hadoop copying the result from hdfs to S3

Question

I've successfully completed my job on Amazon EMR, Now I want to copy results from HDFS to S3,but I have some problem

this is the code (--steps)

{
    "Name":"AAAAA",
    "Type":"CUSTOM_JAR",
    "Jar":"command-runner.jar",
    "ActionOnFailure":"CONTINUE",
    "Args": [
        "s3-dist-cp",
        "--src", "hdfs:///seqaddid_output",
        "--dest", "s3://wuda-notebook/seqaddid"
    ]  
}

this is the logs:

2019-04-12 03:01:23,571 INFO com.amazon.elasticmapreduce.s3distcp.S3DistCp (main): Running with args: -libjars /usr/share/aws/emr/s3-dist-cp/lib/commons-httpclient-3.1.jar,/usr/share/aws/emr/s3-dist-cp/lib/commons-logging-1.0.4.jar,/usr/share/aws/emr/s3-dist-cp/lib/guava-18.0.jar,/usr/share/aws/emr/s3-dist-cp/lib/s3-dist-cp-2.10.0.jar,/usr/share/aws/emr/s3-dist-cp/lib/s3-dist-cp.jar --src hdfs:///seqaddid_output/ --dest s3://wuda-notebook/seqaddid 2019-04-12 03:01:24,196 INFO com.amazon.elasticmapreduce.s3distcp.S3DistCp (main): S3DistCp args: --src hdfs:///seqaddid_output/ --dest s3://wuda-notebook/seqaddid 2019-04-12 03:01:24,203 INFO com.amazon.elasticmapreduce.s3distcp.S3DistCp (main): Using output path 'hdfs:/tmp/4f93d497-fade-4c78-86b9-59fc3da35b4e/output' 2019-04-12 03:01:24,263 INFO com.amazon.elasticmapreduce.s3distcp.S3DistCp (main): GET http://169.254.169.254/latest/meta-data/placement/availability-zone result: us-east-1f 2019-04-12 03:01:24,664 FATAL com.amazon.elasticmapreduce.s3distcp.S3DistCp (main): Failed to get source file system java.io.FileNotFoundException: File does not exist: hdfs:/seqaddid_output at org.apache.hadoop.hdfs.DistributedFileSystem$27.doCall(DistributedFileSystem.java:1444) at org.apache.hadoop.hdfs.DistributedFileSystem$27.doCall(DistributedFileSystem.java:1437) at org.apache.hadoop.fs.FileSystemLinkResolver.resolve(FileSystemLinkResolver.java:81) at org.apache.hadoop.hdfs.DistributedFileSystem.getFileStatus(DistributedFileSystem.java:1452) at com.amazon.elasticmapreduce.s3distcp.S3DistCp.run(S3DistCp.java:795) at com.amazon.elasticmapreduce.s3distcp.S3DistCp.run(S3DistCp.java:705) at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:76) at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:90) at com.amazon.elasticmapreduce.s3distcp.Main.main(Main.java:22) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at org.apache.hadoop.util.RunJar.run(RunJar.java:234) at org.apache.hadoop.util.RunJar.main(RunJar.java:148)

Putti Putti · Accepted Answer · 2019-04-12T09:17:39

It looks like the bug is caused by a race condition when CopyFilesReducer uses multiple CopyFilesRunable instances to download the files from S3. The problem is that it uses the same temp directory in multiple threads, and the threads delete the temp directory when they're done. Hence, when one thread completes before another it deletes the temp directory that another thread is still using.

I've reported the problem to AWS, but in the mean time you can work around the bug by forcing the reducer to use a single thread by setting the variable s3DistCp.copyfiles.mapper.numWorkers to 1 in your job config.

hadoop copying the result from hdfs to S3

2 Answers