| *** markvoelker has quit IRC | 00:02 | |
| *** cshen has joined #openstack-ansible | 00:46 | |
| *** cshen has quit IRC | 00:50 | |
| *** cshen has joined #openstack-ansible | 00:59 | |
| *** Berthe01 has quit IRC | 01:02 | |
| *** cshen has quit IRC | 01:04 | |
| *** mgariepy has quit IRC | 01:29 | |
| *** mgariepy has joined #openstack-ansible | 01:43 | |
| *** gyee has quit IRC | 02:04 | |
| *** mubix has quit IRC | 02:36 | |
| *** vblando has quit IRC | 02:36 | |
| *** Open10K8S has quit IRC | 02:37 | |
| *** Open10K8S has joined #openstack-ansible | 02:37 | |
| *** mubix has joined #openstack-ansible | 02:39 | |
| *** vblando has joined #openstack-ansible | 02:40 | |
| *** udesale has joined #openstack-ansible | 03:52 | |
| *** markvoelker has joined #openstack-ansible | 04:23 | |
| *** markvoelker has quit IRC | 04:28 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-ansible | 04:33 | |
| *** rh-jelabarre has quit IRC | 04:50 | |
| *** luksky has joined #openstack-ansible | 06:00 | |
| *** miloa has joined #openstack-ansible | 06:05 | |
| *** NewJorg has quit IRC | 06:19 | |
| *** markvoelker has joined #openstack-ansible | 06:24 | |
| *** NewJorg has joined #openstack-ansible | 06:25 | |
| *** markvoelker has quit IRC | 06:29 | |
| *** cshen has joined #openstack-ansible | 06:54 | |
| *** yolanda has joined #openstack-ansible | 07:26 | |
| *** sshnaidm|afk is now known as sshnaidm|ruck | 07:28 | |
| *** tosky has joined #openstack-ansible | 07:28 | |
| *** stingrayza has joined #openstack-ansible | 07:35 | |
| *** also_stingrayza has quit IRC | 07:38 | |
| admin0 | morning | 07:43 |
|---|---|---|
| admin0 | in one platform, still on rocky, neutron plugin base has -vpnaas listed.. but i don't see the vpn button in horizon .. | 07:46 |
| admin0 | and nothing failed in the deployment | 07:47 |
| admin0 | 18.1.8 tag | 07:47 |
| *** arkan has joined #openstack-ansible | 08:06 | |
| arkan | hi guys, I have my running openstack, everything was ok. I wanted to install Load Balancer from this page https://docs.openstack.org/openstack-ansible/12.2.6/install-guide/configure-lbaas.html | 08:11 |
| arkan | so I added the code for v2, and now in Horizon my vms disappears | 08:12 |
| arkan | and some warning messages appeared | 08:12 |
| arkan | I'm using openstack-ansible "train" | 08:13 |
| *** arkan has quit IRC | 08:15 | |
| *** arkan_ has joined #openstack-ansible | 08:16 | |
| arkan_ | and in "/project/" url it says "Something went wrong!, unexpected error has occurred. Try refreshing the page. If that doesn't help, contact your local administrator." | 08:16 |
| arkan_ | when I ran the os-neutron-install.yml I did not have errors | 08:17 |
| arkan_ | I also tried to run os-horizon-install.yml after I ran os-neutron-install.yml and also I did not have any errors | 08:18 |
| CeeMac | admin0: did you also add 'horizon_enable_neutron_vpnaas: True' to user_variables? | 08:25 |
| *** markvoelker has joined #openstack-ansible | 08:25 | |
| CeeMac | i think there are some kernel specific modules to be added too iirc | 08:26 |
| CeeMac | i started looking at this in my test cluster but didn't get around to validating it by setting up an actual VPN | 08:26 |
| *** markvoelker has quit IRC | 08:30 | |
| *** ioni has quit IRC | 08:38 | |
| *** masterpe has quit IRC | 08:38 | |
| arkan_ | I removed the code from https://docs.openstack.org/openstack-ansible/12.2.6/install-guide/configure-lbaas.html for load balancer v2 and re-run os-neutron-install.yml and now my vms appeared | 08:40 |
| arkan_ | I will try to install octavia https://docs.openstack.org/openstack-ansible-os_octavia/train/configure-octavia.html | 08:41 |
| *** namrata has joined #openstack-ansible | 08:43 | |
| *** masterpe has joined #openstack-ansible | 08:47 | |
| masterpe | Last week we did a upgrade from rocky to stein to OSA version 19.1.0, now when we deploy a neutron router, it sometimes does not get it configured competently. | 08:47 |
| jrosser | arkan_: that is a very old set of documenation | 08:47 |
| arkan_ | jrosser: I did not know about it https://docs.openstack.org/openstack-ansible/12.2.6/install-guide/configure-lbaas.html | 08:48 |
| jrosser | arkan_: that is from when the lbaas was part of neutron, which does not exist any more. these days you should be looking at the octavia service | 08:48 |
| masterpe | Is there a known bug that causes this? | 08:48 |
| arkan_ | jrosser: is this ok https://docs.openstack.org/openstack-ansible-os_octavia/train/configure-octavia.html ? | 08:48 |
| *** arkan_ is now known as arkan | 08:49 | |
| jrosser | arkan_: yes and there are 3 parts there, you need a provider network for octavia, the service to be setup, and then you need the loadbalancer VM image (amphora) | 08:50 |
| jrosser | so theres a few more moving parts than just deploying the API endpoints | 08:50 |
| arkan | jrosser: is the page sufficient to follow, or should I read other docs, related to installing the vm image (amphora) | 08:51 |
| arkan | jrosser: setting the provider net is easy | 08:51 |
| jrosser | the OSA page is not instructions for octavia really | 08:52 |
| jrosser | the os_octavia role can deal with it all for you i think | 08:52 |
| *** ioni has joined #openstack-ansible | 08:52 | |
| jrosser | if you want to understand a working deployment then there is an 'octavia' scenario in the OSA all-in-one | 08:53 |
| arkan | jrosser: I mean the default values in the os-role is sufficient | 08:53 |
| arkan | jrosser: ok I will check | 08:53 |
| arkan | jrosser: thank you as always | 08:53 |
| jrosser | you have the choice of os_octaiva doing the provider network and amphora image for you, or you can make that setup yourself | 08:54 |
| jrosser | depends what is appropriate in your environment | 08:54 |
| arkan | jrosser: I want to choose os_octavia to do it for me | 08:57 |
| arkan | I can provide in openstack_user_config.yml the provider network for LB | 08:57 |
| arkan | jrosser: so If I provide the provider network for octavia and run os_octavia, is this sufficient ? | 08:59 |
| jrosser | the network config is here https://github.com/openstack/openstack-ansible-os_octavia/blob/master/defaults/main.yml#L326-L354 | 08:59 |
| jrosser | and it will use br-lbaas to connect to octavia containers on the infra hosts so you'll need to create those | 09:00 |
| jrosser | and its up to you however you want vlan or flat or other networking type on the network/compute nodes | 09:00 |
| arkan | jrosser: I will create "- network: " (I will use vlan) section in openstack_user_config.yml and set the needed values from what you've given me, and run os-octavia ansible role | 09:02 |
| namrata | Hi, I am at stable/train branch and doing a fresh installation and and when I run os-placement-install.yml my installation fails on Task Create database of serviceTASK [os_placement : Create database for service] ********************************************************************************************************************fatal: | 09:06 |
| namrata | [alice_placement_container-d59c2be8]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'None' has no attribute 'name'\n\nThe error appears to be in '/etc/ansible/roles/os_placement/tasks/db_setup.yml': line 28, column 7, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe | 09:06 |
| namrata | offending line appears to be:\n\n block:\n - name: Create database for service\n ^ here\n"} | 09:06 |
| *** sshnaidm|ruck has quit IRC | 09:14 | |
| *** sshnaidm has joined #openstack-ansible | 09:23 | |
| *** cshen has quit IRC | 09:29 | |
| *** cshen has joined #openstack-ansible | 09:30 | |
| admin0 | CeeMac, i think i have not done that :) .. will do and check | 09:36 |
| *** sshnaidm has quit IRC | 09:42 | |
| arkan | jrosser: I set the required vars, now I want to add octavia in the inventory in order to run os-octavia | 09:42 |
| arkan | jrosser: is it ok to run openstack-ansible setup-infrastructure.yml --syntax-check ? | 09:43 |
| CeeMac | admin0: good luck :) | 09:43 |
| CeeMac | you'll need to run the horizon playbook again after setting the variable of course | 09:43 |
| arkan | jrosser: If I run os-octavia it will result "skipping: no hosts matched", because it is not in the inventory | 09:44 |
| admin0 | CeeMac, does it appear under neutron_plugin_base: -vpnaas .. | 09:44 |
| admin0 | or it should be - neutron_vpnaas | 09:45 |
| CeeMac | - vpnaas' | 09:46 |
| admin0 | main.yml says # Please add the 'vpnaas' to the neutron_plugin_base list .. so i think i did it correctly .. ran the neutron and the horizon .. still don't see the vpn under network | 09:46 |
| CeeMac | yes, plugin list looks fine. and you added the horizon variable too? | 09:47 |
| admin0 | yep | 09:47 |
| admin0 | horizon_enable_neutron_vpnaas: True | 09:47 |
| CeeMac | do you have anything set for openstack_host_specific_kernel_modules: | 09:48 |
| CeeMac | just wondering if its failed on a dependency | 09:48 |
| admin0 | if I do this: grep -ri horizon_enable_neutron /etc/ansible/roles/* . i do not see that variable anywhere | 09:49 |
| admin0 | all i see is _fwaas and _lbaas | 09:49 |
| admin0 | grep -ri horizon_enable /etc/ansible/roles/* | grep vpn -- no hits | 09:49 |
| CeeMac | admin0: https://pasted.tech/pastes/2cba21450a3d416e2f92736a088722b534294fae | 09:50 |
| CeeMac | is what I had set | 09:50 |
| CeeMac | let me dig back through my notes a second | 09:51 |
| CeeMac | I was using this as a reference : https://docs.openstack.org/openstack-ansible-os_neutron/rocky/configure-network-services.html | 09:52 |
| admin0 | i was in the same page :) | 09:52 |
| CeeMac | step 3 has the kernel overrides :) | 09:52 |
| admin0 | ERROR! the playbook: openstack_hosts-config could not be found .. wrong tags i guess | 09:54 |
| *** sshnaidm has joined #openstack-ansible | 09:55 | |
| CeeMac | ah | 09:57 |
| CeeMac | one minute | 09:57 |
| *** markvoelker has joined #openstack-ansible | 09:58 | |
| CeeMac | found my docs, there is some work to do, just need to dive into a meeting then will pull together something for you admin0 | 10:00 |
| admin0 | thanks CeeMac .. will wait | 10:01 |
| *** markvoelker has quit IRC | 10:03 | |
| *** mgariepy has quit IRC | 10:04 | |
| *** mugsie has quit IRC | 10:07 | |
| *** mugsie has joined #openstack-ansible | 10:10 | |
| CeeMac | admin0: https://pasted.tech/pastes/09e022aef5eea50ad7f25e5217f96b6b895641b9 | 10:23 |
| CeeMac | pretty sure thats a condensed list of what I had to do to get it installed | 10:23 |
| CeeMac | actually you can skip the other neutron settings at the top there, I should have pruned them out | 10:24 |
| CeeMac | https://pasted.tech/pastes/2881be560e445801598cd14b7c831b8a4a06c210 | 10:25 |
| CeeMac | is better | 10:25 |
| admin0 | CeeMac, thanks .. i will follow this | 10:38 |
| admin0 | quick quesiton .. is this workaround only for rocky .... if i want to do vpn for the latest one, is this the same workaround ? | 10:39 |
| CeeMac | admin0: I think maybe just rocky | 10:54 |
| CeeMac | I haven't tried any newer install yet | 10:55 |
| CeeMac | Also not sure of the neutron xaas plugins now, they lost a lot of support on them | 10:55 |
| *** udesale_ has joined #openstack-ansible | 11:04 | |
| *** udesale has quit IRC | 11:06 | |
| *** NewJorg has quit IRC | 11:08 | |
| *** NewJorg has joined #openstack-ansible | 11:14 | |
| namrata | Can somebody help me with this error while installation of os_placement-install.yml I am seeing | 11:20 |
| namrata | http://paste.openstack.org/show/795347/ | 11:20 |
| namrata | i am working on stable/train branch | 11:21 |
| *** markvoelker has joined #openstack-ansible | 11:55 | |
| *** gregwork has quit IRC | 12:00 | |
| *** rh-jelabarre has joined #openstack-ansible | 12:06 | |
| arkan | jrosser: I updated my inventory to include octavia container host, then I ran openstack-ansible setup-hosts.yml but it failed, I log the errors, and I can see it's using openvswitch | 12:09 |
| arkan | openstack_user_config.yml: http://paste.openstack.org/show/795349/ | 12:10 |
| arkan | user_variables.yml: http://paste.openstack.org/show/795350/ | 12:10 |
| arkan | error log: http://paste.openstack.org/show/795348/ | 12:10 |
| *** mgariepy has joined #openstack-ansible | 12:15 | |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible master: Add integrated test for Ubuntu Bionic using ML2/OVN driver https://review.opendev.org/733017 | 12:25 |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible master: Add integrated test for Ubuntu using ML2/OVN driver https://review.opendev.org/733017 | 12:26 |
| jrosser | arkan: i dont think it is using openvswitch. did you create br-lbaas? | 12:32 |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible master: Add integrated test for Ubuntu using ML2/OVN driver https://review.opendev.org/733017 | 12:32 |
| arkan | jrosser: I added it in the openstack_user_config.yml, I will use the same method and use br-ext as I've done it for the external network provider | 12:33 |
| jrosser | arkan: you know that you have to create any bridges that you specify in openstack_user_config? | 12:34 |
| arkan | jrosser: yes, and I think I made a mistake because br-lbass is not created | 12:35 |
| arkan | jsrosser: I will create it now | 12:35 |
| jrosser | ok | 12:35 |
| mgariepy | anything that needs review ? | 12:35 |
| jrosser | mgariepy: train is broken somehow, i don't know where we are with centos8 really | 12:36 |
| arkan | jrosser: I will create the bridge only on the machines that use octavia specified in my openstack_user_config.yml (my controller node), and I will leave the other node (compute node) without creating the bridge | 12:39 |
| namrata | jrosser mgariepy can you help me with the failed installation of os-placement-install.yml this is the error:http://paste.openstack.org/show/795347/ | 12:41 |
| jrosser | namrata: if you comment out the no_log here https://github.com/openstack/openstack-ansible-os_placement/blob/master/tasks/db_setup.yml#L32 | 12:44 |
| jrosser | and rerun with -vv then you should see which variable is empty | 12:44 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - lxd containers https://review.opendev.org/738599 | 12:51 |
| *** namrata has quit IRC | 12:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Remove support for Ubuntu 16.04 https://review.opendev.org/737978 | 13:00 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Do not install python2 or its dev package on Ubuntu Bionic hosts https://review.opendev.org/737979 | 13:00 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-lxc_hosts stable/ussuri: Do not install python2 or its dev package on Ubuntu Bionic hosts https://review.opendev.org/738607 | 13:02 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-lxc_hosts stable/ussuri: Remove support for Ubuntu 16.04 https://review.opendev.org/738608 | 13:05 |
| *** namrata has joined #openstack-ansible | 13:28 | |
| arkan | jrosser: to include octavia | 13:42 |
| arkan | 1. Add bridge br-lbaas (ok) | 13:42 |
| arkan | 2. openstack-ansible setup-hosts.yml (success) | 13:42 |
| arkan | 3. openstack-ansible os-octavia-install.yml (failed) | 13:42 |
| arkan | 4. openstack-ansble os-neutron-install.yml (success) | 13:42 |
| arkan | 5. openstack-ansible os-octavia-install.yml (success) | 13:42 |
| arkan | The problem was after adding 'lbaas' it could not be found by octavia install, so I need to update | 13:42 |
| arkan | neutron plugin which is /etc/neutron/plugins/ml2# vi ml2_conf.ini by running os-neutron-install.yml | 13:42 |
| arkan | And after running it (os-neutron-install.yml), the file updated to includ lbaas vlan | 13:42 |
| arkan | [ml2_type_vlan] | 13:42 |
| arkan | network_vlan_ranges = provider:40:40,lbaas:60:60 | 13:42 |
| ioni | you only need to create br-lbaas on controller | 13:44 |
| jrosser | if you edit the neutron ml2_conf.ini your changes will be removed next time you run the os_neutron playbook | 13:44 |
| *** tacco has joined #openstack-ansible | 13:45 | |
| ioni | you want lbaas network to be flat and vlan ? | 13:45 |
| arkan | jrosser: this is what I've done as I mentioned in step nr. 4 | 13:45 |
| tacco | hi there.. anyone knows where the volumes_attached from "openstack server show " comes from? | 13:45 |
| arkan | ioni: it's vlan with id 60 | 13:45 |
| ioni | what type of network do you have now into the cloud | 13:45 |
| tacco | because i have a volume that still apear there.. but is marked as deleted in the database and does not exists anymore | 13:45 |
| arkan | ioni: I can see in the 'Admin' section there is something new added | 13:46 |
| arkan | Project: service, Network Name: br-lbaas, br-lbaas-subnet 192.168.60.0/24 | 13:47 |
| arkan | ioni: DHCP Agents: 0, Shared: No, External: No, Status: Active, Admin State: UP, Availability Zones: - | 13:48 |
| ioni | you do not need br-lbass on compute nodes | 13:48 |
| ioni | the idea of br-lbaas is to have containers with ips from the same network/subnet | 13:48 |
| ioni | containers ip(api octavia) needs to talk with amphora instances that have ips from that subnet | 13:49 |
| arkan | ioni: In my openstack_user_config.yml, I included octavia only to use Infra host not compute nodes | 13:49 |
| arkan | ioni: also compute node does not have this bridge | 13:49 |
| arkan | ioni: it's only available on the controller node | 13:50 |
| jrosser | arkan: octavia created virtual machines that run the loadbalancers | 13:50 |
| ioni | br-lbaas configuration https://paste.xinu.at/uTm/ | 13:50 |
| jrosser | the octavia service is on the infra nodes, the lxc on those nodes connects to br-lbaas | 13:50 |
| ioni | in your case is tagged 60 not 669 like in my configuration | 13:50 |
| jrosser | br-lbaas must be connected to the provider network that goes to the compute nodes so that the octavia service can commicate with the loadbalancer VM | 13:51 |
| arkan | ioni: yes and I'm not using bond, and it's only on my controller node (my infra host) | 13:51 |
| jrosser | arkan: if you have used the values from os_octavia defaults/main.yml without overriding them, it is set up for a flat network, not vlan | 13:52 |
| jrosser | https://github.com/openstack/openstack-ansible-os_octavia/blob/master/defaults/main.yml#L331 | 13:52 |
| arkan | I've used mine | 13:52 |
| arkan | I overridden some of them | 13:52 |
| jrosser | i'm not sure you need to be adjusting ml2.conf them | 13:52 |
| jrosser | *then | 13:52 |
| jrosser | it's just another vlan tag on your existing vlan type provider network? | 13:53 |
| ioni | arkan: https://paste.xinu.at/G2Attd/ | 13:56 |
| *** spatel has joined #openstack-ansible | 14:24 | |
| namrata | jrosser the error of placement_api install after -vv is http://paste.openstack.org/show/795361/ | 14:28 |
| arkan | jrosser, ioni: sorry I was outside with the workers | 14:32 |
| jrosser | arkan: you have an example from ioni now of a proper config for using the existing vlan provider network | 14:33 |
| arkan | ioni: openstack_user_config.yml http://paste.openstack.org/show/795363/ | 14:33 |
| *** cshen has quit IRC | 14:33 | |
| arkan | ioni: user_variables.yml http://paste.openstack.org/show/795364/ | 14:34 |
| arkan | ioni: thanks I will use your example | 14:36 |
| arkan | but I will change vlan seg: to be 60, yours is 669 | 14:36 |
| *** miloa has quit IRC | 14:37 | |
| arkan | also the subnet, 192.168.60.0/24 instead of 10.100.103.0/24 | 14:37 |
| namrata | jrosser http://paste.openstack.org/show/795361/ is this a bug | 14:59 |
| jrosser | namrata: i don't know :) you will need to determine which variable is undefined using normal ansible debugging techniqes | 15:03 |
| jrosser | but we run that code many many times in CI so it does work with a standard AIO config | 15:03 |
| namrata | @jrosser I removed no_log and tried with -vvv | 15:04 |
| namrata | I am not able to find the variable | 15:04 |
| noonedeadpunk | namrata: maybe you have _oslodb_databases overriden somewhere? | 15:06 |
| *** andrea15 has quit IRC | 15:07 | |
| spatel | namrata: if you run with -vvvv might give you some hints | 15:07 |
| namrata | noonedeadpunk this is a fresh installation I haven;t made any changes | 15:08 |
| namrata | spatel the error log which I pasted is with -vvv only | 15:08 |
| jrosser | namrata: everyone is trying to show you the different things available to help debug | 15:09 |
| noonedeadpunk | so obviously it fails here https://opendev.org/openstack/openstack-ansible-os_placement/src/branch/master/tasks/db_setup.yml#L30 | 15:09 |
| jrosser | if there isnt much useful info in the output then you should increase the debug/verbose level and try again | 15:09 |
| noonedeadpunk | but for placement we have name explicitly set https://opendev.org/openstack/openstack-ansible-os_placement/src/branch/master/tasks/main.yml#L50 | 15:10 |
| *** mgariepy has quit IRC | 15:34 | |
| *** gyee has joined #openstack-ansible | 15:38 | |
| arkan | Ioni: I've done your setup for octavia, just some modifications for vlan seg. http://paste.openstack.org/show/795375/ | 15:44 |
| arkan | then I've run os-neutron-install, os-octavia-install and os-horizon-install | 15:45 |
| arkan | now I can see the load balancer in my horizon | 15:45 |
| arkan | but it is giving Alerts: "Error: Unable to retrieve load balancers." | 15:46 |
| arkan | ioni: and this is the added part in the network section http://paste.openstack.org/show/795376/ | 15:49 |
| *** sshnaidm has quit IRC | 15:51 | |
| *** sshnaidm has joined #openstack-ansible | 15:53 | |
| arkan | [wsgi:error] [pid 18497:tid 139770147088128] [remote 172.29.236.10:50092] urllib3.exceptions. | 15:54 |
| arkan | MaxRetryError: HTTPConnectionPool(host='172.29.236.210', port=9876): Max retries exceeded with url: /v2.0/lbaas/loadbalancers?project_id=d491172334d84e1c8ad5d231b1935e20 (Caused by NewConn | 15:54 |
| arkan | ectionError('<urllib3.connection.HTTPConnection object at 0x7f1eaa57a3c8>: Failed to establish a new connection: [Errno 111] Connection refused',)) | 15:54 |
| arkan | also | 16:00 |
| arkan | [remote 172.29.236.10:54524] No policy rules for | 16:00 |
| arkan | service 'load-balancer' in /openstack/venvs/horizon-20.1.1/lib/python3.6/site-packages/openstack_dashboard/conf/octavia_policy.yaml | 16:00 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 16:03 |
| openstack | Meeting started Tue Jun 30 16:03:45 2020 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:03 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:03 |
| *** openstack changes topic to " (Meeting topic: openstack_ansible_meeting)" | 16:03 | |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:03 |
| noonedeadpunk | #topic office hours | 16:03 |
| *** openstack changes topic to "office hours (Meeting topic: openstack_ansible_meeting)" | 16:03 | |
| noonedeadpunk | \o/ | 16:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-plugins master: WIP add lxd support https://review.opendev.org/738657 | 16:04 |
| noonedeadpunk | so | 16:05 |
| noonedeadpunk | have we backported everything we wanted regarding focal? | 16:05 |
| noonedeadpunk | jrosser: btw regarding this... I was thinking if we should just rename containers in dynamic_inventory? | 16:07 |
| jrosser | o/ hello | 16:07 |
| jrosser | oh you mean _ vs - ? | 16:08 |
| noonedeadpunk | yeah | 16:08 |
| jrosser | oh yes there is probably something nice we need to do there | 16:08 |
| jrosser | i sort of put it to one side because theres a bigger piece of work to do with the group names too | 16:08 |
| jrosser | i did a reasonable look across the master/ussuri branches for stuff to backport for focal | 16:10 |
| jrosser | it looked in reasonable shape | 16:10 |
| noonedeadpunk | I also had a brief look and didn't find anything at glance | 16:10 |
| noonedeadpunk | ok, cool, so will place a bump today | 16:11 |
| noonedeadpunk | a nasty thing is that train bump seems broken... | 16:11 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/735469 | 16:11 |
| *** mgariepy has joined #openstack-ansible | 16:12 | |
| noonedeadpunk | maybe it's because of some ceph-ansible update... | 16:12 |
| jrosser | yes i was looking at that a bit | 16:15 |
| jrosser | and i was also thinking that there was something not right with radosgw authentication | 16:16 |
| noonedeadpunk | I thougnth it might be swift tempest plugin, but it seems that's a part of tempest core. | 16:16 |
| jrosser | there is ceph 14.2.10 i think | 16:16 |
| jrosser | thats a very recent release which is what could have broken | 16:16 |
| jrosser | theres an error in the tempest log about 404 instead of 401, then you can see the auth trail looking all reasonable through radosgw logs and keystone log | 16:17 |
| jrosser | and then i didnt really know what was breaking | 16:17 |
| *** sshnaidm is now known as sshnaidm|ruck | 16:18 | |
| noonedeadpunk | Like interesting part that tempest was expecting unathorized anyway | 16:18 |
| noonedeadpunk | but got 404 instead. | 16:19 |
| noonedeadpunk | so this thing wasn't expected to finish with 200 or smth | 16:19 |
| noonedeadpunk | so yeah, probably ceph has changed smth in terms of return codes and now it's not fully compatible with swift in terms of return codes... | 16:20 |
| jrosser | there was an interesting string in one of the log messages - don't have that open right now | 16:23 |
| jrosser | and it took me to a RH bug about ResellerAdmin needing to be in the radosgw config | 16:24 |
| noonedeadpunk | so here where it fails https://opendev.org/openstack/tempest/src/branch/master/tempest/scenario/test_object_storage_basic_ops.py#L65 | 16:26 |
| *** mgariepy has quit IRC | 16:27 | |
| noonedeadpunk | so we're checking that things fail here eventually | 16:27 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - lxd containers https://review.opendev.org/738599 | 16:28 |
| jrosser | and that code is not touched in several years | 16:29 |
| jrosser | i don't know if it's possible to pin ceph back to the previous point release and see what happens | 16:30 |
| noonedeadpunk | yeah, so I think it's ceph who just returns 404 instead of 401 for unathorized | 16:30 |
| *** cshen has joined #openstack-ansible | 16:30 | |
| noonedeadpunk | btw I guess that's not happening for O | 16:31 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - lxd containers https://review.opendev.org/738599 | 16:32 |
| jrosser | right, which is wierd, as you'd expect all the changes on N to be backports from O | 16:32 |
| *** cshen has quit IRC | 16:35 | |
| noonedeadpunk | I don't see anything obvious here https://github.com/ceph/ceph/compare/v14.2.9...v14.2.10 | 16:38 |
| *** udesale_ has quit IRC | 16:39 | |
| *** mgariepy has joined #openstack-ansible | 16:41 | |
| noonedeadpunk | also interesting if old bump is going to pass now https://review.opendev.org/#/c/735469/ | 16:42 |
| *** jbadiapa has quit IRC | 16:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Do not install development packages on the target https://review.opendev.org/731459 | 16:52 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_rally stable/ussuri: Do not install development packages on the target https://review.opendev.org/738676 | 16:52 |
| noonedeadpunk | #endmeeting | 17:06 |
| *** openstack changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: https://bit.ly/2SAcGAn" | 17:06 | |
| openstack | Meeting ended Tue Jun 30 17:06:01 2020 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:06 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-06-30-16.03.html | 17:06 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-06-30-16.03.txt | 17:06 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-06-30-16.03.log.html | 17:06 |
| ioni | arkan: run the haproxy install playbook, pretty sure it doesn't have an entry point for octavia service | 17:22 |
| *** namrata has quit IRC | 17:36 | |
| arkan | ioni: ok | 17:36 |
| *** cshen has joined #openstack-ansible | 17:43 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - lxd containers https://review.opendev.org/738599 | 17:45 |
| arkan | ioni: now it has, but horizon it is not working as expected, some pages give 503 Service Unavailable | 17:48 |
| arkan | ioni: in my controller node with journalctl I'm seeing this: "2020-06-30 17:52:20.697 568784 ERROR neutron.plugins.ml2.drivers.linuxbridge.agent.linuxbridge_neutron_agent [-] Parsing physical_interface_mappings failed: Value eno1 in mapping: 'provider:eno1' not unique. Agent terminated!" | 17:53 |
| arkan | ioni: this is my current file in neutron container http://paste.openstack.org/show/795395/ | 17:55 |
| jamesdenton | what does linuxbridge_agent.ini look like | 17:56 |
| arkan | ioni: path /etc/neutron/plugins/ml2/ml2_conf.ini | 17:56 |
| arkan | jamesdenton: where to find this file ? in which container ? | 17:57 |
| arkan | jamesdenton: in ls /etc/neutron | 17:59 |
| jamesdenton | yes | 17:59 |
| arkan | only these are : | 17:59 |
| arkan | api-paste.ini api-paste.ini.original neutron.conf plugins rootwrap.conf rootwrap.conf.original rootwrap.d | 17:59 |
| jamesdenton | in plugins/ml2 | 18:00 |
| arkan | ls /etc/neutron/plugins/ml2/ | 18:00 |
| arkan | ml2_conf.ini | 18:00 |
| arkan | there is no such file | 18:00 |
| jamesdenton | is this the neutron-server container? or a compute? | 18:00 |
| arkan | it's a container | 18:01 |
| arkan | root@controller1-neutron-server-container-24d28d27 | 18:01 |
| arkan | in the controller node | 18:01 |
| jamesdenton | ok, so look for the file where you got that log message from | 18:01 |
| jamesdenton | probably on the controller itself, not inside the neutron-server container | 18:02 |
| arkan | jamesdenton: that message was in the journalctl from the controller node (not container) | 18:02 |
| arkan | jamesdenton: yes it is there | 18:03 |
| arkan | in the controller node (not container) | 18:03 |
| arkan | I will paste it | 18:03 |
| jamesdenton | yep | 18:03 |
| arkan | http://paste.openstack.org/show/795397/ | 18:04 |
| arkan | I can see there are 2 mappings | 18:04 |
| arkan | this is because of my - network section in my openstack_user_config.yml | 18:04 |
| arkan | the provider network is using it | 18:05 |
| arkan | and lbaas is also using it | 18:05 |
| jamesdenton | yes, you won't be able to use eno1 for lbaas. but you could use eno1.100, if vlan 100 was configured | 18:05 |
| arkan | as Host_bind_override: "eno1" | 18:05 |
| arkan | I made eno1.60 | 18:05 |
| jamesdenton | you can configure, say eno1.100, map it to lbaas, and then the corresponding neutron neutron for lbaas mgmt would be a 'flat' network | 18:05 |
| arkan | it's also configured in the router | 18:06 |
| jamesdenton | for lbaas? | 18:06 |
| arkan | yes | 18:06 |
| arkan | but I need to update my openstack_user_config | 18:06 |
| jamesdenton | ok, so change your mapping to eno1.60 on the openstack_user_config.yml | 18:06 |
| arkan | jamesdenton: this is my current config http://paste.openstack.org/show/795398/ | 18:07 |
| jamesdenton | change host_bind_override: "eno1" to host_bind_override: "eno1.60" for the lbaas section | 18:07 |
| arkan | yes | 18:08 |
| arkan | after that what to run ? | 18:08 |
| arkan | os-neutron-install.yml | 18:08 |
| arkan | os-octavia-install.yml | 18:08 |
| arkan | os-horizon-install.yml | 18:08 |
| arkan | should I run also haproxy-install.yml? | 18:09 |
| jamesdenton | os-neutron-install | 18:20 |
| arkan | jamesdenton: os-neutron-install run successfully | 18:20 |
| jamesdenton | check your agent log now | 18:20 |
| arkan | I'm running now os-octavia-install | 18:20 |
| arkan | ok | 18:20 |
| arkan | journalctl -xf | 18:21 |
| arkan | there are no errors | 18:21 |
| -openstackstatus- NOTICE: Due to a flood of connections from random prefixes, we have temporarily blocked all AS4837 (China Unicom) source addresses from access to the Git service at opendev.org while we investigate further options. | 18:21 | |
| arkan | jamesdenton: I've finished installing os-neutron, os-octavia, haproxy-install, os-horizon | 18:27 |
| arkan | now some of the working pages are not working | 18:27 |
| arkan | like floating ips | 18:28 |
| arkan | ok, now it worked | 18:28 |
| arkan | maybe it needed time to work | 18:28 |
| jamesdenton | yes, it can take a little bit of time to be ready | 18:29 |
| arkan | the Load Balancer page, gave 1 warning "Warning: Policy check failed." | 18:29 |
| arkan | and "Error: Unable to retrieve load balancers." | 18:29 |
| jamesdenton | there may be a hint in octavia logs | 18:29 |
| arkan | I will check | 18:29 |
| jamesdenton | you might also try cli | 18:29 |
| arkan | I saw in journalctl one big red error | 18:30 |
| admin0 | does osa work on ubuntu 20.04 ? | 18:30 |
| arkan | http://paste.openstack.org/show/795399/ | 18:31 |
| arkan | jamesdenton: this is a repeated error (periodic) | 18:32 |
| arkan | jamesdenton: do you have some cli commands to get more info about why there is error ? | 18:33 |
| arkan | I'm inside the container controller1_octavia_server_container-2e9b4f31 | 18:34 |
| jamesdenton | not really, no. can you try to spin up a vm normally? | 18:34 |
| arkan | I will check | 18:34 |
| arkan | jamesdenton: hummm now I can see strange message in the Launch Instance dialog box | 18:35 |
| arkan | in the Source* section | 18:36 |
| arkan | "There are no allowed boot sources. If you think this is wrong please contact your administrator." | 18:36 |
| jamesdenton | see if you can upload an image, and create a flavor | 18:36 |
| arkan | jamesdenton: my openstack was fully functional, just today I wanted to install os-octavia | 18:36 |
| arkan | after I tried it, then I got sort of errors | 18:37 |
| jamesdenton | what did that process look like? Sorry, i think missed that here | 18:37 |
| arkan | jamesdenton: no even I can not create flavors | 18:38 |
| jamesdenton | are you familiar w/ the openstack cli? can you try there? | 18:38 |
| arkan | jamesdenton: yes | 18:38 |
| arkan | I had created 7 vms with terraform, and other vms | 18:39 |
| arkan | playing around | 18:39 |
| arkan | also teleport for teleporting with ssh between the machines | 18:40 |
| arkan | and bastion vm | 18:40 |
| arkan | all were working | 18:40 |
| arkan | today I needed to install octavia | 18:40 |
| arkan | because I need a load balancer | 18:40 |
| arkan | I also installed kubernetes | 18:40 |
| arkan | I don't know why it's giving errors after installing os-octavia | 18:41 |
| jamesdenton | ultimately, what did the process look like for installing octavia? updating openstack_user_config.yml and maybe user_overrides.yml and then installing which playbooks? | 18:41 |
| jamesdenton | i assume setup-hosts and setup-infra were run, too? to create the octavia api container? | 18:41 |
| arkan | jamesdenton: on moment I will explain you | 18:42 |
| arkan | 1. I've added the bridge | 18:43 |
| arkan | http://paste.openstack.org/show/795401/ | 18:43 |
| arkan | 2. I have the variables into user_variables.yml http://paste.openstack.org/show/795403/ | 18:44 |
| arkan | (section octavia above) | 18:44 |
| arkan | 3. I added the cidr, network section, and octavia-infra_hosts http://paste.openstack.org/show/795404/ | 18:46 |
| arkan | 4. I ran inventory/dyanmic_inventory.py | 18:46 |
| arkan | 5. I ran openstack-ansible setup-hosts.yml | 18:47 |
| arkan | then I tried to run other os-* like os-neutron, os-octavia, os-horizon, os-keyston, haproxy-install | 18:48 |
| arkan | jamesdenton: I did not run setup-infrastructure.yml | 18:49 |
| arkan | jamesdenton: should I run setup-infrastructure.yml ? | 18:50 |
| jamesdenton | no, its ok | 18:50 |
| jamesdenton | there is some mismatch here for the octavia bits, around vlan tagging and whatnot. i need find a working example | 18:50 |
| jamesdenton | i'm just a bit tied up at the moment | 18:50 |
| arkan | ok | 18:50 |
| arkan | jamesdenton: ioni gave me this one https://paste.xinu.at/G2Attd/ and I adapted it to my situation | 18:52 |
| jrosser | arkan: i think you are getting several different sets if info telling you different things, and your config is now mixed up | 18:53 |
| arkan | jrosser: I did not copy paste the values, I adapted it in my current setup, like I've changed vlan, cidr, network name, ...etc | 18:54 |
| jrosser | right, but now you have eno1.60 and type=vlan, thats not right because it's a mix of flat and vlan | 18:55 |
| arkan | but I don't know why octavia is not satisfied today | 18:55 |
| jamesdenton | did ioni happen to provide the openstack_user_config.yml provider network definition, too? | 18:55 |
| arkan | I don't have flat | 18:55 |
| jrosser | yes | 18:55 |
| arkan | jrosser: I did not use flat in my setup | 18:56 |
| jrosser | if you don't want to use a flat network then don't specify eno1.60 becasue that is a de-tagged vlan 60 -> flat | 18:56 |
| jrosser | and previously the issue was that you had eno1 twice in the neutron config | 18:57 |
| jrosser | imho this can be simpler | 18:57 |
| jrosser | just one vlan provider network, that you already have | 18:57 |
| jrosser | and make the *neutron* lbaas network be a vlan with segmentation_id 60 | 18:58 |
| arkan | jrosser: hummm this is strange eno1.60 about flat, look into my current netplan file http://paste.openstack.org/show/795405/ | 18:58 |
| noonedeadpunk | jrosser: yeah old train bump also fails with 14.2.10 ceph... | 18:58 |
| jrosser | arkan: you have host_bind_override: "eno1.60" | 18:59 |
| arkan | all the vlan section are used through sub interface eno1.10, eno1.20 ...etc and they are tagged, and communicate through the router | 18:59 |
| jrosser | that is the neutron pyhsical network config | 18:59 |
| arkan | jrosser: so that one "host_bind_override" is for flat ? | 18:59 |
| jrosser | no, it's a pyhsical interface for neutron | 19:00 |
| arkan | ok | 19:00 |
| arkan | I used it for the provider network and I specified "eno1" which is working | 19:00 |
| *** cshen has quit IRC | 19:00 | |
| arkan | but I don't know what to do now for octavia lbaas | 19:01 |
| arkan | I have only one cable NIC to the router | 19:01 |
| *** sshnaidm|ruck is now known as sshnaidm|bbl | 19:02 | |
| jrosser | thats fine | 19:02 |
| arkan | jrosser: my physical network in my controller is "eno1" | 19:02 |
| noonedeadpunk | and we have no option to use 14.2.9 at the moment:( | 19:03 |
| arkan | this is where the cable is attached | 19:03 |
| *** Julos20 has joined #openstack-ansible | 19:03 | |
| jrosser | arkan: on the controller you have br-lbaas which is hooked to vlan 60 through your netplan config | 19:03 |
| *** cshen has joined #openstack-ansible | 19:03 | |
| arkan | yes | 19:03 |
| jrosser | the octavia containers are connected to that becasue of *part* of the config in openstack_user_config | 19:04 |
| arkan | yes | 19:04 |
| jrosser | container_interface: "eth14" for example, makes eth14 in the octavia container get connected to br-lbaas | 19:04 |
| ebbex | noonedeadpunk: https://review.opendev.org/#/c/673778/ should I be seeing two "admin" projects in my project list, and three domains (Default, heat, magnum) on a fresh train install? | 19:04 |
| *** yolanda has quit IRC | 19:05 | |
| jrosser | then quite separately neutron needs to know how to connect octavia VM to the lbaas network, i.e vlan 60 | 19:05 |
| jrosser | and there are many ways to do that depending on your setup, which is why there is no "correct" answer for this | 19:05 |
| jrosser | arkan: do you have the same eno1 on your compute node? | 19:06 |
| arkan | jrosser: so container (eth14) <--------> br-lbaas bridge | 19:06 |
| arkan | jrosser: yes | 19:06 |
| arkan | I mean compute node is another physical host (separate host) | 19:07 |
| arkan | also it has one NIC which is named "eno1" | 19:07 |
| jrosser | ok, so you already have a config in neutron for the vlan provider network | 19:07 |
| arkan | the provider network is running perfectly | 19:07 |
| arkan | no doubt about it | 19:07 |
| jrosser | but you can't have a second config for it | 19:08 |
| arkan | i can even create floating ips | 19:08 |
| arkan | and access the machines | 19:08 |
| arkan | the problem is happened today when I tried to install octavia by adding vlan 60 | 19:08 |
| jrosser | i think you need to remove some of the config from the openstack_user_config for the lbaas network | 19:08 |
| jrosser | so that it does not try to make a second neutron physical network mapping to eno1, becasue you already have one | 19:09 |
| arkan | what string to use for host_bind_override: ? | 19:09 |
| jrosser | that field is optional | 19:09 |
| arkan | yeah ? | 19:09 |
| arkan | good | 19:10 |
| arkan | then I can delete it | 19:10 |
| arkan | so now we will have container eth14 <--------connected to ------> br-lbaas | 19:10 |
| arkan | and br-lbaas is inside the controller node (infra) | 19:11 |
| jrosser | there is an annotated example file https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.example#L153-L162 | 19:11 |
| arkan | great this is optional | 19:12 |
| jrosser | arkan: and you should also have eth14 <----> controller br-lbaas <----> compute node br-ext(vlan tag 60) | 19:12 |
| arkan | jrosser: aha | 19:13 |
| arkan | jrosser: I though that I needed only for controller node | 19:13 |
| arkan | ok, I will create br-lbaas in compute node also | 19:14 |
| jrosser | no no :) | 19:14 |
| jrosser | br-lbaas is only on the controller | 19:14 |
| arkan | jrosser: aha gr8 | 19:15 |
| jrosser | when octavia creates a amphora VM (i think) there is a network ID in the config file (i forget exactly what) | 19:15 |
| jrosser | neutron needs to know what to connect the vm to | 19:15 |
| jrosser | so in your case the neutron network/subnet would be on physical network name "provider" and segmentation id 60 | 19:16 |
| arkan | yes | 19:16 |
| arkan | no | 19:16 |
| arkan | my working OS is using "provider" network with vlan seg. 40 | 19:17 |
| arkan | Network Type: vlan | 19:18 |
| arkan | Physical Network: provider | 19:18 |
| arkan | Segmentation ID: 40 | 19:18 |
| arkan | Name | 19:18 |
| arkan | provider | 19:18 |
| arkan | ID | 19:18 |
| arkan | cbc85df0-b906-4912-9b2a-bfb7dc60ccc7 | 19:18 |
| arkan | this is the one that I create floating ips from | 19:18 |
| jrosser | that is a logical network created in neutron called "provider" | 19:19 |
| jrosser | and it's using a physical network also (confusingly) called "provider" | 19:19 |
| jrosser | you can have another logical network on a different vlan tag on the same neturon physical network | 19:19 |
| arkan | and I have it in my netplan | 19:20 |
| arkan | and it's configured in my router also as tagged vlan 40 | 19:20 |
| arkan | jrosser: and how to do it ? :) | 19:21 |
| noonedeadpunk | ebbex: 2 admins yes, but they're gonna be in diferent domains. https://docs.openstack.org/heat/pike/install/install-ubuntu.html | 19:21 |
| noonedeadpunk | as well as 2 domains | 19:22 |
| noonedeadpunk | cant instantly recall about magnum, but probably yes | 19:22 |
| *** viks____ has quit IRC | 19:22 | |
| jrosser | arkan: you allow vlan 60 on whatever trunk interface between the compute and controller | 19:22 |
| noonedeadpunk | but I believe there's a bug somewhere, as some permissions are missing | 19:22 |
| noonedeadpunk | so like http://paste.openstack.org/show/794358/ fixes them (thanks to dpaclt) | 19:22 |
| jrosser | then you make sure that the neutron octavia network is using physical network "provider" and segmentation id 60 | 19:22 |
| jrosser | and you make sure that network is referenced in the octavia config file | 19:23 |
| noonedeadpunk | But I didn't manage to verify which exact is missing and why | 19:23 |
| arkan | jrosser: my question is now I've deleted "host_bind_override" for lbaas and ran os-neutron | 19:24 |
| noonedeadpunk | btw we can totally simplify that now https://review.opendev.org/#/c/673778/7/tasks/heat_service_setup.yml | 19:24 |
| arkan | can I keep my old setup and use lbaas without using host_bind_override ? | 19:24 |
| arkan | because now I saw that if I have 2 places in my openstack_user_config that use host_bind_override with the same value, then I will see errors | 19:26 |
| jrosser | hopefully i've just shown you that you can remove one of those | 19:27 |
| arkan | and now I know that host_bind_override is a physical interface (the NIC where the cable is attached) | 19:27 |
| jrosser | that ends up in the physical network mappings in the neutron config | 19:28 |
| arkan | jrosser: yes and it can not have values to the same value like provider:eno1 and lbaas:eno1, this will give errors | 19:29 |
| ebbex | noonedeadpunk: Thanks, I'll look into these docs some more :) Never actually managed to get magnum to work, and never seen two "admin" projects before train (messed up some scripts i had adding stuff to the now conflictingly named "admin" project), but the domain stuff seems legit. | 19:29 |
| noonedeadpunk | it's really pretty weird comparing to other project and totally not readable in terms of that patch... | 19:30 |
| noonedeadpunk | we can improve somehow though... | 19:30 |
| noonedeadpunk | but really worth checking these permissions as dpaclt claimed they fixed heat for him | 19:31 |
| noonedeadpunk | magnum from the other side is pretty tough itself and you'll need to pickup it's specific version to make it work... | 19:31 |
| noonedeadpunk | (you can probably ask guilhermesp about it) | 19:32 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/738357 | 19:33 |
| noonedeadpunk | jrosser: I've just blacklisted this swift test as for now... ^ | 19:33 |
| *** cshen has quit IRC | 19:42 | |
| arkan | jrosser: I caught this error on controller node, I think I need to create the bridge also on compute node | 19:43 |
| arkan | on host compute1 for vnic_type normal using segments [{'id': 'cac5b1a5-b0e9-4883-9a27-fe301bce6c0c', 'network_type': 'vlan', 'physical_network': 'lbaas', 'segmentation_id': 60, 'network_id': 'a681a436-36ea-44de-b1b8-672c45d936a0'}] | 19:43 |
| jrosser | 'physical_network': 'lbaas', | 19:44 |
| jrosser | that doesnt look right according to what you pasted before | 19:44 |
| arkan | yes, why it is physical ? | 19:44 |
| arkan | now my config is modified | 19:44 |
| arkan | and I ran os-neutron, os-octavia, haproxy-install and os-horizon | 19:45 |
| arkan | I've deleted the line host_bind_override: "eno1" | 19:45 |
| arkan | so why it's saying "physical_network" | 19:46 |
| jrosser | there is a logical network defined in neutron | 19:47 |
| arkan | and also it's mentioning the compute node | 19:47 |
| jrosser | openstack network list | 19:47 |
| jrosser | openstack network show <id-of-octavia-network> | 19:47 |
| jrosser | you need to get that correct as well | 19:47 |
| arkan | jrosser: my openstack now got sick, now it is not working | 19:47 |
| arkan | it was working before running os-octavia | 19:48 |
| arkan | now it throws some bunch of error with ImportError: No module named queue | 19:48 |
| arkan | in my horizon, I can see lbaas-mgmt | 19:49 |
| arkan | hummm | 19:49 |
| arkan | Network Type: vlan | 19:49 |
| arkan | Physical Network: lbaas | 19:49 |
| arkan | Segmentation ID: 60 | 19:49 |
| jrosser | yes this is wrong | 19:49 |
| arkan | it's there with physical network | 19:49 |
| arkan | I will delete it | 19:49 |
| jrosser | you don't have a pysical network called lbaas | 19:50 |
| jrosser | override the variables in os_octavia defaults/main.yml to represent the configuration you have | 19:50 |
| arkan | well my openstack horizon, is destroyed now | 19:51 |
| arkan | :))) | 19:51 |
| arkan | a big black circle | 19:51 |
| arkan | ok, now it got logged in again after some refreshings | 19:52 |
| arkan | I overrided with these | 19:54 |
| arkan | octavia_enable_anti_affinity: True | 19:54 |
| arkan | octavia_legacy_policy: True | 19:54 |
| arkan | octavia_provider_network_name: lbaas | 19:54 |
| arkan | octavia_provider_network_type: vlan | 19:54 |
| arkan | octavia_provider_segmentation_id: 60 | 19:54 |
| arkan | octavia_container_interface: "eth14" | 19:54 |
| arkan | octavia_container_network_name: "lbaas_address" | 19:54 |
| arkan | octavia_management_net_dhcp: "False" | 19:54 |
| arkan | octavia_management_net_subnet_cidr: 192.168.60.0/24 | 19:54 |
| arkan | octavia_management_net_subnet_allocation_pools: "192.168.60.20-192.168.60.200" | 19:54 |
| *** dave-mccowan has quit IRC | 19:55 | |
| arkan | is there a magic value that I need to use ? :) | 19:56 |
| arkan | on octavia container I caught an error | 19:57 |
| arkan | 2020-06-30 19:57:17.991 18300 ERROR octavia.controller.worker.v1.controller_worker [-] Failed to create an amphora due to: Failed to build compute instance due to: Network a681a436-36ea-44de-b1b8-672c45d936a0 could not be found. (HTTP 400) (Request-ID: req-2467c352-69fe-4a80-bac9-0e163f471e8d): octavia.common.exceptions.ComputeBuildException: Failed to build compute instance due to: Network a681a436-36ea-44de-b1b8-672c45d936a0 could not | 19:58 |
| arkan | be found. (HTTP 400) (Request-ID: req-2467c352-69fe-4a80-bac9-0e163f471e8d) | 19:58 |
| arkan | because I've deleted it the network | 19:58 |
| *** dave-mccowan has joined #openstack-ansible | 19:58 | |
| arkan | now octavia container is crying | 19:59 |
| arkan | if I will give it the network, then I will cry | 19:59 |
| *** noonedeadpunk has quit IRC | 20:02 | |
| guilhermesp | yeah i mean.. i'd say working with magnum is an adventure :P | 20:03 |
| guilhermesp | we used to be pointing to specific shas for whatever version mangum works | 20:03 |
| guilhermesp | what we should keep in mind is that we need to use coreos images ( can generate them here https://github.com/stackhpc/magnum-terraform/blob/master/upload-coreos.sh ) | 20:04 |
| *** yolanda has joined #openstack-ansible | 20:11 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/ussuri: Bump SHAs for stable/ussuri https://review.opendev.org/738689 | 20:19 |
| *** sshnaidm|bbl has quit IRC | 20:34 | |
| *** sshnaidm|bbl has joined #openstack-ansible | 20:35 | |
| *** cshen has joined #openstack-ansible | 20:35 | |
| *** sshnaidm|bbl is now known as sshnaidm|afk | 20:46 | |
| *** jbadiapa has joined #openstack-ansible | 20:54 | |
| *** markvoelker has quit IRC | 21:32 | |
| *** rh-jelabarre has quit IRC | 21:37 | |
| *** luksky has quit IRC | 21:55 | |
| *** spatel has quit IRC | 21:57 | |
| *** Berthe01 has joined #openstack-ansible | 21:57 | |
| *** Julos20 has quit IRC | 22:00 | |
| openstackgerrit | Erik Berg proposed openstack/openstack-ansible-os_heat master: add description to domain during creation https://review.opendev.org/738712 | 22:33 |
| *** tosky has quit IRC | 22:57 | |
| *** markvoelker has joined #openstack-ansible | 23:33 | |
| *** markvoelker has quit IRC | 23:38 | |
| *** markvoelker has joined #openstack-ansible | 23:48 | |
| *** cshen has quit IRC | 23:53 | |
| *** markvoelker has quit IRC | 23:53 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!