GitHub user jonsnowseven edited a discussion: Iceberg Catalogs & S3 access
Hello all. I am doing a PoC with Gravitino. I noticed that for Iceberg catalogs, we can define access to S3: https://gravitino.apache.org/docs/1.1.0/lakehouse-iceberg-catalog#s3. * Are there other options for S3 authentication (web identity token)? * Do we really need to define those S3 configurations or can we do something like: ``` gravitino.bypass.iceberg.rest-catalog.vended-credentials-enabled=true gravitino.bypass.s3.region=${ENV:AWS_REGION} gravitino.bypass.s3.endpoint=${ENV:S3_ENDPOINT} gravitino.bypass.s3.sts.region=${ENV:AWS_REGION} gravitino.bypass.s3.sts.endpoint=${ENV:STS_ENDPOINT} gravitino.bypass.s3.use-web-identity-token-credentials-provider=true ``` ? I would assume these would be provided to each Iceberg catalog configuration. * Does Gravitino itself need access to S3? Or only the Iceberg catalog needs it? I don't know exactly what Gravitino does behind the scenes so I apologize if my questions are nonsense. Thank you in advance GitHub link: https://github.com/apache/gravitino/discussions/9663 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
