towerops/k8s/deployment.yaml
Graham McIntire 8d00a04187
feat: add FluxCD image automation, fix Preseem duplicate AP crash
Switch deployment pipeline from GitLab Agent kubectl to FluxCD image
automation. CI now only builds and pushes tagged images to Forgejo
registry; FluxCD detects new tags and updates the deployment manifest.

Fix Ecto.MultipleResultsError on /dashboard when a device has multiple
Preseem access point records by using limit(1) instead of Repo.get_by.
2026-02-14 10:30:12 -06:00

147 lines
4.6 KiB
YAML

---
apiVersion: apps/v1
kind: Deployment
metadata:
name: towerops
namespace: towerops
spec:
replicas: 2 # Run 3 replicas for high availability
strategy:
type: RollingUpdate
rollingUpdate:
maxSurge: 100% # Start all new pods at once for fast rollouts
maxUnavailable: 0 # Keep all pods running during rollout
minReadySeconds: 10 # Wait 10s after pod is ready before considering it available
selector:
matchLabels:
app: towerops
template:
metadata:
labels:
app: towerops
spec:
# Use system-cluster-critical priority to ensure towerops starts after CNI is ready
priorityClassName: system-cluster-critical
terminationGracePeriodSeconds: 90 # Allow 90s for graceful shutdown and connection draining
imagePullSecrets:
- name: forgejo-registry
securityContext:
runAsNonRoot: true
runAsUser: 65534
fsGroup: 65534
seccompProfile:
type: RuntimeDefault
containers:
- name: towerops
image: git.mcintire.me/graham/towerops-web:latest # {"$imagepolicy": "flux-system:towerops-web"}
imagePullPolicy: IfNotPresent
securityContext:
allowPrivilegeEscalation: false
runAsNonRoot: true
runAsUser: 65534
capabilities:
drop:
- ALL
seccompProfile:
type: RuntimeDefault
ports:
- containerPort: 4000
name: http
- containerPort: 4369
name: epmd
- containerPort: 9000
name: dist
lifecycle:
preStop:
exec:
command:
- /bin/sh
- -c
- |
# Stop accepting new connections by failing readiness probe
# Give load balancer time to remove pod from rotation
# Allow existing connections to complete
sleep 20
env:
- name: POD_IP
valueFrom:
fieldRef:
fieldPath: status.podIP
- name: POD_NAME
valueFrom:
fieldRef:
fieldPath: metadata.name
- name: POD_NAMESPACE
valueFrom:
fieldRef:
fieldPath: metadata.namespace
- name: ETCD_ENABLED
value: "true"
- name: RELEASE_DISTRIBUTION
value: "name"
- name: RELEASE_NODE
value: "towerops@$(POD_IP)"
- name: PORT
value: "4000"
- name: PHX_HOST
value: "towerops.net"
- name: MIX_ENV
value: "prod"
- name: RELEASE_COOKIE
valueFrom:
secretKeyRef:
name: towerops-secrets
key: RELEASE_COOKIE
- name: SECRET_KEY_BASE
valueFrom:
secretKeyRef:
name: towerops-secrets
key: SECRET_KEY_BASE
- name: CLOAK_KEY
valueFrom:
secretKeyRef:
name: towerops-secrets
key: CLOAK_KEY
- name: DATABASE_SSL
value: "true"
- name: DATABASE_SSL_VERIFY
value: "false"
envFrom:
# Redis connection configured via towerops-redis secret
- secretRef:
name: towerops-redis
- secretRef:
name: towerops-db
- secretRef:
name: towerops-aws
resources:
requests:
memory: "1.5Gi"
cpu: "100m"
limits:
memory: "3Gi"
cpu: "500m"
startupProbe:
httpGet:
path: /health
port: 4000
initialDelaySeconds: 10
periodSeconds: 5
timeoutSeconds: 3
failureThreshold: 12 # 10s + (12 * 5s) = 70s max startup time
livenessProbe:
httpGet:
path: /health
port: 4000
periodSeconds: 10
timeoutSeconds: 3
failureThreshold: 3
readinessProbe:
httpGet:
path: /health
port: 4000
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 2
successThreshold: 1 # Mark ready after first successful check
failureThreshold: 2 # Allow 1 failure before marking not ready