| *** chyka has joined #openstack-ansible | 00:02 | |
| *** chyka has quit IRC | 00:06 | |
| *** pmannidi has joined #openstack-ansible | 00:10 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Improve execution time by using dynamic groups https://review.openstack.org/513876 | 00:15 |
|---|---|---|
| *** jwitko has quit IRC | 00:21 | |
| *** thorst has joined #openstack-ansible | 00:29 | |
| *** chyka has joined #openstack-ansible | 00:38 | |
| *** dxiri has quit IRC | 00:38 | |
| *** newmember has quit IRC | 00:42 | |
| *** chyka has quit IRC | 00:42 | |
| *** thorst has quit IRC | 00:46 | |
| *** kylek3h has quit IRC | 00:55 | |
| *** kylek3h has joined #openstack-ansible | 00:56 | |
| *** brad[] has quit IRC | 00:59 | |
| *** brad[] has joined #openstack-ansible | 01:00 | |
| *** chyka has joined #openstack-ansible | 01:07 | |
| *** chyka has quit IRC | 01:07 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 01:16 |
| *** chigang_ has joined #openstack-ansible | 01:17 | |
| *** persia has quit IRC | 01:21 | |
| *** gus has quit IRC | 01:21 | |
| *** gus has joined #openstack-ansible | 01:22 | |
| *** persia has joined #openstack-ansible | 01:22 | |
| *** jwitko_ has joined #openstack-ansible | 01:25 | |
| *** thorst has joined #openstack-ansible | 01:29 | |
| *** thorst has quit IRC | 01:34 | |
| *** jwitko_ has quit IRC | 01:40 | |
| *** smatzek has joined #openstack-ansible | 01:49 | |
| *** kiennt26 has joined #openstack-ansible | 01:51 | |
| *** esberglu has joined #openstack-ansible | 01:52 | |
| *** kylek3h_ has joined #openstack-ansible | 01:53 | |
| *** kiennt26 has quit IRC | 01:53 | |
| *** kiennt26 has joined #openstack-ansible | 01:54 | |
| *** smatzek has quit IRC | 01:54 | |
| *** kylek3h has quit IRC | 01:56 | |
| *** esberglu has quit IRC | 01:57 | |
| *** hw_wutianwei has joined #openstack-ansible | 02:07 | |
| *** smatzek has joined #openstack-ansible | 02:18 | |
| *** jhesketh_ is now known as jhesketh | 02:22 | |
| *** woodard has quit IRC | 02:30 | |
| *** smatzek has quit IRC | 02:41 | |
| *** smatzek has joined #openstack-ansible | 02:41 | |
| *** smatzek has quit IRC | 02:42 | |
| *** smatzek has joined #openstack-ansible | 02:42 | |
| *** thorst has joined #openstack-ansible | 02:46 | |
| *** smatzek has quit IRC | 02:46 | |
| *** thorst has quit IRC | 02:50 | |
| *** dave-mccowan has quit IRC | 02:59 | |
| *** dxiri has joined #openstack-ansible | 03:18 | |
| *** esberglu has joined #openstack-ansible | 03:20 | |
| *** dxiri has quit IRC | 03:22 | |
| *** dxiri has joined #openstack-ansible | 03:24 | |
| *** esberglu has quit IRC | 03:25 | |
| *** dxiri has quit IRC | 03:36 | |
| *** gkadam has joined #openstack-ansible | 03:38 | |
| SamYaple | is OSA going to use the new peer discovery stuff in rabbitmq 3.7.0 (aka autocluster) | 03:51 |
| *** nshetty has joined #openstack-ansible | 03:51 | |
| SamYaple | i just setup autocluster + etcd and i like it | 03:51 |
| *** kiennt26 has quit IRC | 04:07 | |
| *** poopcat has quit IRC | 04:07 | |
| *** esberglu has joined #openstack-ansible | 04:16 | |
| *** esberglu has quit IRC | 04:20 | |
| *** ivveh- is now known as ivveh | 04:31 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 04:31 |
| *** thorst has joined #openstack-ansible | 04:51 | |
| *** udesale has joined #openstack-ansible | 04:54 | |
| *** thorst has quit IRC | 04:56 | |
| *** mummer has joined #openstack-ansible | 04:59 | |
| *** mummer has quit IRC | 04:59 | |
| *** Trident has quit IRC | 05:03 | |
| *** hw_wutianwei has quit IRC | 05:08 | |
| *** hw_wutianwei has joined #openstack-ansible | 05:09 | |
| *** hw_wutianwei has quit IRC | 05:12 | |
| *** hw_wutianwei has joined #openstack-ansible | 05:15 | |
| *** nwonknu has quit IRC | 05:22 | |
| *** armaan has joined #openstack-ansible | 05:32 | |
| *** cloudnull has quit IRC | 05:32 | |
| *** eglute has quit IRC | 05:33 | |
| *** eglute has joined #openstack-ansible | 05:34 | |
| *** cloudnull has joined #openstack-ansible | 05:34 | |
| *** nwonknu has joined #openstack-ansible | 05:35 | |
| *** korean101 has left #openstack-ansible | 05:42 | |
| *** cloudnull has quit IRC | 05:48 | |
| *** cloudnull has joined #openstack-ansible | 05:51 | |
| *** armaan_ has joined #openstack-ansible | 06:03 | |
| *** esberglu has joined #openstack-ansible | 06:03 | |
| *** armaan has quit IRC | 06:04 | |
| *** armaan has joined #openstack-ansible | 06:05 | |
| *** armaan_ has quit IRC | 06:06 | |
| *** kiennt26 has joined #openstack-ansible | 06:07 | |
| *** esberglu has quit IRC | 06:07 | |
| *** Oku_OS-away is now known as Oku_OS | 06:07 | |
| *** yolanda_ has joined #openstack-ansible | 06:12 | |
| *** gouthamr has quit IRC | 06:14 | |
| *** jvidal has joined #openstack-ansible | 06:19 | |
| *** acormier has joined #openstack-ansible | 06:22 | |
| *** acormier has quit IRC | 06:27 | |
| *** Miouge has joined #openstack-ansible | 06:30 | |
| *** DanyC has joined #openstack-ansible | 06:35 | |
| *** pcaruana has joined #openstack-ansible | 06:39 | |
| *** DanyC has quit IRC | 06:40 | |
| *** yolanda_ has quit IRC | 06:40 | |
| *** rodolof has joined #openstack-ansible | 06:43 | |
| *** SmearedBeard has joined #openstack-ansible | 06:50 | |
| *** thorst has joined #openstack-ansible | 06:53 | |
| *** SmearedBeard has quit IRC | 06:56 | |
| evrardjp | SamYaple: interesting | 06:56 |
| *** huxinhui_ has joined #openstack-ansible | 06:56 | |
| evrardjp | We have thought about something else recently but MQ reforming is high on our lists | 06:57 |
| *** thorst has quit IRC | 06:57 | |
| evrardjp | SamYaple: If you could tell us more that would be great | 06:57 |
| *** sxc731 has joined #openstack-ansible | 07:06 | |
| *** rodolof has quit IRC | 07:07 | |
| *** rodolof has joined #openstack-ansible | 07:08 | |
| *** maybebuggy has joined #openstack-ansible | 07:16 | |
| *** esberglu has joined #openstack-ansible | 07:16 | |
| *** esberglu has quit IRC | 07:21 | |
| *** taseer1 has joined #openstack-ansible | 07:25 | |
| *** Taseer has quit IRC | 07:26 | |
| *** taseer2 has joined #openstack-ansible | 07:26 | |
| *** taseer1 has quit IRC | 07:30 | |
| *** vnogin has joined #openstack-ansible | 07:35 | |
| *** yolanda_ has joined #openstack-ansible | 07:42 | |
| *** yolanda_ is now known as yolanda | 07:43 | |
| *** esberglu has joined #openstack-ansible | 07:53 | |
| *** esberglu has quit IRC | 07:54 | |
| *** esberglu has joined #openstack-ansible | 07:54 | |
| *** esberglu has quit IRC | 07:54 | |
| *** DanyC has joined #openstack-ansible | 07:55 | |
| *** armaan has quit IRC | 07:59 | |
| *** armaan has joined #openstack-ansible | 08:00 | |
| *** sxc731 has quit IRC | 08:01 | |
| *** madorn has quit IRC | 08:02 | |
| evrardjp | odyssey4me: where are the apt sources configured in lxc_container_create? | 08:03 |
| *** madorn has joined #openstack-ansible | 08:04 | |
| evrardjp | oh maybe it's on lxc_hosts in the base template? | 08:05 |
| evrardjp | ok | 08:05 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-lxc_hosts master: SUSE: Improve openSUSE repositories handling https://review.openstack.org/514850 | 08:06 |
| *** pbandark has joined #openstack-ansible | 08:14 | |
| *** thorst has joined #openstack-ansible | 08:14 | |
| *** thorst has quit IRC | 08:19 | |
| openstackgerrit | Kevin Lefevre proposed openstack/openstack-ansible-os_nova master: Enable KSM support for Nova https://review.openstack.org/507862 | 08:20 |
| *** taseer2 is now known as Taseer | 08:25 | |
| *** nshetty_ has joined #openstack-ansible | 08:32 | |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible master: Integrate Congress with OSA. https://review.openstack.org/503971 | 08:33 |
| *** electrofelix has joined #openstack-ansible | 08:34 | |
| *** nshetty has quit IRC | 08:35 | |
| *** nshetty__ has joined #openstack-ansible | 08:40 | |
| *** nshetty__ has quit IRC | 08:41 | |
| *** nshetty_ has quit IRC | 08:43 | |
| *** taseer1 has joined #openstack-ansible | 08:43 | |
| *** Taseer has quit IRC | 08:43 | |
| *** taseer2 has joined #openstack-ansible | 08:44 | |
| *** nshetty has joined #openstack-ansible | 08:44 | |
| *** taseer1 has quit IRC | 08:48 | |
| *** br0k3n has joined #openstack-ansible | 08:51 | |
| *** br0k3n has quit IRC | 08:55 | |
| *** br0k3n has joined #openstack-ansible | 08:55 | |
| *** pmannidi has quit IRC | 08:59 | |
| *** armaan has quit IRC | 09:16 | |
| *** armaan has joined #openstack-ansible | 09:16 | |
| *** armaan has quit IRC | 09:16 | |
| *** SmearedBeard has joined #openstack-ansible | 09:17 | |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible-os_cinder master: Fix indentation of container_vars for ceph doc https://review.openstack.org/514977 | 09:25 |
| Miouge | The doc for ceph + cinder says to use both “storage_hosts” and “storage-infra_hosts” (see https://docs.openstack.org/openstack-ansible-os_cinder/latest/configure-cinder.html#configuring-cinder-to-use-ceph ) | 09:34 |
| Miouge | Espicially the “container_vars” section just below the “storage-infra_hosts” then later it ends up in “storage_hosts” | 09:34 |
| odyssey4me | evrardjp they are not - they are configured in the base by copying from the host | 09:34 |
| evrardjp | yeah | 09:34 |
| evrardjp | I have seen that | 09:34 |
| *** rodolof has quit IRC | 09:35 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514753 | 09:36 |
| *** hw_wutianwei has quit IRC | 09:38 | |
| *** SmearedBeard has quit IRC | 09:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Correct regex for disk size https://review.openstack.org/514996 | 09:47 |
| *** rcarrill1 is now known as rcarrillocruz | 09:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_monasca stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514754 | 09:48 |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514747 | 09:48 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514755 | 09:49 |
| *** rpittau_ has quit IRC | 09:49 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514749 | 09:51 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514751 | 09:55 |
| openstackgerrit | Merged openstack/openstack-ansible-os_zaqar stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514756 | 09:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514752 | 09:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Use new zuul v3 git src location https://review.openstack.org/515006 | 10:02 |
| odyssey4me | evrardjp andymccr ^ do you think we should add a reno to announce the change in default location? | 10:02 |
| evrardjp | no. it's fine. | 10:04 |
| evrardjp | I have read the commit message twice, still don't understand what you meant there. | 10:05 |
| odyssey4me | ok, sorry - maybe I need more coffee | 10:05 |
| odyssey4me | which paragraph is confusing? | 10:05 |
| evrardjp | all of them. | 10:06 |
| odyssey4me | haha | 10:06 |
| evrardjp | I understand you want to simplify developer mode now that we have zuulv3. | 10:06 |
| evrardjp | using /src | 10:06 |
| evrardjp | that's it. | 10:06 |
| *** duonghq has quit IRC | 10:06 | |
| odyssey4me | not simplify developer mode - simplify the job | 10:06 |
| evrardjp | yeah | 10:06 |
| evrardjp | well -- reuse /src basically | 10:07 |
| odyssey4me | yes | 10:07 |
| *** kiennt26 has quit IRC | 10:07 | |
| odyssey4me | previously we had to make sure that 'keystone', for example, was cloned into the right place | 10:07 |
| odyssey4me | now we don't | 10:07 |
| odyssey4me | that's the first paragraph in a nutshell | 10:07 |
| odyssey4me | https://i.imgur.com/Y2Rrixn.gif | 10:10 |
| odyssey4me | damn, that took a lot longer than it should have | 10:11 |
| *** udesale has quit IRC | 10:13 | |
| *** Miouge has quit IRC | 10:21 | |
| *** Miouge has joined #openstack-ansible | 10:22 | |
| *** Miouge has quit IRC | 10:25 | |
| *** Miouge has joined #openstack-ansible | 10:25 | |
| *** stuartgr has joined #openstack-ansible | 10:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Update test image for nova-lxd https://review.openstack.org/515020 | 10:33 |
| *** Miouge has quit IRC | 10:33 | |
| *** pcaruana has quit IRC | 10:37 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Use zuul v3 src location to install ARA https://review.openstack.org/515022 | 10:46 |
| *** chhavi has joined #openstack-ansible | 10:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Use zuul v3 src location to install ARA https://review.openstack.org/515023 | 10:49 |
| *** armaan has joined #openstack-ansible | 11:02 | |
| *** ansibler has joined #openstack-ansible | 11:06 | |
| *** taseer2 is now known as Taseer | 11:08 | |
| odyssey4me | SamYaple yeah, that autocluster stuff looks good - nice to see it being adopted by pivotal as official now | 11:09 |
| odyssey4me | oh, actually I see that autocluster is superceded now: http://next.rabbitmq.com/cluster-formation.html | 11:10 |
| odyssey4me | well, for rabbitmq 3.7 | 11:11 |
| *** rodolof has joined #openstack-ansible | 11:12 | |
| Taseer | odyssey4me: Hi, can you help with this => http://logs.openstack.org/71/503971/13/check/openstack-tox-linters/e12ab77/job-output.txt.gz#_2017-10-25_08_48_18_023130 | 11:14 |
| *** armaan has quit IRC | 11:17 | |
| *** armaan has joined #openstack-ansible | 11:17 | |
| odyssey4me | Taseer yeah, you've added some inventory groups and haven't added the group names to the test which validates that all the groups are there | 11:18 |
| odyssey4me | it's actually telling you which file they're in | 11:18 |
| odyssey4me | look in tests/test_inventory.py and it'll be pretty easy to find | 11:19 |
| *** yolanda has quit IRC | 11:19 | |
| *** nshetty has quit IRC | 11:19 | |
| odyssey4me | evrardjp is this an error you've seen elsewhere? | 11:21 |
| odyssey4me | http://logs.openstack.org/06/515006/1/check/openstack-ansible-functional-centos-7/493ec89/job-output.txt.gz#_2017-10-25_10_48_45_675316 | 11:21 |
| *** thorst has joined #openstack-ansible | 11:21 | |
| evrardjp | that's a new one to me | 11:22 |
| *** nshetty has joined #openstack-ansible | 11:25 | |
| *** thorst has quit IRC | 11:26 | |
| *** smatzek has joined #openstack-ansible | 11:28 | |
| odyssey4me | well, say goodbye to the legacy jobs: https://review.openstack.org/514592 | 11:28 |
| *** dave-mccowan has joined #openstack-ansible | 11:31 | |
| ansibler | Hi, I am guessing this is a bug https://pastebin.com/raw/ZwcricqM I can repeat that if starting install from scratch. Could be some kind of race condition issue. | 11:35 |
| odyssey4me | ansibler that's a purposefully ignored error - it doesn't cause failure | 11:35 |
| ansibler | so far i see that "content" field differs between the runs, but hope someone with deeper understanding can have a look | 11:36 |
| ansibler | I am confused with need to run setup-infrastructure twice. I did run ok on newton, but now with pike I cannot get it wirking with first run | 11:37 |
| ansibler | I have pasted how task run looks in first iteration and also in second one. First time 2 tasks fail, second time succeeed | 11:39 |
| *** pcaruana has joined #openstack-ansible | 11:43 | |
| ansibler | could cope with running setup-infrastructure.yml 2x, but after setup-openstack.yml I end up with cluster where rabbitmq refuses service requests from other services, so I am guessing it could be related to not configuring rabbitmq with first run. I am not changing user_secrets between runs. | 11:43 |
| *** PTO has joined #openstack-ansible | 11:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Document reasonable radosgw timeout https://review.openstack.org/514750 | 11:47 |
| odyssey4me | evrardjp oh bother, that centos issue is showing up in all tests that use memcached today | 11:48 |
| odyssey4me | oanson is there something holding https://review.openstack.org/514268 back from being merged given that it's non-voting? | 11:51 |
| *** PaulusNL has joined #openstack-ansible | 11:56 | |
| odyssey4me | hwoarang any progress with the log uploads? | 11:58 |
| hwoarang | odyssey4me: have you seen my e-mail? it's a zuul problem and nobody contacted me since yesterday | 11:58 |
| hwoarang | so i presume nobody fixed anything | 11:58 |
| odyssey4me | hwoarang yeah, saw it - I was just wondering whether you've had further feedback | 11:58 |
| hwoarang | no :( | 11:59 |
| odyssey4me | cool, I'll follow up with jeblair later | 11:59 |
| hwoarang | the thing that Monty uploaded in my patchset should work | 11:59 |
| hwoarang | the problem is that the 'run' phase is executed from stable/newton ! | 11:59 |
| hwoarang | where there is no post-run playbook | 11:59 |
| PaulusNL | hello all, I'm new here and have a question regarding openstack-ansible on CentOS (cause that is our default OS) | 11:59 |
| hwoarang | no clue why and that's the real zuul issue it seems | 12:00 |
| PaulusNL | how experimental is experimental cause I'm not getting it to work for deploying it in the end | 12:00 |
| evrardjp | PaulusNL: what do you mean? | 12:01 |
| evrardjp | it fully depends on what you want to install... | 12:01 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_octavia master: add segmentation id option to octavia mgmt network creation task https://review.openstack.org/515035 | 12:01 |
| PaulusNL | evrardjp: when the documentation is followed and I'm on the stage to run the playbooks | 12:02 |
| PaulusNL | it always breaks at setup-infrastructure.yml | 12:02 |
| PaulusNL | and for example no /var/www directory is created so the rest will not work | 12:03 |
| evrardjp | PaulusNL: without logs, that's hard to say. | 12:03 |
| evrardjp | How many nodes do you have? | 12:03 |
| evrardjp | for the infrastructure nodes? | 12:03 |
| *** pcaruana has quit IRC | 12:03 | |
| PaulusNL | 1 deploy and one controller host | 12:03 |
| *** pcaruana has joined #openstack-ansible | 12:04 | |
| PaulusNL | the controller host is not deployed with the ansible playbook | 12:04 |
| PaulusNL | eventually it will be bigger | 12:04 |
| PaulusNL | but for now this is eventually what is happening | 12:04 |
| odyssey4me | hwoarang that's bizarre - how do you know it's trying to use the newton branch? | 12:05 |
| hwoarang | odyssey4me: http://logs.openstack.org/06/513706/12/check/openstack-ansible-functional-opensuse-423/3aa14c1/job-output.txt.gz#_2017-10-24_12_43_17_517913 | 12:05 |
| hwoarang | ...playbooks/run@stable/newton] | 12:05 |
| odyssey4me | oh I see http://logs.openstack.org/06/513706/12/check/openstack-ansible-functional-ubuntu-xenial/32b4967/job-output.txt.gz#_2017-10-24_12_40_05_323870 | 12:05 |
| odyssey4me | haha, same same | 12:05 |
| hwoarang | so strange... | 12:05 |
| hwoarang | i think that's the issue jeblair is looking into | 12:06 |
| odyssey4me | ok | 12:06 |
| *** masuberu has joined #openstack-ansible | 12:06 | |
| odyssey4me | I'll chase that up today as lack of logs is now holding us back substantially | 12:07 |
| evrardjp | PaulusNL: what issue do you have in your logs? Were there any failure in the playbooks? | 12:07 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_cinder master: Change ceph example backend name https://review.openstack.org/514588 | 12:07 |
| *** thorst has joined #openstack-ansible | 12:08 | |
| evrardjp | I suspect a network issue. | 12:08 |
| PaulusNL | ok | 12:08 |
| PaulusNL | evrardjp: I'm looking in the logs (with 2 other collega's) | 12:08 |
| PaulusNL | is this way of installing openstack good for a production environment? | 12:09 |
| *** masber has quit IRC | 12:09 | |
| evrardjp | yeah, we do have large clouds under production :p | 12:10 |
| evrardjp | that is the whole point | 12:10 |
| PaulusNL | ok clear on that :) | 12:10 |
| PaulusNL | We are so new to openstack | 12:10 |
| PaulusNL | thanks evrardjp for your reaction | 12:11 |
| evrardjp | haha :) | 12:11 |
| PaulusNL | going to look at our network setup | 12:11 |
| evrardjp | Jullie kan het doen! Misschien kan jij ook uw openstack_user_config stuuren :p | 12:12 |
| evrardjp | meer informatie nodig zal zijn om het probleem op te lossen | 12:12 |
| PaulusNL | evrardjp: did you use Google Translate? :P | 12:12 |
| evrardjp | just rusty | 12:12 |
| evrardjp | :p | 12:13 |
| PaulusNL | ah ok | 12:13 |
| PaulusNL | but I understand | 12:13 |
| evrardjp | haha | 12:13 |
| *** smatzek has quit IRC | 12:13 | |
| PaulusNL | can I post it to you in a pm? | 12:13 |
| evrardjp | what was the thing that gave me? | 12:13 |
| evrardjp | stuuren ipv zenden? | 12:13 |
| *** smatzek has joined #openstack-ansible | 12:14 | |
| evrardjp | Nah you can paste a link here, everyone can help you that way | 12:14 |
| PaulusNL | evrardjp: the sentences itself and also stuuren is sturen | 12:15 |
| PaulusNL | zenden is just another word | 12:15 |
| evrardjp | yeah | 12:15 |
| evrardjp | darn my grammar is not as good as it was! | 12:16 |
| PaulusNL | but I'm not a native English speaker | 12:16 |
| evrardjp | neither do I :p | 12:16 |
| PaulusNL | ik wil je wel helpen door Nederlands te typen | 12:16 |
| PaulusNL | ;) | 12:16 |
| Taseer | odyssey4me: okay, thanks | 12:17 |
| evrardjp | Niet nodig PaulusNL ;) | 12:17 |
| PaulusNL | ok | 12:17 |
| evrardjp | :D | 12:17 |
| PaulusNL | :P | 12:17 |
| evrardjp | I accept though, even if not necessary here. :p | 12:17 |
| *** br0k3n1 has joined #openstack-ansible | 12:18 | |
| *** smatzek has quit IRC | 12:18 | |
| *** br0k3n has quit IRC | 12:18 | |
| PaulusNL | evrardjp: I'm not going to post the config public. I think that the problem is going to be tackeld because of your remark about network! | 12:19 |
| PaulusNL | ansible playbook want to go to one container that does not exist | 12:19 |
| evrardjp | PaulusNL: well, for now it's just a gut feeling | 12:19 |
| PaulusNL | ... i think a rookie error | 12:19 |
| odyssey4me | PaulusNL the initial config is a bit hard due to all the networking setup - but once you get it right it'll be smooth from there | 12:20 |
| PaulusNL | ok | 12:20 |
| evrardjp | PaulusNL: yeah maybe some step was skipped? | 12:20 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible master: Integrate Congress with OSA. https://review.openstack.org/503971 | 12:21 |
| *** thorst_ has joined #openstack-ansible | 12:24 | |
| PaulusNL | evrardjp: I think so... we are looking with 3 persons now :) | 12:24 |
| evrardjp | haha | 12:25 |
| PaulusNL | we think that this option is our problem | 12:25 |
| PaulusNL | internal_lb_vip_address: | 12:26 |
| evrardjp | same as external_lb_vip_address ? | 12:26 |
| *** thorst has quit IRC | 12:26 | |
| PaulusNL | is this an address that does not exist | 12:26 |
| oanson | odyssey4me, no holdup. It's just waiting for review. | 12:26 |
| evrardjp | PaulusNL: yeah, that could be a problem if you don't make it appear | 12:26 |
| oanson | I can press the core reviewers if there's a hurry | 12:27 |
| evrardjp | PaulusNL: either manual config, or automatic with keepalived for example | 12:27 |
| *** yolanda has joined #openstack-ansible | 12:27 | |
| PaulusNL | so the internal and external must be the same and a now existing host? | 12:27 |
| PaulusNL | *none | 12:27 |
| PaulusNL | I have defined the following | 12:28 |
| PaulusNL | global_overrides: | 12:28 |
| PaulusNL | internal_lb_vip_address: 192.168.114.19 | 12:28 |
| PaulusNL | external_lb_vip_address: os.visc.surfsara.nl | 12:28 |
| odyssey4me | oanson ah ok - I've made a note for a patch that will make it actually do what's expected... right now the cross-repo test won't actually be testing patches to dragonflow... only the merged head - the patch I mentioned fixes that | 12:29 |
| odyssey4me | oanson ping us if you need assistance with anything | 12:29 |
| odyssey4me | I have a patch depending on your patch to allow us to completely remove all the legacy jobs from zuul... so I was just following up. | 12:30 |
| *** esberglu has joined #openstack-ansible | 12:31 | |
| *** nshetty_ has joined #openstack-ansible | 12:31 | |
| PaulusNL | we are running the playbooks again :) | 12:32 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/pike: lxc_cache_preparation: Convert the LXC cache preparation to async task https://review.openstack.org/514776 | 12:33 |
| *** mbuil has joined #openstack-ansible | 12:33 | |
| *** nshetty has quit IRC | 12:33 | |
| evrardjp | PaulusNL: they should not be the same | 12:35 |
| evrardjp | Surfsara! | 12:35 |
| evrardjp | Interesting :) | 12:35 |
| evrardjp | I guess you're interested by our shibboleth integration :p | 12:36 |
| PaulusNL | :P | 12:36 |
| mhayden | happy wednesday | 12:36 |
| PaulusNL | first things first | 12:36 |
| evrardjp | PaulusNL: I did the first implementation for Belnet, so you are not alone :p | 12:36 |
| evrardjp | happy wednesday mhayden! | 12:36 |
| PaulusNL | evrardjp: so you need to just pick two random ip's? | 12:37 |
| odyssey4me | morning mhayden - howdy doodie? | 12:37 |
| evrardjp | "random" is not a good idea :p | 12:37 |
| odyssey4me | FYI I'm seeing http://logs.openstack.org/06/515006/1/check/openstack-ansible-functional-centos-7/493ec89/job-output.txt.gz#_2017-10-25_10_48_45_675316 pop up in centos builds all over the place... trying to see if it replicates outside of infra | 12:38 |
| MikeW | Hey guys I'm having issues with rabbitmq routing neutron events to designate-sink... is it ok that each one of my services has their own virtualhost or is that what's messing me up? | 12:38 |
| evrardjp | PaulusNL: I'd suggest you to have one on your external network, and one on your management network. | 12:38 |
| evrardjp | PaulusNL: they need to be resolvable by the hosts in your cloud | 12:38 |
| PaulusNL | ah ok | 12:38 |
| evrardjp | I'd just say, for internal lb vip address, just use one ip in your br-mgmt, not yet used. | 12:39 |
| PaulusNL | evrardjp: belnet cool :) | 12:39 |
| evrardjp | (put it in the list of the reserved ips) | 12:39 |
| evrardjp | PaulusNL: I am not working there anymore :) | 12:39 |
| PaulusNL | evrardjp: we are doing this right now... | 12:40 |
| evrardjp | for the external lb vip address: you can keep your FQDN, just make sure it resolves by all the possible clients of your cloud, including internal ones | 12:40 |
| evrardjp | that is never good to have bad dns resolution. | 12:40 |
| evrardjp | if you keep the two IPs separate, the better, you can have better isolation of the traffic, and it's easier to configure too | 12:40 |
| *** armaan has quit IRC | 12:41 | |
| PaulusNL | ok | 12:41 |
| *** armaan has joined #openstack-ansible | 12:41 | |
| evrardjp | the external lb vip address will basically be terminated at the haproxy node | 12:41 |
| evrardjp | so the haproxy node will have to get this IP somehow | 12:41 |
| PaulusNL | evrardjp: is it a problem to have them both in the same network... for testing now? | 12:41 |
| evrardjp | no it's not | 12:41 |
| evrardjp | for testing :) | 12:42 |
| PaulusNL | ok thanks! | 12:42 |
| evrardjp | keep 2 ips, this way you don't have overlap in binding | 12:42 |
| evrardjp | you can probably keep the FQDN for the external one, and just use a different IP when on production | 12:42 |
| PaulusNL | can I just run the playbooks again... or do I need to reinstall the environment? | 12:42 |
| evrardjp | as long as this FQDN -> IP will be mapped to the haproxy node you'll do fine | 12:43 |
| evrardjp | re-run | 12:43 |
| evrardjp | :p | 12:43 |
| *** markvoelker has joined #openstack-ansible | 12:43 | |
| evrardjp | let's see how it goes | 12:43 |
| PaulusNL | ok :D | 12:43 |
| PaulusNL | running :D | 12:43 |
| evrardjp | if it doesn't tell us what's wrong | 12:43 |
| evrardjp | we'll work it out together | 12:43 |
| PaulusNL | this will take a while but thank you so very much! | 12:44 |
| MikeW | PaulusNL The guys in this chat are the best in all of IRC (I'm not one of them they've helped me a ton) ;) | 12:45 |
| PaulusNL | MikeW: good to know :) | 12:46 |
| PaulusNL | evrardjp: thank you from all the guys here ^_^ (needed to let you know) | 12:47 |
| evrardjp | PaulusNL: haha no worries :) | 12:47 |
| *** udesale has joined #openstack-ansible | 12:48 | |
| odyssey4me | evrardjp it looks like not even our console logs for tests are getting uploading into elasticsearch :/ | 12:48 |
| evrardjp | MikeW: that's great to hear! It's because you're part of it, with that mindset, that it can stay the best :D | 12:48 |
| odyssey4me | I'll follow that up with infra | 12:48 |
| evrardjp | odyssey4me: yeah but it's been a while | 12:48 |
| tnogisto | not sure if this is the correct chat, but if neutron crashes on loop with: CRITICAL neutron [req-ed23388b-1769-4fab-b191-3f4f1aec2089 - - - - -] Unhandled error: UnmappedInstanceError: Class 'neutron.objects.plugins.ml2.vlanallocation.VlanAllocation' is not mapped | 12:48 |
| evrardjp | odyssey4me: I am clueless on why | 12:48 |
| tnogisto | any ideas what do to? | 12:48 |
| evrardjp | Like I said, I know nothing about logstash.o.o too. | 12:48 |
| odyssey4me | evrardjp when we removed the legacy tests, that's when they stopped working | 12:48 |
| evrardjp | yeah | 12:49 |
| evrardjp | but the logs are still there | 12:49 |
| evrardjp | including periodics | 12:49 |
| evrardjp | tnogisto: that sounds like a bad bad thing. | 12:49 |
| odyssey4me | but those were all legacy jobs - now that they're gone... | 12:49 |
| odyssey4me | where're you seeing the logs? | 12:49 |
| *** esberglu has quit IRC | 12:49 | |
| evrardjp | odyssey4me: which ones? | 12:50 |
| evrardjp | periodics? | 12:50 |
| evrardjp | logs.openstack.org/periodic/git.openstack.org/openstack/openstack-ansible/ | 12:50 |
| tnogisto | Im thinking so too, but I dont know what happened to it | 12:50 |
| odyssey4me | oh, that's interesting | 12:50 |
| evrardjp | odyssey4me: well I don't know if that was by design or not. | 12:51 |
| tnogisto | it crashes after trying to initialize vlan driver | 12:51 |
| odyssey4me | heh, I like the new folder format | 12:51 |
| evrardjp | I have still the https://etherpad.openstack.org/p/osa-zuulv3 where it was shown working | 12:51 |
| evrardjp | I think I have the link there | 12:51 |
| evrardjp | let me check | 12:51 |
| evrardjp | tnogisto: I think it's maybe due to a bad nic mapping? | 12:52 |
| evrardjp | have you double checked your configuration? | 12:52 |
| evrardjp | neutron configuration + openstack_user_config | 12:52 |
| evrardjp | tnogisto: ^ | 12:52 |
| tnogisto | Im trying to get self-provisioned networks to work, so probably they are wrong, but it's still interesting that it crashes that horribly :) | 12:53 |
| *** sxc731 has joined #openstack-ansible | 12:55 | |
| tnogisto | https://pastebin.com/06JVM5NP | 12:56 |
| tnogisto | here is the ml2_conf.ini | 12:57 |
| tnogisto | and its the neutron server container | 12:57 |
| odyssey4me | xgerman_ the scenario you implemented for testing octavia in the integrated gate was only ever implemented as experimental. You never progressed it to a non voting test. https://github.com/openstack/openstack-ansible/blob/master/zuul.d/project.yaml#L26-L29 | 13:01 |
| evrardjp | mhayden: just to make sure we are talking about the same thing: when you said vendor in the gpg key, you meant giving a file to that module, right? http://docs.ansible.com/ansible/latest/rpm_key_module.html | 13:01 |
| *** Miouge has joined #openstack-ansible | 13:01 | |
| evrardjp | mhayden: because https://github.com/openstack/openstack-ansible-pip_install/blob/9372b42afa18d2e5d744de62fb07c2213ca835fa/vars/redhat-7.yml#L38 is the thing that can be slow in gates | 13:01 |
| odyssey4me | evrardjp I think it was me who said that perhaps vendoring in the keys was a good idea, and yes that means having the key in-repo as a file. | 13:02 |
| evrardjp | tnogisto: wow you're on ovs? | 13:02 |
| evrardjp | yeah I am fine with that in a file | 13:02 |
| evrardjp | it doesn't move quite often | 13:02 |
| evrardjp | my question is where does it matter | 13:03 |
| evrardjp | mhayden: is that the proper var I am showing? | 13:03 |
| odyssey4me | evrardjp mhayden heh, in my own test on centos I got the same memcache error | 13:03 |
| evrardjp | that sounds bad | 13:04 |
| evrardjp | that didn't happen yesterday so it's new | 13:05 |
| tnogisto | yas, I want to get OVS working | 13:05 |
| tnogisto | it almost works :) | 13:06 |
| tnogisto | almost | 13:06 |
| mhayden | odyssey4me: yeah, this is really strange | 13:06 |
| odyssey4me | mhayden hmm, it's trying to get the package from the 'openstack-pike' repository | 13:06 |
| evrardjp | tnogisto: haha great | 13:07 |
| tnogisto | I had it working with my not ansible deployment but currently I want to get ansible deployed selfprovisioned networks | 13:07 |
| mgariepy | morning everyone | 13:07 |
| evrardjp | not many ppl are using ovs | 13:07 |
| mhayden | odyssey4me: that's where it probably should be getting it from | 13:07 |
| mhayden | most likely a newer version there | 13:07 |
| mhayden | so my OSA AIO went through just fine | 13:07 |
| gunix | evrardjp: so i shouldn't event ry to learn ovs. thank you. :D | 13:08 |
| mhayden | i wonder if it's the repos in the image biting us again | 13:08 |
| odyssey4me | mhayden sure - so I suspect that maybe the key setup failed silently or something | 13:08 |
| odyssey4me | oh hum, gpgcheck is enabled, but not key in the .repo file - is that normal? | 13:08 |
| evrardjp | gunix: haha don't get me wrong, ovs is fun | 13:08 |
| *** PaulusNL has quit IRC | 13:09 | |
| mhayden | odyssey4me: that's okay -- rpm is smart enough to look through its keys ot find the right one | 13:09 |
| evrardjp | it used to work when someone was actively working on it. I think I heard more and more ppl wanting to use it again | 13:09 |
| gunix | evrardjp: fun or not, it's another week of reading forums, wikis and trying stuff out. i would rather use that time to learn ceph | 13:09 |
| mhayden | so my installation on plain old centos 7 worked fine | 13:09 |
| evrardjp | gunix: well it's not that hard. | 13:09 |
| *** esberglu has joined #openstack-ansible | 13:09 | |
| evrardjp | but I understand why you'd want to invest time on ceph. | 13:10 |
| odyssey4me | mhayden I just did the 'functional' tox env on a plain centos7 using the openstack-ansible-tests repo and it failed with the same issue | 13:10 |
| mhayden | odyssey4me: so i'm not seeing those extra repos being disabled -> http://logs.openstack.org/63/501263/5/check/openstack-ansible-functional-centos-7/a349e07/job-output.txt.gz | 13:11 |
| mhayden | i only see epel getting enabled | 13:11 |
| gunix | evrardjp: what are the technical advantages of using ovs? | 13:11 |
| * mhayden goes back to look at the tests repo | 13:11 | |
| mhayden | hmm yeah, all of the old repos are still enabled it seems | 13:12 |
| odyssey4me | mhayden well, all our job does is executes './run_tests.sh {tox env}' now | 13:13 |
| odyssey4me | https://github.com/openstack/openstack-ansible-tests/blob/master/zuul.d/playbooks/run.yml#L20 | 13:13 |
| odyssey4me | that's how bindep is used, etc... so our own scripts should be sorting that out | 13:13 |
| *** PaulusNL has joined #openstack-ansible | 13:14 | |
| evrardjp | gunix: flow management | 13:14 |
| odyssey4me | it appears that only the epel and centos keys are installed | 13:14 |
| mhayden | hmm well let me fire up a test with openstack-ansible-tests then | 13:14 |
| evrardjp | if you don't use flows, I don't see the point | 13:14 |
| PaulusNL | evrardjp: we are going to reinstall the machines and will go further from there | 13:14 |
| gunix | evrardjp: i have no ideea what you mean with flows | 13:14 |
| evrardjp | odyssey4me: mhayden I am working on a patch to vendor in keys, so let's not redo the wheel, I am already doing that :) | 13:14 |
| evrardjp | PaulusNL: ok | 13:15 |
| evrardjp | before re-deploy, ensure that your networking is fine | 13:15 |
| PaulusNL | evrardjp: I think we will do more on Friday for our openstack test environment | 13:15 |
| mhayden | evrardjp: okay | 13:15 |
| evrardjp | PaulusNL: if you want a quick PoC, that's one line of shell, but I guess that's not what you are looking for :p | 13:15 |
| odyssey4me | evrardjp I'm debugging here - not trying to make a patch. | 13:16 |
| odyssey4me | PaulusNL if you're building a test lab out, you might find https://github.com/openstack/openstack-ansible-ops/tree/master/multi-node-aio interesting | 13:16 |
| PaulusNL | evrardjp: no not really... we want to have a real test for production | 13:17 |
| PaulusNL | labs we already have :) | 13:17 |
| PaulusNL | odyssey4me: thanks for letting know | 13:17 |
| mhayden | odyssey4me: the tasks seem to show that the key is already present -> http://logs.openstack.org/11/514711/1/check/openstack-ansible-functional-centos-7/4ada47d/job-output.txt.gz#_2017-10-25_02_30_10_781970 | 13:18 |
| odyssey4me | mhayden yes, applying the command used in the task, I see the key | 13:19 |
| odyssey4me | however I don't see the key when doing rpm -q gpg-pubkey --qf '%{NAME}-%{VERSION}-%{RELEASE}\t%{SUMMARY}\n' | 13:19 |
| mhayden | that's unusual | 13:19 |
| odyssey4me | well, let's put it this way - I see the hash in the list when using: rpm -vv -q centos-release | 13:19 |
| *** armaan has quit IRC | 13:20 | |
| *** mrch has joined #openstack-ansible | 13:20 | |
| *** armaan has joined #openstack-ansible | 13:20 | |
| *** sxc731 has quit IRC | 13:21 | |
| odyssey4me | mhayden the key shown in the install error is different though: Header V4 RSA/SHA1 Signature, key ID 764429e6 | 13:21 |
| odyssey4me | heh, oh, it isn't | 13:21 |
| *** jafeha has joined #openstack-ansible | 13:21 | |
| jafeha | hello everybody | 13:21 |
| oanson | odyssey4me, Thanks! It seems now the gate fails earlier in the process. I've added openstack/requirements to the required projects, but if that solves the issue maybe it should be pushed to the parent? | 13:23 |
| odyssey4me | mhayden ok, if I run this then it shows up properly: rpm --define '%_hkp_keyserver http://pool.sks-keyservers.net' --import 0x764429E6 | 13:24 |
| odyssey4me | so it seems to me that somehow two databases are not in sync | 13:24 |
| mhayden | weird | 13:25 |
| mhayden | well evrardjp is going to vendor those keys, so hopefully we can avoid these shenanigans in the future | 13:25 |
| odyssey4me | so the plan is to vendor them, and just use ansible to import them idempotently? | 13:26 |
| *** snowman4839 has quit IRC | 13:26 | |
| jafeha | i have a question regarding an ovs setup. we've switched from a bridged setup to ovs and have built all the ovs-bridges for our compute nodes. my deployment is currently broken as the pattern check for CONFIG_OPENVSWITCH fails. | 13:27 |
| mhayden | i believe so, but i'll yield to evrardjp on that | 13:27 |
| * mhayden is about to scoot for a mtg | 13:27 | |
| jafeha | i guess this checks if the necessary kernel modules are loaded. do i have to configure ovs-bridges on my mgmt nodes too or does installing the ovs packages do the trick here? | 13:27 |
| mhayden | 90 minutes straight of meetings, actually | 13:27 |
| odyssey4me | mhayden evrardjp ok, so the root cause of the key fail for centos tests has now been figured out - as you're working on a solution for it anyway I'll leave that one to you | 13:28 |
| *** SerenaFeng has joined #openstack-ansible | 13:31 | |
| evrardjp | yeah, but mine is probably a big patch, so it's maybe a good idea to split it out. Let me do two commits then. | 13:32 |
| *** PTO has quit IRC | 13:32 | |
| *** armaan has quit IRC | 13:32 | |
| evrardjp | so the problem is that it's two different keys in two keyservers? | 13:32 |
| evrardjp | what are the odds for that | 13:33 |
| gunix | tempest is a playbook that is part of osa, not part of default openstack, right? | 13:33 |
| *** ansibler has quit IRC | 13:33 | |
| *** galstrom_zzz is now known as galstrom | 13:33 | |
| mgariepy | anyone here tested manila ? | 13:33 |
| gunix | nevermind, i found the tempest role within ansible. i got what it does | 13:37 |
| gunix | opensource is so cool. i can just read the code to understand | 13:37 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Hide yum-config-manager output https://review.openstack.org/514838 | 13:41 |
| evrardjp | gunix: :) | 13:44 |
| evrardjp | sorry for the late answer, you found it yourself :D | 13:44 |
| *** galstrom is now known as galstrom_zzz | 13:49 | |
| dmsimard | odyssey4me: thanks a lot for working on the ara OSA job! I'll fix the pep8 thing today, it was introduced by a pep8 update recently | 13:49 |
| odyssey4me | dmsimard sure thing :) | 13:49 |
| dmsimard | with a bit of luck, I should also be able to tag ara 0.14.5 which will resolve the ansible.cfg issue with ansible 2.4, it will however blacklist 2.4.0 | 13:50 |
| dmsimard | hoping ansible doesn't delay 2.4.1 any further | 13:50 |
| *** galstrom_zzz is now known as galstrom | 13:50 | |
| odyssey4me | evrardjp the issue is that the key is somehow half imported.... so it'd be better if we can just vendor the key and import it idempotently | 13:50 |
| evrardjp | ok | 13:50 |
| evrardjp | I am indeed partly working on that, so it's good | 13:51 |
| *** gus has quit IRC | 13:51 | |
| *** gus has joined #openstack-ansible | 13:51 | |
| odyssey4me | evrardjp I can do a quick hack to unblock gates while you work on something better - but that'll only be worth it if your patch is still going to take another day or so. | 13:52 |
| *** nshetty__ has joined #openstack-ansible | 13:52 | |
| *** smatzek has joined #openstack-ansible | 13:52 | |
| cloudnull | mornings | 13:53 |
| *** smatzek has quit IRC | 13:53 | |
| *** smatzek has joined #openstack-ansible | 13:54 | |
| *** smatzek has quit IRC | 13:54 | |
| *** nshetty_ has quit IRC | 13:54 | |
| *** smatzek has joined #openstack-ansible | 13:55 | |
| *** nshetty__ has quit IRC | 13:55 | |
| *** nshetty__ has joined #openstack-ansible | 13:55 | |
| *** PaulusNL has quit IRC | 13:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_aodh stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/514748 | 13:56 |
| evrardjp | odyssey4me: don't worry I will make that patch | 13:57 |
| *** mrch has quit IRC | 13:57 | |
| *** jwitko has joined #openstack-ansible | 13:59 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 14:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 14:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 14:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 14:07 |
| *** mbuil has quit IRC | 14:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/515082 | 14:13 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/515082 | 14:15 |
| *** a0x3e8 has joined #openstack-ansible | 14:16 | |
| *** SerenaFeng has quit IRC | 14:17 | |
| *** a0x3e8 has quit IRC | 14:17 | |
| admin0 | hi all . how to fix all routers going into standby ? | 14:19 |
| admin0 | is there a way to make 1 master ? | 14:19 |
| *** BjoernT has joined #openstack-ansible | 14:21 | |
| *** udesale has quit IRC | 14:26 | |
| *** acormier has joined #openstack-ansible | 14:26 | |
| *** acormier has quit IRC | 14:27 | |
| *** acormier has joined #openstack-ansible | 14:28 | |
| *** yolanda has quit IRC | 14:28 | |
| *** udesale has joined #openstack-ansible | 14:30 | |
| *** acormier has quit IRC | 14:31 | |
| mgariepy | admin0, how many neutron l3 agent do you have ? | 14:36 |
| admin0 | i see 4 but all on standby | 14:37 |
| admin0 | tried to remove from 3, restart l3 on the remaining one, but it did not helped .. setting it to down, set mode to ha=false helped, but its now a single point of failure .. maybe there is a way in ansible to lock it to 3 max or 2 max | 14:38 |
| mgariepy | like : max_l3_agents_per_router = 2 | 14:39 |
| mgariepy | admin0, https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/neutron.conf.j2#L73 | 14:40 |
| mgariepy | admin0, i would suggest to try i at 2. | 14:41 |
| admin0 | if i have set all my compute nodes to also be network agents, so say 10 compute, 10 l3, how big do they grow ? | 14:41 |
| *** TxGirlGeek has joined #openstack-ansible | 14:41 | |
| admin0 | set max_l3_router = num_agent if num_agent > 2 -- means 1 network can also be in 10 nodes ? | 14:41 |
| *** gouthamr has joined #openstack-ansible | 14:41 | |
| mgariepy | admin0, i was having it to 3 on mitaka but it was causing issue. | 14:42 |
| mgariepy | it took very long to stabilize. so I switched back to no-ha :) | 14:43 |
| *** dxiri has joined #openstack-ansible | 14:45 | |
| admin0 | no-ha is a spof that boss finds hard to accept, ha does not work :D | 14:45 |
| jafeha | is there any dev here who was involved in implementing ovs in osa? | 14:45 |
| *** woodard has joined #openstack-ansible | 14:45 | |
| admin0 | jafeha, i have ovs working in osa, but i am not a dev | 14:45 |
| mgariepy | hmm. i'm a bit confused about the logic in :https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/neutron.conf.j2#L68 | 14:47 |
| jafeha | admin0: could you have a look at your config and tell me "how" you've set the pattern in your user_variables.yml file? i had it first configured including the "=" but ran into issues. removing the character seems to work | 14:48 |
| mgariepy | if there is only 1 agent it's set to 2 ? that is not sounding right haha | 14:48 |
| jafeha | this might be bug in either the documentation or the check itself. i'd say the documentation. | 14:49 |
| admin0 | jafeha, i have no idea what you are refering to | 14:50 |
| jafeha | according the documentation there should be a config line "pattern: "CONFIG_OPENVSWITCH="" | 14:52 |
| jafeha | https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-openvswitch.html | 14:52 |
| admin0 | aha | 14:53 |
| admin0 | hold | 14:53 |
| jafeha | tnx | 14:53 |
| admin0 | pattern: "CONFIG_OPENVSWITCH=" | 14:53 |
| admin0 | let me pastebin | 14:53 |
| jafeha | strange | 14:53 |
| odyssey4me | admin0 so you have OVS in your deployment? | 14:54 |
| admin0 | yes | 14:54 |
| admin0 | with some admin0 hacks and workarounds | 14:54 |
| *** jvidal has quit IRC | 14:54 | |
| odyssey4me | any chance that you could put together a patch for the deploy guide to give a full example configuration? In the format given in https://docs.openstack.org/project-deploy-guide/openstack-ansible/draft/app-config-test.html | 14:54 |
| admin0 | but the core networking is in ovs | 14:54 |
| *** br0k3n1 has quit IRC | 14:55 | |
| evrardjp | after a very long work on testing rpm cli, I realized that rpm_key is properly idempotent with the same commands I was using. Time loss: 1 hour. :/ | 14:56 |
| evrardjp | at least I know how it works. | 14:56 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible master: Integrate Congress with OSA. https://review.openstack.org/503971 | 14:56 |
| odyssey4me | hwoarang evrardjp FYI apparently there's a patch that's merged to solve the log issue - they're restarting zuul now, after which we should be able to test our patch again | 14:58 |
| evrardjp | odyssey4me: the issue you reported was for memcached_server right? | 14:58 |
| odyssey4me | please hold - your logs are important to us | 14:58 |
| evrardjp | great | 14:58 |
| odyssey4me | evrardjp the issue was for openstack-ansible-tests | 14:58 |
| odyssey4me | the memcache_server role tests fine, because it doesn't use the openstack repositories | 14:58 |
| evrardjp | let me check where is the key managed there | 14:58 |
| odyssey4me | evrardjp the key is managed in openstack_hosts and pip_install | 14:58 |
| evrardjp | yeah ok | 14:59 |
| evrardjp | but that was memcached_server that failed then? | 14:59 |
| evrardjp | if that, it means I have to patch hosts more urgently | 14:59 |
| evrardjp | then pip install | 14:59 |
| odyssey4me | yeah, memcache failed to install from the RDO repo because the key was improperly imported | 14:59 |
| odyssey4me | yes, hosts and pip install - with that we should be good | 15:00 |
| *** weezS has joined #openstack-ansible | 15:00 | |
| *** BjoernT has quit IRC | 15:02 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts master: Vendor in the RDO GPG keys to install https://review.openstack.org/515101 | 15:02 |
| tnogisto | oki, got my OVS to work, is it correct that openstack-ansible deployed console doesnt work out of the box? | 15:03 |
| *** snowman4839 has joined #openstack-ansible | 15:03 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts master: Vendor in the RDO GPG keys to install https://review.openstack.org/515101 | 15:03 |
| admin0 | odyssey4me, my workaround was due to setup-hosts not working due to the script expecting linuxbride, i run all network containers on metal on some of the hosts | 15:04 |
| admin0 | so in my case, compute node == network node | 15:04 |
| jafeha | this is really strange. the openstack_kernel_modules.yml explicitly checks for '=y' or '=m' for loaded modules. https://github.com/openstack/openstack-ansible-openstack_hosts/blob/master/tasks/openstack_kernel_modules.yml | 15:04 |
| jafeha | but anyway, i found a workaround for the moment | 15:04 |
| jafeha | odyssey4me: is there still someone around who i could ask abount the openstack_kernel_modules.yml code? | 15:05 |
| *** woodard has quit IRC | 15:06 | |
| admin0 | jafeha, https://pastebin.com/yZGS1dph | 15:06 |
| *** woodard has joined #openstack-ansible | 15:06 | |
| admin0 | jafeha, that is for OVS on the user_variables | 15:07 |
| jafeha | admin0: thank you! i'll have a closer look and check my config | 15:07 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-pip_install master: Vendor in the RDO GPG keys to install https://review.openstack.org/515103 | 15:07 |
| admin0 | odyssey4me, i am currently redoing this platform in pike .. i will paste the full configs in that same manner for ovs | 15:08 |
| admin0 | but give me some days for that | 15:08 |
| odyssey4me | jafeha that stuff was implemented by some intel guys some time back - unfortunately they're not around any more | 15:12 |
| odyssey4me | jafeha it'd be better to just ask the questions to the channel, and hopefully someone can respond... we have some people around who know there way around kernel configs here and there | 15:12 |
| *** Oku_OS is now known as Oku_OS-away | 15:12 | |
| evrardjp | jafeha: what's your issue? | 15:13 |
| odyssey4me | tnogisto it does work out the box - what issue are you seeing? bear in mind that by default you're using a self-signed cert, so your browser might block the wss being tunnelled through it | 15:13 |
| evrardjp | I am refactoring the openstack_hosts role right now, and I refactored those taks. | 15:13 |
| odyssey4me | evrardjp posed a question in https://review.openstack.org/#/c/515101/2 and a suggestion | 15:14 |
| evrardjp | answered the question... the suggestion is duely noted: it will be done differently with my patch | 15:16 |
| evrardjp | basically handling distros the same way | 15:16 |
| evrardjp | I was not ready to go on a crusade for changing those modules. | 15:17 |
| evrardjp | maybe I should just write a quick action plugin and vendor it in too, but I have the feeling we would increase our tech debt | 15:17 |
| tnogisto | ya, it does work, but haproxy thinks its backend should be https, but spice thinks it wants to serve http. | 15:17 |
| evrardjp | that's probably a variable to change tnogisto | 15:18 |
| tnogisto | also horizon links the https site as http | 15:18 |
| tnogisto | true | 15:18 |
| evrardjp | is there no docs for that? | 15:18 |
| tnogisto | I was initialy confused by the behaviour :) | 15:19 |
| odyssey4me | tnogisto that doesn't sound normal | 15:19 |
| evrardjp | true that! | 15:19 |
| odyssey4me | evrardjp another question in the same patch - thanks :) | 15:19 |
| tnogisto | now I know what causes it and will fix it soon to work properly, sorry to bother you guys :) | 15:19 |
| tnogisto | I can show you the conf, but thats how it worked out of the box, didnt make any special changes to it yet. | 15:21 |
| mhayden | xgerman_: do you have time to get openstack-ansible-os_octavia working with zuulv3 | 15:21 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Move log collection to 'post' stage https://review.openstack.org/513711 | 15:22 |
| odyssey4me | mhayden there's already a zuulv3 patch up - but it seems the functional test is failing | 15:23 |
| evrardjp | mhayden: opinion on the location of gpg keys? We'd need an experienced centos admin: https://review.openstack.org/#/c/515101/2 | 15:24 |
| *** nshetty__ has quit IRC | 15:24 | |
| mhayden | evrardjp: i'd suggest /tmp since it's best to let rpm put the files down witht he right permissions and SELinux contexts | 15:25 |
| evrardjp | yeah that's what I thought first | 15:25 |
| evrardjp | tmp being generally a noexec place and all | 15:25 |
| * mhayden commented | 15:25 | |
| evrardjp | thanks | 15:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Move log collection to 'post' stage https://review.openstack.org/513711 | 15:29 |
| jafeha | odyssey4me: thank you | 15:29 |
| odyssey4me | evrardjp mhayden thanks! | 15:31 |
| evrardjp | thank you! | 15:32 |
| evrardjp | with you in capital letters | 15:32 |
| *** galstrom is now known as galstrom_zzz | 15:33 | |
| jafeha | evrardjp: the kernel module check does not work for me when i enable the CONFIG_OPENVSWITCH kernel module as written in the documentation. | 15:34 |
| *** thorst_ has quit IRC | 15:34 | |
| *** rodolof has quit IRC | 15:34 | |
| evrardjp | ok | 15:35 |
| *** thorst has joined #openstack-ansible | 15:35 | |
| evrardjp | so you mean the pattern is wrong | 15:35 |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible master: Add Neutron variables shared with Octavia https://review.openstack.org/515110 | 15:35 |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Improve event streaming with a better Octavia verison https://review.openstack.org/514852 | 15:36 |
| jafeha | evrardjp: there is a pattern check in the code for "$MODULE" the documentation refers to "$MODULE=" | 15:37 |
| jafeha | evrardjp: for admin0 works the option "pattern: "CONFIG_OPENVSWITCH="" for me works the option "pattern: "CONFIG_OPENVSWITCH"" ;) | 15:38 |
| evrardjp | yeah makes sense | 15:38 |
| evrardjp | that is a doc change I guess :) | 15:38 |
| evrardjp | could you patch it? | 15:38 |
| evrardjp | that would be nice | 15:38 |
| admin0 | jafeha, if both works, whats the one to use :) | 15:39 |
| jafeha | thats the question :D | 15:39 |
| *** thorst has quit IRC | 15:39 | |
| jafeha | other implementations refer to "$MODULE" https://docs.openstack.org/openstack-ansible-openstack_hosts/latest/ | 15:39 |
| jafeha | i've never made a doc patch, but if we agree to the change, i'll post a patch tomorrow | 15:41 |
| admin0 | so modules and also config_openvswitch has multiple working examples :D | 15:42 |
| jafeha | the thing is: i do not fully understand the checks in the openstack_kernel_modules.yml. admin0 might have passed another condition? https://github.com/openstack/openstack-ansible-openstack_hosts/blob/master/tasks/openstack_kernel_modules.yml | 15:43 |
| *** jwitko has quit IRC | 15:44 | |
| admin0 | jafeha, me too .. i came here, asked around, tinkered with it, it worked .. moved on to do something else :D | 15:44 |
| *** jwitko has joined #openstack-ansible | 15:44 | |
| jafeha | there are four when clauses, two check for "=y" and "=m" | 15:45 |
| admin0 | thing is, OVS example/usage is less with osa .. so there is no fixed "this is how we are going to do it" road (yet) | 15:46 |
| *** smatzek_ has joined #openstack-ansible | 15:46 | |
| admin0 | if they decide to gate check in both ovs/lb - i am sure a lot of things will improve | 15:47 |
| jafeha | admin0: as we're deploying it, we might be able to help | 15:47 |
| admin0 | yes .. but we are ending up in 2 configs .. lets sit down for a week and see how many more we can gather | 15:47 |
| admin0 | after 4 hours will be more people here ( from the US ) whom also have some ovs-osa deployment | 15:48 |
| admin0 | can ask around for the config they use | 15:48 |
| jafeha | sure. there seems to be much work necessary for the "default" ovs deployment, but my config is not that different in the end | 15:48 |
| *** pcaruana has quit IRC | 15:49 | |
| jafeha | sun sets in berlin, not sure how long i'll be around today ;) | 15:49 |
| *** smatzek has quit IRC | 15:49 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Run security hardening when leaping a deployment https://review.openstack.org/515114 | 15:50 |
| mhayden | evrardjp: looks like https://review.openstack.org/515101 needs a small adjustment | 15:54 |
| evrardjp | oh let me have a look | 15:54 |
| mhayden | just missing an underscore on _add_yum_keys | 15:54 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts master: Vendor in the RDO GPG keys to install https://review.openstack.org/515101 | 15:55 |
| evrardjp | very silly | 15:55 |
| evrardjp | yeah | 15:55 |
| mhayden | evrardjp: i still think you're pretty cool | 15:55 |
| mhayden | regardless of what odyssey4me says | 15:55 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-pip_install master: Vendor in the RDO GPG keys to install https://review.openstack.org/515103 | 15:55 |
| evrardjp | lol | 15:55 |
| mhayden | evrardjp: this one has the same error -> https://review.openstack.org/515103 | 15:56 |
| mhayden | bam! | 15:56 |
| evrardjp | when did he say I am not cool? | 15:56 |
| *** huxinhui_ has quit IRC | 15:56 | |
| evrardjp | :p | 15:56 |
| admin0 | so i create some instances using heat :) .. i pre-specify the ssh key .. does not work .. i rebuild the instances again using rebuild ( and not delete the stack) .. ssh keys work | 15:56 |
| * mhayden whistles | 15:56 | |
| * evrardjp brings forks | 15:56 | |
| evrardjp | :) | 15:56 |
| mhayden | odyssey4me: feel free to re-deuce https://review.openstack.org/515103 and https://review.openstack.org/515101 | 15:56 |
| *** weezS has quit IRC | 15:57 | |
| mhayden | cloudnull: isn't this run as part of `setup-hosts.yml` in newton? https://review.openstack.org/#/c/515114/ | 15:57 |
| cloudnull | it is however we're not running setup-hosts in leap | 15:58 |
| mhayden | oh, i see right above there | 15:58 |
| *** weezS has joined #openstack-ansible | 15:58 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/newton: Remove unified queue from gate pipeline https://review.openstack.org/514724 | 15:58 |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening master: Docs/Reno fixes https://review.openstack.org/514790 | 15:59 |
| *** mpranjic has joined #openstack-ansible | 16:01 | |
| *** wspthr has joined #openstack-ansible | 16:02 | |
| *** Miouge has quit IRC | 16:05 | |
| *** wspthr has left #openstack-ansible | 16:15 | |
| *** wspthr has joined #openstack-ansible | 16:15 | |
| evrardjp | odyssey4me: you wanted to vendor in the uca jeys too? | 16:23 |
| evrardjp | keys* | 16:23 |
| evrardjp | that's a weird lapsus. | 16:23 |
| odyssey4me | evrardjp if we're going to adopt this (which is good for offline installs), we should do it everywhere | 16:23 |
| evrardjp | k | 16:23 |
| evrardjp | good | 16:23 |
| odyssey4me | rabbitmq, mariadb, uca, etc | 16:23 |
| evrardjp | will do | 16:23 |
| evrardjp | yeah that's included in my patch :p | 16:23 |
| evrardjp | well | 16:24 |
| evrardjp | not 100% | 16:24 |
| evrardjp | but | 16:24 |
| evrardjp | it can be an override | 16:24 |
| odyssey4me | it should be our default | 16:24 |
| odyssey4me | IMO | 16:24 |
| evrardjp | why should we ship all the keys for all the containers? | 16:24 |
| evrardjp | I understand shipping uca everywhere | 16:24 |
| evrardjp | but mariadb and rabbitmq keys... meh | 16:24 |
| evrardjp | well why not... | 16:25 |
| odyssey4me | they're all reaching out to the internet, so they're all a risk of failure for the same reasn | 16:25 |
| evrardjp | oh yeah, I thought you meant shipping to all the containers all the keys | 16:26 |
| odyssey4me | oh no | 16:27 |
| *** thorst has joined #openstack-ansible | 16:27 | |
| odyssey4me | just each role that adds a repo, should add the key for the repo | 16:27 |
| evrardjp | yeah ok | 16:27 |
| evrardjp | in the same way | 16:27 |
| odyssey4me | yup | 16:27 |
| evrardjp | sounds good | 16:27 |
| evrardjp | I may even have a good idea for once | 16:28 |
| odyssey4me | you have many good dieas | 16:28 |
| evrardjp | well let's see what it gives in actual code | 16:28 |
| *** armaan has joined #openstack-ansible | 16:29 | |
| *** maybebuggy has quit IRC | 16:34 | |
| *** udesale has quit IRC | 16:35 | |
| *** DanyC has quit IRC | 16:39 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Correct regex for disk size https://review.openstack.org/514996 | 16:50 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Add ceph-ansible v3 support https://review.openstack.org/512655 | 16:50 |
| openstackgerrit | Byron McCollum proposed openstack/openstack-ansible-os_horizon master: IMAGES_ALLOW_LOCATION Wrongly Templated As String https://review.openstack.org/515133 | 16:55 |
| *** snowman4839 has quit IRC | 16:58 | |
| *** snowman4839 has joined #openstack-ansible | 16:59 | |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening master: Docs/Reno fixes https://review.openstack.org/514790 | 17:00 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Hide yum-config-manager output https://review.openstack.org/514835 | 17:01 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Update number of workers to optimize gate memory https://review.openstack.org/511051 | 17:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Move log collection to 'post' stage https://review.openstack.org/513711 | 17:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Move log collection to 'post' stage https://review.openstack.org/513711 | 17:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Correct regex for disk size https://review.openstack.org/515138 | 17:06 |
| *** phalmos has joined #openstack-ansible | 17:20 | |
| *** smatzek_ is now known as smatzek | 17:21 | |
| *** dmsimard is now known as dmsimard|afk | 17:23 | |
| *** vnogin has quit IRC | 17:24 | |
| *** phalmos_ has joined #openstack-ansible | 17:25 | |
| *** phalmos has quit IRC | 17:27 | |
| *** germs has joined #openstack-ansible | 17:34 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova stable/newton: Ensure nova is started after everything else https://review.openstack.org/515145 | 17:34 |
| *** germs1 has joined #openstack-ansible | 17:36 | |
| *** germs has quit IRC | 17:38 | |
| *** TxGirlGeek has quit IRC | 17:43 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install master: Vendor in the RDO GPG keys to install https://review.openstack.org/515103 | 17:45 |
| *** TxGirlGeek has joined #openstack-ansible | 17:49 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Run security hardening when leaping a deployment https://review.openstack.org/515114 | 17:50 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Vendor in the RDO GPG keys to install https://review.openstack.org/515101 | 17:53 |
| *** gkadam has quit IRC | 17:54 | |
| *** poopcat has joined #openstack-ansible | 17:57 | |
| *** electrofelix has quit IRC | 17:57 | |
| *** smatzek_ has joined #openstack-ansible | 17:59 | |
| *** smatzek has quit IRC | 18:02 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-openstack_hosts stable/pike: Vendor in the RDO GPG keys to install https://review.openstack.org/515154 | 18:08 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-pip_install stable/pike: Vendor in the RDO GPG keys to install https://review.openstack.org/515155 | 18:08 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible master: Integrate Congress with OSA. https://review.openstack.org/503971 | 18:13 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests stable/pike: Hide yum-config-manager output https://review.openstack.org/514837 | 18:14 |
| *** chhavi has quit IRC | 18:17 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Hide yum-config-manager output https://review.openstack.org/514835 | 18:29 |
| *** stuartgr has quit IRC | 18:35 | |
| *** dxiri_ has joined #openstack-ansible | 18:35 | |
| *** tasker has quit IRC | 18:38 | |
| *** dxiri has quit IRC | 18:38 | |
| *** dxiri_ has quit IRC | 18:40 | |
| *** yolanda has joined #openstack-ansible | 18:40 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests stable/pike: Add fedora to run_tests_common.sh https://review.openstack.org/514711 | 18:45 |
| *** newmember has joined #openstack-ansible | 18:50 | |
| xgerman_ | this confuses me: http://logs.openstack.org/67/514767/3/check/openstack-ansible-functional-ubuntu-xenial/a4ca403/job-output.txt.gz#_2017-10-24_19_41_06_106126 I am shutting down the bridge but then it’s still up. Did something change with the test code? | 19:02 |
| *** vnogin has joined #openstack-ansible | 19:11 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova stable/newton: Ensure nova is started after everything else https://review.openstack.org/515145 | 19:13 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova stable/newton: Change checkout for nova-lxd to use the EOL tag https://review.openstack.org/515178 | 19:13 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Correct regex for disk size https://review.openstack.org/515138 | 19:13 |
| *** mummer has joined #openstack-ansible | 19:16 | |
| mhayden | yay | 19:21 |
| mgariepy | mhayden, is the STIG stuff talks about KSM anywhere ? | 19:23 |
| mhayden | kvm shared memory? | 19:24 |
| mhayden | i don't believe so | 19:24 |
| mgariepy | yep | 19:24 |
| mhayden | not sure if i would use KSM on a multi-tenant env | 19:24 |
| mgariepy | it's active by default. | 19:24 |
| mhayden | or a multi security level env | 19:24 |
| mhayden | hah | 19:24 |
| mgariepy | at least on ubuntu | 19:24 |
| mhayden | i thought you had to run the ksmtuned daemon for it | 19:24 |
| mhayden | oh | 19:24 |
| mhayden | it's off by default on centos | 19:24 |
| mgariepy | http://paste.openstack.org/show/624642/ | 19:25 |
| mgariepy | qemu-kvm set it to auto on 16.04. | 19:26 |
| mgariepy | hrm (old host ) | 19:26 |
| mhayden | interesting | 19:26 |
| mgariepy | sadly. | 19:26 |
| mhayden | ubuntu continues to surprise me | 19:26 |
| spotz | weird one is on by default and the other off by default | 19:26 |
| mgariepy | hwoarang, what's about opensuse ? | 19:27 |
| mgariepy | that's the things with defaults .. | 19:27 |
| mhayden | dang, looks like evrardjp's GPG key patches weren't enough | 19:27 |
| evrardjp | ? | 19:27 |
| evrardjp | what did I miss? | 19:28 |
| mhayden | http://logs.openstack.org/35/514835/3/check/openstack-ansible-functional-centos-7/a080e79/job-output.txt.gz#_2017-10-25_19_01_20_381416 | 19:28 |
| *** woodard has quit IRC | 19:28 | |
| * mhayden builds an AIO to dig into it | 19:29 | |
| *** marst has joined #openstack-ansible | 19:32 | |
| xgerman_ | so when I try to run the functional test on my own vm with tox -efunctional it works — but on the zuul host it blows up bringing up the br-lbaas interface… | 19:35 |
| xgerman_ | very confused… | 19:36 |
| evrardjp | mhayden: does memcached need another key? | 19:37 |
| mhayden | evrardjp: firing up a AIO | 19:38 |
| evrardjp | mhayden: oh no | 19:38 |
| evrardjp | no need | 19:38 |
| evrardjp | :) | 19:38 |
| evrardjp | got it | 19:38 |
| mhayden | hah, okay | 19:38 |
| evrardjp | well I think | 19:38 |
| evrardjp | https://github.com/openstack/openstack-ansible-memcached_server/blob/master/meta/main.yml#L38-L40 | 19:38 |
| evrardjp | no pip install in deps | 19:39 |
| evrardjp | so we rely on openstack_hosts | 19:39 |
| mhayden | OOOOOH | 19:39 |
| *** TxGirlGeek has quit IRC | 19:39 | |
| evrardjp | but check the host you're targetting | 19:39 |
| mhayden | didn't we change that to include_role somewhere? | 19:39 |
| evrardjp | it's not the host | 19:39 |
| evrardjp | mhayden: I am busy doing that | 19:39 |
| evrardjp | maybe I missed it? | 19:39 |
| evrardjp | well I am completely refactoring openstack_hosts role right now :p | 19:39 |
| evrardjp | but in the meantime | 19:39 |
| evrardjp | we should know why the container didn't inherit from the host | 19:40 |
| evrardjp | ok let me check real quick then | 19:41 |
| evrardjp | if you have a host that should be easier to find out | 19:42 |
| *** pbandark has quit IRC | 19:42 | |
| evrardjp | https://github.com/openstack/openstack-ansible-lxc_hosts/blob/master/vars/redhat-7.yml#L61 | 19:42 |
| evrardjp | so | 19:42 |
| evrardjp | what Jesse said was correct. | 19:42 |
| evrardjp | we should maybe have put the file there | 19:43 |
| mhayden | but pip_install should set that up properly | 19:43 |
| evrardjp | pip install adds the key properly on the lxc_host | 19:43 |
| mhayden | it won't be right in the container cache, but it will b eright when pip_install comes along | 19:43 |
| evrardjp | which is fine | 19:43 |
| mhayden | which i guess is why memcached is broken | 19:44 |
| evrardjp | mhayden: that's what I said earlier | 19:44 |
| evrardjp | memcached doesn't need pip | 19:44 |
| mhayden | yup | 19:44 |
| mhayden | true | 19:44 |
| evrardjp | so we don't have that | 19:44 |
| evrardjp | so we should rely on the base container that behaves properly | 19:44 |
| mhayden | we could try to match the permissions of other gpg keys in there | 19:44 |
| evrardjp | OR alternatively | 19:44 |
| mhayden | and run restorecon on the roles to make sure selinux contexts are good | 19:44 |
| evrardjp | add the key into memcached | 19:44 |
| evrardjp | OR | 19:44 |
| evrardjp | alternatively | 19:44 |
| evrardjp | wait for my next commit | 19:44 |
| * mhayden waits | 19:45 | |
| evrardjp | to openstack_hosts :p | 19:45 |
| evrardjp | it's not ready and it's late | 19:45 |
| evrardjp | what do you think should we do in the meantime? | 19:45 |
| evrardjp | I'd say move the key into that location | 19:45 |
| mhayden | seems okay | 19:45 |
| mhayden | want me to do it? in a meeting for the next 15 min | 19:45 |
| evrardjp | well you can do it after your meeting | 19:46 |
| evrardjp | :) | 19:46 |
| mhayden | okay, can do | 19:46 |
| evrardjp | I'd be happy to see a patch indeed, because I want to stop working for today :D | 19:46 |
| mhayden | go enjoy a beer sir | 19:46 |
| evrardjp | but I will work on this refactor tomorrow | 19:46 |
| evrardjp | in the meantime, all you do there with the proper permissions and selinux context is good, because I can reuse it | 19:47 |
| spotz | later evrardjp | 19:48 |
| evrardjp | see you later spotz! | 19:48 |
| *** woodard has joined #openstack-ansible | 20:01 | |
| mhayden | so it looks like ansible *thinks* that 764429E6 is in place, but it isn't | 20:02 |
| *** Miouge has joined #openstack-ansible | 20:08 | |
| *** hachi_ has joined #openstack-ansible | 20:09 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:11 | |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible-os_cinder master: Fix indentation of container_vars for ceph doc https://review.openstack.org/514977 | 20:12 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-memcached_server master: Apply pip_install to memcached role https://review.openstack.org/515194 | 20:14 |
| *** phalmos_ has quit IRC | 20:15 | |
| *** acormier has joined #openstack-ansible | 20:19 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Hide yum-config-manager output https://review.openstack.org/514835 | 20:20 |
| *** smatzek_ has quit IRC | 20:21 | |
| *** smatzek has joined #openstack-ansible | 20:22 | |
| *** acormier has quit IRC | 20:25 | |
| *** smatzek has quit IRC | 20:26 | |
| *** thorst has quit IRC | 20:27 | |
| *** vnogin has quit IRC | 20:37 | |
| *** thorst has joined #openstack-ansible | 20:48 | |
| *** thorst has quit IRC | 20:48 | |
| *** thorst has joined #openstack-ansible | 20:49 | |
| mhayden | evrardjp: this fixes it --> https://review.openstack.org/515194 (for your review tomorrow) | 20:49 |
| mhayden | although i don't like applying pip_install to memcached, it does work | 20:49 |
| mhayden | and brings that container in line with the others | 20:50 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-memcached_server stable/pike: Apply pip_install to memcached role https://review.openstack.org/515205 | 20:50 |
| evrardjp[m] | yes, that's an alternative | 20:50 |
| evrardjp[m] | wasn't it possible to move the gpg file instead ? | 20:50 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests stable/pike: Hide yum-config-manager output https://review.openstack.org/514837 | 20:51 |
| *** masuberu has quit IRC | 20:51 | |
| mhayden | evrardjp[m]: that is a possibility | 20:52 |
| *** thorst has quit IRC | 20:53 | |
| evrardjp[m] | sounded less terrible than adding pip install but.. meh ;) fine for me for now... | 20:53 |
| evrardjp[m] | i was worried about selinux contexts | 20:54 |
| *** yolanda has quit IRC | 20:56 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-openstack_hosts master: Copy RPM keys into correct place https://review.openstack.org/515206 | 21:00 |
| *** Miouge has quit IRC | 21:00 | |
| *** vnogin has joined #openstack-ansible | 21:01 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/newton: use Newton sha's for cross repo versions https://review.openstack.org/515207 | 21:02 |
| *** mummer has quit IRC | 21:02 | |
| *** hachi_ has quit IRC | 21:03 | |
| *** mummer has joined #openstack-ansible | 21:03 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-pip_install master: Copy RPM keys into correct place https://review.openstack.org/515208 | 21:06 |
| *** dmsimard|afk is now known as dmsimard | 21:08 | |
| *** mummer has quit IRC | 21:10 | |
| *** mummer has joined #openstack-ansible | 21:13 | |
| *** TxGirlGeek has quit IRC | 21:15 | |
| *** jwitko has quit IRC | 21:21 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 21:24 |
| *** mwynne has quit IRC | 21:28 | |
| *** esberglu has quit IRC | 21:29 | |
| *** mwynne has joined #openstack-ansible | 21:30 | |
| openstackgerrit | Victor Morales proposed openstack/openstack-ansible master: Add proxy environment vars to boostrap-aio.yml https://review.openstack.org/515214 | 21:31 |
| *** esberglu has joined #openstack-ansible | 21:43 | |
| *** esberglu has quit IRC | 21:48 | |
| *** weezS has quit IRC | 21:54 | |
| *** marst has quit IRC | 21:58 | |
| openstackgerrit | Frank Zhang proposed openstack/openstack-ansible-ops master: Automate Elasticsearch upgrade in leap https://review.openstack.org/515219 | 22:08 |
| *** TxGirlGeek has joined #openstack-ansible | 22:08 | |
| *** smatzek has joined #openstack-ansible | 22:10 | |
| *** smatzek has quit IRC | 22:15 | |
| *** weezS has joined #openstack-ansible | 22:21 | |
| *** mwynne has quit IRC | 22:29 | |
| *** pbandark has joined #openstack-ansible | 22:31 | |
| *** pbandark has quit IRC | 22:33 | |
| *** thorst has joined #openstack-ansible | 22:35 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 22:40 |
| *** thorst has quit IRC | 22:42 | |
| *** pbandark has joined #openstack-ansible | 22:42 | |
| *** thorst has joined #openstack-ansible | 22:43 | |
| *** thorst has quit IRC | 22:46 | |
| *** pmannidi has joined #openstack-ansible | 22:50 | |
| *** dxiri has joined #openstack-ansible | 22:57 | |
| *** dxiri has quit IRC | 22:57 | |
| *** dxiri has joined #openstack-ansible | 22:58 | |
| *** alanmeadows has quit IRC | 23:03 | |
| *** alanmeadows has joined #openstack-ansible | 23:03 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 23:05 |
| *** masuberu has joined #openstack-ansible | 23:05 | |
| *** weezS has quit IRC | 23:08 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_freezer master: Updated from global requirements https://review.openstack.org/514258 | 23:08 |
| *** gouthamr has quit IRC | 23:22 | |
| *** gouthamr has joined #openstack-ansible | 23:22 | |
| *** TxGirlGeek has quit IRC | 23:23 | |
| *** thorst has joined #openstack-ansible | 23:45 | |
| *** thorst has quit IRC | 23:46 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
