Stellar-IndigoPay/gitops/argo-rollouts-canary.yaml at main · malaysiaonelove/Stellar-IndigoPay · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
# gitops/argo-rollouts-canary.yaml
#
# Argo Rollouts STEPPED canary strategy for the backend. Replaces the
# `Deployment` in k8s/backend.yaml with a `Rollout` that performs a
# stepped rollout via pod replacement (20% → 50% → 100% of new-revision
# pods, with 5-min analysis pauses between steps).
#
# IMPORTANT — this is Argo Rollouts' DEFAULT mode. Both `canaryService`
# and `stableService` point at the same `backend-svc`, so the Rollout
# does NOT split live HTTP traffic between two Service revisions. It
# replaces pods in-place at each `setWeight` step. For real traffic-
# split canary (e.g. 5% of user traffic to the new revision), add a
# `trafficRouting` block — Istio, NGINX, ALB, or Ambassador. The
# bottom of this file has an Istio example commented out.
#
# Each new release:
#   1. Deploys 20% of pods with the new revision.
#   2. Watches the canary metrics for 5 minutes; aborts on regression.
#   3. Graduates to 50%, then 100%, with a 5-minute pause between
#      steps.
#   4. On full rollout, scales down the old revision.
#
# Prerequisite: install argo-rollouts in the cluster.
#   kubectl create namespace argo-rollouts
#   kubectl apply -n argo-rollouts -f https://github.com/argoproj/argo-rollouts/releases/latest/download/install.yaml
#
# Migration: when you adopt this, delete the existing Deployment
# `backend` (the Rollout takes over the same selector). The Service
# continues to select `app: backend`; no service change is required.
#
# Switch the Deployment→Rollout via:
#   kubectl apply -f gitops/argo-rollouts-canary.yaml
#   kubectl delete deployment backend -n indigopay

apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: backend
  namespace: stellar-indigopay
  labels:
    app: backend
spec:
  replicas: 2
  revisionHistoryLimit: 5
  selector:
    matchLabels:
      app: backend
  strategy:
    canary:
      canaryService: backend-svc
      stableService: backend-svc
      steps:
        - setWeight: 20
        - pause: { duration: 5m }
        - setWeight: 50
        - pause: { duration: 5m }
        - setWeight: 100
      analysis:
        templates:
          - templateName: backend-success-rate
        startingStep: 1
        # Re-run analysis on each weight step. If the template returns
        # `Failed`, the rollout aborts and traffic reverts to stable.
        args:
          - name: service-name
            value: backend-svc
  template:
    metadata:
      labels:
        app: backend
        app.kubernetes.io/component: api
        app.kubernetes.io/part-of: stellar-indigopay
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/port: "4000"
        prometheus.io/path: "/metrics"
    spec:
      terminationGracePeriodSeconds: 35
      containers:
        - name: backend
          image: stellar-indigopay/backend:latest
          imagePullPolicy: IfNotPresent
          ports:
            - name: http
              containerPort: 4000
              protocol: TCP
          envFrom:
            - configMapRef:
                name: stellar-indigopay-config
            - secretRef:
                name: stellar-indigopay-secrets
          startupProbe:
            httpGet:
              path: /api/health
              port: http
            initialDelaySeconds: 1
            periodSeconds: 1
            failureThreshold: 30
          livenessProbe:
            httpGet:
              path: /api/health
              port: http
            initialDelaySeconds: 0
            periodSeconds: 10
            timeoutSeconds: 3
            failureThreshold: 3
          readinessProbe:
            httpGet:
              path: /api/readyz
              port: http
            initialDelaySeconds: 5
            periodSeconds: 5
            timeoutSeconds: 5
            failureThreshold: 3
          lifecycle:
            preStop:
              exec:
                command: ["sh", "-c", "sleep 5"]
          resources:
            requests:
              cpu: "100m"
              memory: "128Mi"
            limits:
              cpu: "500m"
              memory: "512Mi"
---
apiVersion: argoproj.io/v1alpha1
kind: AnalysisTemplate
metadata:
  name: backend-success-rate
  namespace: stellar-indigopay
spec:
  args:
    - name: service-name
  metrics:
    - name: http-success-rate
      interval: 60s
      count: 5
      successCondition: result[0] >= 0.99
      failureLimit: 2
      provider:
        prometheus:
          address: http://prometheus-operated.monitoring.svc:9090
          query: |
            sum(rate(http_requests_total{job="stellar-indigopay-backend",status_code!~"5.."}[2m]))
              /
            sum(rate(http_requests_total{job="stellar-indigopay-backend"}[2m]))