Migrate workloads from Zero to Smaug

[HumairAK]

Let's use this issue to track all PR's related to Smaug migration. Please edit/flesh out this issue more to account for everything that needs to be migrated.

ODH:

• [x] JupyterHub / Idle culler (let's quota the JH namespace as well) (https://github.com/operate-first/apps/pull/1054)
• [x] Trino/Superset/Cloudbeaver/hue
• [x] Seldon Operator (https://github.com/operate-first/apps/pull/1068)
• [x] ~Argo Workflows~ (never got used, so we don't need to migrate it until it's needed again)
• [x] Grafana (https://github.com/operate-first/apps/pull/1062)
• [x] Grafana Dashboards (https://github.com/operate-first/apps/pull/1020)
• [x] Kafka (https://github.com/operate-first/apps/pull/1054)

Others:

• [x] Application ServiceMonitors (https://github.com/operate-first/apps/pull/1069)

• [x] CLO Forwarding

• [x] Observatorium (tracking here instead: https://github.com/operate-first/apps/issues/1210 https://github.com/operate-first/apps/issues/909)

• [x] Vector (https://github.com/operate-first/apps/pull/1288/files)

• [x] ACME operator (https://github.com/operate-first/apps/pull/936)

• [x] ~Kfpipelines~ (see: https://github.com/operate-first/support/issues/435)

• [x] Identify what users/teams are still that used Zero still need a namespace, then add these namespaces to Smaug (notes by @tumido) (Update: we are not planning to migrate any of the old projects from zero unless the team specifically requests for it.)

  • [x] pulp (pulp-operator, thoth-pulp-experiments) https://github.com/operate-first/apps/pull/1121
  • [ ] ~b4mad-minecraft (probably yes)~
  • [ ] ~open-aiops ?~
  • [x] Mesh for Data (mesh-for-data, m4d-*) (probably no, since the project was renamed)
  • [ ] ~fde-audio-decoder-demo ?~
  • [ ] ~esi-operate-first-project ?~
  • [ ] ~enterprise-neurosystem-collaboration (probably yes)~
  • [ ] ~codait-advo ?~
  • [ ] ~chi-rhug ?~
  • [ ] ~adoptium-project ?~
  • [x] aicoe-meteor https://github.com/operate-first/apps/pull/1122
  • [x] apicurio-apicurio-registry (https://github.com/operate-first/apps/pull/1258)
  • [ ] ~Pixel (pixel-*) ?~
  • [ ] ~quarkus-extension-registry (probably yes)~
  • [ ] ~redhat-cto-ns ?~
  • [ ] ~sdap ?~
  • [ ] ~sigstore ?~
  • [ ] ~sa-dach-* ?~
  • [ ] ~uky-hpc-workload-generator ?~

Docs:

• [x] update all URL to point to smaug rather than zero, for example: https://www.operate-first.cloud/users/apps/docs/odh/prometheus/README.md

Documentation:

• Identify all links to managed services/docs that are zero specific that need to be updated to point to Smaug or better yet, made generic enough so we don't need to do this in the future.

[4n4nd]

We need to deploy kfpipelines as well, since it is a service that we want to provide.

[HumairAK]

@4n4nd can you add this to the OP

[goern]

@riekrh

[Gregory-Pereira]

So I went to migrate these and argocd hasn't been migrated in operate-first/apps. I also wasn't aware that this was something we need to do. Should I migrate the app as well?

[pacospace]

what is ETA for kfp?

[Shreyanand]

Is there an ETA on migration of the unchecked services (Observatorium, Grafana dashboards)? Also, will the JH user PVCs data and the thanos metrics data on the Zero cluster be migrated as well?

[paoloantinori]

Confirming yes for Apicurio Project!

[4n4nd]

Update: we are not planning to migrate any of the old user projects from zero unless the team specifically requests for it.