[ 
https://issues.apache.org/jira/browse/HDDS-1185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16781100#comment-16781100
 ] 

Jitendra Nath Pandey commented on HDDS-1185:
--------------------------------------------

A few comments:
 # In OzoneFileStatus, blockSize is hardcoded.
 # Set permissions to null, instead of hdfs defaults, as they are not supported 
yet.
 # Owner cannot be UserGroupInformation.getCurrentUser(). Most likely it needs 
to be set explicitly. I would expect it to be part of OMKeyInfo, or should come 
from OM metadata.
 # 

{code:java}
if (dirKeyInfo != null) {
  return new OzoneFileStatus(dirKeyInfo, true);
}

List<OmKeyInfo> keys = metadataManager.listKeys(volumeName, bucketName,
    null, dirKey, 1);
if (keys.iterator().hasNext()) {{code}
If the dirKeyInfo is null, can we still expect listKeys to return some keys?
 The creation time of the first key returned in this list is used as the 
creation-time of the directory. The keys returned may not be in the same order 
as they were created.

 

> Optimize GetFileStatus in OzoneFileSystem by reducing the number of rpc call 
> to OM.
> -----------------------------------------------------------------------------------
>
>                 Key: HDDS-1185
>                 URL: https://issues.apache.org/jira/browse/HDDS-1185
>             Project: Hadoop Distributed Data Store
>          Issue Type: Sub-task
>          Components: Ozone Filesystem
>    Affects Versions: 0.4.0
>            Reporter: Mukul Kumar Singh
>            Assignee: Mukul Kumar Singh
>            Priority: Major
>             Fix For: 0.4.0
>
>         Attachments: HDDS-1185.001.patch, HDDS-1185.002.patch
>
>
> GetFileStatus sends multiple rpc calls to Ozone Manager to fetch the file 
> status for a given file. This can be optimized by performing all the 
> processing on the OzoneManager for this.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org

Reply via email to