rock
1bfe08eeb8
The header of this file always said 2 instances while the spec said 3. The third never ran: gotify-db-3's volume held a single replica that could not start, because talos-cp-2 disk2 is at 2.9% free. That left the cluster permanently in 'Creating a new replica' with one degraded Longhorn volume. Two replicas on two different nodes is the redundancy that actually exists.
40 lines
1.3 KiB
YAML
40 lines
1.3 KiB
YAML
# CNPG Postgres for Gotify. Lightweight — 2 instances, 2Gi storage.
|
|
# CNPG generates secret `gotify-db-app` + service `gotify-db-rw` in ns notifications.
|
|
apiVersion: postgresql.cnpg.io/v1
|
|
kind: Cluster
|
|
metadata:
|
|
name: gotify-db
|
|
namespace: notifications
|
|
annotations:
|
|
argocd.argoproj.io/sync-options: SkipDryRunOnMissingResource=true
|
|
spec:
|
|
# 2, not 3. The file's own header always said 2, and the third instance never
|
|
# existed as a running pod: its volume only ever held a replica that could not
|
|
# start, because talos-cp-2 disk2 has no free space. A third replica on a disk
|
|
# that cannot host it is not redundancy, it is a permanent degraded volume.
|
|
instances: 2
|
|
imageName: ghcr.io/cloudnative-pg/postgresql:16.2
|
|
bootstrap:
|
|
initdb:
|
|
database: gotify
|
|
owner: app
|
|
encoding: UTF8
|
|
localeCollate: C
|
|
localeCType: C
|
|
enableSuperuserAccess: false
|
|
resources:
|
|
requests: { memory: "256Mi", cpu: "100m" }
|
|
limits: { memory: "512Mi", cpu: "500m" }
|
|
storage:
|
|
size: 2Gi
|
|
storageClass: longhorn-cnpg
|
|
monitoring:
|
|
enablePodMonitor: true
|
|
affinity:
|
|
podAntiAffinityType: preferred
|
|
topologyKey: kubernetes.io/hostname
|
|
tolerations:
|
|
- key: node-role.kubernetes.io/control-plane
|
|
operator: Exists
|
|
effect: NoSchedule
|