| *** jawad_axd has joined #openstack-ansible | 00:04 | |
| *** jawad_axd has quit IRC | 00:08 | |
| *** persia has quit IRC | 00:19 | |
| *** cshen has joined #openstack-ansible | 00:21 | |
| cshen | \ | 00:22 |
|---|---|---|
| *** persia has joined #openstack-ansible | 00:24 | |
| *** ivve has quit IRC | 00:25 | |
| *** cshen has quit IRC | 00:37 | |
| *** cshen has joined #openstack-ansible | 00:40 | |
| *** jawad_axd has joined #openstack-ansible | 00:46 | |
| *** dave-mccowan has joined #openstack-ansible | 00:47 | |
| *** gyee has quit IRC | 00:48 | |
| *** persia has quit IRC | 00:48 | |
| *** macz has quit IRC | 00:49 | |
| *** jawad_axd has quit IRC | 00:50 | |
| *** persia has joined #openstack-ansible | 00:54 | |
| *** nicolasbock has joined #openstack-ansible | 00:57 | |
| *** jawad_axd has joined #openstack-ansible | 01:07 | |
| *** nicolasbock1 has joined #openstack-ansible | 01:07 | |
| *** nicolasbock has quit IRC | 01:07 | |
| *** nicolasbock1 is now known as nicolasbock | 01:07 | |
| *** jawad_axd has quit IRC | 01:11 | |
| *** nicolasbock has quit IRC | 01:14 | |
| snadge | fatal: [aio1_horizon_container-3262784b]: FAILED! => {"changed": false, "msg": "http://download.fedoraproject.org/pub/epel/7/x86_64/repodata/56d9b42d7df7fcdcf6101bb2757920071fe893cb3bb6020b45861a071a968be6-updateinfo.xml.bz2: [Errno 14] HTTP Error 404 - Not Found | 01:29 |
| snadge | looking into it.. the other containers passed.. and it seems that one of the mirrors in "download.fedoraproject.org" is missing a file | 01:30 |
| *** nicolasbock has joined #openstack-ansible | 01:31 | |
| snadge | so I'm not really sure what the fix is here.. retry if there is a 404.. or force using a specific mirror instead of a generic download like "download.fedoraproject.org" which seems to assign a mirror somewhat randomly | 01:31 |
| *** dave-mccowan has quit IRC | 01:43 | |
| *** jawad_axd has joined #openstack-ansible | 01:48 | |
| *** persia has quit IRC | 01:48 | |
| *** nicolasbock has quit IRC | 01:51 | |
| *** jawad_axd has quit IRC | 01:53 | |
| *** persia has joined #openstack-ansible | 01:54 | |
| *** nicolasbock has joined #openstack-ansible | 01:58 | |
| *** dave-mccowan has joined #openstack-ansible | 01:59 | |
| *** schwicht has joined #openstack-ansible | 02:03 | |
| *** dave-mccowan has quit IRC | 02:08 | |
| openstackgerrit | Qitao proposed openstack/openstack-ansible-os_swift master: Update invaild link for Conf https://review.opendev.org/682874 | 02:18 |
| openstackgerrit | Qitao proposed openstack/openstack-ansible-os_swift master: Update invaild link for Conf https://review.opendev.org/682874 | 02:19 |
| openstackgerrit | Erik Berg proposed openstack/openstack-ansible-os_neutron stable/train: l3 agent on network_hosts do dvr_snat, anywhere else dvr https://review.opendev.org/692939 | 02:26 |
| *** nurdie has joined #openstack-ansible | 02:32 | |
| *** nurdie_ has joined #openstack-ansible | 02:33 | |
| *** nurdie has quit IRC | 02:37 | |
| *** cshen has quit IRC | 02:41 | |
| *** cshen has joined #openstack-ansible | 02:44 | |
| *** persia has quit IRC | 02:48 | |
| *** nurdie has joined #openstack-ansible | 02:49 | |
| *** jawad_axd has joined #openstack-ansible | 02:50 | |
| *** nurdie_ has quit IRC | 02:52 | |
| *** nurdie has quit IRC | 02:53 | |
| *** persia has joined #openstack-ansible | 02:53 | |
| *** jawad_axd has quit IRC | 02:55 | |
| *** cshen has quit IRC | 02:57 | |
| *** cshen has joined #openstack-ansible | 02:57 | |
| *** lixingxing has joined #openstack-ansible | 03:07 | |
| snadge | why ask questions when you can just grep the source code.. set variable centos_epel_mirror ;) | 03:16 |
| *** cshen has quit IRC | 03:16 | |
| *** cshen has joined #openstack-ansible | 03:21 | |
| *** nicolasbock1 has joined #openstack-ansible | 03:22 | |
| *** nicolasbock has quit IRC | 03:23 | |
| *** nicolasbock1 is now known as nicolasbock | 03:23 | |
| *** cshen has quit IRC | 03:26 | |
| *** cshen has joined #openstack-ansible | 03:33 | |
| *** lixingxing has quit IRC | 03:43 | |
| *** lixingxing has joined #openstack-ansible | 03:46 | |
| *** cshen has quit IRC | 03:47 | |
| *** cshen has joined #openstack-ansible | 03:50 | |
| *** rohit02 has joined #openstack-ansible | 04:02 | |
| *** cshen has quit IRC | 04:12 | |
| *** cshen has joined #openstack-ansible | 04:17 | |
| *** nicolasbock has quit IRC | 04:19 | |
| *** lixingxing has quit IRC | 04:19 | |
| *** cshen has quit IRC | 04:32 | |
| *** pcaruana has joined #openstack-ansible | 05:41 | |
| *** cshen has joined #openstack-ansible | 05:46 | |
| *** persia has quit IRC | 05:48 | |
| *** persia has joined #openstack-ansible | 05:48 | |
| *** gokhani has joined #openstack-ansible | 05:54 | |
| *** cshen has quit IRC | 06:26 | |
| *** nicolasbock has joined #openstack-ansible | 06:39 | |
| *** cshen has joined #openstack-ansible | 06:41 | |
| *** gokhani has quit IRC | 06:52 | |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_keystone master: Add possibility to overwrite public repo https://review.opendev.org/692872 | 06:55 |
| *** nicolasbock has quit IRC | 06:56 | |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-repo_server master: Add possibility to overwrite nginx public repo https://review.opendev.org/692874 | 06:58 |
| *** nicolasbock has joined #openstack-ansible | 07:00 | |
| *** cshen has quit IRC | 07:02 | |
| *** cshen has joined #openstack-ansible | 07:02 | |
| *** cshen has quit IRC | 07:07 | |
| *** prometheanfire has quit IRC | 07:08 | |
| *** prometheanfire has joined #openstack-ansible | 07:09 | |
| *** jawad_axd has joined #openstack-ansible | 07:15 | |
| *** lixingxing has joined #openstack-ansible | 07:19 | |
| *** shyamb has joined #openstack-ansible | 07:20 | |
| *** cshen has joined #openstack-ansible | 07:21 | |
| *** shyamb has quit IRC | 07:30 | |
| *** shyam89 has joined #openstack-ansible | 07:30 | |
| *** lixingxing has quit IRC | 07:38 | |
| *** luksky has joined #openstack-ansible | 07:39 | |
| *** lixingxing has joined #openstack-ansible | 07:41 | |
| *** shyamb has joined #openstack-ansible | 07:54 | |
| *** int-0x21 has joined #openstack-ansible | 07:56 | |
| int-0x21 | Hi is branch 20.0.0.0rc1 train or stein ? | 07:56 |
| *** cshen has quit IRC | 07:57 | |
| *** shyam89 has quit IRC | 07:58 | |
| *** shyamb has quit IRC | 08:00 | |
| *** cshen has joined #openstack-ansible | 08:10 | |
| *** miloa has joined #openstack-ansible | 08:18 | |
| rohit02 | hi team,to disable cinder volume lvm active active configuration error we have disabled the cluster= ceph option in cinder.conf but RBD volumes wont be created | 08:18 |
| rohit02 | in openstack-ansible stein...plzz help on this | 08:19 |
| jrosser | int-0x21: 20.0.0.0rc1 is the point at which the train branch was created | 08:25 |
| jrosser | so it is what the state of master branch was at that point, without any of the housekeeping patches needed after that point on the train branch | 08:26 |
| int-0x21 | jrosser, ah ok is there a better point if i want to test train ? | 08:29 |
| int-0x21 | or is it master then ? | 08:29 |
| *** tosky has joined #openstack-ansible | 08:30 | |
| *** cshen has quit IRC | 08:30 | |
| jrosser | master will be ussuri | 08:31 |
| int-0x21 | Or its stable/train maybe :P | 08:32 |
| openstackgerrit | ShangXiao proposed openstack/openstack-ansible-os_cinder master: Fix typos in the yml files https://review.opendev.org/692956 | 08:33 |
| miloa | Morning | 08:33 |
| jrosser | stable/train of Openstack-ansible repo *today* will use the SHA of all the ansible roles from the day the train branch was cut, basically | 08:33 |
| jrosser | and that’s not what you want | 08:34 |
| jrosser | so you won’t get all the patches to do things like switch Ubuntu cloud archive over from stein to train, for example | 08:35 |
| int-0x21 | then im confused, so if i want to do train what branch do i checkout :P | 08:36 |
| *** zbr|ooo is now known as zbr | 08:38 | |
| openstackgerrit | ShangXiao proposed openstack/openstack-ansible-os_heat master: Fix a typo https://review.opendev.org/692959 | 08:41 |
| jrosser | int-0x21: perhaps just check what the release model is for openstack projects https://releases.openstack.org/reference/release_models.html#cycle-trailing | 08:42 |
| *** gillesMo has joined #openstack-ansible | 08:42 | |
| jrosser | it should be clear that a deployment tool project like openstack-ansible cannot make a release untl after the main projects (nova/cinder/...) have all finalised their own releases and new stable branches | 08:43 |
| jrosser | in addition we have to wait for Ubuntu Cloud Archive and Redhat RDO repos to become available before we can really start work in OSA on a concrete release for Train | 08:43 |
| jrosser | plus the volunteer effort also has to be available to make that all happen and test it | 08:44 |
| jrosser | so currently, OSA has cut a stable/train branch, which implicitly makes the 20.0.0.0rc1 tag | 08:44 |
| jrosser | and we have a checklist of tasks as part of making a release which are being worked through | 08:45 |
| int-0x21 | Yea i just thought to make the time while i test out openstack and at the same time maybe give some feedback | 08:45 |
| jrosser | right - sure, so if you check out stable/train today you will find things not quite as you expect | 08:46 |
| int-0x21 | I was trying out master before but relised i lacked some openstack basic knowlage so i went and did some manual installation and now im back to retry openstack-ansible | 08:46 |
| int-0x21 | So basicly if i want to give some comunity feedback and testing at the same time its best if i go for master ? | 08:47 |
| int-0x21 | bug hunting while i learn more about openstack basicly | 08:47 |
| openstackgerrit | ShangXiao proposed openstack/openstack-ansible-os_magnum master: Fix a typo https://review.opendev.org/692960 | 08:48 |
| jrosser | int-0x21: there is nothing wrong with learning openstack using master | 08:49 |
| int-0x21 | excelent | 08:49 |
| jrosser | this is just an odd time on the stable/train branch because it it right between our rc1 and rc2 for train | 08:49 |
| int-0x21 | Ah | 08:50 |
| jrosser | and here is why https://github.com/openstack/openstack-ansible/blob/stable/train/ansible-role-requirements.yml#L59 | 08:51 |
| int-0x21 | I have a little setup with a deployment host running maas and 8 servers so i reroll pretty fast for clean tests | 08:51 |
| jrosser | for example, on the train branch that pulls in an ansible role by a specific SHA | 08:51 |
| jrosser | and i know that SHA does not include updating UCA to Train | 08:52 |
| jrosser | at rc2 all of those SHA will get updated to bring in all the work we have done on the ansible roles *after* rc1 | 08:53 |
| int-0x21 | I do like openstack asnible as a deployment method, installing openstack manualy is quite time consuming and non trivial | 08:55 |
| int-0x21 | juju was nice as a test but for production it feels bit unstable, juju can quite easility get into some broken loops | 08:56 |
| int-0x21 | tripleo i think i almost fell asleep reading the documentation so that might be good but sooooooo unstructured i died a little on the inside :P | 08:57 |
| int-0x21 | So far ansible is the most reproducible and straightforward deployment method so far :P | 08:57 |
| jrosser | noonedeadpunk: what do you think we still need to do for rc2? | 08:58 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible stable/train: Run setup module prior to bootstrapping aio https://review.opendev.org/692962 | 09:07 |
| *** nicolasbock has quit IRC | 09:08 | |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_keystone master: Add possibility to overwrite public repo https://review.opendev.org/692872 | 09:09 |
| *** lixingxing has quit IRC | 09:11 | |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-repo_server master: Add possibility to overwrite nginx public repo https://review.opendev.org/692874 | 09:11 |
| *** tosky has quit IRC | 09:13 | |
| *** rgogunskiy has joined #openstack-ansible | 09:14 | |
| *** arxcruz is now known as arxcruz|off | 09:24 | |
| *** prometheanfire has quit IRC | 09:26 | |
| *** tosky has joined #openstack-ansible | 09:35 | |
| *** DanyC has joined #openstack-ansible | 09:36 | |
| *** DanyC has quit IRC | 09:37 | |
| *** DanyC has joined #openstack-ansible | 09:39 | |
| *** DanyC has quit IRC | 09:40 | |
| *** DanyC has joined #openstack-ansible | 09:41 | |
| *** sum12 has quit IRC | 09:44 | |
| *** shyamb has joined #openstack-ansible | 10:03 | |
| *** aedc has quit IRC | 10:03 | |
| *** sshnaidm|rover|b is now known as sshnaidm|rover | 10:13 | |
| openstackgerrit | Mikael Loaec proposed openstack/openstack-ansible-os_horizon stable/rocky: [WIP]Cleanup pyc and pyo panels files https://review.opendev.org/692804 | 10:15 |
| *** radeks has joined #openstack-ansible | 10:21 | |
| *** rgogunskiy has quit IRC | 10:31 | |
| *** CobHead has joined #openstack-ansible | 10:38 | |
| CobHead | Hi everyone! I have a question: Say I want to test this patch with OpenStack Ansible: https://review.opendev.org/#/c/691748/ Since the patch has not yet been merged into stable/rocky, it is not part of a tree. As far as I know, OSA does not have any way to cherry-pick a certain commit to in the repo-* playbooks. If anyone could prove me otherwise, that would make me _very_ happy. | 10:41 |
| jrosser | CobHead: that is possible | 10:42 |
| CobHead | Sounds good! I tried pointing repo_packages/openstack_services.yml to that commit ID, but the repo build then fails since the commit is not in a tree. | 10:43 |
| jrosser | ok, so you found that, cool | 10:44 |
| CobHead | Been hacking my way trying to find a way to patch using OSA, you see ;) | 10:44 |
| jrosser | if you go to https://review.opendev.org/#/c/691748/ | 10:44 |
| jrosser | and look at the top right, "download" | 10:45 |
| CobHead | Yes | 10:45 |
| jrosser | you can use the repo and branch git fetch https://review.opendev.org/openstack/octavia refs/changes/48/691748/1 | 10:46 |
| jrosser | argh | 10:46 |
| jrosser | repo https://review.opendev.org/openstack/octavia | 10:46 |
| *** ivve has joined #openstack-ansible | 10:46 | |
| CobHead | In openstack_services.yml? Or do I specifically have to fetch that commit and somehow point it locally? | 10:46 |
| jrosser | and i think the branch would be refs/changes/48/691748/1 | 10:46 |
| jrosser | actually here is an example - i did exactly this to test an unmerged changed to keystone https://review.opendev.org/#/c/682460/3/playbooks/defaults/repo_packages/openstack_services.yml | 10:47 |
| jrosser | so thats one way - alternatively, fork the repo onto github or local gitlab for example, and apply that patch yourself on a test branch | 10:48 |
| jrosser | and update openstack_services.yml to point to ut | 10:48 |
| jrosser | *it | 10:48 |
| CobHead | I'll try it and let you know after I rerun the build :d | 10:48 |
| CobHead | I should probably create a workflow to clone certain commits I want to check out and merge them to a test branch sometime, yeah. | 10:51 |
| CobHead | Would make things easier for me :d | 10:51 |
| CobHead | Deploying now | 10:52 |
| jrosser | CobHead: oh actually there is also another way to override these things | 10:53 |
| jrosser | like this http://paste.openstack.org/show/785797/ in your user_variables.yml | 10:54 |
| CobHead | Ahh | 10:54 |
| CobHead | Makes sense | 10:54 |
| CobHead | jrosser, unfortunately, the first suggestion did not work ;( | 11:02 |
| CobHead | Did not match any files known to git | 11:02 |
| jrosser | boo :( | 11:02 |
| jrosser | that approach did work for my test patch for keystone | 11:04 |
| jrosser | but on master | 11:04 |
| CobHead | I believe that will break the requirements in Rocky, as master is currently using Python 3.x while Rocky is at Python 2.x | 11:05 |
| CobHead | I tried pulling the upstream commit for that fix, but that made OSA err out when building the wheels | 11:05 |
| CobHead | This is not ideal, as it is a blocker for testing patches without having to merge this into a test repo of my own making :d | 11:07 |
| jrosser | I really meant the technique for taking the repo and branch from gerrit works | 11:10 |
| jrosser | that ought to work, you may need to delete the git repo from inside the repo. server first | 11:12 |
| jrosser | but it’s been a while since I did stuff like this with rocky :) | 11:13 |
| CobHead | Gonna try taking the commit ID and apply that then :d | 11:13 |
| CobHead | I also noticed that the repo build exits when it fails to clone any of the repositories. Found that interesting as it was protesting about neutron-lbaas, which in practice doesn't exist anymore | 11:15 |
| CobHead | So I just added ignore_errors: yes to the roles | 11:15 |
| * CobHead shrugs | 11:15 | |
| CobHead | Let's see how this fares then.. | 11:17 |
| CobHead | So now I have octavia_git_repo: https://review.opendev.org/openstack/octavia and octavia_git_install_branch: <commit-id-of-patch> | 11:23 |
| CobHead | That made OSA err out too. | 11:23 |
| CobHead | Because it is not a tree. | 11:24 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_keystone master: Add possibility to overwrite public repo https://review.opendev.org/692872 | 11:26 |
| jrosser | CobHead: urgh.... perhaps your own branch on GitHub or somewhere? | 11:29 |
| CobHead | Seems like that is the only option, although I sincerely wish it was a way to patch earlier versions using OSA :( | 11:30 |
| CobHead | From the official OS repos, that is. | 11:31 |
| *** shyamb has quit IRC | 11:39 | |
| jrosser | CobHead: well it does, overriding the git repos and branches is exactly that mechanism | 11:42 |
| jrosser | and if you need something more complex, then you should host your own forked repos | 11:43 |
| *** cshen has joined #openstack-ansible | 11:45 | |
| *** cshen has quit IRC | 11:51 | |
| CobHead | jrosser: It does work, assuming that the patch is merged into the stable branch/master | 11:58 |
| CobHead | So I forked the repository, pulled that fix and will try again. | 11:58 |
| *** luksky has quit IRC | 12:13 | |
| *** schwicht has quit IRC | 12:26 | |
| *** nsmeds has joined #openstack-ansible | 12:27 | |
| noonedeadpunk | jrosser: for rc2 I need to do manual bumps, as automatical one faild because of spice repo (which wasn't working during weekends) | 12:27 |
| noonedeadpunk | I'm working on them right now | 12:28 |
| *** born2bake has joined #openstack-ansible | 12:34 | |
| jrosser | noonedeadpunk: this is needed for anyone doing ceph bigger than an AIO https://review.opendev.org/#/c/692921/ | 12:39 |
| noonedeadpunk | oh, that's what ThiagoCMC faced with yestarday? | 12:40 |
| jrosser | yes it is | 12:40 |
| jrosser | still i have not investigated why the AIO does not break with this | 12:40 |
| born2bake | Hi guys, I have a question related to networking per-configuration. I want to deploy openstack on 3 different machines. I am not networking guy but those 3 machines were preconfigured with vlans http://prntscr.com/psq4j6 ; When I was deploying openstack before, I know that I need minimum 2 network interfaces. One for provider network and second | 12:41 |
| born2bake | network interface without ip address for floating ip. With these network configuration http://prntscr.com/psq4j6 with VLAN 10,20,30; What interface/vlan I shall use for what provider network and can I use any vlan with ip for raw net-interface for floating ip? Thank you | 12:41 |
| noonedeadpunk | oh, guilhermesp already voted :) | 12:41 |
| guilhermesp | yep just voted :P | 12:41 |
| *** tosky_ has joined #openstack-ansible | 12:42 | |
| jrosser | born2bake: openstack-ansible lets you decide how you want to do the networking, but there are some minimum requirements | 12:43 |
| born2bake | The thing is that I cant touch networking configuration on OS machines | 12:44 |
| born2bake | I just need to adapt openstack deployment using these vlans | 12:44 |
| jrosser | i think you will have trouble then, becasue there is an expectation from openstack-ansible that you provide br-mgmt, br-storage, br-vxlan, br-vlan | 12:45 |
| *** tosky has quit IRC | 12:45 | |
| born2bake | Their idea is VLAN 10 - provider, VLAN 20 - overlay, VLAN 30 - openstack | 12:45 |
| jrosser | born2bake: just to double check you are wanting to deploy openstack using openstack-ansible? | 12:46 |
| born2bake | ..kolla-ansible :( | 12:47 |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Bump SHAs for stable/rocky https://review.opendev.org/692749 | 12:47 |
| jrosser | you will get better advice in #openstack-kolla for that, we are working on a different ansible based openstack deployment tool here | 12:48 |
| *** schwicht has joined #openstack-ansible | 12:50 | |
| *** Garyx has quit IRC | 12:53 | |
| *** tosky_ is now known as tosky | 12:59 | |
| *** schwicht has quit IRC | 13:21 | |
| CobHead | jrosser: Is there a cache for Git that OSA deploys somewhere? Even though I specified for it to use a repo I made, I can still see that it uses the official rocky branch | 13:30 |
| jrosser | yes, i said earlier you may have to look in the repo container and delete the cached repo from in there | 13:31 |
| jrosser | particularly if you are changing the upstrem repo location entirely | 13:31 |
| CobHead | The entire repo folder which is under var/www? | 13:31 |
| jrosser | no, just the repo for the service that you are forking | 13:31 |
| jrosser | likey located here https://github.com/openstack/openstack-ansible-repo_build/blob/stable/rocky/defaults/main.yml#L49 | 13:32 |
| *** int-0x21 has quit IRC | 13:32 | |
| jrosser | and this variable may also be helpful https://github.com/openstack/openstack-ansible-repo_build/blob/stable/rocky/defaults/main.yml#L56 | 13:33 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/stein: Bump SHAs for stable/stein https://review.opendev.org/693002 | 13:33 |
| CobHead | jrosser: repo_build_git_reclone is set to yes in user_variables | 13:33 |
| jrosser | you should be able to check in /var/www/openstackgit/<service> if it has been cloned from the right place and checked out to the point you want | 13:34 |
| jrosser | if not, delete the repo and run the repo_build playbook | 13:34 |
| *** lixingxing has joined #openstack-ansible | 13:35 | |
| CobHead | Deleted and will attempt rebuild. | 13:35 |
| *** luksky has joined #openstack-ansible | 13:37 | |
| CobHead | jrosser: Also, out of curiosity - OSA seems to only get the repos on one repo container, whereas I have 3 of them in my setup | 13:39 |
| CobHead | Any particular reason as to why it doesn't just do it for all 3 of them? | 13:39 |
| jrosser | CobHead: rocky is a long time ago :) | 13:41 |
| jrosser | but perhaps lsyncd is set up to syncronise those repos to all of the repo containers | 13:41 |
| jrosser | repo[0] is the master and lsyncd/rsync organises replicating the content to the others | 13:41 |
| *** cshen has joined #openstack-ansible | 13:44 | |
| *** lixingxing has quit IRC | 13:48 | |
| *** lixingxing has joined #openstack-ansible | 13:48 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/train: Fix a-r-r trackbranches https://review.opendev.org/692838 | 13:52 |
| *** macz has joined #openstack-ansible | 13:55 | |
| CobHead | After deleting the repo and rerunning the repo_build playbook, I can still not see the changes on the repo container. It looks to me that it is still pulling from the Openstack repos, even though it should not. | 14:01 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/693008 | 14:06 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible master: Fix requirements_git_url call https://review.opendev.org/693011 | 14:12 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/train: l3 agent on network_hosts do dvr_snat, anywhere else dvr https://review.opendev.org/692939 | 14:13 |
| *** goldyfruit_ has joined #openstack-ansible | 14:19 | |
| jamesdenton | good morning | 14:20 |
| *** sum12 has joined #openstack-ansible | 14:24 | |
| *** nurdie has joined #openstack-ansible | 14:25 | |
| CobHead | I found the source of my misery, jrosser | 14:28 |
| CobHead | octavia_venv_download_url: http://127.0.0.1/venvs/untagged/ubuntu/octavia.tgz | 14:28 |
| CobHead | I believed that OSA took the actual repo name and renamed the variable accordingly. Sure doesn't seem that way. | 14:29 |
| CobHead | Going to change that now. | 14:29 |
| *** goldyfruit___ has joined #openstack-ansible | 14:34 | |
| *** goldyfruit_ has quit IRC | 14:36 | |
| *** lixingxing has quit IRC | 14:42 | |
| *** rohit02 has quit IRC | 14:43 | |
| *** schwicht has joined #openstack-ansible | 14:45 | |
| *** admin0 has joined #openstack-ansible | 14:54 | |
| *** jawad_axd has quit IRC | 15:00 | |
| *** jawad_axd has joined #openstack-ansible | 15:04 | |
| *** schwicht has quit IRC | 15:04 | |
| *** jawad_axd has quit IRC | 15:09 | |
| CobHead | jrosser: I finally fixed it. I want to thank you for your time and assistance in wrapping my head around this :) | 15:12 |
| *** jamesdenton has quit IRC | 15:13 | |
| *** jamesdenton has joined #openstack-ansible | 15:15 | |
| *** jawad_axd has joined #openstack-ansible | 15:18 | |
| *** macz has quit IRC | 15:23 | |
| *** macz has joined #openstack-ansible | 15:25 | |
| *** admin0 has quit IRC | 15:33 | |
| openstackgerrit | Erik Berg proposed openstack/openstack-ansible-os_neutron stable/stein: l3 agent on network_hosts do dvr_snat, anywhere else dvr https://review.opendev.org/693031 | 15:34 |
| *** admin0 has joined #openstack-ansible | 15:49 | |
| openstackgerrit | Dave Wilde (d34dh0r53) proposed openstack/openstack-ansible-os_keystone master: Standardize on nginx-extras https://review.opendev.org/693033 | 15:51 |
| *** admin0 has quit IRC | 16:02 | |
| *** jawad_axd has quit IRC | 16:02 | |
| *** cshen has quit IRC | 16:10 | |
| *** schwicht has joined #openstack-ansible | 16:11 | |
| *** jawad_axd has joined #openstack-ansible | 16:14 | |
| *** jawad_axd has quit IRC | 16:19 | |
| *** prometheanfire has joined #openstack-ansible | 16:19 | |
| *** hamzaachi has joined #openstack-ansible | 16:26 | |
| *** luksky has quit IRC | 16:31 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Restart mysql when package is installed https://review.opendev.org/693043 | 16:32 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Switch upgrade jobs to Ussuri cycle https://review.opendev.org/692509 | 16:34 |
| *** jawad_axd has joined #openstack-ansible | 16:35 | |
| *** jawad_axd has quit IRC | 16:39 | |
| noonedeadpunk | wow. with this migration from summer time to normal one I've forgot about osa meeting.... | 16:52 |
| noonedeadpunk | I was just thinking about starting it when I've realized it's already passed | 16:53 |
| openstackgerrit | Merged openstack/openstack-ansible stable/train: Run setup module prior to bootstrapping aio https://review.opendev.org/692962 | 16:54 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/693008 | 16:55 |
| jrosser | CobHead: excellent good news you got it to work | 16:55 |
| *** luksky has joined #openstack-ansible | 16:57 | |
| *** nurdie has quit IRC | 16:58 | |
| *** nurdie has joined #openstack-ansible | 16:58 | |
| *** nurdie has quit IRC | 17:04 | |
| *** nurdie has joined #openstack-ansible | 17:08 | |
| *** radeks has quit IRC | 17:12 | |
| *** aedc has joined #openstack-ansible | 17:32 | |
| *** gyee has joined #openstack-ansible | 17:33 | |
| *** DanyC_ has joined #openstack-ansible | 17:36 | |
| *** DanyC has quit IRC | 17:39 | |
| *** jamesdenton has quit IRC | 17:46 | |
| *** jamesdenton has joined #openstack-ansible | 17:49 | |
| *** also_stingrayza has joined #openstack-ansible | 17:52 | |
| *** stingrayza has quit IRC | 17:53 | |
| *** goldyfruit___ has quit IRC | 17:53 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-rabbitmq_server master: Ansible 'search' is a test not a filter https://review.opendev.org/693061 | 17:55 |
| noonedeadpunk | jrosser: dunno how ^ is passing on master..... | 17:55 |
| noonedeadpunk | * 2.9 passing on master without that patch | 17:56 |
| noonedeadpunk | btw I've just reproduce failure on rabbitmq during upgrade | 17:58 |
| *** goldyfruit___ has joined #openstack-ansible | 17:58 | |
| *** jamesdenton has quit IRC | 17:59 | |
| *** miloa has quit IRC | 18:03 | |
| *** miloa has joined #openstack-ansible | 18:03 | |
| *** prometheanfire has quit IRC | 18:04 | |
| *** prometheanfire has joined #openstack-ansible | 18:06 | |
| *** schwicht has quit IRC | 18:08 | |
| miloa | I think i found a problem in os_horizon for distrib install stable/rocky in task "Enable project panels" in horizon_post_install.yml it works for source install but for distro the "item.state" is "file" so basically it checks if the file exists but do not copy the file if not and the panel is enabled. For source install it use "link". | 18:13 |
| miloa | https://github.com/openstack/openstack-ansible-os_horizon/blob/stable/rocky/tasks/horizon_post_install.yml#L55 | 18:13 |
| noonedeadpunk | jrosser: they did really smth wrong with this runpy.py as instead of error message generated by tool they just throw exception that reported error is not in json | 18:13 |
| *** schwicht has joined #openstack-ansible | 18:16 | |
| noonedeadpunk | miloa: for distro installs I guess it just checks that it exists | 18:16 |
| noonedeadpunk | as there's no need in symlinking anything - distro should install panels into correct path for them to work out of the box | 18:16 |
| noonedeadpunk | ah, in rocky | 18:18 |
| noonedeadpunk | Not sure about rocky:) I guess I've re-written the way how dashboards are included in stein | 18:19 |
| *** jamesdenton has joined #openstack-ansible | 18:19 | |
| miloa | noonedeadpunk: yes i saw that :) | 18:20 |
| miloa | noonedeadpunk: the use case is, first i set heat_ui to false, then to true and as there is no file because we set it first to false it fail :) | 18:21 |
| *** hamzaachi has quit IRC | 18:22 | |
| noonedeadpunk | ah, I see. Yeah, that's possible:) | 18:22 |
| noonedeadpunk | you may kinda drop lxc container or venv and re-deploy it :p | 18:23 |
| noonedeadpunk | but yeh, I see what you mean. | 18:23 |
| miloa | noonedeadpunk: i've to go, I will try to find a way tomorrow perhaps using "link" instead of "file" or as you propose drop horizon lxc containers and recreate it. Thanks :) | 18:25 |
| noonedeadpunk | link is probably not going to work... | 18:25 |
| *** cshen has joined #openstack-ansible | 18:26 | |
| miloa | noonedeadpunk: btw, there is an othe problem when putting heat_ui or magnum or LaaBS to false the pyc and pyo are still present so the panels do not disappear, so i had to remove pyc and pyo by hand. https://review.opendev.org/#/c/692804/ :) | 18:28 |
| *** miloa has quit IRC | 18:31 | |
| *** cshen has quit IRC | 18:31 | |
| *** also_stingrayza has quit IRC | 18:35 | |
| *** stingrayza has joined #openstack-ansible | 18:36 | |
| noonedeadpunk | miloa I think it's not going to work that way - pyc file is only created after panel accessed first time | 18:37 |
| noonedeadpunk | it passes though... | 18:38 |
| *** CeeMac has quit IRC | 18:45 | |
| *** goldyfruit_ has joined #openstack-ansible | 18:46 | |
| *** goldyfruit___ has quit IRC | 18:49 | |
| jrosser | noonedeadpunk: I wonder if rabbitmq_upgrade is set properly? | 18:51 |
| jrosser | it would skip that task and not use the search test if any of the previous conditions fail | 18:52 |
| jrosser | afaik they are evaluated in sequence | 18:52 |
| noonedeadpunk | I guess yes and for me lookss like problem in module itself.... | 18:52 |
| noonedeadpunk | so it fails on that json.loads https://github.com/ansible/ansible/blob/devel/lib/ansible/modules/messaging/rabbitmq/rabbitmq_policy.py#L191 | 18:52 |
| noonedeadpunk | but policy_data is ['vhost', 'name', 'pattern', 'apply-to', 'definition', 'priority'] | 18:53 |
| noonedeadpunk | while http://paste.openstack.org/show/785816/ | 18:53 |
| noonedeadpunk | I need to run away - if you wish I may place your ssh key on a vm | 18:53 |
| noonedeadpunk | But I'm going to continue digging in what's going on here | 18:54 |
| noonedeadpunk | so for me looks like it takes wrong line | 18:55 |
| noonedeadpunk | jrosser: yeah, they take from 1st element https://github.com/ansible/ansible/blob/devel/lib/ansible/modules/messaging/rabbitmq/rabbitmq_policy.py#L137 | 18:56 |
| noonedeadpunk | looks like in 3.8.0 they've added one more header:) | 18:57 |
| noonedeadpunk | I'll try to patch it then later | 18:57 |
| jrosser | noonedeadpunk: cool sounds like you have good understanding there | 18:58 |
| noonedeadpunk | as they can do much simplier http://paste.openstack.org/show/785817/ | 18:58 |
| jrosser | I’m just travelling now so also I’m busy for a few hours | 18:59 |
| noonedeadpunk | oh, sorry for pinging you:) | 18:59 |
| jrosser | oh don’t worry | 18:59 |
| jrosser | sat at the train station | 18:59 |
| noonedeadpunk | most likely I'll take care about it only tomorrow... As will need to check previous versions of rabbitmq output | 18:59 |
| guilhermesp | noonedeadpunk: jrosser I suspect when we re-run ceph-install it created the ceph dir with ceph. ownership https://github.com/ceph/ceph-ansible/blob/54365389f816ec9285063a5d976c6d9ac350afa0/roles/ceph-config/tasks/main.yml#L66 | 19:00 |
| ThiagoCMC | Hey guys, the `stable/train` branch is ignoring my "/etc/openstack_deploy/env.d/neutron.yml" file. Contents: https://pastebin.com/4CB66Stf | 19:01 |
| guilhermesp | with our ceph_client we create the keyring for services with the correct ownership for the services https://github.com/openstack/openstack-ansible-ceph_client/blob/1fc2b7a7edaad0ca761fcfc3369b82bf2d476d5f/tasks/ceph_auth_extra.yml#L60 | 19:01 |
| ThiagoCMC | The idea of this file (in stable/stein) is to isolate the neutron_server container within the controllers. It worked. | 19:01 |
| ThiagoCMC | Not, it isn't working! | 19:01 |
| ThiagoCMC | The neutron_server container is appearing within the Network Node! Where I don't want it. There, only the neutron-*-agents will be running, on bare-metal. | 19:02 |
| ThiagoCMC | Any idea? | 19:02 |
| *** hwoarang has quit IRC | 19:04 | |
| *** aedc has quit IRC | 19:07 | |
| *** hwoarang has joined #openstack-ansible | 19:08 | |
| *** hamzaachi has joined #openstack-ansible | 19:11 | |
| *** goldyfruit_ has quit IRC | 19:14 | |
| *** sshnaidm|rover is now known as sshnaidm|off | 19:16 | |
| ThiagoCMC | Maybe the new "/opt/openstack-ansible/inventory/env.d/neutron.yml" is overriding my old "/etc/openstack_deploy/env.d/neutron.yml" file? | 19:17 |
| *** hamzaachi_ has joined #openstack-ansible | 19:21 | |
| *** hamzaachi has quit IRC | 19:21 | |
| *** radeks has joined #openstack-ansible | 19:29 | |
| jrosser | ThiagoCMC: I don’t think we have changed anything really to do with env.d between stein and train | 19:29 |
| jrosser | is this an upgrade or a fresh deploy? | 19:30 |
| jrosser | ThiagoCMC: this is from my notes http://paste.openstack.org/show/777611/ | 19:32 |
| *** aedc has joined #openstack-ansible | 19:33 | |
| ThiagoCMC | This is a fresh deployment. | 19:39 |
| ThiagoCMC | Something is defintely different... | 19:40 |
| ThiagoCMC | Thanks for that pastebin note! | 19:40 |
| *** schwicht has quit IRC | 19:41 | |
| jrosser | it is always possible that there is some new bug py3 related with the inventory | 19:43 |
| jrosser | I’d probably be messing with a ansible-inventory and running dynamic-inventory.py manually and examining the output next | 19:43 |
| ThiagoCMC | Nice tips! I'll try that "network-infra_containers" or even just remove my custom file. | 19:47 |
| ThiagoCMC | And also to play with the dynamic-inventory.py ^_^ | 19:53 |
| ThiagoCMC | I just noticed that the neutron_server container is also appearing in all my Compute Nodes! O_O | 19:57 |
| ThiagoCMC | Removing the "/opt/openstack-ansible/inventory/env.d/neutron.yml" file didn't worked. | 19:58 |
| ThiagoCMC | I'll start this from scratch again... =) | 20:00 |
| *** jawad_axd has joined #openstack-ansible | 20:13 | |
| *** strattao has joined #openstack-ansible | 20:15 | |
| *** tosky has quit IRC | 20:16 | |
| *** hamzaachi_ has quit IRC | 20:19 | |
| *** hamzaachi_ has joined #openstack-ansible | 20:21 | |
| *** hamzaachi_ has quit IRC | 20:23 | |
| *** hamzaachi_ has joined #openstack-ansible | 20:24 | |
| *** zbr is now known as zbr|ooo | 20:29 | |
| *** hamzaachi_ has quit IRC | 20:29 | |
| *** nurdie has quit IRC | 20:30 | |
| *** nurdie has joined #openstack-ansible | 20:31 | |
| *** goldyfruit_ has joined #openstack-ansible | 20:33 | |
| *** radeks has quit IRC | 20:34 | |
| *** nurdie has quit IRC | 20:35 | |
| *** schwicht has joined #openstack-ansible | 20:50 | |
| *** aedc has quit IRC | 21:03 | |
| *** weshay is now known as weshay_ptg | 21:50 | |
| *** macz has joined #openstack-ansible | 21:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/stein: Add Centos GPG key custom url https://review.opendev.org/692462 | 21:56 |
| *** cshen has joined #openstack-ansible | 21:56 | |
| *** nurdie has joined #openstack-ansible | 21:59 | |
| *** nurdie_ has joined #openstack-ansible | 21:59 | |
| *** cshen has quit IRC | 22:01 | |
| *** pcaruana has quit IRC | 22:03 | |
| *** nurdie has quit IRC | 22:04 | |
| *** hamzaachi has joined #openstack-ansible | 22:11 | |
| *** jawad_axd has quit IRC | 22:12 | |
| *** hamzaachi has quit IRC | 22:12 | |
| *** cshen has joined #openstack-ansible | 22:18 | |
| *** schwicht has quit IRC | 22:19 | |
| *** cshen has quit IRC | 22:23 | |
| *** JoyBoy7 has joined #openstack-ansible | 22:25 | |
| JoyBoy7 | Has anyone had a similar problem when installing CEPH in OSA; | 22:26 |
| JoyBoy7 | fatal: [controller1_ceph-mon_container-c597603d]: FAILED! => {"msg": "template error while templating string: no filter named 'ips_in_ranges'. String: {{ _monitor_addresses | default([]) + [{ 'name': item, 'addr': hostvars[item]['ansible_all_ipv4_addresses'] | ips_in_ranges(hostvars[item]['monitor_address_block'].split(',')) | first }] }}"}fatal: | 22:26 |
| JoyBoy7 | [controller2_ceph-mon_container-0bc78030]: FAILED! => {"msg": "template error while templating string: no filter named 'ips_in_ranges'. String: {{ _monitor_addresses | default([]) + [{ 'name': item, 'addr': hostvars[item]['ansible_all_ipv4_addresses'] | ips_in_ranges(hostvars[item]['monitor_address_block'].split(',')) | first }] }}"} | 22:26 |
| *** JoyBoy762 has joined #openstack-ansible | 22:29 | |
| *** JoyBoy762 has left #openstack-ansible | 22:29 | |
| jrosser | JoyBoy7: https://review.opendev.org/#/c/692921/ | 22:30 |
| jamesdenton | jrosser is the MNAIO still a thing? | 22:33 |
| jrosser | should be | 22:33 |
| jrosser | i don't think theres anything there which will have rotted too much | 22:34 |
| jamesdenton | well, that was my next question. i'm happy to fix just didn't want it to be in vain | 22:34 |
| jamesdenton | running into some issues but i may have a PR tomorrow | 22:34 |
| jrosser | tbh i think jesse was doing a ton of work with it but maybe not many others | 22:34 |
| jrosser | if you have nice quick giant node it is a useful tool | 22:34 |
| jamesdenton | agreed | 22:36 |
| jrosser | actually theres some cool stuff got added | 22:36 |
| jrosser | the snapshot / restore of the VM would be super cool for rinse/repeat testing of deployments | 22:37 |
| jrosser | it was always pxebooting the VM that took forever | 22:37 |
| jamesdenton | prob worth updating to python3, too, which i'm in the middle of | 22:37 |
| jrosser | if you have a good use for the MNAIO i'd say it was worth fixing it up | 22:38 |
| jrosser | there is a lot of good stuff there | 22:38 |
| jamesdenton | *thumbs up* | 22:38 |
| *** nicolasbock has joined #openstack-ansible | 22:43 | |
| *** nicolasbock1 has joined #openstack-ansible | 22:46 | |
| *** nicolasbock1 has quit IRC | 22:47 | |
| *** nicolasbock has quit IRC | 22:48 | |
| JoyBoy7 | in my oenstack-ansible.rc i have this line export ANSIBLE_FILTER_PLUGINS="${ANSIBLE_FILTER_PLUGINS:-/etc/ansible/roles/plugins/filter}"its enough ir i should add another to fox my error | 22:48 |
| *** nicolasbock has joined #openstack-ansible | 22:49 | |
| *** luksky has quit IRC | 22:54 | |
| *** cshen has joined #openstack-ansible | 22:59 | |
| *** lixingxing has joined #openstack-ansible | 22:59 | |
| *** JoyBoy7 has quit IRC | 23:00 | |
| *** zbr|ooo has quit IRC | 23:00 | |
| *** nicolasbock1 has joined #openstack-ansible | 23:01 | |
| *** nicolasbock has quit IRC | 23:02 | |
| *** nicolasbock has joined #openstack-ansible | 23:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Drop common-db tag from db_sync task https://review.opendev.org/692161 | 23:04 |
| *** yoctozepto has quit IRC | 23:04 | |
| *** yoctozepto has joined #openstack-ansible | 23:04 | |
| *** nicolasbock1 has quit IRC | 23:06 | |
| *** cshen has quit IRC | 23:10 | |
| *** nurdie_ has quit IRC | 23:16 | |
| *** lixingxing has quit IRC | 23:17 | |
| *** nicolasbock1 has joined #openstack-ansible | 23:34 | |
| *** nicolasbock has quit IRC | 23:36 | |
| *** nicolasbock1 is now known as nicolasbock | 23:36 | |
| *** nicolasbock has quit IRC | 23:46 | |
| *** nicolasbock has joined #openstack-ansible | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!