"Clean leftover kubernetes loadbalancers" (#656) is a breaking change on our environment

[Kumm-Kai]

How to categorize this issue?
/area networking
/kind bug
/platform openstack

What happened:
Since the introduced changes in #656, infrastructure deletion isn't able to complete in our openstack environment.
It fails with the following error: No suitable endpoint could be found in the service catalog.

This is happening because our openstack environment doesn't provide endpoints for the octavia API, so it is more or less expected to fail.

What you expected to happen:
Successfull infrastructure deletion

How to reproduce it (as minimally and precisely as possible):
Good question. (Will followup if necessary)

Anything else we need to know?:
If you have a proposal for a "fix" I could implement it 🙂

Environment:

• Extension version: v1.36.0
• Cloud provider or hardware configuration: STACKIT

[kon-angelo]

Hello @Kumm-Kai. If I understand the issue, it is because we have "embedded" calls in our codebase that make assumptions about the LB provider (octavia in this case) and break compatibility with your environment (yawol).

[Kumm-Kai]

Hello @Kumm-Kai. If I understand the issue, it is because we have "embedded" calls in our codebase that make assumptions about the LB provider (octavia in this case) and break compatibility with your environment (yawol).

Partly correct.
Compatibility with our own LB (yawol) is working as expected (and is unrelated to this).
But due to the "Clean leftover kubernetes loadbalancers" code this breaks deletion in our openstack as it doesn't provide endpoints for the ocatvia API.

As said in the issue description, if you have an idea how this could be solved (without us removing the commit from our fork^^) I would be happy to implement it.