| @maloute:matrix.org | Ok, so it seems that it finally works, I had to:modify the populate_registry script to point it to registry.central:9001. Prior to that, I also needed to put the certificate of the docker registry in a specific directory (as it was a self signed certificate), and a docker login for authenticating to the docker directory | 07:58 |
|---|---|---|
| @maloute:matrix.org | * Ok, so it seems that it finally works, I had to: | 07:59 |
| - I needed to put the ca certificate of the docker registry in a specific directory (as it was a self signed certificate) | ||
| - modify the populate\_registry script to point it to registry.central:9001. | ||
| - launching a docker login for authenticating to the docker registry | ||
| - and lauching the script with a long list of docker images needed | ||
| @maloute:matrix.org | I'm not sure why I needed that, perhaps there is something I missed ? As I am not sure, I don't know if need to open a bug report... | 08:27 |
| @maloute:matrix.org | * I'm not sure why I needed that, perhaps there is something I missed ? As I am not sure, I don't know if I need to open a bug report... | 08:27 |
| @brunomuniz:matrix.org | It looks like something that needs to be better documented, maybe? | 11:59 |
| @brunomuniz:matrix.org | Both: | 12:00 |
| - the fact that you need to account for docker images when dealing with a subcloud; | ||
| - the docker images that the openstack application requires | ||
| @brunomuniz:matrix.org | Maybe? | 12:00 |
| @maloute:matrix.org | Well I've made an internal doc so I could open a bug report on the STX docs with the additions I made, and the STX team will decide what to do with that. | 12:04 |
| @brunomuniz:matrix.org | There's a huge task of bringing the virtual install guides up-to-date (they're outdated compared to the bare metal guides), which involves abstracting common things to a central location and using in both installation methods... so this is a good time if there's anything that needs fixing, I believe. | 12:53 |
| @brunomuniz:matrix.org | BTW, there's a Docs channel (https://matrix.to/#/#starlingx-docs:opendev.org)... but in this case it looks like that opening a bug report should be the way to go. | 12:54 |
| @maloute:matrix.org | Well I also found that the virtual install was not updated (made some bug report to enhance the docs) but in this case, I'm using 3 bare metal servers (2 AIO-DX for central cloud, and one AIO-SX for the subcloud) so that's why it seems strange. | 12:56 |
| @ldeataid:matrix.org | > <@maloute:matrix.org> I'm not sure why I needed that, perhaps there is something I missed ? As I am not sure, I don't know if I need to open a bug report... | 13:35 |
| STX-Openstack don't test the application on subcloud as we don't have the hardware for it, so stuff is probably outdated, but good to know that you got it working! | ||
| @ldeataid:matrix.org | > <@maloute:matrix.org> I'm not sure why I needed that, perhaps there is something I missed ? As I am not sure, I don't know if I need to open a bug report... | 13:35 |
| * STX-Openstack team don't test the application on subcloud as we don't have the hardware for it, so stuff is probably outdated, but good to know that you got it working! | ||
| @ldeataid:matrix.org | * STX-Openstack team don't test the application on subclouds (DC environment) as we don't have the hardware for it, so stuff is probably outdated, but good to know that you got it working! | 13:35 |
| @ldeataid:matrix.org | * STX-Openstack team don't test the application on subclouds (DC environment) as we don't have the hardware for it, so stuff is most likely outdated, but good to know that you got it working! | 13:36 |
| @ldeataid:matrix.org | If you do open a launchpad, let us know | 13:44 |
| @maloute:matrix.org | Lucas de Ataides: here you go: https://bugs.launchpad.net/starlingx/+bug/2039220 | 19:57 |
| @maloute:matrix.org | Sorry to bother you again. | 21:06 |
| On my AIO-SX server in a DC architecture, I'm facing an error when I'm creating a new instance on openstack, it fails with the following error: nova.exception.NoValidHost: No valid host was found. | ||
| I think My AIO-SX controller has the correct labels: | ||
| ``` | ||
| [sysadmin@controller-0 ~(keystone_admin)]$ system host-label-list controller-0 | ||
| +--------------+-------------------------+-------------+ | ||
| | hostname | label key | label value | | ||
| +--------------+-------------------------+-------------+ | ||
| | controller-0 | openstack-compute-node | enabled | | ||
| | controller-0 | openstack-control-plane | enabled | | ||
| | controller-0 | openvswitch | enabled | | ||
| +--------------+-------------------------+-------------+ | ||
| ``` | ||
| But when I'm view the openstack host list, I don't have a compute service: | ||
| ``` | ||
| [sysadmin@controller-0 ~(keystone\_admin)\]$ openstack host list | ||
| +---------------------------------+-----------+----------+ | ||
| | Host Name | Service | Zone | | ||
| +---------------------------------+-----------+----------+ | ||
| | nova-conductor-7b49c988cd-5s452 | conductor | internal | | ||
| | nova-scheduler-695fdbdf48-t87tw | scheduler | internal | | ||
| +---------------------------------+-----------+----------+ | ||
| ``` | ||
| And when i'm listing compute service, I see a status disabled and state down for the compute service: | ||
| ``` | ||
| \[sysadmin@controller-0 ~(keystone\_admin)\]$ openstack compute service list | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| | ID | Binary | Host | Zone | Status | State | Updated At | | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| | 2 | nova-conductor | nova-conductor-7b49c988cd-5s452 | internal | enabled | up | 2023-10-12T20:51:23.000000 | | ||
| | 3 | nova-scheduler | nova-scheduler-695fdbdf48-t87tw | internal | enabled | up | 2023-10-12T20:51:24.000000 | | ||
| | 5 | nova-compute | controller-0 | nova | disabled | down | 2023-10-12T20:51:32.000000 | | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| ``` | ||
| I tried to enable it with no success, the state is still down. I think it's because the host does not have the correct hostname, because my compute pod has this name: | ||
| ``` | ||
| \[sysadmin@controller-0 ~(keystone\_admin)\]$ kubectl get po -A | grep compute | ||
| openstack nova-compute-controller-0-937646f6-zzhdm 2/2 Running 8 (4h26m ago) `22h` | ||
| ``` | ||
| I'm not sure about my diagnostic and how to resolve this ... If someone has an idea 😀 | ||
| @maloute:matrix.org | * Sorry to bother you again. | 21:07 |
| On my AIO-SX server in a DC architecture, I'm facing an error when I'm creating a new instance on openstack, it fails with the following error: nova.exception.NoValidHost: No valid host was found. | ||
| I think My AIO-SX controller has the correct labels: | ||
| ``` | ||
| [sysadmin@controller-0 ~(keystone_admin)]$ system host-label-list controller-0 | ||
| +--------------+-------------------------+-------------+ | ||
| | hostname | label key | label value | | ||
| +--------------+-------------------------+-------------+ | ||
| | controller-0 | openstack-compute-node | enabled | | ||
| | controller-0 | openstack-control-plane | enabled | | ||
| | controller-0 | openvswitch | enabled | | ||
| +--------------+-------------------------+-------------+ | ||
| ``` | ||
| But when I'm view the openstack host list, I don't have a compute service: | ||
| ``` | ||
| [sysadmin@controller-0 ~(keystone\_admin)\]$ openstack host list | ||
| +---------------------------------+-----------+----------+ | ||
| | Host Name | Service | Zone | | ||
| +---------------------------------+-----------+----------+ | ||
| | nova-conductor-7b49c988cd-5s452 | conductor | internal | | ||
| | nova-scheduler-695fdbdf48-t87tw | scheduler | internal | | ||
| +---------------------------------+-----------+----------+ | ||
| ``` | ||
| And when i'm listing compute service, I see a status disabled and state down for the compute service: | ||
| ``` | ||
| \[sysadmin@controller-0 ~(keystone\_admin)\]$ openstack compute service list | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| | ID | Binary | Host | Zone | Status | State | Updated At | | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| | 2 | nova-conductor | nova-conductor-7b49c988cd-5s452 | internal | enabled | up | 2023-10-12T20:51:23.000000 | | ||
| | 3 | nova-scheduler | nova-scheduler-695fdbdf48-t87tw | internal | enabled | up | 2023-10-12T20:51:24.000000 | | ||
| | 5 | nova-compute | controller-0 | nova | disabled | down | 2023-10-12T20:51:32.000000 | | ||
| +----+----------------+---------------------------------+----------+----------+-------+----------------------------+ | ||
| ``` | ||
| I tried to enable it with no success, the state is still down. I think it's because the host does not have the correct hostname, because my compute pod has this name: | ||
| ``` | ||
| \[sysadmin@controller-0 ~(keystone\_admin)\]$ kubectl get po -A | grep compute | ||
| openstack nova-compute-controller-0-937646f6-zzhdm 2/2 Running 8 (4h26m ago) `22h` | ||
| ``` | ||
| I'm not sure about my diagnostic and how to resolve this ... If someone has an idea, thank you ! 😀 | ||
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!