AWS Discussion Forum

Full Version: Target group health check failed
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Folks we are facing the below issue , can anyone help me here? 

Issue: Target group health check failing .

We configured blue green ecs deployment . When we have a single ec2 instance in the cluster and when we create ECS service with desired task count as 1 - the task will place a container on port 8080. 

Next time when we do a deployment - the service fails since the port 8080 is already in use in the container instance . So we decided to use the dynamic port mapping in task definition but in that case the target group health check is failing with the settings like in the image.
Few question to add some context here.

1. So, with just the initial deployment, the target groups works and only when you do additional deployments its failing? Or does the target group never work?
2. Is the inbound from the LB SG allowed on the EC2 SG?
3. In ECS, what network mode are you using?

Also, please confirm if all instruction on this link was follwed correctly including the exceptions to watch out for: https://aws.amazon.com/premiumsupport/kn...pping-ecs/
Yes first time when we deploy after giving host and container port the same i.e. 8080 it gets deployed and runs fine(only one ec2 instance exists at that time). When we increase the desired count to 2, the new one fails saying port is already in use(obvious).
We have allowed the inbound traffic for all ports on EC2 SG from ALB SG
We have not mentioned any network mode so its automatically taking bridge mode. other modes have some issues with dynamic port mapping so we are not specifying any network modes.

We rechecked the article and found that all the recommend settings are already in place as specified in the link.