ingress-gce VS application-gateway-kubernetes-ingress

You would first need to ensure you have an Ingress controller running. There are several available, such as nginx, Traefik, or the one provided by a cloud provider like GKE's Ingress-GCE or AWS's ALB Ingress Controller.

ingress-gce (Google Cloud only)

Yes. Google's ingress controller talks to the GCP API and spins up an HTTP LB for you. The AWS LB Controller handles provisioning ALB's for you. And ... I guess other clouds exist too but these are the ones I am familiar with.

At the end of the blog entry where I posted about how to fix a 502 error returned by GKE Ingress, I described an issue regarding health check configuration that has been opened for over three years. The one I mentioned here to support multiple addresses it's been opened for over four years. I think GKE Ingress is a remarkable resource. It can help you quickly release an application using K8S and cloud-native features wrapped in abstracted manifests, but it seems a bit left aside in some aspects.

There is a considerable debate on this very topic in the issue Ingress Healthcheck Configuration on kubernetes/ingress-gce repository. Looking through issues on GitHub is an excellent way to understand how an application or service is evolving. It's been more than three years, and it's not resolved yet.

Google supports its own ingress controller, GCE, which you can use in place of an nginx controller. However, it has some drawbacks, for instance, lack of rewrite rules support, at least at the moment of writing.

That ALB Controller is for AWS specifically. For Google Cloud you can use https://github.com/kubernetes/ingress-gce

Unfortunately in some cases it works poorly. For more insights, read this - https://github.com/Azure/application-gateway-kubernetes-ingress/issues/1124

My main concern is about zero-downtime deployments - https://github.com/Azure/application-gateway-kubernetes-ingress/blob/master/docs/how-tos/minimize-downtime-during-deployments.md

However, you better read and understand some specific behaviors of AGIC which are caused by its architecture - https://github.com/Azure/application-gateway-kubernetes-ingress/blob/master/docs/how-tos/minimize-downtime-during-deployments.md & https://github.com/Azure/application-gateway-kubernetes-ingress/issues/1124

application-gateway-kubernetes-ingress (Azure only)

Probably something like this https://azure.github.io/application-gateway-kubernetes-ingress/

AKS Application Gateway Ingress Controller is an ingress controller that configures the Azure Application Gateway.

AGIC has a lack of annotations that I find nginx+ offers like server snippets and custom headers. granted you can write re-write rules for AGIC with the help of the portal, it resets everything if one pod fails and this is a major showstopper for me. More info about the exact issue I'm talking about is in this issue. Apart from this, like the previous comments, the extra WAF rules and compliances offered makes AGIC a good choice if you don't have any extra header addition requirements like i do.

If you want Azure to manage ingress, you can use Azure application gateway and application gateway ingress controller (agic). However, if you want a public/private combination, fox example if you have 8 microservices out of which you want 6 to have private endpoint and 2 to have public endpoint over https, then it cannot be dones as agic only supports one ip per port. Also one agic can only use one application gateway (With auto management, meaning it automatically manages the listeners, https settings, backendpools, probes etc. on app gateway). If you it as shared gateway, then you need to manage application gateway manually (with az cli). Also, its github page has good how tos (https://github.com/Azure/application-gateway-kubernetes-ingress).