| *** gyee has quit IRC | 00:11 | |
| *** cshen has joined #openstack-ansible | 00:19 | |
| *** cshen has quit IRC | 00:23 | |
| *** dmsimard1 has joined #openstack-ansible | 00:50 | |
| *** dmsimard has quit IRC | 00:51 | |
| *** dmsimard1 is now known as dmsimard | 00:52 | |
| *** rh-jelabarre has quit IRC | 01:53 | |
| *** cshen has joined #openstack-ansible | 02:19 | |
| *** cshen has quit IRC | 02:23 | |
| *** d34dh0r53 has joined #openstack-ansible | 03:56 | |
| *** d34dh0r53 has quit IRC | 04:02 | |
| *** d34dh0r53 has joined #openstack-ansible | 04:02 | |
| *** d34dh0r53 has joined #openstack-ansible | 04:04 | |
| *** d34dh0r53 has quit IRC | 04:16 | |
| *** evrardjp has quit IRC | 04:35 | |
| *** evrardjp has joined #openstack-ansible | 04:35 | |
| *** udesale has joined #openstack-ansible | 05:10 | |
| *** udesale_ has joined #openstack-ansible | 05:21 | |
| *** udesale has quit IRC | 05:24 | |
| *** thuydang has joined #openstack-ansible | 05:58 | |
| *** udesale_ has quit IRC | 06:14 | |
| *** cshen has joined #openstack-ansible | 06:21 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest stable/stein: Pin neutron-tempest-plugin to 0.9.0 for Stein branch. https://review.opendev.org/721320 | 06:22 |
|---|---|---|
| jrosser | noonedeadpunk: on https://review.opendev.org/#/c/721319/1/vars/main.yml we could make it use master on py > 2.7, but i thought the complexity wasnt worth it, and just pinning to 0.9.0 regardless of python version would be ok | 06:24 |
| jrosser | interested to know if you think there is a better way there - perhaps we could define a var in vars/{distro}.yml which sets the version to use? | 06:25 |
| jrosser | i think it't centos on 2.7 and debian-stable on 3.5 which fail with 1.0.0 | 06:25 |
| *** udesale has joined #openstack-ansible | 06:37 | |
| *** udesale has quit IRC | 06:43 | |
| noonedeadpunk | I don't see really anything bad about just bumping version that way... | 06:49 |
| noonedeadpunk | It's simple and clean now, so I'd say it's ok | 06:50 |
| *** this10nly has joined #openstack-ansible | 06:56 | |
| *** rpittau|afk is now known as rpittau | 07:03 | |
| *** cshen_ has joined #openstack-ansible | 07:06 | |
| *** cshen has quit IRC | 07:09 | |
| *** yolanda has joined #openstack-ansible | 07:13 | |
| jrosser | i'm just trying to see why the master SHA bump fails everywhere - it's the same tempest keystone error that we fixed with this https://review.opendev.org/#/c/720698/ for bionic/distro | 07:26 |
| jrosser | whats not so clear is why thats failing still everywhere else even though the tempest SHA has been bumped | 07:26 |
| noonedeadpunk | jrosser: yeah, that's what I didn't get actually. As SHA you've provided should fix tempest and make this test not fail | 07:27 |
| jrosser | i just looked in the venv build log for the utility container and it seems to build the right SHA there | 07:28 |
| *** fghaas has joined #openstack-ansible | 07:28 | |
| noonedeadpunk | yeah, just thought to check that actually:) | 07:28 |
| noonedeadpunk | ofc we can adjust 720698 to disable this test everywhere.... | 07:29 |
| jrosser | i'm assuming this is doing the right thing https://zuul.opendev.org/t/openstack/build/a6635b2a0b664da88eb4942f63544d6f/log/logs/openstack/aio1_utility_container-ab91122a/python_venv_build.log.txt#21534-21563 | 07:31 |
| jrosser | something is failing here https://zuul.opendev.org/t/openstack/build/a6635b2a0b664da88eb4942f63544d6f/log/logs/openstack/aio1_utility_container-ab91122a/python_venv_build.log.txt#40455-40484 | 07:38 |
| noonedeadpunk | huh | 07:39 |
| noonedeadpunk | I think that https://opendev.org/openstack/tempest/src/commit/30638b53a4f9a277a821f7988d8c18b6de52bf77/tempest/api/identity/admin/v3/test_groups.py is just not supposed to work lol | 07:39 |
| noonedeadpunk | so we're deleting object in for cycle | 07:39 |
| noonedeadpunk | but for comparison use original object | 07:39 |
| jrosser | oh what | 07:39 |
| jrosser | you think the actual tempest code is broken? | 07:39 |
| noonedeadpunk | Yeah, how is that supposed to work? https://opendev.org/openstack/tempest/src/commit/30638b53a4f9a277a821f7988d8c18b6de52bf77/tempest/api/identity/admin/v3/test_groups.py#L123 | 07:40 |
| noonedeadpunk | so in 'g' this key won't be defined, but I think it won't reflect user_groups? | 07:40 |
| noonedeadpunk | or will... | 07:40 |
| noonedeadpunk | sorry. my fault | 07:41 |
| noonedeadpunk | Just forgot that it really operates on this object and don't create it's copy... | 07:43 |
| noonedeadpunk | so disregard | 07:43 |
| *** tosky has joined #openstack-ansible | 07:44 | |
| noonedeadpunk | huh, you know, despite it seems that the correct version of tempest gets isntalled, it fails on L118 https://zuul.opendev.org/t/openstack/build/a6635b2a0b664da88eb4942f63544d6f/log/logs/openstack/aio1_utility_container-ab91122a/utility/tempest_run.log.txt#39 | 07:46 |
| noonedeadpunk | and L118 is exactly where assert was before that patch https://opendev.org/openstack/tempest/commit/30638b53a4f9a277a821f7988d8c18b6de52bf77 | 07:47 |
| jrosser | yes membership_expires_at is still ther | 07:47 |
| jrosser | e | 07:47 |
| noonedeadpunk | the point is, that if we had this patch on board, it would fail on L124 | 07:48 |
| jrosser | yes i see | 07:48 |
| jrosser | theres something else i noticed which is not cool | 07:49 |
| jrosser | like, wtf https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L692 | 07:49 |
| noonedeadpunk | o_O | 07:50 |
| jrosser | i'm wondering if this prevents the wheel we build getting installed, because ours is too new | 07:50 |
| noonedeadpunk | Like according to log is's supposed to get isntalled.... but dunno | 07:50 |
| noonedeadpunk | *it's | 07:50 |
| jrosser | but as you say the assert is on the line for the *old* version..... | 07:51 |
| jrosser | or maybe what you get from pypi.... | 07:51 |
| noonedeadpunk | yeah, so it feels that we somehow get an old version.... but I'm not really sure how.... | 07:52 |
| noonedeadpunk | so we kinda supposed to drop u-c for master but have them for stable branches... | 07:53 |
| noonedeadpunk | (in terms of tempest) | 07:53 |
| jrosser | i feel that may break things | 07:54 |
| jrosser | in other surprising ways | 07:54 |
| *** threestrands has quit IRC | 07:55 | |
| jrosser | this is odd enough it deserves an AIO i think | 07:55 |
| *** thuydang has quit IRC | 07:56 | |
| *** KeithMnemonic has quit IRC | 08:05 | |
| *** udesale has joined #openstack-ansible | 08:16 | |
| *** udesale_ has joined #openstack-ansible | 08:22 | |
| *** udesale has quit IRC | 08:22 | |
| *** udesale has joined #openstack-ansible | 08:22 | |
| *** gshippey has joined #openstack-ansible | 09:10 | |
| *** sshnaidm|afk is now known as sshnaidm | 09:54 | |
| *** thuydang has joined #openstack-ansible | 09:58 | |
| *** jbadiapa has joined #openstack-ansible | 10:12 | |
| *** aedc has joined #openstack-ansible | 10:20 | |
| *** aedc has quit IRC | 10:30 | |
| *** rpittau is now known as rpittau|bbl | 10:31 | |
| *** aedc has joined #openstack-ansible | 10:32 | |
| *** thuydang has quit IRC | 10:40 | |
| jrosser | noonedeadpunk: doh! http://paste.openstack.org/show/792457/ | 10:45 |
| *** thuydang has joined #openstack-ansible | 10:58 | |
| *** udesale has quit IRC | 11:09 | |
| openstackgerrit | amolkahat proposed openstack/openstack-ansible-os_tempest master: Added tempest ironic resources setup. https://review.opendev.org/720705 | 11:52 |
| *** rh-jelabarre has joined #openstack-ansible | 11:53 | |
| *** thuydang has quit IRC | 11:57 | |
| jrosser | noonedeadpunk: i found something else wrong in the tempest role, https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install.yml#L43 | 12:02 |
| jrosser | that var appears to apply to the other times we use python_venv_build in that role | 12:03 |
| noonedeadpunk | oh, so you meen we upgrade tempest once install stackviz... | 12:06 |
| noonedeadpunk | ......... | 12:06 |
| noonedeadpunk | what a mess... | 12:06 |
| noonedeadpunk | why do we do this separatyle | 12:07 |
| noonedeadpunk | to fit both source and distro isntalls I guess... | 12:07 |
| jrosser | i'm not sure what is going on right now tbh | 12:10 |
| jrosser | first off that var being set to 'no' means we are not using the repo server at all for the tempest wheel build | 12:11 |
| noonedeadpunk | I think it's just stackviz build | 12:12 |
| noonedeadpunk | As we don install tempest in tempest_install_source.yml | 12:12 |
| noonedeadpunk | (which is included previously) | 12:12 |
| jrosser | no no | 12:12 |
| jrosser | i added debug | 12:12 |
| jrosser | that value 'no' is also appying to all the things in here https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install.yml#L32-L34 | 12:13 |
| jrosser | we have overridden the defaults/main.yml expression for venv_wheel_build_enable with a static 'no' for the whole play i think\ | 12:13 |
| noonedeadpunk | hm, but we don't have it defined here https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install_source.yml#L16-L28 | 12:16 |
| jrosser | i just added venv_wheel_build_enable: "{{ venv_build_host != inventory_hostname }}" to there | 12:17 |
| jrosser | and it makes the behaviour as expected | 12:17 |
| noonedeadpunk | and we have no on stackviz as we install already build package | 12:17 |
| jrosser | yes | 12:17 |
| jrosser | so now i am going to force rebuild of the tempest venv and see what version we get in there | 12:17 |
| jrosser | this feels like a bad use of ansible we have | 12:18 |
| jrosser | using something that is defined in defaults/main.yml also somewhere in vars: | 12:18 |
| noonedeadpunk | Out of http://codesearch.openstack.org/?q=venv_wheel_build_enable it doesn't seem we could somewhere "globally" override it for tempest job... | 12:19 |
| noonedeadpunk | so it's totally weird | 12:19 |
| jrosser | but i think this is a general problem when trying to include a role more than one time in a particular play | 12:19 |
| jrosser | the vars: you set on each time you include the role are not local to that include | 12:19 |
| jrosser | i neater solution would be to break the stackviz install out into its own play, rahter that just it's own tasks | 12:20 |
| noonedeadpunk | what if... to use include instead of import? Wouldn't it change behaviour? | 12:21 |
| noonedeadpunk | btw was wondering to test speed of import vs include - maybe we have problems because of that... | 12:22 |
| *** ygk_12345 has joined #openstack-ansible | 12:24 | |
| jrosser | ok so now i see the right wheel has been built on the repo server for tempest | 12:26 |
| jrosser | but we do not get the right wheel installed into the venv | 12:27 |
| *** rpittau|bbl is now known as rpittau | 12:31 | |
| jrosser | oh hello Uninstalling tempest-23.0.1.dev92:\n Successfully uninstalled tempest-23.0.1.dev92\nSuccessfully installed cinder-tempest-plugin-0.3.2.dev5 tempest-23.0.0\n" | 12:40 |
| ygk_12345 | can someone look into this https://bugs.launchpad.net/openstack-ansible/+bug/1874067 | 12:43 |
| openstack | Launchpad bug 1874067 in openstack-ansible "Adding new infra host is failing" [Undecided,New] | 12:43 |
| jrosser | noonedeadpunk: so it all gores wrong here https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install_source.yml#L43-L57 | 12:44 |
| jrosser | when we install the plugins, it downgrades tempest due to u-c | 12:45 |
| *** _kukacz has joined #openstack-ansible | 12:45 | |
| *** _kukacz has quit IRC | 12:51 | |
| *** _kukacz has joined #openstack-ansible | 12:52 | |
| *** jbadiapa has quit IRC | 13:00 | |
| *** thuydang has joined #openstack-ansible | 13:01 | |
| noonedeadpunk | I don't think we can omit using u-c.... And I don;t see a solution either... | 13:07 |
| noonedeadpunk | except asking why tempest is in u-c:) | 13:07 |
| noonedeadpunk | jrosser: I'm not really sure, but what if we just place "tempest" somewhere here? https://opendev.org/openstack/openstack-ansible/src/branch/master/global-requirement-pins.txt | 13:08 |
| noonedeadpunk | would it mean to install jsut whatever version we want? | 13:08 |
| jrosser | what i was just looking at was combining the installing of the plugins from source into the same task as installing tempest | 13:09 |
| jrosser | in some ways the problem is this https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_install_source.yml#L43-L57 | 13:09 |
| jrosser | we use u-c there | 13:09 |
| noonedeadpunk | yeah, I got it. But I'm not sure we can easily combine it into a single task? | 13:10 |
| noonedeadpunk | While u-c are kinda needed for plugins (otherwise they were failing to install iirc) | 13:10 |
| jrosser | so i was thinking to add another variable to here https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_install_source.yml#L24 | 13:11 |
| jrosser | something like venv_pip_packages: "{{ tempest_pip_packages + tempest_pip_plugins }}" | 13:12 |
| noonedeadpunk | oh, why don't we use tempest_git_constraints https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_install_source.yml#L50 ? | 13:13 |
| noonedeadpunk | So it would be the same contraints we use for venv build | 13:13 |
| noonedeadpunk | https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/defaults/main.yml#L51-L53 | 13:13 |
| noonedeadpunk | and temepst will be isntalled from correct SHA then | 13:14 |
| jrosser | that will still include u-c though | 13:14 |
| jrosser | i can try that easily though | 13:15 |
| kleini | sorry for interrupting tempest discussion. what do I wrong in configuration if the tunnel network address of network hosts is not correctly read for the dynamic inventory? The network host in openstack_inventory.json does not have a address for the network tunnel_address. and that makes the deployment of the ML2 plugin wrong in regards to VXLAN tunnels. | 13:15 |
| noonedeadpunk | yes, but tempest egg should have precedence I guess | 13:15 |
| jrosser | kleini: normally that is for you to set the ip address on br-vxlan yourself as part of provisioning the hosts | 13:16 |
| jrosser | the only thing that the inventory really knows about is the ip address that ansible uses to ssh to things | 13:17 |
| noonedeadpunk | ygk_12345: that is pretty weird, as lxcbr is smth that lxc has by default... | 13:17 |
| noonedeadpunk | do you have that bridge at all? | 13:17 |
| ygk_12345 | noonedeadpunk yes its lxcbr0 | 13:17 |
| kleini | okay, so the IP address for the tunnel network is read from the br-vxlan bridge? | 13:18 |
| noonedeadpunk | kleini: or from whatever interface you provided for vxlan network | 13:18 |
| jrosser | kleini: you should check out how this is working http://codesearch.openstack.org/?q=dynamic-address-fact&i=nope&files=&repos= | 13:20 |
| noonedeadpunk | ygk_12345: tbh I has never seen such behaviour... can only think that it's because of the limit.... | 13:20 |
| ygk_12345 | noonedeadpunk what limit ? | 13:20 |
| noonedeadpunk | `--limit "c3v*"` | 13:21 |
| ygk_12345 | noonedeadpunk thats the new host | 13:21 |
| ygk_12345 | c3v | 13:21 |
| noonedeadpunk | so we generally include lxc_hosts group tyo the limit while creating containers | 13:21 |
| kleini | I already read the source of that common-task but I still don't understand, why it resolves to the IP on the management interface. it must be some of the fallbacks | 13:26 |
| ygk_12345 | noonedeadpunk the issue still persists | 13:32 |
| noonedeadpunk | ygk_12345: I think you should delete created containers before trying to re-run | 13:32 |
| noonedeadpunk | (on the new host) | 13:32 |
| ygk_12345 | noonedeadpunk the similar hosts with the same OS version are working fine | 13:32 |
| ygk_12345 | noonedeadpunk i have indeed deleted the continaers and also the /var/lib/lxc folder as well on the host | 13:33 |
| noonedeadpunk | hm.... | 13:33 |
| noonedeadpunk | what if you try to re-run without limit at all? I'm not sure this is an issue, but still.... | 13:34 |
| ygk_12345 | it would fail as well | 13:34 |
| noonedeadpunk | huh... | 13:34 |
| ygk_12345 | when it encounters these contianers | 13:34 |
| *** spatel has joined #openstack-ansible | 13:35 | |
| ygk_12345 | noonedeadpunk let me delete the contianers and reboot the host | 13:35 |
| ygk_12345 | noonedeadpunk one final try before rerunning the playbook | 13:36 |
| kleini | br-vxlan has an IP address configured but dynamic-address-fact is not able to read it. I even debugged Ansible and it reads from the interface its MTU and MAC address correctly but not the IP address. What else can I do? | 13:43 |
| *** KeithMnemonic has joined #openstack-ansible | 13:44 | |
| ygk_12345 | noonedeadpunk still the same situation. its not adding the container's eth0 and assiging the IP | 13:45 |
| kleini | as far as I understand the Ansible code, it falls back from metal_query to non_metal_query and there to ansible_host. | 13:45 |
| noonedeadpunk | ygk_12345: do you have /var/lib/lxc/c3v*/eth0.[ini|hwaddr]? | 13:47 |
| ygk_12345 | noonedeadpunk yes its there as well | 13:47 |
| noonedeadpunk | huh. I think you should look into lxc logs maybe | 13:48 |
| noonedeadpunk | As it's kinda up to lxc to add interfaces into bridge on stratup | 13:48 |
| noonedeadpunk | ygk_12345: oh, and do you have lxc.include of eth0.ini in lxc config? | 13:48 |
| ygk_12345 | noonedeadpunk yes its there. | 13:49 |
| *** fghaas has left #openstack-ansible | 13:49 | |
| noonedeadpunk | so I'd say there's defenitely smth going on with lxc, as from what you say I feel like they have been configured correctly... | 13:50 |
| *** d34dh0r53 has joined #openstack-ansible | 13:53 | |
| *** d34dh0r53 has quit IRC | 13:54 | |
| *** d34dh0r53 has joined #openstack-ansible | 13:54 | |
| ygk_12345 | noonedeadpunk got the issue now. its the dnsmaq lxc service thats is faulty. i restarted it . now its working fine | 13:55 |
| jrosser | kleini: have you defined the tunnel network in cidr_networks in openstack_user_config.yml? | 13:55 |
| mgariepy | will this one merge at some point ?https://review.opendev.org/#/c/718786/ | 13:58 |
| kleini | yes it is defined according to openstack_user_config.yml.prod-ceph.example | 13:58 |
| mgariepy | or do we need to trigger a recheck ? | 13:58 |
| *** jbadiapa has joined #openstack-ansible | 13:59 | |
| jrosser | mgariepy: i am not sure what is happening there | 13:59 |
| noonedeadpunk | We just needed to push it a bit :p | 13:59 |
| mgariepy | add +w.. | 13:59 |
| mgariepy | might help a bit. | 13:59 |
| jrosser | kleini: have you got access to an all-in-one build as a reference? | 14:00 |
| kleini | http://paste.openstack.org/show/792476/ <- this is the top part of my openstack_user_config.yml | 14:01 |
| kleini | nope, but I can setup an AIO | 14:02 |
| kleini | but AIO does not use separate compute and network hosts and therefore I doubt, it establishes an vxlan tunnel | 14:02 |
| kleini | but let's see | 14:02 |
| jrosser | yes it does, becasue it can run a full tempest test with VM attached by vxlan | 14:04 |
| *** this10nly has quit IRC | 14:04 | |
| kleini | will setup AIO | 14:04 |
| jrosser | reason i ask is that i am currently debugging other stuff in an AIO but straight away i can see that the tunnel_address variable is present http://paste.openstack.org/show/792477/ | 14:10 |
| *** ygk_12345 has quit IRC | 14:45 | |
| *** mgariepy has quit IRC | 15:04 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Refactor tempest plugin install from source https://review.opendev.org/721645 | 15:05 |
| jrosser | noonedeadpunk: ^ this gives the right repeat version in the vent locally here | 15:07 |
| jrosser | *tempest version | 15:07 |
| *** viks____ has quit IRC | 15:07 | |
| spatel | how are you guys!! | 15:07 |
| spatel | Quick question, we are building new datacenter and planning to build on new openstack release (does train stable enough for production?) | 15:08 |
| noonedeadpunk | jrosser: so you didin't get contraints working? or they just were not? | 15:08 |
| spatel | its been a while its out.. so assuming its | 15:08 |
| jrosser | they were not | 15:08 |
| *** gshippey has quit IRC | 15:09 | |
| jrosser | python_venv_build carefully splices together the supplied constraints and any that are necessary for things from source | 15:09 |
| noonedeadpunk | but shouldn't we then also refactor "Install tempest plugins from packages" as well? | 15:09 |
| *** gshippey has joined #openstack-ansible | 15:09 | |
| jrosser | well - i guess for distro packages you just have to trust that the packaging is good? | 15:11 |
| *** mgariepy has joined #openstack-ansible | 15:17 | |
| *** yolanda has quit IRC | 15:21 | |
| noonedeadpunk | jrosser: no, since we provide the same contraints | 15:22 |
| noonedeadpunk | I guess there packages was meant as pypi packges | 15:23 |
| noonedeadpunk | not distro ones lol | 15:23 |
| noonedeadpunk | this one https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install_source.yml#L59-L73 | 15:24 |
| *** udesale_ has quit IRC | 15:41 | |
| *** gyee has joined #openstack-ansible | 15:52 | |
| *** yolanda has joined #openstack-ansible | 15:52 | |
| redrobot | \o | 16:01 |
| redrobot | Is the weekly meeting still happening today? | 16:01 |
| jrosser | noonedeadpunk: oh i think you are right with those other package installs....... distro vs pypi vs source | 16:02 |
| jrosser | redrobot: it should be i guess..... | 16:05 |
| * redrobot waits patiently for possible meeting ... | 16:09 | |
| *** _kukacz has quit IRC | 16:14 | |
| redrobot | mnaser, no meeting today? | 16:15 |
| noonedeadpunk | sorry | 16:20 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 16:20 |
| openstack | Meeting started Tue Apr 21 16:20:59 2020 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:21 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:21 |
| *** openstack changes topic to " (Meeting topic: openstack_ansible_meeting)" | 16:21 | |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:21 |
| noonedeadpunk | #topic office hours | 16:21 |
| *** openstack changes topic to "office hours (Meeting topic: openstack_ansible_meeting)" | 16:21 | |
| noonedeadpunk | jrosser: redrobot sorry for the delay - with this summer time got a bit lost in time | 16:21 |
| redrobot | noonedeadpunk, no worries | 16:22 |
| noonedeadpunk | so, what do we have except broken tempest thing (again)? | 16:22 |
| redrobot | I was hoping mnaser would be here so we can talk about | 16:23 |
| redrobot | #link https://review.opendev.org/#/c/721348/ | 16:23 |
| mnaser | hi redrobot , sorry its exactly at 12pm so i grabbed something quick to eat | 16:23 |
| * mnaser clicks | 16:23 | |
| mnaser | oh yes hi | 16:23 |
| * redrobot waves all around | 16:24 | |
| mnaser | redrobot: yeah maybe the ansible sig might be a good home for those, it's not that i "Dont want them" but just feel like its not fair for y'all considering we don't really touch/see those roles :) | 16:24 |
| redrobot | Hehe, no worries | 16:24 |
| redrobot | I _think_ we originally asked y'all because it was all Ansible | 16:24 |
| redrobot | but I would not be opposed to moving them all under the Barbican umbrella | 16:24 |
| noonedeadpunk | I think those days there was no ansible sig, but now it is. So indeed I agree with evrardjp and mnaser that it's probably better to place these roles under ansible sig ubrella (as it is all about ANsible :)) | 16:26 |
| noonedeadpunk | as we kinda know ansible but not realy aware of the functional part of these hsm roles | 16:27 |
| noonedeadpunk | so really hard to judge on them while voing on patches | 16:28 |
| noonedeadpunk | *voting | 16:28 |
| redrobot | Ansible SIG != OpenStack Ansible Team? | 16:28 |
| redrobot | I think ultimately the Barbican team will need to have core review permissions on these since we have access to the hardware to test them. | 16:28 |
| noonedeadpunk | nope. OpenStack-Ansible is project for deploying openstack with ansible :) | 16:29 |
| noonedeadpunk | so it's more like deployment tool project along with tripleo, kolla, etc | 16:30 |
| redrobot | gotcha | 16:30 |
| redrobot | https://wiki.openstack.org/wiki/Ansible_SIG then? | 16:30 |
| noonedeadpunk | yeah and #openstack-ansible-sig | 16:31 |
| redrobot | cool, I'll talk to them and see if that might be a good home for our *-hsm roles. | 16:31 |
| redrobot | thanks! :D | 16:31 |
| noonedeadpunk | sure, no problems:) | 16:31 |
| noonedeadpunk | ok, what I though about today, if should we offer providing a zookeeper cluster installation as a part of infra role? | 16:33 |
| noonedeadpunk | Like some of the projects make use it for coordination (not only telemetry ones) so it might be a nice idea having it. | 16:34 |
| *** evrardjp has quit IRC | 16:35 | |
| *** evrardjp has joined #openstack-ansible | 16:35 | |
| noonedeadpunk | Like I know that the roles is already exist, so as for implementation we will just need to add env.d file and sample to documentation | 16:35 |
| noonedeadpunk | As a second step we might add defaults to the roles so in case of existing zookepeer projects would use it for coordination by default | 16:36 |
| *** cyberpear has quit IRC | 16:38 | |
| *** rpittau is now known as rpittau|afk | 16:39 | |
| *** jrosser has quit IRC | 16:41 | |
| *** mnaser has quit IRC | 16:41 | |
| *** vblando has quit IRC | 16:42 | |
| *** Open10K8S has quit IRC | 16:42 | |
| *** cyberpear has joined #openstack-ansible | 16:42 | |
| *** soren has quit IRC | 16:42 | |
| *** Open10K8S has joined #openstack-ansible | 16:42 | |
| *** gshippey has quit IRC | 16:43 | |
| *** cyberpear has quit IRC | 16:43 | |
| *** jrosser has joined #openstack-ansible | 16:44 | |
| *** vblando has joined #openstack-ansible | 16:44 | |
| *** mnaser has joined #openstack-ansible | 16:44 | |
| *** gshippey has joined #openstack-ansible | 16:44 | |
| *** cyberpear has joined #openstack-ansible | 16:44 | |
| *** soren has joined #openstack-ansible | 16:45 | |
| *** mnaser has quit IRC | 16:46 | |
| *** cyberpear has quit IRC | 16:47 | |
| *** mnaser has joined #openstack-ansible | 16:47 | |
| *** cyberpear has joined #openstack-ansible | 16:47 | |
| *** mnaser has quit IRC | 16:49 | |
| *** cyberpear has quit IRC | 16:49 | |
| *** mnaser has joined #openstack-ansible | 16:50 | |
| *** cyberpear has joined #openstack-ansible | 16:50 | |
| *** jbadiapa has quit IRC | 17:02 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Refactor tempest plugin install from source https://review.opendev.org/721645 | 17:15 |
| jrosser | noonedeadpunk: i have massively simplified the tempest source install now | 17:15 |
| noonedeadpunk | #endmeeting | 17:15 |
| *** openstack changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: https://bit.ly/2SAcGAn" | 17:15 | |
| openstack | Meeting ended Tue Apr 21 17:15:51 2020 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:15 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-04-21-16.20.html | 17:15 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-04-21-16.20.txt | 17:15 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-04-21-16.20.log.html | 17:15 |
| noonedeadpunk | jrosser: so we all in all ended up that we need to define venv_wheel_build_enable even if it's default one? | 17:17 |
| *** thuydang has quit IRC | 17:17 | |
| jrosser | that is still set to 'no' otherwise becasue of the setting in the stackviz install | 17:17 |
| jrosser | i can double check that again | 17:17 |
| noonedeadpunk | jrosser: hm wouldn't tempest_pip_plugins be overriden by another set_fact? | 17:18 |
| noonedeadpunk | As we kinda supposed to have mix of repo and package | 17:18 |
| jrosser | it should append as it takes the existing value from the previous set_fact | 17:19 |
| jrosser | thats pretty much how the set_fact / with_items works anyway | 17:19 |
| noonedeadpunk | oh, yeah, sorry) | 17:19 |
| jrosser | hopefully this is ok as i'm very happy to remove all that previous stuff | 17:19 |
| jrosser | i've only deployed tempest in my AIO, not the rest of openstack | 17:20 |
| jrosser | in the repo server /var/www/repo/os-releases/20.1.0.dev192/tempest-20.1.0.dev192-source-constraints.txt i have http://paste.openstack.org/show/792494/ | 17:22 |
| jrosser | and the source installs look correct in the tempest venv also http://paste.openstack.org/show/792495/ | 17:23 |
| noonedeadpunk | I hope that we've dropped isolated flag is ok eventually | 17:24 |
| * jrosser runs away | 17:25 | |
| noonedeadpunk | but yeah, it's much simplified and I think it might also be some speed up because no need in cloning repos in advance | 17:25 |
| jrosser | it moves the cloning into the wheel build | 17:25 |
| *** yolanda has quit IRC | 17:33 | |
| noonedeadpunk | yeah, I mean maybe it's just a bit more fast than would be with ansible (due to connects) | 17:42 |
| noonedeadpunk | anyway we don't have them in required-projects | 17:42 |
| *** cshen_ has quit IRC | 17:46 | |
| *** cshen has joined #openstack-ansible | 17:48 | |
| *** cshen has quit IRC | 17:52 | |
| *** hamzy__ has quit IRC | 17:59 | |
| *** cshen has joined #openstack-ansible | 18:20 | |
| *** cshen has quit IRC | 18:25 | |
| *** gshippey has quit IRC | 19:07 | |
| jrosser | noonedeadpunk: we need to make a plan to merge all this tempest stuff | 19:11 |
| jrosser | bionic distro job is still broken and needs your patch to the integrated repo to fix it | 19:11 |
| jrosser | and thats also needed to land my patch to os_tempest to fix all the other jobs | 19:12 |
| jrosser | so perhaps for now we make bionic/distro non-voting for os_tempest? | 19:12 |
| *** cshen has joined #openstack-ansible | 20:02 | |
| *** idlemind has joined #openstack-ansible | 20:04 | |
| *** cshen has quit IRC | 20:06 | |
| *** mgariepy has quit IRC | 20:09 | |
| *** mgariepy has joined #openstack-ansible | 20:12 | |
| *** thuydang has joined #openstack-ansible | 20:53 | |
| *** spatel has quit IRC | 21:30 | |
| *** thuydang has quit IRC | 21:41 | |
| *** keeper has left #openstack-ansible | 22:35 | |
| *** spatel has joined #openstack-ansible | 22:53 | |
| *** tosky has quit IRC | 23:03 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Refactor tempest plugin install from source https://review.opendev.org/721645 | 23:26 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!