Feature/implementation support request: cluster graph integration

[ghost]

I deploy K8s/dashboard via kops, and I would like to integrate kubernetes-ui/graph into the dashboard. It would show up on the condition that cluster-insight is running.

Is there currently a plugin/component system for dashboard? Could someone who has architectural overview of dashboard help me make a rough outline of the work that has to be done to integrate the cluster graph into dashboard?

[bryk]

There's no extension mechanism at the moment. So this is all do-it-yourself. I think, though, this is the right moment to start thinking about the extensions framework for UI for real.

[bryk]

@floreks @maciaszczykm @cheld WDYT of doing this via one-off compile time hardcoding versus extensions framework?

[cheld]

The scope plugin uses an Iframe. #1949

The helm plugin we are thinking to integrate directly. Not sure about the best approach

In addition @floreks is thinking about a global settings page, backed by a config map to enable/disable plugins - like in chrome

[bryk]

Yeah. So there's a lot of these things appearing recently. I expect to see more. We could tackle this problem in a generic way to help everyone in one sweep. This is a large project, though.

[floreks]

It would be nice to have some extensions framework, but there are many things to consider. We need to discuss that first and think how to provide such a mechanism in a generic and dynamic way that will not require direct dashboard changes.

What I'm working on is not a plugin mechanism but integrations framework for our backend. This means that still we will need to introduce all integrations related code upstream. It will be just easier to integrate with dashboard.

[bryk]

Do you have any designs for the integrations framework already? Can we talk about this? :)

[floreks]

It's still an early sketch. I've been preparing other parts of backend for such framework. I want to minimize work needed to introduce new integrations later on so API still changes. Also I need to introduce some kind of storage provider with storage sync that will keep config up to date and integrate that with integrations manager.

There are many things to consider so I'm trying not to rush anything. It would be nice to discuss it. We can do that tomorrow.

[ghost]

Hi, just checking in on the extension framework/cluster graph integration, is there anything I could do to help further this integration?

@floreks, could you perhaps write a roadmap of what you think the issues to discuss are before implementing a extension framework?

[floreks]

First part can be seen in #2017 pull request. I still need to clean up some code but overall structure is already there. Right now I don't have clear roadmap. For sure after this part I will be working on some storage sync to be able to save per user config.

Unfortunately right now I don't see any tasks that could be done in parallel. Base part has to finished first and then new integrations i.e. prometheus for metrics can be added.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

Prevent issues from auto-closing with an /lifecycle frozen comment.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or @fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten
/remove-lifecycle stale

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close