[ 
https://issues.apache.org/jira/browse/HDFS-11150?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15702487#comment-15702487
 ] 

Rakesh R commented on HDFS-11150:
---------------------------------

Thank you [~yuanbo] for the patch. Just few comments,

# Need to verify persistence in {{fsimage}}, SPS xattributes can be correctly 
loaded from fsimage. Following is just plain thoughts(sample test snippet) for 
your reference, probably you can refine while writing test code.
{code}
    // save namespace and restart NN again to make sure SPS
    // entries can be correctly loaded from fsimage
    hdfsCluster.restartDataNodes();
    NameNodeAdapter.enterSafeMode(hdfsCluster.getNameNode(), false);
    NameNodeAdapter.saveNamespace(hdfsCluster.getNameNode());
    NameNodeAdapter.leaveSafeMode(hdfsCluster.getNameNode());
    hdfsCluster.restartNameNode(0);
{code}
# {{testtestSatisfyDirWithPersistence}} please modify to 
{{testSatisfyDirWithPersistence}}
# IIUC, this code is handling duplicate satisfyStoragePolicy(path) call, 
right?. Could you add test case to verify the same. Also, could you re-phrase 
the exception message {{"File/Directory has been already added for satisfying 
storage policy: " + inode.getFullPathName()}}
{code}
+    if (removed.size() > 0) {
+      throw new IOException("File/Directory has been "
+          + "added into movement queue: " + inode.getFullPathName());
+    }
{code}
# Any reason to not including {{TestOfflineEditsViewer#editsStored}} changes?


> [SPS]: Provide persistence when satisfying storage policy.
> ----------------------------------------------------------
>
>                 Key: HDFS-11150
>                 URL: https://issues.apache.org/jira/browse/HDFS-11150
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: datanode, namenode
>            Reporter: Yuanbo Liu
>            Assignee: Yuanbo Liu
>         Attachments: HDFS-11150-HDFS-10285.001.patch
>
>
> Provide persistence for SPS in case that Hadoop cluster crashes by accident. 
> Basically we need to change EditLog and FsImage here.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
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