This message was deleted.

You talking the alb ingress controller? That should be managed via the helm template

That’s not relevant.

Can you explain your issue further? I do not work for Pulumi, just trying to help from a community stand point.

Well, it’s a resource created outside the scope of Pulumi, but it needs to be reigned back in. Since we’re using Pulimi as infrastructure code (same issue with Terraform though).

Right but the alb ingress controller is where you manage your routes and connections not directly through building your own alb

Since the ALB Controller creates a load balancer, how can we reference that load balancer in Pulimi and how do we know it’s ready?

So depending how its deployed, public facing or internal you could have a post hook to test if the sites accesible. I would still manage this through k8 provider

Again, to reiterate, you don’t know what the LB is called, nor do you have a way to reference it, not to mention knowing if its up. That’s the whole issue dude.

Ill keep this thread for notifications. You technically dont need to pull it into pulumi sinces its a manged service with an iam role to manage your alb eks stood up.

Dude are you for real? You do, if you need access to the zoneID for Route53, or to create metrics in Cloud Watch, or in general if you want to know if something went wrong during the CI step which is its main purpose.

Yeah I am aware of external-dns, again we need it for other stuff, but you still don’t understand: There are dependencies on making sure the ALB is up and running and in a valid state, and we need to setup metrics that needs access to the ALB, Cloud Watch is one aspect, prometheus is another, and etc. These aren’t solutions.