Copyright by:
- Ha Long Do
- John Wannerkawahara
- Pavan Karnati
- Arsalan Ahmed
demo.mp4
To run your own client run the following on a system with docker installed
docker run -d -p 3000:3000 ghcr.io/halo93/pub-sub-client:latest
Visit http://localhost:3000
Using the [Amazon ECS first run wizard][first-run], create an ECS cluster, Use the task definition from file ecs-task.json
โ \ Step-by-step Instructions
-
Go to the AWS Management Console, click Services then select Elastic Container Service under Compute. If you've never used the service before, you'll see a Getting started button which will take you to the first run wizard. If you do not see this button, [navigate to the wizard directly][first-run].
-
The first page of the wizard outlines configuration details to build our new task definition. Select Custom. Enter the docker image link as ghcr.io/halo93/pub-sub-client:latest Note that the task will run using the FARGATE launch type and it is configured for 0.5GB of RAM and 1/4th of a vCPU. Click Next.
-
The next page outlines configuration details for our service. By default it will create an app called simple-app-service running a single task in a new security group. Change it to pub-sub-client Note we can optionally create a load balancer here to distribute traffic across tasks in our service. We won't do this here but will later in the workshop. Leave the defaults and click Next.
-
The last page of the wizard allows us to name our cluster. Enter
pub-sub
into Cluster name. Click Next. -
Review the details of the new task definition, service, and cluster on the next page and click Create.
-
The first run wizard will create the task definition, service, and cluster and all supporting infrastructure. It will show you its progress as it runs:
Once complete, click View service.
-
Click on the Tasks tab to see the tasks running by the service. The task status will transition from PROVISIONING to PENDING to RUNNING as the task is placed within AWS Fargate, the container is pulled and started. Click the refresh button while the task starts. Open the public ip address on port 3000 to access the client.