| opendevreview | Merged openstack/openstack-ansible master: Add Octavia OVN Provider repo requirements https://review.opendev.org/c/openstack/openstack-ansible/+/870834 | 00:19 |
|---|---|---|
| opendevreview | Merged openstack/openstack-ansible stable/zed: Add Glance tempest plugin repo to testing SHA pins list https://review.opendev.org/c/openstack/openstack-ansible/+/870777 | 00:26 |
| opendevreview | chandan kumar proposed openstack/openstack-ansible-os_tempest master: [DNM] whitelist neutron pkg debug https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/870884 | 07:13 |
| opendevreview | Merged openstack/openstack-ansible-os_nova master: Support configuration of resource providers with config files https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/870559 | 09:32 |
| opendevreview | Merged openstack/ansible-role-pki master: Update variables gathering to use vars/varnames lookups https://review.opendev.org/c/openstack/ansible-role-pki/+/869664 | 09:54 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Add variable to control no_log in mq_setup role https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/869602 | 10:08 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Add variable to control no_log in service_setup role https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/869604 | 10:15 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Fix no_log variable templating in db_setup role. https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/870843 | 10:15 |
| opendevreview | Merged openstack/openstack-ansible-galera_server master: Remove "warn" parameter from command module https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/869656 | 10:19 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Disable dhcp-agent and metadata-agent for OVN https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870854 | 10:30 |
| opendevreview | Merged openstack/openstack-ansible master: Add tempest and tempest plugins to required jobs for source deploys https://review.opendev.org/c/openstack/openstack-ansible/+/870839 | 10:50 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Allow git servers for openstack services and tempest to be overridden https://review.opendev.org/c/openstack/openstack-ansible/+/869748 | 11:17 |
| *** dviroel|afk is now known as dviroel | 11:19 | |
| *** dviroel|afk is now known as dviroel | 11:54 | |
| admin1 | hi all .. if I already have a server as network node. and then removed it from network hosts, how can I make it such that it does not add it back as network node again during redploy | 12:54 |
| admin1 | i have one node which i also removed from inventory, but everytime i run the playbook, it adds it back | 12:54 |
| noonedeadpunk | admin1: so you need to remove from inventory and from openstack-user-config (or conf.d) as well | 12:56 |
| jrosser | admin1: how did you remove it from the inventory? | 12:57 |
| admin1 | manually by hand | 12:57 |
| admin1 | also if its possible to uninstall ( neutron ) related stuff cleanly from the host | 12:58 |
| jrosser | i mean did you use the inventory_manage script, or did you just edit openstack_user_config? | 12:59 |
| admin1 | i could not use the script .. but did a manual config . because i wanted to remove the l3 agents and dhcp agents from controllers | 13:01 |
| admin1 | network_hosts -> *infrastructure_hosts =>> *compute_hosts | 13:02 |
| admin1 | agents were in infra before , now they run in compute | 13:03 |
| admin1 | and want to clean infra hosts of all neutron agents not being deployed there again | 13:03 |
| jrosser | you know that just editing openstack_user_config is not removing it from the inventory | 13:04 |
| mgariepy | another point is if it were deployed in lxc container you could just delete the container and be done with the cleaning. | 13:06 |
| mgariepy | i don't think we do have anything that clean services from a host. | 13:06 |
| mgariepy | noonedeadpunk, added a comment on https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870854 | 13:44 |
| noonedeadpunk | ah damn, I was thinking about that but forgot to include in patch | 13:53 |
| jamesdenton | FWIW, DHCP Agent may still be needed for OVN+Ironic in certain scenarios | 13:55 |
| jamesdenton | can prob cross that bridge later | 13:55 |
| mgariepy | i didn't saw your update.. and i was looking where the hell that task was coming from lol | 13:55 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Disable dhcp-agent and metadata-agent for OVN https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870854 | 13:58 |
| opendevreview | James Denton proposed openstack/openstack-ansible-os_ironic master: Create local log path for Ironic Python Agent https://review.opendev.org/c/openstack/openstack-ansible-os_ironic/+/870961 | 14:18 |
| jrosser | jamesdenton: i had something along those lines here https://review.opendev.org/c/openstack/openstack-ansible-os_ironic/+/867003 | 14:20 |
| jrosser | but not configurable | 14:20 |
| jamesdenton | oh, ok. i missed that since i'm on Yoga in this env | 14:20 |
| jrosser | so that is a very good question about ironic + dhcp + ovn | 14:20 |
| jamesdenton | i'll scratch mine! | 14:20 |
| jrosser | i assume that the ovn dhcp thing really only has concern for things on chassis it runs on? | 14:21 |
| jamesdenton | i believe it did, but has been extended to support aa new baremetal port type | 14:23 |
| jamesdenton | here's one related patch | 14:23 |
| jamesdenton | https://bugs.launchpad.net/neutron/+bug/1971431 | 14:23 |
| jamesdenton | https://review.opendev.org/c/openstack/neutron/+/840316 | 14:23 |
| jamesdenton | https://review.opendev.org/c/openstack/neutron/+/840287 | 14:23 |
| jamesdenton | looks like the port gets scheduled to one of the controllers (ovn) which is then responsible for responding | 14:24 |
| jamesdenton | jrosser have you built a disk image lately for ironic? and can you share your command? | 14:28 |
| jrosser | it's in my docs patch :) | 14:28 |
| jamesdenton | well then. lemme go looking | 14:29 |
| jamesdenton | and reviewing :( | 14:29 |
| jrosser | i gave an example for aarch64 https://review.opendev.org/c/openstack/openstack-ansible-os_ironic/+/867547/11/doc/source/configure-ironic-multiarch.rst | 14:30 |
| jrosser | but actually thats totally portable | 14:30 |
| jamesdenton | sure | 14:30 |
| jrosser | oh and thats efiboot as well, as you'd need for arm | 14:31 |
| jrosser | there is also a not-whole-disk image example in https://github.com/openstack/openstack-ansible-os_ironic/blob/master/doc/source/configure-ironic.rst | 14:32 |
| jrosser | but i did get in a mess with separate initird/root images and stuck with whole-disk in the end | 14:33 |
| jrosser | though it was a mess of my own creating, it was jsut more confusing | 14:34 |
| jamesdenton | right, ok. | 14:38 |
| jamesdenton | whole disk is what i've been doing, but running into this issue: Installing GRUB2 boot loader to device /dev/sda failed wi | 14:39 |
| jamesdenton | th Unexpected error while running command. | 14:39 |
| jrosser | well, getting the IPA log (and making sure it is in debug) was specatcularly useful | 14:39 |
| jamesdenton | i feel like i've run into that before and don't recall the fix | 14:39 |
| jrosser | kernel_append_params | 14:40 |
| jrosser | https://review.opendev.org/c/openstack/openstack-ansible-os_ironic/+/867547/11/doc/source/configure-ironic-debugging.rst#68 | 14:40 |
| jrosser | thats where to enable debugging even if you don't want the login thing | 14:40 |
| jamesdenton | thanks | 14:41 |
| jrosser | it was also super useful to change the node state to maintainance=on and the ironic state machine will freeze | 14:41 |
| jrosser | otherwise it gets in some bad state and gives up / powers off and you lose the ability to debug | 14:41 |
| jamesdenton | thanks, yeah maint mode trips me up sometimes | 14:53 |
| jamesdenton | me: "why does this server keep powering down" | 14:53 |
| mgariepy | jrosser, root:neutron for ssl cert won't work. it's 600 for the key so group neutron won't help for reading the file. | 14:53 |
| *** dviroel is now known as dviroel|lunch | 15:32 | |
| *** dviroel|lunch is now known as dviroel | 16:23 | |
| jamesdenton | jrosser user error; i had kernel and ramdisk assigned to my whole disk image :| | 16:32 |
| jrosser | jamesdenton: I think I managed to do the opposite | 16:32 |
| jrosser | set the image as a whole disk one but make whatever setting it is that has ironic think it’s kernel/ramdisk | 16:33 |
| jamesdenton | i have kernel and ramdisk assigned to the baremetal node for booting, then had made a whole disk image and mistakenly applied to the image, too. IPA did not want to write the bootloader. | 16:34 |
| jrosser | jamesdenton: when i got stuck with this i asked in #ironic and this ended up in the docs https://docs.openstack.org/ironic/latest/admin/troubleshooting.html#why-do-i-have-an-error-that-an-nvme-partition-is-not-a-block-device | 16:54 |
| jamesdenton | oh interesting, i read that, too | 16:55 |
| jamesdenton | things are working great now that i adjusted the image. just a dummy move | 16:56 |
| jrosser | it was suuuuper unclear what was going on | 16:56 |
| jrosser | and in my case `img_type` was set when it shouldnt have been | 16:56 |
| jamesdenton | i didnt set that, either | 16:56 |
| jrosser | there are many ways to shoot the foot :) | 16:56 |
| jrosser | noonedeadpunk: we can probably do like this patch for OSA https://review.opendev.org/c/openstack/ansible-collections-openstack/+/867202 | 17:02 |
| noonedeadpunk | oh, yes, would make sense | 17:03 |
| noonedeadpunk | But I don't think we're seing too much galaxy-related issues lately? | 17:04 |
| noonedeadpunk | It's surprisengly stable lately... | 17:04 |
| *** gmann is now known as gmann_afk | 17:29 | |
| noonedeadpunk | openssh_keypair module is sooooooooo weird..... | 17:36 |
| noonedeadpunk | chmod makes it fail with "Unable to read the key. The key is protected with a passphrase or broken. Will not proceed." | 17:36 |
| noonedeadpunk | more hilarious, if set mode to 0644 for module itself it also crashes on second run | 17:37 |
| noonedeadpunk | https://paste.openstack.org/show/bqBh6uvEUgffLhsazBR6/ | 17:37 |
| *** gmann_afk is now known as gmann | 17:41 | |
| *** gmann is now known as gmann_afk | 18:06 | |
| mgariepy | lol | 18:10 |
| * noonedeadpunk submitted bug report to collection | 18:13 | |
| mgariepy | close, works as design ? | 18:15 |
| mgariepy | :P | 18:15 |
| mgariepy | jrosser, have you seen the graph on this page? https://docs.openstack.org/networking-ovn/latest/admin/refarch/refarch.html | 18:18 |
| mgariepy | it does explains most of the inner working of ovn deploy. | 18:19 |
| mgariepy | how they are connected i mean and what part is responsible for doing what. | 18:20 |
| jrosser | that is very useful | 18:21 |
| mgariepy | just missing the dhcp ;p haha | 18:22 |
| jrosser | I was wondering if it was yet possible to have multiple active gateway nodes | 18:24 |
| jrosser | there was some priority iirc rather than sharing | 18:25 |
| jrosser | diagram also missing bgp agent for ipv6 | 18:26 |
| noonedeadpunk | fwiw networking-ovn was last updated in 2018 and was completely deprecated in April 2020 | 18:27 |
| noonedeadpunk | So no idea how this is relevant to be frank | 18:27 |
| jamesdenton | well, i think it was just rolled into neutron | 18:27 |
| noonedeadpunk | Yes, but also couple of changes landed to neutron as well regarding ovn | 18:27 |
| mgariepy | with the diag you see where each service takes it's info and where it runs | 18:28 |
| jamesdenton | https://docs.openstack.org/neutron/latest/admin/ovn/refarch/refarch.html | 18:28 |
| noonedeadpunk | Just saying that docs might be partially obsolete | 18:28 |
| jamesdenton | fixed. | 18:28 |
| noonedeadpunk | yeah. well, it's almost same | 18:30 |
| noonedeadpunk | so, I've missed - do we need neutron-dhcp-agent for ovn ?:) | 18:31 |
| noonedeadpunk | or https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870854 is good to go? | 18:31 |
| mgariepy | we don't need it for the vms. | 18:32 |
| mgariepy | the dhcp is handled ovn direclty | 18:32 |
| noonedeadpunk | what I don't like a lot, that setup_ovs_ovn.yml is not idempotent and raise planty of changed each time.... | 18:38 |
| noonedeadpunk | but we can iterate o nthat | 18:38 |
| jrosser | what is the branch name going to be for AA? | 18:52 |
| *** gmann_afk is now known as gmann | 18:56 | |
| noonedeadpunk | 2023.1 | 19:13 |
| noonedeadpunk | * stable/2023.1 I assume | 19:13 |
| noonedeadpunk | btw, that's the bug report with answer to my issue https://github.com/ansible-collections/community.crypto/issues/564 | 19:13 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-plugins master: Use cryptography backend for openssh_keypair https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/870997 | 19:27 |
| admin1 | has anyone experienced playbooks hanging/failing on os_keystone : Add service project | 19:31 |
| noonedeadpunk | well, saw that in AIO until we increased threads | 19:32 |
| noonedeadpunk | never outside of aio | 19:33 |
| admin1 | i am using a lab env to test something .. its a 12 threads, 32gb .. | 19:33 |
| admin1 | load is 0.38 | 19:33 |
| admin1 | its the controller itself | 19:33 |
| admin1 | not an AIO | 19:33 |
| noonedeadpunk | threads in terms of apache/uwsgi | 19:34 |
| admin1 | oh | 19:34 |
| noonedeadpunk | as mpm was prohibiting to spawn more | 19:34 |
| admin1 | curl http(s) endpoint:5000 replies fine | 19:34 |
| noonedeadpunk | well, I think it all depends how parallel connections it can handle | 19:35 |
| admin1 | how to increase the threads ? just to rule it out | 19:35 |
| noonedeadpunk | we did smth like that https://opendev.org/openstack/openstack-ansible/commit/078c82b03456d46641a3ec05e3d14bd3ac6d1cd5 | 19:35 |
| noonedeadpunk | But I assume that default values should be fine | 19:35 |
| jrosser | you could try similar from the utility container with the cli | 19:35 |
| jrosser | because (sort of) that’s what the ansible is doing | 19:36 |
| jrosser | maybe also creating the Keystone Service project is the first use of admin creds? there might be something wrong there | 19:37 |
| admin1 | changed the connections .. retrying | 19:40 |
| admin1 | set the therad count to 2 | 19:40 |
| admin1 | https://paste.openstack.org/show/bkZIDUbbPIPtcrwMEw2Y/ - i see gateway timeout .. but no idea what it means | 19:47 |
| admin1 | curl http://172.29.236.11:5000/ from util container responds just fine | 19:49 |
| noonedeadpunk | well, that might be aligned with amount of processes/threads | 19:56 |
| noonedeadpunk | as curl is jsut single request, but multiple at the same time might cause issues. But I'm not sure that's it | 19:56 |
| noonedeadpunk | As default is calculated based on cpu cores iirc | 19:56 |
| noonedeadpunk | so that's more then enough | 19:56 |
| mgariepy | is the openstack client works ? | 19:57 |
| admin1 | well, its the first playbook in a new install | 19:57 |
| noonedeadpunk | `keystoneauth1.exceptions.http.GatewayTimeout: Gateway Timeout (HTTP 504)` hm | 19:58 |
| admin1 | stuck at Making authentication request to http://172.29.236.11:5000/v3/auth/tokens => Resetting dropped connection: 172.29.236.11 | 19:58 |
| noonedeadpunk | what's in keystone / apapche logs? | 19:58 |
| noonedeadpunk | As that might be mysql connection issue actually | 19:58 |
| mgariepy | are all revelant endpoint up in haproxy ? | 19:59 |
| jrosser | http? | 20:00 |
| jrosser | dont you get 504 when a http thing tries to talk to an https endpoint? | 20:01 |
| jrosser | well anyway - i thought we default the internal vip to https | 20:02 |
| admin1 | https:// is only on the public side | 20:02 |
| admin1 | internally its all http only | 20:02 |
| admin1 | the openrc in the util has export OS_AUTH_URL=http://172.29.236.11:5000/v3 | 20:03 |
| admin1 | all endpoints are up | 20:03 |
| admin1 | i am going to set keystone_wsgi_threads: and keystone_wsgi_processes: to 12 ( number of cpu threads ) in user_variables, delete the keystone and repo container and retry again | 20:04 |
| admin1 | keystone service backend shows green in haproxy status gui | 20:05 |
| admin1 | and curl https:// on public and http on private endpoint on 5000 both give the reply | 20:05 |
| jrosser | is there no_log on the task? | 20:05 |
| admin1 | will enable log as well | 20:06 |
| admin1 | i have to test moving one of the cluster running on 24.2.4 from ceph running with ceph-ansible to ceph with cephadm, so was trying to recreate the same setup | 20:07 |
| admin1 | i think no_log is there by default | 20:07 |
| admin1 | the only change i have made is the rabbitmq version change | 20:07 |
| jrosser | you will have to comment out the no_log | 20:08 |
| jrosser | but i guess the auth parameters are all in openrc / clouds.yml anyway | 20:08 |
| admin1 | yeah .. | 20:09 |
| admin1 | git checkout the tag, ran setup hosts and infra .. validated all endpoints are up, mysql working from utility, utility can reach the endpoints .. then ran os-keystone setup and got stuck | 20:09 |
| jrosser | i think i might be checking that keystone service logs | 20:10 |
| admin1 | i already nuked the keystone container .and creating it again | 20:10 |
| jrosser | 504 means that haproxy waited too long for the thing behind it to reply | 20:10 |
| admin1 | this time, thread count is 12 | 20:10 |
| admin1 | i did not checked what it was by default though | 20:11 |
| admin1 | as i had never been stuck in this state before | 20:11 |
| mgariepy | jamesdenton, https://pasteboard.co/IYHYnNjokewN.png | 20:11 |
| mgariepy | what do you think about that ? | 20:11 |
| jamesdenton | big improvement | 20:11 |
| mgariepy | i can change the network connection to match the other png and add it to your patch. | 20:12 |
| *** prometheanfire is now known as Guest1769 | 20:12 | |
| mgariepy | does it reflect the correctly how it's wired up you think ? | 20:13 |
| mgariepy | it sure depend on a bunch of other stuff but. | 20:13 |
| mgariepy | any png paste site you guys prefer ? i don't paste png often. | 20:15 |
| jamesdenton | it looks good, except br-tun isn't really connected directly to a vlan, while br-ex is. So i'm not sure how to reflect that | 20:19 |
| mgariepy | yeah indeed. | 20:22 |
| jamesdenton | but i really dig the changes, thank you | 20:22 |
| jamesdenton | and i've never used a png share, so this looks fine | 20:22 |
| mgariepy | it's not a real netowrk it's just a bunch of flow in ovs lol | 20:22 |
| admin1 | ps ax | grep uwsgi | grep -v grep | wc -l => 14 threads running | 20:22 |
| admin1 | inside the keystone container | 20:23 |
| mgariepy | i'll redraw it like this one : https://review.opendev.org/c/openstack/openstack-ansible/+/867577/5/doc/source/reference/figures/networking-openvswitch-cn.drawio.png | 20:26 |
| admin1 | what you see is ( journalctl -f ) .. https://paste.openstack.org/show/bBzQwlnQaOUBlQgnklg1/ that the container suddenly restarts during os_keystone : Add service project | 20:34 |
| jrosser | admin1: the threads thing is a red herring, it was specific to CI jobs where we had deliberately turned the number right down, too much so it turned out | 20:34 |
| jamesdenton | mgariepy sure, that could work. we might even add comments to the page | 20:35 |
| admin1 | it happened again | 20:38 |
| jamesdenton | Well, gonna give this Yoga->Zed upgrade with OVN a go. Wish me luck | 20:47 |
| admin1 | best of luck .. | 20:47 |
| admin1 | yoga(ovs) -> zed (ovn ) ? | 20:48 |
| jamesdenton | ovn both ways | 20:48 |
| mgariepy | tell us how it goes. | 20:48 |
| jamesdenton | will do! | 20:48 |
| jrosser | jamesdenton: stable/zed roles still point to the point we branched | 20:50 |
| jrosser | in ansible-role-requirements | 20:50 |
| spatel | noonedeadpunk what do you use for snapshot of images, i meant regularly or schedulers? | 20:50 |
| jamesdenton | ok, so some backports may be missing? | 20:51 |
| spatel | I am looking at Freezer but not sure its best solution | 20:51 |
| jrosser | jamesdenton: you won't get anything merged to roles since the release | 20:51 |
| jamesdenton | good point, maybe i'll wait. | 20:51 |
| mgariepy | maybe wait the next point rel. | 20:52 |
| mgariepy | lol | 20:52 |
| jrosser | i think we are super close to making 26.1.0 | 20:52 |
| jrosser | +/- ovn stuff :) | 20:52 |
| jamesdenton | then i'll wait for that, thanks for the heads up | 20:52 |
| noonedeadpunk | spatel: we have in-house solution for that... But eventually it's freezer (which is barely alive) or mistral that can be leveraged for automation of snapshots | 20:53 |
| jrosser | i never understood why cinder-backup does not have any concept of schedule | 20:54 |
| spatel | Thanks for heads up on freezer :) | 20:54 |
| noonedeadpunk | yep, I'm quite eager to release 26.1.0 but each time I'm about to push bump, we find a bug | 20:54 |
| spatel | I will look into mistral | 20:54 |
| mnaser | jrosser: https://github.com/vexxhost/staffeln | 20:57 |
| jamesdenton | a wild mnaser appears | 20:57 |
| mnaser | i'm always lurkin | 20:57 |
| jamesdenton | good to see you | 20:57 |
| mnaser | but staffeln is an open source cinder-volume scheduler for lack of a better word | 20:57 |
| mnaser | jamesdenton: ty, you too, hope to see you in vancouver as well | 20:58 |
| jamesdenton | hope to be there! | 20:58 |
| mnaser | having network fun? | 20:58 |
| spatel | ovn fun :) | 20:58 |
| jrosser | mnaser: that is interesting - thanks | 21:00 |
| spatel | so it works with cinder-backup api | 21:00 |
| jrosser | with something automatic like that i would properly build something on nfs to sit next to ceph | 21:01 |
| jrosser | i have a small thing like that for cinder-backup now but hardly anyone uses it because its really not user friendly | 21:02 |
| jrosser | but if it were scheduled/automatic thats a whole different game | 21:02 |
| spatel | Wish there is a doc or example to set it up | 21:05 |
| noonedeadpunk | so basically most important part from the dead freezer | 21:06 |
| noonedeadpunk | which is cool actually | 21:07 |
| *** dviroel is now known as dviroel|afk | 21:09 | |
| jrosser | spatel: we make a role for it? :) | 21:10 |
| spatel | haha! indeed, first lets give it a shot and see if its going to fulfill our need | 21:11 |
| spatel | I will give it a try and see.. I am exploring solution for automatic backup of customer vm (i wish horizon has a option to schedule backup or similar) | 21:12 |
| spatel | Did you ever see this error in mysql - https://paste.opendev.org/show/bd6nfhdsFRfMbfol7BUK/ | 21:13 |
| spatel | This is my very old openstack running on Queens :D | 21:13 |
| jrosser | have you followed the "maintainance tasks" in the docs to check the cluster status | 21:14 |
| spatel | Every few days getting OOM error on mysql | 21:14 |
| spatel | Cluster is health (syncd + showing 3 nodes) | 21:14 |
| spatel | Google saying ignore that error but i think this is something serious. because only that node getting OOM. other two running fine. | 21:15 |
| spatel | server load is ok. | 21:16 |
| spatel | This is how it started - https://paste.opendev.org/show/bBEwJJiXiVHUIEILsi0y/ | 21:16 |
| spatel | Did anyone try ChatGPT ? | 21:21 |
| opendevreview | Marc Gariépy proposed openstack/openstack-ansible master: Update documentation for LXC/metal and LXB/OVS/OVN https://review.opendev.org/c/openstack/openstack-ansible/+/867577 | 21:23 |
| opendevreview | Marc Gariépy proposed openstack/openstack-ansible master: Update documentation for LXC/metal and LXB/OVS/OVN https://review.opendev.org/c/openstack/openstack-ansible/+/867577 | 21:24 |
| *** Guest1769 is now known as prometheanfire | 21:24 | |
| mgariepy | something like that that need a bit of your touch jamesdenton ;) | 21:24 |
| jamesdenton | lol | 21:25 |
| jamesdenton | i will take a look later today, thank you for doing that | 21:25 |
| mgariepy | you are welcome. | 21:25 |
| mgariepy | i just want to document it a bit for internal and internet uses ;D | 21:25 |
| mgariepy | maybe i try to put too much in the drawing.. | 21:26 |
| mgariepy | have a nice evening | 21:26 |
| jamesdenton | well, i think more detail is better here | 21:26 |
| jamesdenton | you too! | 21:26 |
| mgariepy | yeah more detail but the issue is that we might need to split it on a couple of drawing | 21:27 |
| jamesdenton | agreed | 21:28 |
| opendevreview | Merged openstack/openstack-ansible-os_neutron master: Disable dhcp-agent and metadata-agent for OVN https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870854 | 22:07 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Add facility to rewrite source URLs for ansible collections during bootstrap https://review.opendev.org/c/openstack/openstack-ansible/+/870820 | 22:07 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron stable/zed: Disable dhcp-agent and metadata-agent for OVN https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/870907 | 22:08 |
| prometheanfire | ^ had to do this manually on my install, I thought it was becaues ovs/linuxbridge was set up first | 22:27 |
| admin1 | changed from 24.4.2 -> 24.4.3 and it works ¯\_(ツ)_/¯ | 22:37 |
| admin1 | is there a easy way to enable mfa/2fa for keystone when using osa ? | 22:45 |
| jrosser | admin1: it is a post deployment step so not really in scope, but for which user? | 22:47 |
| admin1 | horizon | 22:48 |
| admin1 | i think i have to read more about it on | 22:49 |
| jrosser | horizon user? or user of horizon? | 22:49 |
| jrosser | the 2fa settings are per user in keystone, not really to do with horizon | 22:49 |
| jrosser | https://docs.openstack.org/keystone/latest/admin/multi-factor-authentication.html#multi-factor-authentication | 22:50 |
| admin1 | jrosser, many thanks | 22:56 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
