Improve nginx deployment (#4085)

1. For 60 seconds, check if no containers in the pod crash loop and
only then declare that the pod ready.

2. Add readiness probe. Traffic will be forwarded only to pods that
pass the readiness check.

3. Adjust probes to start only after 60 seconds to match the change
mention in 1.

4. During upgrade, allow an additional pod to be scheduled for the
new replicaset.

5. Don't allow the deployment to have less that 3 pod ready, even during
upgrade.

Signed-off-by: gbenhaim <gbenhaim@redhat.com>

Author: gbenhaim

components/ui/base/proxy/proxy.yaml

@@ -12,10 +12,11 @@ spec:
   selector:
     matchLabels:
       app: proxy
+  minReadySeconds: 60
   strategy:
     rollingUpdate:
-      maxSurge: 25%
-      maxUnavailable: 25%
+      maxSurge: 1
+      maxUnavailable: 0
   template:
     metadata:
       labels:
@@ -62,10 +63,20 @@ spec:
             path: /health
             port: 8080
             scheme: HTTP
-          initialDelaySeconds: 10
+          initialDelaySeconds: 60
           periodSeconds: 60
           successThreshold: 1
           timeoutSeconds: 1
+        readinessProbe:
+          failureThreshold: 3
+          httpGet:
+            path: /health
+            port: 8080
+            scheme: HTTP
+          initialDelaySeconds: 60
+          periodSeconds: 30
+          successThreshold: 3
+          timeoutSeconds: 1
         ports:
         - containerPort: 8080
           name: web