idlemind | nope, you can't neutron only lets one network be 1 interface | 20:44 |
| idlemind | br-vlan gets used in the vlan type network | 20:44 |
| idlemind | there really needs to be a "flat" interface on the host | 20:45 |
| idlemind | it can't be a bridge though | 20:45 |
| idlemind | so, in your case w/eth0 (or enp<whatever>) | 20:45 |
| idlemind | you can make a VLAN that represents your FLAT network on the real LAN side | 20:45 |
| idlemind | say VLAN 718 | 20:45 |
| idlemind | eth0.718 | 20:45 |
| dim__ | ok | 20:46 |
| idlemind | and that vlan interface can be put in host_override | 20:46 |
| idlemind | like i said though i'm not sure why container_bridge is br_vlan for that and how it works but hey. in my case. i just use the VLAN type to get my external connectivity via QinQ trunking | 20:46 |
| idlemind | maybe one of the experts can tell us why br-vlan is used as the container_bridge for flat in the docs | 20:47 |
| idlemind | (my case, i just commend out the stock flat network definition - at least for now) | 20:47 |
| dim__ | I might sound stupid, sorry if so | 20:48 |
| dim__ | but is this what I need to do? | 20:49 |
| dim__ | http://pastebin.ca/3969580 | 20:49 |
| *** dave-mcc_ has quit IRC | 20:49 | |
| shananigans | You can have an access vlan(default vlan) set up on the switch port to use for the flat network as well as trunked vlans to use for various vlan networks. | 20:49 |
| dim__ | well in this case I have vlan30 which is trunked to all the nodes01-03 | 20:50 |
| *** dave-mccowan has joined #openstack-ansible | 20:50 | |
| dim__ | VLAN30 is used as a flat network(br-mgmt) between all three nodes, VLAN31 Tunnel (VXLAN) Network is setup on node02(compute), VLAN32 - Storage between node02 (compute) and node03 (storage) | 20:51 |
| mhayden | evrardjp: i got https://review.openstack.org/541371 to work ;) | 20:54 |
| dim__ | idle mind could you please double check the configuration changes I made are correct? | 20:54 |
| dim__ | http://pastebin.ca/3969580 | 20:54 |
| mhayden | mgariepy: just an FYI -- SELinux is enforcing by default in ci jobs now on centos ;) | 20:54 |
| idlemind | dim__ sure post is invalid tho | 20:54 |
| dim__ | if not what would be a correct configuration, also should I add that VLAN to all nodes? | 20:54 |
| idlemind | dim__ repost your user_config the pastebin link was messed up or something | 20:55 |
| dim__ | hold on a sec | 20:56 |
| dim__ | http://pastebin.ca/3969582 | 20:56 |
| idlemind | dim__ yup as long as enp1s0f0.718 exists on the node02 host | 20:56 |
| mgariepy | ok thanks mhayden | 20:57 |
| evrardjp | mhayden: ahah ok | 20:57 |
| evrardjp | thanks | 20:57 |
| idlemind | (it can be any interface that exists and isn't being used by one of the other networks) | 20:57 |
| evrardjp | Found another repo for mariadb! | 20:57 |
| evrardjp | wooot | 20:57 |
| evrardjp | not documented | 20:57 |
| dim__ | Should enp1s0f0.718 be connected to some bridge? | 20:58 |
| dim__ | idlemind should enp1s0f0.718 be connected to the bridge? like br-718 as an example? | 20:58 |
| idlemind | dim__ for flat it can't be a bridge | 20:58 |
| idlemind | (from what i'm told) | 20:58 |
| dim__ | ok | 20:58 |
| dim__ | thank you | 20:58 |
| dim__ | so then I need to have that VLAN on all nodes including the deployment host? | 20:58 |
| evrardjp | mhayden: ok so I have the choice: Either we ship something with CVEs -- or we ship something that can be clustered. | 20:58 |
| evrardjp | What would mhayden do? | 20:59 |
| evrardjp | (if it was late and bed is close) | 20:59 |
| idlemind | i think the vlan only needs to be on the network_hosts one | 20:59 |
| dim__ | ok, so once I have it on neutron hosts and, node01 and node02 it should work | 21:02 |
| mhayden | evrardjp: haha, not sure i understand the context | 21:02 |
| dim__ | or I need to re-run sensible playbook for hosts, infrastructure and openstack deployment | 21:02 |
| dim__ | ? | 21:02 |
| idlemind | dim__ ya or alternatively, just comment out that provider network and re-run the os-neutron playbook to see if you can get an instance up | 21:02 |
| evrardjp | mhayden: galera_server clustering is broken. Gates for galera_server are broken. | 21:03 |
| evrardjp | mhayden: mariadb released a new package yesterday | 21:03 |
| mhayden | oh, i was running that manually right now to figure out WTF was happening | 21:03 |
| evrardjp | that new package is fixing cves | 21:03 |
| evrardjp | https://review.openstack.org/#/c/541855/1 | 21:03 |
| evrardjp | mhayden: Please read what I had written there, if you can continue my work that would be great, I'm highly tired. | 21:04 |
| evrardjp | got that bug real late so... | 21:04 |
| * mhayden ganders | 21:04 | |
| evrardjp | it's probably something we do with our config (mysql/systemd), because I doubt they would release a package that can't pass clustering tests. | 21:05 |
| dim__ | the whole thing like this? http://pastebin.ca/3969583 | 21:05 |
| evrardjp | mhayden: in the meantime I am testing to change the url real quick | 21:05 |
| nurdie | is it common that infra2_galera joins the WSREP cluster, but then fails to start because of a broken pipe? | 21:06 |
| nurdie | this is killing me | 21:06 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 21:06 |
| cloudnull | odyssey4me: ^ | 21:06 |
| evrardjp | nurdie: new | 21:07 |
| evrardjp | nurdie: mariadb just released a package that breaks clustering | 21:08 |
| evrardjp | nurdie: do you run 10.1.31? | 21:08 |
| mhayden | yeah, it's weird | 21:08 |
| mhayden | just hanging | 21:08 |
| evrardjp | I am working on it | 21:08 |
| evrardjp | yeah | 21:08 |
| cloudnull | evrardjp: did you get that pin to work ? | 21:08 |
| nurdie | evrardjp, yes | 21:09 |
| evrardjp | pin could work, but we don't have another package available in repo, so it's using it. | 21:09 |
| cloudnull | oh... | 21:09 |
| evrardjp | nurdie: welcome to the club. | 21:09 |
| dim__ | idlemind I have commented all section out as stated here: http://pastebin.ca/3969583 | 21:09 |
| cloudnull | that's a bummer | 21:09 |
| nurdie | evrardjp, no | 21:09 |
| nurdie | lol | 21:09 |
| evrardjp | cloudnull: https://review.openstack.org/#/c/541855/1 | 21:09 |
| nurdie | evrardjp, wait jesus idunno | 21:09 |
| evrardjp | I am using what I commented there | 21:09 |
| evrardjp | I'm testing if it fixes it temporarily. | 21:09 |
| dim__ | and running os-neutron setup with this configuration | 21:09 |
| idlemind | cool cool that should at least get an instance up and running | 21:10 |
| nurdie | evrardjp, yep, 10.1.31 -__- | 21:10 |
| cloudnull | evrardjp: the repo in your comment seems like the way to go | 21:11 |
| evrardjp | nurdie: welcome to the club | 21:11 |
| evrardjp | cloudnull: yeah probably | 21:11 |
| evrardjp | I am testing | 21:11 |
| evrardjp | locally | 21:11 |
| evrardjp | then we can figure out what's wrong with our config | 21:11 |
| evrardjp | same applies to centos and suse ofc | 21:11 |
| *** SmearedBeard has quit IRC | 21:12 | |
| nurdie | evrardjp, lol thanks. | 21:12 |
| nurdie | evrardjp, i had the same issues this morning with 10.2, since they broke their 10.1 repos | 21:12 |
| evrardjp | nurdie: so pick your poison: cves or no clustering? | 21:13 |
| nurdie | 10.1 centos repos* | 21:13 |
| evrardjp | yeah on top of that. | 21:13 |
| nurdie | cves | 21:13 |
| nurdie | lol | 21:13 |
| evrardjp | haha | 21:13 |
| nurdie | is there a known version for ocata OSA that works? | 21:14 |
| nurdie | 10.0? | 21:14 |
| evrardjp | 10.1.30 works | 21:14 |
| nurdie | ah | 21:14 |
| evrardjp | it's just the version released yesterday that sucks | 21:14 |
| evrardjp | at least with our configuration | 21:14 |
| evrardjp | ok it just passed my test | 21:14 |
| nurdie | thanks i'm going to nuke my OSA and force 10.1.30 >_< | 21:15 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-tests master: Add os_gnocchi role to required projects https://review.openstack.org/541902 | 21:15 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541270 | 21:16 |
| mnaser | can i get really quick eyes on the openstack-ansible-tests change above, it's trivial | 21:16 |
| dim__ | looks better now @inlemind http://pastebin.ca/3969588 | 21:16 |
| dim__ | @idlemind | 21:16 |
| idlemind | dim__ yay now you can tinker w/your cloud. as you get more comfortable you can start looking at bringing that FLAT back in over a different physical (eth12 in the docs) or vlan (of eth0 - enp1...) | 21:17 |
| dim__ | the VM is now running | 21:20 |
| dim__ | Active nova None Running | 21:20 |
| dim__ | thank you so much!!!! | 21:20 |
| dim__ | I have another problem though, glance behaves in a weird way letting me upload the image from the CLI only | 21:21 |
| dim__ | and if I start a new instance it's not attaching any image to it, unless U run nova command from the console | 21:21 |
| *** acormier has quit IRC | 21:22 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server master: Prevent latest mariadb version https://review.openstack.org/541855 | 21:22 |
| evrardjp | nurdie: cloudnull mhayden ^ | 21:23 |
| evrardjp | this should probably temporaily workaround the issue. | 21:23 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server master: Prevent latest mariadb version https://review.openstack.org/541855 | 21:24 |
| evrardjp | mhayden: could you babysit this or alternatively find the root cause of the issue? ^ | 21:26 |
| mhayden | already looking for root cause | 21:26 |
| nurdie | evrardjp, thanks. I just hardcoded the 10.1.30/centos repo for mine | 21:26 |
| mhayden | galera sits there listening on port 44444 and doesn't ever get anything | 21:26 |
| mhayden | so it just hangs forever | 21:26 |
| mhayden | err 4444 | 21:26 |
| nurdie | mhayden, yup | 21:26 |
| nurdie | I joins the cluster first | 21:27 |
| mhayden | if i telnet to the port and say gibberish, it hangs up on me | 21:27 |
| *** rama_y has joined #openstack-ansible | 21:27 | |
| nurdie | someone should tcpdump it. i already killed my deployment to use 10.1.30 | 21:27 |
| dim__ | idlemind thank you for all your help | 21:28 |
| idlemind | np | 21:28 |
| idlemind | have a good night | 21:28 |
| dim__ | by the way did you have any problems with the console access??? | 21:28 |
| idlemind | dim__ using spice (the default) i have to open it in a new tab for most vm's and the windows image is upside down | 21:28 |
| idlemind | i'll probably change to novnc ... it seems that fixes a lot of it | 21:28 |
| idlemind | just haven't had a chance to run the os-nova playbook w/the change again | 21:29 |
| dim__ | is this needs to be specified in the config? | 21:29 |
| dim__ | even in a new tab it shows the following "The connection was reset" | 21:29 |
| idlemind | hmm might be a certificate thing | 21:31 |
| idlemind | try to accept the certificate | 21:31 |
| idlemind | open in new tab and add the exception | 21:31 |
| idlemind | then reload and you should get a console displayed | 21:31 |
| dim__ | here is another problem | 21:36 |
| dim__ | 2018-02-07 15:35:25.952 1152 INFO nova.console.websocketproxy [-] 192.168.30.11 - - [07/Feb/2018 15:35:25] code 404, message No such file | 21:36 |
| dim__ | 2018-02-07 15:35:25.954 1152 INFO nova.console.websocketproxy [-] handler exception: [Errno 104] Connection reset by peer | 21:36 |
| dim__ | root@infra1-nova-console-container-e5931327:~# | 21:36 |
| *** armaan has quit IRC | 21:37 | |
| *** armaan has joined #openstack-ansible | 21:38 | |
| dim__ | @idlemind is this normal? | 21:38 |
| dim__ | http://pastebin.ca/3969591 | 21:38 |
| cloudnull | evrardjp: lgtm | 21:38 |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-ops stable/pike: Zuul: Remove project name https://review.openstack.org/541913 | 21:39 |
| *** eumel8 has quit IRC | 21:39 | |
| *** threestrands has joined #openstack-ansible | 21:39 | |
| *** threestrands has quit IRC | 21:39 | |
| *** threestrands has joined #openstack-ansible | 21:39 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 21:51 |
| idlemind | dim__ seems kinda strange to me | 21:57 |
| dim__ | your config looks the same? | 21:58 |
| dim__ | ```root@infra1-nova-console-container-e5931327:~# less /etc/nova/nova.conf | 21:58 |
| dim__ | ``` | 21:58 |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-ceph_client stable/newton: Zuul: Remove project name https://review.openstack.org/541919 | 21:59 |
| *** woodard has quit IRC | 21:59 | |
| mhayden | so, on centos at least, we get the mariadb packages from mariadb's mirrors, but the galera RPM comes from the RDO repo | 22:03 |
| mhayden | which is likely not a good thing | 22:03 |
| * mhayden is testing said hypothesis | 22:03 | |
| dim__ | I found the problem | 22:06 |
| dim__ | it's due to VIP address | 22:06 |
| dim__ | I used another HA IP and it worked | 22:07 |
| dim__ | http://192.168.30.11:6082/spice_auto.html?token=9251e603-0afb-4208-853d-8c99e43259d5&title=myserver1(4db49a67-79ed-43fa-84cc-55d0ef9698e8) | 22:07 |
| dim__ | instead of 192.168.30.10 | 22:07 |
| *** TxGirlGeek has quit IRC | 22:08 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:08 | |
| *** hamza21 has quit IRC | 22:12 | |
| *** Guy has joined #openstack-ansible | 22:12 | |
| idlemind | dim__ ahh gotcha | 22:15 |
| idlemind | good times | 22:15 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-galera_server master: Set proper yum repo priority on CentOS 7 https://review.openstack.org/541924 | 22:15 |
| dim__ | thank you so much | 22:15 |
| idlemind | not a problem now you get to help the next guy up :) | 22:15 |
| *** sar has quit IRC | 22:18 | |
| *** TxGirlGeek has quit IRC | 22:20 | |
| dim__ | Awesome group I will try to help as well sometimes | 22:22 |
| *** vnogin has joined #openstack-ansible | 22:23 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:24 | |
| *** TxGirlGeek has quit IRC | 22:26 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:26 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Collect logs to a subfolder properly https://review.openstack.org/541426 | 22:30 |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Add a status line for SELinux status https://review.openstack.org/541371 | 22:30 |
| *** TxGirlGeek has quit IRC | 22:30 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:38 | |
| *** dariko has quit IRC | 22:40 | |
| *** dariko has joined #openstack-ansible | 22:41 | |
| *** woodard has joined #openstack-ansible | 22:45 | |
| *** tux_ has quit IRC | 22:46 | |
| *** tux_ has joined #openstack-ansible | 22:47 | |
| *** TxGirlGeek has quit IRC | 22:47 | |
| *** tux_ has quit IRC | 22:51 | |
| *** dariko has quit IRC | 22:53 | |
| *** acormier has joined #openstack-ansible | 22:53 | |
| *** acormier has quit IRC | 22:53 | |
| *** acormier has joined #openstack-ansible | 22:54 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:55 | |
| *** pmannidi has joined #openstack-ansible | 22:55 | |
| openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_octavia stable/pike: Zuul: Remove project name https://review.openstack.org/541932 | 22:59 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Zuul: Remove project name https://review.openstack.org/541093 | 23:05 |
| *** masber has joined #openstack-ansible | 23:27 | |
| *** acormier has quit IRC | 23:34 | |
| *** acormier has joined #openstack-ansible | 23:35 | |
| *** dave-mccowan has quit IRC | 23:45 | |
| *** markvoelker_ has quit IRC | 23:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Replace virtualenv-tools by a script https://review.openstack.org/533648 | 23:46 |
| *** dave-mccowan has joined #openstack-ansible | 23:48 | |
| *** acormier has quit IRC | 23:50 | |
| *** acormier has joined #openstack-ansible | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!