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
duckdb documentation describes how to create multiple secrets for the same service, e.g. if we have two S3 buckets that need different credentials. The directions indicating scoping based on bucket name.
However, I commonly encounter the case where I need to access two different S3 services at different endpoints -- for instance, S3 buckets data.source.coop , S3 buckets on a local minio instance, S3 buckets on AWS. It would be natural to scope these by the endpoint name, not the bucket name -- I may have the same credentials for multiple buckets, and I may have the same bucket name on different endpoints.
I think it is something of a bug that scope focuses on bucket name but ignores endpoint. Would you consider adding support for this?
The text was updated successfully, but these errors were encountered:
duckdb documentation describes how to create multiple secrets for the same service, e.g. if we have two S3 buckets that need different credentials. The directions indicating scoping based on bucket name.
However, I commonly encounter the case where I need to access two different S3 services at different endpoints -- for instance, S3 buckets data.source.coop , S3 buckets on a local minio instance, S3 buckets on AWS. It would be natural to scope these by the endpoint name, not the bucket name -- I may have the same credentials for multiple buckets, and I may have the same bucket name on different endpoints.
I think it is something of a bug that scope focuses on bucket name but ignores endpoint. Would you consider adding support for this?
The text was updated successfully, but these errors were encountered: