Hello Folks, I am trying to understand how to use an AWS IAM role for reading data from S3. My starrocks instance is on a self-managed K8s cluster in AWS and I use

kube2iam

for managing credentials. This is what I observe: • I am running into

403

errors when I try to query a single parquet file in s3 using

select

and

files()

StarRocks > CREATE TABLE table_test as SELECT * FROM FILES ("path"="s3://<bucket>/<filename>.parquet", "format"="parquet", "aws.s3.region"="us-east-1");
ERROR 2013 (HY000): Lost connection to MySQL server during query
No connection. Trying to reconnect...
Connection id:    16777259
Current database: *** NONE ***

ERROR 1064 (HY000): Access storage error. Error message: failed to get file schema, path: s3://<bucket>/<filename>.parquet, error: [Init parquet reader fail. IOError: BE access S3 file failed, SdkResponseCode=403, SdkErrorType=15, SdkErrorMessage=No response body., filename: s3://<bucket>/<filename>.parquet]

◦ AK and SK are not an option we have to rely on

kube2iam

• I know my pods are annotated correctly and have the right permissions because: ◦ I am able to

exec

into the container install

awscli

and copy the file directly ◦ I can also verify the caller identity and it has the same IAM role as the pod annotation Did anyone have luck in running starrocks with IAM role and

kube2iam

? Thanks.