Migrate forgejo's in-pod Postgres onto CloudNativePG (later) #328

Closed
opened 2026-06-15 00:18:07 +00:00 by coilysiren · 1 comment
Owner

Optional, lower-priority follow-up: migrate forgejo's in-pod postgres:17 (deploy/forgejo.yml:99) onto the shared CloudNativePG-managed cluster, so all app Postgres lives under one operator with consistent backups/upgrades.

Why later, not now

Forgejo is the source-of-truth git server - its Postgres works and is isolated today. No urgency. Worth doing only once CloudNativePG is proven in-cluster (see the parent task) so the migration is low-risk.

Scope (when picked up)

  • Provision a forgejo database/role on the shared CloudNativePG cluster
  • Plan + test a dump/restore (or logical replication) cutover with forgejo quiesced
  • Update deploy/forgejo.yml to point at the managed cluster and drop the in-pod postgres container
  • Verify backups + restore path

Refs

  • Depends on: CloudNativePG operator landing first (parent backlog #326)
Optional, lower-priority follow-up: migrate forgejo's in-pod `postgres:17` (`deploy/forgejo.yml:99`) onto the shared CloudNativePG-managed cluster, so all app Postgres lives under one operator with consistent backups/upgrades. ## Why later, not now Forgejo is the source-of-truth git server - its Postgres works and is isolated today. No urgency. Worth doing only once CloudNativePG is proven in-cluster (see the parent task) so the migration is low-risk. ## Scope (when picked up) - [ ] Provision a `forgejo` database/role on the shared CloudNativePG cluster - [ ] Plan + test a dump/restore (or logical replication) cutover with forgejo quiesced - [ ] Update `deploy/forgejo.yml` to point at the managed cluster and drop the in-pod postgres container - [ ] Verify backups + restore path ## Refs - Depends on: CloudNativePG operator landing first (parent backlog https://forgejo.coilysiren.me/coilyco-flight-deck/infrastructure/issues/326)
Author
Owner

Backlog burndown 2026-06-17: closing low-priority (P3/P4) to bring the open count to a manageable level. Nothing lost — reopen if this resurfaces. Batch tag: burndown-2026-06.

Backlog burndown 2026-06-17: closing low-priority (P3/P4) to bring the open count to a manageable level. Nothing lost — reopen if this resurfaces. Batch tag: `burndown-2026-06`.
coilysiren 2026-06-17 08:22:31 +00:00
Sign in to join this conversation.
No description provided.