| openstackgerrit | Merged openstack/openstack-ansible master: Include package pins for OpenStack services https://review.openstack.org/647813 | 00:21 |
|---|---|---|
| *** cmart has joined #openstack-ansible | 00:45 | |
| *** markvoelker has joined #openstack-ansible | 00:51 | |
| *** markvoelker has quit IRC | 01:03 | |
| *** markvoelker has joined #openstack-ansible | 01:18 | |
| *** jbadiapa has quit IRC | 01:23 | |
| *** jbadiapa has joined #openstack-ansible | 01:23 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_sahara master: Update role for new source build process https://review.openstack.org/647850 | 01:45 |
| *** cmart has quit IRC | 01:45 | |
| *** ianychoi has quit IRC | 02:03 | |
| *** ianychoi has joined #openstack-ansible | 02:03 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_neutron master: Convert dynamic includes to static imports https://review.openstack.org/637026 | 02:07 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add the manila service to OSA https://review.openstack.org/645653 | 02:09 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add the manila service to OSA https://review.openstack.org/645653 | 02:09 |
| *** mgagne has quit IRC | 02:15 | |
| *** mgagne has joined #openstack-ansible | 02:15 | |
| *** charz has quit IRC | 02:16 | |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client master: Templatize rsyslog configuration files https://review.openstack.org/624805 | 02:20 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_magnum master: Update role for new source build process https://review.openstack.org/647979 | 02:24 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Allow user-defined extra distro packages https://review.openstack.org/594160 | 02:24 |
| openstackgerrit | Merged openstack/openstack-ansible-os_congress master: Update role for new source build process https://review.openstack.org/647930 | 02:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Update role for new source build process https://review.openstack.org/647830 | 02:45 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Convert dynamic includes to static imports https://review.openstack.org/647303 | 02:49 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Update role for new source build process https://review.openstack.org/647932 | 02:55 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Fix polling namespace on compute nodes https://review.openstack.org/607202 | 02:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: fix tox python3 overrides https://review.openstack.org/606756 | 02:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Update role for new source build process https://review.openstack.org/647889 | 03:03 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Switch to import_tasks when tags should be applied https://review.openstack.org/629139 | 03:03 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Set appropriate branch key in _horizon_translations_pull https://review.openstack.org/639978 | 03:04 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_manila master: Update manila so that it works with cephfs https://review.openstack.org/643616 | 03:11 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_manila master: Update manila so that it works with cephfs https://review.openstack.org/643616 | 03:13 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/rocky: Remove vif_plugin_is_fatal and vif_plugin_timeout values https://review.openstack.org/641458 | 03:17 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Update role for new source build process https://review.openstack.org/647838 | 03:34 |
| *** udesale has joined #openstack-ansible | 03:46 | |
| *** raukadah is now known as chandankumar | 04:02 | |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Add manila as a configured component https://review.openstack.org/645723 | 04:07 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Convert dynamic includes to static imports https://review.openstack.org/637026 | 04:43 |
| *** udesale has quit IRC | 04:44 | |
| *** udesale has joined #openstack-ansible | 04:46 | |
| *** shyam89 has joined #openstack-ansible | 04:58 | |
| *** shyam89 has quit IRC | 05:03 | |
| *** shyam89 has joined #openstack-ansible | 05:15 | |
| *** markvoelker has quit IRC | 06:00 | |
| *** markvoelker has joined #openstack-ansible | 06:31 | |
| *** miloa has joined #openstack-ansible | 06:56 | |
| miloa | Morning | 06:56 |
| *** SmearedBeard has joined #openstack-ansible | 06:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible master: Imported Translations from Zanata https://review.openstack.org/647662 | 07:00 |
| *** shyam89 has quit IRC | 07:00 | |
| *** shyam89 has joined #openstack-ansible | 07:04 | |
| *** SmearedBeard has quit IRC | 07:12 | |
| *** ppetit has quit IRC | 07:20 | |
| *** DanyC has joined #openstack-ansible | 07:20 | |
| *** cshen has joined #openstack-ansible | 07:20 | |
| *** shyam89 has quit IRC | 07:31 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Update role for new source build process https://review.openstack.org/648019 | 07:36 |
| *** kopecmartin|off is now known as kopecmartin | 07:43 | |
| *** ivve has joined #openstack-ansible | 07:47 | |
| *** shyam89 has joined #openstack-ansible | 07:48 | |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Update role for new source build process https://review.openstack.org/648026 | 07:58 |
| *** luksky has joined #openstack-ansible | 08:03 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_gnocchi master: Update role for new source build process https://review.openstack.org/648029 | 08:09 |
| *** DanyC has quit IRC | 08:11 | |
| *** shyam89 has quit IRC | 08:16 | |
| *** tosky has joined #openstack-ansible | 08:17 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Update role for new source build process https://review.openstack.org/648019 | 08:18 |
| *** hamzaachi has joined #openstack-ansible | 08:26 | |
| *** gillesMo has joined #openstack-ansible | 08:32 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Update role for new source build process https://review.openstack.org/648036 | 08:46 |
| CeeMac | morning | 08:48 |
| *** zhongjun2 has joined #openstack-ansible | 08:49 | |
| *** shardy has joined #openstack-ansible | 08:51 | |
| zhongjun2 | Hi, anyone familiar with this issue https://github.com/ansible/ansible/issues/27423? | 08:51 |
| *** cyberpear has joined #openstack-ansible | 08:54 | |
| *** pcaruana has joined #openstack-ansible | 08:57 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Added Openstack Masakari role for the deployment https://review.openstack.org/645186 | 09:05 |
| *** shyam89 has joined #openstack-ansible | 09:12 | |
| *** priteau has joined #openstack-ansible | 09:15 | |
| *** DanyC has joined #openstack-ansible | 09:16 | |
| *** DanyC_ has joined #openstack-ansible | 09:19 | |
| *** DanyC has quit IRC | 09:20 | |
| *** Smeared_Beard has joined #openstack-ansible | 09:26 | |
| admin0 | i am getting apparomor related issues in hypervisors .. is it safe to disable apparmor ? | 09:49 |
| odyssey4me | o/ | 09:51 |
| admin0 | hi | 09:57 |
| *** shyam89 has quit IRC | 10:08 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add panko to integrated build https://review.openstack.org/647898 | 10:09 |
| admin0 | if a hypervisor apt upgrades qemu etc, do we need to run the playbooks again ? | 10:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add telemetry service bundle to AIO tests https://review.openstack.org/647921 | 10:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add telemetry service bundle to AIO tests https://review.openstack.org/647921 | 10:09 |
| *** shyam89 has joined #openstack-ansible | 10:21 | |
| *** rgogunskiy has joined #openstack-ansible | 10:24 | |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Update role for new source build process https://review.openstack.org/648026 | 10:26 |
| *** rgogunskiy has quit IRC | 10:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 10:35 |
| odyssey4me | admin0 you might need to, yes - because the libvirt python lib location may have changed | 10:36 |
| *** electrofelix has joined #openstack-ansible | 10:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 10:43 |
| *** aedc has joined #openstack-ansible | 10:46 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Add option to disable container builds on specific hosts https://review.openstack.org/647884 | 10:47 |
| openstackgerrit | Merged openstack/openstack-ansible master: Imported Translations from Zanata https://review.openstack.org/647662 | 10:47 |
| *** shyam89 has quit IRC | 10:53 | |
| *** hughsaunders has quit IRC | 11:03 | |
| noonedeadpunk | morning folks | 11:07 |
| noonedeadpunk | i think I have to look for trailing whitespace linter for my IDE.... | 11:09 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_gnocchi master: Update role for new source build process https://review.openstack.org/648029 | 11:10 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Update role for new source build process https://review.openstack.org/648026 | 11:14 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Update role for new source build process https://review.openstack.org/648036 | 11:17 |
| *** cshen has quit IRC | 11:18 | |
| *** aedc has quit IRC | 11:19 | |
| cyberpear | any ideas on how to pass the equivalent of 'validate_certs: no' to the ansible openstack inventory plugin? | 11:23 |
| odyssey4me | cyberpear in clouds.yaml: https://docs.openstack.org/os-client-config/latest/user/configuration.html#ssl-settings | 11:27 |
| odyssey4me | cyberpear an example: https://gist.github.com/odyssey4me/fd340be50301a8b7d6eca46f65866caa#file-clouds-yaml-L28-L40 | 11:28 |
| *** shyam89 has joined #openstack-ansible | 11:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 11:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 11:35 |
| noonedeadpunk | odyssey4me: check commit message in https://review.openstack.org/648073 | 11:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 11:36 |
| odyssey4me | noonedeadpunk good catch, thanks | 11:36 |
| noonedeadpunk | hm, how to reset drafts in gerrit now?:) | 11:37 |
| noonedeadpunk | ok, needed to descend to previous patch) | 11:39 |
| odyssey4me | noonedeadpunk ceilometer no longer needs a service in the keystone service catalog, right? | 11:43 |
| odyssey4me | noonedeadpunk well, I see that https://github.com/openstack/openstack-ansible-os_ceilometer/blob/master/tasks/ceilometer_service_setup.yml sets up a service user and role - are those still needed? | 11:44 |
| noonedeadpunk | I think it doesn't. | 11:45 |
| noonedeadpunk | Will check it and patch then | 11:45 |
| odyssey4me | noonedeadpunk thanks - http://logs.openstack.org/51/647851/2/check/openstack-ansible-deploy-aio_telemetry_metal-centos-7/2116371/job-output.txt.gz#_2019-03-26_22_35_15_451564 is the failure I'm hitting in https://review.openstack.org/647851, which lead me to ask that question | 11:47 |
| *** cshen has joined #openstack-ansible | 11:48 | |
| noonedeadpunk | inventory_hostname == groups['ceilometer_central_container'][0]' might be replaced by run_once, but as ceilometer_service_setup.yml is not required at all, so it'll be better to get rid of it | 11:49 |
| odyssey4me | noonedeadpunk yep | 11:49 |
| odyssey4me | I dunno if run_once will auto select the first play host though? | 11:50 |
| odyssey4me | but anyway, if it's not required - better to be rid of those tasks | 11:50 |
| noonedeadpunk | not sure if that matters in this context | 11:50 |
| odyssey4me | 'This directive forces the task to attempt execution on the first host in the current batch and then applies all results and facts to all the hosts in the same batch.' | 11:51 |
| odyssey4me | Well, that's good - that means that a lot of our 'if this is the first host, then execute' conditionals can be changed to run_once | 11:51 |
| noonedeadpunk | I mean, that in general _service_setup.yml are delegated, so we just need to run this once, and don't really care wether it was delegated from first or second host | 11:51 |
| odyssey4me | well, it does matter if there's any serialisation or any dependency for the service because that thing must be done before a service comes up, otherwise the service fails | 11:52 |
| noonedeadpunk | yeah, in case of serialisation it does... | 11:52 |
| odyssey4me | although for serialisation: 'When used together with “serial”, tasks marked as “run_once” will be run on one host in each serial batch. If it’s crucial that the task is run only once regardless of “serial” mode, use when: inventory_hostname == ansible_play_hosts[0] construct.' | 11:53 |
| noonedeadpunk | odyssey4me service_credentials are required and seems that used for pollsters to ask cinder, nova, glace, etc about some info | 12:00 |
| odyssey4me | noonedeadpunk ah ok, so those things are needed - fair enough, so we switch to run_once then :) | 12:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_masakari master: Test using integrated build https://review.openstack.org/647571 | 12:01 |
| noonedeadpunk | but [keystone_authtoken] in template should not even exists I think | 12:01 |
| noonedeadpunk | https://github.com/openstack/openstack-ansible-os_ceilometer/blob/master/templates/ceilometer.conf.j2#L114 | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Test using integrated build https://review.openstack.org/648104 | 12:01 |
| noonedeadpunk | at least since queens... | 12:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Test using integrated build https://review.openstack.org/648104 | 12:04 |
| odyssey4me | noonedeadpunk I trust that you'll do the right thing - I'm not really aware of all that, but do recall there being some discussion around that recently. I think it was on the ML? | 12:05 |
| noonedeadpunk | I guess I've missed it... | 12:06 |
| noonedeadpunk | will try to find it then | 12:07 |
| *** strattao has joined #openstack-ansible | 12:09 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_blazar master: Test using integrated build https://review.openstack.org/648106 | 12:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_blazar master: Test using integrated build https://review.openstack.org/648106 | 12:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_blazar master: Test using integrated build https://review.openstack.org/648106 | 12:11 |
| noonedeadpunk | I guess, that keystone_authtoken section in config was required by gnocchi at some step, up to the point it became a separate project... | 12:14 |
| noonedeadpunk | I can't find any mentioning of that section in ceilometer itself since ocata | 12:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Test using integrated build https://review.openstack.org/648108 | 12:15 |
| *** markvoelker has quit IRC | 12:21 | |
| *** strattao has quit IRC | 12:32 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 12:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Update role for new source build process https://review.openstack.org/647979 | 12:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_panko master: Remove extras https://review.openstack.org/648112 | 12:34 |
| *** shyam89 has quit IRC | 12:34 | |
| *** gillesMo has left #openstack-ansible | 12:35 | |
| *** pcaruana has quit IRC | 12:39 | |
| *** gillesMo has joined #openstack-ansible | 12:39 | |
| *** pcaruana has joined #openstack-ansible | 12:42 | |
| *** pcaruana has quit IRC | 12:42 | |
| *** pcaruana has joined #openstack-ansible | 12:43 | |
| chandankumar | odyssey4me: Hello | 12:45 |
| chandankumar | odyssey4me: is it possible to override specific key value here https://github.com/openstack/openstack-ansible-os_tempest/blob/master/defaults/main.yml#L333 ? | 12:45 |
| *** shyam89 has joined #openstack-ansible | 12:46 | |
| *** udesale has quit IRC | 12:48 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Cleanup config from keystone_authtoken section https://review.openstack.org/648115 | 12:49 |
| *** udesale has joined #openstack-ansible | 12:49 | |
| noonedeadpunk | odyssey4me I think that should work ^ | 12:50 |
| *** gillesMo has quit IRC | 12:56 | |
| *** jhesketh has quit IRC | 13:10 | |
| openstackgerrit | Martin Kopec proposed openstack/openstack-ansible-os_tempest master: User guide https://review.openstack.org/644570 | 13:11 |
| openstackgerrit | Martin Kopec proposed openstack/openstack-ansible-os_tempest master: User guide https://review.openstack.org/644570 | 13:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 13:12 |
| *** aedc has joined #openstack-ansible | 13:13 | |
| odyssey4me | chandankumar if you use config_template, sure | 13:13 |
| odyssey4me | chandankumar erm, no - that won't work... but using a combine filter might work if you provide for another var to use... I doubt it though, that's one of the issues with using dicts | 13:13 |
| mnaser | good morning | 13:19 |
| * mnaser came into the office with a dead mouse battery so is sitting on timeout | 13:20 | |
| mnaser | silly upside down charging mouses | 13:20 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_sahara master: Update role for new source build process https://review.openstack.org/647850 | 13:20 |
| chandankumar | odyssey4me: may be I thinking to reuse config template and add one more task to reuse the value and convert it | 13:21 |
| chandankumar | https://github.com/openstack/openstack-ansible-os_tempest/blob/master/defaults/main.yml#L300 | 13:21 |
| *** gillesMo has joined #openstack-ansible | 13:22 | |
| chandankumar | let me propose a patch | 13:22 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Update role for new source build process https://review.openstack.org/648036 | 13:24 |
| *** shyam89 has quit IRC | 13:27 | |
| *** aedc has quit IRC | 13:27 | |
| *** shyam89 has joined #openstack-ansible | 13:27 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Update role for new source build process https://review.openstack.org/647833 | 13:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update the congress service egg name to openstack_congress https://review.openstack.org/648122 | 13:30 |
| *** shyam89 has quit IRC | 13:31 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_magnum master: cleanup: move venv install into main to avoid extra include https://review.openstack.org/648124 | 13:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Update the congress service egg name to openstack_congress https://review.openstack.org/648125 | 13:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Update the congress service egg name to openstack_congress https://review.openstack.org/648125 | 13:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Test using integrated build https://review.openstack.org/648108 | 13:33 |
| *** shyam89 has joined #openstack-ansible | 13:33 | |
| noonedeadpunk | mnaser: regarding https://review.openstack.org/#/c/648029 - I was going to try to work on gnocchi installation from packages. So then gnocchi_install.yml will be utilised again. | 13:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_mistral master: Update role for new source build process https://review.openstack.org/647841 | 13:38 |
| noonedeadpunk | So is there sense in cleanup now to return everything back in the future? | 13:38 |
| noonedeadpunk | and it's kinda unification, that all roles have _install.yml | 13:39 |
| noonedeadpunk | in terms of following code it might make sense | 13:39 |
| *** priteau has quit IRC | 13:44 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Update role for new source build process https://review.openstack.org/648036 | 13:46 |
| zhongjun2 | Hi, anyone familiar with this issue https://github.com/ansible/ansible/issues/27423? | 13:46 |
| zhongjun2 | is this still an issue？ | 13:47 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_gnocchi master: Update role for new source build process https://review.openstack.org/648029 | 13:48 |
| zhongjun2 | It fixed in shade to solve the performance issue. https://bugs.launchpad.net/shade/+bug/1709577 Support to get resource by id to avoid finding in list loop | 13:48 |
| openstack | Launchpad bug 1709577 in Shade "Support to get resource by id to avoid finding in list loop" [Undecided,Fix committed] - Assigned to Rui Chen (kiwik-chenrui) | 13:48 |
| zhongjun2 | But we use openstacksdk instead of shade. I am not sure whether it resolved in openstacksdk? | 13:54 |
| *** jeh has joined #openstack-ansible | 13:59 | |
| *** shananigans has joined #openstack-ansible | 13:59 | |
| admin0 | CeeMac, ovs is working fine .. one is in 16.04 and one is in 18.04 | 14:04 |
| CeeMac | admin0: I was wondering how you're deploying openvswitch from host network backing perspective in bionic? | 14:04 |
| CeeMac | fall back on ifupdown or have you got it working with netplan or networkd? | 14:04 |
| admin0 | the netplan does not work, so fallback to ifupdown | 14:04 |
| CeeMac | have you tried plan networkd at all? | 14:05 |
| CeeMac | *plain | 14:05 |
| CeeMac | admin0: are you running HA DVR at all? | 14:07 |
| *** shyam89 has quit IRC | 14:12 | |
| *** spatel has joined #openstack-ansible | 14:13 | |
| spatel | Morning folks!!! | 14:13 |
| spatel | Hope everyone doing great! | 14:14 |
| admin0 | no dvr | 14:15 |
| cloudnull | mornings all | 14:15 |
| admin0 | the way I do is make each compute also a network host | 14:15 |
| admin0 | that way, there is no few points of failure in the system | 14:15 |
| admin0 | and the networking stuff is distributed | 14:15 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Update role for new source build process https://review.openstack.org/648019 | 14:16 |
| admin0 | so 20 compute nodes are also 20 network nodes .. 2 nodes are disabled as compute and just used as network .. so that in case a compute node dies, we have some reserved capacity to boot up the instances faster | 14:16 |
| CeeMac | that makes sense. Are you running the legacy routers in HA then? | 14:18 |
| CeeMac | hi spatel / cloudnull | 14:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 14:21 |
| admin0 | CeeMac, before i ran on HA, but i found them in split brain ( the public ip is in one router, and the private gateway is in another router ) , sometimes all are active, sometimes 2 are masters, sometimes all are slaves | 14:22 |
| admin0 | do i disabled HA | 14:23 |
| admin0 | it takes just a few seconds to move them from one node to another, so its not that noticable | 14:23 |
| CeeMac | split brain isn't great like. | 14:23 |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Add Leap 15 jobs https://review.openstack.org/647479 | 14:24 |
| CeeMac | do the routers automatically spin up on another node if a network/compute node fails, or is that something you have in a kill/spawn script? | 14:24 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Reuse tempest_tempest_conf_overrides in tempestconf override https://review.openstack.org/648137 | 14:26 |
| *** mmercer has quit IRC | 14:27 | |
| chandankumar | odyssey4me: ^^ something like this, is it possible to call combine filter only when extra overrides and enable_tempestcinf is true and how can I test in tempestocnf job itslef? | 14:27 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 14:28 |
| odyssey4me | chandankumar the path you're taking makes debugging obscure - sometimes it's better just to not do things like this, even if you can make it work | 14:29 |
| spatel | CeeMac: DVR is beast it in self.. its hard to troubleshoot sometime.. i have played in past and it need serious planning. | 14:29 |
| odyssey4me | what's the issue with just providing the whole dict override in the first place? | 14:29 |
| odyssey4me | lol, mnaser guilhermesp jrosser noonedeadpunk now I understand why no-one took up the changes for the nova and neutron roles... | 14:30 |
| odyssey4me | they're... complicated | 14:30 |
| CeeMac | spatel: oh? | 14:30 |
| CeeMac | hmm | 14:30 |
| noonedeadpunk | :D | 14:31 |
| CeeMac | i'm in a bit of a bind, as I think our use case will call for them, and I don't fancy the idea of a rip and replace down the line | 14:31 |
| spatel | CeeMac: This is 4 year ago stuff now it may improved, also it ate up lots of public IPs because every DVR need own public IP for SNAT traffic. | 14:32 |
| CeeMac | but at the same time, I'm not keen on introducing potential instabilty at the get go | 14:32 |
| CeeMac | yeah, that was one note of cautin i saw, regarding the PIPs | 14:32 |
| CeeMac | admin0: was there a particular reason you went with ovs over lxb if you're not using DVR? | 14:33 |
| noonedeadpunk | I just didn't feel really confident about them, so decided to leave to someone more experieced, and it seemed to be the right decision | 14:33 |
| spatel | Now i am using VLAN base networking where my Core router is gateway for all instances.. | 14:33 |
| CeeMac | as in your physical core router? | 14:33 |
| spatel | Yes my Cisco Nexus 9000 L3 switch | 14:34 |
| CeeMac | i'm going to be running vxlan and I'm note sure I have my core fully kitted up for VTEP to facilitate that at this point | 14:34 |
| CeeMac | plenty of grunt in there then :D | 14:34 |
| CeeMac | so many decisions! | 14:35 |
| spatel | Openstack networking is really serious topic and planning, it all depend on what you trying to do. | 14:35 |
| odyssey4me | has anyone around used barbican? - using the integrated build test results in a fairly general failure: http://logs.openstack.org/04/648104/2/check/openstack-ansible-deploy-aio_metal_barbican-ubuntu-bionic/c13f8e0/logs/ara-report/result/b6372a9a-d9c5-40d0-98dd-3d5f8011155c/ | 14:36 |
| CeeMac | as is the case with networking in most cases :D | 14:36 |
| CeeMac | the goal is to have full virtualisation of the tenant environment | 14:36 |
| spatel | I learnt lots of lesson and finally now i am in good spot | 14:36 |
| CeeMac | we already run a vlan based multi-tenant environment with VMware | 14:37 |
| CeeMac | NSX is a deal breaker from a cost perspective | 14:37 |
| CeeMac | so openstack | 14:37 |
| CeeMac | vxlan is the preference to avoid vlan exhaustion | 14:37 |
| spatel | we lost many thousands $$ because of poor network performance and breaching SLA of customers :( | 14:37 |
| CeeMac | although I have the capacity now to do service provider bridging, I don't want to mix the context on the PE routers to that extent | 14:38 |
| CeeMac | i feel your pain :( | 14:38 |
| CeeMac | its always the networks fault :( | 14:38 |
| noonedeadpunk | Or not really right - then I'd know much more about these role caveats | 14:38 |
| CeeMac | but yeah, virtual routers would be the preference as the intention for this cloud is to be fully self service | 14:38 |
| spatel | agreed++ | 14:39 |
| odyssey4me | noonedeadpunk mostly we just make it up as we go along ;) | 14:39 |
| CeeMac | I'm running Juniper in the core/edge so the plan ultimately is to use their plugins to pass the Router and FW creation to VRs on the physical kit | 14:39 |
| spatel | in my case we are telco company and high latency has zero room :( | 14:39 |
| CeeMac | just not there yet as I need to migrate L2 to EVPN across the DCIs and WAN to get away from MSTP | 14:40 |
| CeeMac | for real | 14:40 |
| CeeMac | then i can repurpose my core fabric for VTEP | 14:40 |
| spatel | Yeah! there are lot to explore in network layer for openstack | 14:40 |
| CeeMac | i'm loving that aspect | 14:41 |
| CeeMac | unfortunately, the downside to such available variety is the amount of options/questions it raises lol | 14:41 |
| CeeMac | I've got LXB basic routers in RC. Was planning on at least HA routers in prod | 14:41 |
| odyssey4me | CeeMac with great power comes great responsibility ;) | 14:41 |
| CeeMac | haha odyssey4me: indeed, as a great man once said :) | 14:42 |
| CeeMac | however, if HA routes are going to split brain, thats not going to end well :/ | 14:42 |
| CeeMac | iirc you can't instance migrate between LXB and OVS tenant networks right? | 14:43 |
| CeeMac | without downing the VM, removing its NIC, giving it new NIC, bringing it up again at the least? | 14:43 |
| spatel | CeeMac: I believe you can't migrate LXB to OVS without downtime ( I would tag jamesdenton on this ) | 14:45 |
| *** priteau has joined #openstack-ansible | 14:45 | |
| CeeMac | spatel: i'm pretty sure I had this conversation with jamesdenton already, and the general feeling was rip and replace | 14:46 |
| CeeMac | nuke/rebuild from a network perspective | 14:46 |
| spatel | I am in same boat :) | 14:46 |
| CeeMac | which is why I find myself in a quandry | 14:46 |
| spatel | I am running LinuxBridge and at some point i need OVS | 14:46 |
| CeeMac | but then if ovs isn't giving any particular benefit at this point in time, its probablt not worth the pain? lol | 14:47 |
| spatel | LinuxBridge are so simple to troubleshoot but i need advance feature in near future so i have to go with OVS | 14:47 |
| CeeMac | saying that, I'm looking at a BackupaaS product that also can perform cloud migrations, so it might be a case of just building a new cloud and lift/shift if it comes to it | 14:47 |
| spatel | My reason to use OVS is to use DPDK | 14:48 |
| CeeMac | right, yeah, i was going to ask | 14:48 |
| spatel | Otherwise i am happy with LinuxBridge :) | 14:48 |
| CeeMac | yeah, with how you have it deployed that makes sense | 14:49 |
| CeeMac | might need to get a few more ppls feedback on HA/VRRP routers before I decide to go down that route | 14:49 |
| CeeMac | jamesdenton/jrosser may weigh in at some point :) | 14:50 |
| spatel | Right now i am running all my compute with SR-IOV networking but i am loosing live migration and bonding redendency :( | 14:50 |
| CeeMac | no live migration or bonding? | 14:50 |
| CeeMac | is that a symptom of the SR-IOV or legacy config? | 14:50 |
| spatel | SR-IOV doesn't support that feature at present | 14:50 |
| CeeMac | I just got live-migration working in my RC | 14:50 |
| spatel | DPDK support Bonding / Live migration | 14:50 |
| CeeMac | oh, thats good to know | 14:50 |
| CeeMac | aha, hence your requirement :) | 14:51 |
| spatel | DPDK required OVS | 14:51 |
| CeeMac | indeed | 14:51 |
| CeeMac | and we're back full circle :D | 14:51 |
| CeeMac | its like I'm speaking to a future self, lol | 14:51 |
| brtknr | Any feedback on this: https://github.com/ansible/ansible/pull/53757/files | 14:52 |
| CeeMac | hmm | 14:52 |
| CeeMac | so, if i go into prod with ovs, but legacy routers | 14:52 |
| brtknr | Is this even the right place to ask about this: https://github.com/ansible/ansible/pull/53757/files? | 14:52 |
| CeeMac | theres nothing to stop a tenant adding a DVR to the network should they wish once I've enabled that functionality? | 14:52 |
| spatel | LXB support DVR | 14:53 |
| CeeMac | brb | 14:53 |
| CeeMac | it does? | 14:53 |
| CeeMac | I thought DVR had ovs dependancy? | 14:53 |
| spatel | I believe you can do with LinuxBridge | 14:56 |
| spatel | let me check hold on | 14:56 |
| mgariepy | not from this table: https://docs.openstack.org/neutron/rocky/feature_classification/general_feature_support_matrix.html | 14:57 |
| admin0 | CeeMac, i have lxb in some and ovs in some .. the ones having ovs, they also have some epa/numa/nfv workload | 14:57 |
| admin0 | waiting for dpdk support via osa ( hint: less hacks, more like enable_dpkd: true ) type of setting .. | 14:59 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add the manila service to OSA https://review.openstack.org/645653 | 14:59 |
| spatel | CeeMac: I was wrong here - DVR required OVS ( doesn't work with LXB ) | 15:01 |
| CeeMac | admin0: seems like a sensible plan | 15:01 |
| CeeMac | spatel: thanks for confirming | 15:01 |
| spatel | admin0: i like your dpdk support idea... ha..ha.. | 15:02 |
| * admin0 plans to send jamesdenton pizza when he fixes this :D | 15:02 | |
| CeeMac | admin0: you were saying earlier that it only took a couple of seconds to move if a node fails, is that an automated recovery or something you've scripted? | 15:02 |
| admin0 | scripted .. u can shift dhcp and routers around | 15:03 |
| CeeMac | ah, ok | 15:03 |
| CeeMac | I'll need to make a note of that, i need to look at scripting VM recovery on host failure etc | 15:03 |
| admin0 | CeeMac, https://www.openstackfaq.com/openstack-migrate-routers-and-dhcp/ | 15:04 |
| CeeMac | I don't suppose you've got a blog already? | 15:04 |
| CeeMac | oh | 15:04 |
| CeeMac | that helps, thanks! | 15:04 |
| spatel | CeeMac: you should check out jamesdenton new networking book, worth looking at - LEARNING OPENSTACK NETWORKING - THIRD EDITION | 15:04 |
| admin0 | basically if a node dies that had your router, you just remove from the old agent and add in the new agent | 15:04 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add the manila service to OSA https://review.openstack.org/645653 | 15:04 |
| admin0 | osa will recreate the router there | 15:04 |
| admin0 | and all back to normal | 15:04 |
| odyssey4me | admin0 are you using https://github.com/openstack/openstack-ansible-os_neutron/blob/master/defaults/main.yml#L403-L408 for that, or something else? | 15:04 |
| admin0 | odyssey4me, i did not know of this : | 15:05 |
| admin0 | mine was ping router, if down: bash commands | 15:06 |
| odyssey4me | admin0 you can see the source that gets laid down here: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/neutron-ha-tool.py.j2 | 15:06 |
| CeeMac | odyssey4me, is that an automated recovery for non-HA routers? | 15:07 |
| spatel | CeeMac: one more advantage of OVS, it support sflow :) | 15:07 |
| odyssey4me | CeeMac yep, it gets run on cron once per minute: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tasks/neutron_l3_ha.yml#L54 | 15:07 |
| CeeMac | spatel: yeah that was on my list of "good" features | 15:07 |
| CeeMac | odyssey4me, awesome. thanks :) | 15:08 |
| odyssey4me | CeeMac depending on your SLA requirements, it works OK... it's not perfect, but it does the job | 15:08 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Cleanup config from keystone_authtoken section https://review.openstack.org/648115 | 15:08 |
| CeeMac | yeah, I'll need to have an internal discussion with the project team / SMT around SLAs but this definately helps | 15:08 |
| CeeMac | admin0: do you have your process wrapped up in your monitoring tools, or its a standalone watcher? | 15:09 |
| admin0 | CeeMac, with what odyssey4me just pointed out, this throws away the need to have monitoring tool or the script | 15:11 |
| admin0 | it does it itself via cron | 15:11 |
| CeeMac | indeed, was just curious :D | 15:12 |
| CeeMac | think I'll have a play with OVS and see how it feels from a build out perspective, then at least I can add features easier if/when required | 15:14 |
| CeeMac | there isn't any impact on the neutron XaaS plugins for OVS vs LXB they're supported/compatible with both? | 15:15 |
| CeeMac | specifically FW and VPN, but also LB | 15:15 |
| mnaser | calls all morning | 15:15 |
| * mnaser wants to hide | 15:15 | |
| mnaser | :P | 15:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 15:16 |
| noonedeadpunk | odyssey4me: placed update for https://review.openstack.org/648115 - I don't see really better option that intersect to be honest | 15:16 |
| mnaser | odyssey4me: im wondering if we did _something_ that is causing centos to slow down so much more | 15:17 |
| *** shardy has quit IRC | 15:17 | |
| chandankumar | odyssey4me: ah you are correct, I can directly use the dict | 15:18 |
| odyssey4me | noonedeadpunk commented | 15:21 |
| chandankumar | odyssey4me: https://review.openstack.org/#/c/648026/ please have a look at this changes | 15:23 |
| odyssey4me | mnaser I'm not sure. It is odd. | 15:24 |
| mnaser | odyssey4me: im comparing runs, looks like setup-hosts takes at least 10 minutes more, so maybe I can find something in common there | 15:24 |
| mnaser | 2m33s to create container on lentos, 56s on ubuntu | 15:25 |
| odyssey4me | mnaser are you ready to ditch containers for centos/suse yet? ;) | 15:25 |
| mnaser | odyssey4me: tbh, I'd love to.. but I think it's a bit too late to do that right now. I know we're internally working at dropping that though | 15:26 |
| mnaser | though, I doubt we'll be running stein on containers based on our stuff | 15:26 |
| mnaser | which means we totally could drop support but I think there is people like spatel who rely on it | 15:26 |
| mnaser | cache prep took 2 minutes I wonder if we're pulling in a lot of packages and making our containers much bigger | 15:27 |
| admin0 | what is broken in neutron to get this on ip netns Error: Peer netns reference is invalid. | 15:29 |
| admin0 | if anyone has seen it before and knows what the fix is | 15:29 |
| mnaser | oh ubuntu uses eatmydata | 15:30 |
| mnaser | in its cache prep | 15:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 15:31 |
| mnaser | ok setup-openstack took 51 minutes on ubuntu, 1h17m n centos | 15:32 |
| spatel | mnaser: sorry i am not following you guys, ditch containers for centos/suse ??? | 15:32 |
| mnaser | spatel: ya, for deploying on Baremetal only on centos | 15:34 |
| spatel | But why? | 15:35 |
| *** udesale has quit IRC | 15:35 | |
| mnaser | because they are slow, hack-y on centos and hard to maintain | 15:35 |
| mnaser | hacky because centos natively doesn't have lxc containers | 15:35 |
| mnaser | so we're using some copr build | 15:35 |
| odyssey4me | I think you mean 'slow to deploy' | 15:35 |
| mnaser | correct yes | 15:36 |
| spatel | But its ok for "slow to deploy" right ? we are not going to build cloud every single day | 15:36 |
| mnaser | in our CI we do :) | 15:36 |
| mnaser | and the centos job takes 2h20 minutes now .. for who knows what reason it changed | 15:36 |
| mnaser | and I have to sit and debug it because our gates are broken and developers can't merge code | 15:37 |
| mnaser | it looks like every operation is taking twice as long | 15:37 |
| spatel | If we going to drop support then i have no choice and need to find alternative solution :( | 15:38 |
| spatel | we are 100% CentOS shop :( | 15:38 |
| mnaser | no, we are not dropping centos support | 15:38 |
| mnaser | it would be dropping centos support for deploying using lxc containers | 15:38 |
| mnaser | you can still deploy, directly on the baremetal | 15:38 |
| odyssey4me | spatel I think the discussion at this point is to make it opt-in, rather than opt-out... and the plan has always been to switch to nspawn instead of LXC anyway... but we're struggling to get traction to finish that work | 15:38 |
| mnaser | (again, this is being discussed) | 15:39 |
| spatel | Ah! that make sense | 15:39 |
| mnaser | yeah, and that too | 15:39 |
| noonedeadpunk | odyssey4me: I've tested in in sandbox by adding tag ceilometer-service-setup to the ceilometer_service_setup.yml include. Got the following output. So seems, like run_once worked. http://paste.openstack.org/show/748486/ | 15:41 |
| odyssey4me | hmm, ceilometer is being built, whether there's configuration for it or not | 15:41 |
| spatel | mnaser: yes i can deploy on baremetal but i will loose good feature like snapshot or quick upgrade etc.. :( | 15:42 |
| mnaser | spatel: not sure what you mean by snapchat or quick upgrade | 15:43 |
| odyssey4me | noonedeadpunk ok, one more nit there - if you could fix that I think we're good to go | 15:43 |
| noonedeadpunk | but I'm not sure how it may allocate to another host, while task is included only for some host group. It may be situation, when host group has changed during runtime, but not sure that it won't cause other problems | 15:43 |
| spatel | mnaser: if i deploy every openstack component on baremetal without container in that case. | 15:44 |
| mnaser | upgrade would be faster on centos =P | 15:44 |
| mnaser | anything would be faster in that case to be honest | 15:44 |
| spatel | hmm! | 15:45 |
| spatel | after 1 year i am so used to with container, can't think of that.. i can sure give it a try on baremetal :) | 15:45 |
| odyssey4me | spatel *if* it drops, it won't happen in a hurry - the primary discussion for us right now is what we test in CI | 15:46 |
| odyssey4me | CI has pretty strict restrictions | 15:47 |
| odyssey4me | mnaser we could perhaps increase the timeout, just to get us over the hump until we can clean up the legacy build mechanism | 15:47 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Cleanup config from keystone_authtoken section https://review.openstack.org/648115 | 15:47 |
| mnaser | odyssey4me: I think that.. or move it to non voting, I *personally* feel ok with metal jobs reporting state of integration | 15:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Test using integrated build https://review.openstack.org/647851 | 15:49 |
| odyssey4me | mnaser so after https://review.openstack.org/647813 merged with a <2hr build time, the only patches that have merged which affect a standard AIO directly are https://review.openstack.org/647303, https://review.openstack.org/647889, https://review.openstack.org/629139, https://review.openstack.org/639978, https://review.openstack.org/647838, https://review.openstack.org/637026 | 15:54 |
| *** cshen has quit IRC | 15:56 | |
| mnaser | odyssey4me: I'm wondering if its cloud provider specific, looks like some of those happen with ovh and there seems to be issued there | 15:56 |
| *** Smeared_Beard has quit IRC | 15:58 | |
| *** SmearedBeard has joined #openstack-ansible | 16:00 | |
| *** KeithMnemonic has quit IRC | 16:01 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Extend AIO build timeout to 2.5hrs https://review.openstack.org/648167 | 16:02 |
| *** KeithMnemonic has joined #openstack-ansible | 16:04 | |
| odyssey4me | mnaser the only AIO patch that's not yet up for bp/python-build-simplification is the one for keystone, which you allocated to yourself... can you get it done quickly - the sooner we can get it merged, the sooner I can start cleaning up the old mechanism | 16:04 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Reuse tempest_tempest_conf_overrides in tempestconf extra args https://review.openstack.org/648137 | 16:05 |
| * chandankumar needs some help on python-build-simplicaiton os_tempest patch https://review.openstack.org/#/c/648026/ when free please have a look! | 16:07 | |
| chandankumar | it is leading to test failure on centos source jobs | 16:07 |
| chandankumar | might be related to my patch | 16:08 |
| *** cmart has joined #openstack-ansible | 16:09 | |
| odyssey4me | chandankumar I've just been looking at it | 16:10 |
| chandankumar | odyssey4me: can we store the pip list stuff at one place in logs? | 16:11 |
| chandankumar | for each venv in a job? | 16:12 |
| odyssey4me | chandankumar it's already there: http://logs.openstack.org/26/648026/3/check/openstack-ansible-functional-centos-7/80ddc57/logs/ara-report/result/4c3bccec-1b0e-4d51-b4ae-91881a8e5c4d/ | 16:13 |
| odyssey4me | that's stackviz | 16:13 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Update role for new source build process https://review.openstack.org/648029 | 16:14 |
| chandankumar | odyssey4me: not like that but in a file? | 16:14 |
| chandankumar | at one place | 16:14 |
| chandankumar | like fooo venv -> pip list ; bar venv -> pip list | 16:15 |
| odyssey4me | chandankumar that will happen with the integrated build testing | 16:16 |
| chandankumar | ok, cool! | 16:16 |
| *** gyee has joined #openstack-ansible | 16:17 | |
| *** KeithMnemonic has quit IRC | 16:18 | |
| odyssey4me | chandankumar the fail is in tempest execution though, and looks like it might be nova related? http://logs.openstack.org/26/648026/3/check/openstack-ansible-functional-centos-7/80ddc57/logs/ara-report/result/887e5d25-f51f-4aee-aef9-77def27532de/ | 16:18 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Reuse tempest_tempest_conf_overrides in tempestconf extra args https://review.openstack.org/648137 | 16:19 |
| chandankumar | odyssey4me: in other tempestconf job, it failed with nova issue what i think got fixed long time a go | 16:20 |
| chandankumar | in tempestconf | 16:20 |
| chandankumar | odyssey4me: one more thing https://review.openstack.org/#/c/648026/3/tasks/tempestconf.yml@25 | 16:22 |
| chandankumar | is it possible to move this stuff in tempest source install itself? | 16:22 |
| chandankumar | it is just a venv name change | 16:22 |
| odyssey4me | chandankumar anything is possible, but I dunno if it's advisable | 16:23 |
| odyssey4me | I don't really have enough head space to deal with this all right now. I'll give it some thought and can maybe get back to you tomorrow afternoon. | 16:24 |
| chandankumar | odyssey4me: that will work! | 16:24 |
| odyssey4me | I'm looking at the whole plugin install thing and wondering whether than can consolidate down too. | 16:25 |
| chandankumar | odyssey4me: yes that too also! | 16:25 |
| *** chandankumar is now known as raukadah | 16:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 16:28 |
| *** shyamb has joined #openstack-ansible | 16:30 | |
| goldenfri | Hi, I have another question if someone has a moment, I am deploying rocky with ceph and during the glance install it fails at "Retrieve keyrings for openstack clients from ceph cluster" because it looks like ceph isn't installed in the glance container on the infra nodes | 16:30 |
| goldenfri | I assume it should be installed on the glance container at some point? | 16:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update role for new source build process https://review.openstack.org/648071 | 16:38 |
| masterpe | Hi, I'm try to install OSA where are the apt files comming from local mirror. To create to mirror I use aptly. At creating the lxc-cache I hit a bug that it creates a /var/lib/machines/ubuntu-bionic-amd64/etc/apt/trusted.gpg.d/trusted.gpg.d/ I solved this with: https://github.com/openstack/openstack-ansible-lxc_hosts/pull/2 | 16:39 |
| odyssey4me | masterpe as you'll see in the PR, openstack requires a different development workflow - if you can submit a patch through that workflow it'd be great, otherwise please register a bug with details about the issue and the solution that worked for you | 16:46 |
| *** KeithMnemonic has joined #openstack-ansible | 16:48 | |
| guilhermesp | afternooons | 16:49 |
| guilhermesp | hum maybe tosky has a clue about these failing tests ? http://logs.openstack.org/50/647850/5/check/openstack-ansible-functional-centos-7/fcca2e3/logs/ara-report/result/8ee58bc6-62c8-4660-a038-b824706d7e26/ | 16:49 |
| cloudnull | o/ guilhermesp | 16:49 |
| guilhermesp | hello hello cloudnull | 16:49 |
| tosky | guilhermesp: the plugins are missing, but I added code for that | 16:51 |
| tosky | unless it was reverted | 16:51 |
| tosky | guilhermesp: http://git.openstack.org/cgit/openstack/openstack-ansible-os_sahara/commit/?id=ed7ccad8987e89953e349c9dc9cd4b30232f9425 | 16:53 |
| masterpe | odyssey4me: thnks I was already reading up | 16:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_masakari master: Clean up testing https://review.openstack.org/648183 | 16:54 |
| *** Darcidride has joined #openstack-ansible | 16:56 | |
| tosky | guilhermesp: and yeah, you removed venv_pip_packages which took care of also including the plugins | 16:56 |
| tosky | so it's a no go | 16:56 |
| guilhermesp | yeah | 16:56 |
| guilhermesp | I was just seeing this | 16:57 |
| guilhermesp | going to fix it | 16:57 |
| guilhermesp | thanks tosky | 16:57 |
| *** shyamb has quit IRC | 16:58 | |
| *** hamzaachi has quit IRC | 16:58 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_sahara master: Update role for new source build process https://review.openstack.org/647850 | 17:00 |
| *** hamzaachi has joined #openstack-ansible | 17:02 | |
| *** hamzaachi_ has joined #openstack-ansible | 17:04 | |
| *** hamzaachi has quit IRC | 17:07 | |
| *** kopecmartin is now known as kopecmartin|off | 17:10 | |
| cyberpear | Will openstacksdk merge config from a clouds.yaml and environment variables? | 17:17 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-repo_server master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648186 | 17:17 |
| cyberpear | There is no environment variable for "verify=no" or "insecure=yes" that I can see | 17:17 |
| *** KeithMnemonic has quit IRC | 17:17 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 17:18 |
| cyberpear | (btw, thanks odyssey4me for the hint earlier) | 17:18 |
| *** cmart has quit IRC | 17:18 | |
| odyssey4me | cyberpear yes, IIRC env vars override yaml config entries | 17:19 |
| cyberpear | I tried "{clouds: {envvars: {verify: no}}}" but the openstack inventory plugin for ansible is still failing | 17:20 |
| cyberpear | If I set OS_CACERT env var, things work, but I don't want to force end users to download the self-signed cert | 17:21 |
| odyssey4me | cyberpear on, I dunno about those envvars | 17:23 |
| cyberpear | the modules use a different interface (sdk.connect) than the inventory plugin (OpenStackInventory)... the modules support verify=no directly independent of clouds.yaml | 17:24 |
| cyberpear | it's unrelated, but you review a pull request for me? https://github.com/ansible/ansible/pull/53367 | 17:24 |
| cyberpear | it removes the requirement of the anisble inventory file for openstack to end in .yaml (or .yml) | 17:25 |
| *** priteau has quit IRC | 17:27 | |
| *** KeithMnemonic has joined #openstack-ansible | 17:29 | |
| odyssey4me | cyberpear maybe my updates to https://gist.github.com/odyssey4me/fd340be50301a8b7d6eca46f65866caa help? | 17:31 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_cinder master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648188 | 17:31 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_nova master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648189 | 17:31 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_ironic master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648190 | 17:31 |
| * cyberpear looks | 17:38 | |
| *** KeithMnemonic has quit IRC | 17:42 | |
| *** priteau has joined #openstack-ansible | 17:44 | |
| *** cshen has joined #openstack-ansible | 17:50 | |
| *** cshen has quit IRC | 17:55 | |
| *** miloa has quit IRC | 17:55 | |
| *** DanyC_ has quit IRC | 18:03 | |
| *** DanyC has joined #openstack-ansible | 18:03 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add the manila service to OSA https://review.openstack.org/645653 | 18:05 |
| *** DanyC has quit IRC | 18:08 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_trove master: Update role for new source build process https://review.openstack.org/648204 | 18:22 |
| *** electrofelix has quit IRC | 18:27 | |
| *** hamzaachi_ has quit IRC | 18:31 | |
| *** KeithMnemonic has joined #openstack-ansible | 18:39 | |
| *** priteau has quit IRC | 18:43 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_trove master: Update role for new source build process https://review.openstack.org/648204 | 18:43 |
| goldenfri | Does any one know why I would get this failure running openstack-ansible os-glance-install.yml in rocky? http://paste.openstack.org/show/748502/ I'm at a loss at this point | 18:45 |
| *** KeithMnemonic has quit IRC | 18:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Update role for new source build process https://review.openstack.org/647833 | 18:49 |
| *** CeeMac_ has joined #openstack-ansible | 18:52 | |
| *** cmart has joined #openstack-ansible | 18:53 | |
| *** SmearedBeard has quit IRC | 18:55 | |
| *** CeeMac has quit IRC | 18:56 | |
| admin0 | anyone seen this before: 'Connection aborted.', BadStatusLine('No status line received - the server has closed the connection' | 19:06 |
| admin0 | this is coming up on heat stack deletion | 19:06 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Update role for new source build process https://review.openstack.org/648215 | 19:06 |
| admin0 | Unable to establish connection to https://endpoint:8004/v1/xxxxx/stacks | 19:06 |
| noonedeadpunk | odyssey4me: I guess this ^ blocked https://review.openstack.org/#/c/647851/3 | 19:06 |
| *** Darcidride has quit IRC | 19:07 | |
| *** Darcidride has joined #openstack-ansible | 19:08 | |
| noonedeadpunk | not sure wether I do everything correctly in this patch, but hope so | 19:08 |
| noonedeadpunk | just really missed this task during placement of the first ceilometer patch... | 19:09 |
| *** gillesMo has quit IRC | 19:15 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_trove master: Update role for new source build process https://review.openstack.org/648204 | 19:26 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_keystone master: Update role for new source build process https://review.openstack.org/648222 | 19:33 |
| *** hamzaachi has joined #openstack-ansible | 19:35 | |
| *** Darcidride has quit IRC | 19:36 | |
| *** Darcidride has joined #openstack-ansible | 19:36 | |
| *** cshen has joined #openstack-ansible | 19:50 | |
| mnaser | cores: can we land https://review.openstack.org/#/c/648167/1 ? work is being blocked by timeouts | 19:51 |
| cloudnull | done | 19:52 |
| *** DanyC has joined #openstack-ansible | 19:53 | |
| *** pcaruana has quit IRC | 19:53 | |
| *** cshen has quit IRC | 19:55 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_aodh master: Test using integrated build https://review.openstack.org/647851 | 19:59 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_octavia master: Update role for new source build process https://review.openstack.org/648231 | 20:12 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible master: barbican: run without SSL inside gate for metal jobs https://review.openstack.org/648234 | 20:19 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_barbican master: Test using integrated build https://review.openstack.org/648104 | 20:19 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_rally master: Update role for new source build process https://review.openstack.org/648236 | 20:27 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_keystone master: Update role for new source build process https://review.openstack.org/648222 | 20:28 |
| *** SmearedBeard has joined #openstack-ansible | 20:32 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Cleanup config from keystone_authtoken section https://review.openstack.org/648115 | 20:35 |
| *** DanyC has quit IRC | 20:37 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_blazar master: Updated from OpenStack Ansible Tests https://review.openstack.org/639522 | 20:41 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_blazar master: Update role for new source build process https://review.openstack.org/647835 | 20:41 |
| *** mmercer has joined #openstack-ansible | 20:42 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_blazar master: Correct notification driver https://review.openstack.org/635673 | 20:43 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_blazar master: fix tox python3 overrides https://review.openstack.org/606810 | 20:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: cleanup: move venv install into main to avoid extra include https://review.openstack.org/648124 | 20:43 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_blazar master: missing document start "---" https://review.openstack.org/614097 | 20:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara master: Update role for new source build process https://review.openstack.org/647850 | 20:56 |
| openstackgerrit | Merged openstack/openstack-ansible-os_blazar master: Add bugs url link to README https://review.openstack.org/622881 | 20:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Update role for new source build process https://review.openstack.org/648073 | 21:09 |
| *** DanyC has joined #openstack-ansible | 21:14 | |
| *** aedc has joined #openstack-ansible | 21:20 | |
| mnaser | hi there: https://review.openstack.org/#/c/648222/ if possible :) | 21:26 |
| cloudnull | done | 21:28 |
| cloudnull | mnaser trade you https://review.openstack.org/#/c/645653/ ? | 21:30 |
| cloudnull | :) | 21:30 |
| *** cyberpear has quit IRC | 21:32 | |
| mnaser | cloudnull: did you get a the role tests running using integrated? | 21:33 |
| mnaser | looks like its not happy https://review.openstack.org/#/c/647856/ | 21:33 |
| guilhermesp | hum mnaser I suppose here we are not going to use rally_git_repo https://review.openstack.org/#/c/648236/1 | 21:34 |
| guilhermesp | then we are not going to use rally_git_constraints? | 21:34 |
| mnaser | guilhermesp: does rally not use upper constraints anymore? | 21:35 |
| guilhermesp | the code seems to be using, but there is that part the just comments the git repo and use the pip package to install the stable rally verisions https://github.com/openstack/openstack-ansible-os_rally/blob/9fdc3ab1784e418496ba59566f7da85a5b0a10f0/defaults/main.yml#L28 | 21:39 |
| *** spatel has quit IRC | 21:42 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Update role for new source build process https://review.openstack.org/648215 | 21:49 |
| *** ivve has quit IRC | 21:50 | |
| *** cshen has joined #openstack-ansible | 21:50 | |
| *** SmearedBeard has quit IRC | 21:54 | |
| *** cshen has quit IRC | 21:55 | |
| *** shananigans has quit IRC | 21:56 | |
| *** aedc has quit IRC | 22:12 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_masakari master: Update role for new source build process https://review.openstack.org/648036 | 22:12 |
| cloudnull | mnaser no its not happy yet . however I believe the integrated tests need the role to be supported in the integrated repo before that will work. | 22:19 |
| cloudnull | however i'm not sure. | 22:19 |
| * cloudnull basing that off of https://review.openstack.org/#/c/647898/ | 22:20 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Update role for new source build process https://review.openstack.org/648222 | 22:20 |
| *** hamzaachi has quit IRC | 22:23 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_barbican master: cleanup: stop installing git for barbican run-time https://review.openstack.org/648256 | 22:27 |
| *** DanyC has quit IRC | 22:29 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Extend AIO build timeout to 2.5hrs https://review.openstack.org/648167 | 22:29 |
| openstackgerrit | Merged openstack/openstack-ansible master: Update the congress service egg name to openstack_congress https://review.openstack.org/648122 | 22:29 |
| *** DanyC has joined #openstack-ansible | 22:29 | |
| *** dxiri has joined #openstack-ansible | 22:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Update role for new source build process https://review.openstack.org/648204 | 22:38 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_manila master: Update manila so that it works with cephfs https://review.openstack.org/643616 | 22:39 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_manila master: Update manila so that it works with cephfs https://review.openstack.org/643616 | 22:39 |
| *** jeh has quit IRC | 22:41 | |
| *** Darcidride has quit IRC | 22:44 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add zero-container AIO distro install jobs https://review.openstack.org/647840 | 22:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Remove experimetal scenarios https://review.openstack.org/647897 | 22:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add panko to integrated build https://review.openstack.org/647898 | 22:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add telemetry service bundle to AIO tests https://review.openstack.org/647921 | 22:48 |
| openstackgerrit | Merged openstack/openstack-ansible-os_manila master: Update manila so that it works with cephfs https://review.openstack.org/643616 | 22:59 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/rocky: Update workspace tempest.conf on changes https://review.openstack.org/638030 | 23:18 |
| *** tosky has quit IRC | 23:27 | |
| *** cshen has joined #openstack-ansible | 23:51 | |
| *** cshen has quit IRC | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!