Fix CNPG webhook drift for WAL timeout parameters #14

Merged

forgejo_admin merged 1 commit from 13-deploy-cloudnativepg-operator-cluster-to into main

2026-03-02 21:17:43 +00:00

forgejo_admin commented

2026-03-02 21:15:08 +00:00

Owner

Summary

Pin wal_receiver_timeout and wal_sender_timeout CNPG defaults in the CNPG Cluster manifest
CNPG's admission webhook injects these values (5min) when absent, causing Terraform provider "inconsistent result after apply" errors on every tofu apply

Changes

terraform/main.tf: added wal_receiver_timeout = "5min" and wal_sender_timeout = "5min" to the postgresql.parameters block in the kubernetes_manifest.cnpg_cluster resource

Test Plan

tofu fmt -check passes
tofu validate passes
tofu plan shows no drift for the CNPG cluster resource (requires live cluster)

Review Checklist

No secrets committed
No unnecessary file changes
Commit messages are descriptive
tofu fmt and tofu validate pass

Forgejo issue: #13
plan-2026-02-26-tf-modularize-postgres -- Phase 2

## Summary - Pin `wal_receiver_timeout` and `wal_sender_timeout` CNPG defaults in the CNPG Cluster manifest - CNPG's admission webhook injects these values (`5min`) when absent, causing Terraform provider "inconsistent result after apply" errors on every `tofu apply` ## Changes - `terraform/main.tf`: added `wal_receiver_timeout = "5min"` and `wal_sender_timeout = "5min"` to the `postgresql.parameters` block in the `kubernetes_manifest.cnpg_cluster` resource ## Test Plan - [x] `tofu fmt -check` passes - [x] `tofu validate` passes - [ ] `tofu plan` shows no drift for the CNPG cluster resource (requires live cluster) ## Review Checklist - [x] No secrets committed - [x] No unnecessary file changes - [x] Commit messages are descriptive - [x] `tofu fmt` and `tofu validate` pass ## Related Notes - Forgejo issue: #13 - `plan-2026-02-26-tf-modularize-postgres` -- Phase 2

forgejo_admin added 1 commit

2026-03-02 21:15:08 +00:00

Fix CNPG admission webhook drift for WAL timeout parameters 721a27e650

CNPG's admission webhook injects wal_receiver_timeout and
wal_sender_timeout defaults into the Cluster spec, causing
Terraform provider drift errors on every apply. Pin the
CNPG default values (5min) in the manifest.

Closes #13

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>