Skip to content

Bump up RHDH to 1.9 and RHDH chart to 5.0.2#78

Open
thepetk wants to merge 5 commits intoredhat-ai-dev:mainfrom
thepetk:bump-up-rhdh-chart
Open

Bump up RHDH to 1.9 and RHDH chart to 5.0.2#78
thepetk wants to merge 5 commits intoredhat-ai-dev:mainfrom
thepetk:bump-up-rhdh-chart

Conversation

@thepetk
Copy link
Collaborator

@thepetk thepetk commented Feb 4, 2026

What does this PR do?

The PR bumps up the RHDH version to 1.9 and the RHDH Chart version to 5.0.2:

  • Version 5.0.2 has been selected looking at the rhdh-chart chart.yaml for release-1.9 branch.
  • A refinement (also grouping) of all plugins has been done.
  • We have removed the plugin suffix as it's no longer required. For example the oci://ghcr.io/redhat-developer/rhdh-plugin-export-overlays/red-hat-developer-hub-backstage-plugin-lightspeed-backend:bs_1.45.3__1.1.1!red-hat-developer-hub-backstage-plugin-lightspeed-backend is now oci://ghcr.io/redhat-developer/rhdh-plugin-export-overlays/red-hat-developer-hub-backstage-plugin-lightspeed-backend:bs_1.45.3__1.1.1
  • In 5.0.2 the nameOverride for the backstage deployment is changed to developer-hub. In order to keep the same deployment resource I'm keeping the basktage one. This will just rollout a new update of the existing deployment on openshift -> instead of replacing with a new one.
  • A tmp volume has been created as the scaffolder actions where complaining upon testing

Which issue(s) does this PR fix

Fixes RHIDP-11152 & RHDHBUGS-2502

How to test changes / Special notes to the reviewer

I have tested the suggested changes on a test cluster of mine thoroughly:

  • Lightspeed works as expected
  • Software templates can be deployed and tekton pipelines work fine
  • Header and Homepage are aligned with the vanilla one. The only change is on the application-launcher dropdown

I have also tested the exact deployment scenario -> meaning I have created an instance on RHDH 1.8 with RHDH Chart 3.1.1 and then went up to RHDH 1.9 & RHDH-chart 5.0.2

The changes in the plugins are based on:

@thepetk
Copy link
Collaborator Author

thepetk commented Feb 4, 2026

Replaces #77

@thepetk
Copy link
Collaborator Author

thepetk commented Feb 4, 2026

Sharing here screenshots from the tests:

  • OCP Deployment successful:
image
  • Global header and Homepage aligned to vanilla except the custom dropdown
image
  • Using RHDH 1.9 and Backstage version 1.45.3
image
  • Created software template app and pipelines work as expected:
image
  • Lightpseed works as expected
image

Copy link
Contributor

@gabemontero gabemontero left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

any chance of pointing your test env to the dev cluster's RHOAI endpoints and seeing if the model catalog bridge / openshift ai connector works @thepetk ?

@thepetk
Copy link
Collaborator Author

thepetk commented Feb 4, 2026

any chance of pointing your test env to the dev cluster's RHOAI endpoints and seeing if the model catalog bridge / openshift ai connector works

@gabemontero I've tried something different without pointing the connector to the devCluster:

  • Verified my Openshift AI Operator is at the same version with the one we currently have on devCluster (2.25.1)
  • Registered a new model through RHOAI
  • I'm able to see this model registered in RHDH
  • Logs from the model catalog pods look OK.

Happy to perform the test you're suggesting, for sure I'll need a bit more time to make it happen. Just in case let me know if my test above covers your concerns

@thepetk thepetk requested a review from gabemontero February 4, 2026 15:28
@gabemontero
Copy link
Contributor

any chance of pointing your test env to the dev cluster's RHOAI endpoints and seeing if the model catalog bridge / openshift ai connector works

@gabemontero I've tried something different without pointing the connector to the devCluster:

  • Verified my Openshift AI Operator is at the same version with the one we currently have on devCluster (2.25.1)
  • Registered a new model through RHOAI
  • I'm able to see this model registered in RHDH
  • Logs from the model catalog pods look OK.

Happy to perform the test you're suggesting, for sure I'll need a bit more time to make it happen. Just in case let me know if my test above covers your concerns

that test is totally a viable alternative @thepetk .... simply tried to convey the "simplest" option when putting forth that option

disabled: false
- package: oci://quay.io/tpetkos/customized-sign-in-page:v0.1.0!red-hat-developer-hub-backstage-plugin-customized-sign-in-page
##### Custom sign in page plugin #####
- package: oci://quay.io/tpetkos/customized-sign-in-page:v0.1.0
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

by no means wanted to gate the review of this on the comment, but I wonder @thepetk @johnmcollier if we should push a copy of quay.io/tpetkos/customized-sign-in-page:v0.1.0 to quay.io/redhat-ai-dev

wdyt?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, let's push it under redhat-ai-dev

Copy link
Contributor

@johnmcollier johnmcollier left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Changes look good

@thepetk Can you confirm if the Tool calling/trail UI shows up with the updated Lightspeed?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants