You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There are some containers that are not available immediately after start. Because of this, ECS blue/green deploys can actually cause a service to become unavailable if, after a deployment, blue is taken out of service before green is actually available.
Are there plans for adding time delays for deployments so that blue will be kept in service time amount before it is removed? Or better yet, are there plans to introduce health-checking at the service level such that blue will remain until green passes the health check?
The text was updated successfully, but these errors were encountered:
@djenriquez The easiest way to do this today is to use an ALB. ECS listens to ELB/ALB health checks and will ensure that newly launched tasks become healthy before stopping enough tasks to drop your service below the minimum healthy percent.
I'm going to close this in favor of #534, which is about health checking using Docker's health check feature.
There are some containers that are not available immediately after start. Because of this, ECS blue/green deploys can actually cause a service to become unavailable if, after a deployment, blue is taken out of service before green is actually available.
Are there plans for adding time delays for deployments so that blue will be kept in service
time
amount before it is removed? Or better yet, are there plans to introduce health-checking at the service level such that blue will remain until green passes the health check?The text was updated successfully, but these errors were encountered: