How do I run traefik behind Kubernetes on Google Container Engine? - kubernetes

Question

So I have Traefik "running" on Kubernetes:
apiVersion: v1
kind: Service
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: '{"kind":"Service","apiVersion":"v1","metadata":{"name":"traefik","namespace":"kube-system","creationTimestamp":null,"labels":{"k8s-app":"traefik-ingress-lb"}},"spec":{"ports":[{"name":"http","port":80,"targetPort":80},{"name":"https","port":443,"targetPort":443}],"selector":{"k8s-app":"traefik-ingress-lb"},"type":"LoadBalancer"},"status":{"loadBalancer":{}}}'
creationTimestamp: 2016-11-30T23:15:49Z
labels:
k8s-app: traefik-ingress-lb
name: traefik
namespace: kube-system
resourceVersion: "9672"
selfLink: /api/v1/namespaces/kube-system/services/traefik
uid: ee07b957-b752-11e6-88fa-42010af00083
spec:
clusterIP: 10.11.251.200
ports:
- name: http
nodePort: 30297
port: 80
protocol: TCP
targetPort: 80
- name: https
nodePort: 30247
port: 443
protocol: TCP
targetPort: 443
selector:
k8s-app: traefik-ingress-lb
sessionAffinity: None
type: LoadBalancer
status:
loadBalancer:
ingress:
- ip: # IP THAT IS ALLOCATED BY k8s BUT NOT ASSIGNED
---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "1"
kubectl.kubernetes.io/last-applied-configuration: '###'
creationTimestamp: 2016-11-30T22:59:07Z
generation: 2
labels:
k8s-app: traefik-ingress-lb
name: traefik-ingress-controller
namespace: kube-system
resourceVersion: "23438"
selfLink: /apis/extensions/v1beta1/namespaces/kube-system/deployments/traefik-ingress-controller
uid: 9919ff46-b750-11e6-88fa-42010af00083
spec:
replicas: 1
selector:
matchLabels:
k8s-app: traefik-ingress-lb
strategy:
rollingUpdate:
maxSurge: 1
maxUnavailable: 1
type: RollingUpdate
template:
metadata:
creationTimestamp: null
labels:
k8s-app: traefik-ingress-lb
name: traefik-ingress-lb
version: v1.1
spec:
containers:
- args:
- --web
- --kubernetes
- --configFile=/etc/config/traefik.toml
- --logLevel=DEBUG
image: gcr.io/myproject/traefik
imagePullPolicy: Always
name: traefik-ingress-lb
ports:
- containerPort: 80
hostPort: 80
protocol: TCP
- containerPort: 8080
protocol: TCP
resources:
limits:
cpu: 200m
memory: 30Mi
requests:
cpu: 100m
memory: 20Mi
terminationMessagePath: /dev/termination-log
volumeMounts:
- mountPath: /etc/config
name: config-volume
- mountPath: /etc/traefik
name: traefik-volume
dnsPolicy: ClusterFirst
restartPolicy: Always
securityContext: {}
terminationGracePeriodSeconds: 60
volumes:
- configMap:
defaultMode: 420
name: traefik-config
name: config-volume
- emptyDir: {}
name: traefik-volume
status:
observedGeneration: 2
replicas: 1
unavailableReplicas: 1
updatedReplicas: 1
My problem is that the external IP assigned by Kubernetes does not actually forward to Traefik; in fact, the IP assigned does not even show up in my Google Cloud Platform console. How do I get Traefik working with a load-balancer on Google Container Engine?

Answer 1

If you do
kubectl describe svc/traefik
Do the endpoints match the IPs from:
kubectl get po -lk8s-app=traefik-ingress-lb -o wide
What happens when you do hit the LB IP? Does it load indefinitely, or do you get a different service?

Answer 2

There was a new fork of Traefik that supposedly fixes some kubernetes issues. However, I did further reading and talking about Traefik and uncovered some (recent) advisories of potential instability. While this is to be expected in new software, I decided to switch to NGINX to handle my reverse proxy. It's been working wonderfully, so I'm going to go ahead and close this question.