-
Notifications
You must be signed in to change notification settings - Fork 274
Description
I'm trying to spin the NTH up using queue mode in an EKS cluster that doesn't have direct internet access but does have VPC endpoints created for (hopefully) every required service.
This isn't the first controller that I've spun up; every previous one I'm using requires patching in the environment variable AWS_STS_REGIONAL_ENDPOINTS=regional
in order to make them use sts.<region>.amazonaws.com
instead of the global sts.amazonaws.com
endpoint. The former endpoint then resolves to the VPC endpoint I've created and authentication is successful.
I'm getting the following logged when running the NTH:
2021/04/26 17:09:25 WRN There was a problem monitoring for events error="WebIdentityErr: failed to retrieve credentials\ncaused by: RequestError: send request failed\ncaused by: Post \"https://sts.amazonaws.com/\": dial tcp 52.94.241.129:443: i/o timeout" event_type=SQS_TERMINATE
2021/04/26 17:11:25 WRN There was a problem monitoring for events error="WebIdentityErr: failed to retrieve credentials\ncaused by: RequestError: send request failed\ncaused by: Post \"https://sts.amazonaws.com/\": dial tcp 52.46.134.192:443: i/o timeout" event_type=SQS_TERMINATE
I've set the above environment variable, and also tried removing the AWS_ENDPOINT
and AWS_REGION
variables just in case they were confusing things but the Deployment/Pod doesn't appear to be respecting it.