Scenario 01 - Considering microservice replication across multiple locations with replication within each cluster
Diagram
Testing Steps
- Install ISTIO - Deploy istio control plane in each cluster. (NOTE - For testing use common root CA). Reference
- Deploy the application is both the clusters as shown in the above figure. (NOTE - Make sure that the istio sidecars are injected to the Service pods) For example use - server- httpbin, client - sleep
- Configure DNS - To provide resolution of service from remote clusters, istio uses its own DNS called istiocoredns which provides the resolution of remote istio services
NOTE - In order to utilize istiocoredns, Kubernetes DNS' must be configured to stub a domain in a specific format
3. Use configMap to add a stub domain to the Kubernetes's DNS.
kind: ConfigMap
data:
Corefile: |
.:53 {
log
errors
health
kubernetes cluster.local in-addr.arpa ip6.arpa {
pods insecure
upstream
fallthrough in-addr.arpa ip6.arpa
}
prometheus :9153
proxy . /etc/resolv.conf
cache 30
loop
reload
loadbalance
}
global:53 {
errors
cache 30
proxy . 10.43.57.78
}
4. Add Istio service entry with details of the remote servers (server service 03 and server service 04) to the cluster where the client is running (For ISTIO multi-cluster communication usage of SNI ports is mandatory at both ends on istio-ingressgateway)
apiVersion: networking.istio.io/v1alpha3
kind: ServiceEntry
metadata:
name: servicename04-bar
spec:
hosts:
# template for the remote service name - <servicename.namespace.global>
- serverservice04.bar.global
# Treat remote cluster services as part of the service mesh
# as all clusters in the service mesh share the same root of trust.
location: MESH_INTERNAL
ports:
- name: http1
number: 8000
protocol: http
resolution: DNS
addresses:
# the IP address to which httpbin.bar.global will resolve to
# must be unique for each remote service, within a given cluster.
# This address need not be routable. Traffic for this IP will be captured
# by the sidecar and routed appropriately.
- 240.0.0.2
endpoints:
# This is the routable address of the ingress gateway in cluster2 that
# sits in front of sleep.foo service. Traffic from the sidecar will be
# routed to this address.
- address: 172.25.55.50
ports:
http1: 15443
5. Create istio virtual service on the client cluster with all the destination server that it wants to connect. The API calls from the client can be load balanced by assigned weight to each server.
apiVersion: networking.istio.io/v1alpha3
kind: VirtualService
metadata:
name: reviews-route
spec:
hosts:
- service01.bar.svc.cluster.local
http:
- match:
- uri:
prefix: "/headers"
route:
- destination:
host: serviceserver04.bar.global
port:
number: 8000
weight: 50
- destination:
host: serviceserver01.bar.svc.cluster.local
port:
number: 8000
weight: 25
- destination:
host: serviceserver02.bar.svc.cluster.local
port:
number: 8000
weight: 25
6. Verify that client pod is sending requests to servers in the order of assigned weight by running the below bash script in the client service pod
#!/bin/bash COUNTER=0 while [ $COUNTER -lt 10 ]; do curl -v service01.bar.svc.cluster.local/headers sleep 2 done
NOTE - Although the service running in external cluster can be accessed by following the steps above, the failover mechanism is not yet supported by ISTIO.
IN PROGRESS......