Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?
We (and a number of AWS customers) that use the aws ecs wait services-stable command (or similar commands within SDKs) may have deployments that take > 10 minutes to become stable. There is no native way to use the waiter for a deployment that takes > 10 minutes.
Are you currently working around this issue?
I have seen a number of different ways of working around this:
wrapping the AWS waiter with code so the waiter itself is called repeatedly
using a different method for determining if/when a deployment is complete (EventBridge?)
not utilizing a waiter (and assuming the deployment succeeded)
using a different method for deploying ECS Service updates
something else?
Proposed Solution(s)
Near Term Solution: change this to 15, 30 or 60 minutes
Long Term Solution: expose the waiter's delay and maxAttempts values to the SDK(s) and the AWS CLI (you can see this ticket here: https://github.com/aws/aws-cli/issues/1295 where there are three comments about this)
Additional context
This feels like something (near-term, changing the values for delay and maxAttempts) that would be relatively easy to do (I submitted a PR for botocore: https://github.com/boto/botocore/issues/3305)
There is some AWS documentation that would need to be updated, most notably:
If you think you might have additional information that you'd like to include via an attachment, please do - we'll take a look. (Remember to remove any personally-identifiable information.)
Community Note
Tell us about your request
I would like to propose that AWS increase the "ServicesStable" waiter time (link for botocore as of 2024-11-18: https://github.com/boto/botocore/blob/develop/botocore/data/ecs/2014-11-13/waiters-2.json#L42-L72) to some amount of time greater than 10 minutes.
Which service(s) is this request for?
ECS, ECS Fargate
Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?
We (and a number of AWS customers) that use the
aws ecs wait services-stable
command (or similar commands within SDKs) may have deployments that take > 10 minutes to become stable. There is no native way to use the waiter for a deployment that takes > 10 minutes.Are you currently working around this issue?
I have seen a number of different ways of working around this:
Proposed Solution(s)
Additional context
Attachments
If you think you might have additional information that you'd like to include via an attachment, please do - we'll take a look. (Remember to remove any personally-identifiable information.)