| *** cshen has joined #openstack-ansible | 00:23 | |
| *** cshen has quit IRC | 00:59 | |
| *** sum12 has quit IRC | 01:04 | |
| *** sum12 has joined #openstack-ansible | 01:13 | |
| *** cshen has joined #openstack-ansible | 01:36 | |
| *** cshen has quit IRC | 01:41 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_panko master: remove unnecessary '=' https://review.opendev.org/587296 | 02:49 |
|---|---|---|
| openstackgerrit | Merged openstack/openstack-ansible-os_panko master: format rst code https://review.opendev.org/614429 | 02:49 |
| snadge | i wish i understood rabbit more, neutron is failing to connect to rabbit, giving me this error.. "AMQPLAIN login refused: user 'neutron' - invalid credentials" | 03:24 |
| snadge | the controller ran out of disk space, i had to delete some journals, now rabbit seems completely broken | 03:25 |
| snadge | i did a queue reset, and now this authentication problem | 03:25 |
| *** cshen has joined #openstack-ansible | 03:37 | |
| *** cshen has quit IRC | 03:41 | |
| *** cshen has joined #openstack-ansible | 03:52 | |
| *** cshen has quit IRC | 03:57 | |
| *** evrardjp has quit IRC | 04:03 | |
| *** dave-mccowan has quit IRC | 04:06 | |
| *** evrardjp has joined #openstack-ansible | 04:10 | |
| *** evrardjp has quit IRC | 04:36 | |
| *** evrardjp has joined #openstack-ansible | 04:36 | |
| snadge | ack.. it seems each roll, eg neutron, nova, cinder etc, installs its own rabbitmq user.. so I pretty much have to blow everything away and start again? | 04:53 |
| *** cshen has joined #openstack-ansible | 05:01 | |
| *** cshen has quit IRC | 05:05 | |
| *** cshen has joined #openstack-ansible | 05:17 | |
| *** cshen has quit IRC | 05:21 | |
| *** udesale has joined #openstack-ansible | 05:28 | |
| openstackgerrit | Merged openstack/ansible-role-systemd_service master: Missing document start "---" https://review.opendev.org/715105 | 06:00 |
| *** cshen has joined #openstack-ansible | 06:15 | |
| *** udesale_ has joined #openstack-ansible | 06:23 | |
| *** udesale has quit IRC | 06:25 | |
| *** udesale_ has quit IRC | 06:27 | |
| *** miloa has joined #openstack-ansible | 06:42 | |
| *** thuydang has joined #openstack-ansible | 06:44 | |
| *** thuydang has quit IRC | 07:15 | |
| *** cshen has quit IRC | 07:18 | |
| *** DanyC has joined #openstack-ansible | 07:21 | |
| *** cshen has joined #openstack-ansible | 07:22 | |
| *** DanyC has quit IRC | 07:22 | |
| *** DanyC has joined #openstack-ansible | 07:23 | |
| *** tosky has joined #openstack-ansible | 07:25 | |
| *** DanyC has quit IRC | 07:27 | |
| *** cshen has quit IRC | 07:27 | |
| *** rpittau|afk is now known as rpittau | 07:34 | |
| *** arxcruz|off is now known as arxcruz | 07:35 | |
| *** cshen has joined #openstack-ansible | 07:37 | |
| *** DanyC has joined #openstack-ansible | 07:38 | |
| *** jbadiapa has joined #openstack-ansible | 07:47 | |
| *** jbadiapa has quit IRC | 07:50 | |
| *** jbadiapa has joined #openstack-ansible | 07:51 | |
| *** shyamb has joined #openstack-ansible | 08:15 | |
| chandankumar | noonedeadpunk, thanks, last review for tripleo c8 https://review.opendev.org/#/c/715367/ | 08:23 |
| *** shyam89 has joined #openstack-ansible | 08:38 | |
| *** udesale has joined #openstack-ansible | 08:38 | |
| *** shyamb has quit IRC | 08:41 | |
| *** shyam89 has quit IRC | 08:53 | |
| *** shyam89 has joined #openstack-ansible | 08:54 | |
| jrosser | o/ morning | 08:57 |
| noonedeadpunk | mornings | 09:01 |
| *** shyam89 has quit IRC | 09:30 | |
| noonedeadpunk | jrosser: we have broken rally for train and earlier... And the problem is that it's smth not easy fix with contraints because of https://github.com/openstack/rally-openstack/blob/master/requirements.txt#L6... | 09:45 |
| noonedeadpunk | we probably need to add some local contraints | 09:46 |
| jrosser | we could pin rally like we did for tempest back to stable releases? | 09:46 |
| jrosser | becasue similarly it is branchless | 09:47 |
| *** gshippey has joined #openstack-ansible | 09:47 | |
| noonedeadpunk | yes, but we isntall rally-openstack, but not rally... | 09:47 |
| noonedeadpunk | and rally-openstack has rally requirement in it | 09:47 |
| noonedeadpunk | which is rally>=0.11.0 for previous version... | 09:48 |
| noonedeadpunk | and 3.0.0 will be isntalled anyway.... | 09:48 |
| noonedeadpunk | or I didn't get what you mean... | 09:49 |
| noonedeadpunk | oh, we can pin here? https://opendev.org/openstack/openstack-ansible/src/branch/master/global-requirement-pins.txt | 09:50 |
| noonedeadpunk | yeah, I think we can, nice! thanks! | 09:53 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Fix CentOS python executable https://review.opendev.org/715215 | 09:57 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Fix CentOS python executable https://review.opendev.org/715215 | 09:58 |
| *** cshen has quit IRC | 10:13 | |
| *** shyamb has joined #openstack-ansible | 10:39 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_panko master: Trivial: Fix the pep8 warning https://review.opendev.org/589055 | 10:55 |
| *** rpittau is now known as rpittau|bbl | 10:56 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.opendev.org/486156 | 11:04 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests stable/stein: Set requirements_git_url during functional tests https://review.opendev.org/715915 | 11:07 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests stable/stein: Do not compress log files before uploading https://review.opendev.org/714414 | 11:08 |
| *** DanyC has quit IRC | 11:20 | |
| noonedeadpunk | jrosser: oh.. can you remind me why did we need to specify branch here? https://review.opendev.org/#/c/715915/1/test-ansible-env-prep.sh | 11:25 |
| noonedeadpunk | Like in current state (without /branch/) it work with both providing SHA and branch... | 11:26 |
| noonedeadpunk | but actually once we set only branch here as for now.... we can avoid extra redirect... | 11:27 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Added tempest_tempestconf_profile_ specific vars https://review.opendev.org/714601 | 11:30 |
| *** DanyC has joined #openstack-ansible | 11:31 | |
| *** DanyC has quit IRC | 11:37 | |
| *** DanyC has joined #openstack-ansible | 11:38 | |
| *** DanyC has joined #openstack-ansible | 11:39 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Make train to use python2 https://review.opendev.org/715215 | 11:40 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Fix python2 support https://review.opendev.org/715215 | 11:42 |
| jrosser | noonedeadpunk: i did change that to avoid a redirect | 11:47 |
| jrosser | but does that break it for a SHA? | 11:47 |
| noonedeadpunk | yeah | 11:47 |
| jrosser | oh hmm | 11:47 |
| noonedeadpunk | as you provide branch you cat set only branch | 11:47 |
| noonedeadpunk | while redirect can distinguish between sha barnch and tag I guess | 11:48 |
| jrosser | lets fix that | 11:48 |
| noonedeadpunk | but I guess in test we do hardcode branch as for now? | 11:48 |
| jrosser | we do, yes - i think the tests repo only deals in terms of branches there | 11:49 |
| jrosser | different for the integrated repo | 11:49 |
| noonedeadpunk | so we maybe can leave as is. will fix once we'll need sha to be used there :p | 11:49 |
| noonedeadpunk | or as you wish actually | 11:50 |
| jrosser | for the tests repo i think it can stay - becasue that value comes from .gitreview defaultbranch | 11:51 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Remove python2 or python3 problem pip packages https://review.opendev.org/713849 | 11:55 |
| *** thuydang has joined #openstack-ansible | 11:59 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Fix python2 support https://review.opendev.org/715215 | 12:00 |
| *** jamesden_ has joined #openstack-ansible | 12:02 | |
| jrosser | this is going to need attention https://review.opendev.org/#/c/710251/ | 12:03 |
| jrosser | not the patch itself but the failures | 12:03 |
| *** rh-jelabarre has joined #openstack-ansible | 12:04 | |
| noonedeadpunk | is there apparmour for centos at all... | 12:05 |
| noonedeadpunk | hm, and lxc jobs do not fail... So weird... | 12:06 |
| noonedeadpunk | ah, so it's not actually basic lxc jobs which fail... | 12:07 |
| *** macz_ has joined #openstack-ansible | 12:09 | |
| *** macz_ has quit IRC | 12:10 | |
| *** macz_ has joined #openstack-ansible | 12:11 | |
| openstackgerrit | Sorin Sbarnea proposed openstack/openstack-ansible-os_tempest master: Enable placeholder openstack-tox-molecule job https://review.opendev.org/673859 | 12:14 |
| *** shyamb has quit IRC | 12:16 | |
| chandankumar | noonedeadpunk, jrosser we are working on unifing tempest skip list and here it the proposal https://hackmd.io/twHmV2f2SGCMvfHjLMoGoA love to hear your feedback | 12:16 |
| noonedeadpunk | chandankumar: I think tests to skip might jsut be added to tempest_test_blacklist https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/defaults/main.yml#L101 ? | 12:22 |
| chandankumar | noonedeadpunk, yes, but same tests starts failing at multiple jobs then we need to add the same tests to blacklist at multiple places | 12:24 |
| noonedeadpunk | but I probably missing problem basics bacause of my lack of knowldge of tripleo basics | 12:24 |
| chandankumar | noonedeadpunk, I will rewrite the problem statement | 12:25 |
| chandankumar | noonedeadpunk, will ping back for review | 12:25 |
| jrosser | chandankumar: i think it is possible even now to take what you describe in "known_failures", extract the 'test' field filtered by 'job' and pass that to tempest_test_blacklist? | 12:27 |
| noonedeadpunk | tempest_test_blacklist accepts their format now | 12:27 |
| noonedeadpunk | https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_post_install.yml#L169-L175 | 12:28 |
| noonedeadpunk | I guess the problems is that they want to maintain exceptions in a separate repo which will be included by os_temepst | 12:29 |
| noonedeadpunk | and will fulfil tempest_test_blacklist based on that... | 12:29 |
| noonedeadpunk | or smth like this... | 12:29 |
| jrosser | yes - that sounds totally fine, except it maybe prefereable to consume that repo then pass a var to os_tempest | 12:29 |
| noonedeadpunk | yeah | 12:30 |
| *** andrewbonney has joined #openstack-ansible | 12:31 | |
| jrosser | this pattern is quite attractive https://github.com/openstack/openstack-ansible/blob/master/playbooks/common-tasks/dynamic-address-fact.yml | 12:32 |
| jrosser | the the repo containing the skip list was a role it could set_fact a variable whose name was defined in the calling playbook | 12:33 |
| *** rpittau|bbl is now known as rpittau | 12:46 | |
| *** redrobot has quit IRC | 12:55 | |
| *** Guest43440 has joined #openstack-ansible | 12:56 | |
| *** Guest43440 is now known as redrobot | 12:58 | |
| *** thuydang has quit IRC | 13:02 | |
| chandankumar | noonedeadpunk, jrosser we will be putting openstack tempest skiplist problem statement as a tomorrow meeting item, arxcruz will be leading that | 13:04 |
| arxcruz | yes, i will :) | 13:04 |
| noonedeadpunk | ok, good | 13:08 |
| *** tlunkw has joined #openstack-ansible | 13:08 | |
| *** rh-jlabarre has joined #openstack-ansible | 13:12 | |
| *** irclogbot_0 has quit IRC | 13:13 | |
| *** guilhermesp has quit IRC | 13:15 | |
| *** gundalow has quit IRC | 13:15 | |
| *** mmercer has quit IRC | 13:15 | |
| *** rh-jelabarre has quit IRC | 13:15 | |
| *** tinwood has quit IRC | 13:15 | |
| *** mmercer has joined #openstack-ansible | 13:17 | |
| *** tinwood has joined #openstack-ansible | 13:17 | |
| *** irclogbot_1 has joined #openstack-ansible | 13:17 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Install pcre-devel to build pyScss2 https://review.opendev.org/715652 | 13:19 |
| *** thuydang has joined #openstack-ansible | 13:20 | |
| *** gundalow has joined #openstack-ansible | 13:26 | |
| *** mgariepy has quit IRC | 13:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/stein: Set requirements_git_url during functional tests https://review.opendev.org/715915 | 13:42 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/stein: Do not compress log files before uploading https://review.opendev.org/714414 | 13:42 |
| *** mgariepy has joined #openstack-ansible | 13:43 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Fix python2 support https://review.opendev.org/715215 | 13:47 |
| *** kleini has joined #openstack-ansible | 13:47 | |
| *** miloa has quit IRC | 13:52 | |
| *** thuydang has quit IRC | 14:20 | |
| jrosser | noonedeadpunk: this fails, should it be "{{ item|bool }}" instead? | 14:22 |
| jrosser | https://review.opendev.org/#/c/713849/ | 14:22 |
| noonedeadpunk | oh, probably yes... | 14:24 |
| noonedeadpunk | I think youre right | 14:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Remove python2 or python3 problem pip packages https://review.opendev.org/713849 | 14:25 |
| *** guilhermesp has joined #openstack-ansible | 14:29 | |
| *** tlunkw has quit IRC | 14:34 | |
| kleini | I have a problem with the latest stable/stein deployment. It fails when the service project should be created in Keystone. Keystone and HAproxy work without any problems. I can access the API from the deployment host with the openstack client. Unfortunately the openstacksdk used to create the service project tells me: "Cloud default was not found." Where in OpenStack-Ansible is this cloud default configured? | 14:35 |
| jrosser | the service setup will normally happen on the utility container | 14:35 |
| jrosser | you should find an openrc / clouds.yaml in there | 14:36 |
| kleini | the os_project task is run on the deployment host. I found the clouds.yml there in /root/.config/openstack/clouds.yml and it is configured correctly. Now I need to find, why it does not work for Keystone deployment. | 14:38 |
| jrosser | if you have any output to share you can put it at paste.openstack.org | 14:39 |
| kleini | Can I output some maybe faulty task_vars with Ansible debugger? How is this clouds.yml passed in Ansible to the openstacksdk? | 14:39 |
| noonedeadpunk | kleini: generally os_project tasks and etc should be delegagted_to utiity container, so yeah, output is worth seeing | 14:40 |
| kleini | http://paste.openstack.org/show/791342/ <- this is normal output | 14:40 |
| kleini | If I enable more debugging with -vvvv option, I see, that the task is delegated from controller_keystone_container to localhost. | 14:41 |
| kleini | Again, I don't see any TCP traffic on HAproxy or Keystone nginx WSGI. So it fails already in OpenStack-Ansible somewhere with resolving the configuration for the 'default' cloud. | 14:43 |
| noonedeadpunk | kleini: try to uncomment this https://opendev.org/openstack/openstack-ansible/src/branch/master/etc/openstack_deploy/user_variables.yml#L33 | 14:43 |
| noonedeadpunk | or you can get clouds.yaml from utility container and place it to the deploy host | 14:43 |
| noonedeadpunk | jrosser: hm, maybe we should also place clouds.yaml to the deploy host if we set localhost as a default? | 14:44 |
| kleini | as already said, clouds.yml in root user on deployment host looks totally correct. | 14:44 |
| noonedeadpunk | oh | 14:44 |
| jrosser | i do wonder if we should default it to the utility container always | 14:45 |
| noonedeadpunk | do you run openstack-ansible with sudo or from root user directrly? | 14:45 |
| jrosser | becasue the environment is well controlled | 14:45 |
| noonedeadpunk | or that yes | 14:45 |
| kleini | I run it with sudo | 14:46 |
| noonedeadpunk | kleini: also you may try placing clouds.yaml under /etc/openstack folder... | 14:46 |
| kleini | maybe that is the problem. damn | 14:46 |
| openstackgerrit | Duncan Martin Walker proposed openstack/openstack-ansible-ops master: Expose Journalbeat queue configuration https://review.opendev.org/715969 | 14:47 |
| noonedeadpunk | kleini: I guess you have $USER != root in your env, so clouds.yaml then be under your user, not root | 14:47 |
| noonedeadpunk | or globally under /etc/openstack/clouds.yaml | 14:47 |
| *** yoctozepto has quit IRC | 15:07 | |
| *** yoctozepto has joined #openstack-ansible | 15:08 | |
| *** bverschueren has quit IRC | 15:09 | |
| *** Miouge has quit IRC | 15:09 | |
| *** arxcruz has quit IRC | 15:09 | |
| *** bverschueren has joined #openstack-ansible | 15:10 | |
| *** EmilienM has quit IRC | 15:10 | |
| *** arxcruz has joined #openstack-ansible | 15:10 | |
| *** Miouge has joined #openstack-ansible | 15:11 | |
| kleini | thank you very much. using sudo caused the problem. learned one more very important thing. | 15:16 |
| kleini | Now the service catalog from Keystone seems to be empty, which breaks creating the service project. | 15:17 |
| openstackgerrit | Merged openstack/openstack-ansible-os_panko master: Trivial: Fix the pep8 warning https://review.opendev.org/589055 | 15:19 |
| noonedeadpunk | hm... not sure how good this advice is, but maybe you can drop keystone db if you didn't go futher keystone isntallation and re-run os-keystone-install playbook? | 15:23 |
| kleini | yes, I think the setup of keystone is broken somehow | 15:23 |
| noonedeadpunk | or even db+keystone containers jsut to be sure | 15:23 |
| noonedeadpunk | you run in from glance | 15:23 |
| noonedeadpunk | like openstack-ansible /opt/openstack-ansible/playbooks/containers-lxc-destroy.yml --limit keystone_all to drop them | 15:25 |
| noonedeadpunk | and openstack-ansible /opt/openstack-ansible/playbooks/containers-lxc-create.yml --limit keystone_all,lxc_hosts to create them again | 15:25 |
| kleini | I know. But I use nspawn containers ;-) | 15:26 |
| noonedeadpunk | oh. let us know if they | 15:26 |
| noonedeadpunk | re working :p | 15:26 |
| noonedeadpunk | as they were left half-backed several releases ago | 15:27 |
| noonedeadpunk | s/backed/baked/ | 15:27 |
| *** jgwentworth is now known as melwitt | 15:33 | |
| noonedeadpunk | btw it would be gret if we had them revived and in CI I guess... | 15:33 |
| kleini | With stable/stein I often deployed now a working OpenStack instance with nspawn containers. I have some tricky configurations like existing Ceph, that I want to use for images and volumes, but generally it works good. | 15:43 |
| noonedeadpunk | oh, good to know | 15:44 |
| *** fridtjof[m] has quit IRC | 15:45 | |
| *** soren has quit IRC | 15:45 | |
| *** donnyd has quit IRC | 15:45 | |
| *** gixx has quit IRC | 15:45 | |
| *** masterpe has quit IRC | 15:45 | |
| *** rpittau has quit IRC | 15:45 | |
| *** Anticimex has quit IRC | 15:45 | |
| *** fridtjof[m] has joined #openstack-ansible | 15:45 | |
| *** soren has joined #openstack-ansible | 15:45 | |
| *** donnyd has joined #openstack-ansible | 15:45 | |
| *** gixx has joined #openstack-ansible | 15:45 | |
| *** masterpe has joined #openstack-ansible | 15:45 | |
| *** rpittau has joined #openstack-ansible | 15:45 | |
| *** Anticimex has joined #openstack-ansible | 15:45 | |
| *** fridtjof[m] has quit IRC | 15:48 | |
| *** masterpe has quit IRC | 15:48 | |
| kleini | okay, still the same issue with Keystone, that service catalog is empty | 15:54 |
| noonedeadpunk | can you post output? | 15:54 |
| kleini | will wipe my whole setup and start from scratch | 15:54 |
| kleini | http://paste.openstack.org/show/791345/ | 15:55 |
| noonedeadpunk | Hm, I guess this task is supposed to bootstrap keystone... https://opendev.org/openstack/openstack-ansible-os_keystone/src/branch/stable/stein/tasks/keystone_service_bootstrap.yml#L26 | 15:58 |
| noonedeadpunk | oh, run it with -e keystone_service_setup=true | 15:59 |
| noonedeadpunk | it = playbook | 15:59 |
| noonedeadpunk | kleini: ^ | 15:59 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Fix paging link protocol when behind haproxy https://review.opendev.org/706090 | 16:01 |
| jrosser | noonedeadpunk: i think rally is still failing on stretch - i wonder if we just disable it for anywhere < py3.6 | 16:06 |
| noonedeadpunk | I'd say it's almost passing | 16:06 |
| noonedeadpunk | https://review.opendev.org/#/c/715215/ | 16:07 |
| jrosser | http://paste.openstack.org/show/791346/ | 16:07 |
| noonedeadpunk | yeah, debian failed.... | 16:08 |
| noonedeadpunk | but centos passed.... | 16:09 |
| *** rpittau is now known as rpittau|afk | 16:12 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Bump SHAs for master https://review.opendev.org/714308 | 16:21 |
| openstackgerrit | Merged openstack/openstack-ansible stable/stein: Bump SHAs for stable/stein https://review.opendev.org/714309 | 16:21 |
| *** udesale has quit IRC | 16:29 | |
| *** evrardjp has quit IRC | 16:36 | |
| *** evrardjp has joined #openstack-ansible | 16:36 | |
| *** thuydang has joined #openstack-ansible | 16:39 | |
| *** Anticimex has quit IRC | 16:43 | |
| *** Anticimex has joined #openstack-ansible | 16:44 | |
| *** prometheanfire has quit IRC | 16:45 | |
| *** fridtjof[m] has joined #openstack-ansible | 16:48 | |
| noonedeadpunk | huh, debian uses https://opendev.org/openstack/rally-openstack/raw/branch/master/upper-constraints.txt | 16:48 |
| noonedeadpunk | so it's kinda py3??? | 16:49 |
| * noonedeadpunk deploys sandbox to see | 16:49 | |
| jrosser | 3.5 on stretch isnt it? | 16:49 |
| jrosser | which no doubt fails a bunch of stuff on openstack now needing >=3.6 | 16:49 |
| noonedeadpunk | yeah and osprofiler is one of it... | 16:50 |
| jrosser | as here https://pypi.org/project/rally/ | 16:50 |
| noonedeadpunk | we need to force stretch to use py2? | 16:50 |
| jrosser | and so on for lots of things, so i think just fixing one pin in global pins isn't going to work | 16:50 |
| noonedeadpunk | the thing is that venv is py2 | 16:51 |
| jrosser | isnt it the same issue as tempest, that it's branchless | 16:51 |
| noonedeadpunk | it's worse as don't follow u-c | 16:51 |
| jrosser | right | 16:51 |
| jrosser | i wonder what the situation on our stein jobs is | 16:52 |
| jrosser | that will show whats happening under py2.7 | 16:52 |
| noonedeadpunk | I think we probably don't include rally for stein? | 16:56 |
| noonedeadpunk | yeah, rally is not in setup-openstack.yml | 16:57 |
| jrosser | imho we should just disable the deployment for py < 3.6 | 16:59 |
| * jrosser has nightmares about all that tempest trouble | 17:00 | |
| *** DanyC has quit IRC | 17:08 | |
| *** DanyC has joined #openstack-ansible | 17:08 | |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_horizon master: Config options around federation URLs https://review.opendev.org/716026 | 17:11 |
| *** DanyC has quit IRC | 17:13 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add default value for user_role_file when cloning git repos https://review.opendev.org/715186 | 17:15 |
| *** prometheanfire has joined #openstack-ansible | 17:23 | |
| velmeran | I'm having an issue with a deployment, getting to the point where its checking for available compute nodes and not finding any: http://paste.openstack.org/show/791351/ | 17:31 |
| kleini | my problem remains the same. Keystone response results in the error message, that the service catalog is empty. | 17:32 |
| velmeran | I have a single compute node I'm targetting, its just a VM, so can't really do true virtualization, but I'm really just trying to test a simple deployment so don't need it to really do any vm work. | 17:32 |
| jrosser | velmeran: do you have placement deployed? | 17:32 |
| jrosser | there is a failure in the output you pasted (it’s an ascii table) | 17:33 |
| velmeran | not sure, I'm pretty new to all this. Is there a specific service I need to add to the openstack_user_config.yml file? | 17:34 |
| jrosser | which branch are you deploying | 17:34 |
| noonedeadpunk | kleini: have you ran keystone-setup playbook with keystone_service_setup=true? | 17:34 |
| noonedeadpunk | I guess this is kind of var that is cached somewhere | 17:35 |
| noonedeadpunk | either on deploy or controller nodes | 17:35 |
| kleini | I see admin user and endpoint in the database but Keystone does not return them. | 17:35 |
| kleini | I see admin user and Keystone endpoint in the database but Keystone does not return them. | 17:36 |
| noonedeadpunk | hm. and whats it logs? | 17:36 |
| noonedeadpunk | *in | 17:36 |
| jrosser | velmeran: there was an error in out documentation which was fixed with this patch https://review.opendev.org/#/c/702934/ for both master and stable/train | 17:37 |
| velmeran | I'm on the train branch | 17:37 |
| kleini | keystone runs as wsgi behind nginx. there are no keystone logs | 17:37 |
| noonedeadpunk | yes, but wsgi should log stuff to nginx | 17:38 |
| *** masterpe has joined #openstack-ansible | 17:38 | |
| *** openstacking_123 has joined #openstack-ansible | 17:39 | |
| velmeran | If I just grabbed the latest git yesterday, would I have that patch already? | 17:39 |
| noonedeadpunk | velmeran: here's the thing not about fix but about the way of configuration of openstack-user-config.yml | 17:40 |
| noonedeadpunk | so placement-infra_hosts should be defined there | 17:40 |
| *** andrewbonney has quit IRC | 17:41 | |
| noonedeadpunk | but yes, docs should contain the neccesity of it's defenition | 17:41 |
| velmeran | okay, and that should be targeted at the compute node, or just the shared-infra_host? | 17:42 |
| noonedeadpunk | for ifra hosts | 17:43 |
| kleini | http://paste.openstack.org/show/791354/ <- keystone has some very expressive warnings | 17:49 |
| velmeran | okay. If I add that in, so I need to restart the deploy at the setup host, setup infra, or setup openstack part? | 17:53 |
| *** thuydang has quit IRC | 17:55 | |
| noonedeadpunk | velmeran: you ned to run os-placement-install.yml actually | 17:57 |
| noonedeadpunk | or setup-openstack | 17:57 |
| noonedeadpunk | kleini: what OS are you deploying on? | 17:59 |
| kleini | ubuntu 18.04.4 LTS | 18:01 |
| noonedeadpunk | hm | 18:01 |
| noonedeadpunk | and osa release? (will try to spawn a sandbox) | 18:02 |
| kleini | install_method is source. maybe deploying keystone from scratch is broken somehow in stable/stein | 18:02 |
| mnaser | noonedeadpunk: have you looked into the train centos-7 failures by any chance? | 18:02 |
| kleini | stable/stein or 19.0.11 | 18:02 |
| noonedeadpunk | mnaser: yep, I did https://review.opendev.org/#/c/715215/ | 18:03 |
| noonedeadpunk | spawning stretch sandbox now | 18:03 |
| noonedeadpunk | to see what's wrong there | 18:03 |
| noonedeadpunk | (or we can eventyually do not deploy rally on py<3.6 as jrosser suggested) | 18:04 |
| * mnaser looks | 18:18 | |
| *** DanyC has joined #openstack-ansible | 18:27 | |
| *** DanyC has quit IRC | 18:32 | |
| *** spatel has joined #openstack-ansible | 18:38 | |
| spatel | ioni: hey, are you there? | 18:38 |
| noonedeadpunk | kleini: sandbox passed keystone without any issues | 18:59 |
| noonedeadpunk | and already deployed galnce anf cinder... | 18:59 |
| kleini | so maybe it is nspawn related or whatever | 19:00 |
| noonedeadpunk | I did nspawn sandbox | 19:01 |
| noonedeadpunk | on bionic and 19.0.11 | 19:01 |
| noonedeadpunk | so it's smth deployment specific.... | 19:01 |
| kleini | is it possibly related to an LDAP Users domain? | 19:01 |
| kleini | will check that configuration | 19:01 |
| noonedeadpunk | hm, that may be, as I was using just some base keystone | 19:02 |
| *** thuydang has joined #openstack-ansible | 19:05 | |
| *** DanyC has joined #openstack-ansible | 19:06 | |
| *** DanyC has quit IRC | 19:11 | |
| *** jbadiapa has quit IRC | 19:14 | |
| *** DanyC has joined #openstack-ansible | 19:38 | |
| kleini | Still the same problem without any special configured domains. | 19:38 |
| *** DanyC has quit IRC | 19:43 | |
| *** mugsie has quit IRC | 19:51 | |
| *** mugsie has joined #openstack-ansible | 19:54 | |
| *** kleini has quit IRC | 20:05 | |
| *** DanyC has joined #openstack-ansible | 20:07 | |
| *** sshnaidm is now known as sshnaidm|afk | 20:07 | |
| *** macz_ has quit IRC | 20:14 | |
| *** macz_ has joined #openstack-ansible | 20:27 | |
| *** gshippey has quit IRC | 20:52 | |
| velmeran | back again, after getting that placement thing fixed, I'm now getting this: ERROR: Could not find a version that satisfies the requirement rally===3.0.0 I'm on Centos7, also seeing a lot of pip needs to be updated messages during the setup-openstack script, would it break things if I update pip manually? | 21:15 |
| jrosser | velmeran: that is an outstanding bug currently | 21:17 |
| jrosser | comment out this line https://github.com/openstack/openstack-ansible/blob/master/playbooks/setup-openstack.yml#L48 | 21:18 |
| jrosser | you don't need rally for a working cloud | 21:18 |
| jrosser | on the question of pip versions that is locked down here https://github.com/openstack/openstack-ansible/blob/master/global-requirement-pins.txt | 21:19 |
| jrosser | new versions may contain surprised so i'd just leave that alone | 21:20 |
| jrosser | *surprises | 21:20 |
| velmeran | yea, that was always my experiance with pip, upgrade to fix one thing, break everything else... | 21:20 |
| jrosser | there is a carefully constructed set of virtualenvs that we try to make the same in CI as on your deployment | 21:21 |
| velmeran | okay, commented it out, lets see if I can get this going. | 21:21 |
| jrosser | its the last playbook - i think you are done :) | 21:21 |
| velmeran | those are some famous last words... | 21:21 |
| jrosser | what you've probably got now is a set of 'sensible defaults' | 21:22 |
| jrosser | things like external networking almost certainly need extra work/config as thats pretty much unique to each deployment | 21:23 |
| jrosser | i'm done for today, but if you get stuck just ask | 21:24 |
| *** spatel has quit IRC | 22:02 | |
| *** tosky has quit IRC | 22:20 | |
| *** DanyC has quit IRC | 22:37 | |
| *** DanyC has joined #openstack-ansible | 22:38 | |
| snadge | hmm.. my train cluster was working properly, until a few days ago when it ran out of space.. now the rabbit queue just randomly stops responding | 22:38 |
| *** DanyC has quit IRC | 22:42 | |
| snadge | but only to neutron.. i want to try and figure out why, but it might be faster to just blow everything away and start again | 22:44 |
| *** thuydang has quit IRC | 23:13 | |
| *** maharg101 has quit IRC | 23:17 | |
| *** maharg101 has joined #openstack-ansible | 23:21 | |
| *** macz_ has quit IRC | 23:45 | |
| *** rh-jlabarre has quit IRC | 23:53 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!