| *** yann-kaelig has quit IRC | 00:06 | |
| *** macz_ has quit IRC | 00:10 | |
| *** tosky has quit IRC | 00:15 | |
| *** macz_ has joined #openstack-ansible | 00:51 | |
| *** macz_ has quit IRC | 00:56 | |
| *** dwilde has quit IRC | 02:04 | |
| *** macz_ has joined #openstack-ansible | 02:04 | |
| *** dwilde has joined #openstack-ansible | 02:07 | |
| *** dwilde has quit IRC | 02:07 | |
| *** dwilde has joined #openstack-ansible | 02:08 | |
| *** macz_ has quit IRC | 02:09 | |
| *** dwilde has quit IRC | 02:44 | |
| *** d34dh0r53 has quit IRC | 03:32 | |
| *** d34dh0r53 has joined #openstack-ansible | 03:33 | |
| *** d34dh0r53 has quit IRC | 03:35 | |
| *** d34dh0r53 has joined #openstack-ansible | 03:35 | |
| *** d34dh0r53 has quit IRC | 04:19 | |
| *** d34dh0r53 has joined #openstack-ansible | 04:20 | |
| *** d34dh0r53 has quit IRC | 04:22 | |
| *** d34dh0r53 has joined #openstack-ansible | 04:22 | |
| *** ChiTo has joined #openstack-ansible | 04:54 | |
| *** d34dh0r53 has quit IRC | 04:59 | |
| *** d34dh0r53 has joined #openstack-ansible | 04:59 | |
| *** d34dh0r53 has quit IRC | 05:01 | |
| *** d34dh0r53 has joined #openstack-ansible | 05:01 | |
| *** d34dh0r53 has quit IRC | 05:03 | |
| *** d34dh0r53 has joined #openstack-ansible | 05:03 | |
| *** d34dh0r53 has quit IRC | 05:03 | |
| *** d34dh0r53 has joined #openstack-ansible | 05:04 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #openstack-ansible | 05:33 | |
| *** macz_ has joined #openstack-ansible | 06:06 | |
| *** macz_ has quit IRC | 06:11 | |
| *** gyee has quit IRC | 06:17 | |
| *** miloa has joined #openstack-ansible | 07:01 | |
| *** pto has joined #openstack-ansible | 07:05 | |
| *** pto has quit IRC | 07:05 | |
| *** pto has joined #openstack-ansible | 07:06 | |
| *** dave-mccowan has quit IRC | 07:30 | |
| *** rpittau|afk is now known as rpittau | 08:08 | |
| *** andrewbonney has joined #openstack-ansible | 08:17 | |
| *** tosky has joined #openstack-ansible | 08:22 | |
| *** openstackgerrit has joined #openstack-ansible | 08:32 | |
| openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible-os_zun master: Add support for kata container runtime https://review.opendev.org/c/openstack/openstack-ansible-os_zun/+/769555 | 08:32 |
|---|---|---|
| openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible master: Add service for zun-wsproxy console access https://review.opendev.org/c/openstack/openstack-ansible/+/769142 | 08:41 |
| openstackgerrit | Denys Mishchenko proposed openstack/openstack-ansible-os_neutron master: Added tag neutron-policy-overrides to just update neutron policy https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/770248 | 08:47 |
| *** MrClayPole has joined #openstack-ansible | 08:47 | |
| *** maharg101 has joined #openstack-ansible | 09:05 | |
| *** MrClayPole has left #openstack-ansible | 09:05 | |
| *** MrClayPole has joined #openstack-ansible | 09:08 | |
| openstackgerrit | Denys Mishchenko proposed openstack/openstack-ansible-os_cinder master: Added tags for policy override task https://review.opendev.org/c/openstack/openstack-ansible-os_cinder/+/770251 | 09:14 |
| *** MickyMan77 has joined #openstack-ansible | 09:15 | |
| noonedeadpunk | prometheanfire: I have the way to return them back (keypresses) but it's not automated yet | 09:36 |
| noonedeadpunk | https://gist.github.com/noonedeadpunk/77b92077c1bd7ae365835be61c45100c | 09:36 |
| noonedeadpunk | the issue with spice was that is has been very laggy on graphic environments, like windows | 09:37 |
| noonedeadpunk | maybe it's not the case anymore, no idea... | 09:38 |
| noonedeadpunk | andrewbonney: this one should be rebased now :( https://review.opendev.org/c/openstack/openstack-ansible-os_zun/+/769143 | 09:40 |
| andrewbonney | No problem, I've noticed an issue in it so I've got a fix for that which will go in too | 09:40 |
| noonedeadpunk | oh, ok | 09:41 |
| pto | I have deinfed rsyslog_client_user_defined_targets: ... ... ... and i dont see rsyslog is configured in urssuri. Have something been changed? | 09:54 |
| pto | * on ubuntu 20.04 | 09:55 |
| openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible-os_zun master: Add configuration for zun-wsproxy service https://review.opendev.org/c/openstack/openstack-ansible-os_zun/+/769143 | 10:29 |
| jrosser | noonedeadpunk: i have had a first look at the new pip resolver | 11:01 |
| jrosser | this is one approach we could take, which seems to work at least for os_keystone http://paste.openstack.org/show/801552/ | 11:01 |
| jrosser | for that role it seems to be sufficient to move the git requirement into an actual requirement rather than a constraint | 11:03 |
| *** rpittau has quit IRC | 11:06 | |
| *** rpittau has joined #openstack-ansible | 11:07 | |
| noonedeadpunk | hm, which kind of means that we don't need to adjust python_venv_build at all? | 11:13 |
| noonedeadpunk | I hoped tbh that we would be able to just adjust the single role to make all of them happy, but that diff makes sense to me | 11:14 |
| jrosser | yeah, for the time being yes | 11:14 |
| jrosser | there is probably legacy stuff in python_venv_build working around old pip bugs which we can clean up | 11:14 |
| jrosser | i guess we could do some hack in python_venv_build which moves anything with git+https out of the constraints and into the requirements | 11:16 |
| jrosser | but thats nasty | 11:16 |
| *** structz has joined #openstack-ansible | 11:18 | |
| openstackgerrit | Erik Berg proposed openstack/openstack-ansible-os_neutron master: Allow overriding firewall_driver for ovs https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/768522 | 11:25 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_keystone master: Move keystone pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/770271 | 11:34 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_cinder master: Move cinder pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_cinder/+/770272 | 11:36 |
| openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible-os_zun master: Reduce number of processes on small systems https://review.opendev.org/c/openstack/openstack-ansible-os_zun/+/770273 | 11:45 |
| openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible master: Restrict zun thread/process count for AIO deployments https://review.opendev.org/c/openstack/openstack-ansible/+/770274 | 11:52 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron master: Move neutron pip packages from constraints to requirements https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/770276 | 12:01 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_nova master: Move nova pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/770279 | 12:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_placement master: Move placement pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_placement/+/770280 | 12:05 |
| *** structz has quit IRC | 12:07 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Move tempest pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/770281 | 12:09 |
| *** yann-kaelig has joined #openstack-ansible | 12:09 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_horizon master: Move horizon pip packages from constraints to requirements https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/770283 | 12:18 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Update pip/setuptools/wheel to latest version https://review.opendev.org/c/openstack/openstack-ansible/+/770284 | 12:23 |
| *** fresta has quit IRC | 12:30 | |
| *** fresta has joined #openstack-ansible | 12:30 | |
| noonedeadpunk | jrosser: yep, that is nasty and I'd rather just patch every role instead | 12:32 |
| jrosser | noonedeadpunk: could you take a look at this patch? https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/768117 | 12:56 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_placement master: Move placement pip package from a constraint to a requirement https://review.opendev.org/c/openstack/openstack-ansible-os_placement/+/770280 | 12:57 |
| noonedeadpunk | yeah, I don't like it that way... | 13:32 |
| *** rfolco has joined #openstack-ansible | 13:41 | |
| *** yann-kaelig has quit IRC | 13:43 | |
| *** rh-jelabarre has quit IRC | 13:56 | |
| *** rh-jelabarre has joined #openstack-ansible | 14:03 | |
| jrosser | something is wrong with the debian-buster builds | 14:10 |
| *** spatel has joined #openstack-ansible | 14:16 | |
| spatel | morning folks | 14:16 |
| spatel | noonedeadpunk: look like we didn't release anything after RC1 - https://docs.openstack.org/releasenotes/openstack-ansible/victoria.html | 14:18 |
| jrosser | spatel: still trying https://review.opendev.org/c/openstack/openstack-ansible/+/770073 ....... | 14:26 |
| spatel | +1 thanks | 14:27 |
| spatel | assuming its going to be RC2 | 14:28 |
| jrosser | it will be the release, we are out of time | 14:46 |
| jrosser | if anyone has any clue what is broken with the debian jobs it would be really helpful https://zuul.openstack.org/builds?job_name=openstack-ansible-deploy-aio_metal-debian-buster | 14:49 |
| jrosser | i've got a better formatted version of the problem here http://paste.openstack.org/show/801557/ | 14:50 |
| *** pto has quit IRC | 14:57 | |
| andrewbonney | jrosser: do we pull in buster-backports? Looks like there was a recent build there: https://metadata.ftp-master.debian.org/changelogs//main/q/qemu/qemu_5.2+dfsg-3~bpo10+1_changelog | 14:59 |
| jrosser | oh interesting | 14:59 |
| jrosser | yes we do https://zuul.openstack.org/build/9973f0779cf6479aa70e6dbc573eb02e/log/logs/etc/host/apt/sources.list.d/backports.list.txt | 15:00 |
| *** yann-kaelig has joined #openstack-ansible | 15:00 | |
| jrosser | not sure where Depends: libfdt1 (>= 1.6.0) is coming from though | 15:01 |
| *** pcaruana has quit IRC | 15:03 | |
| *** yann-kaelig has quit IRC | 15:03 | |
| andrewbonney | Think it's in https://packages.debian.org/buster-backports/qemu-system-x86 | 15:04 |
| *** pcaruana has joined #openstack-ansible | 15:12 | |
| jrosser | ahha and here it is https://packages.debian.org/buster-backports/libfdt1 | 15:15 |
| jrosser | and it seems to be in the mirror https://mirror.bhs1.ovh.opendev.org/debian/pool/main/d/device-tree-compiler/ | 15:25 |
| andrewbonney | I'm guessing it may be a conflict with a package further down the tree that apt isn't showing | 15:26 |
| jrosser | libftd1 only depends on libc | 15:27 |
| mgariepy | what does depend on libftd1 ? | 15:28 |
| *** d34dh0r53 has quit IRC | 15:51 | |
| *** macz_ has joined #openstack-ansible | 15:51 | |
| prometheanfire | noonedeadpunk: thanks for the link :D | 15:51 |
| *** macz_ has joined #openstack-ansible | 15:52 | |
| prometheanfire | noonedeadpunk: that a ctrl+alt+delete patch? | 15:52 |
| noonedeadpunk | yep | 15:52 |
| prometheanfire | there was a button for me to click for that | 15:53 |
| *** d34dh0r53 has joined #openstack-ansible | 15:53 | |
| prometheanfire | my issue with it was that I had keypresses being eaten when the shift key was unpressed (meaning released) | 15:54 |
| noonedeadpunk | uh, nasty as well. but honstly it seems like project is more dead then alive | 15:55 |
| noonedeadpunk | oh, yes, it has been added with +- recent patch - awesome https://github.com/freedesktop/spice-html5/commit/56ddb3005e39a9e76670fec9676e667d44167bf3 | 15:55 |
| noonedeadpunk | I think it's a time to spawn aio? | 15:56 |
| noonedeadpunk | in case nobody has done already that... | 15:57 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 16:00 |
| openstack | Meeting started Tue Jan 12 16:00:18 2021 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| *** openstack changes topic to " (Meeting topic: openstack_ansible_meeting)" | 16:00 | |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:00 |
| noonedeadpunk | #topic bug triage | 16:00 |
| *** openstack changes topic to "bug triage (Meeting topic: openstack_ansible_meeting)" | 16:00 | |
| noonedeadpunk | hey! have been a while since last meeting and there are a lot of new stuff in launchpad... | 16:00 |
| noonedeadpunk | https://bugs.launchpad.net/openstack-ansible/+bug/1908510 | 16:01 |
| openstack | Launchpad bug 1908510 in openstack-ansible "Keystone federation part is not executed during play" [Undecided,New] | 16:01 |
| noonedeadpunk | I think jrosser you have laready covered that? | 16:01 |
| noonedeadpunk | oh, ok, fix is not working | 16:04 |
| noonedeadpunk | specificly this part https://review.opendev.org/c/openstack/openstack-ansible/+/760431/1/playbooks/os-keystone-install.yml | 16:05 |
| jrosser | o/ hello | 16:06 |
| noonedeadpunk | but was also fixed afterwards.... | 16:06 |
| noonedeadpunk | uh, this broken gerrit->launchpad integgration makes me mad | 16:07 |
| jrosser | it is fixed here i think https://review.opendev.org/c/openstack/openstack-ansible/+/767513 | 16:07 |
| noonedeadpunk | yeah, alredy found :( | 16:08 |
| noonedeadpunk | ok, next thing https://bugs.launchpad.net/openstack-ansible/+bug/1908544 | 16:08 |
| openstack | Launchpad bug 1908544 in openstack-ansible "ROLE lxc_hosts: variable lxc_hosts_container_image_url has no effect" [Undecided,New] | 16:08 |
| jrosser | oh this is always true https://opendev.org/openstack/openstack-ansible-lxc_hosts/src/branch/master/defaults/main.yml#L204 | 16:10 |
| jrosser | so this test does not work https://opendev.org/openstack/openstack-ansible-lxc_hosts/src/branch/master/tasks/lxc_cache_prestage.yml#L52 | 16:11 |
| noonedeadpunk | but it should be always false then? | 16:12 |
| noonedeadpunk | ah, because of the default, yeah | 16:12 |
| noonedeadpunk | that does not make sense actually... | 16:13 |
| noonedeadpunk | we don;t need to set_fact at all I guess | 16:13 |
| jrosser | for debian and redhat it's always set to something https://opendev.org/openstack/openstack-ansible-lxc_hosts/src/branch/master/vars/debian.yml#L16 | 16:14 |
| noonedeadpunk | that is totally broken ,confirming | 16:14 |
| jrosser | the set fact is required for ubuntu i think to resolve the container name? | 16:15 |
| jrosser | the later when: clauses are supposed to only make it run for ubuntu? | 16:15 |
| noonedeadpunk | you mean that checksum is not defined? | 16:18 |
| *** d34dh0r53 has quit IRC | 16:19 | |
| *** d34dh0r53 has joined #openstack-ansible | 16:19 | |
| noonedeadpunk | I think yI could broke that when was fixing ubuntu image somehow | 16:19 |
| jrosser | certainly the vars are set up differently between ubuntu and the other distros | 16:20 |
| noonedeadpunk | anyway, it should be length test at least, not bool | 16:20 |
| jrosser | and we must change the default for lxc_hosts_container_image_url to be just empty | 16:21 |
| noonedeadpunk | yeah, agree | 16:21 |
| noonedeadpunk | or not - I think that will break everythin except ubuntu? | 16:21 |
| jrosser | the whole thing needs overhauling i guess | 16:22 |
| jrosser | again :( | 16:22 |
| jrosser | it might be better to define a bool in vars/<distro>.yml that says if the magic code to resolve the latest container is needed or not | 16:23 |
| jrosser | the its just simpler and 'true' for ubuntu and false for everything else | 16:23 |
| noonedeadpunk | `_lxc_hosts_container_image_url` for ubuntu should not be defined, but I think empty string will give false | 16:23 |
| noonedeadpunk | well, yes, we might go this way as well | 16:24 |
| jrosser | if _lxc_hosts_container_image_url | length = 0 and _lxc_hosts_container_image_resolve .... | 16:24 |
| jrosser | i.e only do it if there is no provided URL and the OS requires that we resolve | 16:24 |
| jrosser | +/- an underscore there | 16:25 |
| noonedeadpunk | yeah, ok, agreed | 16:25 |
| noonedeadpunk | https://bugs.launchpad.net/openstack-ansible/+bug/1908703 | 16:26 |
| openstack | Launchpad bug 1908703 in openstack-ansible "federation domain not configured correct" [Undecided,New] | 16:26 |
| noonedeadpunk | I really was trying not to touch that part of the code as never used it with my own | 16:27 |
| ThiagoCMC | Hey guys, which hosts are supposed to be under 'masakari-monitor_hosts:' ? The controllers or the compute nodes? The Pacemaker and Corosync were installed at the baremetal! Outside of the Masakari container. Is this expected? | 16:27 |
| noonedeadpunk | these are compute hosts | 16:27 |
| ThiagoCMC | Damn... ok | 16:27 |
| noonedeadpunk | I think this has been fixed with https://opendev.org/openstack/openstack-ansible-os_keystone/commit/4c9642765e632d409c6d9028aae26f4514af7033 | 16:29 |
| noonedeadpunk | uh, no, it's different probably... | 16:32 |
| noonedeadpunk | but I can hardly understand what goes wrong here.... | 16:32 |
| ThiagoCMC | I'm wondering about the releasenotes for 22.0.0.0b1, where it says: "Support is added for deploying OpenStack on Ubuntu Focal (20.04) with source based installs. Ubuntu Cloud Archive is not available for Focal at this point so it is not possible to offer distro package based installs for Focal." - But Cloud Archive is available to focal! | 16:32 |
| noonedeadpunk | yes but do not contain Ussuri packeges - I guess this reno is from U. | 16:34 |
| ThiagoCMC | Hmmm... Got it | 16:34 |
| *** gshippey has joined #openstack-ansible | 16:35 | |
| *** ChiTo has joined #openstack-ansible | 16:35 | |
| jrosser | gshippey: any ideas on https://bugs.launchpad.net/openstack-ansible/+bug/1908703 | 16:35 |
| openstack | Launchpad bug 1908703 in openstack-ansible "federation domain not configured correct" [Undecided,New] | 16:36 |
| ChiTo | Hi openstack-ansible team, happy new year! | 16:36 |
| gshippey | thank you - was literally about to go look on eavesdrop! | 16:36 |
| noonedeadpunk | nice to see you around gshippey :) | 16:36 |
| jrosser | this task https://opendev.org/openstack/openstack-ansible-os_keystone/src/branch/master/tasks/keystone_federation_sp_idp_setup.yml#L124 | 16:38 |
| jrosser | takes trusted_id_list.domain_id | 16:39 |
| jrosser | i think that federated_identities is dealt with completely somewhere else, so i'm really not sure what the bug is about | 16:40 |
| ChiTo | I am trying to install Octavia for my cloud but I am hitting this issue: FAILED! => { "msg": "'dict object' has no attribute 'address'", am I missing some specific value? it is hit on the post-installation task | 16:40 |
| jrosser | as the data and task highlighted do not seem to be related | 16:40 |
| ChiTo | I have tried with vlan as well as flat types and both raise the error, my deployment does not use lxc, and it is installed on bare metal (i mean the neutron components and octavia ones) | 16:41 |
| jrosser | ChiTo: can we look at this after the meeting? | 16:41 |
| noonedeadpunk | pto is not around so... | 16:41 |
| ChiTo | jrosser: Sure thing, thanks for your reply | 16:42 |
| jrosser | i think we should query if the task is related to the data | 16:42 |
| jrosser | becasue domain can be specified in several places | 16:42 |
| noonedeadpunk | can I ask that somebody from team picked this up? As I have no experience in that area at all :( | 16:43 |
| jrosser | sure i'll see if we can try to find out whats going on there | 16:44 |
| noonedeadpunk | Thanks | 16:46 |
| noonedeadpunk | next https://bugs.launchpad.net/openstack-ansible/+bug/1872169 | 16:46 |
| openstack | Launchpad bug 1872169 in openstack-ansible "train failed deploy setup_openstack.yaml " [Undecided,New] | 16:46 |
| ThiagoCMC | I did something bad yesterday, where I configured the 'masakari-monitor_hosts:' with the controllers host/ip, now, even after cleaning it up the 'openstack_deploy/conf.d/masakari.yml' with just the compute nodes, the playbooks are still trying to install pacemaker/corosync at the controllers! I'm guessing that the inventory is now polluted, how to clean it up? | 16:46 |
| noonedeadpunk | I guess it's because `"/openstack/venvs/rally-20.1.1.dev1/bin/pip", "install", "-U", "--constraint", "/openstack/venvs/rally-20.1.1.dev1/global-constraints.txt", "--constraint", "/openstack/venvs/rally-20.1.1.dev1/constraints.txt", "--log", "/var/log/python_venv_build.log", "--find-links", "http://localhost/os-releases/20.1.1.dev1/ubuntu-18.04-x86_64", "--trusted-host", "localhost", "pip", "setuptools", "wheel"` | 16:46 |
| *** sshnaidm|ruck is now known as sshnaidm|afk | 16:47 | |
| jrosser | connection refused to localhost:80 | 16:48 |
| noonedeadpunk | yeah, it should not be localhost | 16:48 |
| noonedeadpunk | and we define it here https://opendev.org/openstack/openstack-ansible/src/branch/master/playbooks/defaults/source_install.yml#L27 | 16:48 |
| jrosser | i wonder why localhost is an inventory target, thats not right | 16:49 |
| noonedeadpunk | it's not a target, but it's a utility host? | 16:50 |
| noonedeadpunk | localhost is there because it's default for https://opendev.org/openstack/ansible-role-python_venv_build/src/branch/master/defaults/main.yml#L79 | 16:51 |
| ThiagoCMC | The Senlin endpoint on Victoria is being installed under "RegionOne", it's ignoring the configured 'service_region' name. Is there a quick fix available? I tried to declare 'senlin_service_region: "{{ service_region }}"' within 'inventory/group_vars/senlin.yml' but no effect. | 16:54 |
| noonedeadpunk | I really don't see how this might happen even for 20.0.0 | 16:55 |
| *** openstackgerrit has quit IRC | 16:55 | |
| noonedeadpunk | um, you should be able to set senlin_service_region jsut in user_variables | 16:56 |
| noonedeadpunk | or openstack_deploy/group_vars/senlin_all.yml | 16:57 |
| jrosser | if someone makes internal_lb_vip_address: localhost then that would go all kinds of wrong | 16:57 |
| jrosser | perhaps for that bug we need to ask to either 1) join IRC and chat or 2) attach o_u_c.yml | 16:58 |
| noonedeadpunk | I can't imagin somebody doing that... but.... | 16:58 |
| noonedeadpunk | +1 | 16:58 |
| jrosser | we should talk about pip resolver and also whatever is broken with debian right now | 16:59 |
| noonedeadpunk | #topic office hours | 16:59 |
| *** openstack changes topic to "office hours (Meeting topic: openstack_ansible_meeting)" | 16:59 | |
| noonedeadpunk | yeah, my fault... | 17:00 |
| noonedeadpunk | regarding debian - I read through, and I started deploying aio to look closer | 17:00 |
| jrosser | ok thats cool i'd not got that far | 17:01 |
| jrosser | backports repo is there by default in the CI images i think | 17:01 |
| noonedeadpunk | we need it for the newer libvirt iirc | 17:02 |
| jrosser | right, and thats where it all starts to go wrong | 17:02 |
| noonedeadpunk | as otherwise nova fails | 17:02 |
| noonedeadpunk | well, yes :( | 17:02 |
| jrosser | i expect taking the paste i made and trying to apt-install the stuff off there might get you to the same error quickly | 17:02 |
| jrosser | rather than run the whole AIO | 17:02 |
| jrosser | i looked in the infra mirror and the libfdt1 seems to be there with a new enough version | 17:03 |
| noonedeadpunk | uh https://opendev.org/openstack/openstack-ansible-openstack_hosts/src/branch/master/vars/debian.yml#L89 | 17:08 |
| noonedeadpunk | eventually need to end meeting I guess and continue to check out debian | 17:09 |
| noonedeadpunk | #endmeeting | 17:09 |
| *** openstack changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: http://bit.ly/osa-review-board-v3" | 17:09 | |
| openstack | Meeting ended Tue Jan 12 17:09:25 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:09 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2021/openstack_ansible_meeting.2021-01-12-16.00.html | 17:09 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2021/openstack_ansible_meeting.2021-01-12-16.00.txt | 17:09 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2021/openstack_ansible_meeting.2021-01-12-16.00.log.html | 17:09 |
| jrosser | i made a topic for the pip work https://review.opendev.org/q/topic:%22osa-new-pip%22+(status:open%20OR%20status:merged) | 17:10 |
| *** d34dh0r53 has quit IRC | 17:10 | |
| noonedeadpunk | that is really not correct https://zuul.opendev.org/t/openstack/build/048d23bc26f84ab68c715b77e87f2aea/log/logs/etc/host/apt/sources.list.d/osbpo.list.txt | 17:10 |
| jrosser | completely not right | 17:11 |
| ThiagoCMC | Does Victoria supports Senlin's Horizon dashboard ui ? | 17:17 |
| gshippey | @noonedeadpunk o/ hello! Sorry i've been quiet, need to get back into the swing of things after the holidays | 17:18 |
| *** miloa has quit IRC | 17:19 | |
| noonedeadpunk | no worries at all! | 17:19 |
| ThiagoCMC | Is it ok to mix nova_compute (qemu) with zun_compute on the same host? | 17:20 |
| noonedeadpunk | I think no dashboard support yet | 17:21 |
| ThiagoCMC | Ok | 17:21 |
| ThiagoCMC | I tried to enable Zun for the first time, with 'zun-infra_hosts:' but playbook fails: 'The group `zun_compute` is undefined.' - Wondering if I can put it at the same qemu host...? | 17:21 |
| noonedeadpunk | to share zun with compute you might need https://review.opendev.org/c/openstack/openstack-ansible-os_zun/+/769116 I gues | 17:22 |
| noonedeadpunk | jrosser: on my VM I don't see any package conflict.... | 17:23 |
| jrosser | thats with an AIO run or with just apt-get.... | 17:23 |
| noonedeadpunk | with just apt-get for now | 17:23 |
| noonedeadpunk | oh, hm | 17:24 |
| noonedeadpunk | I guess I missed smth | 17:24 |
| noonedeadpunk | ah, yes... | 17:25 |
| ThiagoCMC | Interesting! So, is 'openstack-ansible-os_zun/+/769116' really required to mix Zun with Nova? | 17:26 |
| *** d34dh0r53 has joined #openstack-ansible | 17:27 | |
| *** maharg101 has quit IRC | 17:27 | |
| noonedeadpunk | yeah, plain debian is really broken with updates repo... | 17:31 |
| *** rpittau is now known as rpittau|afk | 17:32 | |
| noonedeadpunk | but I see the problem | 17:33 |
| jrosser | ThiagoCMC: if you don't set the var from that patch then zun and nova will seperatly account for the CPU usage on the compute host in placement | 17:35 |
| jrosser | so you do double accounting | 17:35 |
| ThiagoCMC | Awesome, thank you for clarifying it for me! | 17:37 |
| noonedeadpunk | jrosser: https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/770457 that should help | 17:38 |
| jrosser | oh nice, yes so we allowed qemu* but not its depends-on | 17:39 |
| jrosser | gerritbot seems to have vanished | 17:39 |
| noonedeadpunk | it does | 17:39 |
| noonedeadpunk | well, there were no requirement for these packages to be installed before... | 17:40 |
| noonedeadpunk | I'm not really sure if I'm missing smth else | 17:40 |
| noonedeadpunk | like libpmem1 or libslirp0 | 17:41 |
| noonedeadpunk | Uh, I guess I'd better create a list... | 17:41 |
| noonedeadpunk | eventually these are new packages that are installed during upgrade to qemu provided with backports repo | 17:43 |
| *** yann-kaelig has joined #openstack-ansible | 17:44 | |
| ChiTo | jrosser: I was looking for some value for 'address' and I just noticed that the octavia_hm_hosts variable tries to get the IP address for setting the controllers and pass it to the jinja template but my concern is that I am not using lxc, but bare metal only and still getting the: "msg": "'dict object' has no attribute 'container_interface'" | 17:51 |
| ChiTo | I am sorry it is not the container_interface but the address one | 17:51 |
| jrosser | ChiTo: here https://opendev.org/openstack/openstack-ansible-os_octavia/src/branch/master/defaults/main.yml#L342 | 17:54 |
| ChiTo | jrosser: exactly there is where I nnoticed the reference to address key | 17:54 |
| ChiTo | my question is, should I set then my IP address directly to my br-lbaas on the ctrl/neutron nodes? | 17:55 |
| jrosser | i think it is not really a question of containers or not really | 17:55 |
| jrosser | container_networks is a variable that talks about networks, it's in the inventory | 17:55 |
| ChiTo | this is my network part on the opesntack_user_config.yaml: http://paste.openstack.org/show/801562/ | 17:56 |
| ChiTo | and my cidr_networks refer to a lbaas: variable with a 172.28.99.0/24 net | 17:56 |
| jrosser | run the dynamic inventory and see what is created ./inventory/dynamic_inventory.py | less | 17:56 |
| jrosser | search through for container_networks | 17:56 |
| ChiTo | sure, let me do it now | 17:57 |
| jrosser | i have a SECNARIO=aio_metal AIO here and container_networks is defined and contains what i'd expect, no entries for lbaas though because it's not got octavia deployed | 17:58 |
| ChiTo | got i noticed that there is no address set on my br-lbaas: http://paste.openstack.org/show/801563 | 17:59 |
| ChiTo | I have created the br-lbaas through ovs on my centos because I am using ovs for everything not linux bridges, so should I set the address first then? | 18:00 |
| jrosser | well i don't know, you kind of cant have it both ways | 18:02 |
| jrosser | if you have assigned the address yourself then it won't be in the inventory data | 18:03 |
| ChiTo | but my question then is how the inventory getting the address? | 18:03 |
| ChiTo | if I don't set it previously | 18:04 |
| ChiTo | jrosser: instead of setting a br-mgmt then I should set on those servers the IP that will be used for the br-lbaas? | 18:05 |
| jrosser | for a containerised deployment the inventory would assign addresses to eth14 in the lxc containers and those would be present for the jinja to extract | 18:06 |
| ChiTo | and what happen in my case where I have no lxc containers? | 18:07 |
| ChiTo | but everythinig on metal | 18:07 |
| jrosser | ok so i check what is happening on an lxc deployment | 18:09 |
| jrosser | this is what i see in the inventory for container_networks for the octavia container http://paste.openstack.org/show/801564/ | 18:10 |
| jrosser | and address is populated, becasue the inventory has assigned that IP automatically | 18:10 |
| jrosser | this means that for a metal deployment the jinja in the octavia role defaults/main.yml is not appropriate | 18:11 |
| *** d34dh0r53 has quit IRC | 18:13 | |
| ChiTo | jrosser: understood, then my lbaas network should be routed from my br-mgmt right? I have a vlan reserved physically separated for managing the lbaas | 18:14 |
| jrosser | you can do it however you like | 18:14 |
| ChiTo | jrosser: oh sorry I noticed dthe static route is created internally on the lxc and add it the veth | 18:15 |
| jrosser | i would create my own jinja override in user_variables.yml for octavia_hm_hosts which generates the data you want | 18:15 |
| jrosser | if you want to do it routed then that is fine - i think jamesdenton may have an example for that somewhere | 18:15 |
| jrosser | if you want a dedicated interface then thats also fine, you just have to create an expression to gather the right addresses | 18:16 |
| ChiTo | jrosser: got it, the octavia_hm_hosts is a dict? if I have 3 nodes for octavia then it would be: ['172.28.99.10','172.28.99.11',172.28.99.12' | 18:17 |
| ChiTo | sorry I forgot to close the bracket | 18:17 |
| ChiTo | or it is a comma separated list? | 18:17 |
| jrosser | i think a string with commas | 18:18 |
| jrosser | codesearch is a big help here to find how it's used https://codesearch.opendev.org/?q=octavia_hm_hosts | 18:19 |
| ChiTo | jrosser: got it, let me try to override it to set it from my linux conf, thanks for the reference | 18:19 |
| ChiTo | jrosser: I just realized there is a group octavia-infra_hosts but then the API is installed along with the hm, workers, etc. right? | 18:20 |
| ChiTo | can we separate the API from the other components? | 18:20 |
| ChiTo | let's say an octavia_hosts and octavia-infra_hosts groups? | 18:21 |
| jrosser | i'm not sure why you'd want to do that | 18:21 |
| ChiTo | I think it is the same situation as the neutron-infra | 18:21 |
| ChiTo | because I control the APIs with a separated "managing operations team" from their own load balancers | 18:21 |
| jrosser | not really - the actual loadblanacing is done in nova VM that get created on compute nodes | 18:21 |
| ChiTo | I mean the haproxy's separated from the dataplane functionalities | 18:21 |
| jrosser | the haproxy is never on the controller | 18:22 |
| ChiTo | agreed, but that is why my operations team are focusing only on the API level with those nodes and the networking team on the rest of the datapath flow | 18:22 |
| jrosser | you could try splitting it out into different hosts as ansible groups are defined for them all | 18:27 |
| jrosser | would need a bunch of messing in env.d maybe | 18:27 |
| *** yann-kaelig has quit IRC | 18:29 | |
| ChiTo | jrosser: that is true I did a mess actually trying to separate the neutron api from the other components/agents and it took me a lot of time to get it back | 18:29 |
| jrosser | imho splitting the API away from the other parts is not so helpful | 18:30 |
| jrosser | when somehting doesnt work the answer is likley in the log of the thing thats now elsewhere | 18:30 |
| jrosser | it's not really datapath | 18:30 |
| ChiTo | jrosser: I agree, at the end the oeprational practicecs are doing the same methodologies to fix a possible issue | 18:30 |
| *** d34dh0r53 has joined #openstack-ansible | 18:31 | |
| jrosser | anyway - it looks like the os_octavia role may not natively support metal deployments unless i'm missing something | 18:31 |
| jrosser | an override seems to be needed for the lbaas network addresses | 18:32 |
| ChiTo | jrosser: I am running it right now and now it was set: controller_ip_port_list = 172.28.99.21:5555,172.28.99.22:5555,172.28.99.23:5555,172.28.99.24:5555,172.28.99.25:5555,172.28.99.26:5555 | 18:32 |
| ChiTo | on every octavia.conf file | 18:32 |
| ChiTo | is there a clean way to remove the venvs within /openstack for a specific service? | 18:35 |
| ChiTo | since I am on metal I have found is more difficult to tear it down or remove the packages/venv, unless is done manually whereis a lxc it is much more easier by only destroying the container | 18:36 |
| jrosser | thinking about the octavia_hm_hosts variable you should probably run up an AIO with SCENARIO=aio_metal_octavia and see how it works there | 18:38 |
| jrosser | becasue we test octavia/metal deployment in CI but just for one node | 18:38 |
| jrosser | to recreate the venv you can use -e venv_rebuild=yes | 18:39 |
| ChiTo | thanks for the tip | 18:39 |
| jrosser | but yes one of the downsides of the metal deploy is stuff all spread on the host that you can't just delete | 18:39 |
| ChiTo | so the SCENARIO I just pass it as a shell variable right? | 18:40 |
| jrosser | yes before doing ./scripts/boostrap-aio.sh | 18:40 |
| jrosser | it decides which components get deploys | 18:40 |
| jrosser | also you said you are on centos | 18:40 |
| jrosser | kind of don't know what to say there tbh | 18:40 |
| ChiTo | jrosser: exactly that is my concern, on metal the things gets worst when it comes to all the stuff spread as you mentioned | 18:42 |
| ChiTo | jrosser: I think I need to get back to the lxc approach, which now I find it 100% architected better, ,sorry for my confusion | 18:43 |
| *** ianychoi_ has quit IRC | 18:46 | |
| ChiTo | I am goinng to try the bootstrap with aio_metal_octavia on my devel env and to keep you updated to test it as well | 18:47 |
| ChiTo | with multiple nodes | 18:47 |
| jrosser | ChiTo: though seriously though for Centos, if you are doing a new deployment right now, as things currently stand OSA has no answer for Centos Stream | 18:47 |
| ChiTo | jrosser: thanks for the advise, I am mcurrently on train, and my cconcern now is with future openstack versions and centos 8.x/stream, I need to make a decision by EOY | 18:48 |
| jrosser | we are completely reliant on things from EPEL, and to date theres no equivalents for Stream | 18:48 |
| ChiTo | jrosser: Since Centos 7.X is getting its EOL until 2024, are you planning keep maintenance to the existing code at least to centos 7.x? | 18:49 |
| ChiTo | or would you drop it eentually? | 18:49 |
| jrosser | ChiTo: Centos-7 is already dropped in stable/victoria for OSA | 18:51 |
| jrosser | by some miracle we made it work for stable/ussuri but the rest of openstack i think dropped it before then | 18:52 |
| ChiTo | jrosser: understood thanks for the news, then I thinn the good path to go is Ubuntu and Victoria right? | 18:52 |
| jrosser | well all i can say is we don't have a good answer for what happens after Centos-8 EOL | 18:53 |
| jrosser | ideally someone will step forward as a maintainer for the Centos stuff | 18:53 |
| ChiTo | maybe rocky linux? | 18:53 |
| jrosser | it's the same story, someone would have to commit to support that | 18:54 |
| jrosser | like actually dive in and fix the CI every time it breaks | 18:54 |
| jrosser | we've had support for other OS, Suse and so in the past and they have become too much tech debt to carry without committed maintainers | 18:55 |
| ThiagoCMC | Hey guys, I just made something really simple but super cool! A Heat Template that deploys OSA AIO with 1 command! Check it out! https://github.com/tmartinx/svauto/blob/main/misc/os-heat-templates/osa-aio1.yaml | 18:55 |
| ChiTo | jrosser: got it, and I assume Victoria was also dropped for CentOS 8.x on OSA right? | 18:56 |
| ThiagoCMC | You launch this template and after about 1 hour, Horizon will be available at the Stack's Floating IP! | 18:56 |
| jrosser | ChiTo: Victoria will release with Centos-8 support but the EOL for Centos-8 will come around much quicker than the EOL for Victoria, so thats a big problem | 18:57 |
| ChiTo | understood, in my opinion then I would keep now with CentOS 7.x until understanding what is going to happen with CentOS stream and a possible maintenance path in the future | 19:00 |
| ChiTo | it is very likely I have to move my cloud to Ubuntu | 19:00 |
| ChiTo | s/migrate it | 19:00 |
| *** gyee has joined #openstack-ansible | 19:02 | |
| *** andrewbonney has quit IRC | 19:07 | |
| avagi | hi, | 19:21 |
| avagi | I am fighting with the switft proxy-server starting issue | 19:21 |
| avagi | STDERR: The option "xxxxxx" is not known to keystonemiddleware | 19:21 |
| avagi | later, when the server is receiving requests: | 19:21 |
| avagi | STDERR: ERROR:oslo.messaging._drivers.impl_rabbit:Connection failed: Server unexpectedly closed connection | 19:21 |
| avagi | environment: OSA 21.2.0, OS Rocky | 19:21 |
| avagi | do you have any suggestions, how to fix this? | 19:21 |
| avagi | thanks for your help in advance! | 19:21 |
| ThiagoCMC | noonedeadpunk, the 'TASK [python_venv_build : Install python packages into the venv]' - Error: raise Exception("pkg-config binary is required to compile libvirt-python") | 19:22 |
| ThiagoCMC | Sorry, masakari monitros at compute | 19:22 |
| *** maharg101 has joined #openstack-ansible | 19:23 | |
| *** spatel has quit IRC | 19:24 | |
| ThiagoCMC | I'm not sure if I apt install pkg-config or pkgconf ... Any suggestion? | 19:25 |
| ThiagoCMC | I'll go pkg-config lol | 19:27 |
| *** maharg101 has quit IRC | 19:28 | |
| ThiagoCMC | The package 'libvirt-dev' is also necessary for masakari-monitors at the compute nodes... | 19:34 |
| jrosser | ThiagoCMC: that sounds like it's trying to compile something, wouldnt that ususally be on the repo server if it's python wheels? | 19:35 |
| ThiagoCMC | I'm wondering about that too! | 19:35 |
| ThiagoCMC | But this is what it's happening with 'stable/victoria' | 19:35 |
| ThiagoCMC | The 'os-masakari-install.yml' is building things at the compute nodes. | 19:36 |
| *** spatel has joined #openstack-ansible | 19:36 | |
| *** spatel has quit IRC | 19:36 | |
| ThiagoCMC | I can see it at '/var/log/python_venv_build.log' | 19:36 |
| *** spatel has joined #openstack-ansible | 19:36 | |
| ThiagoCMC | The compute node log shows: "Installing collected packages: libvirt-python, masakari-monitors", "Successfully installed libvirt-python-6.10.0 masakari-monitors-10.0.0" | 19:38 |
| jrosser | it should be straightforward to figure out whats happening | 19:39 |
| jrosser | just compare what happens with the venv build to some other role like glance | 19:40 |
| ThiagoCMC | Ok... It is installed now at the compute nodes! Whew | 19:41 |
| ThiagoCMC | But my controllers are now polluted because I make a mistake and tried to install the masakari-monitors at the Controllers, instead of the Computes... I'm so stupid hehe - Now I don't know how to clean 'openstack_deploy/openstack_inventory.json' :-( | 19:42 |
| ThiagoCMC | Done. \o/ | 19:52 |
| ThiagoCMC | The Masakari playbooks are all good but, I'm seeing an error at the monitors: | 19:55 |
| ThiagoCMC | masakari-hostmonitor: WARNING masakarimonitors.hostmonitor.host_handler.handle_host [-] Corosync communication using 'br-mgmt' is failed.: oslo_concurrency.processutils.ProcessExecutionError: Unexpected error while running command. | 19:56 |
| ThiagoCMC | ERROR masakarimonitors.hostmonitor.host_handler.handle_host [-] Corosync communication is failed. | 19:56 |
| ThiagoCMC | Pacemaker cluster doesn't have quorum. | 19:56 |
| ThiagoCMC | :-( | 19:56 |
| ThiagoCMC | =P | 19:56 |
| ThiagoCMC | The `crm_verify -L -V` shows: error: Resource start-up disabled since no STONITH resources have been defined - Either configure some or disable STONITH with the stonith-enabled option - Might be it... | 19:59 |
| ThiagoCMC | The corosync daemon looks ok | 20:00 |
| *** yann-kaelig has joined #openstack-ansible | 20:02 | |
| *** arxcruz has quit IRC | 20:14 | |
| *** openstackgerrit has joined #openstack-ansible | 20:47 | |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_nova master: Add extra packages from Debian updates https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/770457 | 20:47 |
| spatel | folks, any idea what is this - http://paste.openstack.org/show/801568/ | 21:11 |
| spatel | I am trying to do "openstack-ansible os-horizon-install.yml --tags horizon-config" | 21:20 |
| spatel | look like tags is causing that issue. | 21:20 |
| *** gshippey has quit IRC | 21:24 | |
| *** maharg101 has joined #openstack-ansible | 21:24 | |
| *** rfolco has quit IRC | 21:25 | |
| *** maharg101 has quit IRC | 21:29 | |
| *** yann-kaelig has quit IRC | 21:37 | |
| *** rh-jelabarre has quit IRC | 22:04 | |
| *** spatel has quit IRC | 22:35 | |
| ThiagoCMC | noonedeadpunk, I think that I found a problem with the generated corosync.conf (masakari-monitors), look: http://paste.openstack.org/show/801571/ - The same IP is repeated at ring0_addr! But I believe that it should point to the other 2 (.21 and .22) as well... Right? | 22:59 |
| ThiagoCMC | Aha! Now I see 'corosync: [QUORUM] Members[2]: 1 3' | 23:02 |
| ThiagoCMC | 'masakari-hostmonitor ... Corosync communication using 'br-mgmt' is normal.' Wheee! Getting there... :-D | 23:04 |
| ThiagoCMC | But, still... 'Pacemaker cluster doesn't have quorum.' :-/ | 23:05 |
| ThiagoCMC | noonedeadpunk, BTW, thank you for this: https://github.com/noonedeadpunk/ansible-pacemaker-corosync !! lol | 23:06 |
| *** arxcruz has joined #openstack-ansible | 23:09 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_nova master: Add extra packages from Debian updates https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/770457 | 23:11 |
| *** spatel has joined #openstack-ansible | 23:46 | |
| ThiagoCMC | Aha! I had to `apt install pcs -y ; pcs property set stonith-enabled=false` - Pacemaker has quorum! I also restarted the other pcs nodes as well. | 23:53 |
| *** spatel has quit IRC | 23:54 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!