[ 
https://issues.apache.org/jira/browse/HBASE-19969?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16390526#comment-16390526
 ] 

Ted Yu commented on HBASE-19969:
--------------------------------

{code}
+      Path tmpBackupDir = new 
Path(HBackupFileSystem.getBackupTmpDirForBackupId(backupRoot, mergedBackupId));
{code}
I looked at the calls to HBackupFileSystem.getBackupTmpDirForBackupId where 
return value is wrapped by Path.
It would be cleaner if HBackupFileSystem.getBackupTmpDirForBackupId() returns 
Path.
{code}
+  protected void copyFile(FileSystem fs, Path p, Path newPath) throws 
IOException {
{code}
Can you add javadoc to the above method ? It is easier to understand what the 
two Paths are for.
{code}
+    boolean exists = fs.exists(newPath);
+    LOG.debug("copyFile: "+ newPath+" exists="+exists);
{code}
The method throws IOException. If newPath doesn't exist, throwing IOException 
is better.
{code}
+    Path pp = p;
{code}
Unused variable.



> Improve FT in merge operation
> -----------------------------
>
>                 Key: HBASE-19969
>                 URL: https://issues.apache.org/jira/browse/HBASE-19969
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: Vladimir Rodionov
>            Assignee: Vladimir Rodionov
>            Priority: Major
>         Attachments: HBASE-19969-v1.patch
>
>
> Some file system operations are not fault tolerant during merge. We delete 
> backup data in a backup file system, then copy new data over to backup 
> destination. Deletes can be partial, copy can fail as well



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

Reply via email to