xichen01 commented on code in PR #4188:
URL: https://github.com/apache/ozone/pull/4188#discussion_r1551354188
##########
hadoop-ozone/client/src/main/java/org/apache/hadoop/ozone/client/rpc/RpcClient.java:
##########
@@ -1195,7 +1196,8 @@ public OzoneOutputStream createKey(
.setReplicationConfig(replicationConfig)
.addAllMetadata(metadata)
.setAcls(getAclList())
- .setLatestVersionLocation(getLatestVersionLocation);
+ .setLatestVersionLocation(getLatestVersionLocation)
+ .setOwnerName(ownerName);
Review Comment:
@smengcl Thank for your suggestion.
- Necessity for the Owner
"Owner" is a fundamental feature of file systems, including S3 and
Filesystem, and is used to explicitly record who created the current Object.
This is important for permissions management (Linux UGO and ACL Permission).
So it is necessary to store a real Owner for each Object. This is the only
way to achieve better permissions control and so on.
- For the `chown` command
When we `chown` a directory, we do not recursively `chown` every file in the
directory by default, unless the -R option is specified.
- For the performance
In fact, we ran a performance test after adding an Owner, and there was no
observable performance degradation.
- For the owner configuration
Creating a file in Linux, or creating an Object via S3, doesn't have an
option to create a file/object with an empty owner, so I think we probably
don't need it.
--
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]
