| *** macza has quit IRC | 00:06 | |
| *** mathlin has joined #openstack-ansible | 00:11 | |
| *** cshen has joined #openstack-ansible | 00:15 | |
| *** cshen has quit IRC | 00:20 | |
| *** fatdragon has joined #openstack-ansible | 00:38 | |
| *** mathlin has quit IRC | 00:45 | |
| *** fatdragon has joined #openstack-ansible | 01:05 | |
| *** mathlin has joined #openstack-ansible | 01:08 | |
| *** strobelight has joined #openstack-ansible | 01:14 | |
| *** cshen has joined #openstack-ansible | 01:40 | |
| *** mathlin has quit IRC | 01:41 | |
| *** hwoarang has joined #openstack-ansible | 01:44 | |
| *** cshen has quit IRC | 01:46 | |
| *** hwoarang_ has quit IRC | 01:46 | |
| *** mino_ has joined #openstack-ansible | 01:50 | |
| *** macza has joined #openstack-ansible | 01:55 | |
| *** shardy has quit IRC | 01:57 | |
| *** hwoarang has quit IRC | 02:01 | |
| *** hwoarang has joined #openstack-ansible | 02:02 | |
| *** shardy has joined #openstack-ansible | 02:04 | |
| *** mathlin has joined #openstack-ansible | 02:05 | |
| *** DanyC has joined #openstack-ansible | 02:13 | |
| *** DanyC has quit IRC | 02:17 | |
| *** macza has quit IRC | 02:30 | |
| *** mathlin has quit IRC | 02:39 | |
| *** mathlin has joined #openstack-ansible | 03:02 | |
| *** hwoarang_ has joined #openstack-ansible | 03:30 | |
| *** hwoarang has quit IRC | 03:31 | |
| *** macza has joined #openstack-ansible | 03:32 | |
| *** mathlin has quit IRC | 03:36 | |
| *** macza has quit IRC | 03:36 | |
| *** shardy has quit IRC | 03:52 | |
| *** shardy has joined #openstack-ansible | 03:53 | |
| *** mathlin has joined #openstack-ansible | 03:59 | |
| *** mino_ has quit IRC | 04:03 | |
| *** udesale has joined #openstack-ansible | 04:12 | |
| *** spsurya has joined #openstack-ansible | 04:22 | |
| *** mathlin has quit IRC | 04:32 | |
| *** strobelight has quit IRC | 04:37 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone master: Cleanup files and templates using smart sources https://review.openstack.org/588960 | 04:48 |
|---|---|---|
| *** spsurya has quit IRC | 05:10 | |
| *** spsurya has joined #openstack-ansible | 05:13 | |
| *** cshen has joined #openstack-ansible | 05:15 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Synced tempest plugins names and services https://review.openstack.org/628926 | 05:17 |
| *** cshen has quit IRC | 05:20 | |
| *** mathlin has joined #openstack-ansible | 05:22 | |
| *** cshen has joined #openstack-ansible | 05:28 | |
| *** shyamb has joined #openstack-ansible | 05:37 | |
| *** cshen has quit IRC | 05:37 | |
| *** cshen has joined #openstack-ansible | 05:38 | |
| *** shyamb has quit IRC | 05:43 | |
| *** radeks has joined #openstack-ansible | 05:48 | |
| *** shyamb has joined #openstack-ansible | 05:51 | |
| *** cshen has quit IRC | 05:54 | |
| *** hwoarang has joined #openstack-ansible | 05:54 | |
| *** mathlin has quit IRC | 05:56 | |
| *** hwoarang_ has quit IRC | 05:57 | |
| *** hwoarang_ has joined #openstack-ansible | 06:08 | |
| *** hwoarang has quit IRC | 06:11 | |
| *** radeks has quit IRC | 06:11 | |
| *** fatdragon has quit IRC | 06:13 | |
| *** radeks has joined #openstack-ansible | 06:20 | |
| *** cshen has joined #openstack-ansible | 06:23 | |
| *** gyee has quit IRC | 06:24 | |
| *** cshen has quit IRC | 06:27 | |
| *** mathlin has joined #openstack-ansible | 06:46 | |
| *** shyamb has quit IRC | 06:50 | |
| *** shyamb has joined #openstack-ansible | 07:03 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-lxc_hosts master: Increasing LXC container default shutdown delay https://review.openstack.org/625998 | 07:04 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-lxc_hosts master: Use libeatmydata during LXC cache prep https://review.openstack.org/621509 | 07:05 |
| *** shyamb has quit IRC | 07:05 | |
| *** shyamb has joined #openstack-ansible | 07:06 | |
| *** hamzaachi has joined #openstack-ansible | 07:09 | |
| *** DanyC has joined #openstack-ansible | 07:13 | |
| *** DanyC has quit IRC | 07:18 | |
| *** mathlin has quit IRC | 07:20 | |
| *** cshen has joined #openstack-ansible | 07:24 | |
| *** shyamb has quit IRC | 07:25 | |
| *** cshen has quit IRC | 07:29 | |
| *** cshen has joined #openstack-ansible | 07:30 | |
| *** fatdragon has joined #openstack-ansible | 07:30 | |
| *** shyamb has joined #openstack-ansible | 07:32 | |
| fnpanic | hi | 07:39 |
| fnpanic | odyssey4me: Thanks for looking at the proxy patches for rocky | 07:39 |
| *** fatdragon has quit IRC | 07:42 | |
| *** pcaruana has joined #openstack-ansible | 07:42 | |
| *** jbadiapa has joined #openstack-ansible | 07:43 | |
| *** mathlin has joined #openstack-ansible | 07:43 | |
| *** shyamb has quit IRC | 07:46 | |
| *** gkadam has joined #openstack-ansible | 07:48 | |
| *** gkadam is now known as gkadam-afk | 07:50 | |
| *** kopecmartin|off is now known as kopecmartin | 08:00 | |
| chandankumar | jrosser: \o. | 08:04 |
| chandankumar | jrosser: https://review.openstack.org/628926 is good to go :-) | 08:05 |
| *** mwhahaha has quit IRC | 08:05 | |
| *** jungleboyj has quit IRC | 08:06 | |
| *** jungleboyj has joined #openstack-ansible | 08:06 | |
| *** mwhahaha has joined #openstack-ansible | 08:06 | |
| *** hamzaachi_ has joined #openstack-ansible | 08:08 | |
| *** hamzaachi has quit IRC | 08:10 | |
| *** priteau has joined #openstack-ansible | 08:12 | |
| *** mathlin has quit IRC | 08:17 | |
| *** shardy has quit IRC | 08:19 | |
| *** shardy has joined #openstack-ansible | 08:20 | |
| *** DanyC has joined #openstack-ansible | 08:21 | |
| *** gkadam-afk is now known as gkadam | 08:24 | |
| *** DanyC has quit IRC | 08:26 | |
| *** cshen has quit IRC | 08:34 | |
| *** tosky has joined #openstack-ansible | 08:37 | |
| jrosser | chandankumar: nice! done | 08:38 |
| jrosser | hopefully that makes the os_barbican and other role tests go# | 08:39 |
| *** mathlin has joined #openstack-ansible | 08:40 | |
| *** cshen has joined #openstack-ansible | 08:40 | |
| *** hamzaachi__ has joined #openstack-ansible | 08:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: cleanup: remove metadata checksum fixes tasks https://review.openstack.org/627792 | 08:46 |
| *** hamzaachi_ has quit IRC | 08:47 | |
| *** fatdragon has joined #openstack-ansible | 08:48 | |
| *** evrardjp_ has joined #openstack-ansible | 08:48 | |
| *** evrardjp__ has joined #openstack-ansible | 08:49 | |
| *** evrardjp has quit IRC | 08:51 | |
| *** evrardjp_ has quit IRC | 08:52 | |
| *** evrardjp__ has quit IRC | 08:53 | |
| *** fatdragon has quit IRC | 08:55 | |
| *** shyamb has joined #openstack-ansible | 08:56 | |
| *** thuydang has joined #openstack-ansible | 08:57 | |
| *** thuydang has quit IRC | 09:01 | |
| *** evrardjp has joined #openstack-ansible | 09:03 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_keystone master: Cleanup files and templates using smart sources https://review.openstack.org/588960 | 09:05 |
| *** mathlin has quit IRC | 09:14 | |
| *** sawblade6 has joined #openstack-ansible | 09:18 | |
| *** DanyC has joined #openstack-ansible | 09:20 | |
| openstackgerrit | Merged openstack/ansible-hardening master: cleanup: use updated conditionals https://review.openstack.org/629022 | 09:20 |
| odyssey4me | jrosser we may still have to tweak the roles a bit, but that'll help start things to move them towards a better working state | 09:29 |
| *** thuydang has joined #openstack-ansible | 09:31 | |
| *** mkuf_ has joined #openstack-ansible | 09:31 | |
| *** mkuf has quit IRC | 09:34 | |
| jrosser | evrardjp: re that arm patch being broken on metal, i think all the metal jobs are brok on master | 09:40 |
| evrardjp | mmm | 09:41 |
| evrardjp | sad | 09:41 |
| jrosser | i believe this may be to blame https://github.com/openstack/openstack-ansible-galera_client/commit/30cb1e72d6736bae9dacbed0c3e6ef8fb174aaeb | 09:41 |
| evrardjp | but I guess nobody cares. | 09:41 |
| jrosser | the current fail is galera server tripping over things that u | 09:42 |
| jrosser | *used to have been fiddled in the galera client role | 09:42 |
| evrardjp | I see | 09:42 |
| evrardjp | well I sadly don't have time to fix that rn | 09:43 |
| jrosser | i didnt know if it was a mistake removing that stuff from galera client or that the same thing needs doing in the server role to match | 09:44 |
| evrardjp | I don't know how it passed in the first place | 09:44 |
| evrardjp | well I have an idea but ... hey | 09:44 |
| evrardjp | I am not fully aware of the context of this patch in galera-client | 09:45 |
| jrosser | the galera client role test doesnt test on-metal, nor will it co-install galera_server role | 09:45 |
| odyssey4me | jrosser if galera_Server requires that logic, then we should move it there | 09:46 |
| *** vnogin has joined #openstack-ansible | 09:46 | |
| jrosser | it was in both - now only in one | 09:46 |
| odyssey4me | interesting - which metal job is now failing? | 09:47 |
| odyssey4me | both centos & opensuse | 09:47 |
| odyssey4me | not ubuntu | 09:47 |
| odyssey4me | not surprising, because all that stuff was to make centos/suse look like ubuntu | 09:48 |
| *** mathlin has joined #openstack-ansible | 09:49 | |
| *** Miouge has left #openstack-ansible | 09:50 | |
| *** Miouge has joined #openstack-ansible | 09:51 | |
| openstackgerrit | Arx Cruz proposed openstack/openstack-ansible-os_tempest master: Better tempest black and whitelist management https://review.openstack.org/621605 | 09:54 |
| jrosser | odyssey4me: should we just revert that change from galera_client? | 09:56 |
| odyssey4me | jrosser yeah, I haven't any more time to spend on it | 09:57 |
| jrosser | no me neither - i did prod mnaser about this a day or two ago | 09:58 |
| odyssey4me | I'll do a revert now. | 09:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client master: Revert "cleanup: stop managing files inside /etc" https://review.openstack.org/629122 | 09:59 |
| *** shyamb has quit IRC | 10:00 | |
| odyssey4me | well, we can make any roles use the integrated build for testing now - we just can't set them to do any different affinities (to test clustering) just yet | 10:01 |
| odyssey4me | I think perhaps I should get working on that, because that will open the door to switching most roles over to have deploy and upgrade tests. | 10:01 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Move role data in share/ansible/roles https://review.openstack.org/629127 | 10:03 |
| odyssey4me | jrosser oh, it looks like the issue could be solved quite simply - lemme push up an alternative patch | 10:04 |
| *** shyamb has joined #openstack-ansible | 10:05 | |
| evrardjp | odyssey4me: interesting , will wait for the patch. Have voted on the first one, should the new approach fail. | 10:06 |
| *** thuydang has quit IRC | 10:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Switch galera_install_dnf back to being a symlink https://review.openstack.org/629131 | 10:10 |
| odyssey4me | that fixes one issue I noticed... :p | 10:11 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Synced tempest plugins names and services https://review.openstack.org/628926 | 10:11 |
| *** mkuf has joined #openstack-ansible | 10:12 | |
| * jrosser rechecks a designate and barbican review now that tempest change is in ^ | 10:15 | |
| *** mkuf_ has quit IRC | 10:16 | |
| *** priteau has quit IRC | 10:17 | |
| *** mathlin has quit IRC | 10:22 | |
| Miouge | It would be cool to get some reviews on this Pike cherry pick: https://review.openstack.org/#/c/623233/ Rocky and Queens are already merged | 10:27 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Replace /etc/my.cnf.d if it exists for SUSE/CentOS https://review.openstack.org/629135 | 10:28 |
| odyssey4me | jrosser evrardjp ^ that should so the trick | 10:28 |
| *** mkuf has quit IRC | 10:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/rocky: Resolve btrfs backing store variable inconsistency https://review.openstack.org/629137 | 10:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/queens: Resolve btrfs backing store variable inconsistency https://review.openstack.org/629138 | 10:30 |
| openstackgerrit | Christian Zunker proposed openstack/openstack-ansible-os_horizon master: Switch to import_tasks when tags should be applied https://review.openstack.org/629139 | 10:30 |
| *** shyamb has quit IRC | 10:33 | |
| *** cshen has quit IRC | 10:34 | |
| *** kopecmartin is now known as kopecmartin|afk | 10:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Use the inventory to enable/disable services by default https://review.openstack.org/628979 | 10:35 |
| *** cshen has joined #openstack-ansible | 10:37 | |
| *** mkuf has joined #openstack-ansible | 10:39 | |
| odyssey4me | evrardjp could you also vote on this fix please https://review.openstack.org/#/c/629131/1 | 10:39 |
| *** cshen has quit IRC | 10:42 | |
| *** priteau has joined #openstack-ansible | 10:44 | |
| *** mathlin has joined #openstack-ansible | 10:46 | |
| *** fatdragon has joined #openstack-ansible | 10:52 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: cleanup: remove metadata checksum fixes tasks https://review.openstack.org/627792 | 10:53 |
| *** udesale has quit IRC | 10:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-hardening master: Switch to using import_tasks for static inclusion https://review.openstack.org/614329 | 10:59 |
| *** fatdragon has quit IRC | 11:01 | |
| admin0 | \o | 11:12 |
| *** cshen has joined #openstack-ansible | 11:12 | |
| *** dcdamien has joined #openstack-ansible | 11:13 | |
| admin0 | what does this mean ? TASK [os_heat : Create DB for service] = FAILED! => {"censored": "the output has been hidden due to the fact that 'no_log: true' was specified for this result", "changed": false} | 11:16 |
| *** thuydang has joined #openstack-ansible | 11:16 | |
| *** cshen has quit IRC | 11:17 | |
| odyssey4me | admin0 it means that something went wrong, but the log output has been surpressed because there is 'no_log: true' set on that task | 11:18 |
| *** mathlin has quit IRC | 11:19 | |
| *** thuydang has quit IRC | 11:22 | |
| *** cshen has joined #openstack-ansible | 11:42 | |
| *** mathlin has joined #openstack-ansible | 11:42 | |
| admin0 | for internal cloud, how do I populate this variable: horizon_keystone_multidomain_dropdown ? horizon_keystone_multidomain_dropdown: { 'Default', 'domain.com' } ?? | 11:44 |
| odyssey4me | admin0 that's a boolean - so it's true/false | 11:47 |
| odyssey4me | it populates from the keystone service catalog, as far as I know | 11:48 |
| *** shyamb has joined #openstack-ansible | 11:48 | |
| odyssey4me | if not, then it seems like https://docs.openstack.org/django_openstack_auth/latest/configuration/index.html#openstack-keystone-domain-choices might be needed - and there's no support for that right now | 11:50 |
| odyssey4me | perhaps noonedeadpunk can advise when he's back online | 11:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-hardening master: Switch to using import_tasks for static inclusion https://review.openstack.org/614329 | 11:54 |
| *** noonedeadpunk has joined #openstack-ansible | 11:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-hardening stable/rocky: cleanup: use updated conditionals https://review.openstack.org/629158 | 11:58 |
| dcdamien | odyssey4me: I hoped that dependent change would resolve issue with tempest on centos but it doesn't look like -> https://review.openstack.org/#/c/628785/3 | 11:59 |
| *** noonedeadpunk has quit IRC | 12:03 | |
| odyssey4me | dcdamien nope, I didn't think it would - because nothing is setting 'tempest_service_available_designate: true' so that it installs the package | 12:03 |
| odyssey4me | that's what https://review.openstack.org/#/c/628979/4/defaults/main.yml is for, but there's a bug there it seems | 12:03 |
| *** noonedeadpunk has joined #openstack-ansible | 12:04 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Use the inventory to enable/disable services by default https://review.openstack.org/628979 | 12:04 |
| *** thuydang has joined #openstack-ansible | 12:11 | |
| *** thuydang has quit IRC | 12:13 | |
| chandankumar | odyssey4me: Hello | 12:15 |
| chandankumar | odyssey4me: I am hitting this failure http://logs.openstack.org/82/627482/27/check/python-tempestconf-tempest-os-tempest/6fd8f64/job-output.txt.gz#_2019-01-08_12_00_54_483665 | 12:15 |
| chandankumar | odyssey4me: here is the playbook https://review.openstack.org/#/c/627482/ | 12:15 |
| *** mathlin has quit IRC | 12:17 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Update all SHAs for 16.0.25 https://review.openstack.org/627083 | 12:17 |
| *** cshen has quit IRC | 12:17 | |
| odyssey4me | chandankumar it looks like the path is incomplete | 12:19 |
| odyssey4me | no, that's not it - see http://logs.openstack.org/82/627482/27/check/python-tempestconf-tempest-os-tempest/6fd8f64/ara-report/result/72ec7db5-65c7-4aef-a36a-5ab23f135cbf/ | 12:21 |
| odyssey4me | chandankumar the issue is that the parent directory for that path doesn't exist | 12:21 |
| *** cshen has joined #openstack-ansible | 12:21 | |
| fnpanic | hi | 12:21 |
| chandankumar | odyssey4me: I need to replace that path with something else then | 12:22 |
| jamesdenton | @mbuil ping me when you're around, please | 12:22 |
| fnpanic | when i want to install elatiflow (https://github.com/robcowart/elastiflow) manual to a elastic cluster installed with the elk_metrics_6x playbooks from the ops repo, could this work? | 12:23 |
| odyssey4me | chandankumar well, I'm wondering whether it makes sense for the python-venv-build role to create the parent - or whether the calling role should create it | 12:23 |
| chandankumar | odyssey4me: I think python-venv-build role should create the parent directory | 12:23 |
| chandankumar | let me propose a patch there | 12:23 |
| odyssey4me | chandankumar yeah, seems sensible to me - thanks | 12:24 |
| *** cshen has quit IRC | 12:27 | |
| *** gkadam has quit IRC | 12:29 | |
| *** shyamb has quit IRC | 12:30 | |
| *** mathlin has joined #openstack-ansible | 12:39 | |
| openstackgerrit | Chandan Kumar proposed openstack/ansible-role-python_venv_build master: Create venv_install_destination_path parent directory https://review.openstack.org/629167 | 12:41 |
| chandankumar | odyssey4me: ^^ | 12:41 |
| *** shyamb has joined #openstack-ansible | 12:42 | |
| *** ansmith has joined #openstack-ansible | 12:42 | |
| mbuil | jamesdenton: back from lunch | 12:45 |
| *** ansmith has quit IRC | 12:48 | |
| *** hamzaachi__ has quit IRC | 12:50 | |
| *** vnogin has quit IRC | 12:54 | |
| openstackgerrit | Chandan Kumar proposed openstack/ansible-role-python_venv_build master: Create venv_install_destination_path parent directory https://review.openstack.org/629167 | 12:55 |
| *** dcdamien has quit IRC | 12:57 | |
| *** cshen has joined #openstack-ansible | 12:58 | |
| *** iw_zah has joined #openstack-ansible | 13:00 | |
| *** cshen has quit IRC | 13:02 | |
| *** szaher has joined #openstack-ansible | 13:08 | |
| arxcruz | odyssey4me: hey, happy new year :) | 13:11 |
| arxcruz | odyssey4me: I hope now https://review.openstack.org/#/c/621605/ is as you wanted :) | 13:11 |
| openstackgerrit | Merged openstack/openstack-ansible stable/queens: Update all SHAs for 17.1.7 https://review.openstack.org/627081 | 13:12 |
| *** mathlin has quit IRC | 13:13 | |
| chandankumar | arxcruz: mnaser https://review.openstack.org/#/c/629167/ | 13:19 |
| iw_zah | Is there a room for some basic problems deploying OS using QSA? I dont want to bother | 13:20 |
| *** udesale has joined #openstack-ansible | 13:20 | |
| odyssey4me | iw_zah what is QSA? | 13:20 |
| iw_zah | OSA, sorry | 13:21 |
| odyssey4me | iw_zah this is the right place | 13:21 |
| iw_zah | Deploying Openstack in a tagged infrastructure should be possible, isn't it? | 13:22 |
| odyssey4me | arxcruz I'm still confused - why the need to move them into vars files at all - I get the sense that you're not understanding how using overrides in ansible works? | 13:24 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_horizon master: Revert "Rename SUSE vpnaas-ui package" https://review.openstack.org/629177 | 13:24 |
| arxcruz | odyssey4me: I believe it's more easyt read than keep everything in the defaults/main.yml but please, enlight me about your idea here | 13:26 |
| arxcruz | odyssey4me: you want me to put int in var_files/all, or keep it in the defaults/main.yml ? | 13:27 |
| odyssey4me | arxcruz if tripleo wants to curate their own list, then that can be done - the defaults in the role can all be overrideen by any other variable precedence - by moving those vars into vars/main.yml, you're removing all the override capabilities making it much harder to override | 13:27 |
| odyssey4me | these should stay in defaults, because role defaults are the lowest precedence, giving *all* options to override | 13:28 |
| odyssey4me | https://docs.ansible.com/ansible/latest/user_guide/playbooks_variables.html#variable-precedence-where-should-i-put-a-variable | 13:28 |
| arxcruz | odyssey4me: so, can't we use the same list? | 13:28 |
| arxcruz | odyssey4me: my idea was add the installer type and we can share the list | 13:29 |
| odyssey4me | arxcruz yes, that's what I suggested right from the start | 13:29 |
| arxcruz | odyssey4me: okay, so i'll move it to defaults/main.yml | 13:29 |
| odyssey4me | well, as I mentioned before, why don't be just use the same list - and if you really want to override it then do that | 13:29 |
| arxcruz | odyssey4me: because on tripleo we have different list based on release | 13:30 |
| odyssey4me | there's no need to distinguish between what we want in the whitelist/blacklist and what tripleo does as far as I can see - if openstack is broken, it's broken for everyone | 13:30 |
| arxcruz | odyssey4me: not reallyin the way tripleo works | 13:30 |
| odyssey4me | based on release? you mean based on series? | 13:30 |
| odyssey4me | ie rocky vs stein? | 13:30 |
| arxcruz | odyssey4me: we test pike, queens, rocky, stein | 13:30 |
| arxcruz | not sure if we still test ocata | 13:31 |
| odyssey4me | ok, then that's a different issue which we need ot think about | 13:31 |
| arxcruz | odyssey4me: that's why i move to a file, we can just override the variable file name in vars/ | 13:31 |
| arxcruz | like blacklist_rocky.yml | 13:32 |
| arxcruz | then it will load rocky black list | 13:32 |
| arxcruz | blacklist_{{ release }}.yml | 13:32 |
| odyssey4me | ok, but you can still do that -but do it in the playbook that you use to run this | 13:32 |
| arxcruz | odyssey4me: https://github.com/openstack/tripleo-quickstart-extras/blob/master/roles/validate-tempest/tasks/pre-tempest.yml#L27 | 13:33 |
| odyssey4me | leave the defaults there with the master list, which targets master - then for stable branch tests override the default in the playbook | 13:33 |
| odyssey4me | is there a thing that calls this os_tempest role which is being worked on for tripleo? | 13:34 |
| arxcruz | odyssey4me: https://review.openstack.org/#/c/628415 | 13:35 |
| arxcruz | chandankumar: is working on that | 13:35 |
| *** plobo has joined #openstack-ansible | 13:38 | |
| odyssey4me | I've commented how you would import an override for the vars you want in there. | 13:38 |
| odyssey4me | The role doesn't have to be changed at all to accommodate what you want to do. | 13:38 |
| evrardjp | cores: I have discussed with Dirk (from rpm-packaging) about the issues we had this cycle with the SUSE packages. I think, from now on, if we see an issue in SUSE packages that break a build, we can (and should) ping in the #openstack-rpm-packaging to find out the root cause of the issues together. | 13:38 |
| odyssey4me | ok, thansk evrardjp | 13:39 |
| arxcruz | odyssey4me: ok, thanks! | 13:40 |
| *** plobo has quit IRC | 13:40 | |
| odyssey4me | arxcruz so the only thing to do is get the data structure the way you'd like it so that it has the annotations, which I love :) | 13:40 |
| arxcruz | odyssey4me: yup, too fast, i was typing about that :P | 13:40 |
| *** trident has quit IRC | 13:41 | |
| arxcruz | chandankumar: ^ | 13:41 |
| *** trident has joined #openstack-ansible | 13:42 | |
| *** jhesketh has quit IRC | 13:45 | |
| *** dcdamien has joined #openstack-ansible | 13:46 | |
| *** szaher has quit IRC | 13:47 | |
| *** jhesketh has joined #openstack-ansible | 13:47 | |
| cloudnull | mornings | 13:48 |
| *** ansmith has joined #openstack-ansible | 13:49 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Provide support for network interface mappings without override https://review.openstack.org/623620 | 13:51 |
| *** szaher has joined #openstack-ansible | 13:52 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat master: Only implement policy.json if an override is configured https://review.openstack.org/629191 | 13:55 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Switch galera_install_dnf back to being a symlink https://review.openstack.org/629131 | 13:57 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Replace /etc/my.cnf.d if it exists for SUSE/CentOS https://review.openstack.org/629135 | 13:57 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone master: Cleanup files and templates using smart sources https://review.openstack.org/588960 | 13:57 |
| odyssey4me | cores - it appears that policy-in-code was done for heat in queens, and we're still carrying our own policy file, which may cause problems and potential security issues - we should get https://review.openstack.org/#/c/629191/1 merged and ported back asap | 13:59 |
| cloudnull | done | 14:02 |
| *** trident has quit IRC | 14:03 | |
| cloudnull | fnpanic that should work, though I've never done it so hard to say 100% | 14:03 |
| *** mathlin has joined #openstack-ansible | 14:03 | |
| *** trident has joined #openstack-ansible | 14:03 | |
| *** kopecmartin|afk is now known as kopecmartin | 14:06 | |
| cloudnull | fnpanic looks like one the elk cluster is online you can start the installation here https://github.com/robcowart/elastiflow/blob/master/INSTALL.md#2-add-and-update-required-logstash-plugins and it should "Just Work-tm" | 14:06 |
| jrosser | mnaser: lxc_hosts seems to be wedged up on centos still - have you seen anything like this? http://logs.openstack.org/78/613078/19/check/openstack-ansible-lxc-btrfs-centos-7/7782654/job-output.txt.gz#_2019-01-07_21_49_54_811367 | 14:09 |
| mnaser | click | 14:09 |
| cloudnull | odyssey4me i'm getting a weird error with python_venv_build - https://pasted.tech/pastes/1004aa3c07952aafd3229ba0baad3ae76ad1c0e9.raw - encountered when re-running the cinder role local tests . | 14:10 |
| mnaser | hmm | 14:10 |
| jrosser | there was a btrfs fix went in earlier today | 14:10 |
| jrosser | but sadly thats just revealed that underneath | 14:10 |
| cloudnull | have you seen anything like that before ? | 14:10 |
| mnaser | is that the btrfs job | 14:10 |
| jrosser | it does it the same on btrsf and overlayfs | 14:10 |
| cloudnull | specifically - ` [Errno 2] No such file or directory: '/src/local-package-constraints-master.txt'` ? | 14:10 |
| mnaser | ok so this just blows up on btrfs and overlayfs only jrosser ? or all of the above | 14:11 |
| odyssey4me | cloudnull yes, I've seen that - it's because you don't have the right env vars set, because you're not using ./run_tests.sh - if you set the WORKING_DIR env var I think it'll solve it | 14:11 |
| cloudnull | ok. | 14:11 |
| jrosser | mnaser: like this https://review.openstack.org/#/c/613078/ | 14:11 |
| mnaser | lemme check sys logs | 14:12 |
| odyssey4me | cloudnull given that cinder is only using the standard functional test jobs - maybe we should switch it over to using the integrated repo for tests now - lemme push up a patch for it | 14:12 |
| mnaser | i dont use any overlay drivers tbh | 14:12 |
| cloudnull | odyssey4me ++ | 14:12 |
| jrosser | just wondered if there was any centos-ism rang a bell there - i was wondering if this all started around the 7.5->7.6 point | 14:13 |
| mnaser | jrosser: http://logs.openstack.org/78/613078/19/check/openstack-ansible-lxc-btrfs-centos-7/7782654/logs/host/lxc/lxc-container1.log.txt.gz | 14:13 |
| mnaser | Failed to rename network device "vethRWE53H" to "eth0": File exists | 14:13 |
| jrosser | it's upset about the networking, yes | 14:13 |
| mnaser | upset about networking when the storage driver changes | 14:14 |
| * mnaser flips table | 14:14 | |
| odyssey4me | mnaser are we happy to go ahead with https://review.openstack.org/#/c/623253/ ? that'll make the ceph_client role get tested by the integrated build | 14:14 |
| mnaser | ooooou | 14:14 |
| mnaser | yes i love that | 14:14 |
| mnaser | odyssey4me: i was also thinking that for the roles we do metal tests, might be quicker overall, keep the container ones for integrated repo | 14:15 |
| odyssey4me | it doesn't take away the existing role tests using the tests repo - I don't think we should remove those functional tests until we've got better penetration | 14:15 |
| mnaser | but that's an optimization we can do later | 14:15 |
| mnaser | yeah and because we just project templates, it's just 1 change so.. | 14:16 |
| * mnaser goes back to silly centos network issue | 14:16 | |
| odyssey4me | well, not exactly for all roles - because for each role that's not in the AIO we'll need to add an in-repo job which adds the extra service | 14:17 |
| mnaser | ah yes | 14:17 |
| odyssey4me | here's an example - in the cinder role, do we want to just test cinder & keystone... or do we want to test a full AIO? | 14:18 |
| odyssey4me | my suggestion would be just cinder with keystone - that'll be quick and simpler to spot issues for | 14:19 |
| mnaser | isnt an aio just keystone+nova+cinder anyways? | 14:19 |
| mnaser | well i worry that if we break something in the volume attach path | 14:19 |
| admin0 | TASK [os_heat : Add service/heat user] -- always fails .. is this a known issue .... 18.1.1 -> 18.1.2 | 14:19 |
| odyssey4me | aio = keystone, cinder, nova, neutron, horizon, galera, rabbitq | 14:19 |
| odyssey4me | oh, and glance: https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/vars/main.yml#L24 | 14:20 |
| admin0 | aio = swift also ? | 14:20 |
| admin0 | it has been useful | 14:20 |
| guilhermesp | mornings...would we have a meeting today? | 14:21 |
| odyssey4me | no, the AIO no longer has swift in master/rocky | 14:21 |
| odyssey4me | it's easily added, but not there by default | 14:22 |
| mnaser | guilhermesp: yes we will! | 14:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/rocky: Revert "Rename SUSE vpnaas-ui package" https://review.openstack.org/629194 | 14:24 |
| openstackgerrit | Arx Cruz proposed openstack/openstack-ansible-os_tempest master: Better tempest blacklist management https://review.openstack.org/621605 | 14:24 |
| arxcruz | I hope now it's okay | 14:24 |
| arxcruz | odyssey4me: ^ | 14:24 |
| arxcruz | probably won't be | 14:25 |
| admin0 | jamesdenton, reading your article on the mnaio, this weekend i am trying 2 nvme on raid0 to see if the time can be brought down even lower | 14:26 |
| jamesdenton | awesome! | 14:29 |
| *** shyamb has quit IRC | 14:29 | |
| jamesdenton | i have a samsung 970 evo showing up tomorrow and will see how much lower it'll go | 14:29 |
| jamesdenton | anything to speed up these tests | 14:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Test using integrated build https://review.openstack.org/629195 | 14:31 |
| chandankumar | cloudnull: mnaser small fix https://review.openstack.org/#/c/629167/ for venv_build | 14:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Test using integrated build https://review.openstack.org/629195 | 14:32 |
| admin0 | same .. 970 evo x2 goes into my crosshair hero VII .. should be able to get full speeds | 14:32 |
| odyssey4me | jamesdenton admin0 you know, if you have enough RAM on the host you can use a ramdisk instead :p | 14:33 |
| *** priteau has quit IRC | 14:33 | |
| admin0 | i only have 64 | 14:34 |
| jamesdenton | odyssey4me i remember you mentioning that. 76GB looks like it would handle the base VMs + host | 14:37 |
| *** mathlin has quit IRC | 14:37 | |
| jrosser | if you had some dedupe on the filesystem it may well be significantly less | 14:38 |
| jrosser | given the basic VM are pretty much all the same | 14:38 |
| *** shardy has quit IRC | 14:41 | |
| *** ygk_12345 has joined #openstack-ansible | 14:43 | |
| ygk_12345 | hi all I want to enable log in this context | 14:43 |
| ygk_12345 | how can I do that ? | 14:43 |
| ygk_12345 | when running os-heat-install playbook I am seeing this "the output has been hidden due to the fact that 'no_log: true" | 14:43 |
| jamesdenton | i retract my statement about 76GB. That was RAM not disk. my bad | 14:44 |
| ygk_12345 | FAILED - RETRYING: Add service/heat user (5 retries left). | 14:44 |
| ygk_12345 | i wan t to see the exact duplicate entry and delete it and try again | 14:44 |
| ygk_12345 | but I dont know what it is | 14:44 |
| ygk_12345 | so I want to see the log | 14:44 |
| ygk_12345 | how to enable it in ansible ? | 14:44 |
| *** hamzaachi has joined #openstack-ansible | 14:45 | |
| mnaser | odyssey4me: btw, what do you think of this? https://review.openstack.org/#/c/627784/ | 14:45 |
| ygk_12345 | this sis the bug I am facing | 14:45 |
| ygk_12345 | https://bugs.launchpad.net/openstack-ansible/+bug/1807346 | 14:45 |
| openstack | Launchpad bug 1807346 in openstack-ansible "[heat] Installations fails during Update Q->R" [Undecided,New] - Assigned to Guilherme Steinmuller Pimentel (guilhermesp) | 14:45 |
| *** cshen has joined #openstack-ansible | 14:48 | |
| mnaser | Moved network device "vethSFKG8R"/"(null)" to network namespace of 1973 | 14:48 |
| mnaser | i wonder why it says (null) | 14:48 |
| odyssey4me | ygk_12345 at this time the only way is to remark out the no_log line in the task file | 14:49 |
| mnaser | ok | 14:50 |
| mnaser | there is no 'tox_env' called 'btrfs' | 14:51 |
| *** shardy has joined #openstack-ansible | 14:51 | |
| mnaser | in lxc_hosts role | 14:51 |
| *** shardy has quit IRC | 14:51 | |
| mnaser | ..so what are those jobs doing | 14:51 |
| *** shardy has joined #openstack-ansible | 14:51 | |
| odyssey4me | mnaser it's running a test in lxc_containers_create | 14:51 |
| odyssey4me | it's a cross-repo test | 14:51 |
| mnaser | gah | 14:51 |
| mnaser | duh | 14:51 |
| mnaser | okay that makes sense | 14:51 |
| mnaser | i was losing it | 14:51 |
| mnaser | we really should have those jobs voting | 14:53 |
| mnaser | if we're going to have them cross-gate lxc_hosts | 14:53 |
| odyssey4me | mnaser sure, although it's not hard to end up with a circular failure there | 14:53 |
| mnaser | yeah | 14:54 |
| mnaser | equal part awful | 14:54 |
| *** cshen has quit IRC | 14:54 | |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build master: Create venv_install_destination_path parent directory https://review.openstack.org/629167 | 14:54 |
| mnaser | http://logs.openstack.org/82/628982/3/check/openstack-ansible-btrfs-centos-7/20eaf30/job-output.txt.gz#_2019-01-07_17_37_23_867545 | 14:55 |
| ygk_12345 | odyssey4me: how to do that ? | 14:55 |
| mnaser | shouldn't that be a hard fail? | 14:55 |
| ygk_12345 | odyssey4me: I mean which file exactly ? | 14:56 |
| odyssey4me | ygk_12345 grep for the task name in that role and you'll find it | 14:56 |
| ygk_12345 | odyssey4me: ok | 14:56 |
| cloudnull | general question, has anyone seen this before: https://pasted.tech/pastes/7c4f4463a2c0382b08fab19920a9b03ac772a2b8.raw - in cinder it seems to be throwing SQL errors, though in general the CLI is working (create|show|delete volumes is all working), I'm curious if this is an issue with my smart_sources patch for cinder or just a general issue with c | 14:57 |
| cloudnull | inder at the moment? | 14:57 |
| odyssey4me | mnaser according to the comments in code and previous reviews for those tasks, no - apparently centos doesn't always do the right thing with btrfs | 14:57 |
| *** fatdragon has joined #openstack-ansible | 14:58 | |
| fnpanic | cloudnull: Thanks! I will try it and report back! | 14:58 |
| cloudnull | ++ | 14:58 |
| odyssey4me | cloudnull looks like it passed some tests recently: https://review.openstack.org/628197 | 14:59 |
| odyssey4me | not sure if something broken since then upstream or downstream | 14:59 |
| ygk_12345 | odyssey4me: the os-heat playbook is failing here "failed: [c1b_heat_api_container-45a01df4 -> localhost] (item={u'project': u'admin', u'role': u'admin', u'user': u'stack_domain_admin'}) => {"attempts": 5, "changed": false, "item": {"project": "admin", "role": "admin", "user": "stack_domain_admin"}, "msg": "User stack_domain_admin is not valid"}" | 14:59 |
| ygk_12345 | odyssey4me: TASK [os_heat : Add service user to roles] | 15:00 |
| *** mathlin has joined #openstack-ansible | 15:00 | |
| odyssey4me | ygk_12345 and if you look at the tasks before that, should that user have been created - or if you look in keystone, can you see the user? | 15:01 |
| odyssey4me | can you see what might be wrong there? | 15:01 |
| chandankumar | odyssey4me: http://logs.openstack.org/82/627482/28/check/python-tempestconf-tempest-os-tempest/c6e0056/job-output.txt.gz#_2019-01-08_13_48_10_280399 | 15:01 |
| chandankumar | odyssey4me: this time permission error | 15:01 |
| cloudnull | odyssey4me interesting, looks like its doing the same thing there, but the tests are passing: http://logs.openstack.org/97/628197/1/check/openstack-ansible-functional-ubuntu-bionic/02b6f68/logs/openstack/cinder/cinder-scheduler.service.journal.log.txt.gz#_Jan_03_15_24_33 | 15:01 |
| mnaser | ok, creating a container with backing store btrfs seems *not* to create a btrfs subvolume at /var/lib/lxc/centos-7-amd64 | 15:01 |
| ygk_12345 | odyssey4me: I see that user. but I deleted them before running that playbook | 15:02 |
| chandankumar | odyssey4me: https://review.openstack.org/#/c/627482/28/playbooks/python-tempestconf-os-tempest.yaml does adding here become:true will help? | 15:02 |
| *** fatdragon has quit IRC | 15:03 | |
| odyssey4me | chandankumar well, it's a remote host and you're accessing it via the root user - so I'd guess not... unless that's a pseudonym for localhost, in which case yes - you need to become | 15:03 |
| odyssey4me | cloudnull how nice :/ | 15:03 |
| odyssey4me | ygk_12345 I don't see how deleting the user is helpful. Inspecting the user to see whether it's in the right domain, etc might be useful. | 15:04 |
| chandankumar | odyssey4me: thanks, updated :-) | 15:04 |
| odyssey4me | chandankumar you could also given it a path that the user has access to - the /openstack path is a little OSA specific, and really just a legacy thing | 15:05 |
| odyssey4me | cloudnull "Access denied for user 'cinder'@'cinder.openstack.local' (using password: YES)") (Background on this error at: http://sqlalche.me/e/e3q8)" | 15:05 |
| ygk_12345 | odyssey4me: the user 'stack_domain_admin' is in heat domain | 15:06 |
| odyssey4me | cloudnull it looks like those errors are until the migrations run, then it has access and starts | 15:06 |
| ygk_12345 | odyssey4me: this is how the error looks | 15:06 |
| ygk_12345 | odyssey4me: {"attempts": 5, "c hanged": false, "item": {"project": "admin", "role": "admin", "user": "stack_domain_admin"}, "msg": "User stack_domain_admin is not valid"} | 15:06 |
| openstackgerrit | Michael Vollman proposed openstack/openstack-ansible-os_manila master: Basic working os_manila role https://review.openstack.org/611930 | 15:06 |
| odyssey4me | ygk_12345 I'm not really in a position to trace it all in detail right now - if you inspect the tasks and the variables fed to it, you should be able to see whether each task is doing the right thing or whether there's an issue elsewhere. | 15:07 |
| odyssey4me | ygk_12345 right, so if the user exists and the role add task is failing - what does that tell you? | 15:07 |
| cloudnull | hum... ok ima try and reproduce the tempest results and see whats what. but first multiple hours of meetings :) | 15:07 |
| cloudnull | bbl | 15:08 |
| odyssey4me | ygk_12345 try doing it via the CLI to see if it works with the same parameters - it might be a bug in ansible | 15:08 |
| mnaser | hm | 15:12 |
| mnaser | looks like xenial fails too | 15:13 |
| mnaser | so i wonder if this feature is something that needs a newer kernel? | 15:13 |
| mnaser | or a newer release of lxc | 15:13 |
| mnaser | it looks like rhel dropped support for btrfs back in 7.4 | 15:16 |
| jrosser | i wonder what backends folks are actually using on centos | 15:18 |
| ygk_12345 | can someone help me with this error please | 15:21 |
| ygk_12345 | {"attempts": 5, "c hanged": false, "item": {"project": "admin", "role": "admin", "user": "stack_domain_admin"}, "msg": "User stack_domain_admin is not valid"} | 15:21 |
| ygk_12345 | this is heappening when running the heat playbook | 15:22 |
| *** markvoelker has joined #openstack-ansible | 15:23 | |
| mnaser | http://logs.openstack.org/78/613078/19/check/openstack-ansible-lxc-overlayfs-centos-7/dc8f90c/logs/host/lxc/lxc-container1.log.txt.gz | 15:25 |
| mnaser | what the | 15:25 |
| mnaser | jrosser: i wonder how many people actually even change the backend beyond 'dir' :\ | 15:25 |
| mnaser | it copied things into `/usr/lib64/lxc/rootfs/` for some reason | 15:26 |
| mnaser | Overlayfs: Mounted /var/lib/lxc/centos-7-amd64/rootfs onto /usr/lib64/lxc/rootfs with options upperdir=/var/lib/lxc/container1/delta0,lowerdir=/var/lib/lxc/centos-7-amd64/rootfs,workdir=/var/lib/lxc/container1/olwork. | 15:26 |
| *** markvoelker has quit IRC | 15:26 | |
| evrardjp | mnaser: if you have question about how people deploy with OSA, maybe we can now leverage the openstack survey. Though I wouldn't use it for low level questions ... But I think we are not using it enough. | 15:27 |
| *** markvoelker has joined #openstack-ansible | 15:27 | |
| *** markvoelker has quit IRC | 15:27 | |
| mnaser | in my personal experience, we've seen people do all sorts of things with OSA but.. rarely use a different backend | 15:27 |
| mnaser | though we have a few that experiment | 15:27 |
| evrardjp | My experience is about the same | 15:28 |
| *** dcdamien has quit IRC | 15:28 | |
| mnaser | taken up an hour and a half of my time trying to solve something that i won't ever use and i (have a feeling) not a lot of people use :( | 15:28 |
| *** cshen has joined #openstack-ansible | 15:30 | |
| mnaser | https://github.com/lxc/lxc/pull/2656 | 15:30 |
| mnaser | i think this is it | 15:30 |
| mnaser | https://github.com/lxc/lxc/issues/2655 | 15:30 |
| *** ygk_12345 has quit IRC | 15:30 | |
| evrardjp | mnaser: worth asking cloudnull :p He was the one toying with btrfs and lxc, IIRC, because it's close to nspawn + btrfs | 15:33 |
| mnaser | it looks like any cow stuff is not working on centos-7 | 15:33 |
| mnaser | probably because of an lxc-copy bug | 15:33 |
| mnaser | i dont know how much time cloudnull has to look into it but i'm all knowledged-out | 15:33 |
| *** mathlin has quit IRC | 15:34 | |
| *** cshen has quit IRC | 15:35 | |
| odyssey4me | If we think that it's better to reduce the support matrix, because supporting everything isn't all that feasible, then we should push up a patch to hard stop if any bad combinations are used, and we should remove the tests | 15:35 |
| cloudnull | is it just btrfs on centos? if so maybe we just take that one out ? | 15:36 |
| * cloudnull haven't been following a long | 15:36 | |
| odyssey4me | it's a little odd that btrfs isn't working, but machinectl is? | 15:36 |
| mnaser | cloudnull: its btrfs and overlayfs | 15:37 |
| mnaser | at least overlayfs i've identified one weird thing | 15:37 |
| cloudnull | machinectl doesn't use the lxc mechanisms to create a copy | 15:37 |
| mnaser | "/var/lib/lxc/centos-7-amd64/rootfs onto /usr/lib64/lxc/rootfs" | 15:37 |
| mnaser | which is that it tries to copy to some weird path | 15:37 |
| odyssey4me | cloudnull ah, ok | 15:37 |
| cloudnull | we've had lots of random issues with overlayfs over the years. | 15:37 |
| cloudnull | could be some new issues is cropping up ? | 15:37 |
| odyssey4me | perhaps overlayfs should just be ditched as a back-end | 15:38 |
| cloudnull | maybe its best to make those non-voting/remove them on cent? | 15:38 |
| jrosser | its wierd | 15:38 |
| odyssey4me | then btrfs just made unsupported on centos | 15:38 |
| jrosser | overlayfs: lxc_network - network.c:lxc_network_move_created_netdev_priv:2417 - Moved network device "veth6GW2PC"/"(null)" to network namespace of 21788 | 15:38 |
| jrosser | machinectl: lxc_network - network.c:lxc_network_move_created_netdev_priv:2417 - Moved network device "veth3M0LWR"/"eth1" to network namespace of 353 | 15:39 |
| cloudnull | ^ odyssey4me we've seen that one before, no ? | 15:39 |
| cloudnull | overlayfs and vethpair issues? | 15:39 |
| odyssey4me | it rings a bell, but for the life of me I can't recall from what | 15:39 |
| jrosser | btrfs: lxc_network - network.c:lxc_network_move_created_netdev_priv:2417 - Moved network device "vethW0I3U4"/"(null)" to network namespace of 22106 | 15:39 |
| jrosser | same ^ | 15:39 |
| odyssey4me | was this perhaps something we saw in nspawn with the namespace pid conflict? | 15:40 |
| odyssey4me | something something old systemd nonsense | 15:40 |
| jrosser | :) | 15:40 |
| *** cshen has joined #openstack-ansible | 15:40 | |
| odyssey4me | if it is the case, then it will be inconsistent and it will happen for any backend | 15:41 |
| cloudnull | Oh that could be ^ | 15:42 |
| odyssey4me | we see that on xenial, but on bionic it's fine | 15:43 |
| mnaser | honestly i'd really love to try and fix it and keep feature parity | 15:43 |
| cloudnull | if its related to the issues we saw in glibc then that's a bummer. | 15:43 |
| mnaser | but given that btrfs is removed from rhel7.4 and dropped in rhel8 | 15:43 |
| chandankumar | odyssey4me: cloudnull https://review.openstack.org/629127 please have a look on this when free thanks :-) | 15:43 |
| mnaser | that one can safely be removed | 15:43 |
| jrosser | the top 3 here are all the same i think https://review.openstack.org/#/q/project:openstack/openstack-ansible-lxc_hosts+status:open | 15:43 |
| mnaser | i think overlayfs is supposed to stick around | 15:43 |
| cloudnull | mnaser ++ | 15:43 |
| mnaser | but at least the issue in overlayfs is obvious | 15:44 |
| mnaser | it's copying to the wrong folder | 15:44 |
| odyssey4me | I'm stuck in meetings for a while - bbl. | 15:44 |
| cloudnull | RHEL is removing BTRFS entirely in RHEL8. while we still use cent7 maybe its time to sunset that backend on cent | 15:44 |
| mnaser | cloudnull: and it was dropped support in 7.4 too | 15:45 |
| odyssey4me | where does the btrfs removal put us for machinectl/nspawn? | 15:45 |
| cloudnull | in an odd place. | 15:45 |
| mnaser | "Btrfs is available as a Technology Preview feature in Red Hat Enterprise Linux 7 but has been deprecated since the Red Hat Enterprise Linux 7.4 release. It will be removed in a future major release of Red Hat Enterprise Linux." | 15:45 |
| *** cshen has quit IRC | 15:45 | |
| mnaser | honestly i've been slowly more and more inching towards metal deployments | 15:45 |
| openstackgerrit | Arx Cruz proposed openstack/openstack-ansible-os_tempest master: Better tempest blacklist management https://review.openstack.org/621605 | 15:46 |
| mnaser | and unless i can start getting a bit more help in centos, that might be the only scenario i can help keep supported :\ | 15:46 |
| evrardjp | mnaser: :) | 15:46 |
| evrardjp | I might now have another padawan! :P | 15:46 |
| evrardjp | maybe will have* | 15:46 |
| mnaser | we discussed at the ptg making containers opt-in | 15:47 |
| evrardjp | simplify things, let people come with their own method of machine containers/virtual machines | 15:47 |
| openstackgerrit | Arx Cruz proposed openstack/openstack-ansible-os_tempest master: Better tempest blacklist management https://review.openstack.org/621605 | 15:47 |
| mnaser | and having metal be default | 15:47 |
| cloudnull | odyssey4me - with all of the things rhel is removing for rhel8 we're honestly going to have our hands full with it as a supported OS. | 15:48 |
| mnaser | cloudnull: around containers i assume/ | 15:48 |
| mnaser | but for the metal deployments, we're probably okay? | 15:48 |
| cloudnull | networking, supportined backens, etc. | 15:48 |
| cloudnull | file systems... | 15:48 |
| evrardjp | cloudnull: systemd-network won't be supported? | 15:49 |
| odyssey4me | cloudnull mnaser then perhaps we should just bite the bullet this cycle and have centos remove all container techs and be metal only | 15:49 |
| cloudnull | I hope everyone like network manager :) | 15:49 |
| cloudnull | evrardjp presently, nope | 15:49 |
| odyssey4me | oh holy hell | 15:49 |
| cloudnull | only network manager, NMCLI for the win... | 15:49 |
| mnaser | we don't touch networking in OSA unless we're doing containers (and then we touch the networking inside contaienrs afaik?) | 15:49 |
| mnaser | filesystems is only an issue for containers too | 15:49 |
| evrardjp | well if we move to consider metal first for centos, it simplify things, as we say that people have to configure their own networking first. | 15:49 |
| evrardjp | mnaser: yup exactly | 15:50 |
| odyssey4me | fair enough | 15:50 |
| odyssey4me | all storage and networking is in the deployer's hands | 15:50 |
| mnaser | i guess the only place we really touch networking is in os_neutron but thats only ovs and/or linuxbridge | 15:50 |
| odyssey4me | we just put openstack down | 15:50 |
| mnaser | and i dont think they're removing that bahaha | 15:50 |
| evrardjp | mnaser: that's already conditional,so we can have different "backends" for wiring up the networking | 15:50 |
| evrardjp | it only requires good ansible modules for it | 15:51 |
| cloudnull | its still early but it seems rhel is wanting to support geneveve as their supported network platform when it comes to openstack. -cc jamesdenton | 15:51 |
| evrardjp | mnaser: removing linux bridge would be a gutsy approach :) | 15:51 |
| cloudnull | so i suspect there will be issues there too | 15:51 |
| jamesdenton | geneve w/ OVN | 15:51 |
| cloudnull | ^ | 15:51 |
| chandankumar | arxcruz: https://review.openstack.org/#/c/621605/ since all projects are moving to storyboard can we add entry about storyboard also? | 15:51 |
| *** strobelight has joined #openstack-ansible | 15:52 | |
| mnaser | i mean | 15:52 |
| mnaser | geneve probably needs ovs to operate | 15:52 |
| mnaser | and i doubt they'd rip out linuxbridge | 15:52 |
| mnaser | so i think we're still...ok | 15:52 |
| jamesdenton | yeah, OVN is based on OVS, GENEVE just replaces VXLAN | 15:53 |
| mnaser | odyssey4me: i know you mentioned earlier you're in meetings so feel free to answer this later | 15:54 |
| cloudnull | as RHEL 8 gets out of beta it'll be interesting to see what we have to do to keep is going. | 15:54 |
| mnaser | but i think we can combine efforts to bring integrated jobs down to roles and move to metal at the same time for centos | 15:54 |
| cloudnull | ok, back to being gone. see you all soon. | 15:54 |
| mnaser | that way we don't have to rewrite all of the functional jobs that leveraged containers | 15:54 |
| *** shardy has quit IRC | 15:54 | |
| cloudnull | https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8-beta/html/8.0_beta_release_notes/removed_functionality -- for your reading enjoyment :) | 15:56 |
| *** mathlin has joined #openstack-ansible | 15:57 | |
| mnaser | thats a fun read | 15:59 |
| mnaser | cloudnull, DimGR, d34dh0r53, hughsaunders, b3rnard0, palendae, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, scarlisle, luckyinva, ntt, javeriak, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, bgmccollum, darrenc, JRobinson__, colinmcnamara, thorst, adreznec, eil397, | 15:59 |
| mnaser | qwang,nishpatwa_, cathrichardson, drifterza, hwoarang, cshen, ullbeking, mnaser, nicolasbock, jrosser, cjloader, antonym, dcdamien | 15:59 |
| mnaser | 1 minute warning :) | 15:59 |
| spotz | o/ | 15:59 |
| mnaser | #startmeeting openstack_ansible_meeting | 16:00 |
| openstack | Meeting started Tue Jan 8 16:00:01 2019 UTC and is due to finish in 60 minutes. The chair is mnaser. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| *** openstack changes topic to " (Meeting topic: openstack_ansible_meeting)" | 16:00 | |
| *** hamerins has joined #openstack-ansible | 16:00 | |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:00 |
| mnaser | #topic rollcall | 16:00 |
| mnaser | o/ | 16:00 |
| *** openstack changes topic to "rollcall (Meeting topic: openstack_ansible_meeting)" | 16:00 | |
| jamesdenton | o/ | 16:00 |
| odyssey4me | o/ | 16:00 |
| evrardjp | o/ | 16:00 |
| spotz | o/ | 16:00 |
| mattt | o/ | 16:00 |
| chandankumar | \o/ | 16:01 |
| mnaser | anyone has any last week highlights to bring up? i know we've missed 2 weeks because of holidays :) | 16:01 |
| cloudnull | +/- /o | 16:01 |
| odyssey4me | holidays were a highlight :) | 16:01 |
| guilhermesp | o/ | 16:01 |
| *** udesale has quit IRC | 16:02 | |
| mnaser | that's the best highlgiht | 16:03 |
| mnaser | #topic Bug triage | 16:03 |
| *** openstack changes topic to "Bug triage (Meeting topic: openstack_ansible_meeting)" | 16:03 | |
| mnaser | catch up time | 16:03 |
| mnaser | #link https://bugs.launchpad.net/openstack-ansible/+bug/1810584 | 16:03 |
| openstack | Launchpad bug 1810584 in openstack-ansible "openstack-ansible setup-hosts.yml fails in task: lxc_hosts Ensure image has been pre-staged" [Undecided,New] | 16:03 |
| mnaser | i can curl that file | 16:04 |
| evrardjp | 3 ppl saying it's a problem , I suppose it's real | 16:04 |
| mnaser | "Running the same command a second time, the command succeeds." | 16:04 |
| evrardjp | weirdly it works in gates in one go | 16:04 |
| mnaser | in my experience, cdimage.ubuntu.com isn't the most reliable | 16:04 |
| mnaser | yeah i've never seen gate failures around that | 16:05 |
| evrardjp | maybe we are using an infra mirror? | 16:05 |
| mnaser | 2 people on the same time too | 16:05 |
| mnaser | i don't think we are for this | 16:05 |
| jrosser | these are multinode though arent they? | 16:05 |
| evrardjp | jrosser: they seem to | 16:05 |
| evrardjp | but IIRC odyssey4me told me RAX has multinode daily periodics now, which would have shown up the issue | 16:06 |
| mnaser | _lxc_hosts_container_image_url: "http://cdimage.ubuntu.com/ubuntu-base/releases/18.04/release/ubuntu-base-18.04.1-base-{{ lxc_cache_map.arch }}.tar.gz" | 16:06 |
| odyssey4me | yep, we don't see failures there for MNAIO tests as far as I know | 16:06 |
| evrardjp | we maybe need to document this as a known issue in case of people not overriding the value | 16:07 |
| mnaser | lxc_hosts_container_image_url: "{{ _lxc_hosts_container_image_url }}" | 16:07 |
| evrardjp | odyssey4me: I suppose you're using RAX mirrors for that | 16:07 |
| odyssey4me | or perhaps we should increase the timeout | 16:07 |
| mnaser | odyssey4me: the thing is if you look at the logs | 16:07 |
| mnaser | 0B downloaded | 16:07 |
| odyssey4me | nope, we're using the upstream sources every time | 16:07 |
| mnaser | so.. it's just not downloading ever | 16:07 |
| odyssey4me | well, that's nice :/ | 16:07 |
| evrardjp | oh god | 16:07 |
| mnaser | 64MiB/s later on the same download on the rerun | 16:07 |
| evrardjp | look at the nice can of worms? | 16:07 |
| mnaser | the 2nd person reported on the same day | 16:08 |
| *** szaher has quit IRC | 16:08 | |
| odyssey4me | well, we can remove the async and add retries I guess - or figure out another way of having async + retries | 16:08 |
| odyssey4me | or we can ditch containers :p | 16:08 |
| mnaser | :D | 16:08 |
| mnaser | for the sakes of this bug, i hate to say it but i guess its confirmed because we don't have a retry mechanism | 16:09 |
| mnaser | even though it's not really our fault, but we don't have recovery from this simple failure | 16:09 |
| *** szaher has joined #openstack-ansible | 16:09 | |
| mnaser | confirmed/low ? | 16:10 |
| mnaser | because it's really handling a third party failure | 16:10 |
| mnaser | i guess ill go for that. | 16:10 |
| evrardjp | yeah I'm fine with that triage | 16:10 |
| mnaser | #link https://bugs.launchpad.net/openstack-ansible/+bug/1810538 | 16:11 |
| openstack | Launchpad bug 1810538 in openstack-ansible "keepalived.service is not enabled" [Undecided,New] | 16:11 |
| evrardjp | but we need to explain it's not really our fault in the bug :p | 16:11 |
| mnaser | uh | 16:11 |
| mnaser | i'll add a note | 16:11 |
| evrardjp | that's weird. It should not pass my functional testing if it's not started | 16:12 |
| prometheanfire | o/ | 16:12 |
| evrardjp | oh wait | 16:12 |
| mnaser | evrardjp: enabled not started | 16:12 |
| evrardjp | after reboot | 16:12 |
| evrardjp | yeah I misread | 16:12 |
| evrardjp | I don't test that | 16:12 |
| evrardjp | but it's not really my fault if the module is not doing appropriately | 16:12 |
| evrardjp | :p | 16:12 |
| mnaser | https://github.com/evrardjp/ansible-keepalived/blob/master/tasks/main.yml#L197-L200 | 16:13 |
| evrardjp | curious | 16:14 |
| mnaser | i somehow doubt | 16:14 |
| mnaser | that `enabled: yes` is broken | 16:14 |
| mnaser | and i have a feeling something else is breaking it | 16:14 |
| *** iw_zah has quit IRC | 16:14 | |
| mnaser | anyways, i cant reproduce because i've always seen keepalive go back up | 16:14 |
| evrardjp | I can't remember if keepalived can be socket activated | 16:14 |
| mnaser | and i can see it up here `systemctl is-enabled keepalived.service` on a c7 deploy | 16:15 |
| evrardjp | in that case a wrong net config would never activate the socket | 16:15 |
| evrardjp | but i doubt it's that | 16:15 |
| mnaser | it would still be enabled but stopped | 16:15 |
| evrardjp | we need a systemd unit log | 16:15 |
| evrardjp | I think there is something else going on | 16:16 |
| mnaser | asked for more info | 16:16 |
| evrardjp | the enablement of the service resulting in a failure should be in the journal | 16:16 |
| evrardjp | thx | 16:16 |
| mnaser | incomplete | 16:16 |
| mnaser | medium | 16:16 |
| mnaser | for now.. | 16:16 |
| mnaser | #link https://bugs.launchpad.net/openstack-ansible/+bug/1810537 | 16:17 |
| openstack | Launchpad bug 1810537 in openstack-ansible "volume creation fails after successful installation" [Undecided,New] | 16:17 |
| mnaser | hey isn't that what you were working on cloudnull ? | 16:17 |
| mnaser | the fix is icnldued | 16:17 |
| mnaser | cc odyssey4me ^ | 16:17 |
| mnaser | https://review.openstack.org/#/c/628197/ | 16:17 |
| cloudnull | mnaser that maybe the issue I was seeing. | 16:18 |
| mnaser | cloudnull: can i assign to you? maybe https://review.openstack.org/#/c/628197/ is the fix | 16:18 |
| mnaser | it certainly has to do with it | 16:18 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_cinder master: Adds resource_filters.json distribution https://review.openstack.org/628197 | 16:19 |
| mnaser | added closes-bug | 16:19 |
| cloudnull | sure | 16:19 |
| mnaser | evrardjp: i just did it, lets +2? | 16:19 |
| evrardjp | haha | 16:19 |
| cloudnull | this was the error from the cinder-api log I was seeing | 16:19 |
| cloudnull | https://pasted.tech/pastes/61c4496978d40841ddaf22d0e3ca49936f269a3a.raw | 16:19 |
| mnaser | considering noonedeadpunk had it fixed for a while now and we haven't done much :) | 16:19 |
| cloudnull | when running tempest.api.volume.test_volumes_list.VolumesListTestJSON | 16:19 |
| mnaser | cloudnull: while its not the same traceback, it complains about `common.get_enabled_resource_filters` | 16:19 |
| cloudnull | ++ | 16:20 |
| mnaser | which seems pretty darn close so.. | 16:20 |
| cloudnull | so it totally could be the same issue, will tinker in a bit | 16:20 |
| mnaser | keep https://review.openstack.org/#/c/628197/2 in mind and ill go onto the next | 16:20 |
| mnaser | #link https://bugs.launchpad.net/openstack-ansible/+bug/1810533 | 16:20 |
| *** pcaruana has quit IRC | 16:20 | |
| openstack | Launchpad bug 1810533 in openstack-ansible "openstsack-ansible behind a proxy fails when calling apt-key" [Undecided,New] | 16:20 |
| noonedeadpunk | Oh, never knew, that there's a bug for this:) | 16:20 |
| mnaser | noonedeadpunk: no problem =) | 16:20 |
| mnaser | oh look a proxy issue | 16:20 |
| * mnaser looks at jrosser | 16:20 | |
| evrardjp | :) | 16:21 |
| jrosser | didnt we change these to be in the repo now? | 16:21 |
| mnaser | im going to guess this might be happening in cache prep stage | 16:21 |
| mnaser | unfortunately it doesn't mention which role | 16:21 |
| odyssey4me | MOAR WORKFLOW | 16:21 |
| evrardjp | jrosser: that's true we are carrying things | 16:21 |
| evrardjp | odyssey4me: haha I laughed too | 16:22 |
| mnaser | fine ill join in too | 16:22 |
| evrardjp | BUFFEROVERWORKFLOW | 16:22 |
| mnaser | i dont see any 'apt-key' references using codesearch.openstack.org | 16:22 |
| mnaser | i guess i can ask where that change was done? | 16:22 |
| odyssey4me | oh yeah, I'll take that - I've done all the patches except the rocky patch for rabbitmq_server | 16:22 |
| odyssey4me | master is all don | 16:22 |
| odyssey4me | *done | 16:22 |
| evrardjp | odyssey4me: thanks | 16:23 |
| spotz | hehe | 16:23 |
| mnaser | okay cool, ill assign then odyssey4me :) | 16:23 |
| mnaser | #link https://bugs.launchpad.net/openstack-ansible/+bug/1810319 | 16:23 |
| openstack | Launchpad bug 1810319 in openstack-ansible "Can't set gateway for provider network" [Undecided,New] | 16:23 |
| odyssey4me | https://review.openstack.org/#/q/(topic:vendor-gpg-keys+OR+topic:vendor-gpg-keys-stable/rocky)+(status:open+OR+status:merged) | 16:23 |
| mnaser | looks like that's dcdamien | 16:24 |
| jrosser | i wasnt sure about this one - "specific default gateway" sounds a bit bogus? | 16:24 |
| jamesdenton | i seem to recall looking at the playbooks could not find any reference to 'gateway' other than docs | 16:24 |
| mnaser | sounds like the user wants a static route | 16:24 |
| jrosser | ^ that, yes | 16:24 |
| mnaser | was it lxc_container_create that made network configs? | 16:25 |
| noonedeadpunk | I don't think, that it ever created default routes. But several times I've faced with issue of missing nat rules on controller nodes after setup-hosts.yml | 16:26 |
| noonedeadpunk | but it's probably not related | 16:26 |
| mnaser | i dunno | 16:27 |
| evrardjp | long ago we could inject routes to the process. I remember, I wrote it. | 16:27 |
| mnaser | i mean we have it listed there | 16:27 |
| jrosser | theres an example in the tests https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/tests/group_vars/all_containers.yml#L17-L26 | 16:27 |
| mnaser | for static routes yeah | 16:28 |
| mnaser | not default | 16:28 |
| jamesdenton | static could easily be 0.0.0.0, no? | 16:28 |
| odyssey4me | missing rules on the host is fixed by lxc-system-manage | 16:28 |
| mnaser | well it would conflict with the lxc one | 16:28 |
| jamesdenton | ahh | 16:28 |
| mnaser | the 10.8.0.0 or whatever | 16:28 |
| evrardjp | jamesdenton: having two 0.0.0.0/0 seems to be problematic | 16:28 |
| evrardjp | for some ppl :p | 16:28 |
| evrardjp | not sure what we are talking about anymore | 16:29 |
| odyssey4me | unicorns! | 16:29 |
| mnaser | i added a comment | 16:29 |
| mnaser | and i'll set to invalid i guess | 16:29 |
| mnaser | linking to jrosser example | 16:29 |
| evrardjp | we are talking about two different things | 16:29 |
| evrardjp | IMO | 16:29 |
| mnaser | user wants to define a gateway => not possible | 16:30 |
| mnaser | we're pretty sure user just wants a static route pointing ?? somewhere ?? | 16:30 |
| mnaser | the default route is always 10.8.0.1 or whatever the lxc host ends up with | 16:30 |
| mnaser | we use that for natting and all the other fun stuff we need to do to make sure things work | 16:30 |
| mnaser | also | 16:30 |
| mnaser | don't we run haproxy inside metal? | 16:31 |
| mnaser | i'm pretty sure we do.. right? keepalived and haproxy | 16:31 |
| *** mathlin has quit IRC | 16:31 | |
| noonedeadpunk | Yep | 16:31 |
| evrardjp | https://github.com/openstack/openstack-ansible-lxc_hosts/blob/a8b96e2e37ffea4b7c3e055b1310b10bb95a7b2a/defaults/main.yml#L106 | 16:31 |
| evrardjp | let me backtrack this into the inventory | 16:31 |
| odyssey4me | haproxy/keepalived are installed on the host for an AIO, yes | 16:32 |
| mnaser | so i guess this seems to be a specific scenario that the user came up with | 16:32 |
| evrardjp | it seems it's not in the inventory anymore. | 16:32 |
| mnaser | running haproxy in containers but then wanting the container to not be wired to the internet lxc network but wired to the public network in this case | 16:33 |
| jrosser | i think this is running containerised haproxy, needing eth0 to be natted to install (guess) and the external interface on this new eth14 | 16:33 |
| evrardjp | one could run haproxy and keepalived in containers, and use whatever network interface he wants | 16:33 |
| evrardjp | I think it's a valid issue | 16:34 |
| mnaser | not really, because our architecture has a default route of the host | 16:35 |
| mnaser | which means that traffic goes in eth14 | 16:35 |
| evrardjp | we probably removed that feature at some point, kept the feature in lxc_hosts, and probably forgot to edit the default template of the inventory. | 16:35 |
| mnaser | but on the way out, it hits the default route | 16:35 |
| jrosser | imho this is addressed with a static route as i linked | 16:35 |
| mnaser | i'l leave this and lets wait for the user to comment on next week | 16:36 |
| evrardjp | mnaser: not sure to understand | 16:36 |
| mnaser | evrardjp: just because traffic enters from one interface, does not mean it will exit from the same one | 16:36 |
| evrardjp | IMO you could have your own NIC that's not natted in the container, and that would require a default route for reason x | 16:36 |
| mnaser | if your default route is the physical host that runs the container | 16:36 |
| evrardjp | well no I meant, if you don't run nat at all | 16:37 |
| evrardjp | you could have just bridges on the host, and ignore lxc nat. | 16:37 |
| mnaser | you could, but that's a whole another use case | 16:37 |
| mnaser | anyways, i think we've taken abit of time on this | 16:37 |
| mnaser | i dont wanna burn everyone out with all this stuff for now | 16:37 |
| jrosser | evrardjp: yes this is exactly what i do in the new http proxy test, no default route https://review.openstack.org/#/c/625523/ | 16:38 |
| evrardjp | haha true :) | 16:38 |
| mnaser | i feel like bug triage drains everyone out and we lose people :( | 16:38 |
| mnaser | so enough of that for today | 16:38 |
| *** gyee has joined #openstack-ansible | 16:38 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cinder master: Cleanup files and templates using smart sources https://review.openstack.org/588953 | 16:38 |
| mnaser | rhel 8 was an interesting subject cloudnull brought up and odyssey4me evrardjp was discussing earlier | 16:38 |
| mnaser | a lot of stuff has been removed which makes containers even harder | 16:38 |
| * mnaser is using more and more non-containerized centos deploys | 16:39 | |
| evrardjp | mnaser: I checked the link you gave above | 16:39 |
| guilhermesp | with this, cinder as backend for glance is going to be supported by us but just when glance is METAL | 16:39 |
| evrardjp | (might worth giving it here) | 16:39 |
| mnaser | oh yeah cloudnull posted that | 16:39 |
| mnaser | cloudnull> https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8-beta/html/8.0_beta_release_notes/removed_functionality -- for your reading enjoyment :) | 16:39 |
| evrardjp | mnaser: I am scared of the LVM Python bindings have been removed | 16:39 |
| mnaser | yeah guilhermesp also caught a really annoying issue about iscsiadm and containers | 16:40 |
| mnaser | things like running glance in containers with cinder backend doesn't work for example | 16:40 |
| odyssey4me | yep | 16:40 |
| odyssey4me | nothing using iscsi will work in a container | 16:40 |
| mnaser | learned that the hard way :) | 16:40 |
| evrardjp | :) | 16:40 |
| guilhermesp | yep hahaha | 16:40 |
| prometheanfire | that's been known for a while iirc (we learned the hard way too) | 16:40 |
| guilhermesp | I suffered a bit but that's ok | 16:40 |
| *** kopecmartin is now known as kopecmartin|off | 16:40 | |
| mnaser | so given all that stuff, i'd like to propose moving centos off containers into metal | 16:41 |
| mnaser | as a path to eventually maybe make containers an opt-in | 16:41 |
| mnaser | but making centos more of a canary to see what happens in there | 16:41 |
| prometheanfire | a second cent gate for it? | 16:41 |
| evrardjp | mnaser: that sounds fair. I like the approach. | 16:41 |
| mnaser | prometheanfire: we already gate metal and lxc | 16:41 |
| mnaser | and they're both somewhat reliable, i guess. | 16:42 |
| evrardjp | mnaser: do we have fine tuned the default listen interfaces of all the services? | 16:42 |
| prometheanfire | just curious what it'd do to coverage is all | 16:42 |
| mnaser | evrardjp: i'll have to pick up that work and maybe JUST maybe multinode metal jobs | 16:42 |
| evrardjp | if there is no 0.0.0.0 anymore, we can pretty much simplify more things | 16:42 |
| jrosser | evrardjp: you need this https://github.com/openstack/openstack-ansible/blob/master/playbooks/listening-port-report.yml | 16:43 |
| mnaser | so i was thinking we can combine some of the efforts that odyssey4me has been doing in order to run integrated for roles | 16:43 |
| mnaser | so that we don't have to rewrite most functional tests to throw them away later | 16:43 |
| evrardjp | mnaser: for now I'd say the only part is haproxy because he listens to the same port as some services, so if you put multinode, just put haproxy on a single node, then all the rest into an infra node, then a compute node | 16:43 |
| evrardjp | a 3 node perfection :) | 16:43 |
| mnaser | evrardjp: yea but ideally i'd like haproxy to be colocatable | 16:43 |
| mnaser | most deployments don't have a dedicated haproxy node | 16:43 |
| mnaser | in my experience at least | 16:44 |
| mnaser | people want to reuse their controllers | 16:44 |
| evrardjp | that's sad, because it helps :p | 16:44 |
| mnaser | running everything on it's own machine is nice too because that helps, but then we can do vms to make it easier cause they don't need resources .. but machine containers work easier too cause they're lighter | 16:44 |
| mnaser | oh. | 16:44 |
| mnaser | wait. | 16:44 |
| mnaser | :) | 16:44 |
| evrardjp | :) | 16:45 |
| odyssey4me | maybe we should just use kata containers instead :p | 16:45 |
| mnaser | hey | 16:45 |
| evrardjp | k8s? | 16:45 |
| mnaser | i'm not gonna lie | 16:45 |
| mnaser | i've thought about using docker containers to run some stuff | 16:45 |
| evrardjp | mnaser: it makes sense | 16:46 |
| mnaser | like imagine how nice it'd be just to pull down memcache, the same way, across all systems | 16:46 |
| evrardjp | but then ppl will say it's memcache with centos, oh no it's memcache on suse I want, oh no it's memcache on ubuntu I want, OH NO it's not Alpine! | 16:46 |
| jrosser | we should learn carefully from the tripleo experience there - seems to have turned messy | 16:47 |
| evrardjp | I think run everything on metal is simple | 16:47 |
| evrardjp | if ppl want their own things, they can | 16:47 |
| prometheanfire | not going to make all people happy all the time | 16:48 |
| evrardjp | for example, they set their ubuntu nodes with lxd networks, and then install on lxd nodes | 16:48 |
| evrardjp | prometheanfire: that's true | 16:48 |
| evrardjp | prometheanfire: but we don't have to | 16:48 |
| *** fatdragon has joined #openstack-ansible | 16:48 | |
| mnaser | i've used every single deployment tool so far by now lol | 16:48 |
| prometheanfire | evrardjp: just making sure we were not trying to, that path leads to destruction :P | 16:48 |
| evrardjp | if we deliver the minimum amount of code it's easier to maintain in the long run, and people will be happy, even if we lack features | 16:48 |
| noonedeadpunk | but I'd prefer not to completel reject containers, in favor of metal | 16:49 |
| mnaser | noonedeadpunk: nope, we don't want to do that at all | 16:49 |
| evrardjp | mnaser: fuel too? | 16:49 |
| mnaser | evrardjp: fuel, mos, tripleo, kolla-{ansible,k8s}, puppet openstack | 16:49 |
| evrardjp | noonedeadpunk: it's not about rejecting it, because they make sense... It's about giving them as opt-in | 16:49 |
| mnaser | even the tripleo before redhat revived it :) | 16:49 |
| evrardjp | mnaser: wow, the HP thing? | 16:49 |
| mnaser | yep | 16:49 |
| mnaser | noonedeadpunk: when you say not rejecting containers, you mean machine containers or app containers | 16:50 |
| evrardjp | like odyssey4me would have said, at that time dinosaurs were roaming on earth | 16:50 |
| noonedeadpunk | mnaser I mean machine containers | 16:50 |
| cloudnull | jrosser ++ totally agree. | 16:50 |
| mnaser | yeah, we don't aim on dropping it | 16:50 |
| mnaser | we have users that are happy with it | 16:51 |
| noonedeadpunk | Yeah, then bare metal is really nice. But kata looks also pretty interesting. But I don't like docker to be honest.. | 16:51 |
| jrosser | i'm also here primarily becasue of the attractive architecture / approach | 16:51 |
| mnaser | kata is nothing but a docker runner | 16:51 |
| mnaser | it just runs docker containers in vms, thats' all | 16:51 |
| odyssey4me | it makes sense to me to have lxc only running on ubuntu, which formally supports it - and nspawn/lxc opt-in for those who want it | 16:51 |
| mnaser | yeah, the lxc stuff is pretty hacked up too | 16:51 |
| mnaser | because we depend on some other external repo | 16:51 |
| mnaser | and the more we stray away from upstream tooling the harder it gets | 16:52 |
| evrardjp | the urgency is remove it for centos | 16:52 |
| evrardjp | when is 8 official? | 16:52 |
| mnaser | having discussed all of this, is anyone not too opposed to removing containers from centos to be able to support rhel 8 and make it a bit more maintainable, i guess | 16:52 |
| spotz | Can we blame mhayden for something here?:) | 16:52 |
| mhayden | always | 16:53 |
| cloudnull | I'd really love folks to give the nspawn bits a spin. they need folks to use it and report where its all broken for them and their environments. | 16:53 |
| spotz | heheh, we miss you:) | 16:53 |
| *** shardy has joined #openstack-ansible | 16:53 | |
| prometheanfire | mhayden: we should do tacos this week | 16:53 |
| mnaser | cloudnull: i really tried to get nspawn working at gate, 17 patch sets later and i was still struggling | 16:53 |
| FrankZhang | wild Major found | 16:54 |
| prometheanfire | cloudnull: I need to bug you about the state of that | 16:54 |
| mnaser | and unfortunately we haven't been getting traction on it | 16:54 |
| evrardjp | FrankZhang: :) | 16:54 |
| mnaser | my knowledge in it is limited | 16:54 |
| odyssey4me | to my mind, given that we have a very limited centos/suse support base and people supporting/developing, perhaps we should consider scaling back their support to only without containers | 16:54 |
| evrardjp | odyssey4me: I would be fine with that. | 16:54 |
| jrosser | mnaser: theres an important distinction between "we removed this stuff becasue rhel8 broke the world" vs. "we removed this stuff becasue it's the general direction of travel of OSA" | 16:54 |
| *** mathlin has joined #openstack-ansible | 16:54 | |
| odyssey4me | jrosser yep, fair point | 16:54 |
| jrosser | i think we have enough confusion already between containers/metal, with very few actually using metal, and we carry the overhead | 16:55 |
| mnaser | jrosser: i agree. i don't want to take away from major users who's benefit is containers and the architecture we provide | 16:55 |
| evrardjp | odyssey4me: to be honest, lxc is supposed to be phased out in favor of new lxd/lxc bindings, right? | 16:55 |
| cloudnull | mnaser maybe we could spend some time on a mnaio or some env to get nspawn up and running and answer questions. | 16:55 |
| * cloudnull has access to hardware to do that | 16:55 | |
| odyssey4me | if we move to containers being opt-in, then we should change the current dynamic inventory to a simpler inventory plugin which can be easily enabled/disabled | 16:55 |
| mnaser | well, i'd be just happy with passing jobs to start with :( | 16:55 |
| spotz | I know we don't want to wait that long to make a decision but this could be a great Forum or PTG discussioon | 16:56 |
| evrardjp | odyssey4me: indeed | 16:56 |
| prometheanfire | odyssey4me: would we still need the current one to support people if they do opt in to containers? | 16:56 |
| evrardjp | spotz: I tried that though, it didn't bring many supporters | 16:56 |
| mnaser | also i think making our default deploy tooling simpler but leaving the complex one available is better | 16:56 |
| evrardjp | that's why I never worked on the removal :p | 16:56 |
| mnaser | most users don't want to throw a /24 at an openstack control plane deployment, they want 3 ips for each of their controllers | 16:56 |
| evrardjp | mnaser: couldn't it be on the side, in ops repo? | 16:56 |
| spotz | evrardjp: Weird, cause the in[ut would help make a design for direction:( | 16:56 |
| mnaser | *if* someone wants to do that, they can do it (and they probably have the knowledge to do it) | 16:57 |
| evrardjp | spotz: oh no I meant the lack of willingness to change was an input in itself | 16:57 |
| prometheanfire | mnaser: swich the control plane to v6, never run out of addresses :P | 16:57 |
| cloudnull | ^ :D | 16:57 |
| mnaser | baha | 16:57 |
| prometheanfire | then just use haproxy for 624 | 16:57 |
| mnaser | like or maybe as an idea | 16:57 |
| prometheanfire | :P | 16:57 |
| mnaser | we can decouple the dynamic inventory away out | 16:57 |
| odyssey4me | we've run around this circle many times, but until it matters enough to someone it's not going to happen | 16:57 |
| prometheanfire | ^ | 16:58 |
| mnaser | odyssey4me: the ipv6 circle or the metal circle | 16:58 |
| odyssey4me | for now it's easy to make centos metal only, and perhaps also work on some plays to transition any container deployments to metal when doing the upgrade | 16:58 |
| spotz | evrardjp: hehe, then they need to provide help!:) | 16:58 |
| odyssey4me | perhaps for suse we do the same given that we have a low support base and user base | 16:58 |
| mnaser | if folks are ok with that, i will do the work | 16:58 |
| odyssey4me | we leave ubuntu as-is until it matter enough to anyone to change up how that's all done | 16:59 |
| cloudnull | +1 | 16:59 |
| evrardjp | lgtm | 16:59 |
| *** macza has joined #openstack-ansible | 16:59 | |
| spotz | +1 | 16:59 |
| mnaser | awesome | 16:59 |
| evrardjp | and thanks mnaser for the work | 16:59 |
| evrardjp | it shouldn't be that hard if we keep lxc around | 16:59 |
| mnaser | thank you for your patience with my shenanigans dealing with dinosaurs | 16:59 |
| mnaser | :) | 16:59 |
| evrardjp | for ubuntu | 16:59 |
| cloudnull | would be sad to see SUSE support / test matrix reduced, but i understand | 16:59 |
| mnaser | anyone else have anything in mind? | 16:59 |
| odyssey4me | mnaser I think we can pretty simply just remove centos/suse from the openstack-ansible-tests templates, then add cross-repo integrated build tests to all roles with a metal build | 17:00 |
| evrardjp | as we won't change inventory, methods of deployments, but still simplify code for centos | 17:00 |
| mnaser | odyssey4me: that was my goal pretty much :) | 17:00 |
| mnaser | and then figure out upgrades | 17:00 |
| evrardjp | well | 17:00 |
| mnaser | anyways, we're kinda at time | 17:00 |
| prometheanfire | mnaser: I have an item to bring up for notice (after this) :P | 17:00 |
| evrardjp | it's not enough for clustering roles | 17:00 |
| mnaser | whats up prometheanfire ? | 17:00 |
| prometheanfire | the barbican role seems broken | 17:00 |
| mnaser | evrardjp: i had an idea for that too :) | 17:00 |
| mnaser | orly? | 17:01 |
| evrardjp | mnaser: multinode? | 17:01 |
| mnaser | evrardjp: yep | 17:01 |
| evrardjp | mnaser: good | 17:01 |
| mnaser | oh thats a lot of red | 17:01 |
| prometheanfire | at least in master | 17:01 |
| mnaser | fatal: [infra1]: FAILED! => {"changed": false, "cmd": "set -e\n if [ -d /opt/tempest-testing/bin ];\n then\n . /opt/tempest-testing/bin/activate\n fi\n tempest run --whitelist-file /root/workspace/etc/tempest_whitelist.txt", "delta": "0:00:02.464257", "end": "2019-01-03 18:26:46.166077", "msg": "non-zero return code", "rc": 1, "start": "2019-01-03 18:26:43.701820", "stderr": "", "stderr_lines": [], "stdout": "The | 17:01 |
| mnaser | specified regex doesn't match with anything", "stdout_lines": ["The specified regex doesn't match with anything"]} | 17:01 |
| jrosser | that is due to changes in tempest | 17:01 |
| mnaser | mayb arxcruz or chandankumar can help us with that | 17:01 |
| jrosser | and there was a patch this morning merged as the first part of addressing that | 17:01 |
| odyssey4me | yep, I've been doing some work on that front too | 17:01 |
| mnaser | okay cool, so maybe that is a good canary patch to see if it works or not :) | 17:02 |
| prometheanfire | mnaser: k, thanks | 17:02 |
| odyssey4me | https://review.openstack.org/628979 has been a bit of a work in progress | 17:02 |
| chandankumar | mnaser: can you point me the log url | 17:02 |
| mnaser | chandankumar: i looked at https://review.openstack.org/#/c/625634/ | 17:02 |
| evrardjp | I also have a few things to add for the record of the meeting: | 17:02 |
| jrosser | this https://github.com/openstack/openstack-ansible-os_tempest/commit/25b5533c30e328c80d29348dff0cfc0f2ac5e88f | 17:02 |
| jrosser | even if that doesnt fix barbican/designate it's the first step | 17:03 |
| evrardjp | 1) If there is a problem with SUSE packaging, please query in #openstack-rpm-packaging | 17:03 |
| odyssey4me | the previous issue was that distro installs didn't install the plugins to do the tests - now that's happening, nothing's setting the var that enables the plugin to be installed | 17:03 |
| chandankumar | jrosser: does barbican/designate failure does not got fixed? | 17:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Use the inventory to enable/disable services by default https://review.openstack.org/628979 | 17:04 |
| mnaser | evrardjp: anything else for the record of the meeting? or we can wrap up and keep discussing :) | 17:04 |
| evrardjp | 2) I am bringing the idea of gating OBS Staging->OBS repo using OSA for more stability | 17:04 |
| mnaser | yay | 17:05 |
| odyssey4me | that should solve it | 17:05 |
| evrardjp | thanks mnaser | 17:05 |
| evrardjp | and everyone! | 17:05 |
| odyssey4me | awesome, thanks evrardjp | 17:05 |
| odyssey4me | even just trying to install openstack using anythign would e a good start | 17:05 |
| mnaser | ++ | 17:06 |
| mnaser | okay, we're over but i think we covered most things | 17:06 |
| mnaser | i shall end :) | 17:06 |
| mnaser | thank you everyone!! | 17:06 |
| mnaser | #endmeeting | 17:06 |
| *** openstack changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: http://bit.ly/2xA1eZC" | 17:06 | |
| spotz | Thanks! | 17:06 |
| openstack | Meeting ended Tue Jan 8 17:06:17 2019 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:06 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2019/openstack_ansible_meeting.2019-01-08-16.00.html | 17:06 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2019/openstack_ansible_meeting.2019-01-08-16.00.txt | 17:06 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2019/openstack_ansible_meeting.2019-01-08-16.00.log.html | 17:06 |
| mnaser | odyssey4me: jrosser btw, regarding keeping containers around | 17:06 |
| mnaser | i have thought about the idea of decoupling the tooling that lays down the containers/networking stack and osa | 17:06 |
| mnaser | that way, you can just ansible-playbook -i osa-inventory foo.yml OR ansible-playbook -i your-static-file foo.yml | 17:07 |
| prometheanfire | cloudnull: is the networkd stuff merged? (aka, can I move forward on the gentoo stuff)? | 17:07 |
| mnaser | and you can just run a set of playbooks *before* OSA to lay down all of your network/containers | 17:07 |
| mnaser | and feed that into OSA | 17:07 |
| evrardjp | mnaser: I had that using LXD :p | 17:07 |
| odyssey4me | sure, all that it needs is the lxc/nspawn roles and an ansible iventory plugin - and none of that needs to be in OSA specifically, even if it is curated by this community | 17:07 |
| mnaser | so openstack ansible stays an openstack installer first, and it just deploys on different targets | 17:07 |
| mnaser | odyssey4me: yep! and that way it can even be used for anything else, not necessarily just openstack | 17:08 |
| odyssey4me | and yeah, the lxc stuff could evolve to use lxd - whereas the nspawn stuff can continue in its own right | 17:08 |
| evrardjp | the good thing with lxd,odyssey4me, is that it has a nice api you can use | 17:08 |
| mnaser | really, the same way we say: setup your physical hardware the way you want it, because we don't touch it | 17:08 |
| evrardjp | yeah | 17:09 |
| mnaser | we can do the same about containers, here's all the containers, now deploy openstack | 17:09 |
| *** DanyC has quit IRC | 17:09 | |
| *** hamzaachi has quit IRC | 17:09 | |
| mnaser | and then we can just have jobs that run that role first and feed the inventory to osa | 17:09 |
| evrardjp | it also simplifies ansible quite a lot, we don't need our wrapper anymore for example | 17:09 |
| *** DanyC has joined #openstack-ansible | 17:09 | |
| evrardjp | and we can have our own cli for parallel runs | 17:09 |
| jrosser | i have a whole pile of LXD work in the bag which could be appropriated | 17:09 |
| evrardjp | I would totally love to work on that, but my employer don't give me a second for it :p | 17:10 |
| mnaser | but my concern is for the bigger deployers with more specific use cases, does the separation of concerns still make things workable? | 17:10 |
| *** shardy has quit IRC | 17:10 | |
| chandankumar | jrosser: can we add a dependson https://review.openstack.org/#/c/628979/ on this one https://review.openstack.org/#/c/625634/ | 17:11 |
| chandankumar | jrosser: from above review I am seeing only keystone tempest plugin is getting installed | 17:11 |
| *** DanyC has quit IRC | 17:12 | |
| chandankumar | jrosser: http://logs.openstack.org/34/625634/1/check/openstack-ansible-functional-distro_install-centos-7/66d08ea/logs/ara-report/result/85edc814-2096-4012-b7a3-bf163f893479/ | 17:12 |
| prometheanfire | chandankumar: that's my patch, so yes :P | 17:12 |
| evrardjp | mnaser: what do you mean ? | 17:12 |
| chandankumar | prometheanfire: let me do that | 17:13 |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible-os_barbican master: Force force-tlsv12 only https://review.openstack.org/625634 | 17:13 |
| prometheanfire | already done :| | 17:13 |
| chandankumar | prometheanfire: cool | 17:13 |
| evrardjp | mnaser: I think it's easier if you're a bigger cloud: Finer tuning of the infra | 17:13 |
| evrardjp | the only thing that matters between the two parts is the inventory | 17:13 |
| odyssey4me | mnaser evrardjp jrosser I think it's something we can do nicely if we combine efforts - but it will need to be a combined effort, not bits and pieces. I'm suggesting it would be a series goal, with no other goals. | 17:14 |
| evrardjp | which is easily solvable if you think of writing an inventory of the created containers. | 17:14 |
| evrardjp | odyssey4me: agreed | 17:14 |
| odyssey4me | Perhaps people could hack a few things here and there, but then at the PTG we pull it together. | 17:14 |
| evrardjp | this is a refactor of the mindset | 17:14 |
| evrardjp | this cannot be done in one cycle | 17:14 |
| evrardjp | it needs to be spread in multiple steps in at least two cycles | 17:15 |
| odyssey4me | A way to make this easily workable, for example, is to put together a way to use playbooks/roles in a more pluggable way. | 17:15 |
| odyssey4me | So if you want LXC things, you pull them in. If you don't, then it doesn't have any of that. | 17:15 |
| evrardjp | I was lacking creativity for the user experience tbh. | 17:16 |
| jrosser | dividing the inventory away sounds nice | 17:17 |
| evrardjp | we can now, as there is no reliance on the inventory anymore (except if that was brought back) | 17:17 |
| odyssey4me | the only reliance right now is container builds | 17:18 |
| odyssey4me | if it's implemented as a plugin, rather than a dynamic inventory, then we'd be able to use standard ansible all the way for all group_vars, host_vars, overrides, etc and the inventory plugin would only be handling the container things | 17:19 |
| evrardjp | yeah so decoupling can be done in multiple phases: Being bare metal first in roles is the first step | 17:19 |
| jrosser | it would be interesting to understand pain points in more complex deployments | 17:19 |
| odyssey4me | and it'd be easy to whitelist/blacklist to enable/disable | 17:19 |
| jrosser | having a sane world where you have seperate OSA and ceph deploys | 17:19 |
| jrosser | but your ops repo ELK stack is across both of those | 17:19 |
| jrosser | ^ these kind of things are where i struggle | 17:20 |
| evrardjp | jrosser: I think this is fine already | 17:20 |
| evrardjp | as you can now chain inventoriues | 17:20 |
| evrardjp | -u | 17:20 |
| odyssey4me | I think most of the pain comes from having the inventory being a dynamic inventory at the moment, rather than a plugin | 17:20 |
| evrardjp | if you have no reliance over openstack-ansible CLI, it's just super easy to have that | 17:20 |
| evrardjp | odyssey4me: what kind of plugin you mean? | 17:21 |
| odyssey4me | because in this way you have to do a bunch of clunky things to string inventories together | 17:21 |
| odyssey4me | if we convert the dynamic inventory into an inventory plugin then using it will be far simpler | 17:21 |
| evrardjp | yeah but it really doesn't make sense to have that IMO | 17:22 |
| odyssey4me | sure, but some want it | 17:22 |
| evrardjp | they want the easiness of a yaml edit that manages it all | 17:22 |
| evrardjp | but that can be changed using native ansible inventory nowadays | 17:22 |
| noonedeadpunk | yeah, getting rid of openstack-ansible cli is nice as gives opportunity to use ansible runner safely for your python scripts | 17:22 |
| evrardjp | it's just that ansible yaml inventory is far more ugly than what we have :) | 17:24 |
| noonedeadpunk | but inventory migration during some upgrade will be probably a nightmare... | 17:25 |
| *** mathlin has quit IRC | 17:28 | |
| odyssey4me | evrardjp you're forgetting that the dynamic inventory creates containers names and IP's - it's convenient... having that done by a plugin keeps the convenience for those who want it | 17:28 |
| cloudnull | mnaser noonedeadpunk: The issue with JSON test failing (https://bugs.launchpad.net/openstack-ansible/+bug/1810537) is fixed with https://review.openstack.org/#/c/628197 | 17:29 |
| openstack | Launchpad bug 1810537 in openstack-ansible "volume creation fails after successful installation" [Undecided,In progress] - Assigned to Mohammed Naser (mnaser) | 17:29 |
| odyssey4me | and makes it easier to disable for those who don't | 17:29 |
| evrardjp | I like the idea of the toggling tbh | 17:29 |
| evrardjp | and the separation of concerns | 17:29 |
| evrardjp | because if you don't want generation of IPs or are running IPv6, you don't need that :p | 17:30 |
| cloudnull | Just locally ran https://review.openstack.org/#/c/588953 and all is good with that patch applied | 17:30 |
| evrardjp | odyssey4me: there is this kind of nice "nmap" inventory that's good enough :D | 17:30 |
| odyssey4me | something broke between dec 6 and today - check the zuul logs for the cross-repo integrated test: https://review.openstack.org/#/c/623253/ | 17:30 |
| cloudnull | seems the resource_filters.json is not included when the venv is built. | 17:31 |
| evrardjp | if you auto assign ips using dhcp, you don't need that feature odyssey4me :) | 17:31 |
| odyssey4me | cloudnull yeah, that's cos https://github.com/openstack/cinder/blob/master/setup.cfg#L28-L30 doesn't include https://github.com/openstack/cinder/blob/master/etc/cinder/resource_filters.json - we should fix that | 17:32 |
| cloudnull | ah ha! | 17:33 |
| *** kaiokmo has quit IRC | 17:33 | |
| cloudnull | https://review.openstack.org/629272 odyssey4me | 17:39 |
| odyssey4me | boom, thanks cloudnull | 17:40 |
| cloudnull | so when that, or something like it, and the smart_sources commit for cinder are merged we can drop our template | 17:40 |
| cloudnull | thank you , it looks like your the original committer for the data_files input into cinder :) | 17:41 |
| *** cshen has joined #openstack-ansible | 17:41 | |
| odyssey4me | cloudnull yep, it's all part of the journey to make our lives simpler :) | 17:42 |
| noonedeadpunk | cloudnull odyssey4me oh, great! I've was wondering about how configs gets into venv about for a week:) Glad, that I've got the answer. And thanks for this patch! | 17:43 |
| ansmith | odyssey4me: hello, circling around tempest failure with qdrouterd role, same tempest test passes in oslo.messaging ci with ovs, if i wanted to try ovs in place of lxb, is there more to it than changing neutron_plugin_type? | 17:45 |
| ansmith | odyssey4me: I see there may be a difference to how notifications take place with the plugins which may explain things | 17:46 |
| odyssey4me | ansmith I honestly have no idea - mnaser may be able to advise there, or jamesdenton - they know how to use ovs instead of LXB for neutron only. | 17:47 |
| *** cshen has quit IRC | 17:47 | |
| odyssey4me | ok folks, I'm out for the night - cheerio! | 17:47 |
| jamesdenton | ansmith it is pretty straightforward: https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-openvswitch.html | 17:47 |
| ansmith | jamesdenton: thanks, will give it a whirl | 17:48 |
| jamesdenton | sure, i'll be around if you run into issues | 17:48 |
| *** DanyC has joined #openstack-ansible | 17:50 | |
| *** mathlin has joined #openstack-ansible | 17:51 | |
| *** DanyC has quit IRC | 17:54 | |
| *** cshen has joined #openstack-ansible | 17:54 | |
| prometheanfire | chandankumar: https://review.openstack.org/#/c/625634/ finished it's testing (failed again) | 18:08 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Adds resource_filters.json distribution https://review.openstack.org/628197 | 18:09 |
| *** hamerins has quit IRC | 18:09 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_cinder stable/rocky: Adds resource_filters.json distribution https://review.openstack.org/629278 | 18:10 |
| *** hamerins has joined #openstack-ansible | 18:11 | |
| openstackgerrit | Frank Zhang proposed openstack/openstack-ansible-os_barbican master: Removing stale policy file https://review.openstack.org/629280 | 18:22 |
| *** mathlin has quit IRC | 18:25 | |
| *** partlycloudy has joined #openstack-ansible | 18:33 | |
| *** thuydang has joined #openstack-ansible | 18:39 | |
| * mnaser read sbuffer | 18:42 | |
| mnaser | cool, everyone is been helped :) | 18:44 |
| partlycloudy | Hi everyone, I have a question: I want to use the same network for the external API endpoints and the provider network. How does OSA assign external IPs on the load balancer nodes? Would it start from the {{ external_lb_vip_address }} and then goes down continuesouly? | 18:45 |
| *** mathlin has joined #openstack-ansible | 18:48 | |
| cloudnull | partlycloudy in config you specify a CIDR and OSA will pull from that at will. All host addresses will automatically be excluded from the available IP range and should you need there's an additional exclude list where you can add more IPs/Ranges a pool of unavailable IPs. | 18:51 |
| mnaser | partlycloudy: however for load balancer, haproxy specifically, it runs on in baremetal | 18:51 |
| mnaser | which means that it will just use the interface that you configured, yourself | 18:51 |
| cloudnull | the inventory code will randomly pull an IP from the CIDR you provide so it's non linear. | 18:51 |
| *** lbragsta_ has joined #openstack-ansible | 18:52 | |
| *** lbragstad has quit IRC | 18:52 | |
| cloudnull | also what mnaser said :) | 18:52 |
| cloudnull | anyone around want to give https://review.openstack.org/#/c/588960/ a review ? | 18:54 |
| cloudnull | and https://review.openstack.org/#/c/607462 | 18:54 |
| *** lbragsta_ is now known as lbragstad | 18:55 | |
| partlycloudy | Thank you, cloudnull and mnaser. Suppose I use the 3 infra nodes as load balancer, do I need to pre-configure external IPs (in the same range as VIP) on the infra nodes? | 18:59 |
| mnaser | partlycloudy: yep, that way keepalived can host the vip on those interfaces | 18:59 |
| mnaser | cloudnull: checkin | 18:59 |
| mnaser | deleting mah code | 19:00 |
| partlycloudy | mnaser: got it. thank you very much! :-) | 19:00 |
| noonedeadpunk | folks, can you please check this cherry-pick? https://review.openstack.org/#/c/629278/ | 19:01 |
| *** kaiokmo has joined #openstack-ansible | 19:01 | |
| spotz | done noonedeadpunk | 19:12 |
| noonedeadpunk | thanks! | 19:16 |
| *** mathlin has quit IRC | 19:22 | |
| *** hamerins has quit IRC | 19:34 | |
| *** hamerins has joined #openstack-ansible | 19:35 | |
| *** cshen has quit IRC | 19:35 | |
| *** SimAloo has joined #openstack-ansible | 19:36 | |
| *** partlycloudy has quit IRC | 19:37 | |
| *** cshen has joined #openstack-ansible | 19:38 | |
| *** thuydang has quit IRC | 19:40 | |
| *** thuydang has joined #openstack-ansible | 19:43 | |
| *** mathlin has joined #openstack-ansible | 19:45 | |
| *** cshen has quit IRC | 19:46 | |
| *** cshen has joined #openstack-ansible | 19:48 | |
| *** cshen has quit IRC | 19:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/rocky: Adds resource_filters.json distribution https://review.openstack.org/629278 | 19:57 |
| *** gary_perkins has quit IRC | 20:01 | |
| *** gary_perkins has joined #openstack-ansible | 20:06 | |
| *** thuydang has left #openstack-ansible | 20:13 | |
| *** thuydang has joined #openstack-ansible | 20:15 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cinder master: Cleanup files and templates using smart sources https://review.openstack.org/588953 | 20:24 |
| *** spsurya has quit IRC | 20:41 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible master: Update inventory generator to support container gateway override https://review.openstack.org/629309 | 20:47 |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-lxc_container_create master: Update container interface template to disable DHCP routes when gateway set https://review.openstack.org/629310 | 20:48 |
| *** tosky has quit IRC | 20:49 | |
| * jrosser looks at this very suspicoiusly https://github.com/ceph/ceph-ansible/commit/416b503476c9859cf38db1d71a64cb77a4117a6b | 20:52 | |
| jamesdenton | evrardjp jrosser re: the gateway override for containers... the static_routes are handled on a per network/interface basis, and unless you disable the route pushed via DHCP on eth0 Ivia lxcbr0) you'll have overlapping default routes, as someone mentioned. https://review.openstack.org/629309 and https://review.openstack.org/629310 a first pass if that's really a feature we want to implement. Otherwise we should | 20:53 |
| jamesdenton | probably remove mention of it from docs | 20:53 |
| jrosser | "leseb merged commit 416b503 into stable-3.2 2 days ago" pretty much when our integrated repo ceph jobs stopped working | 20:53 |
| prometheanfire | chandankumar: can you confirm that os_barbican does not set up a web server in front of uwsgi (relies on haproxy) | 20:59 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [DNM] test rolling back last commit of ceph-ansible stable-3.2 https://review.openstack.org/629312 | 20:59 |
| *** cshen has joined #openstack-ansible | 21:00 | |
| openstackgerrit | Frank Zhang proposed openstack/openstack-ansible-os_barbican master: Removing stale policy file https://review.openstack.org/629280 | 21:02 |
| *** radeks has quit IRC | 21:02 | |
| *** cshen has quit IRC | 21:06 | |
| logan- | jrosser: makes sense, we probably need to run ceph-facts just after the ceph-defaults role | 21:09 |
| *** jbadiapa has quit IRC | 21:10 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Call ceph-facts role as required by changes to ceph-ansible https://review.openstack.org/629317 | 21:14 |
| *** hamerins has quit IRC | 21:14 | |
| jrosser | logan-: something like that ^, can you keep an eye on that and https://review.openstack.org/629312 should give a comparison with the ceph-facts patch rolled out | 21:14 |
| jrosser | i'm done for today | 21:15 |
| logan- | will do | 21:15 |
| jrosser | ty | 21:15 |
| logan- | thanks | 21:15 |
| *** hamerins has joined #openstack-ansible | 21:15 | |
| *** cshen has joined #openstack-ansible | 21:18 | |
| *** cshen has quit IRC | 21:23 | |
| openstackgerrit | Jakob Englisch proposed openstack/ansible-hardening master: Chrony: make ntp server options configurable https://review.openstack.org/629318 | 21:23 |
| *** partlycloudy has joined #openstack-ansible | 21:23 | |
| *** ansmith has quit IRC | 21:26 | |
| *** jamesdenton has quit IRC | 21:35 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cinder master: Cleanup files and templates using smart sources https://review.openstack.org/588953 | 21:39 |
| *** thuydang has quit IRC | 21:42 | |
| mnaser | we reaaaaaaally gotta stop hardcoding config options ;( | 21:47 |
| *** jamesdenton has joined #openstack-ansible | 21:56 | |
| *** thuydang has joined #openstack-ansible | 22:05 | |
| *** DanyC has joined #openstack-ansible | 22:11 | |
| spotz | mnaser: wasn't me!:) | 22:11 |
| *** DanyC has quit IRC | 22:15 | |
| *** thuydang has quit IRC | 22:16 | |
| openstackgerrit | Damian Cikowski (dcdamien) proposed openstack/openstack-ansible-os_designate master: Set project that owns managed resources https://review.openstack.org/628785 | 22:30 |
| openstackgerrit | Jakob Englisch proposed openstack/ansible-hardening master: Chrony: add an option to sync the hardware clock https://review.openstack.org/629330 | 22:46 |
| *** tosky has joined #openstack-ansible | 23:01 | |
| *** strobelight has quit IRC | 23:03 | |
| *** SimAloo has quit IRC | 23:07 | |
| *** cshen has joined #openstack-ansible | 23:08 | |
| *** cshen has quit IRC | 23:13 | |
| *** hamerins has quit IRC | 23:13 | |
| *** l-j has joined #openstack-ansible | 23:37 | |
| cloudnull | mnaser which hard coded op have you found now? | 23:40 |
| mnaser | cloudnull: keystone allowed auth methods | 23:40 |
| mnaser | its token and password | 23:41 |
| mnaser | doesnt include application_credentials | 23:41 |
| cloudnull | oh... | 23:41 |
| cloudnull | that's kinda a bummer. but at least you should be able to override that with config_template? | 23:41 |
| cloudnull | however, i agree, it should be an array that a user can easily tweak | 23:42 |
| cloudnull | if folks can review https://review.openstack.org/#/c/588953 and https://review.openstack.org/#/c/588960 they now work and get rid of some (most) of the config related templates we're carrying. | 23:42 |
| cloudnull | kinda related :) | 23:42 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible master: Call ceph-facts role as required by changes to ceph-ansible https://review.openstack.org/629317 | 23:52 |
| *** ansmith has joined #openstack-ansible | 23:53 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!