| *** tosky has quit IRC | 00:01 | |
| *** cshen has joined #openstack-ansible | 00:03 | |
| *** watersj has quit IRC | 00:06 | |
| *** cshen has quit IRC | 00:08 | |
| *** watersj has joined #openstack-ansible | 00:10 | |
| *** threestrands has joined #openstack-ansible | 00:19 | |
| *** Adri2000 has quit IRC | 00:37 | |
| *** Adri2000 has joined #openstack-ansible | 00:44 | |
| poopcat | @watersj, how large is the image / how long does it take to upload? Horizon timeout is 30 minutes by default so if it takes longer than that, that could be your problem | 01:17 |
|---|---|---|
| watersj | poopcat, its cirros raw, 112mb | 01:18 |
| poopcat | ah yea, that shouldn't be a problem | 01:18 |
| *** gyee has quit IRC | 01:19 | |
| poopcat | the other thought is how large the horizon/glance containers are and if, when you're uploading, they're 100% full at any point | 01:19 |
| poopcat | when you upload via horizon, it saves the image to Horizon container and then uploads to Glance | 01:20 |
| watersj | i looked there too, little small but I have 20g free | 01:20 |
| watersj | also stopped other containers in my multi-node leaving 1 running to keep it simple | 01:21 |
| poopcat | yea, that should suffice.. those are your usual tell-tale signs... I don't have a train environment to replicate so I don't know if this is a bug or a problem with your implementation | 01:21 |
| watersj | poopcat, only other thing is I'm testing and have multinode setup as VMs on standalone KVM box, with nested vms enabled. Perforamce could be better but need to learn w/what I have | 01:27 |
| poopcat | oof, understood. Yes, performance will be pretty poor in the Openstack VMs (that live on VMs :-P ) | 01:29 |
| watersj | are you using stien? | 01:32 |
| poopcat | a bunch... mostly liberty-stein | 01:37 |
| watersj | i'll revert back to stein, and see | 01:39 |
| poopcat | I'll be honest though -- I only test Horizon if someone says something is wrong with it. Otherwise, I have no reason to check it. | 01:40 |
| *** cshen has joined #openstack-ansible | 02:04 | |
| *** cshen has quit IRC | 02:10 | |
| *** weifan has joined #openstack-ansible | 03:06 | |
| *** weifan has quit IRC | 03:14 | |
| *** rh-jelabarre has quit IRC | 04:03 | |
| *** cshen has joined #openstack-ansible | 04:05 | |
| *** cshen has quit IRC | 04:11 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-ansible | 04:33 | |
| *** raukadah is now known as chandankumar | 04:36 | |
| *** cshen has joined #openstack-ansible | 05:00 | |
| *** udesale has joined #openstack-ansible | 05:04 | |
| *** cshen has quit IRC | 05:06 | |
| *** itandops has quit IRC | 05:37 | |
| *** cshen has joined #openstack-ansible | 05:39 | |
| *** gkadam has joined #openstack-ansible | 05:40 | |
| *** gkadam has quit IRC | 05:40 | |
| openstackgerrit | fuzihao proposed openstack/openstack-ansible-os_manila master: Fix pygments style https://review.opendev.org/729513 | 06:14 |
| openstackgerrit | fuzihao proposed openstack/openstack-ansible-os_barbican master: Fix pygments style https://review.opendev.org/729516 | 06:23 |
| *** AJaeger has joined #openstack-ansible | 06:31 | |
| *** jawad_axd has joined #openstack-ansible | 06:31 | |
| AJaeger | jrosser: Btw. the pygments_style change itself is fine since previous openstackdocstheme used 'native' and we do not set this in version 2.2.0 anymore which can lead to different rendering. But such a change should indeed come with proper explanation... | 06:32 |
| AJaeger | and not just copying something without knowing why ;( | 06:33 |
| *** yolanda has joined #openstack-ansible | 06:33 | |
| jrosser | AJaeger: for OSA it's like 30+ repos to do so having no explanation or reference turns a small global chance like that in to a really big deal | 06:34 |
| jrosser | *change | 06:34 |
| AJaeger | jrosser: I agree completely! | 06:35 |
| AJaeger | I just wanted to give you that background why it might be beneficial in the end... but let's see what comes back in the review | 06:35 |
| AJaeger | jrosser: here's my narrative: https://review.opendev.org/#/c/729514/ | 06:36 |
| jrosser | AJaeger: ah interesting theres a bit more going on there | 06:40 |
| AJaeger | jrosser: openstackdocstheme 2.2.0 is a large change - and we'll soon sent an email to the discuss list | 06:41 |
| AJaeger | there were a few surprises in it for us ;( | 06:42 |
| openstackgerrit | fuzihao proposed openstack/openstack-ansible-os_barbican master: Fix pygments style https://review.opendev.org/729516 | 06:42 |
| *** cshen has quit IRC | 06:43 | |
| openstackgerrit | fuzihao proposed openstack/openstack-ansible-os_manila master: Fix pygments style https://review.opendev.org/729513 | 06:45 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 06:50 |
| *** cshen has joined #openstack-ansible | 07:06 | |
| *** tosky has joined #openstack-ansible | 07:30 | |
| jrosser | AJaeger: i wonder if there is a way of making openstackdocs_repo_name and openstackdocs_bug_project be defined outside the docs conf.py | 07:36 |
| AJaeger | it's all python, isn't it? | 07:36 |
| jrosser | if that were the case OSA could move the authoritative copy of these config files into here https://github.com/openstack/openstack-ansible-tests/tree/master/sync | 07:37 |
| jrosser | then the proposal bot would do the hard work | 07:37 |
| AJaeger | mmh, you need to experiment a bit for that, never seen that in OpenStack | 07:38 |
| *** also_stingrayza has joined #openstack-ansible | 07:42 | |
| *** stingrayza has quit IRC | 07:47 | |
| *** cshen_ has joined #openstack-ansible | 07:58 | |
| *** cshen has quit IRC | 08:01 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 08:01 |
| openstackgerrit | Adrien Cunin proposed openstack/openstack-ansible master: Actually destroy container even if keeping data https://review.opendev.org/729533 | 08:01 |
| janno | did someone switch stein from python 2.7 to python 3.x with openstack-ansible? | 08:19 |
| jrosser | janno: i hope not! what are you seeing? | 08:32 |
| janno | jrosser: we didn't either | 08:33 |
| janno | jrosser: i was wondering if this is supported | 08:33 |
| *** gshippey has joined #openstack-ansible | 08:35 | |
| *** redrobot has quit IRC | 08:37 | |
| jrosser | janno: the first time you get some python3 is with Train | 08:38 |
| jrosser | janno: in Train we did a lot of conversion work and added switches all over the place like this https://github.com/openstack/openstack-ansible-os_glance/blob/stable/train/defaults/main.yml#L21 | 08:39 |
| jrosser | so it's "dual stack python" for Train, the roles default to py2 but a global setting in openstack-ansible overrides that to actually deploy as py3 | 08:39 |
| ebbex | Re-running ceph-install.yml on stable/train:20.1.0 seems to do some updates to authentication, and all running vm's lose write access. Anyone else experience this? | 08:39 |
| jrosser | ebbex: there have been a fairly steady trickle of fixes on train for ceph auth https://github.com/openstack/openstack-ansible-ceph_client/commits/stable/train | 08:47 |
| jrosser | could you see if any of those are suspect | 08:47 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: [ussuri][goal] Update contributor documentation https://review.opendev.org/727460 | 08:48 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Allow minimal auditbeat deploy on LXC containers https://review.opendev.org/725284 | 08:48 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: [ussuri][goal] Update contributor documentation https://review.opendev.org/727462 | 08:49 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: [ussuri][goal] Update contributor documentation https://review.opendev.org/727402 | 08:55 |
| *** aedc has joined #openstack-ansible | 09:00 | |
| ebbex | jrosser: thanks, i'll have a look :) | 09:12 |
| janno | jrosser: thanks | 09:30 |
| janno | we are currently trying to deploy a stein cluster for testing purposes and saw many problems with the gnocchi containers. | 09:41 |
| janno | apparently, all the problems were caused by the gnocchi role using the upper constraints from the master, rather than the corresponding commit. | 09:41 |
| janno | which in turn has led to attempts to install packages that are only compatible with Python 3. | 09:42 |
| janno | I wonder if these are known problems. | 09:44 |
| janno | Or if actually nobody uses Stein or Gnocchi anymore. | 09:49 |
| jrosser | janno: it should come from here https://github.com/openstack/openstack-ansible-os_gnocchi/blob/stable/stein/defaults/main.yml#L45 | 10:03 |
| jrosser | i guess i'd be adding a debug: task to the gnochhi role to see what the value was | 10:04 |
| janno | yes, but requirements_git_install_branch and requirements_git_url are not defined there. therefore it falls back to https://git.openstack.org/cgit/openstack/requirements/plain/upper-constraints.txt?h=master | 10:05 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-ceph_client master: Install octopus ceph client https://review.opendev.org/729329 | 10:06 |
| janno | This happens, because os-gnocchi-install.yml does not include openstack_services.yml from repo_packages | 10:08 |
| janno | https://github.com/openstack/openstack-ansible/blob/c936cf259ac1a406be41272d623df08e70575669/playbooks/os-gnocchi-install.yml#L20 | 10:08 |
| noonedeadpunk | jrosser: btw octopus kinda passed but have the same thing as with mysql client and server issue | 10:08 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 10:08 |
| jrosser | noonedeadpunk: yeah i saw | 10:09 |
| jrosser | janno: that looks like a bug, it would be worth checking the later branches to see if there is a patch which didnt get backported to stein | 10:10 |
| * noonedeadpunk reads back | 10:10 | |
| janno | jrosser: It's still missing | 10:11 |
| jrosser | yes, i've just been looking and you're right | 10:11 |
| noonedeadpunk | that's totally a bug and wondering how it was working all this time (at least for me) | 10:12 |
| noonedeadpunk | janno: do you wanna to patch this?:) | 10:23 |
| janno | I can try :) | 10:24 |
| *** itandops has joined #openstack-ansible | 10:46 | |
| *** watersj has quit IRC | 10:50 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-nspawn_hosts master: Return cross repo jobs to voting https://review.opendev.org/729565 | 10:51 |
| *** watersj has joined #openstack-ansible | 10:51 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 11:01 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-nspawn_hosts master: Return cross repo jobs to voting https://review.opendev.org/729565 | 11:16 |
| *** cshen_ has quit IRC | 11:16 | |
| *** cshen has joined #openstack-ansible | 11:17 | |
| *** cshen has quit IRC | 11:22 | |
| *** cshen has joined #openstack-ansible | 11:29 | |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_container_create master: Ensure that facts are gathered for newly created containers https://review.opendev.org/728945 | 11:37 |
| *** sshnaidm is now known as sshnaidm|afk | 11:42 | |
| jrosser | noonedeadpunk: nspawn is now all working again https://review.opendev.org/#/c/729565/ | 11:50 |
| noonedeadpunk | Great! | 11:51 |
| noonedeadpunk | Will bump branching to it | 11:51 |
| openstackgerrit | Jan Marquardt proposed openstack/openstack-ansible master: Include openstack_services.yml as vars file for Gnocchi role https://review.opendev.org/729572 | 11:52 |
| watersj | trying deploy stable/stein on ubuntu 18.04 hosts, during ceph client install I get Packages were downgraded and -y was used without --allow-downgrades. Need use old version of ubuntu 16.04? | 11:53 |
| jrosser | watersj: the caph client role *should* be setting up apt package pinning to ensure that the ceph packages only come from a particular repo | 11:54 |
| jrosser | could you check if that is happening? | 11:55 |
| jrosser | here is the data we set for stein https://github.com/openstack/openstack-ansible-ceph_client/blob/stable/stein/defaults/main.yml#L43 | 11:56 |
| jrosser | you can use apt policy <packagename> to debug whats got installed and what the candidates were | 11:57 |
| watersj | polcy pointing @ download.ceph.com, mimic, priority 1001. | 12:03 |
| *** rh-jelabarre has joined #openstack-ansible | 12:06 | |
| watersj | jrosser, could I override that and use nautilus instead adding "ceph_stable_release: nautilus" to my user_variables.yml | 12:11 |
| watersj | ? | 12:11 |
| jrosser | oh.... well are you deploying ceph with OSA? | 12:11 |
| jrosser | or do you have a standalone ceph cluster that you integrate with? | 12:12 |
| watersj | no, having it setup glance, cinder, nova talk to existing cluster | 12:12 |
| watersj | ceph cluster... | 12:12 |
| jrosser | then you can override the variable for ceph_client no problem | 12:13 |
| jrosser | is is more of a problem when OSA deploys ceph as there is only one ceph release for each OSA release that we test | 12:13 |
| jrosser | but for using just ceph_client to connect to your own ceph cluster you have a lot more freedon | 12:14 |
| jrosser | noonedeadpunk: talking of ceph, are you happy with the nautilus -> octopus change, maybe thinking of manila? | 12:17 |
| noonedeadpunk | Yeah, I think we probably can deploy octopus for U. And what's wrong with manila at the moment? I think N->O in terms of ceph not so mach has changed in terms of architecture? | 12:21 |
| noonedeadpunk | Except I left it half backed at the moment (int terms of container deployments) | 12:22 |
| *** udesale_ has joined #openstack-ansible | 12:23 | |
| noonedeadpunk | Or you mean to add some noop patch for manila to check? | 12:23 |
| jrosser | more that, yes i guess | 12:23 |
| jrosser | just sanity check that it doesnt all blow up with octopus instead of nautilus | 12:24 |
| jrosser | we could merge the changes now, just wanted to check if you were happy doing that | 12:24 |
| *** udesale has quit IRC | 12:25 | |
| jrosser | noonedeadpunk: it looks like the galera_server merge might pass | 12:25 |
| jrosser | that then needs a thorough review to see if i've broken anything | 12:25 |
| jrosser | in particular i'm wanting to look at anything that used to be in the galera_client defaults thats not got properly moved over to the combined role | 12:26 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: DNM https://review.opendev.org/728578 | 12:27 |
| jrosser | the upgrade job worries me too for the galera change, i don't think we're actually testing those changes in the upgrade job | 12:28 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests master: Add ansible openstack collection https://review.opendev.org/718362 | 12:36 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests master: Add ansible openstack collection https://review.opendev.org/718362 | 12:36 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests master: Add ansible openstack collection https://review.opendev.org/718362 | 12:36 |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_hosts master: Return cross repo jobs to voting https://review.opendev.org/729565 | 12:37 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Add ansible openstack collection https://review.opendev.org/718431 | 12:37 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Add ansible openstack collection https://review.opendev.org/718431 | 12:38 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_glance master: Use ansible openstack collection https://review.opendev.org/718473 | 12:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_glance master: Use ansible openstack collection https://review.opendev.org/718473 | 12:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_magnum master: Use ansible openstack collection https://review.opendev.org/718633 | 12:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_keystone master: Use ansible openstack collection https://review.opendev.org/718632 | 12:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_gnocchi master: Use ansible openstack collection https://review.opendev.org/718626 | 12:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible openstack collection https://review.opendev.org/718644 | 12:40 |
| *** itandops has quit IRC | 12:52 | |
| *** sshnaidm|afk is now known as sshnaidm | 12:53 | |
| noonedeadpunk | jrosser: I think we would eventually test once it merge.... but yeah - it might fail eventually | 12:54 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_octavia master: Use ansible openstack collection https://review.opendev.org/718635 | 12:55 |
| jrosser | noonedeadpunk: i was thinking about recording the git sha of all the required_projects repos early in the job | 12:55 |
| jrosser | then in the second part of the upgrade job those could be the target to checkout rather than the branch | 12:56 |
| noonedeadpunk | that is smth I've wrote down for discussion for ptl) | 12:56 |
| noonedeadpunk | *ptg | 12:56 |
| jrosser | maybe even the zuul setup should be doing that for us | 12:57 |
| *** itandops has joined #openstack-ansible | 12:59 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_trove master: Use ansible openstack collection https://review.opendev.org/718641 | 13:00 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_swift master: Use ansible openstack collection https://review.opendev.org/718638 | 13:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Use ansible openstack collection https://review.opendev.org/718639 | 13:05 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_designate master: Use ansible openstack collection https://review.opendev.org/718625 | 13:06 |
| *** cshen has quit IRC | 13:06 | |
| jrosser | chandankumar: arxcruz are you looking at using ansible collections yet, see this in os_tempest https://review.opendev.org/718639 | 13:07 |
| noonedeadpunk | jrosser: so you think we should also move to colelctions in U? | 13:12 |
| jrosser | i am more in favour of that than waiting a cycle i think | 13:13 |
| jrosser | particularly as we would be consuming the openstack modules independant of ansible releases then, and fixing stuff should be much easier | 13:13 |
| jrosser | i'm going to look next at getting depends-on working for the collections | 13:14 |
| jrosser | i think it would be OK to move everything except os_tempest if the tripleo folk are not ready yet | 13:14 |
| *** jamesdenton has joined #openstack-ansible | 13:15 | |
| *** dpaclt has joined #openstack-ansible | 13:15 | |
| dpaclt | Hi All,I am getting Invalid input for operation: physical_network 'lbaas' unknown for flat provider network. error while deploying octavia ..Any thoughts ? | 13:19 |
| jrosser | jamesdenton: hello! i had a question for you about our neutron ovn jobs if you have moment | 13:21 |
| jrosser | dpaclt: octavia needs a management network that connects the octavia containers to the octavia amphora virtual machines | 13:27 |
| jamesdenton | hi jrosser | 13:27 |
| jrosser | you need to provision br-lbaas on your infra nodes however you like (flat or vlan), make sure that gets also to your computes and configure a neutron provider network | 13:28 |
| jrosser | jamesdenton: o/ hello | 13:28 |
| jamesdenton | jrosser been moving and without access for a bit. apologies. whats up? | 13:28 |
| jrosser | i've been fixing up some of our neutron role jobs, the ovn one was broken https://review.opendev.org/#/c/728804/ | 13:28 |
| dpaclt | http://paste.openstack.org/show/793804/ <<<yeah i have defined a network for it ..Please have a look | 13:29 |
| jrosser | the code having moved is not problem and easy to sort out | 13:29 |
| jamesdenton | k | 13:29 |
| jrosser | but this goes badly wrong https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_8db/728804/1/check/openstack-ansible-ovn-ubuntu-bionic/8dbb66a/logs/ara-report/result/ce33bc4e-38b9-440f-a572-f8461ea3fe49/ | 13:29 |
| jamesdenton | yes, i see that | 13:30 |
| jamesdenton | i will see if i can spin up an AIO today and take a closer look | 13:31 |
| jrosser | just asking if you'd seen any neutron db migrations blow up, really | 13:31 |
| openstackgerrit | amolkahat proposed openstack/openstack-ansible-os_tempest master: Added tempest ironic resources setup. https://review.opendev.org/720705 | 13:31 |
| jamesdenton | probably? :D | 13:32 |
| jrosser | jamesdenton: actually that test is not an AIO, it's a "./run_tests.sh ovn" in the os_neutron repo | 13:32 |
| jrosser | i guess it's a decision about how much effort to fix it if its trivial vs. we make the OVN job non voting and never look at it again :) | 13:33 |
| jrosser | given the large work around ovn in neutron i'd rather keep the job if we can | 13:34 |
| jamesdenton | well, i think it's worth making it voting, but we could temporarily disable it. i think we really need to consider OVN as the replacement for ML2/OVS in the not-so-distant future | 13:34 |
| jrosser | is that a PTG topic? | 13:35 |
| jrosser | *should that be... | 13:35 |
| jamesdenton | yes | 13:35 |
| jamesdenton | sorry, i haven't looked at any of that | 13:35 |
| jrosser | no worries :) | 13:35 |
| *** cshen has joined #openstack-ansible | 13:36 | |
| *** cshen has quit IRC | 13:40 | |
| *** Brace has joined #openstack-ansible | 14:08 | |
| *** fghaas has joined #openstack-ansible | 14:12 | |
| *** KeithMnemonic has joined #openstack-ansible | 14:19 | |
| Brace | hi all, I've been looking through the docs to try and figure out how to upgrade the nodes on an OpenStack install from Ubuntu 16.04 to 18.04 | 14:32 |
| Brace | but I can't find much to be honest | 14:32 |
| Brace | can anyone point me at a guide? | 14:32 |
| *** spatel has joined #openstack-ansible | 14:34 | |
| *** threestrands has quit IRC | 14:34 | |
| jamesdenton | i'm not sure if anything like that exists publicly, to be honest | 14:37 |
| spatel | Not sure how important its but i create dirty virsh wrapper script to get more info during troubleshooting - http://paste.openstack.org/show/793808/ | 14:49 |
| spatel | Do you think we should push it out to OSA ? | 14:49 |
| jamesdenton | might be good for https://github.com/openstack/openstack-ansible-ops | 14:50 |
| watersj | Brace, are you planning on upgrading you openstack to later version? | 14:50 |
| watersj | Brace, if not you could drop node, upgrade, redeploy | 14:50 |
| spatel | jamesdenton: it has to be on compute nodes to run.. we can run remotely over SSH session if key is pushed out | 14:51 |
| jrosser | spatel: are you using prometheus at all? | 14:55 |
| spatel | jrosser: no i am using influxDB/grafana | 14:56 |
| spatel | jrosser: why? | 14:56 |
| jrosser | we're just starting to roll this out https://github.com/vexxhost/libvirtd_exporter | 14:56 |
| jrosser | which would be sort of similar thing but API endpoint | 14:56 |
| spatel | Hmm.. interesting so how does it work.. all compute will report to prometheus? | 14:58 |
| jrosser | other way round, it makes an http endpoint prometheus (or anythin else) scrapes | 14:58 |
| spatel | prometheus pool data from all compute nodes right? | 15:00 |
| spatel | pull* | 15:00 |
| *** cshen has joined #openstack-ansible | 15:00 | |
| jrosser | looks like you can glue that all together https://github.com/influxdata/telegraf/tree/master/plugins/inputs/prometheus | 15:00 |
| jrosser | spatel: yes thats right, it's a timeseries kind of like infuxdb. main difference is the pull based collection all over http which is becoming a sort de-facto standard | 15:02 |
| spatel | Currently all my compute nodes running telegraf and sending data to influx for all kind of data CPU/MEM and that is how i am monitoring cloud and capacity | 15:02 |
| spatel | I need push base because pull base not scalable sometime :( | 15:02 |
| jrosser | you could put the libvirt collector on and have telegraf collect from localhost | 15:02 |
| jrosser | importantly (for us) the data contains user and project for each vm | 15:03 |
| spatel | jrosser: this kind of data i am sending to influx, using virsh to report CPU usage etc, i don't like ceilometer so all doing home brew stuff - https://imgur.com/a/VcdRKZd | 15:05 |
| spatel | jrosser: also i am not in public domain, this is all private cloud so we don't care about some project related metrics etc.. | 15:07 |
| Brace | watersj: yes its' currently Rocky, once we've gone up to 18.04, we're then going to go Rocky > Stein > Train | 15:16 |
| Brace | watersj: but we need to upgrade all of the nodes in the cluster | 15:18 |
| *** itandops has quit IRC | 15:19 | |
| jrosser | Brace: there is work-in-progress documentation here https://review.opendev.org/#/c/710803/ | 15:20 |
| jrosser | imho folk tend to get in a mess when trying to do a partial upgrade, like upgrade the controllers but only do a subset of computes | 15:21 |
| jrosser | then try to do a point-release OSA upgrade in the middle of that | 15:21 |
| *** itandops has joined #openstack-ansible | 15:22 | |
| jrosser | it's best to try to work through everything and get a uniform OS before tackling any other upgrades | 15:23 |
| watersj | jrosser, stand up new cluster and migrate? | 15:23 |
| jrosser | that should not be necessary, i've done Xenial->Bionic on Rocky, but it was a bunch of work to do | 15:23 |
| Brace | we've got quite a lot of stuff running on the cluster, so while we have talked about building a new cluster, we'd rather do it in place if possible | 15:24 |
| Brace | so do all the controllers and all the compute and then do the upgrades | 15:24 |
| Brace | that is what we were planning on doing | 15:24 |
| jrosser | do the controllers first, one by one, carefully. paying particular attention to keeping rabbitmq and galera clusters healthy | 15:24 |
| Brace | we've got ceph as well, I assume that complicates things | 15:25 |
| jrosser | deployed with OSA? | 15:25 |
| Brace | yes | 15:25 |
| jrosser | more work-in-progress stuff https://review.opendev.org/#/c/711440/ | 15:26 |
| Brace | I assume that you can't just do an in place upgrade of ubuntu | 15:35 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-tests master: Use galera server role to install galera client https://review.opendev.org/725902 | 15:35 |
| Brace | you need to do a clean reinstall of the OS | 15:36 |
| mgariepy | Brace for the controllers, you might want to scale down the number of dhcp agent (if you resolve dns in the vms from the dhcp_agents) and move ports arounds if you don't want the dns resolve to timeout. | 15:37 |
| mgariepy | for each network that is. | 15:37 |
| jrosser | Brace: in-place upgrades may be possible, but again care is needed | 15:39 |
| jrosser | note that the venvs contain python libs which have bindings against system shared libraries, like libvirt and ceph | 15:40 |
| *** gyee has joined #openstack-ansible | 15:40 | |
| mgariepy | Brace, https://etherpad.opendev.org/p/osa-rocky-bionic-upgrade | 15:41 |
| mgariepy | it doesn't covert everything but it's a start. | 15:41 |
| Brace | mgariepy: thanks for that | 15:45 |
| mgariepy | you might want to deactivate the router gently on the neutron nodes (before sapping it..) | 15:47 |
| mgariepy | zapping** | 15:47 |
| mgariepy | when rescheduling 10s or 100s of router it can take a while for stuff to stabilize. not sure how much router/network you have on your nodes but it can be disruptive to do so. | 15:49 |
| jrosser | mgariepy: do you have any tips for if i have a broken network node but the active router is still on the broken node? | 15:49 |
| jrosser | i.e "force node <foo> to be active for all the routers, regardless" | 15:49 |
| jrosser | we had an nvme disk with a pcie error bad enough to break the host (no ssh, no new processes could start) but otherwise everything that was previously running remained | 15:50 |
| mgariepy | i run my router in not-ha mode | 15:51 |
| jrosser | that would do it :) | 15:51 |
| mgariepy | i never had a host failure yet. | 15:51 |
| mgariepy | i move router around gently when i need to. but it's really a mess and when there are a lot of movement that need to be done it takes a really long time. | 15:52 |
| jrosser | if it were a physical router i'd go and fiddle the vrrp weight and magic would happen | 15:52 |
| jrosser | but i couldnt see an equivalent concept for neutron | 15:52 |
| mgariepy | the router in ha are scheduled on all the 3 network node i guess ? | 15:53 |
| jrosser | i have 2 network nodes there, but yes | 15:53 |
| mgariepy | not sure you would be able to edit the router on the broken node. | 15:54 |
| jrosser | perhaps shutting the switchports to the broken node would have made it fail over | 15:54 |
| mgariepy | indeed | 15:54 |
| mgariepy | how many router ? | 15:54 |
| jrosser | oh a very small number, maybe 5 max | 15:55 |
| mgariepy | ha. | 15:55 |
| mgariepy | yep turn the switch port off. | 15:55 |
| mgariepy | haha | 15:55 |
| mgariepy | only 5 would not be a big issue. | 15:56 |
| jrosser | it was irritating because all the processes still seemed to be running, but everything was completely hosed for remote management, even the IPMI KVM | 15:56 |
| jrosser | now we do many many SSD firmware upgrades which is not fun | 15:56 |
| mgariepy | with 100s with 3 agents, the network was getting really confused (router with ha-active on all the nodes, and innactive on all the nodes..) | 15:57 |
| mgariepy | for the controllers it's somewhat easy, no big impact for the clients. | 15:58 |
| noonedeadpunk | jrosser: I think manilla is happy with octopus as well https://review.opendev.org/#/c/728578/ | 15:58 |
| jrosser | noonedeadpunk: cool - shall we merge that? | 15:59 |
| jrosser | noonedeadpunk: also we need to decide on what to do with magnum for centos/disrto | 15:59 |
| noonedeadpunk | except it isntalled nautilus *( | 15:59 |
| jrosser | OH | 15:59 |
| noonedeadpunk | yeah https://zuul.opendev.org/t/openstack/build/df75a523a4ca46a383134430fb638994/log/logs/etc/host/apt/sources.list.d/download_ceph_com_debian_nautilus.list.txt | 15:59 |
| mgariepy | but for the hypervisor it's another story moving vms around unless they are ceph backed is not fun. | 15:59 |
| mgariepy | unless you have pci passthrough or other fancy stuff going on. | 16:00 |
| noonedeadpunk | yeah - this time tests failed in too weird way https://c9ef8b5b39218c9d38a6-083f3205a01a368b196dd0a8486413e5.ssl.cf1.rackcdn.com/727387/2/check/openstack-ansible-deploy-aio_distro_metal-centos-7/cfb8404/logs/openstack/aio1-utility/stestr_results.html | 16:02 |
| noonedeadpunk | let me disable them | 16:02 |
| jrosser | ceph-common and ceph_client have both installed nautilus, something wierd going on there | 16:03 |
| noonedeadpunk | yeah. maybe just depends on didn't worked | 16:04 |
| noonedeadpunk | I was restoring patch and then editing | 16:04 |
| noonedeadpunk | So maybe just try to recheck... | 16:04 |
| jamesdenton | did something change with doc colors? https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-ovn.html | 16:04 |
| jamesdenton | a lot of the code formatting colors are... off | 16:04 |
| jrosser | see ML post just now from AJaeger | 16:04 |
| jrosser | looks like we need to do a bazillion patches there | 16:05 |
| jamesdenton | :facepalm: | 16:05 |
| jamesdenton | ok | 16:05 |
| noonedeadpunk | you mean that one? http://lists.openstack.org/pipermail/openstack-discuss/2020-May/014658.html | 16:06 |
| AJaeger | noonedeadpunk: http://lists.openstack.org/pipermail/openstack-discuss/2020-May/014971.html | 16:07 |
| noonedeadpunk | (not likely) | 16:07 |
| jamesdenton | thanks AJaeger | 16:07 |
| AJaeger | jrosser, jamesdenton, sorry, that was an unexpected side effect with the doc colors - otherwise we would have waited ;( | 16:07 |
| jamesdenton | no worries | 16:08 |
| jrosser | hilariously "invert display colours" doesnt actually help | 16:08 |
| jrosser | its just differently bad | 16:08 |
| AJaeger | you can mark the black box and suddenly it appears ;) | 16:08 |
| jamesdenton | i guess the text is black, too? is that what happened? | 16:09 |
| jamesdenton | i see mention of the theme style (native vs sphinx) | 16:09 |
| AJaeger | do you want me to send a mass update to just change pygments_style? That works with both old and new theme... | 16:10 |
| jrosser | AJaeger: the only reason i -1 that patch from earlier is because i didnt know if there was more broken than just the styling given the patch you showed me to designate | 16:11 |
| * AJaeger can quickly script something for your repos... | 16:11 | |
| AJaeger | jrosser: that commit message was soo... | 16:11 |
| jrosser | well yes :) | 16:11 |
| AJaeger | jrosser: so, yes, that's the right change for you for now. | 16:11 |
| mgariepy | jrosser, it doesnt seems like neutron supports changing the priority for a router keepalived process. | 16:11 |
| AJaeger | My designate change is the best way forward once you're ready for victoria | 16:12 |
| jrosser | AJaeger: if you're happy that just updating the style doesnt leave us with residual trouble in the immediate term, thats cool, patches welcome :) | 16:12 |
| AJaeger | jrosser: yes, I'm happy. Let me push first change - review - and then I push the rest... | 16:12 |
| jrosser | AJaeger: part of the issue i guess is we've not quite branched yet - so as hinted at in the patch this morning we currently have master requirements even though we're technically still finishing U | 16:13 |
| jrosser | so anything that gets merged now will end up on U | 16:14 |
| jrosser | (ish) | 16:14 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_magnum master: Set centos-7 distro jobs as non-voting https://review.opendev.org/729648 | 16:14 |
| noonedeadpunk | Technically we can kinda branch with what we have on our hands right now | 16:15 |
| noonedeadpunk | and backport the rest for rc2 | 16:15 |
| jrosser | yes, its our choice really if we pull the role SHA forward or not | 16:15 |
| noonedeadpunk | As looking at amount of things left I'm worried that we never branch | 16:15 |
| *** redrobot has joined #openstack-ansible | 16:16 | |
| noonedeadpunk | And we continue adding to the list lol | 16:16 |
| jrosser | should we make a list for that? etherpad? | 16:16 |
| AJaeger | jrosser: so, if you branch now and use ussuri constraints, you don't have the problem anymore... | 16:16 |
| noonedeadpunk | let' call it https://etherpad.opendev.org/p/openstack-ansible-ussuri | 16:16 |
| jrosser | AJaeger: but still on master we will have a problem? | 16:17 |
| AJaeger | jrosser: yes, you would - but I would push the full patch (see designate) to fix it there... | 16:17 |
| AJaeger | We can merge the quick-aid now, it does not harm... | 16:17 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Create Ussuri https://review.opendev.org/729258 | 16:19 |
| noonedeadpunk | Ok, I've rolled back ceph-ansible to 4-0 here ^ | 16:20 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Create Ussuri https://review.opendev.org/729258 | 16:21 |
| jrosser | as the changes required for octopus seem quite small (nothing like the work for nautilus) then thats looking like an easy backport | 16:21 |
| noonedeadpunk | yeah | 16:22 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_neutron master: git branch -D pygments_style new-branch.sh pygments_style https://review.opendev.org/729651 | 16:22 |
| noonedeadpunk | I think ceph even have l->o direct upgrade path :p | 16:22 |
| AJaeger | jrosser, noonedeadpunk, jamesdenton ^ | 16:23 |
| AJaeger | if that is fine for you and you need it, I can spam you ;) | 16:23 |
| noonedeadpunk | With that commit msg?:) | 16:23 |
| openstackgerrit | Duncan Martin Walker proposed openstack/openstack-ansible-ops master: Ensure that the beat setup role runs when elk_beat_setup=true https://review.opendev.org/729654 | 16:23 |
| AJaeger | noonedeadpunk: something went wrong on my side ;( | 16:23 |
| jamesdenton | +1 lol | 16:23 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_neutron master: Update docstheme for style and PDF https://review.opendev.org/729651 | 16:24 |
| AJaeger | trying again, guess you spotted my error ;) | 16:24 |
| jrosser | automation-tastic | 16:25 |
| jamesdenton | thanks so much | 16:25 |
| AJaeger | so, want it everywhere? | 16:25 |
| * jrosser wonders if we wait 5 mins and look at the docs preview..... | 16:26 | |
| AJaeger | we can... | 16:27 |
| noonedeadpunk | also we really need to figure out how to merge galera change | 16:29 |
| noonedeadpunk | As https://review.opendev.org/#/c/728434/ not going to pass | 16:29 |
| noonedeadpunk | So we can literally disable all ci for that | 16:29 |
| noonedeadpunk | oh, wait | 16:30 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP] Use galera server role to install galera client https://review.opendev.org/728434 | 16:31 |
| *** aedc_ has joined #openstack-ansible | 16:32 | |
| jrosser | noonedeadpunk: surely that needs some depends-on? | 16:33 |
| noonedeadpunk | but, um, to what? | 16:34 |
| jrosser | omg | 16:34 |
| noonedeadpunk | Like the only one I can see will cause circular one | 16:34 |
| jrosser | /o\ i see what you mean now | 16:34 |
| *** aedc has quit IRC | 16:35 | |
| noonedeadpunk | So maybe we can jsut drop from a-r-r in second path | 16:36 |
| noonedeadpunk | *stage | 16:36 |
| jrosser | i think that is the only way | 16:37 |
| jrosser | i did that just to be absolutely sure the old client role was still called anywhere | 16:38 |
| jrosser | *was not | 16:38 |
| AJaeger | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_5b9/729651/2/check/openstack-tox-docs/5b9ece4/docs/ is ready | 16:41 |
| jrosser | cool | 16:42 |
| AJaeger | jamesdenton: is https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_5b9/729651/2/check/openstack-tox-docs/5b9ece4/docs/app-ovn.html now ok? | 16:42 |
| jamesdenton | yep, looks good, thank you | 16:42 |
| AJaeger | then I'll push the rest... | 16:42 |
| jamesdenton | i am happy to roll thru and +2 additional patches | 16:42 |
| jamesdenton | kk | 16:42 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-config_template master: Update docstheme for style and PDF https://review.opendev.org/729660 | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible master: Update docstheme for style and PDF https://review.opendev.org/729661 | 16:43 |
| jamesdenton | jrosser i think i found the culprit with the OVN stuff and found another issue. Running tests now | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-python_venv_build master: Update docstheme for style and PDF https://review.opendev.org/729662 | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-systemd_mount master: Update docstheme for style and PDF https://review.opendev.org/729663 | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-systemd_networkd master: Update docstheme for style and PDF https://review.opendev.org/729664 | 16:43 |
| AJaeger | https://review.opendev.org/#/q/topic:pygments_style+is:open | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-systemd_service master: Update docstheme for style and PDF https://review.opendev.org/729665 | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-uwsgi master: Update docstheme for style and PDF https://review.opendev.org/729666 | 16:43 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-hardening master: Update docstheme for style and PDF https://review.opendev.org/729667 | 16:44 |
| jamesdenton | jrosser if you don't mind i'll just hijack that patch and make the changes | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-apt_package_pinning master: Update docstheme for style and PDF https://review.opendev.org/729668 | 16:44 |
| jrosser | jamesdenton: absolutely, go for it | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-ceph_client master: Update docstheme for style https://review.opendev.org/729669 | 16:44 |
| AJaeger | sorry, I should have removed "and PDF" from commit message - now updated... | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-galera_client master: Update docstheme for style https://review.opendev.org/729670 | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-galera_server master: Update docstheme for style https://review.opendev.org/729671 | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-haproxy_server master: Update docstheme for style https://review.opendev.org/729672 | 16:44 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-lxc_container_create master: Update docstheme for style https://review.opendev.org/729673 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-config_template master: Update docstheme for style https://review.opendev.org/729660 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-lxc_hosts master: Update docstheme for style https://review.opendev.org/729674 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible master: Update docstheme for style https://review.opendev.org/729661 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-memcached_server master: Update docstheme for style https://review.opendev.org/729675 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/ansible-role-python_venv_build master: Update docstheme for style https://review.opendev.org/729662 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-nspawn_container_create master: Update docstheme for style https://review.opendev.org/729676 | 16:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-nspawn_hosts master: Update docstheme for style https://review.opendev.org/729677 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-openstack_hosts master: Update docstheme for style https://review.opendev.org/729678 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-openstack_openrc master: Update docstheme for style https://review.opendev.org/729679 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-ops master: Update docstheme for style https://review.opendev.org/729680 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_aodh master: Update docstheme for style https://review.opendev.org/729681 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_barbican master: Update docstheme for style https://review.opendev.org/729682 | 16:46 |
| openstackgerrit | amolkahat proposed openstack/openstack-ansible-os_tempest master: Added tempest ironic resources setup. https://review.opendev.org/720705 | 16:46 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_blazar master: Update docstheme for style https://review.opendev.org/729683 | 16:47 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_ceilometer master: Update docstheme for style https://review.opendev.org/729684 | 16:47 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_cinder master: Update docstheme for style https://review.opendev.org/729685 | 16:47 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_cloudkitty master: Update docstheme for style https://review.opendev.org/729686 | 16:47 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_congress master: Update docstheme for style https://review.opendev.org/729687 | 16:47 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_designate master: Update docstheme for style https://review.opendev.org/729688 | 16:48 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_glance master: Update docstheme for style https://review.opendev.org/729689 | 16:48 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_gnocchi master: Update docstheme for style https://review.opendev.org/729690 | 16:48 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_heat master: Update docstheme for style https://review.opendev.org/729691 | 16:48 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_horizon master: Update docstheme for style https://review.opendev.org/729692 | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_ironic master: Update docstheme for style https://review.opendev.org/729693 | 16:49 |
| jrosser | noonedeadpunk: see "Zuul Variables" here https://zuul-ci.org/docs/zuul/reference/jobs.html | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_keystone master: Update docstheme for style https://review.opendev.org/729694 | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_magnum master: Update docstheme for style https://review.opendev.org/729695 | 16:49 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Bump MariaDB version to 10.4 https://review.opendev.org/717830 | 16:49 |
| jrosser | noonedeadpunk: specifically zuul.projects | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_manila master: Update docstheme for style https://review.opendev.org/729696 | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_masakari master: Update docstheme for style https://review.opendev.org/729697 | 16:49 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_mistral master: Update docstheme for style https://review.opendev.org/729698 | 16:50 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_murano master: Update docstheme for style https://review.opendev.org/729700 | 16:50 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_nova master: Update docstheme for style https://review.opendev.org/729701 | 16:50 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_octavia master: Update docstheme for style https://review.opendev.org/729702 | 16:50 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_panko master: Update docstheme for style https://review.opendev.org/729703 | 16:50 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_placement master: Update docstheme for style https://review.opendev.org/729704 | 16:51 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_rally master: Update docstheme for style https://review.opendev.org/729705 | 16:51 |
| noonedeadpunk | jrosser: I think because I did smth inappropriate somewhere but don't catch where... | 16:51 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_sahara master: Update docstheme for style https://review.opendev.org/729706 | 16:51 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_swift master: Update docstheme for style https://review.opendev.org/729707 | 16:51 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_tacker master: Update docstheme for style https://review.opendev.org/729708 | 16:52 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_tempest master: Update docstheme for style https://review.opendev.org/729709 | 16:52 |
| noonedeadpunk | jrosser: or I just didn't catch the context | 16:52 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_trove master: Update docstheme for style https://review.opendev.org/729710 | 16:52 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-os_zun master: Update docstheme for style https://review.opendev.org/729711 | 16:52 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-plugins master: Update docstheme for style https://review.opendev.org/729712 | 16:52 |
| jrosser | oh no sorry, about the upgrade jobs, looks like there is a var passed into the jobs with all the projets and git version info already | 16:52 |
| jrosser | -ECONTEXT ! | 16:52 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-rabbitmq_server master: Update docstheme for style https://review.opendev.org/729713 | 16:53 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-repo_server master: Update docstheme for style https://review.opendev.org/729714 | 16:53 |
| AJaeger | jamesdenton: hope you have a script to +2 or use gertty ;) | 16:53 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-rsyslog_server master: Update docstheme for style https://review.opendev.org/729715 | 16:53 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-rsyslog_client master: Update docstheme for style https://review.opendev.org/729716 | 16:53 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-ansible-tests master: Update docstheme for style https://review.opendev.org/729717 | 16:53 |
| jamesdenton | AJaeger i prefer the most inefficient way possible | 16:53 |
| AJaeger | lol | 16:53 |
| AJaeger | done! have fun ;) | 16:54 |
| jamesdenton | i don't do any of this frequently enough to know the tricks of the trade | 16:55 |
| jamesdenton | so thanks for the suggestion | 16:55 |
| AJaeger | So, https://review.opendev.org/#/q/topic:pygments_style+is:open is the review list for these 59 changes... | 16:55 |
| AJaeger | bbl | 16:55 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Bump MariaDB version to 10.4 https://review.opendev.org/717830 | 16:57 |
| noonedeadpunk | jrosser: oh, that's might work actually | 16:58 |
| *** cshen has quit IRC | 17:00 | |
| *** cshen has joined #openstack-ansible | 17:02 | |
| spatel | folks do you know we can change flavor CPU count without re-creating new flavor? | 17:03 |
| *** schwicht has quit IRC | 17:05 | |
| *** udesale_ has quit IRC | 17:07 | |
| *** cshen has quit IRC | 17:07 | |
| *** aedc has joined #openstack-ansible | 17:12 | |
| *** aedc_ has quit IRC | 17:13 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP][DNM] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 17:14 |
| noonedeadpunk | spatel: I think you just can't | 17:14 |
| noonedeadpunk | The same thing is with images metadata | 17:15 |
| spatel | noonedeadpunk: i am worried about terraform if its using UUID and if i re-create it then might cause issue | 17:15 |
| *** aedc has quit IRC | 17:16 | |
| *** aedc has joined #openstack-ansible | 17:17 | |
| noonedeadpunk | you can just create another flavor I think and update uuid for terraform. OR eventaully delete current flavor and create new one with manually providing the same id | 17:17 |
| noonedeadpunk | But all current instances will be with non-existing one then | 17:17 |
| noonedeadpunk | jrosser: sorry I didn't get your comment https://review.opendev.org/#/c/727425/2 | 17:21 |
| jrosser | noonedeadpunk: i thought this change was trying to add it to the table of contents https://review.opendev.org/#/c/727425/2/doc/source/index.rst | 17:22 |
| jrosser | but then i don't see it here https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_f85/727425/2/check/openstack-tox-docs/f85ff43/docs/ | 17:22 |
| jrosser | but i was a bit unsure if that was supposed to happen or not | 17:22 |
| noonedeadpunk | I was doing like to opposite :P The thing is, that just CONTRIBUTING.rst has been symlinked to the https://review.opendev.org/#/c/727425/2/doc/source/contributor/contributing.rst | 17:23 |
| noonedeadpunk | And as new template do not hav heading, sphinx was throwing warning which caused tests to fail | 17:23 |
| noonedeadpunk | So I either needed to add some different contributing rst or just drop this part of the docs | 17:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-apt_package_pinning master: Update docstheme for style https://review.opendev.org/729668 | 17:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-role-systemd_service master: Update docstheme for style https://review.opendev.org/729665 | 17:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-role-systemd_networkd master: Update docstheme for style https://review.opendev.org/729664 | 17:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Update docstheme for style https://review.opendev.org/729667 | 17:25 |
| *** aedc has quit IRC | 17:25 | |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-role-systemd_mount master: Update docstheme for style https://review.opendev.org/729663 | 17:25 |
| *** aedc has joined #openstack-ansible | 17:25 | |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-role-uwsgi master: Update docstheme for style https://review.opendev.org/729666 | 17:25 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron master: Update docstheme for style https://review.opendev.org/729651 | 17:26 |
| *** itandops has quit IRC | 17:29 | |
| jrosser | noonedeadpunk: ok no problem! so long as its meant to be like that, cool | 17:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Update docstheme for style https://review.opendev.org/729694 | 17:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_mistral master: Update docstheme for style https://review.opendev.org/729698 | 17:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Update docstheme for style https://review.opendev.org/729688 | 17:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_blazar master: Update docstheme for style https://review.opendev.org/729683 | 17:32 |
| *** schwicht has joined #openstack-ansible | 17:32 | |
| *** aedc has quit IRC | 17:32 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_masakari master: Update docstheme for style https://review.opendev.org/729697 | 17:32 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Update docstheme for style https://review.opendev.org/729689 | 17:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_placement master: Update docstheme for style https://review.opendev.org/729704 | 17:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_panko master: Update docstheme for style https://review.opendev.org/729703 | 17:33 |
| *** drokath has joined #openstack-ansible | 17:33 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Update docstheme for style https://review.opendev.org/729705 | 17:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic master: Update docstheme for style https://review.opendev.org/729693 | 17:34 |
| noonedeadpunk | jrosser: zuul.projects won't do the trick, unfortunatelly. they do contain all project, but also contain "checkout: master" for all of them:( | 17:34 |
| jrosser | oh thats sad | 17:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_manila master: Update docstheme for style https://review.opendev.org/729696 | 17:35 |
| jrosser | you've tried with a depends-on? | 17:35 |
| noonedeadpunk | hm. may try. but we're also insterested in like testing called role. | 17:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Update docstheme for style https://review.opendev.org/729685 | 17:36 |
| noonedeadpunk | so right now if we do change for let's say galera, during upgrade it will be still pulled as master | 17:37 |
| noonedeadpunk | so if zuul.projects do not contain specific version.... | 17:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Update docstheme for style https://review.opendev.org/729690 | 17:37 |
| *** cshen has joined #openstack-ansible | 17:37 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Update docstheme for style https://review.opendev.org/729702 | 17:38 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_openrc master: Update docstheme for style https://review.opendev.org/729679 | 17:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Update docstheme for style https://review.opendev.org/729701 | 17:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Update docstheme for style https://review.opendev.org/729684 | 17:40 |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_container_create master: Update docstheme for style https://review.opendev.org/729676 | 17:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Update docstheme for style https://review.opendev.org/729682 | 17:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat master: Update docstheme for style https://review.opendev.org/729691 | 17:40 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Update docstheme for style https://review.opendev.org/729671 | 17:41 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client master: Update docstheme for style https://review.opendev.org/729670 | 17:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_murano master: Update docstheme for style https://review.opendev.org/729700 | 17:42 |
| *** cshen has quit IRC | 17:42 | |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build master: Update docstheme for style https://review.opendev.org/729662 | 17:42 |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_hosts master: Update docstheme for style https://review.opendev.org/729677 | 17:42 |
| openstackgerrit | Merged openstack/openstack-ansible-memcached_server master: Update docstheme for style https://review.opendev.org/729675 | 17:42 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Update docstheme for style https://review.opendev.org/729680 | 17:42 |
| openstackgerrit | Merged openstack/ansible-config_template master: Update docstheme for style https://review.opendev.org/729660 | 17:42 |
| openstackgerrit | Merged openstack/openstack-ansible master: Update docstheme for style https://review.opendev.org/729661 | 17:43 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Update docstheme for style https://review.opendev.org/729678 | 17:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_congress master: Update docstheme for style https://review.opendev.org/729687 | 17:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Update docstheme for style https://review.opendev.org/729692 | 17:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Update docstheme for style https://review.opendev.org/729681 | 17:44 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: Update docstheme for style https://review.opendev.org/729672 | 17:44 |
| *** itandops has joined #openstack-ansible | 17:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Update docstheme for style https://review.opendev.org/729674 | 17:44 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Update docstheme for style https://review.opendev.org/729669 | 17:45 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cloudkitty master: Update docstheme for style https://review.opendev.org/729686 | 17:46 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: Update docstheme for style https://review.opendev.org/729673 | 17:48 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: [ussuri][goal] Update contributor documentation https://review.opendev.org/727425 | 17:48 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP][DNM] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 17:51 |
| noonedeadpunk | jrosser: I think depends-on and stuff processed here https://zuul-ci.org/docs/zuul/reference/jobs.html#var-zuul.items | 17:53 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP][DNM] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 17:54 |
| *** chandankumar is now known as raukadah | 18:06 | |
| *** itandops has quit IRC | 18:19 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 18:20 |
| *** itandops has joined #openstack-ansible | 18:20 | |
| noonedeadpunk | so regarding upgrade jobs I kinda recalled my original idea. So we do have all roles we need with the required versions cloned by zuul. So the only thing we need to do - do not touch them during initial deployment of previous branch and link them before upgrade runs, so that it used them instead of cloning | 18:24 |
| *** fghaas has quit IRC | 18:29 | |
| jrosser | ah right so we clone from the repos in the vm image on the N-1 run and checkout the N-1 branch | 18:35 |
| jrosser | then delete all that and symlink on the upgrade run | 18:35 |
| jrosser | we shouldn’t go back to cloning from opendev.org | 18:37 |
| *** gshippey has quit IRC | 18:52 | |
| *** itandops has quit IRC | 18:54 | |
| noonedeadpunk | I guess not going back should be already implemented? | 18:55 |
| *** itandops has joined #openstack-ansible | 19:10 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 19:28 |
| Brace | mgariepy: we can live with a bit of down time, as long as it recovers in a sensible fashion | 19:33 |
| *** cshen has joined #openstack-ansible | 19:37 | |
| mgariepy | Brace, how many router/network/vms ? | 19:37 |
| mgariepy | are all your vms backed by ceph storage ? | 19:38 |
| *** cshen has quit IRC | 19:41 | |
| mgariepy | ho. also if you do have local ephemeral storage 18.04 default formated ephemeral drives are not quite compatible with 16.04 ones, (default ops differs quite a bit) so if you migrate vms around the ephemeral disk might need some more handling. | 19:42 |
| *** cshen has joined #openstack-ansible | 19:44 | |
| *** schwicht has quit IRC | 19:56 | |
| *** cshen has quit IRC | 20:26 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_magnum master: Cleanup after repo_build and pip_install retirement https://review.opendev.org/727387 | 20:31 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_magnum master: Use ansible openstack collection https://review.opendev.org/718633 | 20:33 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP] Make upgrade jobs get role relevant versions https://review.opendev.org/710849 | 20:40 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_manila master: DNM https://review.opendev.org/728578 | 20:44 |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-os_neutron master: networking-ovn is now integrated into neutron https://review.opendev.org/728804 | 20:55 |
| jamesdenton | jrosser ^^ | 20:55 |
| * jrosser looks | 20:55 | |
| jrosser | oh that makes sense | 20:56 |
| jrosser | was it still grabbing networking-ovn from pip | 20:56 |
| jamesdenton | yep | 20:56 |
| jrosser | doh :) | 20:57 |
| jamesdenton | and then the path changed | 20:57 |
| jamesdenton | anyway, should be OK. tempest passed :D | 20:57 |
| jamesdenton | thanks for the heads up | 20:57 |
| *** cshen has joined #openstack-ansible | 20:57 | |
| jrosser | thanks for digging into that - its good to keep the job alive particularly if its something you thing we should pay some more attention to for V | 20:57 |
| jamesdenton | indeed | 20:58 |
| jamesdenton | i hope to free up some cycles, really need to look at a reference architecture for OVN from some of the other projects | 20:58 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-rsyslog_server master: Cleanup py27 support https://review.opendev.org/719953 | 21:01 |
| *** gyee has quit IRC | 21:01 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_manila master: Cleanup after repo_build and pip_install retirement https://review.opendev.org/727659 | 21:01 |
| *** cshen has quit IRC | 21:02 | |
| *** gyee has joined #openstack-ansible | 21:06 | |
| *** schwicht has joined #openstack-ansible | 21:25 | |
| *** rh-jelabarre has quit IRC | 22:26 | |
| *** rh-jelabarre has joined #openstack-ansible | 22:27 | |
| *** spatel has quit IRC | 22:45 | |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build stable/train: Verify version number for virtualenv>=20.0.0 https://review.opendev.org/718278 | 22:59 |
| *** drokath has quit IRC | 23:04 | |
| *** tosky has quit IRC | 23:48 | |
| *** gyee has quit IRC | 23:48 | |
| *** threestrands has joined #openstack-ansible | 23:49 | |
| *** cshen has joined #openstack-ansible | 23:53 | |
| *** cshen has quit IRC | 23:58 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!