Refactor canary analysis

- move CRD metrics and weight setting to canary analysis
- add max weight to CRD spec

artifacts/rollouts/podinfo.yaml

@@ -2,6 +2,8 @@
 # run tester: kubectl run -n test tester --image=quay.io/stefanprodan/podinfo:1.2.1 -- ./podinfo --port=9898
 # generate latency: watch curl http://podinfo-canary:9898/delay/1
 # generate errors: watch curl http://podinfo-canary:9898/status/500
+# run load test: kubectl run -n test -it --rm --restart=Never hey --image=stefanprodan/loadtest -- sh
+# generate load: hey -z 2m -h2 -m POST -d '{test: 1}' -c 10 -q 5 http://podinfo:9898/api/echo
 apiVersion: apps.weave.works/v1beta1
 kind: Rollout
 metadata:
@@ -12,25 +14,29 @@ metadata:
   namespace: test
 spec:
   targetKind: Deployment
+  virtualService:
+    name: podinfo
   primary:
     name: podinfo
     host: podinfo
   canary:
     name: podinfo-canary
     host: podinfo-canary
-  virtualService:
-    name: podinfo
+  canaryAnalysis:
+    # max traffic percentage routed to canary
+    # percentage (0-100)
+    maxWeight: 100
     # canary increment step
     # percentage (0-100)
-    weight: 10
-  metrics:
-  - name: istio_requests_total
-    # minimum req success rate (non 5xx responses)
-    # percentage (0-100)
-    threshold: 99
-    interval: 1m
-  - name: istio_request_duration_seconds_bucket
-    # maximum req duration P99
-    # milliseconds
-    threshold: 500
-    interval: 1m
+    stepWeight: 10
+    metrics:
+    - name: istio_requests_total
+      # minimum req success rate (non 5xx responses)
+      # percentage (0-100)
+      threshold: 99
+      interval: 1m
+    - name: istio_request_duration_seconds_bucket
+      # maximum req duration P99
+      # milliseconds
+      threshold: 500
+      interval: 30s

artifacts/steerer/crd.yaml

@@ -21,10 +21,18 @@ spec:
       properties:
         spec:
           required:
+            - targetKind
+            - virtualService
             - primary
             - canary
-            - virtualService
+            - canaryAnalysis
           properties:
+            targetKind:
+              type: string
+            virtualService:
+              properties:
+                name:
+                  type: string
             primary:
               properties:
                 name:
@@ -37,22 +45,23 @@ spec:
                   type: string
                 host:
                   type: string
-            virtualService:
+            canaryAnalysis:
               properties:
-                name:
-                  type: string
-                weight:
+                maxWeight:
                   type: number
-            metrics:
-              type: array
-              properties:
-                items:
-                  type: object
+                stepWeight:
+                  type: number
+                metrics:
+                  type: array
                   properties:
-                    name:
-                      type: string
-                    interval:
-                      type: string
-                      pattern: "^[0-9]+(m)"
-                    threshold:
-                      type: number
+                    items:
+                      type: object
+                      properties:
+                        name:
+                          type: string
+                        interval:
+                          type: string
+                          pattern: "^[0-9]+(m)"
+                        threshold:
+                          type: number
+

artifacts/steerer/deployment.yaml

@@ -22,7 +22,7 @@ spec:
       serviceAccountName: steerer
       containers:
       - name: steerer
-        image: stefanprodan/steerer:0.0.1-beta.12
+        image: stefanprodan/steerer:0.0.1-rc.1
         imagePullPolicy: Always
         ports:
         - name: http

artifacts/workloads/deployment-canary.yaml

@@ -57,7 +57,7 @@ spec:
             - http
             - localhost:9898/readyz
           failureThreshold: 3
-          periodSeconds: 10
+          periodSeconds: 3
           successThreshold: 1
           timeoutSeconds: 2
         resources:
@@ -65,7 +65,7 @@ spec:
             cpu: 1000m
             memory: 256Mi
           requests:
-            cpu: 10m
+            cpu: 100m
             memory: 16Mi
         volumeMounts:
         - mountPath: /data

artifacts/workloads/deployment.yaml

@@ -3,16 +3,13 @@ kind: Deployment
 metadata:
   name: podinfo
   namespace: test
-  annotations:
-    apps.weave.works/progressive: "true"
   labels:
     app: podinfo
 spec:
-  replicas: 2
+  replicas: 1
   strategy:
     rollingUpdate:
-      maxSurge: 25%
-      maxUnavailable: 1
+      maxUnavailable: 0
     type: RollingUpdate
   selector:
     matchLabels:
@@ -48,6 +45,7 @@ spec:
             - check
             - http
             - localhost:9898/healthz
+          initialDelaySeconds: 5
           failureThreshold: 3
           periodSeconds: 10
           successThreshold: 1
@@ -59,11 +57,15 @@ spec:
             - check
             - http
             - localhost:9898/readyz
+          initialDelaySeconds: 5
           failureThreshold: 3
-          periodSeconds: 10
+          periodSeconds: 3
           successThreshold: 1
           timeoutSeconds: 1
         resources:
+          limits:
+            cpu: 2000m
+            memory: 512Mi
           requests:
-            cpu: 1m
-            memory: 16Mi
+            cpu: 10m
+            memory: 64Mi

artifacts/workloads/hpa-canary.yaml

@@ -14,6 +14,8 @@ spec:
   - type: Resource
     resource:
       name: cpu
+      # scale up if usage is above
+      # 99% of the requested CPU (100m)
       targetAverageUtilization: 99
   - type: Resource
     resource:

artifacts/workloads/hpa.yaml

@@ -0,0 +1,19 @@
+apiVersion: autoscaling/v2beta1
+kind: HorizontalPodAutoscaler
+metadata:
+  name: podinfo
+  namespace: test
+spec:
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: podinfo
+  minReplicas: 2
+  maxReplicas: 4
+  metrics:
+  - type: Resource
+    resource:
+      name: cpu
+      # scale up if usage is above
+      # 99% of the requested CPU (100m)
+      targetAverageUtilization: 99

artifacts/workloads/virtual-service.yaml

@@ -1,13 +1,10 @@
 apiVersion: networking.istio.io/v1alpha3
 kind: VirtualService
 metadata:
-  annotations:
-    apps.weave.works/progressive-revision: ""
-    apps.weave.works/progressive-status: ""
-  labels:
-    app: podinfo
   name: podinfo
   namespace: test
+  labels:
+    app: podinfo
 spec:
   gateways:
   - public-gateway.istio-system.svc.cluster.local
@@ -26,4 +23,7 @@ spec:
         port:
           number: 9898
       weight: 0
-    timeout: 30s
+    timeout: 10s
+    retries:
+      attempts: 3
+      perTryTimeout: 2s