[HADOOP-11584] s3a file block size set to 0 in getFileStatus

• Type: Sub-task
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 2.6.0
• Fix Version/s: 2.7.0
• Component/s: fs/s3
• Assignee: Brahma Reddy Battula
• Reporter: Daniel Hecht
• Created: 11/Feb/15 20:14
• Updated: 24/Apr/15 22:49
• Resolved: 21/Feb/15 12:30

Description

The consequence is that mapreduce probably is not splitting s3a files in the expected way. This is similar to HADOOP-5861 (which was for s3n, though s3n was passing 5G rather than 0 for block size).

FileInputFormat.getSplits() relies on the FileStatus block size being set:

        if (isSplitable(job, path)) {
          long blockSize = file.getBlockSize();
          long splitSize = computeSplitSize(blockSize, minSize, maxSize);

However, S3AFileSystem does not set the FileStatus block size field. From S3AFileStatus.java:

  // Files
  public S3AFileStatus(long length, long modification_time, Path path) {
    super(length, false, 1, 0, modification_time, path);
    isEmptyDirectory = false;
  }

I think it should use S3AFileSystem.getDefaultBlockSize() for each file's block size (where it's currently passing 0).

Issue Links

• is related to: HADOOP-11601 Enhance FS spec & tests to mandate FileStatus.getBlocksize() >0 for non-empty files
• is related to: HADOOP-11606 intermittent failure of TestS3AFileSystemContract.testRenameRootDirForbidden