captainzmc commented on pull request #915: URL: https://github.com/apache/hadoop-ozone/pull/915#issuecomment-629963699
> For any request to S3, what volume need to be used if we generate from accessKeyID, and if that volume is there in the list use that or fall back to new code. This should work for all the scenarios mentioned above. > > Lets say we got request for > bucket1 from user1 -> volumeName from accessKeyID lets say is xxxv1, we use xxxv1 as volume. > bucket2 from user1 -> volumeName from accessKeyID lets say is xxxv2, we use xxxv2 as volume. > Now we got a new request, which is bucket3, new one which is not from older version, when we generate volume name from accessKeyID, it does not match with list, so fall back. > > So, with this approach, onlything is admin need to make sure that he configure all volumeNames for all users in the cluster, by getting their accessKeyID's. Reuse the AccesskeyId to determine the value of the volume. This conflicts with the design of [hdds-3385](https://github.com/apache/hadoop-ozone/blob/master/hadoop-hdds/docs/content/design/ozone-volume-management.md), which was meant to simplify the implementation by removing the mapping between volume and AccessKeyId. We need more discussion here. cc @elek. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
