Deploying Retool on-prem using the Terraform templates, need help debugging

I followed the instructions here and have successfully spun up the resources with Terraform

However when visiting the DNS Name of the load balancer, which I got from the AWS console, the browser does not load anything. I tried both with and without port 3000. Any ideas how to debug? The only thing I added was an existing VPC and two subnet ids.

provider "aws" {
  region = "us-west-2"
}

terraform {
  backend "s3" {
    bucket  = "testforesitelabs-terraform"
    key     = "retool/terraform.tfstate"
    region  = "us-west-2"
    encrypt = true
  }
}

module "retool" {
  source = "./module"
  aws_region = "us-west-2"
  vpc_id = "vpc-009ef99629dab0fff"
  subnet_ids = [
      "subnet-067ff1fe3a23ce09f",
      "subnet-0512f37af80344aa4"
  ]
  ssh_key_name = "retool-test"
  ecs_retool_image = "2.115.2"
}

I put all the code in the repo into a "module" folder

Hey Ashton! Would you mind sharing any container logs you have? Also, are there any errors in your browser console?

dockerlogs.csv (103.9 KB)

yes, here are the logs from the only EC2 instance, thanks so much! I put the minimum instances as one. Not sure how to get logs for the load balancer. pardon my inexperience with Terraform and Docker.

Any update on this? Any help would be appreciated!

Hey Ashton!

What command are you running to grab those logs? And how many containers do you currently see running? I wonder if you have some conflicts :thinking:

No problem. Has anyone actually succeeded in spinning up Retool via the instructions here? https://github.com/tryretool/terraform-retool-modules/tree/main/modules/aws_ecs_ec2
I didn't change anything except the min_instance_count to 1, so there is one running. Can I book an appointment with someone? I can't keep these resources up and running for days when I wait for a response, I had to destroy them each time. I am sure someone can find out the issue quickly via video call. It is necessary for our organization to be able to spin up Retool via Terraform.

Hey Ashton! Yeah, I believe people have been successful :slight_smile: And I might have a lead for us to follow here :eyes:

Looking at this line in your logs:

level=error time=2023-04-17T17:48:14Z msg="Failed to pull image for container" task="9897219894824b529435dcb91df14486" container="retool-jobs-runner" image="2.115.2" error="Error response from daemon: pull access denied for 2.115.2

It seems like the image tag might need to be tryretool/backend:2.115.2