You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-issues@hadoop.apache.org by GitBox <gi...@apache.org> on 2020/02/19 15:17:24 UTC

[GitHub] [hadoop] sodonnel commented on a change in pull request #1758: HDFS-15052. WebHDFS getTrashRoot leads to OOM due to FileSystem objec…

sodonnel commented on a change in pull request #1758: HDFS-15052. WebHDFS getTrashRoot leads to OOM due to FileSystem objec…
URL: https://github.com/apache/hadoop/pull/1758#discussion_r381356059
 
 

 ##########
 File path: hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/namenode/web/resources/NamenodeWebHdfsMethods.java
 ##########
 @@ -1345,11 +1348,21 @@ protected Response get(
     }
   }
 
-  private static String getTrashRoot(String fullPath,
-      Configuration conf) throws IOException {
-    FileSystem fs = FileSystem.get(conf != null ? conf : new Configuration());
-    return fs.getTrashRoot(
-        new org.apache.hadoop.fs.Path(fullPath)).toUri().getPath();
+  private String getTrashRoot(String fullPath) throws IOException {
+    String user = UserGroupInformation.getCurrentUser().getShortUserName();
+    org.apache.hadoop.fs.Path path = new org.apache.hadoop.fs.Path(fullPath);
+    String parentSrc = path.isRoot() ?
+        path.toUri().getPath() : path.getParent().toUri().getPath();
+    EncryptionZone ez = getRpcClientProtocol().getEZForPath(parentSrc);
+    org.apache.hadoop.fs.Path trashRoot;
+    if (ez != null) {
+      trashRoot = new org.apache.hadoop.fs.Path(
+          new org.apache.hadoop.fs.Path(ez.getPath(), TRASH_PREFIX), user);
+    } else {
+      trashRoot = new org.apache.hadoop.fs.Path(
+          new org.apache.hadoop.fs.Path(USER_HOME_PREFIX, user), TRASH_PREFIX);
+    }
+    return trashRoot.toUri().getPath();
 
 Review comment:
   From the the current implementation, we need to fix the memory leak and the fact it does not work with security enabled, and probably backport that across the active branches. The memory leak especially is a big problem.
   
   Would it make sense to adopt this patch as it stands and push it to all the branches and then open another Jira to implement the namenode RPC suggestion on trunk as a new feature?

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org