Gateway API Canary Deployments

This guide shows you how to use Gateway API and Flagger to automate canary deployments and A/B testing.

Flagger Canary Stages

Prerequisites

Flagger requires a Kubernetes cluster v1.19 or newer and any mesh/ingress that implements the v1beta1 or the v1 version of Gateway API. We'll be using Istio for the sake of this tutorial, but you can use any other implementation.

Install the Gateway API CRDs

kubectl apply -k "github.com/kubernetes-sigs/gateway-api/config/crd?ref=v1.0.0"

Install Istio:

istioctl install --set profile=minimal -y

# Suggestion: Please change release-1.20 in below command, to your real istio version.
kubectl apply -f https://raw.githubusercontent.com/istio/istio/release-1.20/samples/addons/prometheus.yaml

Install Flagger in the flagger-system namespace:

kubectl create ns flagger-system

helm repo add flagger https://flagger.app
helm upgrade -i flagger flagger/flagger \
  --namespace flagger-system \
  --set prometheus.install=false \
  --set meshProvider=gatewayapi:v1 \
  --set metricsServer=http://prometheus.istio-system:9090

Note: The above installation sets the mesh provider to be gatewayapi:v1. If your Gateway API implementation uses the v1beta1 CRDs, then set the --meshProvider value to gatewayapi:v1beta1.

Create a namespace for the Gateway:

kubectl create ns istio-ingress

Create a Gateway that configures load balancing, traffic ACL, etc:

apiVersion: gateway.networking.k8s.io/v1beta1
kind: Gateway
metadata:
  name: gateway
  namespace: istio-ingress
spec:
  gatewayClassName: istio
  listeners:
  - name: default
    hostname: "*.example.com"
    port: 80
    protocol: HTTP
    allowedRoutes:
      namespaces:
        from: All

Bootstrap

Flagger takes a Kubernetes deployment and optionally a horizontal pod autoscaler (HPA), then creates a series of objects (Kubernetes deployments, ClusterIP services, HTTPRoutes for the Gateway). These objects expose the application inside the mesh and drive the canary analysis and promotion.

Create a test namespace:

kubectl create ns test

Create a deployment and a horizontal pod autoscaler:

kubectl apply -k https://github.com/fluxcd/flagger//kustomize/podinfo?ref=main

Deploy the load testing service to generate traffic during the canary analysis:

kubectl apply -k https://github.com/fluxcd/flagger//kustomize/tester?ref=main

Create metric templates targeting the Prometheus server in the flagger-system namespace. The PromQL queries below are meant for Envoy, but you can change it to your ingress/mesh provider accordingly.

apiVersion: flagger.app/v1beta1
kind: MetricTemplate
metadata:
  name: latency
  namespace: flagger-system
spec:
  provider:
    type: prometheus
    address: http://prometheus.istio-system:9090
  query: |
    histogram_quantile(0.99,
      sum(
        rate(
          istio_request_duration_milliseconds_bucket{
            reporter="source",
            destination_workload_namespace=~"{{ namespace }}",
            destination_workload=~"{{ target }}",
          }[{{ interval }}]
        )
      ) by (le)
    )/1000
---
apiVersion: flagger.app/v1beta1
kind: MetricTemplate
metadata:
  name: error-rate
  namespace: flagger-system
spec:
  provider:
    type: prometheus
    address: http://prometheus.istio-system:9090
  query: |
    100 - sum(
      rate(
        istio_requests_total{
          reporter="source",
          destination_workload_namespace=~"{{ namespace }}",
          destination_workload=~"{{ target }}",
          response_code!~"5.*"
        }[{{ interval }}]
      )
    )
    /
    sum(
      rate(
        istio_requests_total{
          reporter="source",
          destination_workload_namespace=~"{{ namespace }}",
          destination_workload=~"{{ target }}",
        }[{{ interval }}]
      )
    )
    * 100

Save the above resource as metric-templates.yaml and then apply it:

kubectl apply -f metric-templates.yaml

Create a canary custom resource (replace "www.example.com" with your own domain):

apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: podinfo
  namespace: test
spec:
  # deployment reference
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: podinfo
  # the maximum time in seconds for the canary deployment
  # to make progress before it is rollback (default 600s)
  progressDeadlineSeconds: 60
  # HPA reference (optional)
  autoscalerRef:
    apiVersion: autoscaling/v2
    kind: HorizontalPodAutoscaler
    name: podinfo
  service:
    # service port number
    port: 9898
    # container port number or name (optional)
    targetPort: 9898
    # Gateway API HTTPRoute host names
    hosts:
     - www.example.com
    # Reference to the Gateway that the generated HTTPRoute would attach to.
    gatewayRefs:
      - name: gateway
        namespace: istio-ingress
  analysis:
    # schedule interval (default 60s)
    interval: 1m
    # max number of failed metric checks before rollback
    threshold: 5
    # max traffic percentage routed to canary
    # percentage (0-100)
    maxWeight: 50
    # canary increment step
    # percentage (0-100)
    stepWeight: 10
    metrics:
    - name: error-rate
      # max error rate (5xx responses)
      # percentage (0-100)
      templateRef:
        name: error-rate
        namespace: flagger-system
      thresholdRange:
        max: 1
      interval: 1m
    - name: latency
      templateRef:
        name: latency
        namespace: flagger-system
      # seconds
      thresholdRange:
         max: 0.5
      interval: 30s
    # testing (optional)
    webhooks:
      - name: smoke-test
        type: pre-rollout
        url: http://flagger-loadtester.test/
        timeout: 15s
        metadata:
          type: bash
          cmd: "curl -sd 'anon' http://podinfo-canary.test:9898/token | grep token"
      - name: load-test
        url: http://flagger-loadtester.test/
        timeout: 5s
        metadata:
          cmd: "hey -z 2m -q 10 -c 2 -host www.example.com http://gateway-istio.istio-ingress/"

Save the above resource as podinfo-canary.yaml and then apply it:

kubectl apply -f ./podinfo-canary.yaml

When the canary analysis starts, Flagger will call the pre-rollout webhooks before routing traffic to the canary. The canary analysis will run for five minutes while validating the HTTP metrics and rollout hooks every minute.

After a couple of seconds Flagger will create the canary objects:

# applied 
deployment.apps/podinfo
horizontalpodautoscaler.autoscaling/podinfo
canary.flagger.app/podinfo

# generated 
deployment.apps/podinfo-primary
horizontalpodautoscaler.autoscaling/podinfo-primary
service/podinfo
service/podinfo-canary
service/podinfo-primary
httproutes.gateway.networking.k8s.io/podinfo

Expose the app outside the cluster

Find the external address of Istio's load balancer:

export ADDRESS="$(kubectl -n istio-ingress get svc/gateway-istio -ojson \
| jq -r ".status.loadBalancer.ingress[].hostname")"
echo $ADDRESS

Configure your DNS server with a CNAME record (AWS) or A record (GKE/AKS/DOKS) and point a domain e.g. www.example.com to the LB address.

Now you can access the podinfo UI using your domain address.

Note that you should be using HTTPS when exposing production workloads on internet. You can obtain free TLS certs from Let's Encrypt, read this guide on how to configure cert-manager to secure Istio with TLS certificates.

If you're using a local cluster via kind/k3s you can port forward the Envoy LoadBalancer service:

kubectl port-forward -n istio-ingress svc/gateway-istio 8080:80

Now you can access podinfo via curl -H "Host: www.example.com" localhost:8080

Automated canary promotion

Trigger a canary deployment by updating the container image:

kubectl -n test set image deployment/podinfo \
podinfod=stefanprodan/podinfo:6.0.1

Flagger detects that the deployment revision changed and starts a new rollout:

kubectl -n test describe canary/podinfo

Status:
  Canary Weight:         0
  Failed Checks:         0
  Phase:                 Succeeded
Events:
  Type     Reason  Age   From     Message
  ----     ------  ----  ----     -------
  Normal   Synced  3m    flagger  New revision detected podinfo.test
  Normal   Synced  3m    flagger  Scaling up podinfo.test
  Warning  Synced  3m    flagger  Waiting for podinfo.test rollout to finish: 0 of 1 updated replicas are available
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 5
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 10
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 15
  Normal   Synced  2m    flagger  Advance podinfo.test canary weight 20
  Normal   Synced  2m    flagger  Advance podinfo.test canary weight 25
  Normal   Synced  1m    flagger  Advance podinfo.test canary weight 30
  Normal   Synced  1m    flagger  Advance podinfo.test canary weight 35
  Normal   Synced  55s   flagger  Advance podinfo.test canary weight 40
  Normal   Synced  45s   flagger  Advance podinfo.test canary weight 45
  Normal   Synced  35s   flagger  Advance podinfo.test canary weight 50
  Normal   Synced  25s   flagger  Copying podinfo.test template spec to podinfo-primary.test
  Warning  Synced  15s   flagger  Waiting for podinfo-primary.test rollout to finish: 1 of 2 updated replicas are available
  Normal   Synced  5s    flagger  Promotion completed! Scaling down podinfo.test

Note that if you apply new changes to the deployment during the canary analysis, Flagger will restart the analysis.

A canary deployment is triggered by changes in any of the following objects:

• Deployment PodSpec (container image, command, ports, env, resources, etc)

• ConfigMaps mounted as volumes or mapped to environment variables

• Secrets mounted as volumes or mapped to environment variables

You can monitor how Flagger progressively changes the weights of the HTTPRoute object that is attahed to the Gateway with:

watch kubectl get httproute -n test podinfo -o=jsonpath='{.spec.rules}'

You can monitor all canaries with:

watch kubectl get canaries --all-namespaces

NAMESPACE   NAME      STATUS        WEIGHT   LASTTRANSITIONTIME
test        podinfo   Progressing   15       2022-01-16T14:05:07Z
prod        frontend  Succeeded     0        2022-01-15T16:15:07Z
prod        backend   Failed        0        2022-01-14T17:05:07Z

Automated rollback

During the canary analysis you can generate HTTP 500 errors and high latency to test if Flagger pauses the rollout.

Trigger another canary deployment:

kubectl -n test set image deployment/podinfo \
podinfod=stefanprodan/podinfo:6.0.2

Exec into the load tester pod with:

kubectl -n test exec -it flagger-loadtester-xx-xx sh

Generate HTTP 500 errors:

watch curl http://podinfo-canary:9898/status/500

Generate latency:

watch curl http://podinfo-canary:9898/delay/1

When the number of failed checks reaches the canary analysis threshold, the traffic is routed back to the primary, the canary is scaled to zero and the rollout is marked as failed.

kubectl -n test describe canary/podinfo

Status:
  Canary Weight:         0
  Failed Checks:         10
  Phase:                 Failed
Events:
  Type     Reason  Age   From     Message
  ----     ------  ----  ----     -------
  Normal   Synced  3m    flagger  Starting canary deployment for podinfo.test
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 5
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 10
  Normal   Synced  3m    flagger  Advance podinfo.test canary weight 15
  Normal   Synced  3m    flagger  Halt podinfo.test advancement error rate 69.17% > 1%
  Normal   Synced  2m    flagger  Halt podinfo.test advancement error rate 61.39% > 1%
  Normal   Synced  2m    flagger  Halt podinfo.test advancement error rate 55.06% > 1%
  Normal   Synced  2m    flagger  Halt podinfo.test advancement error rate 47.00% > 1%
  Normal   Synced  2m    flagger  (combined from similar events): Halt podinfo.test advancement error rate 38.08% > 1%
  Warning  Synced  1m    flagger  Rolling back podinfo.test failed checks threshold reached 10
  Warning  Synced  1m    flagger  Canary failed! Scaling down podinfo.test

Session Affinity

While Flagger can perform weighted routing and A/B testing individually, with Gateway API it can combine the two leading to a Canary release with session affinity. For more information you can read the deployment strategies docs.

Note: The implementation must have support for the ResponseHeaderModifier API.

Create a canary custom resource (replace www.example.com with your own domain):

apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: podinfo
  namespace: test
spec:
  # deployment reference
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: podinfo
  # the maximum time in seconds for the canary deployment
  # to make progress before it is rollback (default 600s)
  progressDeadlineSeconds: 60
  # HPA reference (optional)
  autoscalerRef:
    apiVersion: autoscaling/v2
    kind: HorizontalPodAutoscaler
    name: podinfo
  service:
    # service port number
    port: 9898
    # container port number or name (optional)
    targetPort: 9898
    # Gateway API HTTPRoute host names
    hosts:
     - www.example.com
    # Reference to the Gateway that the generated HTTPRoute would attach to.
    gatewayRefs:
      - name: gateway
        namespace: istio-ingress
  analysis:
    # schedule interval (default 60s)
    interval: 1m
    # max number of failed metric checks before rollback
    threshold: 5
    # max traffic percentage routed to canary
    # percentage (0-100)
    maxWeight: 50
    # canary increment step
    # percentage (0-100)
    stepWeight: 10
    # session affinity config
    sessionAffinity:
      # name of the cookie used
      cookieName: flagger-cookie
      # max age of the cookie (in seconds)
      # optional; defaults to 86400
      maxAge: 21600
    metrics:
    - name: error-rate
      # max error rate (5xx responses)
      # percentage (0-100)
      templateRef:
        name: error-rate
        namespace: flagger-system
      thresholdRange:
        max: 1
      interval: 1m
    - name: latency
      templateRef:
        name: latency
        namespace: flagger-system
      # seconds
      thresholdRange:
         max: 0.5
      interval: 30s
    # testing (optional)
    webhooks:
      - name: smoke-test
        type: pre-rollout
        url: http://flagger-loadtester.test/
        timeout: 15s
        metadata:
          type: bash
          cmd: "curl -sd 'anon' http://podinfo-canary.test:9898/token | grep token"
      - name: load-test
        url: http://flagger-loadtester.test/
        timeout: 5s
        metadata:
          cmd: "hey -z 2m -q 10 -c 2 -host www.example.com http://gateway-istio.istio-ingress/"

Save the above resource as podinfo-canary-session-affinity.yaml and then apply it:

kubectl apply -f ./podinfo-canary-session-affinity.yaml

Trigger a canary deployment by updating the container image:

kubectl -n test set image deployment/podinfo \
podinfod=ghcr.io/stefanprodan/podinfo:6.0.1

You can load www.example.com in your browser and refresh it until you see the requests being served by podinfo:6.0.1. All subsequent requests after that will be served by podinfo:6.0.1 and not podinfo:6.0.0 because of the session affinity configured by Flagger in the HTTPRoute object.

A/B Testing

Besides weighted routing, Flagger can be configured to route traffic to the canary based on HTTP match conditions. In an A/B testing scenario, you'll be using HTTP headers or cookies to target a certain segment of your users. This is particularly useful for frontend applications that require session affinity.

Flagger A/B Testing Stages

Create a canary custom resource (replace "www.example.com" with your own domain):

apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: podinfo
  namespace: test
spec:
  # deployment reference
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: podinfo
  # the maximum time in seconds for the canary deployment
  # to make progress before it is rollback (default 600s)
  progressDeadlineSeconds: 60
  # HPA reference (optional)
  autoscalerRef:
    apiVersion: autoscaling/v2
    kind: HorizontalPodAutoscaler
    name: podinfo
  service:
    # service port number
    port: 9898
    # container port number or name (optional)
    targetPort: 9898
    # Gateway API HTTPRoute host names
    hosts:
     - www.example.com
    # Reference to the Gateway that the generated HTTPRoute would attach to.
    gatewayRefs:
      - name: gateway
        namespace: istio-ingress
  analysis:
    # schedule interval (default 60s)
    interval: 1m
    # max number of failed metric checks before rollback
    threshold: 5
    # max traffic percentage routed to canary
    # percentage (0-100)
    maxWeight: 50
    # canary increment step
    # percentage (0-100)
    stepWeight: 10
    metrics:
    - name: error-rate
      # max error rate (5xx responses)
      # percentage (0-100)
      templateRef:
        name: error-rate
        namespace: flagger-system
      thresholdRange:
        max: 1
      interval: 1m
    - name: latency
      templateRef:
        name: latency
        namespace: flagger-system
      # seconds
      thresholdRange:
         max: 0.5
      interval: 30s
    # testing (optional)
    webhooks:
      - name: smoke-test
        type: pre-rollout
        url: http://flagger-loadtester.test/
        timeout: 15s
        metadata:
          type: bash
          cmd: "curl -sd 'anon' http://podinfo-canary.test:9898/token | grep token"
      - name: load-test
        url: http://flagger-loadtester.test/
        timeout: 5s
        metadata:
          cmd: "hey -z 2m -q 10 -c 2 -host www.example.com -H 'X-Canary: insider' http://gateway-istio.istio-ingress/"

The above configuration will run an analysis for ten minutes targeting those users that have an insider cookie.

Save the above resource as podinfo-ab-canary.yaml and then apply it:

kubectl apply -f ./podinfo-ab-canary.yaml

Trigger a canary deployment by updating the container image:

kubectl -n test set image deployment/podinfo \
podinfod=stefanprodan/podinfo:6.0.3

Flagger detects that the deployment revision changed and starts a new rollout:

kubectl -n test describe canary/podinfo

Status:
  Failed Checks:         0
  Phase:                 Succeeded
Events:
  Type     Reason  Age   From     Message
  ----     ------  ----  ----     -------
  Normal   Synced  3m    flagger  New revision detected podinfo.test
  Normal   Synced  3m    flagger  Scaling up podinfo.test
  Warning  Synced  3m    flagger  Waiting for podinfo.test rollout to finish: 0 of 1 updated replicas are available
  Normal   Synced  3m    flagger  Advance podinfo.test canary iteration 1/10
  Normal   Synced  3m    flagger  Advance podinfo.test canary iteration 2/10
  Normal   Synced  3m    flagger  Advance podinfo.test canary iteration 3/10
  Normal   Synced  2m    flagger  Advance podinfo.test canary iteration 4/10
  Normal   Synced  2m    flagger  Advance podinfo.test canary iteration 5/10
  Normal   Synced  1m    flagger  Advance podinfo.test canary iteration 6/10
  Normal   Synced  1m    flagger  Advance podinfo.test canary iteration 7/10
  Normal   Synced  55s   flagger  Advance podinfo.test canary iteration 8/10
  Normal   Synced  45s   flagger  Advance podinfo.test canary iteration 9/10
  Normal   Synced  35s   flagger  Advance podinfo.test canary iteration 10/10
  Normal   Synced  25s   flagger  Copying podinfo.test template spec to podinfo-primary.test
  Warning  Synced  15s   flagger  Waiting for podinfo-primary.test rollout to finish: 1 of 2 updated replicas are available
  Normal   Synced  5s    flagger  Promotion completed! Scaling down podinfo.test

Traffic mirroring

Flagger Canary Traffic Shadowing

For applications that perform read operations, Flagger can be configured to do B/G tests with traffic mirroring. Gateway API traffic mirroring will copy each incoming request, sending one request to the primary and one to the canary service. The response from the primary is sent back to the user and the response from the canary is discarded. Metrics are collected on both requests so that the deployment will only proceed if the canary metrics are within the threshold values.

Note that mirroring should be used for requests that are idempotent or capable of being processed twice (once by the primary and once by the canary).

You can enable mirroring by replacing stepWeight with iterations and by setting analysis.mirror to true:

apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: podinfo
  namespace: test
spec:
  # deployment reference
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: podinfo
  service:
    # service port number
    port: 9898
    # container port number or name (optional)
    targetPort: 9898
    # Gateway API HTTPRoute host names
    hosts:
     - www.example.com
    # Reference to the Gateway that the generated HTTPRoute would attach to.
    gatewayRefs:
      - name: gateway
        namespace: istio-ingress
  analysis:
    # schedule interval
    interval: 1m
    # max number of failed metric checks before rollback
    threshold: 5
    # total number of iterations
    iterations: 10
    # enable traffic shadowing
    mirror: true
    # Gateway API HTTPRoute host names
    metrics:
      - name: request-success-rate
        thresholdRange:
          min: 99
        interval: 1m
      - name: request-duration
        thresholdRange:
          max: 500
        interval: 1m
    webhooks:
      - name: load-test
        url: http://flagger-loadtester.test/
        timeout: 5s
        metadata:
          cmd: "hey -z 2m -q 10 -c 2 -host www.example.com http://gateway-istio.istio-ingress/"

With the above configuration, Flagger will run a canary release with the following steps:

• detect new revision (deployment spec, secrets or configmaps changes)

• scale from zero the canary deployment

• wait for the HPA to set the canary minimum replicas

• check canary pods health

• run the acceptance tests

• abort the canary release if tests fail

• start the load tests

• mirror 100% of the traffic from primary to canary

• check request success rate and request duration every minute

• abort the canary release if the metrics check failure threshold is reached

• stop traffic mirroring after the number of iterations is reached

• route live traffic to the canary pods

• promote the canary (update the primary secrets, configmaps and deployment spec)

• wait for the primary deployment rollout to finish

• wait for the HPA to set the primary minimum replicas

• check primary pods health

• switch live traffic back to primary

• scale to zero the canary

• send notification with the canary analysis result

The above procedures can be extended with custom metrics checks, webhooks, manual promotion approval and Slack or MS Teams notifications.