| *** smatzek has quit IRC | 00:01 | |
| *** markvoelker_ has quit IRC | 00:03 | |
| *** markvoelker has joined #openstack-ansible | 00:09 | |
| *** markvoelker has quit IRC | 00:13 | |
| *** jwitko has quit IRC | 00:16 | |
| *** jwitko has joined #openstack-ansible | 00:16 | |
| *** markvoelker has joined #openstack-ansible | 00:18 | |
| *** gouthamr has joined #openstack-ansible | 00:20 | |
| *** markvoelker has quit IRC | 00:22 | |
| *** markvoelker has joined #openstack-ansible | 00:27 | |
| *** thorst has joined #openstack-ansible | 00:31 | |
| cloudnull | snowman4839: you can run the lxc-container-destroy.yml playbook | 00:31 |
|---|---|---|
| cloudnull | then remove the /etc/openstack_deploy/openstack-inventory.json file | 00:32 |
| cloudnull | then its back to vanilla. | 00:32 |
| *** markvoelker has quit IRC | 00:32 | |
| snowman4839 | but doesn't that destroy all containers? | 00:34 |
| snowman4839 | lets just say I borked galera, if I destroy the galera containers and them remove all mentions of it in inventory, is that sufficent for it to just act like it was never there? | 00:34 |
| *** kiennt26 has joined #openstack-ansible | 00:40 | |
| *** duonghq has joined #openstack-ansible | 00:45 | |
| *** vnogin has joined #openstack-ansible | 00:52 | |
| *** markvoelker has joined #openstack-ansible | 00:54 | |
| *** woodard_ has joined #openstack-ansible | 00:55 | |
| *** woodard has quit IRC | 00:55 | |
| *** thorst has quit IRC | 00:57 | |
| *** vnogin has quit IRC | 00:57 | |
| *** chyka has joined #openstack-ansible | 00:58 | |
| *** markvoelker has quit IRC | 00:59 | |
| *** masuberu has joined #openstack-ansible | 01:00 | |
| *** chyka has quit IRC | 01:02 | |
| *** masber has quit IRC | 01:04 | |
| *** drifterza has quit IRC | 01:06 | |
| *** markvoelker has joined #openstack-ansible | 01:12 | |
| *** thorst has joined #openstack-ansible | 01:23 | |
| *** drifterza has joined #openstack-ansible | 01:27 | |
| *** thorst has quit IRC | 01:36 | |
| *** masber has joined #openstack-ansible | 01:44 | |
| *** masuberu has quit IRC | 01:47 | |
| *** markvoelker has quit IRC | 01:47 | |
| *** vnogin has joined #openstack-ansible | 01:53 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Add openstack-ansible-linters test https://review.openstack.org/513110 | 01:54 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: Regenerate haproxy.cfg on base config changes https://review.openstack.org/511046 | 01:54 |
| *** vnogin has quit IRC | 01:58 | |
| *** markvoelker has joined #openstack-ansible | 02:00 | |
| *** markvoelker has quit IRC | 02:05 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Update all SHAs https://review.openstack.org/512037 | 02:05 |
| *** markvoelker has joined #openstack-ansible | 02:10 | |
| *** markvoelker has quit IRC | 02:14 | |
| *** hw_wutianwei has joined #openstack-ansible | 02:17 | |
| *** masber has quit IRC | 02:19 | |
| *** markvoelker has joined #openstack-ansible | 02:19 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Add openstack-ansible-linters test https://review.openstack.org/513111 | 02:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Add openstack-ansible-linters test https://review.openstack.org/513113 | 02:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Add openstack-ansible-linters test https://review.openstack.org/513114 | 02:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Add openstack-ansible-linters test https://review.openstack.org/513109 | 02:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat master: Add openstack-ansible-linters test https://review.openstack.org/513116 | 02:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Bypass web server during service setup https://review.openstack.org/510633 | 02:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Add openstack-ansible-linters test https://review.openstack.org/513117 | 02:22 |
| *** markvoelker has quit IRC | 02:23 | |
| *** markvoelker has joined #openstack-ansible | 02:28 | |
| *** markvoelker has quit IRC | 02:33 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Update readme to use force flag instead https://review.openstack.org/512423 | 02:35 |
| *** thorst has joined #openstack-ansible | 02:37 | |
| *** thorst has quit IRC | 02:42 | |
| *** markvoelker has joined #openstack-ansible | 02:46 | |
| *** drifterza has quit IRC | 02:48 | |
| *** jwitko has quit IRC | 02:50 | |
| *** jwitko has joined #openstack-ansible | 02:50 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Fix Ceilometer deployments https://review.openstack.org/513229 | 02:51 |
| *** vnogin has joined #openstack-ansible | 02:54 | |
| *** gouthamr has quit IRC | 02:57 | |
| *** vnogin has quit IRC | 02:59 | |
| *** markvoelker has quit IRC | 03:20 | |
| *** gkadam has joined #openstack-ansible | 03:23 | |
| *** markvoelker has joined #openstack-ansible | 03:25 | |
| *** markvoelker has quit IRC | 03:30 | |
| *** germs has quit IRC | 03:30 | |
| *** germs has joined #openstack-ansible | 03:30 | |
| *** markvoelker has joined #openstack-ansible | 03:34 | |
| *** germs1 has joined #openstack-ansible | 03:36 | |
| *** germs has quit IRC | 03:38 | |
| *** thorst has joined #openstack-ansible | 03:38 | |
| *** markvoelker has quit IRC | 03:39 | |
| *** markvoelker has joined #openstack-ansible | 03:43 | |
| *** thorst has quit IRC | 03:46 | |
| *** markvoelker has quit IRC | 03:48 | |
| *** markvoelker has joined #openstack-ansible | 03:52 | |
| *** vnogin has joined #openstack-ansible | 03:55 | |
| *** markvoelker has quit IRC | 03:57 | |
| *** dave-mccowan has quit IRC | 03:57 | |
| *** nwonknu has quit IRC | 03:57 | |
| *** Tahvok has quit IRC | 03:57 | |
| *** bauruine has quit IRC | 03:57 | |
| *** vnogin has quit IRC | 03:59 | |
| *** markvoelker has joined #openstack-ansible | 04:02 | |
| *** poopcat has quit IRC | 04:02 | |
| *** Tahvok has joined #openstack-ansible | 04:02 | |
| *** bauruine has joined #openstack-ansible | 04:03 | |
| *** markvoelker has quit IRC | 04:06 | |
| *** germs1 has quit IRC | 04:08 | |
| *** nwonknu has joined #openstack-ansible | 04:10 | |
| *** markvoelker has joined #openstack-ansible | 04:10 | |
| *** markvoelker has quit IRC | 04:15 | |
| *** markvoelker has joined #openstack-ansible | 04:20 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Fix Ceilometer deployments https://review.openstack.org/513229 | 04:31 |
| *** jbadiapa_ has joined #openstack-ansible | 04:43 | |
| *** jbadiapa has quit IRC | 04:46 | |
| *** kiennt26 has quit IRC | 04:46 | |
| *** jbadiapa_ has quit IRC | 04:53 | |
| *** markvoelker has quit IRC | 04:54 | |
| *** vnogin has joined #openstack-ansible | 04:56 | |
| *** vnogin has quit IRC | 05:00 | |
| *** markvoelker has joined #openstack-ansible | 05:08 | |
| *** newmember has quit IRC | 05:11 | |
| *** newmember has joined #openstack-ansible | 05:11 | |
| *** markvoelker has quit IRC | 05:12 | |
| *** masber has joined #openstack-ansible | 05:15 | |
| *** markvoelker has joined #openstack-ansible | 05:17 | |
| *** masuberu has joined #openstack-ansible | 05:17 | |
| *** masber has quit IRC | 05:19 | |
| *** markvoelker has quit IRC | 05:22 | |
| *** chyka has joined #openstack-ansible | 05:22 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Consolidate cached container methods https://review.openstack.org/513265 | 05:23 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create master: Fix BTRFS package name https://review.openstack.org/513157 | 05:23 |
| *** markvoelker has joined #openstack-ansible | 05:26 | |
| *** chyka has quit IRC | 05:26 | |
| *** markvoelker has quit IRC | 05:29 | |
| *** markvoelker has joined #openstack-ansible | 05:29 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Consolidate cached container methods https://review.openstack.org/513265 | 05:32 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create master: Fix BTRFS package name https://review.openstack.org/513157 | 05:34 |
| *** woodard_ has quit IRC | 05:36 | |
| *** thorst has joined #openstack-ansible | 05:42 | |
| *** markvoelker has quit IRC | 05:42 | |
| *** markvoelker has joined #openstack-ansible | 05:42 | |
| *** markvoelker has quit IRC | 05:42 | |
| *** thorst has quit IRC | 05:47 | |
| *** snowman4839 has quit IRC | 05:55 | |
| *** vnogin has joined #openstack-ansible | 05:56 | |
| *** snowman4839 has joined #openstack-ansible | 05:58 | |
| *** vnogin has quit IRC | 06:01 | |
| *** woodard has joined #openstack-ansible | 06:08 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Fix Ceilometer deployments https://review.openstack.org/513229 | 06:10 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Fix Ceilometer deployments https://review.openstack.org/513229 | 06:11 |
| *** rich1iu has joined #openstack-ansible | 06:14 | |
| *** yolanda has quit IRC | 06:15 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server master: Add openstack-ansible-linters test https://review.openstack.org/513135 | 06:17 |
| *** Oku_OS-away is now known as Oku_OS | 06:18 | |
| *** cshen has quit IRC | 06:21 | |
| *** drifterza has joined #openstack-ansible | 06:28 | |
| *** drifterza has quit IRC | 06:30 | |
| *** drifterza has joined #openstack-ansible | 06:35 | |
| *** drifterza has quit IRC | 06:39 | |
| rich1iu | one question , is it possible to run openstack-ansible on arm64 platform ? | 06:39 |
| *** drifterza has joined #openstack-ansible | 06:40 | |
| *** drifterza has quit IRC | 06:40 | |
| *** cshen has joined #openstack-ansible | 06:45 | |
| *** drifterza has joined #openstack-ansible | 06:48 | |
| *** drifterza has quit IRC | 06:48 | |
| *** drifterza has joined #openstack-ansible | 06:49 | |
| *** markvoelker has joined #openstack-ansible | 06:49 | |
| *** kiennt26 has joined #openstack-ansible | 06:51 | |
| *** drifterza has quit IRC | 06:53 | |
| *** drifterza has joined #openstack-ansible | 06:53 | |
| *** drifterza has quit IRC | 06:54 | |
| *** vishwana_ has joined #openstack-ansible | 06:54 | |
| *** vishwanathj has quit IRC | 06:54 | |
| *** mbuil has joined #openstack-ansible | 06:55 | |
| *** gunix has joined #openstack-ansible | 06:55 | |
| gunix | how many hosts do you recommand for a POC? | 06:55 |
| *** pcaruana has joined #openstack-ansible | 06:55 | |
| *** drifterza has joined #openstack-ansible | 06:56 | |
| gunix | would 3 controllers, 2 computes, 1 block storage, 2 object storage be ok ? | 06:56 |
| *** drifterza has left #openstack-ansible | 06:56 | |
| *** drifterza has joined #openstack-ansible | 06:56 | |
| *** vnogin has joined #openstack-ansible | 06:57 | |
| *** vishwanathj has joined #openstack-ansible | 06:58 | |
| *** vishwana_ has quit IRC | 06:59 | |
| SamYaple | gunix: it would be ok for a POC yes, assuming block storage is lvm | 07:00 |
| SamYaple | gunix: i would recommend 3 object storage thouhg, you can collocate that on the controllers however | 07:00 |
| gunix | SamYaple: i have no ideea what openstack-ansible deploys for the storage node :D | 07:00 |
| SamYaple | gunix: you can use ceph or lvm as teh common ones. there are a number of plugins that OSA supports though | 07:01 |
| gunix | SamYaple: i would practice with separate storage nodes | 07:01 |
| gunix | SamYaple: can you do any sort of multipath using lvm ? | 07:01 |
| SamYaple | you can, im not sure if OSA supports that. youd have to look/wait for someone in the channel who knows | 07:01 |
| SamYaple | im a ceph guyh myself | 07:02 |
| *** vnogin has quit IRC | 07:02 | |
| gunix | SamYaple: do you know any good online tutorial for ceph? i tried it once but i failed to deploy | 07:05 |
| *** drifterza has quit IRC | 07:07 | |
| *** drifterza has joined #openstack-ansible | 07:07 | |
| *** drifterza has quit IRC | 07:08 | |
| *** newmember has quit IRC | 07:18 | |
| *** newmember has joined #openstack-ansible | 07:19 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Include the Ansible version variable when installing ara https://review.openstack.org/513287 | 07:21 |
| SamYaple | gunix: OSA uses ceph-ansible, so i would start there | 07:22 |
| *** taseer1 has joined #openstack-ansible | 07:22 | |
| *** taseer2 has joined #openstack-ansible | 07:23 | |
| *** Taseer has quit IRC | 07:23 | |
| cloudnull | gunix: yes we have the ability work with multipath storage and nova | 07:24 |
| cloudnull | https://docs.openstack.org/openstack-ansible-os_nova/pike/configure-nova.html#multipath | 07:24 |
| *** jbadiapa has joined #openstack-ansible | 07:24 | |
| cloudnull | the trick would be ensuring it all will work with arm64 from a compute stand point, which it should assuming you're working with qemu/kvm | 07:26 |
| cloudnull | https://github.com/openstack/openstack-ansible-os_nova/tree/master/tasks/drivers | 07:26 |
| SamYaple | o/ cloudnull | 07:26 |
| cloudnull | though I don't know anyone doing this with arm at the moment so there may be underlying issues we've not accounted for. | 07:26 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_ceilometer master: [DNM] Quick Attempt to fix ceilometer https://review.openstack.org/513051 | 07:26 |
| cloudnull | o/ SamYaple | 07:26 |
| cloudnull | hows it going brother? | 07:26 |
| SamYaple | im actually off for the night, but going pretty good over all | 07:27 |
| SamYaple | yourself? | 07:27 |
| *** taseer1 has quit IRC | 07:27 | |
| cloudnull | evrardjp: https://review.openstack.org/#/c/513229/ -- re ceilometer issues | 07:27 |
| cloudnull | SamYaple: fair to midland | 07:27 |
| cloudnull | threatening to go to sleep | 07:27 |
| SamYaple | i wont make it to sydney, got other travel going on. but maybe we can catch up in Dublin? for the next ptg | 07:28 |
| cloudnull | but i think i have a few more min in me :D | 07:28 |
| *** taseer2 is now known as Taser | 07:28 | |
| SamYaple | nice | 07:28 |
| *** Taser is now known as Taseer | 07:28 | |
| SamYaple | ive hit the mental wall | 07:28 |
| cloudnull | we do need to go get some BBQ ! | 07:28 |
| evrardjp | oh I didn't intend to fix ceilometer, just to make sure it can pass some standard ansible stuff :p | 07:28 |
| SamYaple | indeed | 07:28 |
| evrardjp | I will have a look cloudnull | 07:28 |
| evrardjp | I have removed the -2 we discussed yesterday too | 07:28 |
| evrardjp | and hello cloudnull SamYaple! | 07:29 |
| evrardjp | :p | 07:29 |
| SamYaple | hello evrardjp! | 07:29 |
| SamYaple | goodnight evrardjp! | 07:29 |
| cloudnull | take care SamYaple | 07:29 |
| evrardjp | yup! | 07:29 |
| evrardjp | take care! | 07:29 |
| evrardjp | :) | 07:29 |
| cloudnull | evrardjp: if you have a moment, https://review.openstack.org/#/c/513287 I'm seeing this in pike | 07:30 |
| cloudnull | I wanted ara 0.14 and in a local deployment i have and it installed ansible 2.4.0 which breaks everything | 07:30 |
| evrardjp | oh yeah you introduced a fix! | 07:32 |
| evrardjp | yeah I think I filed a bug for that | 07:32 |
| evrardjp | oh it's not only there | 07:32 |
| cloudnull | orly? | 07:32 |
| evrardjp | yeah | 07:32 |
| evrardjp | in role tests | 07:32 |
| evrardjp | if you run run_tests twice | 07:33 |
| evrardjp | even if ansible and ara is already installed, it checks ara==0.14 and sees ansible>=... | 07:33 |
| evrardjp | and then installs 2.4 | 07:33 |
| evrardjp | maybe we should constraint ara. | 07:33 |
| *** DanyC has joined #openstack-ansible | 07:34 | |
| evrardjp | ansible* | 07:34 |
| evrardjp | cloudnull: for my understanding, just adding ansible fix the thing? | 07:35 |
| evrardjp | let me try that on my machine I am surprised | 07:35 |
| cloudnull | evrardjp: yup | 07:37 |
| cloudnull | just adding ansible in fixed it | 07:37 |
| evrardjp | interesting | 07:37 |
| evrardjp | validating | 07:37 |
| cloudnull | well it was the ansible version var | 07:37 |
| cloudnull | so it came out as something like ansible==2.3.2.0 | 07:37 |
| *** DanyC_ has quit IRC | 07:37 | |
| cloudnull | but yes. that fixed it for me | 07:37 |
| evrardjp | we should add a comment there | 07:37 |
| evrardjp | yes seem to work for me | 07:38 |
| evrardjp | wait | 07:38 |
| evrardjp | weird stuff | 07:38 |
| evrardjp | I uninstalled ara, and reinstalled ara | 07:38 |
| evrardjp | with ansible already installed | 07:38 |
| evrardjp | by doing pip install ara==0.14 | 07:39 |
| evrardjp | he didn't try to reinstall ansible | 07:39 |
| evrardjp | which is weird | 07:39 |
| evrardjp | I wonder how to reproduce that then | 07:39 |
| evrardjp | I think I will be happier when we'll build a pre-packaged stuff. | 07:41 |
| evrardjp | in the meantime you got my vote if that fixed it for you. | 07:42 |
| *** thorst has joined #openstack-ansible | 07:43 | |
| evrardjp | hwoarang: thanks on the dashboard update! | 07:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests master: Constrain ansible when installing ARA https://review.openstack.org/513298 | 07:45 |
| cloudnull | evrardjp: ^ that should fix it in tests too | 07:45 |
| cloudnull | we've already got an ansible constraint file in tests so simply using that when we install aara will limit its options | 07:46 |
| evrardjp | yeah | 07:47 |
| evrardjp | I didn't remember that file | 07:47 |
| evrardjp | great | 07:47 |
| *** thorst has quit IRC | 07:47 | |
| cloudnull | feel free to update those commits w/ the bug ID if you ahve it handy, or ping it to me when you find it. | 07:48 |
| cloudnull | otherwise I'm crashing. | 07:48 |
| cloudnull | :D | 07:48 |
| cloudnull | see you in a few hours o/ | 07:49 |
| evrardjp | yeah take care! | 07:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: Constrain ansible when installing ARA https://review.openstack.org/513298 | 07:52 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Add BTRFS cached container task https://review.openstack.org/513265 | 07:52 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Add option to disable xtrabackup compression https://review.openstack.org/504333 | 07:56 |
| *** markvoelker has quit IRC | 08:01 | |
| *** DanyC_ has joined #openstack-ansible | 08:01 | |
| gunix | cloudnull: nice! | 08:01 |
| gunix | cloudnull: thank you for the info regarding multipath | 08:01 |
| *** DanyC has quit IRC | 08:04 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_molteniron master: Add openstack-ansible-linters test https://review.openstack.org/513121 | 08:13 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install master: Add openstack-ansible-linters test https://review.openstack.org/513130 | 08:15 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Updated repo cloner for zuulv3 https://review.openstack.org/512475 | 08:18 |
| *** pbandark has joined #openstack-ansible | 08:20 | |
| *** vnogin has joined #openstack-ansible | 08:27 | |
| *** vnogin has quit IRC | 08:32 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_server stable/pike: [DNM] test gates https://review.openstack.org/513309 | 08:33 |
| *** yolanda has joined #openstack-ansible | 08:34 | |
| *** armaan has joined #openstack-ansible | 08:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_zaqar master: Add openstack-ansible-linters test https://review.openstack.org/513129 | 08:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Add openstack-ansible-linters test https://review.openstack.org/513123 | 08:49 |
| *** electrofelix has joined #openstack-ansible | 08:58 | |
| *** chyka has joined #openstack-ansible | 08:58 | |
| *** chyka has quit IRC | 09:03 | |
| *** vnogin has joined #openstack-ansible | 09:12 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Add openstack-ansible-linters test https://review.openstack.org/513127 | 09:16 |
| *** yolanda has quit IRC | 09:16 | |
| *** yolanda has joined #openstack-ansible | 09:18 | |
| *** armaan has quit IRC | 09:23 | |
| *** armaan has joined #openstack-ansible | 09:23 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-apt_package_pinning stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513326 | 09:26 |
| odyssey4me | rich1iu yes, multiple hardware architectures are supported - and IIRC someone has implemented changes to support arm specifically | 09:26 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-ceph_client stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513328 | 09:26 |
| odyssey4me | evrardjp don't do any more job changes until the current ones have made their way through zuul | 09:26 |
| odyssey4me | zuul is sitting at 12GB/15GB RAM usage | 09:27 |
| evrardjp | OMG. you're kidding? | 09:27 |
| odyssey4me | http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=63979&rra_id=all | 09:27 |
| odyssey4me | we crashed zuul twice yesterday with all our changes | 09:27 |
| evrardjp | mmm. So I keep my changes pending? | 09:27 |
| evrardjp | I am glad I manually reviewed them | 09:28 |
| *** duonghq has quit IRC | 09:28 | |
| evrardjp | so what should I do then? | 09:28 |
| evrardjp | If I can't work on each of the roles | 09:28 |
| evrardjp | I can't work on the proposal bot | 09:28 |
| odyssey4me | proposal bot is known as not working, infra was working on various things related to it last night | 09:29 |
| evrardjp | well I will try to work on that quick, but I guess the proposal bot will burn zuul v3 ... | 09:29 |
| evrardjp | odyssey4me: yeah I know Andreas was working on it, but he is not checking on our jobs, I don't know if you heard a different story | 09:29 |
| evrardjp | but right now there is NO testing on other branches than master. | 09:30 |
| evrardjp | that's why I wanted to add patches... | 09:30 |
| evrardjp | ok I will work on the queens M1 | 09:31 |
| evrardjp | oh no, no need, releases is blocked too. | 09:31 |
| *** vnogin has quit IRC | 09:33 | |
| odyssey4me | yes, releases is blocked because of the issue with signing, which relates to the proposal bot | 09:33 |
| odyssey4me | in other words, be patient | 09:33 |
| evrardjp | ok will fix centos swift. | 09:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Constrain ansible when installing ARA https://review.openstack.org/513298 | 09:36 |
| odyssey4me | ^ adjusted that to work in more than just the tests repo | 09:36 |
| *** yifei has quit IRC | 09:39 | |
| evrardjp | thanks | 09:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Add openstack-ansible-linters test https://review.openstack.org/513115 | 09:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic master: Add openstack-ansible-linters test https://review.openstack.org/513118 | 09:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Add openstack-ansible-linters test https://review.openstack.org/513119 | 09:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_watcher master: Add openstack-ansible-linters test https://review.openstack.org/513128 | 09:43 |
| *** thorst has joined #openstack-ansible | 09:44 | |
| odyssey4me | evrardjp I can't decide whether make the gate pipeline share a queue was a good idea or not | 09:45 |
| odyssey4me | previously we had independent queues - so every repo merged of its own accord, only testing with whatever was available and merged at the time of testing | 09:46 |
| evrardjp | I don't understand | 09:46 |
| evrardjp | oh | 09:46 |
| evrardjp | what was the reason to share the same queue? | 09:47 |
| odyssey4me | now that they share a queue for merging, each role patch goes into the same queue - in other words they are all tested together, and if one fails it is taken out of the line | 09:47 |
| odyssey4me | it increases the quality of testing, which is why I did it -it makes sure that all patches in the gate queue are tested together - not seperately | 09:47 |
| evrardjp | shas are relevant for us | 09:48 |
| evrardjp | this doesn't increase the quality | 09:48 |
| odyssey4me | but it does mean that each test waits for the one ahead of it to pass before it can merge, and it that one fails, the tests restart | 09:48 |
| evrardjp | but it doesn't decrease it either | 09:48 |
| *** thorst has quit IRC | 09:49 | |
| evrardjp | did infra ask for that? | 09:49 |
| evrardjp | to relieve the load? | 09:49 |
| odyssey4me | the shared queue is how nova/cinder/neutron/etc process their gate - testing changes made in each repo together | 09:49 |
| odyssey4me | no, infra did not ask for it - I thought it was a good idea and added it | 09:49 |
| odyssey4me | given, it's only showing up as an issue because we have a lot of patches in the queue | 09:50 |
| evrardjp | I'd like to at least be informed of those changes | 09:50 |
| evrardjp | in the future | 09:50 |
| odyssey4me | you were - you reviewed the patches ;) | 09:50 |
| evrardjp | not with the full context then | 09:50 |
| evrardjp | but ok that's my fault | 09:50 |
| odyssey4me | but yeah, sure - it's not like we had much time to discuss right | 09:50 |
| *** yolanda has quit IRC | 09:51 | |
| evrardjp | no it's alright, it means I badly reviewed. It happens sometimes. | 09:51 |
| evrardjp | :p | 09:51 |
| odyssey4me | I should not have added it, actually. I changed what was previously there without really discussing it. | 09:51 |
| evrardjp | humans after all :) | 09:51 |
| odyssey4me | anyway, too late now | 09:51 |
| evrardjp | nah it's alright :) | 09:51 |
| evrardjp | is it possible to revert easily? | 09:52 |
| odyssey4me | hmm, actually - I think it's the job only - so only tests repo | 09:52 |
| odyssey4me | oh no - it's a change to every repo: https://github.com/openstack/openstack-ansible-tests/blob/master/zuul.d/project.yaml#L25 | 09:52 |
| evrardjp | while you are here, should we discuss the job names? With what we started to discuss yesterday | 09:52 |
| evrardjp | to trigger from outside easily | 09:53 |
| *** taseer1 has joined #openstack-ansible | 09:53 | |
| *** Taseer has quit IRC | 09:53 | |
| *** taseer2 has joined #openstack-ansible | 09:54 | |
| evrardjp | or alternatively, is there a way to say from keystone repo, use osa generic update job _ON THIS repo_ | 09:54 |
| odyssey4me | easy enough - openstack-ansible-keystone-upgrade, then have the required repositories for it.... I just don't know exactly how to make it use our repo as the primary for the test | 09:54 |
| odyssey4me | what we might have to do is setup a workspace in the job, with a symlink to our repo, then go from there | 09:55 |
| *** armaan has quit IRC | 09:55 | |
| evrardjp | with this job name, maybe the whole thing is smart | 09:55 |
| evrardjp | instead of openstack-ansible-upgrade-ubuntu-xenial | 09:55 |
| *** armaan has joined #openstack-ansible | 09:55 | |
| odyssey4me | actually, I think we're confusing everything - it should just work as expected, as long as we change where we expect keystone to be | 09:55 |
| *** hw_wutianwei has quit IRC | 09:55 | |
| odyssey4me | I don't think it's crucial to work on now anyway. | 09:56 |
| odyssey4me | I need to focus on other work today and tomorrow, but can help out again this evening. | 09:57 |
| evrardjp | I'll speak with the team and tell them our status | 09:57 |
| evrardjp | this way they are at least informed | 09:57 |
| evrardjp | yeah | 09:57 |
| odyssey4me | in terms of the changes to jobs - only submit around 5 at a time, and let them complete processing before submitting another batch | 09:57 |
| evrardjp | focus on your day work. don't worry :p | 09:57 |
| odyssey4me | so prioritise the repositories you want tests in, do the urgent ones first | 09:57 |
| evrardjp | ok | 09:58 |
| odyssey4me | this has 5 min update intervals: http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=63979&rra_id=all | 09:58 |
| *** taseer1 has quit IRC | 09:58 | |
| evrardjp | main repo, tests and neutron all branches basically. | 09:58 |
| evrardjp | and keystone | 09:58 |
| odyssey4me | if the used memory goes up to the max (16GB), then zuul will hang and we will have to recheck them all, 5 at a time... so try to avoid that | 09:59 |
| ansibler | odyssey4me, thanks for suggesting removing non OSA roles from /etc/ansible/roles yesterday - it helped and I got over openstack-ansible setup-infrastructure.yml. | 09:59 |
| odyssey4me | so yeah, I would think focus on the defcore roles and infra roles | 09:59 |
| evrardjp | Working only 5 at a time is gonna teach me patience. | 09:59 |
| odyssey4me | and any roles that have patches pending | 09:59 |
| ansibler | Though now stuck in openstack-ansible setup-openstack.yml https://pastebin.com/raw/szPA1nEB Have grep'd /etc and /opt for different ceph paterns but cannot understand what's wrong. Double use of same key? | 09:59 |
| evrardjp | odyssey4me: I just want to get master in the right state right now. For example we could think keystone is fine, but it's not. upgrade is still doing ocata | 10:00 |
| evrardjp | it's been 3 weeks I am trying to merge this _fully working_ patch | 10:01 |
| evrardjp | I can work on the 2.4 support too | 10:02 |
| odyssey4me | well, if there are troublesome tests due to other issues at play, perhaps switch them to non-voting so that we can get tests established and working again | 10:02 |
| odyssey4me | we can always submit another patch to make a non voting test voting | 10:02 |
| evrardjp | that is true. | 10:02 |
| odyssey4me | then use that to buy time | 10:02 |
| evrardjp | but they are technically passing, so why disabling them? It's just temporary issues | 10:02 |
| odyssey4me | temporary issues for three weeks - sounds to me like they might be transient, but they're not temporary | 10:03 |
| evrardjp | anyway, I understand the gist, and can restructure my work. | 10:03 |
| evrardjp | I didn't know the difference, let me google that. | 10:03 |
| odyssey4me | also, non voting doesn't mean that we ignore the results - it just means that it's not required to merge the patch | 10:03 |
| evrardjp | I know that... :) | 10:04 |
| odyssey4me | let's try to get them merged as-is for another day | 10:04 |
| *** kiennt26 has quit IRC | 10:04 | |
| odyssey4me | if they don't work, then this evening we can chat with a broader group to discuss what to do about it | 10:05 |
| odyssey4me | could you review https://review.openstack.org/512746 to get it out of the queue | 10:06 |
| odyssey4me | oh yeah, by the way - that tempest test I put up last night is one of the causes of failures I've seen quite often | 10:07 |
| odyssey4me | the other one is the git issue - I have a plan there and will do a patch for it later | 10:07 |
| evrardjp | google didn't really help me on the difference between temporary and transient. | 10:08 |
| evrardjp | what are you talking about? | 10:08 |
| odyssey4me | temporary is just that, it's there - but it will go away soo | 10:08 |
| evrardjp | tempest? is that the swift issue? | 10:08 |
| odyssey4me | *soon | 10:08 |
| evrardjp | transient seem to be the same thing | 10:08 |
| odyssey4me | transient is sometimes there, sometimes not | 10:08 |
| odyssey4me | it's less deliberate | 10:09 |
| evrardjp | according to google, but that's not oxford | 10:09 |
| evrardjp | ok | 10:09 |
| odyssey4me | oh sure, but that's the intent of my meaning | 10:09 |
| evrardjp | ok understood | 10:09 |
| odyssey4me | temporary is something like a workaround - you control when it's there and when it goes | 10:09 |
| evrardjp | so what did you say with the git and tempest? | 10:09 |
| evrardjp | most of our issues right now are with the lxc images | 10:10 |
| odyssey4me | so there are a few problems I've seen popping up in job failures | 10:10 |
| evrardjp | I am not aware of git ones | 10:10 |
| odyssey4me | one is that tempest fails to download its image | 10:10 |
| evrardjp | oh yeah, but I voted on this patch I think | 10:10 |
| odyssey4me | that should be solved by https://review.openstack.org/513200 | 10:10 |
| evrardjp | isn't it merged yet? | 10:10 |
| evrardjp | oh yeah, we are waiting for tempest to pass the first thing. | 10:10 |
| odyssey4me | you're right to hold it back by the way - I have not functionally tested it | 10:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Use the deployment host to download images for tempest https://review.openstack.org/513200 | 10:11 |
| evrardjp | I am not in a place where I want to pay the price of a blind merge | 10:11 |
| odyssey4me | it's now rebased on the thing with jobs, so it will test | 10:11 |
| evrardjp | you rebased on top? | 10:11 |
| evrardjp | ok | 10:12 |
| evrardjp | good | 10:12 |
| odyssey4me | I'm not asking you to - I'm saying that we shouldn't blind merge it. It needs to be tested because I wrote it blind and it's a big change. | 10:12 |
| odyssey4me | right, another issue I've seen is that when the roles git clone, they fail. | 10:12 |
| odyssey4me | the way around that is going to be to implement a 'pre' stage in our job, because the 'pre' stage gets retried three times before failing | 10:13 |
| *** rodolof has joined #openstack-ansible | 10:13 | |
| odyssey4me | another issue I've seen is more centos/suse specific - I've seen repo updates just hang until timeout | 10:14 |
| odyssey4me | I asked whether there was a local infra suse repo or reverse proxy last night, but didn't get an answer. I think that might be worth persueing. | 10:15 |
| hwoarang | odyssey4me: let me know if you find anything. I have been running local and opnfv tests over and over but I haven't seen any timeouts myself :( | 10:16 |
| odyssey4me | hwoarang yeah, I suspect it may just be a transient issue which yum doesn't handle very well | 10:17 |
| odyssey4me | perhaps something that nothing can handle well | 10:17 |
| odyssey4me | for example, it could be some sort of packet loss due to a bad route appearing | 10:17 |
| *** ansibler has quit IRC | 10:18 | |
| odyssey4me | your approach to using an async task for the cache prep was quite clever, I thought | 10:18 |
| hwoarang | odyssey4me: evrardjp: right i am seeing https://review.openstack.org/#/c/496679/ but the async task failed as expected | 10:19 |
| hwoarang | however it seems no logs were collected :/ | 10:19 |
| hwoarang | i wonder why... | 10:19 |
| evrardjp | yeah the async you added is great. | 10:20 |
| odyssey4me | right, that's an issue we can solve now | 10:20 |
| evrardjp | killing the job | 10:20 |
| odyssey4me | so, much like we could do the git cloning in the 'pre' stage... we can now also do the log collection in the 'post' stage | 10:20 |
| evrardjp | I'd love it to not fail transiently in infra at first, but that's for another time :) | 10:20 |
| odyssey4me | the post stage is always run, regardless of whether the job passed or failed | 10:21 |
| evrardjp | that seems good. | 10:21 |
| hwoarang | odyssey4me: do i understand correctly that we miss something in the post step to collect logs? | 10:21 |
| evrardjp | next community meeting has a point on the agenda to discuss what to keep from logs | 10:21 |
| odyssey4me | so, we're in a place where we can split the one step test run into three steps and have infra execute them slightly differently | 10:21 |
| odyssey4me | hwoarang yeah, right now we only have a run step | 10:21 |
| evrardjp | yeah, the git clone in pre-steps is good. | 10:22 |
| evrardjp | and the prepa of ansible | 10:22 |
| evrardjp | then we just run stuff | 10:22 |
| odyssey4me | https://github.com/openstack/openstack-ansible-tests/blob/master/zuul.d/jobs.yaml#L46 | 10:22 |
| odyssey4me | that's all we do | 10:22 |
| * hwoarang is taking mental notes | 10:23 | |
| odyssey4me | so my general plan is to work on splitting up the way the test runs into pre/run/post stages and to have our job use them that way | 10:23 |
| odyssey4me | evrardjp one thing we could also consider is removing the use of containers from roles that do not need to use them - they add a lot of overhead in terms of the setup process | 10:24 |
| hwoarang | maybe i can submit a patch to use the opensuse mirrors instead of the infra ones. I can't even remember if we use the openstack mirrors. I need to see the logs once they are available | 10:24 |
| odyssey4me | if all a job is doing is setting up a single container - it may as well use the host | 10:24 |
| evrardjp | odyssey4me: I agree, that was my plan | 10:24 |
| evrardjp | I am still working on the everything on metal when I have a few minutes with no infra burning. | 10:25 |
| odyssey4me | hwoarang I was wondering whether there are suse mirrors in infra | 10:25 |
| hwoarang | i am not sure... | 10:25 |
| odyssey4me | hwoarang I don't see anything here to set a mirror for suse: https://github.com/openstack/openstack-ansible-tests/blob/b068c49acabd969b78355d773390edc0094d8235/common-tasks/test-set-nodepool-vars.yml | 10:25 |
| odyssey4me | perhaps that'd be a good thing to make work! | 10:25 |
| hwoarang | plus we stopped collecting /etc/* files it seems so i can't easily check what mirrors are being used | 10:25 |
| hwoarang | odyssey4me: i think the dib images that are loaded by infra are using the infra mirrors | 10:26 |
| hwoarang | and i dont think i wipe them | 10:26 |
| odyssey4me | for that, I'd suggest that we targz the etc contents as they were so that we can still get them, it just requires a download | 10:26 |
| hwoarang | yeah that would be helpful | 10:26 |
| odyssey4me | hwoarang ah ok | 10:26 |
| * hwoarang is rushing to lunch will be back asap | 10:27 | |
| evrardjp | odyssey4me: agreed. fungi has suggested to still reduce what we need to log there | 10:27 |
| evrardjp | I agree with him | 10:27 |
| odyssey4me | evrardjp oh sure, there is a lot in there which is not relevant most of the time | 10:27 |
| evrardjp | exactly | 10:27 |
| odyssey4me | another way to do it could be to only collect on failure | 10:28 |
| evrardjp | so I thought we could add to a tar the folders we want | 10:28 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tacker master: Initial OSA zuul v3 role jobs https://review.openstack.org/512746 | 10:30 |
| evrardjp | swift works perfectly fine on my machine... | 10:30 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Add openstack-ansible-linters test https://review.openstack.org/513120 | 10:33 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins master: Add openstack-ansible-linters test https://review.openstack.org/513132 | 10:33 |
| *** rodolof has quit IRC | 10:34 | |
| odyssey4me | hmm, that's not good: http://logs.openstack.org/00/513200/3/check/openstack-ansible-functional-ubuntu-xenial/4bcc12c/job-output.txt.gz#_2017-10-19_10_25_46_841152 | 10:34 |
| *** rodolof has joined #openstack-ansible | 10:35 | |
| evrardjp | odyssey4me: what are you trying to say? | 10:35 |
| evrardjp | yes we have that all over the place, but do you know why it fails? | 10:35 |
| odyssey4me | all tests for the patch failed with the same issue - the download took too long | 10:35 |
| odyssey4me | well, the async task took too long | 10:36 |
| odyssey4me | that obviously needs debugging | 10:36 |
| odyssey4me | is this happening consistently, or only sometimes? | 10:36 |
| evrardjp | transient :) | 10:36 |
| evrardjp | sometimes this takes like 5 minutes | 10:36 |
| *** taseer2 is now known as Taseer | 10:36 | |
| evrardjp | sometimes this takes like 30 minutes. | 10:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Initial OSA zuul v3 role jobs https://review.openstack.org/512747 | 10:38 |
| evrardjp | woot! | 10:39 |
| evrardjp | ^ milestone reached! | 10:39 |
| evrardjp | hwoarang: did you see the log that odyssey4me has shown | 10:40 |
| evrardjp | I think we should rollback to 118 :p | 10:40 |
| evrardjp | let me try locally | 10:40 |
| evrardjp | odyssey4me: on my machine this lxc download takes less than 5 seconds. | 10:40 |
| odyssey4me | ok, so hwoarang's patch is doing the right thing - obviously it works sometimes | 10:40 |
| odyssey4me | at least this gets us a fail - it's progress | 10:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Add openstack-ansible-linters test https://review.openstack.org/513122 | 10:42 |
| odyssey4me | maybe https://review.openstack.org/512386 will help? | 10:44 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 10:45 |
| odyssey4me | evrardjp do you want to remove the common queue from the backports? | 10:46 |
| evrardjp | oh yeah | 10:47 |
| evrardjp | I should give it a different name right? | 10:47 |
| evrardjp | let me patch that quick | 10:47 |
| odyssey4me | no, just remove the line | 10:47 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 10:47 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 10:48 |
| odyssey4me | yes, that's right | 10:48 |
| odyssey4me | do the same for https://review.openstack.org/513328 and https://review.openstack.org/513326 ? | 10:49 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-apt_package_pinning stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513326 | 10:49 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-ceph_client stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513328 | 10:49 |
| evrardjp | I'd prefer start the whole batch, come back later this way I can do other work, but we are definitely out of ram now. | 10:50 |
| odyssey4me | yep | 10:50 |
| evrardjp | this isn't sustainable | 10:51 |
| evrardjp | I like zuul v3, and all its features, but right now I'd prefer to have v2. | 10:52 |
| *** yifei has joined #openstack-ansible | 10:52 | |
| odyssey4me | infra will work on making it better | 10:52 |
| evrardjp | ask me that in a few hours when I'll have eaten, another answer! | 10:52 |
| odyssey4me | haha, hangry is real | 10:53 |
| evrardjp | so true. | 10:53 |
| evrardjp | and you know I am quite a hobbit. | 10:56 |
| evrardjp | do you think we can simply backport those changes into O and N? | 10:56 |
| *** stuartgr has joined #openstack-ansible | 10:59 | |
| odyssey4me | not as-s, but almost | 10:59 |
| rich1iu | odyssey4me: thanks, I will try it. | 11:00 |
| odyssey4me | ocata has centos support, but no suse support | 11:00 |
| odyssey4me | as it stands right now, I think that perhaps we should just skip centos/suse for ocata/newton | 11:00 |
| odyssey4me | newton also needs trusty tests | 11:00 |
| gunix | https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/deploymenthost.html | 11:01 |
| gunix | no repo for ubuntu 16.04 ? | 11:01 |
| odyssey4me | gunix what do you mean? | 11:01 |
| openstackgerrit | Kevin Lefevre proposed openstack/openstack-ansible-os_nova master: Enable KSM support for Nova https://review.openstack.org/507862 | 11:02 |
| gunix | nevermind, stupid question. i expected to see the openstack repo fro pike | 11:02 |
| gunix | but it's probably somewhere else | 11:02 |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible master: (WIP) Integrate Congress with OSA. https://review.openstack.org/503971 | 11:02 |
| odyssey4me | gunix we add it automatically in the right places for you | 11:02 |
| Taseer | evrardjp: I made some changes to congress tests. I rebased the patch, but it did not seem to trigger the pipelines | 11:05 |
| odyssey4me | Taseer zuul is almost out of memory so it might take a while | 11:06 |
| Taseer | odyssey4me: okay, I'll wait then. Thanks | 11:06 |
| odyssey4me | Taseer I see your change in the queue though | 11:06 |
| odyssey4me | bottom left of http://zuulv3.openstack.org/ | 11:07 |
| Taseer | Yes, I find it | 11:10 |
| *** smatzek has joined #openstack-ansible | 11:12 | |
| *** DanyC_ has quit IRC | 11:13 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Add git and iptables to the list of packages https://review.openstack.org/512631 | 11:20 |
| *** cshen_ has joined #openstack-ansible | 11:25 | |
| *** markvoelker has joined #openstack-ansible | 11:27 | |
| *** cshen_ has quit IRC | 11:29 | |
| *** armaan has quit IRC | 11:31 | |
| *** armaan has joined #openstack-ansible | 11:32 | |
| *** chyka has joined #openstack-ansible | 11:33 | |
| *** smatzek has quit IRC | 11:33 | |
| *** cshen_ has joined #openstack-ansible | 11:34 | |
| gunix | https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/deploymenthost.html | 11:37 |
| *** chyka has quit IRC | 11:37 | |
| gunix | why is here the branch 16.0.1 suggested? | 11:37 |
| gunix | can't i just git clone without -b? | 11:37 |
| gunix | the tutorial is for 16.0.2 :D | 11:38 |
| *** armaan has quit IRC | 11:43 | |
| odyssey4me | gunix because that's the latest released tag | 12:06 |
| gunix | and the head is testing, i guess | 12:06 |
| odyssey4me | if you clone the head of the branch, you get the next proposed version | 12:06 |
| gunix | ? | 12:06 |
| gunix | ok, thank you! | 12:06 |
| odyssey4me | yes, we typically release every two weeks - giving that time in between to test and verify if things are ok | 12:07 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara master: Add openstack-ansible-linters test https://review.openstack.org/513124 | 12:07 |
| *** ansibler has joined #openstack-ansible | 12:11 | |
| *** dave-mccowan has joined #openstack-ansible | 12:12 | |
| *** thorst has joined #openstack-ansible | 12:13 | |
| hwoarang | evrardjp: nope. which one? | 12:29 |
| * hwoarang is reading the backlog slowly | 12:29 | |
| *** jamesdenton has joined #openstack-ansible | 12:30 | |
| *** yifei has quit IRC | 12:33 | |
| hwoarang | odyssey4me: evrardjp: Ohhh i see it now. the same error on ubuntu... that's not good | 12:34 |
| hwoarang | unless it didn't really finish in 5 minutes... maybe i was too optimistic | 12:35 |
| mhayden | morning folks | 12:39 |
| evrardjp | and... my internet connection is back! | 12:39 |
| evrardjp | hwoarang: what were we talking about? | 12:40 |
| *** yolanda has joined #openstack-ansible | 12:41 | |
| hwoarang | evrardjp: you asked me if I say odyssey4me's log. I presume you meant this http://logs.openstack.org/00/513200/3/check/openstack-ansible-functional-ubuntu-xenial/4bcc12c/job-output.txt.gz#_2017-10-19_10_25_46_841152 | 12:44 |
| hwoarang | i am not sure if it's the same issue or maybe the mirrors were slow and 5 minutes were not enough... | 12:45 |
| *** gkadam has quit IRC | 12:45 | |
| evrardjp | oh yes. No I think I just meant did you see the retry seconds | 12:46 |
| evrardjp | it look like it's doing it every 5 seconds | 12:46 |
| evrardjp | isn't it the task we messed up with? | 12:46 |
| hwoarang | yeah :/ | 12:46 |
| hwoarang | 'delay' is lying | 12:47 |
| evrardjp | I think I tried it on my machine and it worked. So now, I wonder what my first amend on your patch wasn't the solution... | 12:50 |
| evrardjp | 118 if you recall :) | 12:50 |
| hwoarang | confusing | 12:50 |
| evrardjp | down from everyone or just me... cacti.openstack.org :/ | 12:50 |
| evrardjp | I didn't do anything wrong! | 12:51 |
| hwoarang | works for me | 12:52 |
| hwoarang | hmm shall we change it to 118? i don't understand what's going on. i guess the docs are lying | 12:52 |
| evrardjp | so my internet is down again then | 12:53 |
| evrardjp | and nobody is getting my irc messages | 12:54 |
| evrardjp | woot! | 12:54 |
| *** hachi has joined #openstack-ansible | 13:00 | |
| evrardjp | great I am back! Darn flaky ipv6 stack | 13:02 |
| *** cshen_ has quit IRC | 13:03 | |
| evrardjp | hwoarang: on my machine it respects the docs | 13:04 |
| hwoarang | :/ | 13:06 |
| hwoarang | bug bug bug | 13:07 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 13:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 13:09 |
| evrardjp | hwoarang: do you think it's possible to have incorrect timings? | 13:11 |
| evrardjp | in the gates I mean | 13:11 |
| evrardjp | what do you see in xci? Everything is downloading fine? | 13:12 |
| odyssey4me | evrardjp perhaps it's time to switch to using debootstrap instead ;) | 13:12 |
| evrardjp | oh fun stuff nova-lxd is broken today | 13:13 |
| evrardjp | yesterday it worked fine. | 13:13 |
| evrardjp | odyssey4me: everywhere? :p | 13:13 |
| *** yolanda has quit IRC | 13:13 | |
| evrardjp | I was one of the first to say ok to debootstrap I think :p | 13:13 |
| *** woodard has quit IRC | 13:14 | |
| evrardjp | but not a prio | 13:14 |
| evrardjp | I may rethink this :p | 13:14 |
| *** woodard has joined #openstack-ansible | 13:14 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova master: Update paste, policy and rootwrap configurations 2017-10-14 https://review.openstack.org/512044 | 13:19 |
| *** tnogisto has joined #openstack-ansible | 13:23 | |
| tnogisto | hello, sorry this might be a stupid question, but are there any examples of ansible-openstack user_config file, where openvswitch with self-provisioned networks is deployed? | 13:33 |
| tnogisto | currently my compute node seems to take the managment IP, not the overlay IP, and Im not sure why that happens | 13:33 |
| tnogisto | network nodes neutron client node has the correct IP | 13:34 |
| *** woodard has quit IRC | 13:36 | |
| *** hachi has quit IRC | 13:41 | |
| *** hachi has joined #openstack-ansible | 13:42 | |
| *** yifei has joined #openstack-ansible | 13:45 | |
| *** hachi_ has joined #openstack-ansible | 13:48 | |
| *** hachi has quit IRC | 13:48 | |
| *** hachi_ has quit IRC | 13:53 | |
| hwoarang | evrardjp: yeah i haven't seen these timeouts :( | 13:58 |
| hwoarang | but we run far less jobs than openstack. Maybe 6 per day but still... | 13:58 |
| hwoarang | and we only upstream mirrors | 13:58 |
| evrardjp | the etherpad is updated if you want to help | 13:59 |
| evrardjp | any of you :) | 13:59 |
| evrardjp | hwoarang: ok | 13:59 |
| evrardjp | well we'll see :) | 13:59 |
| hwoarang | the etherpad re zuul? | 13:59 |
| *** vnogin has joined #openstack-ansible | 14:02 | |
| *** lbragstad has joined #openstack-ansible | 14:04 | |
| *** galstrom_zzz is now known as galstrom | 14:05 | |
| *** esberglu has joined #openstack-ansible | 14:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/pike: tasks: lxc_container_create: Retry LXC start on failures https://review.openstack.org/509910 | 14:13 |
| *** gouthamr has joined #openstack-ansible | 14:20 | |
| cloudnull | mornings | 14:20 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: [WIP] Clean out old mariadb apt sources https://review.openstack.org/513400 | 14:24 |
| evrardjp | morning | 14:29 |
| evrardjp | hwoarang: yes | 14:29 |
| *** gkadam has joined #openstack-ansible | 14:38 | |
| *** weezS has joined #openstack-ansible | 14:38 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 14:41 |
| *** jbadiapa has quit IRC | 14:43 | |
| *** galstrom is now known as galstrom_zzz | 14:45 | |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_tempest master: add var to set a specific public network gateway_ip https://review.openstack.org/513407 | 14:45 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova master: Update upgrade role for Queens from P https://review.openstack.org/496679 | 14:45 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 14:53 |
| *** jvidal has quit IRC | 14:55 | |
| *** rodolof has quit IRC | 14:57 | |
| *** rodolof has joined #openstack-ansible | 14:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513328 | 15:03 |
| *** gkadam is now known as gkadam-away | 15:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-apt_package_pinning stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513326 | 15:05 |
| *** Oku_OS is now known as Oku_OS-away | 15:10 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:10 |
| *** yifei has quit IRC | 15:12 | |
| *** markvoelker has quit IRC | 15:12 | |
| *** markvoelker has joined #openstack-ansible | 15:13 | |
| *** acormier has quit IRC | 15:14 | |
| *** acormier has joined #openstack-ansible | 15:14 | |
| *** newmember has quit IRC | 15:15 | |
| *** markvoelker has quit IRC | 15:17 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:17 |
| *** acormier has quit IRC | 15:18 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:19 |
| odyssey4me | evrardjp you're mixing up periodics and PR jobs though | 15:19 |
| evrardjp | yes | 15:19 |
| evrardjp | same definition | 15:19 |
| odyssey4me | that's not a good idea I don't think | 15:19 |
| evrardjp | I don't think we should have two different definitions for one single thing | 15:19 |
| odyssey4me | the parent defines the base things, then as you go down the tree you get more specific | 15:20 |
| evrardjp | double negation or? | 15:20 |
| evrardjp | and? | 15:20 |
| odyssey4me | also, the distros are better defined lower down | 15:20 |
| odyssey4me | one job, then one variation using the job as parent, then all distros as variations of the second... it cuts down a lot of repeated bits - eg: https://github.com/openstack/openstack-ansible-tests/blob/master/zuul.d/jobs.yaml#L26-L59 | 15:22 |
| odyssey4me | in this case - one job, then a variation would need to be for PR's and another for periodics as they have different triggers | 15:23 |
| odyssey4me | the branches bit should only get added to the periodics one | 15:23 |
| evrardjp | well I kinda repeated on purpose... many things don't need to say deploy | 15:24 |
| evrardjp | it was very readable | 15:24 |
| evrardjp | but I can modify it | 15:24 |
| evrardjp | I don't understand why we need two variations for periodics and prs | 15:25 |
| odyssey4me | do the PR's one first - don't try and tackle them both at once | 15:25 |
| evrardjp | it's the same job content, just different variables | 15:25 |
| odyssey4me | right now all the jobs you're defining will get run for all reviews | 15:25 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:26 |
| evrardjp | why? | 15:26 |
| evrardjp | why all the jobs will run for all reviews? | 15:27 |
| odyssey4me | oh, I just saw the pipeline now | 15:27 |
| evrardjp | sorry the why wasn't very clear | 15:27 |
| odyssey4me | apologies - misread | 15:27 |
| evrardjp | I do many stupid word order mistakes, but except that, the pipeline seem simple | 15:27 |
| odyssey4me | it'd be nice if the jobs in the pipelines were alphanumerically ordered ;) | 15:28 |
| evrardjp | but you're gonna mix apples and oranges? | 15:28 |
| evrardjp | centos with ubuntu? | 15:28 |
| evrardjp | blasphemy! | 15:29 |
| evrardjp | :p | 15:29 |
| evrardjp | ok for the reordering | 15:29 |
| evrardjp | it was clearer for me that way, because I used the old job definition | 15:29 |
| evrardjp | I think it's quite simpler at the end | 15:29 |
| evrardjp | let me re-order them | 15:29 |
| evrardjp | darn OCD you have :p | 15:30 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:32 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 15:34 |
| odyssey4me | it looks like zuul has some memory back - want to recheck a few things again? | 15:35 |
| odyssey4me | looks like they're mostly done - so perhaps backports? | 15:35 |
| evrardjp | oh yeah I didn't do any commit to avoid breaking things | 15:36 |
| evrardjp | I have them pending, just waiting | 15:36 |
| evrardjp | should we finish the big ones first? | 15:37 |
| evrardjp | because what's pending are the big ones | 15:37 |
| evrardjp | in master I mean | 15:37 |
| evrardjp | nova keystone swift... | 15:37 |
| evrardjp | I will do the glance cinder heat neutron | 15:37 |
| evrardjp | well neutron is done | 15:37 |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: L3 Support for ODL deployments https://review.openstack.org/497458 | 15:38 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_glance stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513421 | 15:38 |
| odyssey4me | well, the big ones are failing more often - perhaps focus more on the ones that have more patches pending | 15:38 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_heat stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513422 | 15:38 |
| *** phalmos has joined #openstack-ansible | 15:39 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_cinder stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513423 | 15:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513424 | 15:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_horizon stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513425 | 15:40 |
| evrardjp | 5 ! | 15:41 |
| *** chyka has joined #openstack-ansible | 15:41 | |
| evrardjp | galera lxc hosts | 15:42 |
| evrardjp | let me add them | 15:42 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513426 | 15:43 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_hosts stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513427 | 15:43 |
| *** woodard has joined #openstack-ansible | 15:45 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: Provide support for SFC deployments https://review.openstack.org/510909 | 15:49 |
| *** ianychoi has quit IRC | 16:06 | |
| *** ianychoi has joined #openstack-ansible | 16:07 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible stable/pike: Include the Ansible version variable when installing ara https://review.openstack.org/513434 | 16:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Constrain ansible when installing ARA https://review.openstack.org/513298 | 16:10 |
| odyssey4me | cloudnull apologies - just updated that one again - missed another line | 16:11 |
| cloudnull | no worries :D | 16:11 |
| * cloudnull was tired when I wrote that | 16:11 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible-os_neutron master: L3 Support for ODL deployments https://review.openstack.org/497458 | 16:11 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/pike: Constrain ansible when installing ARA https://review.openstack.org/513437 | 16:13 |
| cloudnull | I backported those changes to pike so I didn't forget. | 16:13 |
| cloudnull | do we need them in ocata too ? | 16:13 |
| odyssey4me | no, no ara implementation in ocata | 16:13 |
| cloudnull | kk | 16:13 |
| cloudnull | odyssey4me: evrardjp: on https://review.openstack.org/#/c/513265 do we want a btrfs, zfs, overlayfs scenario within that role too ? | 16:15 |
| odyssey4me | yeah, for now | 16:15 |
| evrardjp | create just a container, show that CoW works, badaboom | 16:16 |
| odyssey4me | if we're doing back-end things, we may as well test them | 16:16 |
| evrardjp | should be two tasks | 16:16 |
| odyssey4me | cloudnull evrardjp actually... hmm | 16:16 |
| odyssey4me | currently the container create there is a backing container for cow containers created by lxc-container-create | 16:16 |
| cloudnull | we're going to need to implement this similar to lxc-container-create | 16:16 |
| evrardjp | yeah but lxc-container-create has many things right? | 16:17 |
| odyssey4me | adding it in this repo adds quite a bit of overhead | 16:17 |
| evrardjp | isn't it possible to do something very lightweight instead? This way we don't test the functions of the creation of lxc container | 16:17 |
| *** pcaruana has quit IRC | 16:18 | |
| evrardjp | odyssey4me: let me rephrase then | 16:18 |
| odyssey4me | it'd be nicer to actually move the cow base prep to lxc-container-create instead | 16:18 |
| evrardjp | could we trigger a lxc container job? | 16:18 |
| evrardjp | we could technically now. | 16:18 |
| odyssey4me | evrardjp we could | 16:18 |
| odyssey4me | although we'd have to define it in the tests repo | 16:18 |
| evrardjp | no I mean the same way we'd do for keystone | 16:19 |
| odyssey4me | I prefer that idea, that duplicating a bunch of stuff in this repo | 16:19 |
| odyssey4me | evrardjp yes, that's exactly what I mean | 16:19 |
| odyssey4me | I can prep up a review for it if you like? | 16:19 |
| odyssey4me | cloudnull happy with that? | 16:19 |
| *** markvoelker has joined #openstack-ansible | 16:20 | |
| cloudnull | sounds good to me | 16:20 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Add tests for multi-backend https://review.openstack.org/513443 | 16:23 |
| cloudnull | i pitched it over. | 16:24 |
| cloudnull | I had it mostly done. | 16:24 |
| cloudnull | however I'm happy to abandon it if we've a better way | 16:24 |
| *** markvoelker has quit IRC | 16:24 | |
| cloudnull | maybe we should move the zfs and btrfs backend creation playbooks to the tests repo ? | 16:25 |
| cloudnull | odyssey4me: or are you doing that? | 16:25 |
| *** cshen_ has joined #openstack-ansible | 16:26 | |
| odyssey4me | cloudnull I'll show you an even better solution ;) | 16:27 |
| cloudnull | woot! | 16:27 |
| * cloudnull abanding my copy-pasta :D | 16:28 | |
| *** cshen_ has quit IRC | 16:28 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: [WIP] Clean out old mariadb apt sources https://review.openstack.org/513400 | 16:30 |
| *** cshen_ has joined #openstack-ansible | 16:32 | |
| evrardjp | so FYI today is release day for the M1. | 16:36 |
| evrardjp | I will prepare the release after my dinner | 16:36 |
| evrardjp | great time for nova-lxd to fail. | 16:37 |
| evrardjp | not their fault though | 16:37 |
| *** gkadam-away has quit IRC | 16:48 | |
| evrardjp | I will try to fix it, for the greater good! | 16:49 |
| hwoarang | odyssey4me: evrardjp: so i am finally looking at the zuulv3 stuff. after having read the migration guide i am still not sure where our common jobs are defined | 16:51 |
| *** vnogin has quit IRC | 16:52 | |
| hwoarang | so for example, where is openstack-ansible-functional-centos-7 defined these days? | 16:52 |
| *** phalmos_ has joined #openstack-ansible | 16:52 | |
| odyssey4me | openstack-ansible-tests is where the base jobs are defined | 16:53 |
| odyssey4me | scenario's are defined in repositories that require them | 16:53 |
| hwoarang | odyssey4me: and how does each role know that the job is in the tests repo so it can inherit from there? | 16:54 |
| hwoarang | must have missed that bit in the migration guide | 16:54 |
| odyssey4me | hwoarang zuul's load order specifies that repo high up in the load order (we had to make sure of that) | 16:54 |
| *** phalmos has quit IRC | 16:55 | |
| odyssey4me | so the configs in openstack-ansible-tests is loaded before any of other repositories | 16:55 |
| hwoarang | ok but where do you define that? | 16:55 |
| *** vnogin has joined #openstack-ansible | 16:55 | |
| hwoarang | because the project.yml is in-repo know so you have to tell zuul somehow to go look at -tests for the actual job definition | 16:56 |
| odyssey4me | hwoarang trying to find it - hang on | 16:56 |
| odyssey4me | https://review.openstack.org/512678 | 16:57 |
| hwoarang | ohhhhhh | 16:58 |
| hwoarang | sneaky | 16:58 |
| hwoarang | ok now it all makes sense | 16:58 |
| hwoarang | thank you odyssey4me | 16:59 |
| odyssey4me | sure thing :) we're all learning how to ride this bike... slowly | 17:00 |
| *** cshen_ has quit IRC | 17:02 | |
| *** Random__ has joined #openstack-ansible | 17:07 | |
| Random__ | hey everyone, quick question: do i need to create the "lxcbr0" bridg before hand, or will the ansible-openstack playbooks make that one for me | 17:09 |
| *** shardy has quit IRC | 17:12 | |
| *** vakuznet has joined #openstack-ansible | 17:14 | |
| hwoarang | Random__: iirc it will create one for you | 17:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Add cross-repo base job https://review.openstack.org/513453 | 17:17 |
| *** vnogin has quit IRC | 17:18 | |
| vakuznet | hi, gnocchi upgrade from 15.1.9 is failing "/openstack/venvs/gnocchi-15.1.10/bin/gnocchi-upgrad | 17:19 |
| vakuznet | e", "--create-legacy-resource-types"], "delta": "0:00:01.121827", "end": "2017-10-19 17:13:34.314072", "failed": true, "rc": 1, "start": "2017 | 17:19 |
| vakuznet | -10-19 17:13:33.192245", "stderr": "RuntimeError: module compiled against API version 0xb but this version of numpy is 0xa | 17:19 |
| *** smatzek has joined #openstack-ansible | 17:19 | |
| *** phalmos has joined #openstack-ansible | 17:23 | |
| *** vnogin has joined #openstack-ansible | 17:24 | |
| *** mbuil has quit IRC | 17:24 | |
| *** phalmos_ has quit IRC | 17:26 | |
| vakuznet | is it known issue? | 17:26 |
| *** vnogin has quit IRC | 17:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts master: Add lxc-container-create tests https://review.openstack.org/513457 | 17:30 |
| odyssey4me | cloudnull evrardjp ^ hopefully that does the trick, with the dependent patch | 17:31 |
| *** jonmills has joined #openstack-ansible | 17:31 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set OpenStack-Ansible release to 17.0.0.0b1 https://review.openstack.org/513458 | 17:32 |
| *** smatzek has quit IRC | 17:32 | |
| evrardjp | I'll have a look odyssey4me | 17:33 |
| evrardjp | thanks! | 17:33 |
| jonmills | I'm new to openstack-ansible, and in fact new to ansible. not new to openstack though. i wanted to ask you guys if it's possible to make openstack-ansible deploy openstack without using lxc? | 17:33 |
| odyssey4me | zuul has loaded the jobs, so the config is right - we'll see if it functionally works | 17:34 |
| *** lbragstad has quit IRC | 17:35 | |
| evrardjp | jonmills: yes, on bare metal. You may hit a few bugs when doing 100% on bare metal though | 17:35 |
| evrardjp | but we are happy to help you on the way | 17:35 |
| evrardjp | jonmills: isolation is, although, best ensured with containers | 17:35 |
| evrardjp | odyssey4me: so basically the job is the same, you just change into a different folder | 17:36 |
| odyssey4me | evrardjp yes | 17:36 |
| evrardjp | so both folders have to be checked out, which is easy for a-r-r | 17:36 |
| evrardjp | maybe different with keyston | 17:36 |
| evrardjp | e | 17:36 |
| evrardjp | but still easy | 17:36 |
| odyssey4me | the get-a-r-r playbook detects another checkout there and uses it without changing it | 17:37 |
| jonmills | evrardjp: actually a case where I'd prefer to put the control plane in KVM VMs that I manage outside OpenStack. I'd have the networking set up such that there would be a shared control plane vlan seen by openstack hypervisors and the KVM nodes in question. | 17:37 |
| jonmills | evrardjp: essentially i'm not a lxc fan and just prefer kvm (sorry) | 17:37 |
| evrardjp | jonmills: lxc containers are not like docker containers | 17:37 |
| evrardjp | it's alright :) | 17:37 |
| evrardjp | jonmills: would you prefer systemd-nspawn containers? :p | 17:37 |
| odyssey4me | jonmills sure, but with containers when rabbitmq dies, you just destroy and recreate containers - not the whole control plane ;) | 17:37 |
| odyssey4me | but meh, each to their own :) | 17:38 |
| evrardjp | odyssey4me: yeah :) | 17:38 |
| evrardjp | well | 17:38 |
| evrardjp | it's technically possible | 17:38 |
| jonmills | evrardjp: it's a case where I'm running my cluster/baremetal managed by xCAT. xCAT handles KVM quite nicely for me. | 17:38 |
| evrardjp | jonmills: if you are willing to iron out some of the bugs you see on the way, we'd love to see your contributions. | 17:39 |
| evrardjp | jonmills: I'd suggest you to read the deploy guide once | 17:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513344 | 17:39 |
| evrardjp | then start with putting everything on metal | 17:39 |
| evrardjp | WOOOT | 17:39 |
| evrardjp | NEUTRON PIKE MERGED! | 17:39 |
| evrardjp | the first! | 17:39 |
| *** lbragstad has joined #openstack-ansible | 17:39 | |
| evrardjp | jonmills: this way you get familiar with the process, how ansible works, how we do things with it, etc. | 17:40 |
| spotz | We merged?!!! | 17:40 |
| jonmills | evrardjp: i'm sort of mid-way through reading the deployment guide. i'm searching for some sort of toggle or keyword (to put in a playbook?) to basically say "hey, stop caring about lxc" | 17:40 |
| odyssey4me | evrardjp the third actually ;) | 17:40 |
| evrardjp | is_metal | 17:40 |
| jonmills | aha | 17:40 |
| Random__ | Does anyone know if i can just update the etc/network/interfaces file to make bridges or do i have to run brctl addbr also? | 17:41 |
| odyssey4me | jonmills it's not well documented, and not a simple toggle - evrardjp has been working on implementing something like that - or at least an example config | 17:41 |
| evrardjp | Random__: etc network interfaces just define the bridges, you should either restart, or do them manually | 17:41 |
| evrardjp | I'd restart to check that everything is alright upon reboot :p | 17:41 |
| jonmills | Thank you evrardjp and odyssey4me. | 17:42 |
| logan- | you have to set it to 11 to go full metal, but i think only evrardjp has attained that level | 17:42 |
| Random__ | i mean if i change the file and then restart is that all i need to do | 17:42 |
| evrardjp | logan-: you mean I am the only one willing to throw hardware at it :p hahaha | 17:42 |
| *** vakuznet has quit IRC | 17:42 | |
| *** csmart has quit IRC | 17:43 | |
| *** evrardjp[m] has quit IRC | 17:43 | |
| odyssey4me | every time... https://media.giphy.com/media/gui67fZ3xIneM/giphy.gif | 17:44 |
| logan- | how goes the zuuls? my workstation got messed up so just catching up | 17:44 |
| odyssey4me | logan- we crashed it twice | 17:45 |
| odyssey4me | it only has 16GB of memory, and apparently isn't very good at giving it back to the system when processing job changes... and we're doing a lot of changes | 17:45 |
| logan- | haha nice | 17:45 |
| *** germs has joined #openstack-ansible | 17:46 | |
| odyssey4me | IIRC every time it processes a job change, it uses around 100MB memory... when I threw ~40 patches at once into the mix, it crashed | 17:46 |
| odyssey4me | to be fair, it's handling over 3000 repositories apparently | 17:46 |
| logan- | makes sense i guess that the dynamic reconfiguration stuff can be costly when done in bulk | 17:47 |
| odyssey4me | yeah, so we're trying to be kind and make only a few changes at a time | 17:47 |
| odyssey4me | if you'd like to watch the memory cycle, see http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=63979&rra_id=all | 17:47 |
| *** germs1 has joined #openstack-ansible | 17:47 | |
| logan- | are there some slow providers currently? i notice some integrated jobs are going the normal 1h35m-1h40m and then others timeout constantly | 17:48 |
| logan- | guessing its a slower nodepool provider rather than zuul related stuff | 17:48 |
| odyssey4me | ja, I noticed that too -stuff is a bit slow right now... not sure exactly why, been busy trying to make jobs work | 17:48 |
| logan- | ah wow time for a resize on that instance :D | 17:48 |
| *** pbandark has quit IRC | 17:49 | |
| odyssey4me | yup, they know about it | 17:50 |
| *** vakuznet has joined #openstack-ansible | 17:50 | |
| *** pbandark has joined #openstack-ansible | 17:50 | |
| odyssey4me | zuul v4's plans included being able to scale horizontally... for now though it's a vertically scaled monolith | 17:50 |
| evrardjp | if you fancy some votes on the first queens milestone it's here: https://review.openstack.org/#/c/513458/ | 17:50 |
| odyssey4me | the scheduler specifically | 17:50 |
| *** germs has quit IRC | 17:51 | |
| evrardjp | in the meantime, I am gone for playing a board game. I will be back in an hour or so. | 17:51 |
| odyssey4me | evrardjp if that passes the gate, I wonder if it's worth the bother actually merging it - you could just do the release from your checkout | 17:52 |
| odyssey4me | not sure why I didn't think of that before | 17:52 |
| evrardjp | I will do a release from my checkout | 17:53 |
| odyssey4me | oh, because it won't actually release a thing that works - never mind | 17:53 |
| evrardjp | but after :p | 17:54 |
| evrardjp | ok see you laterz ! | 17:54 |
| *** poopcat has joined #openstack-ansible | 18:00 | |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening stable/pike: Zuulv3 migration https://review.openstack.org/513461 | 18:02 |
| cloudnull | Random__: if these are new bridges you can just edit the file | 18:03 |
| cloudnull | and then run ifup | 18:03 |
| cloudnull | if the bridges exist you will need to add them to the file and run (brctl || ip) to make the changes live | 18:04 |
| spotz | Random__ you RandomTech from yesterday? | 18:05 |
| *** albertcard1 has joined #openstack-ansible | 18:05 | |
| *** germs1 is now known as germs | 18:06 | |
| Random__ | yes i put a space in my name and it borked | 18:07 |
| *** poopcat has quit IRC | 18:07 | |
| Random__ | spotz: ^ | 18:07 |
| spotz | Random__: Might want to let Cloudnull the whole plan, he'll be able to help more:) | 18:07 |
| spotz | know the.. can't type | 18:08 |
| cloudnull | :P | 18:08 |
| Random__ | i was just figuring out how to write the task file efficently | 18:08 |
| spotz | cloudnull: He's trying to do an ansible play that will automatically setup our roles and run IF I remember correctly from Random__ yesterday | 18:09 |
| Random__ | ya im making a playbook to aid with the network setup as these blades are being done from scratch | 18:10 |
| *** electrofelix has quit IRC | 18:10 | |
| jrosser | Random__: weve done this and its horrific | 18:10 |
| Random__ | im just talking about the bridges | 18:10 |
| Random__ | i have everything but the task to do the templating | 18:11 |
| jrosser | well yes, but theres a lot of banana skins waiting for you hidden in the ifup/down commands | 18:11 |
| jrosser | we pxeboot a server then set up all the required interfaces and bridges with a playbook | 18:11 |
| jrosser | and its been very hard to make reliable/repeatable | 18:12 |
| Random__ | have you tried doing a networking restart instead of ifup? | 18:12 |
| jrosser | oh yes | 18:12 |
| Random__ | so sudo systemctl restart networking.service | 18:12 |
| jrosser | and had to rewrite that systemd service too | 18:12 |
| jrosser | boxrick: we need to opensource our network role | 18:13 |
| Random__ | what os were you using? | 18:14 |
| spotz | And this is why I suggested joining us over here yesterday Random__. We've got the best/helpful folks around:) | 18:16 |
| jrosser | Random__: ubuntu | 18:17 |
| Random__ | its stange that the openstack-ansible channel isnt joined by default | 18:17 |
| jrosser | if youre in centos im told youll jave a much easier time | 18:17 |
| *** Random__ is now known as RandomTech | 18:17 | |
| cloudnull | Random__: if you want something that can be used more universally without having to deal with net-config you could use something like: https://github.com/osic/osic-clouds/blob/master/nextgen/scripts/vxlan-mesh-create.sh | 18:17 |
| RandomTech | nope ubuntu 16 | 18:17 |
| spotz | Well we're a project just like keystone so if you don't have an OSA related question you may not want to be here | 18:18 |
| jrosser | ok well busy for a couple if hours but let me try to get stuff on github later | 18:18 |
| cloudnull | we used that script in the OSIC to automate a mesh network setup for hosts deployed within ironic | 18:18 |
| jrosser | its mad that weve workedtheough all this and not shared it | 18:18 |
| spotz | Bad jrosser!! | 18:18 |
| logan- | jrosser: have you tried networkd? wonder if things get any easier that way | 18:19 |
| jrosser | perhaps, we got badly stung by there being hiddden state behing the files inimterface.d | 18:19 |
| jrosser | and residual ip addresses needing to be flished manually of interfaces | 18:20 |
| logan- | gotcha | 18:20 |
| jrosser | sorry phone typing, ill try to link to our stuff later | 18:20 |
| RandomTech | jrosser: did you make sure you had bridge-utils first? | 18:20 |
| *** rodolof has quit IRC | 18:23 | |
| *** rodolof has joined #openstack-ansible | 18:24 | |
| boxrick | RandomTech: oh yes, all there. | 18:24 |
| RandomTech | hmm | 18:24 |
| RandomTech | what kind of error did you run into? | 18:24 |
| jrosser | RandomTech: boxrick is your man for this | 18:27 |
| RandomTech | boxrick: what error did you run into? | 18:27 |
| *** rodolof has quit IRC | 18:28 | |
| *** cshen_ has joined #openstack-ansible | 18:33 | |
| odyssey4me | logan- mind giving this a review? https://review.openstack.org/#/c/496678/ | 18:34 |
| logan- | on it | 18:34 |
| odyssey4me | thx | 18:36 |
| *** stuartgr has quit IRC | 18:37 | |
| *** galstrom_zzz is now known as galstrom | 18:37 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts master: Add lxc-container-create tests https://review.openstack.org/513457 | 18:38 |
| odyssey4me | ok, I'm out for the night - cheers all | 18:39 |
| *** cshen_ has quit IRC | 18:40 | |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening stable/pike: Zuulv3 migration https://review.openstack.org/513461 | 18:41 |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening master: Remove queue from zuulv3 gate https://review.openstack.org/513475 | 18:41 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: [WIP] Clean out old mariadb apt sources https://review.openstack.org/513400 | 18:43 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_swift stable/pike: Suppress rsync warnings w/command https://review.openstack.org/501263 | 18:46 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-galera_server stable/pike: Suppress command/shell warnings https://review.openstack.org/511935 | 18:47 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-galera_server stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513426 | 18:47 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-galera_server stable/pike: Suppress command/shell warnings https://review.openstack.org/511935 | 18:47 |
| chyka | hey all, i'm still working on that issue with booting from the provider network. if i console into an instance and manually add the static IP it works. But at boot is jut repeats "trying to raise network interfaces" till it times out | 18:48 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_cinder stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513423 | 18:48 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_tempest stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513424 | 18:48 |
| chyka | and its driving me crazy | 18:48 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_horizon stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513425 | 18:49 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513427 | 18:49 |
| vakuznet | 15.1.10 gnocchi upgrade error: https://pastebin.com/n9drgHE5 | 18:49 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_heat stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513422 | 18:49 |
| cloudnull | chyka: meta data / dhcp getting into the instance? | 18:49 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_glance stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/513421 | 18:49 |
| chyka | yeah, it gets up to dhcprequest but nothing after that | 18:50 |
| chyka | so there's some back an forth, but it doesn't go any further than that | 18:51 |
| jrosser | RandomTech: just working on getting our repo public | 18:51 |
| *** cshen_ has joined #openstack-ansible | 18:53 | |
| *** smatzek has joined #openstack-ansible | 19:05 | |
| *** smatzek has quit IRC | 19:07 | |
| *** smatzek has joined #openstack-ansible | 19:08 | |
| *** smatzek has quit IRC | 19:12 | |
| chyka | yeah there's dhcp requests but then no acknowledgement. then after failing to raise the interface cloud-init doesn't set an IP so the instance cant access any metadata. | 19:13 |
| RandomTech | GHOST RandomTech | 19:16 |
| RandomTech | oops :p | 19:16 |
| RandomTech | i think i have the task writen up for the networking | 19:17 |
| *** pbandark has quit IRC | 19:18 | |
| jrosser | RandomTech: looking at our code, theres a few subtleys. The order things are brought up/down in interfaces.d/ is not what you'd necessarily want or like. | 19:19 |
| jrosser | in addition, if you change any of the contents of interfaces.d and bring interface up/down afterwards you are in for bad times(tm) | 19:20 |
| RandomTech | im not changing interfaces.d im changing interfaces | 19:20 |
| jrosser | none of this generally matters in simple "eth0 dhcp" setups | 19:21 |
| RandomTech | were not using dhcp either | 19:21 |
| jrosser | but if you try to configure a quite complex set of interfaces, bonds and bridges as you might when following the OSA deply guide it becomes very unreliable for the reasons above | 19:21 |
| RandomTech | im following this: https://docs.openstack.org/project-deploy-guide/openstack-ansible/draft/app-config-test.html | 19:22 |
| jrosser | sure | 19:23 |
| *** newmember has joined #openstack-ansible | 19:23 | |
| jrosser | we've done the same, and attempted to entirely automate/ansible-ise the preparation of the host systems before running OSA | 19:24 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Update upgrade role for Queens from P https://review.openstack.org/496678 | 19:25 |
| jrosser | RandomTech: https://github.com/bbc/network_configure | 19:25 |
| jrosser | cloudnull: ^^ | 19:25 |
| jrosser | logan-: ^^ thats what we ended up having to do | 19:29 |
| logan- | thanks jrosser | 19:29 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Add BTRFS cached container task https://review.openstack.org/513265 | 19:36 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: Fix BTRFS package name https://review.openstack.org/513157 | 19:40 |
| *** dave-mccowan has quit IRC | 19:49 | |
| *** dave-mccowan has joined #openstack-ansible | 19:51 | |
| *** inc0 has joined #openstack-ansible | 19:51 | |
| inc0 | hey there, you guys remember discussion about healthchecks on ptg? | 19:52 |
| mhayden | yeah | 19:52 |
| inc0 | wanna do some design of it next week? First project we want to prototype it will be ironic | 19:52 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests stable/pike: Add Fedora support to bindep install step https://review.openstack.org/513487 | 19:55 |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening stable/pike: Zuulv3 migration https://review.openstack.org/513461 | 19:55 |
| openstackgerrit | Daniel Speichert proposed openstack/openstack-ansible-lxc_hosts master: Stop trying to set MTU on LXC bridge interface https://review.openstack.org/512789 | 19:59 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 20:01 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Initial integrated zuulv3 jobs https://review.openstack.org/513406 | 20:02 |
| *** dave-mccowan has quit IRC | 20:02 | |
| *** smatzek has joined #openstack-ansible | 20:04 | |
| *** weezS has quit IRC | 20:06 | |
| *** weezS has joined #openstack-ansible | 20:06 | |
| *** vnogin has joined #openstack-ansible | 20:10 | |
| chyka | any ideas about that provider network? :( | 20:14 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: [TEST] Update Ansible to 2.4.1.0-0.4.rc2 https://review.openstack.org/501814 | 20:14 |
| *** asb47_ has quit IRC | 20:20 | |
| cloudnull | jrosser: boxrick: very nice! | 20:23 |
| *** vnogin has quit IRC | 20:24 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Constrain ansible when installing ARA https://review.openstack.org/513437 | 20:26 |
| *** asb47 has joined #openstack-ansible | 20:27 | |
| *** RandomTech has quit IRC | 20:27 | |
| jrosser | cloudnull: you wouldnt beleive the grief we've had with ubuntu networking :( to the point we've discussed changing to centos just on that basis | 20:28 |
| cloudnull | I can imagine :) | 20:28 |
| jrosser | that role is used on all our boostrap, deploy, infra, ceph, everything so theres some very complex data gets fed in | 20:29 |
| cloudnull | I'm hopeful things like networkd will eventually be good enough to universally replace distro specific network scripts, but until then... its kinda a PITA | 20:29 |
| cloudnull | the role looks super useful. | 20:30 |
| jrosser | we also have other values for device_type like 'ipmi' and 'switch' which are consumed in our dhcp and tftp roles | 20:30 |
| jrosser | but they all just get skipped in that one for host interfaces, which is quite neat | 20:30 |
| SamYaple | cloudnull: ive beed using networkd for a couple years now to replace distro networking | 20:31 |
| SamYaple | cloudnull: im down to *one* systemd oneshot service to implment some ip rule stuff | 20:31 |
| cloudnull | SamYaple: I've tried a couple of times, things like bonding and vlan handling has been a bit suspect in some applications. | 20:31 |
| cloudnull | but ++ i have networkd all running at home. its seems to +/- work | 20:32 |
| SamYaple | suspect as you think its been setup wrong? or it just didnt work | 20:32 |
| cloudnull | it didn't work. | 20:32 |
| SamYaple | yea i know what you mean. its not feature complete yet | 20:32 |
| cloudnull | or it did, until systemd release x.y.z came out and it was broken . | 20:33 |
| SamYaple | well that seems just broken. i do bond+vlan at my house | 20:33 |
| *** vnogin has joined #openstack-ansible | 20:33 | |
| *** jwitko_ has joined #openstack-ansible | 20:33 | |
| cloudnull | its been a ride | 20:33 |
| SamYaple | a fun one though. i really like networkd | 20:33 |
| SamYaple | but agreed, it has some growing up to do still | 20:33 |
| cloudnull | I too like it. | 20:33 |
| cloudnull | I think its going to be great given a couple more iterations | 20:33 |
| *** jwitko has quit IRC | 20:36 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Tune-up the MNAIO for physical and virtual hosts https://review.openstack.org/511985 | 20:36 |
| *** jwitko_ has quit IRC | 20:40 | |
| *** asb47 has quit IRC | 20:41 | |
| *** asb47 has joined #openstack-ansible | 20:42 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set OpenStack-Ansible release to 17.0.0.0b1 https://review.openstack.org/513458 | 20:46 |
| *** woodard has quit IRC | 20:46 | |
| *** cshen_ has quit IRC | 20:46 | |
| Neptu | hej | 20:49 |
| Neptu | Im trying to make a first instalation of openstack-ansible and I have 5 nodes on virtualbox | 20:49 |
| Neptu | im looking to the apendix https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-test.html | 20:49 |
| Neptu | but im not so sure why I need this networks and how to configure them in advance | 20:50 |
| *** snowman48391 has joined #openstack-ansible | 20:55 | |
| *** snowman4839 has quit IRC | 20:55 | |
| spotz | Neptu That's an example config with our 'example' IPs | 21:01 |
| *** jwitko has joined #openstack-ansible | 21:03 | |
| Neptu | spotz: but why 3 bridges are required | 21:03 |
| Neptu | i guess i still a newbie on openstack and those are obvious questions | 21:03 |
| Neptu | but for me they ar enot | 21:03 |
| spotz | Neptu - So the management network is for the hosts to talk amongst themselves. The storage network is for the transfer of storage which can be bandwidth intensive. The vxlan in this case could be considered your 'public; network. The guys will correct me if wrong with that interpretation | 21:06 |
| spotz | neptu - now things aren't going to quite match up here as this doesn't have quite the same networks and no storage network but does this help to visualize? https://docs.openstack.org/newton/networking-guide/deploy-ovs-selfservice.html | 21:11 |
| *** markvoelker has joined #openstack-ansible | 21:13 | |
| *** gouthamr has quit IRC | 21:15 | |
| *** thorst has quit IRC | 21:16 | |
| *** rodolof has joined #openstack-ansible | 21:17 | |
| *** thorst has joined #openstack-ansible | 21:18 | |
| *** dave-mccowan has joined #openstack-ansible | 21:23 | |
| *** thorst has quit IRC | 21:23 | |
| *** jonmills has quit IRC | 21:25 | |
| *** smatzek has quit IRC | 21:26 | |
| *** dave-mccowan has quit IRC | 21:29 | |
| Neptu | spotz: is helping | 21:33 |
| spotz | neptu yeah sometimes a visiual makes things clearer | 21:33 |
| Neptu | spotz: in that sense you are basically spliting the networking to avoid problems | 21:34 |
| Neptu | ok ok | 21:34 |
| Neptu | now i need to understand on how to setup the target hosts | 21:34 |
| spotz | neptu Correct except for the 'public' or tunnel network | 21:34 |
| Neptu | building a vagrant ansible script to pre-setup my test enviroment | 21:34 |
| spotz | Which is going to be your outside IPs/access/etc | 21:35 |
| spotz | I believe you can do multi-node with the AIO and that's documented.. But don't quote me I'd have to look and confirm | 21:35 |
| Neptu | ok so you telling me you have a ipv4 internal and the tunnel should give "external" ips | 21:36 |
| Neptu | for the vms's | 21:36 |
| *** dave-mccowan has joined #openstack-ansible | 21:37 | |
| spotz | Well it will be where your external floating IPs are from | 21:37 |
| spotz | I've been doing a lot of provider networks or simple private/public ones lately so slight chance of misinterpretation | 21:37 |
| Neptu | spotz: the idea is to use this as test... this are my first steps | 21:39 |
| Neptu | gonna try to configure the machines | 21:39 |
| Neptu | I guess I need to define what nodes are doing what | 21:40 |
| spotz | neptu You could start with an AIO then to get a feel then go more multi-node | 21:40 |
| jrosser | spotz: i have found those examples a little confusing becasue they dont explicitly talk about where floating ip etc are allocated from | 21:40 |
| *** woodard has joined #openstack-ansible | 21:40 | |
| Neptu | in my case i have a vagrant script that sizes to n so i guess one controller node one storage and many compute | 21:40 |
| Neptu | AIO?? | 21:40 |
| spotz | jrosser: If you want to bug it I can try to get to it. It'd be wishlisty in my free time | 21:41 |
| spotz | Neptu: AIO = All in One so everything would be on one host | 21:41 |
| jrosser | it may just be that its not set up, but doesnt say that | 21:41 |
| Neptu | spotz: so is like a devstack instance | 21:41 |
| Neptu | I saw a Vagrantfile aswell | 21:41 |
| Neptu | so wonder what that does | 21:42 |
| spotz | jrosser: Or we forgot and no one's mentioned it:) | 21:42 |
| spotz | neptu correct | 21:42 |
| *** dave-mccowan has quit IRC | 21:44 | |
| *** albertcard1 is now known as poopcat | 21:44 | |
| spotz | Neptu probably has an AIO in that | 21:45 |
| *** thorst has joined #openstack-ansible | 21:47 | |
| Neptu | spotz: gonna check the code | 21:47 |
| Neptu | but basically you clone the repo and you run the boostrap-aoi script? | 21:47 |
| Neptu | and that should give you a devstack?? | 21:47 |
| *** markvoelker has quit IRC | 21:48 | |
| spotz | neptu Well not devstack but an all inclusive installation via OSA into LXC containers like you would have if you did an OSA installation | 21:48 |
| spotz | but all on one machine | 21:49 |
| Neptu | cool is a good start | 21:49 |
| *** thorst has quit IRC | 21:51 | |
| Neptu | ./scripts/gate-check-commit.sh | 21:51 |
| Neptu | seems to run this | 21:51 |
| *** csmart has joined #openstack-ansible | 21:57 | |
| *** jwitko has quit IRC | 22:02 | |
| *** markvoelker has joined #openstack-ansible | 22:02 | |
| *** markvoelker has quit IRC | 22:06 | |
| *** markvoelker has joined #openstack-ansible | 22:11 | |
| *** markvoelker has quit IRC | 22:16 | |
| *** markvoelker has joined #openstack-ansible | 22:20 | |
| *** phalmos has quit IRC | 22:23 | |
| *** thorst has joined #openstack-ansible | 22:23 | |
| *** evrardjp[m] has joined #openstack-ansible | 22:23 | |
| *** markvoelker has quit IRC | 22:25 | |
| *** thorst_ has joined #openstack-ansible | 22:26 | |
| *** thorst has quit IRC | 22:27 | |
| *** markvoelker has joined #openstack-ansible | 22:29 | |
| *** thorst_ has quit IRC | 22:30 | |
| *** galstrom is now known as galstrom_zzz | 22:32 | |
| *** snowman48391 has quit IRC | 22:33 | |
| *** markvoelker has quit IRC | 22:34 | |
| *** weezS has quit IRC | 22:34 | |
| *** snowman4839 has joined #openstack-ansible | 22:36 | |
| *** rodolof has quit IRC | 22:38 | |
| *** markvoelker has joined #openstack-ansible | 22:39 | |
| *** markvoelker has quit IRC | 22:43 | |
| *** markvoelker has joined #openstack-ansible | 22:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Constrain ansible when installing ARA https://review.openstack.org/513298 | 22:53 |
| *** dave-mccowan has joined #openstack-ansible | 22:56 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Include the Ansible version variable when installing ara https://review.openstack.org/513287 | 22:57 |
| *** thorst has joined #openstack-ansible | 23:07 | |
| *** weezS has joined #openstack-ansible | 23:20 | |
| *** markvoelker has quit IRC | 23:22 | |
| *** smatzek has joined #openstack-ansible | 23:27 | |
| *** chyka has quit IRC | 23:29 | |
| *** smatzek has quit IRC | 23:31 | |
| *** weezS has quit IRC | 23:34 | |
| *** markvoelker has joined #openstack-ansible | 23:36 | |
| *** markvoelker has quit IRC | 23:40 | |
| *** newmember has quit IRC | 23:41 | |
| *** markvoelker has joined #openstack-ansible | 23:45 | |
| *** smatzek has joined #openstack-ansible | 23:46 | |
| *** markvoelker has quit IRC | 23:49 | |
| *** markvoelker has joined #openstack-ansible | 23:54 | |
| *** markvoelker has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
