[jira] Created: (HADOOP-1453) exists() not necessary before DFS.open

classic Classic list List threaded Threaded
16 messages Options
Reply | Threaded
Open this post in threaded view
|

[jira] Created: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
exists() not necessary before DFS.open
--------------------------------------

                 Key: HADOOP-1453
                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
             Project: Hadoop
          Issue Type: Bug
          Components: dfs
            Reporter: Raghu Angadi
             Fix For: 0.14.0



{code:title=DistributedFileSystem.java:131|borderStyle=solid}
public FSDataInputStream open(Path f, int bufferSize) throws IOException {
      if (!exists(f)) {
        throw new FileNotFoundException(f.toString());
      }

      return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
 }
{code}

{{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.



--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Assigned: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi reassigned HADOOP-1453:
------------------------------------

    Assignee: Raghu Angadi

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Attachment: HADOOP-1453-01.patch

Attached simple patch:

{code}
Index: src/java/org/apache/hadoop/dfs/DistributedFileSystem.java
===================================================================
--- src/java/org/apache/hadoop/dfs/DistributedFileSystem.java   (revision 546652)
+++ src/java/org/apache/hadoop/dfs/DistributedFileSystem.java   (working copy)
@@ -129,10 +129,6 @@
     }
 
     public FSDataInputStream open(Path f, int bufferSize) throws IOException {
-      if (!exists(f)) {
-        throw new FileNotFoundException(f.toString());
-      }
-
       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
     }
 
{code}

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12504078 ]

Raghu Angadi commented on HADOOP-1453:
--------------------------------------

I guess this is not enough. There is an exists() in ChecksumFileSystem.open() as well. Should that be removed as well? :

{code}
  public FSDataInputStream open(Path f, int bufferSize) throws IOException {
    if (!exists(f)) {
      throw new FileNotFoundException(f.toString());
    }
    return new FSDataInputStream(new FSInputChecker(this, f, bufferSize),
                                 bufferSize);
  }
{code}

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12504082 ]

Raghu Angadi commented on HADOOP-1453:
--------------------------------------

This implies there are 3 extra RPCs to NameNode for typical DFS file open(). One in ChecksumFS and one each for file and .file.crc in DFS.
 

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Attachment: HADOOP-1453-02.patch

Attaching an updated patch. I think it is correct and can be checked in if anyone reviews it.


> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12505982 ]

dhruba borthakur commented on HADOOP-1453:
------------------------------------------

Both your patches are for DistributedFileSystem whereas your comment says that there are *two* places where a redundant exists()  call is made: DistributedFileSystem and ChecksumFileSystem. Can you pl clarify?

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Attachment: HADOOP-1453-02.patch

Thanks Dhruba. Now it is fixed. Please review 02.patch.

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12506066 ]

dhruba borthakur commented on HADOOP-1453:
------------------------------------------

+1 code looks good.

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Status: Patch Available  (was: Open)

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12506239 ]

Hadoop QA commented on HADOOP-1453:
-----------------------------------

-1, build or testing failed

2 attempts failed to build and test the latest attachment http://issues.apache.org/jira/secure/attachment/12360067/HADOOP-1453-02.patch against trunk revision r548523.

Test results:   http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/298/testReport/
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/298/console

Please note that this message is automatically generated and may represent a problem with the automation system and not the patch.

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Status: Open  (was: Patch Available)

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Raghu Angadi updated HADOOP-1453:
---------------------------------

    Status: Patch Available  (was: Open)

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12506346 ]

Hadoop QA commented on HADOOP-1453:
-----------------------------------

+1

http://issues.apache.org/jira/secure/attachment/12360067/HADOOP-1453-02.patch applied and successfully tested against trunk revision r548794.

Test results:   http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/308/testReport/
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/308/console

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Updated: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

     [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Doug Cutting updated HADOOP-1453:
---------------------------------

    Resolution: Fixed
        Status: Resolved  (was: Patch Available)

I just committed this.  Thanks, Raghu!

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply | Threaded
Open this post in threaded view
|

[jira] Commented: (HADOOP-1453) exists() not necessary before DFS.open

Tim Allison (Jira)
In reply to this post by Tim Allison (Jira)

    [ https://issues.apache.org/jira/browse/HADOOP-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12506868 ]

Hudson commented on HADOOP-1453:
--------------------------------

Integrated in Hadoop-Nightly #131 (See [http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Nightly/131/])

> exists() not necessary before DFS.open
> --------------------------------------
>
>                 Key: HADOOP-1453
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1453
>             Project: Hadoop
>          Issue Type: Bug
>          Components: dfs
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.14.0
>
>         Attachments: HADOOP-1453-01.patch, HADOOP-1453-02.patch, HADOOP-1453-02.patch
>
>
> {code:title=DistributedFileSystem.java:131|borderStyle=solid}
> public FSDataInputStream open(Path f, int bufferSize) throws IOException {
>       if (!exists(f)) {
>         throw new FileNotFoundException(f.toString());
>       }
>       return new DFSClient.DFSDataInputStream(dfs.open(getPath(f)), bufferSize);
>  }
> {code}
> {{exists(f)}} adds extra namenode interaction that is not really required. Open is a critical DFS call.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.