[Starlingx-discuss] Stx-Openstack Install

Andrew Bays andrew.bays at gmail.com
Thu Aug 13 18:50:45 UTC 2020


Greetings,

I hope you all are well.  I have two virtualized Starling-X distributed
cloud deployments (both which consist of 2 VMs for AIO HA central cloud and
2 VMs for AIO HA subcloud).  One is version 3.0 and the other is 4.0.  Both
were created/installed following the official documentation:

-
https://docs.starlingx.io/deploy_install_guides/r3_release/distributed_cloud/index.html
-
https://docs.starlingx.io/deploy_install_guides/r3_release/openstack/install.html
-
https://docs.starlingx.io/deploy_install_guides/r4_release/distributed_cloud/index.html
-
https://docs.starlingx.io/deploy_install_guides/r4_release/openstack/install.html

My 3.0 deployment has both k8s and Openstack running on the central and
subcloud clouds.  I didn't have any luck with stx-openstack version 1.0-9
"latest" stable at first, so I tried "versioned"
<http://mirror.starlingx.cengn.ca/mirror/starlingx/release/3.0.0/centos/outputs/helm-charts/stx-openstack-1.0-19-centos-stable-versioned.tgz>,
which ended up working for me.  However, once I set my subcloud to
"managed" in the central cloud with dcmanager, I never saw the Openstack
identity sync status reach "in-sync", even after waiting a long time:

[sysadmin at controller-0 ~(keystone_admin)]$ dcmanager subcloud show subcloud2
> +--------------------------------+----------------------------+
> | Field                          | Value                      |
> +--------------------------------+----------------------------+
> | id                             | 8                          |
> | name                           | subcloud2                  |
> ...
> | software_version               | 19.12                      |
> | management                     | managed                    |
> | availability                   | online                     |
> | deploy_status                  | complete                   |
>
...
> | identity_sync_status           | in-sync                    |
> | identity_openstack_sync_status | unknown                    |
> | patching_sync_status           | in-sync                    |
> | platform_sync_status           | in-sync                    |
> +--------------------------------+----------------------------+
>

Could anyone provide guidance on where to look in terms of pods, services,
logs, etc, such that I might find a clue as to what the problem might be?
Am I possibly missing configuration or some other required step to enable
the sync (beyond just setting the subcloud to "managed")?

For my 4.0 distributed cloud, I can't even get stx-openstack 1-0.49 (
"latest"
<http://mirror.starlingx.cengn.ca/mirror/starlingx/release/4.0.1/centos/flock/outputs/helm-charts/stx-openstack-1.0-49-centos-stable-latest.tgz>
or "versioned"
<http://mirror.starlingx.cengn.ca/mirror/starlingx/release/4.0.1/centos/flock/outputs/helm-charts/stx-openstack-1.0-49-centos-stable-versioned.tgz>)
to install.  The keystone-api-proxy pod fails with the following error in
its logs:

$ kubectl logs pods/keystone-api-proxy-f
> + dcorch-api-proxy --config-file=/etc/dcorch/dcorch.conf --type identity
> Traceback (most recent call last):
>   File "/var/lib/openstack/bin/dcorch-api-proxy", line 5, in <module>
>     from dcorch.cmd.api_proxy import main
>   File
> "/var/lib/openstack/lib/python2.7/site-packages/dcorch/cmd/api_proxy.py",
> line 42, in <module>
>     from dcorch.api.proxy.common import utils
>   File
> "/var/lib/openstack/lib/python2.7/site-packages/dcorch/api/proxy/common/utils.py",
> line 26, in <module>
>     from dccommon.drivers.openstack import sdk_platform as sdk
>   File
> "/var/lib/openstack/lib/python2.7/site-packages/dccommon/drivers/openstack/sdk_platform.py",
> line 26, in <module>
>     from dccommon.drivers.openstack.barbican import BarbicanClient
>   File
> "/var/lib/openstack/lib/python2.7/site-packages/dccommon/drivers/openstack/barbican.py",
> line 23, in <module>
>     from barbicanclient import client
> *ImportError: No module named barbicanclient*
>

Has anyone else encountered this issue or know how to fix/work-around it?
It seems like some dependency is missing from the container image, but I'm
not sure.

Any insights/feedback on these two issues would be greatly appreciated.
Thank you!

Regards,
Andrew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20200813/0aae1654/attachment-0001.html>


More information about the Starlingx-discuss mailing list