github/flagger
1
0
Fork 0
mirror of https://github.com/fluxcd/flagger.git synced 2026-04-15 06:57:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #118 from weaveworks/appmesh

Browse Source 
App Mesh refactoring + docs
This commit is contained in:
Stefan Prodan
2019-03-24 13:53:40 +02:00
committed by GitHub
parent 27ab4b08f9 8fa558f124
commit 1caa62adc8
30 changed files with 932 additions and 60 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Makefile
+1 -1
View File
@@ -13,7 +13,7 @@ run:
run-appmesh:
go run cmd/flagger/* -kubeconfig=$$HOME/.kube/config -log-level=info -mesh-provider=appmesh \
-metrics-server=http://ab3325177465c11e9a94c02c4171f346-2027522465.us-west-2.elb.amazonaws.com:9090 \
-metrics-server=http://acfc235624ca911e9a94c02c4171f346-1585187926.us-west-2.elb.amazonaws.com:9090 \
-slack-url=https://hooks.slack.com/services/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 \
-slack-channel="devops-alerts"
artifacts/appmesh/canary.yaml
+50
View File
@@ -0,0 +1,50 @@
apiVersion: flagger.app/v1alpha3
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/v2beta1
kind: HorizontalPodAutoscaler
name: podinfo
service:
# container port
port: 9898
# App Mesh reference
meshName: global.appmesh-system
# define the canary analysis timing and KPIs
canaryAnalysis:
# schedule interval (default 60s)
interval: 10s
# max number of failed metric checks before rollback
threshold: 10
# max traffic percentage routed to canary
# percentage (0-100)
maxWeight: 50
# canary increment step
# percentage (0-100)
stepWeight: 5
# App Mesh Prometheus checks
metrics:
- name: envoy_cluster_upstream_rq
# minimum req success rate (non 5xx responses)
# percentage (0-100)
threshold: 99
interval: 1m
# external checks (optional)
webhooks:
- name: load-test
url: http://flagger-loadtester.test/
timeout: 5s
metadata:
cmd: "hey -z 1m -q 10 -c 2 http://podinfo.test:9898/"
artifacts/appmesh/deployment.yaml
+65
View File
@@ -0,0 +1,65 @@
apiVersion: apps/v1
kind: Deployment
metadata:
name: podinfo
namespace: test
labels:
app: podinfo
spec:
minReadySeconds: 5
revisionHistoryLimit: 5
progressDeadlineSeconds: 60
strategy:
rollingUpdate:
maxUnavailable: 0
type: RollingUpdate
selector:
matchLabels:
app: podinfo
template:
metadata:
annotations:
prometheus.io/scrape: "true"
labels:
app: podinfo
spec:
containers:
- name: podinfod
image: quay.io/stefanprodan/podinfo:1.4.0
imagePullPolicy: IfNotPresent
ports:
- containerPort: 9898
name: http
protocol: TCP
command:
- ./podinfo
- --port=9898
- --level=info
env:
- name: PODINFO_UI_COLOR
value: blue
livenessProbe:
exec:
command:
- podcli
- check
- http
- localhost:9898/healthz
initialDelaySeconds: 5
timeoutSeconds: 5
readinessProbe:
exec:
command:
- podcli
- check
- http
- localhost:9898/readyz
initialDelaySeconds: 5
timeoutSeconds: 5
resources:
limits:
cpu: 2000m
memory: 512Mi
requests:
cpu: 100m
memory: 64Mi
artifacts/appmesh/global-mesh.yaml
+7
View File
@@ -0,0 +1,7 @@
apiVersion: appmesh.k8s.aws/v1alpha1
kind: Mesh
metadata:
name: global
namespace: appmesh-system
spec:
serviceDiscoveryType: dns
artifacts/appmesh/hpa.yaml
+19
View File
@@ -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/appmesh/ingress.yaml
+177
View File
@@ -0,0 +1,177 @@
---
kind: ConfigMap
apiVersion: v1
metadata:
name: ingress-config
namespace: test
labels:
app: ingress
data:
envoy.yaml: |
static_resources:
listeners:
- address:
socket_address:
address: 0.0.0.0
port_value: 80
filter_chains:
- filters:
- name: envoy.http_connection_manager
config:
access_log:
- name: envoy.file_access_log
config:
path: /dev/stdout
codec_type: auto
stat_prefix: ingress_http
http_filters:
- name: envoy.router
config: {}
route_config:
name: local_route
virtual_hosts:
- name: local_service
domains: ["*"]
routes:
- match:
prefix: "/"
route:
cluster: podinfo
host_rewrite: podinfo.test
timeout: 15s
retry_policy:
retry_on: "gateway-error,connect-failure,refused-stream"
num_retries: 10
per_try_timeout: 5s
clusters:
- name: podinfo
connect_timeout: 0.30s
type: strict_dns
lb_policy: round_robin
http2_protocol_options: {}
hosts:
- socket_address:
address: podinfo.test
port_value: 9898
admin:
access_log_path: /dev/null
address:
socket_address:
address: 0.0.0.0
port_value: 9999
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: ingress
namespace: test
labels:
app: ingress
spec:
replicas: 1
selector:
matchLabels:
app: ingress
strategy:
type: RollingUpdate
rollingUpdate:
maxUnavailable: 0
template:
metadata:
labels:
app: ingress
annotations:
prometheus.io/path: "/stats/prometheus"
prometheus.io/port: "9999"
prometheus.io/scrape: "true"
# dummy port to exclude ingress from mesh traffic
# only egress should go over the mesh
appmesh.k8s.aws/ports: "444"
spec:
terminationGracePeriodSeconds: 30
containers:
- name: ingress
image: "envoyproxy/envoy-alpine:d920944aed67425f91fc203774aebce9609e5d9a"
securityContext:
capabilities:
drop:
- ALL
add:
- NET_BIND_SERVICE
command:
- /usr/bin/dumb-init
- --
args:
- /usr/local/bin/envoy
- --base-id 30
- --v2-config-only
- -l
- $loglevel
- -c
- /config/envoy.yaml
ports:
- name: admin
containerPort: 9999
protocol: TCP
- name: http
containerPort: 80
protocol: TCP
- name: https
containerPort: 443
protocol: TCP
livenessProbe:
initialDelaySeconds: 5
tcpSocket:
port: admin
readinessProbe:
initialDelaySeconds: 5
tcpSocket:
port: admin
resources:
requests:
cpu: 100m
memory: 64Mi
volumeMounts:
- name: config
mountPath: /config
volumes:
- name: config
configMap:
name: ingress-config
---
kind: Service
apiVersion: v1
metadata:
name: ingress
namespace: test
spec:
selector:
app: ingress
ports:
- protocol: TCP
name: http
port: 80
targetPort: 80
- protocol: TCP
name: https
port: 443
targetPort: 443
type: LoadBalancer
---
apiVersion: appmesh.k8s.aws/v1alpha1
kind: VirtualNode
metadata:
name: ingress
namespace: test
spec:
meshName: global.appmesh-system
listeners:
- portMapping:
port: 80
protocol: http
serviceDiscovery:
dns:
hostName: ingress.test
backends:
- virtualService:
virtualServiceName: podinfo.test
artifacts/eks/appmesh-prometheus.yaml
+264
View File
@@ -0,0 +1,264 @@
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRole
metadata:
name: prometheus
labels:
app: prometheus
rules:
- apiGroups: [""]
resources:
- nodes
- services
- endpoints
- pods
- nodes/proxy
verbs: ["get", "list", "watch"]
- apiGroups: [""]
resources:
- configmaps
verbs: ["get"]
- nonResourceURLs: ["/metrics"]
verbs: ["get"]
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRoleBinding
metadata:
name: prometheus
labels:
app: prometheus
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: prometheus
subjects:
- kind: ServiceAccount
name: prometheus
namespace: appmesh-system
---
apiVersion: v1
kind: ServiceAccount
metadata:
name: prometheus
namespace: appmesh-system
labels:
app: prometheus
---
apiVersion: v1
kind: ConfigMap
metadata:
name: prometheus
namespace: appmesh-system
labels:
app: prometheus
data:
prometheus.yml: |-
global:
scrape_interval: 5s
scrape_configs:
# Scrape config for AppMesh Envoy sidecar
- job_name: 'appmesh-envoy'
metrics_path: /stats/prometheus
kubernetes_sd_configs:
- role: pod
relabel_configs:
- source_labels: [__meta_kubernetes_pod_container_name]
action: keep
regex: '^envoy$'
- source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_port]
action: replace
regex: ([^:]+)(?::\d+)?;(\d+)
replacement: ${1}:9901
target_label: __address__
- action: labelmap
regex: __meta_kubernetes_pod_label_(.+)
- source_labels: [__meta_kubernetes_namespace]
action: replace
target_label: kubernetes_namespace
- source_labels: [__meta_kubernetes_pod_name]
action: replace
target_label: kubernetes_pod_name
# Exclude high cardinality metrics
metric_relabel_configs:
- source_labels: [ cluster_name ]
regex: '(outbound|inbound|prometheus_stats).*'
action: drop
- source_labels: [ tcp_prefix ]
regex: '(outbound|inbound|prometheus_stats).*'
action: drop
- source_labels: [ listener_address ]
regex: '(.+)'
action: drop
- source_labels: [ http_conn_manager_listener_prefix ]
regex: '(.+)'
action: drop
- source_labels: [ http_conn_manager_prefix ]
regex: '(.+)'
action: drop
- source_labels: [ __name__ ]
regex: 'envoy_tls.*'
action: drop
- source_labels: [ __name__ ]
regex: 'envoy_tcp_downstream.*'
action: drop
- source_labels: [ __name__ ]
regex: 'envoy_http_(stats|admin).*'
action: drop
- source_labels: [ __name__ ]
regex: 'envoy_cluster_(lb|retry|bind|internal|max|original).*'
action: drop
# Scrape config for API servers
- job_name: 'kubernetes-apiservers'
kubernetes_sd_configs:
- role: endpoints
namespaces:
names:
- default
scheme: https
tls_config:
ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
relabel_configs:
- source_labels: [__meta_kubernetes_service_name, __meta_kubernetes_endpoint_port_name]
action: keep
regex: kubernetes;https
# Scrape config for nodes
- job_name: 'kubernetes-nodes'
scheme: https
tls_config:
ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
kubernetes_sd_configs:
- role: node
relabel_configs:
- action: labelmap
regex: __meta_kubernetes_node_label_(.+)
- target_label: __address__
replacement: kubernetes.default.svc:443
- source_labels: [__meta_kubernetes_node_name]
regex: (.+)
target_label: __metrics_path__
replacement: /api/v1/nodes/${1}/proxy/metrics
# scrape config for cAdvisor
- job_name: 'kubernetes-cadvisor'
scheme: https
tls_config:
ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
kubernetes_sd_configs:
- role: node
relabel_configs:
- action: labelmap
regex: __meta_kubernetes_node_label_(.+)
- target_label: __address__
replacement: kubernetes.default.svc:443
- source_labels: [__meta_kubernetes_node_name]
regex: (.+)
target_label: __metrics_path__
replacement: /api/v1/nodes/${1}/proxy/metrics/cadvisor
# scrape config for pods
- job_name: kubernetes-pods
kubernetes_sd_configs:
- role: pod
relabel_configs:
- action: keep
regex: true
source_labels:
- __meta_kubernetes_pod_annotation_prometheus_io_scrape
- source_labels: [ __address__ ]
regex: '.*9901.*'
action: drop
- action: replace
regex: (.+)
source_labels:
- __meta_kubernetes_pod_annotation_prometheus_io_path
target_label: __metrics_path__
- action: replace
regex: ([^:]+)(?::\d+)?;(\d+)
replacement: $1:$2
source_labels:
- __address__
- __meta_kubernetes_pod_annotation_prometheus_io_port
target_label: __address__
- action: labelmap
regex: __meta_kubernetes_pod_label_(.+)
- action: replace
source_labels:
- __meta_kubernetes_namespace
target_label: kubernetes_namespace
- action: replace
source_labels:
- __meta_kubernetes_pod_name
target_label: kubernetes_pod_name
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: prometheus
namespace: appmesh-system
labels:
app: prometheus
spec:
replicas: 1
selector:
matchLabels:
app: prometheus
template:
metadata:
labels:
app: prometheus
annotations:
version: "appmesh-v1alpha1"
spec:
serviceAccountName: prometheus
containers:
- name: prometheus
image: "docker.io/prom/prometheus:v2.7.1"
imagePullPolicy: IfNotPresent
args:
- '--storage.tsdb.retention=6h'
- '--config.file=/etc/prometheus/prometheus.yml'
ports:
- containerPort: 9090
name: http
livenessProbe:
httpGet:
path: /-/healthy
port: 9090
readinessProbe:
httpGet:
path: /-/ready
port: 9090
resources:
requests:
cpu: 10m
memory: 128Mi
volumeMounts:
- name: config-volume
mountPath: /etc/prometheus
volumes:
- name: config-volume
configMap:
name: prometheus
---
apiVersion: v1
kind: Service
metadata:
name: prometheus
namespace: appmesh-system
labels:
name: prometheus
spec:
selector:
app: prometheus
ports:
- name: http
protocol: TCP
port: 9090
artifacts/namespaces/test.yaml
+1
View File
@@ -4,3 +4,4 @@ metadata:
name: test
labels:
istio-injection: enabled
appmesh.k8s.aws/sidecarInjectorWebhook: enabled
charts/flagger/Chart.yaml
+1 -1
View File
@@ -6,7 +6,7 @@ kubeVersion: ">=1.11.0-0"
engine: gotpl
description: Flagger is a Kubernetes operator that automates the promotion of canary deployments using Istio routing for traffic shifting and Prometheus metrics for canary analysis.
home: https://docs.flagger.app
icon: https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/logo/flagger-icon.png
icon: https://raw.githubusercontent.com/weaveworks/flagger/master/docs/logo/flagger-icon.png
sources:
- https://github.com/weaveworks/flagger
maintainers:
charts/grafana/Chart.yaml
+1 -1
View File
@@ -3,7 +3,7 @@ name: grafana
version: 1.0.0
appVersion: 5.4.3
description: Grafana dashboards for monitoring Flagger canary deployments
icon: https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/logo/flagger-icon.png
icon: https://raw.githubusercontent.com/weaveworks/flagger/master/docs/logo/flagger-icon.png
home: https://flagger.app
sources:
- https://github.com/weaveworks/flagger
charts/grafana/README.md
+1 -1
View File
@@ -2,7 +2,7 @@
Grafana dashboards for monitoring progressive deployments powered by Istio, Prometheus and Flagger.
![flagger-grafana](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/grafana-canary-analysis.png)
![flagger-grafana](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/grafana-canary-analysis.png)
## Prerequisites
charts/loadtester/Chart.yaml
+1 -1
View File
@@ -6,7 +6,7 @@ kubeVersion: ">=1.11.0-0"
engine: gotpl
description: Flagger's load testing services based on rakyll/hey that generates traffic during canary analysis when configured as a webhook.
home: https://docs.flagger.app
icon: https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/logo/flagger-icon.png
icon: https://raw.githubusercontent.com/weaveworks/flagger/master/docs/logo/flagger-icon.png
sources:
- https://github.com/weaveworks/flagger
maintainers:
docs/diagrams/flagger-canary-overview.png
BIN
View File
Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 207 KiB

After

Width:  |  Height:  |  Size: 44 KiB

docs/diagrams/flagger-gitops-aws.png
BIN
View File
Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 32 KiB

docs/gitbook/README.md
+2 -2
View File
@@ -1,5 +1,5 @@
---
description: Flagger is an Istio progressive delivery Kubernetes operator
description: Flagger is a progressive delivery Kubernetes operator
---
# Introduction
@@ -13,7 +13,7 @@ Flagger implements a control loop that gradually shifts traffic to the canary wh
indicators like HTTP requests success rate, requests average duration and pods health.
Based on analysis of the **KPIs** a canary is promoted or aborted, and the analysis result is published to **Slack**.
![Flagger overview diagram](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-canary-overview.png)
![Flagger overview diagram](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-canary-overview.png)
Flagger can be configured with Kubernetes custom resources and is compatible with
any CI/CD solutions made for Kubernetes. Since Flagger is declarative and reacts to Kubernetes events,
docs/gitbook/SUMMARY.md
+3 -2
View File
@@ -11,8 +11,9 @@
## Usage
* [Canary Deployments](usage/progressive-delivery.md)
* [A/B Testing](usage/ab-testing.md)
* [Istio Canary Deployments](usage/progressive-delivery.md)
* [Istio A/B Testing](usage/ab-testing.md)
* [App Mesh Canary Deployments](usage/appmesh-progressive-delivery.md)
* [Monitoring](usage/monitoring.md)
* [Alerting](usage/alerting.md)
docs/gitbook/how-it-works.md
+4 -4
View File
@@ -4,7 +4,7 @@
a horizontal pod autoscaler \(HPA\) and creates a series of objects
\(Kubernetes deployments, ClusterIP services and Istio virtual services\) to drive the canary analysis and promotion.
![Flagger Canary Process](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-canary-hpa.png)
![Flagger Canary Process](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-canary-hpa.png)
### Canary Custom Resource
@@ -251,7 +251,7 @@ and for backend HTTP APIs that are accessible only from inside the mesh.
### Canary Stages
![Flagger Canary Stages](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-canary-steps.png)
![Flagger Canary Stages](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-canary-steps.png)
A canary deployment is triggered by changes in any of the following objects:
@@ -579,12 +579,12 @@ Flagger metric checks will fail with "no values found for metric istio_requests_
Flagger comes with a load testing service based on [rakyll/hey](https://github.com/rakyll/hey)
that generates traffic during analysis when configured as a webhook.
![Flagger Load Testing Webhook](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-load-testing.png)
![Flagger Load Testing Webhook](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-load-testing.png)
First you need to deploy the load test runner in a namespace with Istio sidecar injection enabled:
```bash
export REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl -n test apply -f ${REPO}/artifacts/loadtester/deployment.yaml
kubectl -n test apply -f ${REPO}/artifacts/loadtester/service.yaml
docs/gitbook/install/flagger-install-on-eks-appmesh.md
+15 -8
View File
@@ -125,12 +125,6 @@ Install the App Mesh sidecar injector in the `appmesh-system` namespace:
The above script generates a certificate signed by Kubernetes CA,
registers the App Mesh mutating webhook and deploys the injector.
Deploy Prometheus in the `appmesh-system` namespace:
```bash
kubectl apply -f ./prometheus
```
Create a mesh called global in the `appmesh-system` namespace:
```bash
@@ -145,7 +139,20 @@ kubectl -n appmesh-system describe mesh
Status:
Mesh Condition:
Status: True
Type: Active
Type: MeshActive
```
### Install Prometheus
In order to expose the App Mesh metrics to Flagger,
you'll need to use Prometheus to scrapes the Envoy sidecars.
Deploy Prometheus in the `appmesh-system` namespace:
```bash
REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/eks/appmesh-prometheus.yaml
```
### Install Flagger and Grafana
@@ -162,7 +169,7 @@ Deploy Flagger in the _**appmesh-system**_ namespace:
helm upgrade -i flagger flagger/flagger \
--namespace=appmesh-system \
--set meshProvider=appmesh \
--set metricsServer=http://prometheus.appmesh:9090
--set metricsServer=http://prometheus.appmesh-system:9090
```
You can install Flagger in any namespace as long as it can talk to the Istio Prometheus service on port 9090.
docs/gitbook/install/flagger-install-on-google-cloud.md
+4 -4
View File
@@ -2,7 +2,7 @@
This guide walks you through setting up Flagger and Istio on Google Kubernetes Engine.
![GKE Cluster Overview](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-gke-istio.png)
![GKE Cluster Overview](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-gke-istio.png)
### Prerequisites
@@ -208,12 +208,12 @@ stable/cert-manager
### Istio Gateway TLS setup
![Istio Let's Encrypt](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/istio-cert-manager-gke.png)
![Istio Let's Encrypt](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/istio-cert-manager-gke.png)
Create a generic Istio Gateway to expose services outside the mesh on HTTPS:
```bash
REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/gke/istio-gateway.yaml
```
@@ -332,7 +332,7 @@ Because Flagger uses the Istio HTTP metrics to run the canary analysis you have
Prometheus configuration that's similar to the one that comes with the official Istio Helm chart.
```bash
REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/gke/istio-prometheus.yaml
```
docs/gitbook/install/flagger-install-on-kubernetes.md
+1 -1
View File
@@ -132,7 +132,7 @@ helm upgrade -i flagger-loadtester flagger/loadtester \
Deploy with kubectl:
```bash
export REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl -n test apply -f ${REPO}/artifacts/loadtester/deployment.yaml
kubectl -n test apply -f ${REPO}/artifacts/loadtester/service.yaml
docs/gitbook/tutorials/canary-helm-gitops.md
+6 -6
View File
@@ -29,7 +29,7 @@ You can find the chart source [here](https://github.com/stefanprodan/flagger/tre
Create a test namespace with Istio sidecar injection enabled:
```bash
export REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/namespaces/test.yaml
```
@@ -79,7 +79,7 @@ Flagger will route all traffic to the primary pods and scale to zero the `fronte
Open your browser and navigate to the frontend URL:
![Podinfo Frontend](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/demo-frontend.png)
![Podinfo Frontend](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/demo-frontend.png)
Now let's install the `backend` release without exposing it outside the mesh:
@@ -104,7 +104,7 @@ frontend Initialized 0 2019-02-12T17:50:50Z
Click on the ping button in the `frontend` UI to trigger a HTTP POST request
that will reach the `backend` app:
![Jaeger Tracing](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/demo-frontend-jaeger.png)
![Jaeger Tracing](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/demo-frontend-jaeger.png)
We'll use the `/echo` endpoint (same as the one the ping button calls)
to generate load on both apps during a canary deployment.
@@ -155,7 +155,7 @@ You can monitor the canary deployment with Grafana. Open the Flagger dashboard,
select `test` from the namespace dropdown, `frontend-primary` from the primary dropdown and `frontend` from the
canary dropdown.
![Flagger Grafana Dashboard](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/demo-frontend-dashboard.png)
![Flagger Grafana Dashboard](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/demo-frontend-dashboard.png)
Now trigger a canary deployment for the `backend` app, but this time you'll change a value in the configmap:
@@ -213,7 +213,7 @@ Copying backend.test template spec to backend-primary.test
Promotion completed! Scaling down backend.test
```
![Flagger Grafana Dashboard](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/demo-backend-dashboard.png)
![Flagger Grafana Dashboard](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/demo-backend-dashboard.png)
If 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.
@@ -237,7 +237,7 @@ In the [GitOps model](https://www.weave.works/technologies/gitops/),
any change to production must be committed in source control
prior to being applied on the cluster. This way rollback and audit logs are provided by Git.
![Helm GitOps Canary Deployment](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-flux-gitops.png)
![Helm GitOps Canary Deployment](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-flux-gitops.png)
In order to apply the GitOps pipeline model to Flagger canary deployments you'll need
a Git repository with your workloads definitions in YAML format,
docs/gitbook/usage/ab-testing.md
+4 -4
View File
@@ -1,17 +1,17 @@
# A/B Testing
# Istio A/B Testing
This guide shows you how to automate A/B testing with Flagger.
This guide shows you how to automate A/B testing with Istio and Flagger.
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](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/diagrams/flagger-abtest-steps.png)
![Flagger A/B Testing Stages](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/diagrams/flagger-abtest-steps.png)
Create a test namespace with Istio sidecar injection enabled:
```bash
export REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/namespaces/test.yaml
```
docs/gitbook/usage/alerting.md
+2 -2
View File
@@ -15,12 +15,12 @@ helm upgrade -i flagger flagger/flagger \
Once configured with a Slack incoming **webhook**, Flagger will post messages when a canary deployment
has been initialised, when a new revision has been detected and if the canary analysis failed or succeeded.
![flagger-slack](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/slack-canary-notifications.png)
![flagger-slack](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/slack-canary-notifications.png)
A canary deployment will be rolled back if the progress deadline exceeded or if the analysis reached the
maximum number of failed checks:
![flagger-slack-errors](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/slack-canary-failed.png)
![flagger-slack-errors](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/slack-canary-failed.png)
### Prometheus Alert Manager
docs/gitbook/usage/appmesh-progressive-delivery.md
+279
View File
@@ -0,0 +1,279 @@
# App Mesh Canary Deployments
This guide shows you how to use App Mesh and Flagger to automate canary deployments.
You'll need an EKS cluster configured with App Mesh, you can find the install guide
[here](https://docs.flagger.app/install/flagger-install-on-eks-appmesh).
### Bootstrap
Flagger takes a Kubernetes deployment and optionally a horizontal pod autoscaler (HPA),
then creates a series of objects (Kubernetes deployments, ClusterIP services, App Mesh virtual nodes and services).
These objects expose the application on the mesh and drive the canary analysis and promotion.
The only App Mesh object you need to creat by yourself is the mesh resource.
Create a mesh called `global` in the `appmesh-system` namespace:
```bash
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/appmesh/global-mesh.yaml
```
Create a test namespace with App Mesh sidecar injection enabled:
```bash
kubectl apply -f ${REPO}/artifacts/namespaces/test.yaml
```
Create a deployment and a horizontal pod autoscaler:
```bash
kubectl apply -f ${REPO}/artifacts/appmesh/deployment.yaml
kubectl apply -f ${REPO}/artifacts/appmesh/hpa.yaml
```
Deploy the load testing service to generate traffic during the canary analysis:
```bash
helm upgrade -i flagger-loadtester flagger/loadtester \
--namespace=test \
--set meshName=global.appmesh-system \
--set backends[0]=podinfo.test
```
Create a canary custom resource:
```yaml
apiVersion: flagger.app/v1alpha3
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/v2beta1
kind: HorizontalPodAutoscaler
name: podinfo
service:
# container port
port: 9898
# App Mesh reference
meshName: global.appmesh-system
# App Mesh egress (optional)
backends:
- backend.test
# define the canary analysis timing and KPIs
canaryAnalysis:
# schedule interval (default 60s)
interval: 10s
# max number of failed metric checks before rollback
threshold: 10
# max traffic percentage routed to canary
# percentage (0-100)
maxWeight: 50
# canary increment step
# percentage (0-100)
stepWeight: 5
# App Mesh Prometheus checks
metrics:
- name: envoy_cluster_upstream_rq
# minimum req success rate (non 5xx responses)
# percentage (0-100)
threshold: 99
interval: 1m
# external checks (optional)
webhooks:
- name: load-test
url: http://flagger-loadtester.test/
timeout: 5s
metadata:
cmd: "hey -z 1m -q 10 -c 2 http://podinfo.test:9898/"
```
Save the above resource as podinfo-canary.yaml and then apply it:
```bash
kubectl apply -f ./podinfo-canary.yaml
```
After a couple of seconds Flagger will create the canary objects:
```bash
# applied
deployment.apps/podinfo
horizontalpodautoscaler.autoscaling/podinfo
canary.flagger.app/podinfo
# generated Kubernetes objects
deployment.apps/podinfo-primary
horizontalpodautoscaler.autoscaling/podinfo-primary
service/podinfo
service/podinfo-canary
service/podinfo-primary
# generated App Mesh objects
virtualnode.appmesh.k8s.aws/podinfo
virtualnode.appmesh.k8s.aws/podinfo-canary
virtualnode.appmesh.k8s.aws/podinfo-primary
virtualservice.appmesh.k8s.aws/podinfo.test
```
The App Mesh specific settings are:
```yaml
service:
port: 9898
meshName: global.appmesh-system
backends:
- backend1.test
- backend2.test
```
App Mesh blocks all egress traffic by default. If your application needs to call another service, you have to create an
App Mesh virtual service for it and add the virtual service name to the backend list.
### Setup App Mesh ingress (optional)
In order to expose the podinfo app outside the mesh you'll be using an Envoy ingress and an AWS classic load balancer.
The ingress binds to an internet domain and forwards the calls into the mesh through the App Mesh sidecar.
If podinfo becomes unavailable due to a HPA downscaling or a node restart,
the ingress will retry the calls for a short period of time.
Deploy the ingress and the AWS ELB service:
```bash
kubectl apply -f ${REPO}/artifacts/appmesh/ingress.yaml
```
Find the ingress public address:
```bash
kubectl -n test describe svc/ingress | grep Ingress
LoadBalancer Ingress: yyy-xx.us-west-2.elb.amazonaws.com
```
Open your browser and navigate to the ingress address to access podinfo UI.
### Automated canary promotion
Trigger a canary deployment by updating the container image:
```bash
kubectl -n test set image deployment/podinfo \
podinfod=quay.io/stefanprodan/podinfo:1.4.1
```
Flagger detects that the deployment revision changed and starts a new rollout:
```text
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.
During the analysis the canary’s progress can be monitored with Grafana:
![App Mesh Canary Dashboard](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/flagger-grafana-appmesh.png)
You can monitor all canaries with:
```bash
watch kubectl get canaries --all-namespaces
NAMESPACE NAME STATUS WEIGHT LASTTRANSITIONTIME
test podinfo Progressing 15 2019-03-16T14:05:07Z
prod frontend Succeeded 0 2019-03-15T16:15:07Z
prod backend Failed 0 2019-03-14T17:05:07Z
```
If you’ve enabled the Slack notifications, you should receive the following messages:
![flagger-slack](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/slack-canary-notifications.png)
### Automated rollback
During the canary analysis you can generate HTTP 500 errors to test if Flagger pauses the rollout.
Trigger a canary deployment:
```bash
kubectl -n test set image deployment/podinfo \
podinfod=quay.io/stefanprodan/podinfo:1.4.2
```
Exec into the load tester pod with:
```bash
kubectl -n test exec -it flagger-loadtester-xx-xx sh
```
Generate HTTP 500 errors:
```bash
hey -z 1m -c 5 -q 5 http://podinfo.test:9898/status/500
```
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.
```text
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 success rate 69.17% < 99%
Normal Synced 2m flagger Halt podinfo.test advancement success rate 61.39% < 99%
Normal Synced 2m flagger Halt podinfo.test advancement success rate 55.06% < 99%
Normal Synced 2m flagger Halt podinfo.test advancement success rate 47.00% < 99%
Normal Synced 2m flagger (combined from similar events): Halt podinfo.test advancement success rate 38.08% < 99%
Warning Synced 1m flagger Rolling back podinfo.test failed checks threshold reached 10
Warning Synced 1m flagger Canary failed! Scaling down podinfo.test
```
If you’ve enabled the Slack notifications, you’ll receive a message if the progress deadline is exceeded,
or if the analysis reached the maximum number of failed checks:
![flagger-slack-errors](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/slack-canary-failed.png)
docs/gitbook/usage/monitoring.md
+1 -1
View File
@@ -14,7 +14,7 @@ helm upgrade -i flagger-grafana flagger/grafana \
The dashboard shows the RED and USE metrics for the primary and canary workloads:
![canary dashboard](https://raw.githubusercontent.com/stefanprodan/flagger/master/docs/screens/grafana-canary-analysis.png)
![canary dashboard](https://raw.githubusercontent.com/weaveworks/flagger/master/docs/screens/grafana-canary-analysis.png)
### Logging
docs/gitbook/usage/progressive-delivery.md
+2 -2
View File
@@ -1,11 +1,11 @@
# Canary Deployments
# Istio Canary Deployments
This guide shows you how to use Istio and Flagger to automate canary deployments.
Create a test namespace with Istio sidecar injection enabled:
```bash
export REPO=https://raw.githubusercontent.com/stefanprodan/flagger/master
export REPO=https://raw.githubusercontent.com/weaveworks/flagger/master
kubectl apply -f ${REPO}/artifacts/namespaces/test.yaml
```
pkg/controller/observer.go
+7 -5
View File
@@ -114,11 +114,12 @@ func (c *CanaryObserver) GetEnvoySuccessRate(name string, namespace string, metr
var rate *float64
querySt := url.QueryEscape(`sum(rate(` +
metric + `{kubernetes_namespace="` +
namespace + `",app="` +
name + `",envoy_response_code!~"5.*"}[1m])) / sum(rate(` +
namespace + `",kubernetes_pod_name=~"` +
name + `-[0-9a-zA-Z]+(-[0-9a-zA-Z]+)",envoy_response_code!~"5.*"}[` +
interval + `])) / sum(rate(` +
metric + `{kubernetes_namespace="` +
namespace + `",app="` +
name + `"}[` +
namespace + `",kubernetes_pod_name=~"` +
name + `-[0-9a-zA-Z]+(-[0-9a-zA-Z]+)"}[` +
interval + `])) * 100 `)
result, err := c.queryMetric(querySt)
if err != nil {
@@ -152,7 +153,8 @@ func (c *CanaryObserver) GetDeploymentCounter(name string, namespace string, met
querySt := url.QueryEscape(`sum(rate(` +
metric + `{reporter="destination",destination_workload_namespace=~"` +
namespace + `",destination_workload=~"` +
name + `",response_code!~"5.*"}[1m])) / sum(rate(` +
name + `",response_code!~"5.*"}[` +
interval + `])) / sum(rate(` +
metric + `{reporter="destination",destination_workload_namespace=~"` +
namespace + `",destination_workload=~"` +
name + `"}[` +
pkg/controller/scheduler.go
+1 -1
View File
@@ -25,7 +25,7 @@ func (c *Controller) scheduleCanaries() {
current[name] = fmt.Sprintf("%s.%s", canary.Spec.TargetRef.Name, canary.Namespace)
job, exists := c.jobs[name]
// schedule new job for exsiting job with different analysisInterval or non-existing job
// schedule new job for existing job with different analysis interval or non-existing job
if (exists && job.GetCanaryAnalysisInterval() != canary.GetAnalysisInterval()) || !exists {
if exists {
job.Stop()
pkg/router/appmesh.go
+11 -11
View File
@@ -37,21 +37,21 @@ func (ar *AppMeshRouter) Sync(canary *flaggerv1.Canary) error {
// sync virtual node e.g. app-namespace
// DNS app.namespace
err := ar.syncVirtualNode(canary, fmt.Sprintf("%s-%s", targetName, canary.Namespace), primaryHost)
err := ar.syncVirtualNode(canary, targetName, primaryHost)
if err != nil {
return err
}
// sync virtual node e.g. app-primary-namespace
// DNS app-primary.namespace
err = ar.syncVirtualNode(canary, fmt.Sprintf("%s-%s", primaryName, canary.Namespace), primaryHost)
err = ar.syncVirtualNode(canary, primaryName, primaryHost)
if err != nil {
return err
}
// sync virtual node e.g. app-canary-namespace
// DNS app-canary.namespace
err = ar.syncVirtualNode(canary, fmt.Sprintf("%s-%s", canaryName, canary.Namespace), canaryHost)
err = ar.syncVirtualNode(canary, canaryName, canaryHost)
if err != nil {
return err
}
@@ -150,8 +150,8 @@ func (ar *AppMeshRouter) syncVirtualNode(canary *flaggerv1.Canary, name string,
// syncVirtualService creates or updates a virtual service
func (ar *AppMeshRouter) syncVirtualService(canary *flaggerv1.Canary, name string) error {
targetName := canary.Spec.TargetRef.Name
canaryVirtualNode := fmt.Sprintf("%s-canary-%s", targetName, canary.Namespace)
primaryVirtualNode := fmt.Sprintf("%s-primary-%s", targetName, canary.Namespace)
canaryVirtualNode := fmt.Sprintf("%s-canary", targetName)
primaryVirtualNode := fmt.Sprintf("%s-primary", targetName)
// App Mesh supports only URI prefix
routePrefix := "/"
@@ -164,11 +164,11 @@ func (ar *AppMeshRouter) syncVirtualService(canary *flaggerv1.Canary, name strin
vsSpec := &appmeshv1alpha1.VirtualServiceSpec{
MeshName: canary.Spec.Service.MeshName,
VirtualRouter: &appmeshv1alpha1.VirtualRouter{
Name: fmt.Sprintf("%s-%s-router", targetName, canary.Namespace),
Name: fmt.Sprintf("%s-router", targetName),
},
Routes: []appmeshv1alpha1.Route{
{
Name: fmt.Sprintf("%s-%s-route", targetName, canary.Namespace),
Name: fmt.Sprintf("%s-route", targetName),
Http: appmeshv1alpha1.HttpRoute{
Match: appmeshv1alpha1.HttpRouteMatch{
Prefix: routePrefix,
@@ -265,10 +265,10 @@ func (ar *AppMeshRouter) GetRoutes(canary *flaggerv1.Canary) (
targets := vs.Spec.Routes[0].Http.Action.WeightedTargets
for _, t := range targets {
if t.VirtualNodeName == fmt.Sprintf("%s-canary-%s", targetName, canary.Namespace) {
if t.VirtualNodeName == fmt.Sprintf("%s-canary", targetName) {
canaryWeight = int(t.Weight)
}
if t.VirtualNodeName == fmt.Sprintf("%s-primary-%s", targetName, canary.Namespace) {
if t.VirtualNodeName == fmt.Sprintf("%s-primary", targetName) {
primaryWeight = int(t.Weight)
}
}
@@ -301,11 +301,11 @@ func (ar *AppMeshRouter) SetRoutes(
vsClone.Spec.Routes[0].Http.Action = appmeshv1alpha1.HttpRouteAction{
WeightedTargets: []appmeshv1alpha1.WeightedTarget{
{
VirtualNodeName: fmt.Sprintf("%s-canary-%s", targetName, canary.Namespace),
VirtualNodeName: fmt.Sprintf("%s-canary", targetName),
Weight: int64(canaryWeight),
},
{
VirtualNodeName: fmt.Sprintf("%s-primary-%s", targetName, canary.Namespace),
VirtualNodeName: fmt.Sprintf("%s-primary", targetName),
Weight: int64(primaryWeight),
},
},
pkg/router/appmesh_test.go
+2 -2
View File
@@ -38,7 +38,7 @@ func TestAppmeshRouter_Sync(t *testing.T) {
}
// check virtual node
vnName := fmt.Sprintf("%s-%s", mocks.appmeshCanary.Spec.TargetRef.Name, mocks.appmeshCanary.Namespace)
vnName := mocks.appmeshCanary.Spec.TargetRef.Name
vn, err := router.appmeshClient.AppmeshV1alpha1().VirtualNodes("default").Get(vnName, metav1.GetOptions{})
if err != nil {
t.Fatal(err.Error())
@@ -72,7 +72,7 @@ func TestAppmeshRouter_Sync(t *testing.T) {
}
// verify
vnCanaryName := fmt.Sprintf("%s-canary-%s", mocks.appmeshCanary.Spec.TargetRef.Name, mocks.appmeshCanary.Namespace)
vnCanaryName := fmt.Sprintf("%s-canary", mocks.appmeshCanary.Spec.TargetRef.Name)
vnCanary, err := router.appmeshClient.AppmeshV1alpha1().VirtualNodes("default").Get(vnCanaryName, metav1.GetOptions{})
if err != nil {
t.Fatal(err.Error())