ECS service failure in some case

[af6140]

I am authenticated through aws-vault exec with assumed roles, when login to web console I can see all clusters and details without issue.
But when go to ECS screen, I got 400 on ECS: DescribeClsuters http calls. Wondering does claws pass all caller identity when make this call.

[yimsk]

A few questions to help debug:

1. Are you passing -p <profile> to claws when using aws-vault exec?

   Since aws-vault injects credentials via environment variables, -p is not needed.

2. Does it work with the -e option?

   aws-vault exec <profile> -- claws -e
-e ignores ~/.aws/config and uses environment credentials instead.

3. Do these CLI commands work?

   aws-vault exec <profile> -- aws ecs list-clusters
aws-vault exec <profile> -- aws ecs describe-clusters --clusters <ARN>

[yimsk]

If CLI works but claws doesn't, could you share:

1. Full claws command you're running

2. Are you using multi-profile (-p dev,prod) or multi-region (-r us-east-1,ap-northeast-1)?

3. Debug log output:

   aws-vault exec <profile> -- claws -e -l /tmp/claws.log

   Then share the contents of /tmp/claws.log

[af6140]

A few questions to help debug:

1. Are you passing -p <profile> to claws when using aws-vault exec?
   Since aws-vault injects credentials via environment variables, -p is not needed.
   No
2. Does it work with the -e option?
   aws-vault exec <profile> -- claws -e
-e ignores ~/.aws/config and uses environment credentials instead.
   No
3. Do these CLI commands work?
   aws-vault exec <profile> -- aws ecs list-clusters
aws-vault exec <profile> -- aws ecs describe-clusters --clusters <ARN>
   Both works

[af6140]

If CLI works but claws doesn't, could you share:

1. Full claws command you're running

aws-valut exec -- claws

3. Are you using multi-profile (-p dev,prod) or multi-region (-r us-east-1,ap-northeast-1)?

No

4. Debug log output:> aws-vault exec <profile> -- claws -e -l /tmp/claws.log
   Then share the contents of /tmp/claws.log

time=2026-01-25T16:42:16.691-05:00 level=INFO msg="claws started" profiles=[] regions=[] readOnly=false
time=2026-01-25T16:42:18.290-05:00 level=DEBUG msg=navigating clearStack=false stackDepth=0
time=2026-01-25T16:42:18.290-05:00 level=DEBUG msg="loading resources" service=ecs resourceType=clusters profiles=1 regions=[us-east-1] multiProfile=false multiRegion=false
time=2026-01-25T16:42:18.900-05:00 level=WARN msg="describe clusters" error="operation error ECS: DescribeClusters, https response error StatusCode: 400, RequestID: 537ba906-5766-4691-a5c7-35d5a941b9c7, InvalidParameterException: Clusters cannot have more than 100 elements"
time=2026-01-25T16:42:18.901-05:00 level=ERROR msg="failed to list resources" error="describe clusters: operation error ECS: DescribeClusters, https response error StatusCode: 400, RequestID: 537ba906-5766-4691-a5c7-35d5a941b9c7, InvalidParameterException: Clusters cannot have more than 100 elements" duration=610.131709ms

[af6140]

Closed accidentally

[yimsk]

Thanks for the detailed log — that made it easy to identify the issue.

The root cause was that DescribeClusters was being called with all cluster ARNs at once, which exceeds the API's 100-element limit. Fixed by batching the calls into chunks of 100.

Released in v0.14.1. Could you give it a try and let me know if it resolves the issue?

https://github.com/clawscli/claws/releases/tag/v0.14.1

[af6140]

Thanks for the detailed log — that made it easy to identify the issue.

The root cause was that DescribeClusters was being called with all cluster ARNs at once, which exceeds the API's 100-element limit. Fixed by batching the calls into chunks of 100.

Released in v0.14.1. Could you give it a try and let me know if it resolves the issue?

https://github.com/clawscli/claws/releases/tag/v0.14.1

Works now!!!

[yimsk]

Confirmed working by reporter on v0.14.1. Closing.