You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Koji Noguchi (JIRA)" <ji...@apache.org> on 2007/09/10 19:10:29 UTC
[jira] Updated: (HADOOP-1866) distcp requires large heapsize when
copying many files
[ https://issues.apache.org/jira/browse/HADOOP-1866?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Koji Noguchi updated HADOOP-1866:
---------------------------------
Affects Version/s: 0.13.1
bq. Koji, I assume this was on 0.13 or 0.14? The new distcp in 0.15 is almost completely re-written, so the version matters a lot.
Sorry. This was on 0.13.1. Looks like 0.15 distcp handles it much better. (no ArrayList finalPathList that holds all the files)
> distcp requires large heapsize when copying many files
> ------------------------------------------------------
>
> Key: HADOOP-1866
> URL: https://issues.apache.org/jira/browse/HADOOP-1866
> Project: Hadoop
> Issue Type: Bug
> Components: util
> Affects Versions: 0.13.1
> Reporter: Koji Noguchi
> Priority: Minor
>
> Trying to distcp 1.5 million files with 1G client heapsize, failed with outofmemory.
> Exception in thread "main" java.lang.OutOfMemoryError: GC overhead limit
> exceeded
> at java.util.regex.Pattern.compile(Pattern.java:1438)
> at java.util.regex.Pattern.<init>(Pattern.java:1130)
> at java.util.regex.Pattern.compile(Pattern.java:846)
> at java.lang.String.replace(String.java:2208)
> at org.apache.hadoop.fs.Path.normalizePath(Path.java:147)
> at org.apache.hadoop.fs.Path.initialize(Path.java:137)
> at org.apache.hadoop.fs.Path.<init>(Path.java:126)
> at org.apache.hadoop.dfs.DfsPath.<init>(DfsPath.java:32)
> at org.apache.hadoop.dfs.DistributedFileSystem$RawDistributedFileSystem.listPaths(DistributedFileSystem.java:214)
> at org.apache.hadoop.fs.FileSystem.listPaths(FileSystem.java:483)
> at org.apache.hadoop.fs.FileSystem.listPaths(FileSystem.java:496)
> at org.apache.hadoop.fs.ChecksumFileSystem.listPaths(ChecksumFileSystem.java:539)
> at org.apache.hadoop.util.CopyFiles$FSCopyFilesMapper.setup(CopyFiles.java:327)
> at org.apache.hadoop.util.CopyFiles.copy(CopyFiles.java:762)
> at org.apache.hadoop.util.CopyFiles.run(CopyFiles.java:808)
> at org.apache.hadoop.util.ToolBase.doMain(ToolBase.java:189)
> at org.apache.hadoop.util.CopyFiles.main(CopyFiles.java:818)
> It would be nice if distcp doesn't require gigs of heapsize when copying large amount of files.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.