ChenSammi commented on PR #7965: URL: https://github.com/apache/ozone/pull/7965#issuecomment-2693201615
@nandakumar131 , I'm wondering if one major reason of [HDDS-12175](https://issues.apache.org/jira/browse/HDDS-12175)) is because SCM will retry and resend the same batch of blockID to one DN up to thousands of times, so the SCM audit log cannot afford to log detail blockID. Since there is no retry for block allocation in OM, it's safer in OM. Generally OM will allocate one block each time for FS access. While for S3 object put, OM will allocate all blocks one time for the object. @symious , If it's multi-upload, will the block allocation in unit of Ozone block size(256MB), or the multi-upload part size(for example, 4MB)? If it's the Ozone block size(256MB), then it's kind of acceptable, but if it's part size, then that will be too much to log all blockIDs. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
