Skip to content

Conversation

@WongSaang
Copy link

Benefit from OSS compatibility with the S3 API
@deitch
Copy link
Collaborator

deitch commented Aug 10, 2023

Given that it is compatible with s3, why do we need to make any changes? Can we not just use s3://bucketname.host.com/path? The scheme in that URL (s3) is a protocol, so it covers it.

@WongSaang
Copy link
Author

  1. OSS requires a separate configuration: aws configure set s3.addressing_style virtual
  2. If a user needs to store files on both AWS and OSS simultaneously, and their access keys are different, an independent profile needs to be added for OSS

@deitch
Copy link
Collaborator

deitch commented Aug 11, 2023

OSS requires a separate configuration: aws configure set s3.addressing_style virtual

Is that not the default by now?

If a user needs to store files on both AWS and OSS simultaneously, and their access keys are different, an independent profile needs to be added for OSS

Would that not be an issue if I wanted to store it in 2 separate actual AWS S3 buckets with 2 separate profiles as well? If I pass s3://bucket1.amazonaws.com/foo and s3://bucket2.amazonaws.com/bar, and each of those is a different account with different credentials, or different endpoints, I have the same issue? And if I have 2 or 3 distinct OSS credentials?

If so, then this isn't a question of, "how do I provide different credentials to OSS and AWS?", but rather, "how do I support multiple s3-protocol targets, each with different credentials and/or configurations?"

Is that correct?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants