steerer

Istio progressive rollout gated by Prometheus HTTP success rate metric

Usage

Create a test namespace:

kubectl apply -f ./artifacts/namespace/

Create GA and canary deployments, services, hpa and Istio virtual service:

kubectl apply -f ./artifacts/workloads/

Rollout:

Rollout flow:

• scan namespace for deployments marked for rollout
• scan namespace for a corresponding canary deployment (-canary prefix)
• check Istio virtual service routes are mapped to GA and canary ClusterIP services
• check GA and canary deployments status (halt rollout if a rolling update is underway or if pods are unhealthy)
• increase canary traffic weight percentage by 10%
• check canary HTTP success rate (halt rollout if percentage is under the specified threshold)
• advance canary traffic wight by 10% till it reaches 100%
  • halt rollout while canary success rate is under the threshold
  • halt rollout if the GA or canary deployment becomes unhealthy
  • halt rollout while canary deployment is being scaled up/down by HPA
• promote canary to GA (copy canary deployment spec template over GA)
• wait for GA rolling update to finish
• route all traffic to GA
• scale to zero the canary deployment
• mark rollout deployment as finished
• wait for the canary deployment to be updated (revision bump) and start over

HTTP success rate query:

sum(
    rate(
        istio_requests_total{
          reporter="destination",
          destination_workload_namespace=~"$namespace",
          destination_workload=~"$workload",
          response_code!~"5.*"
        }[$interval]
    )
) 
/ 
sum(
    rate(
        istio_requests_total{
          reporter="destination",
          destination_workload_namespace=~"$namespace",
          destination_workload=~"$workload"
        }[$interval]
    )
)