Tabs that make calls to kubeflow API break when using SSL self-signed certs

[gustavosr98]

Bug Description

I enabled SSL for Kubeflow Dashboard with self-signed certs following this guide

The dashboard (partially), Jupyter notebooks and Dex work fine with SSL

Now, I have multiple tabs broken (eg. experiments, runs, etc) with the following message

Error: failed to retrieve list of pipelines. Click Details for more information.

Error occured while trying to proxy to: kubeflow.maas/apis/v2beta1/pipelines?page_token=&page_size=10&sort_by=created_at%20desc&filter=

Is there a step we might be missing to finish configuring SSL?
My assumption is that the pod that sends the request does not trust the SSL CA, how can I make it trust it?

To Reproduce

Deploy Kubeflow
Create a self-signed certificate
Follow this guide to enable SSL
Check the Experiments tab

Environment

Charmed Kubeflow 1.10/stable
Juju 3.6/stable (3.6.5)

Relevant Log Output

# juju debug-log --include kubeflow-dashboard/0 --replay
[..]
unit-kubeflow-dashboard-0: 19:04:00 WARNING unit.kubeflow-dashboard/0.juju-log The ``kubernetes_service_patch v1`` library is DEPRECATED and will be removed in October 2025. For patching the Kubernetes service created by Juju during the deployment of a charm, ``ops.Unit.set_ports`` functionality should be used instead.
unit-kubeflow-dashboard-0: 19:04:00 INFO unit.kubeflow-dashboard/0.juju-log HTTP Request: GET https://10.152.183.1/api/v1/namespaces/kubeflow/services/kubeflow-dashboard "HTTP/1.1 200 OK"
unit-kubeflow-dashboard-0: 19:04:00 INFO unit.kubeflow-dashboard/0.juju-log HTTP Request: PATCH https://10.152.183.1/api/v1/namespaces/kubeflow/services/kubeflow-dashboard "HTTP/1.1 200 OK"
unit-kubeflow-dashboard-0: 19:04:00 INFO unit.kubeflow-dashboard/0.juju-log Kubernetes service 'kubeflow-dashboard' patched successfully
unit-kubeflow-dashboard-0: 19:04:00 WARNING unit.kubeflow-dashboard/0.update-status ./src/charm.py:321: DeprecationWarning: Calling `ops.main.main()` is deprecated, call `ops.main()` instead
unit-kubeflow-dashboard-0: 19:04:00 WARNING unit.kubeflow-dashboard/0.update-status   main(KubeflowDashboardOperator)
unit-kubeflow-dashboard-0: 19:04:00 INFO juju.worker.uniter.operation ran "update-status" hook (via hook dispatching script: dispatch)
unit-kubeflow-dashboard-0: 19:14:17 ERROR juju.worker.dependency "log-sender" manifold worker returned unexpected error: sending log message: websocket: close 1005 (no status): websocket: close sent
unit-kubeflow-dashboard-0: 19:14:17 INFO unit.kubeflow-dashboard/0.juju-log HTTP Request: PATCH https://10.152.183.1/api/v1/namespaces/kubeflow/services/kubeflow-dashboard "HTTP/1.1 200 OK"
unit-kubeflow-dashboard-0: 19:14:17 INFO unit.kubeflow-dashboard/0.juju-log Kubernetes service 'kubeflow-dashboard' patched successfully
unit-kubeflow-dashboard-0: 19:14:17 WARNING unit.kubeflow-dashboard/0.update-status ./src/charm.py:321: DeprecationWarning: Calling `ops.main.main()` is deprecated, call `ops.main()` instead
unit-kubeflow-dashboard-0: 19:14:17 WARNING unit.kubeflow-dashboard/0.update-status   main(KubeflowDashboardOperator)
unit-kubeflow-dashboard-0: 19:14:17 INFO juju.worker.uniter.operation ran "update-status" hook (via hook dispatching script: dispatch)



# date
Fri Jun  6 09:10:04 PM UTC 2025
# kk logs kubeflow-dashboard-0 -c kubeflow-dashboard
2025-06-06T21:08:35.734Z [pebble] GET /v1/notices?after=2025-06-06T19%3A35%3A05.341395242Z&timeout=30s 30.001097476s 200
2025-06-06T21:09:05.736Z [pebble] GET /v1/notices?after=2025-06-06T19%3A35%3A05.341395242Z&timeout=30s 30.001018866s 200
2025-06-06T21:09:35.738Z [pebble] GET /v1/notices?after=2025-06-06T19%3A35%3A05.341395242Z&timeout=30s 30.001107568s 200

Additional Context

No response

[syncronize-issues-to-jira]

Thank you for reporting your feedback to us!

The internal ticket has been created: https://warthogs.atlassian.net/browse/KF-7493.

This message was autogenerated