Aws-load-balancer-controller: Health Checks do not work if using multiple pods on routes

[owenhaynes]

e.g

• host: test.com
  http:
  paths:
  
  
  
  • path: "/api/"
    
    backend:
    
    serviceName: api
    
    servicePort: 80
  
  
  • path: "/"
    
    backend:
    
    serviceName: frontend
    
    servicePort: 80
  
  

You can only set one health-check on the ingress so if service api does not have the same route for a health check as frontend you get a unable system

[owenhaynes]

Maybe do what the google ingress controller does and use the health checks of the pods?

[bigkraig]

Do you have any ideas of how you would like to see it configured?

[gauravarora]

@bigkraig I believe annotations on the service would be the easiest way.

[seandrucker]

All annotations to configure a listener should be done on the service the k8s way.

See SSL support on AWS here: https://kubernetes.io/docs/concepts/services-networking/service/#ssl-support-on-aws

So, just move all the following to the service definition, and if present, override anything listed in the ingress:

alb.ingress.kubernetes.io/backend-protocol
alb.ingress.kubernetes.io/certificate-arn
alb.ingress.kubernetes.io/healthcheck-interval-seconds
alb.ingress.kubernetes.io/healthcheck-path
alb.ingress.kubernetes.io/healthcheck-port
alb.ingress.kubernetes.io/healthcheck-protocol
alb.ingress.kubernetes.io/healthcheck-timeout-seconds
alb.ingress.kubernetes.io/healthy-threshold-count
alb.ingress.kubernetes.io/unhealthy-threshold-count
alb.ingress.kubernetes.io/listen-ports
alb.ingress.kubernetes.io/security-groups
alb.ingress.kubernetes.io/subnets
alb.ingress.kubernetes.io/successCodes
alb.ingress.kubernetes.io/tags

[bigkraig]

@sundryp I've got this covered for the annotations that would effect a TG. Some of the other annotations in your list didn't apply on service resources.

@owenhaynes hopefully this helps!

alb.ingress.kubernetes.io/backend-protocol
alb.ingress.kubernetes.io/healthcheck-interval-seconds
alb.ingress.kubernetes.io/healthcheck-path
alb.ingress.kubernetes.io/healthcheck-port
alb.ingress.kubernetes.io/healthcheck-protocol
alb.ingress.kubernetes.io/healthcheck-timeout-seconds
alb.ingress.kubernetes.io/healthy-threshold-count
alb.ingress.kubernetes.io/unhealthy-threshold-count
alb.ingress.kubernetes.io/successCodes

[pperzyna]

@joshrosso
Can we expect a solution?

[bigkraig]

Supported as of #423, please use 1.0-beta.2 or later. thanks!

[AdeOpe]

Works brilliantly
All i needed to do was move the following from my ingress to my Service definition of the app

alb.ingress.kubernetes.io/healthcheck-path: "/svc/status"
alb.ingress.kubernetes.io/successCodes: "200"