[ECS] [request]: Add container image pull time as a metric
- Dominant language
- Shell
- Stars
- 5.4k
- Forks
- 334
- PR merge metrics
- No merged PRs in 30d
Description
### Community Note
* Please vote on this issue by adding a 👍 [reaction](https://blog.github.com/2016-03-10-add-reactions-to-pull-requests-issues-and-comments/) 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**
Add container image pull time as a metric
**Which service(s) is this request for?**
This could be ECS on EC2 or Fargate
**Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?**
We have a customer who has several accounts and regions pulling from Docker Registry reporting this feature ask. We'd like to see if this is a common community ask so we can prioritize this better.
"Docker registry has a limit on how many concurrent pulls and how much download bandwidth can be used, and throttles image pulls. Tasks during this throttled period are PENDING, and it is not very clear why they are pending. On close look, we can see that PENDING tasks are increasing the pull timer. We need a metric that can be used to trigger alarms when pull times exceed a threshold. "
**Are you currently working around this issue?**
How are you currently solving this problem?
Manually checking. Or use ECR where the throttling limits are more generous where customers can avoid this issue or at least less impacted.
Curious if the community is actually interested in making this a metric regardless if pulling from Docker Hub or ECR or any other public registry.
Contributor guide
Research direction
No files, tests, or code entry points are identified; this is an AWS ECS/Fargate feature request rather than a scoped repository change. Start by clarifying the metric's definition and coverage for EC2/Fargate and Docker Hub, ECR, or other registries, then confirm how pull throttling and PENDING time should be represented. Done means an agreed, implementable metric specification.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- aws, docker
- Domain
- cloud, observability
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100