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
If the cluster is not running in an EC2 instance (for example, I am on bare-metal), this operation blocks and hangs for ~20s because it is unable to resolve the hostname of the metadata server. Since it can't resolve the hostname of the metadata server, it doesn't get an HTTP error since the connection can't even be established and the operation is retried until it has reached its max retry.
It would be good if the operation could either a) not retry, or b) fail quicker when dns name cannot be found.
Hello @derekwaynecarr, thank you for reaching out to us. You can disable retries by setting MaxRetries, in the config, to 0. Going to close this issue. If you have any more questions, please let us know.
In the Kubernetes project, we use the following operation to detect if the cluster is running in an EC2 instance:
https://github.com/aws/aws-sdk-go/blob/master/aws/ec2metadata/api.go#L37
If the cluster is not running in an EC2 instance (for example, I am on bare-metal), this operation blocks and hangs for ~20s because it is unable to resolve the hostname of the metadata server. Since it can't resolve the hostname of the metadata server, it doesn't get an HTTP error since the connection can't even be established and the operation is retried until it has reached its max retry.
It would be good if the operation could either a) not retry, or b) fail quicker when dns name cannot be found.
For history:
kubernetes/kubernetes#22668
The text was updated successfully, but these errors were encountered: