Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
If you are interested in working on this issue or have submitted a pull request, please leave a comment
Tell us about your request
Very similar to #760 but we are not as able to proxy as they were.
Blue/Green Fargate deployments with traffic control that can handle multiple target groups per service.
Which service(s) is this request for?
This could be Fargate, ECS
Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?
We currently run an application in Fargate that receives traffic on 5 different ports, 2 public and 3 privte, going through NLBs. It would make autoscaling and cost optimisation much more effective if we could receive that traffic on a small group of base containers, and scale up when needed, instead of the current situation where we will need a minimum of 5 tasks * autoscaling required for HA operations.
Are you currently working around this issue?
We have 5 services with 2 tasks in each service that only handles 1 of the 5 network ports. This makes scaling very inefficient.
Additional context
This feature could take our application from the current minimum of 9 containers to a minimum of 3.
monelgordillo
commented
1 month ago
Community Note
Tell us about your request Very similar to #760 but we are not as able to proxy as they were. Blue/Green Fargate deployments with traffic control that can handle multiple target groups per service.
Which service(s) is this request for? This could be Fargate, ECS
Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard? We currently run an application in Fargate that receives traffic on 5 different ports, 2 public and 3 privte, going through NLBs. It would make autoscaling and cost optimisation much more effective if we could receive that traffic on a small group of base containers, and scale up when needed, instead of the current situation where we will need a minimum of 5 tasks * autoscaling required for HA operations.
Are you currently working around this issue? We have 5 services with 2 tasks in each service that only handles 1 of the 5 network ports. This makes scaling very inefficient.
Additional context This feature could take our application from the current minimum of 9 containers to a minimum of 3.