Does hive-s3-connector based on hive metastore need to configure fs.s3a.* related parameters in core-site.xml?

[gzusgw]

I use trino to query files on minio, and the hive metastore version used is 4.0.1. This is my configuration

connector.name=hive
hive.metastore.uri=thrift://localhost:9083
fs.native-s3.enabled=true
s3.region=us-east-1
s3.path-style-access=true
s3.endpoint=http://localhost:9000
s3.aws-access-key=NBNDcH3AlLhnpYSN5X8c
s3.aws-secret-key=*************************************
hive.non-managed-table-writes-enabled=true
hive.storage-format=ORC

I found that if I don't configure fs.s3a.* related parameters in the hive metastore, trino query will report an error:

Query failed (#20250418_155152_00000_yzayh): Got exception: java.nio.file.AccessDeniedException s3a://trino/csv: org.apache.hadoop.fs.s3a.auth.NoAuthWithAWSException: No AWS Credentials provided by TemporaryAWSCredentialsProvider SimpleAWSCredentialsProvider EnvironmentVariableCredentialsProvider IAMInstanceCredentialsProvider : com.amazonaws.SdkClientException: Unable to load AWS credentials from environment variables (AWS_ACCESS_KEY_ID (or AWS_ACCESS_KEY) and AWS_SECRET_KEY (or AWS_SECRET_ACCESS_KEY))

Is it necessary to configure in the hive metastore? This means that a hive metastore can only manage metadata on one minio. Why isn't it dynamically controlled by the properties of the catalog in trino?

[ziggekatten]

This is an Hive metastore limitation. The metastore server only reads its runtime config at startup.

Trino uses the credentials to access files, not to alter Hive runtime configs.