Meet Deployment Fix

Problem

The meet application’s infrastructure dependencies were missing from the Kubernetes cluster:

meet-postgresql (Bitnami PostgreSQL) — not deployed
meet-redis (Redis cache) — not deployed
meet-static (Nginx static file server) — not deployed

This caused:

meet-superuser job in CrashLoopBackOff (19 restarts) — database unreachable
meet-migrate job stuck in Running for 103 minutes — waiting for database
meet-frontend pod stuck in Init:0/1 — ConfigMap meet-static-files missing

Root Cause

The meet Helm chart was deployed via helmfile, but the supporting releases (meet-postgresql, meet-redis, meet-static) defined in helmfile/apps/meet/helmfile-child.yaml.gotmpl were never installed. The database host was configured as meet-cluster-rw (the Bitnami PostgreSQL service with fullnameOverride: meet-cluster-rw), but the service didn’t exist.

Fix Applied

Deployed the three missing releases using helmfile with selectors:

helmfile -e demo sync --selector name=meet-postgresql
helmfile -e demo sync --selector name=meet-redis
helmfile -e demo sync --selector name=meet-static

Resulting state (final):

Resource	Status
`meet-postgresql` release → `meet-cluster-rw-0` pod	Running
`meet-redis` release → `meet-redis-master-0` pod	Running
`meet-static` release → `meet-static-nginx-*` pod	Running
`meet-backend` deployment	Running
`meet-frontend` deployment	Running
`meet-migrate` job	Completed
`meet-superuser` job	Completed (“Superuser created successfully.”)