Implement AWS_METADATA_SERVICE_TIMEOUT + AWS_METADATA_SERVICE_NUM_ATTEMPTS

[lifeofguenter]

Would be great if AWS_METADATA_SERVICE_TIMEOUT and AWS_METADATA_SERVICE_NUM_ATTEMPTS could be implemented analogue to boto/python-sdk.

[diehlaws]

Thanks for reaching out to us about this @lifeofguenter. Can you provide some details regarding your use case that necessitates these environment variables? Are you currently unable to set these values in your config without them? Describing any other issues you're running into related to this would help us implement support for these in a useful manner.

[lifeofguenter]

This was recommended by aws-support for boto3/aws-cli due to IMDS being a service with rate limits.

However, in our case, we heavily use terraform which uses the aws-sdk-go library, so it would be great if one could set these settings here as well.

Context: hashicorp/terraform#25835 (comment)

[mdgreenfield]

We have also started running into this issue. It would be great if we could get these env vars supported for parity with boto3.

[github-actions]

We have noticed this issue has not received attention in 1 year. We will close this issue for now. If you think this is in error, please feel free to comment and reopen the issue.

[lifeofguenter]

still relevant?

[abvm659]

Please do not close this - e.g. terraform stopped supporting its custom metadata timeout solution and relies only on AWS_METADATA_SERVICE_TIMEOUT - which is not supported in go aws sdk yet - which is very strange. For large scale usage of terraform and in our go applications, unability to set timeout is causing huge issue. There is no good workaround for that (one can only retrigger the calls if they fail after some random wait time).

[jonathan-russo]

I would like to see this implemented as well. We are getting intermittent IMDS errors on our Jenkins server when attempting to apply Terraform:

failed to refresh cached credentials, no EC2 IMDS role found, operation error ec2imds: GetMetadata, canceled, context deadline exceeded

It would be nice to set AWS_METADATA_SERVICE_NUM_ATTEMPTS and/or AWS_METADATA_SERVICE_TIMEOUT to have the SDK re-attempt the request after the credential rollover.

[github-actions]

Comments on closed issues are hard for our team to see.
If you need more assistance, please either tag a team member or open a new issue that references this one.
If you wish to keep having a conversation with other community members under this issue feel free to do so.