tegwick
a0d4bc9de2
Add Burstable CPU/memory requests and relax liveness/readiness probes so pg_dump and forgejo-backup I/O no longer trip :8000 health checks on the ~4Gi railiance01 node (was BestEffort with 5 probe-driven restarts).
52 lines
No EOL
1.5 KiB
YAML
52 lines
No EOL
1.5 KiB
YAML
---
|
||
# CNPG cluster for Forgejo (railiance01 production forge).
|
||
# Managed by railiance-platform (S3). Operator: cnpg-system.
|
||
#
|
||
# Apply: KUBECONFIG=~/.kube/config-hosteurope make forgejo-db-deploy
|
||
# Status: make forgejo-db-status
|
||
# Backup: make forgejo-backup (pg_dump + forgejo dump → Nextcloud; docs/forgejo-backup.md)
|
||
#
|
||
# Pre-condition: forgejo-db-credentials Secret in databases namespace.
|
||
# See helm/forgejo-db-secret.sops.yaml.template
|
||
apiVersion: postgresql.cnpg.io/v1
|
||
kind: Cluster
|
||
metadata:
|
||
name: forgejo-db
|
||
namespace: databases
|
||
labels:
|
||
app.kubernetes.io/name: forgejo-db
|
||
app.kubernetes.io/component: database
|
||
app.kubernetes.io/managed-by: manual
|
||
railiance.io/layer: s3-platform
|
||
railiance.io/consumer: forgejo
|
||
spec:
|
||
instances: 1
|
||
imageName: ghcr.io/cloudnative-pg/postgresql:16
|
||
# Burstable QoS + headroom for pg_dump / forgejo-backup (was BestEffort → probe
|
||
# timeouts on :8000 under node CPU pressure). Sized like net-kingdom-pg on ~4Gi node.
|
||
resources:
|
||
requests:
|
||
cpu: 200m
|
||
memory: 512Mi
|
||
limits:
|
||
cpu: "1"
|
||
memory: 1Gi
|
||
# Default liveness window is 30s (3×10s); extend so backup I/O stalls do not restart the pod.
|
||
livenessProbeTimeout: 90
|
||
probes:
|
||
liveness:
|
||
timeoutSeconds: 10
|
||
periodSeconds: 15
|
||
failureThreshold: 6
|
||
readiness:
|
||
timeoutSeconds: 10
|
||
periodSeconds: 15
|
||
failureThreshold: 6
|
||
storage:
|
||
size: 10Gi
|
||
bootstrap:
|
||
initdb:
|
||
database: forgejo
|
||
owner: forgejo
|
||
secret:
|
||
name: forgejo-db-credentials |