Snappy compression error in Hadoop 2.x

Question

I've setup a Hadoop cluster using the newly 2.x version. And I installed snappy and hadoop snappy according to this guide, to enable snappy compression in map output.

When running the example wordcount, The error occurred:

[dm@node1 ~]$ hadoop jar /opt/hadoop-2.0.5-alpha/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.0.5-alpha.jar wordcount /in /out
13/09/06 05:09:52 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
13/09/06 05:09:53 INFO service.AbstractService: Service:org.apache.hadoop.yarn.client.YarnClientImpl is inited.
13/09/06 05:09:53 INFO service.AbstractService: Service:org.apache.hadoop.yarn.client.YarnClientImpl is started.
13/09/06 05:10:04 INFO input.FileInputFormat: Total input paths to process : 1
13/09/06 05:10:04 INFO snappy.LoadSnappy: Snappy native library loaded
13/09/06 05:10:04 INFO lzo.GPLNativeCodeLoader: Loaded native gpl library
13/09/06 05:10:04 INFO lzo.LzoCodec: Successfully loaded & initialized native-lzo library [hadoop-lzo rev d0f5a10f99f1b2af4f6610447052c5a67b8b1cc7]
13/09/06 05:10:04 INFO mapreduce.JobSubmitter: number of splits:1
13/09/06 05:10:04 WARN conf.Configuration: mapred.jar is deprecated. Instead, use mapreduce.job.jar
13/09/06 05:10:04 WARN conf.Configuration: mapred.output.value.class is deprecated. Instead, use mapreduce.job.output.value.class
13/09/06 05:10:04 WARN conf.Configuration: mapreduce.combine.class is deprecated. Instead, use mapreduce.job.combine.class
13/09/06 05:10:04 WARN conf.Configuration: mapreduce.map.class is deprecated. Instead, use mapreduce.job.map.class
13/09/06 05:10:04 WARN conf.Configuration: mapred.job.name is deprecated. Instead, use mapreduce.job.name
13/09/06 05:10:04 WARN conf.Configuration: mapreduce.reduce.class is deprecated. Instead, use mapreduce.job.reduce.class
13/09/06 05:10:04 WARN conf.Configuration: mapred.input.dir is deprecated. Instead, use mapreduce.input.fileinputformat.inputdir
13/09/06 05:10:04 WARN conf.Configuration: mapred.output.dir is deprecated. Instead, use mapreduce.output.fileoutputformat.outputdir
13/09/06 05:10:04 WARN conf.Configuration: mapred.map.tasks is deprecated. Instead, use mapreduce.job.maps
13/09/06 05:10:04 WARN conf.Configuration: mapred.output.key.class is deprecated. Instead, use mapreduce.job.output.key.class
13/09/06 05:10:04 WARN conf.Configuration: mapred.working.dir is deprecated. Instead, use mapreduce.job.working.dir
13/09/06 05:10:05 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1378415309099_0001
13/09/06 05:10:06 INFO client.YarnClientImpl: Submitted application application_1378415309099_0001 to ResourceManager at node1/192.168.56.101:60832
13/09/06 05:10:06 INFO mapreduce.Job: The url to track the job: http://node1:60888/proxy/application_1378415309099_0001/
13/09/06 05:10:06 INFO mapreduce.Job: Running job: job_1378415309099_0001
13/09/06 05:10:32 INFO mapreduce.Job: Job job_1378415309099_0001 running in uber mode : false
13/09/06 05:10:32 INFO mapreduce.Job:  map 0% reduce 0%
13/09/06 05:10:53 INFO mapreduce.Job:  map 100% reduce 0%
13/09/06 05:11:02 INFO mapreduce.Job: Task Id : attempt_1378415309099_0001_m_000000_0, Status : FAILED
Error: org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z
Container killed by the ApplicationMaster.


13/09/06 05:11:03 INFO mapreduce.Job:  map 0% reduce 0%
13/09/06 05:11:07 INFO mapreduce.Job: Task Id : attempt_1378415309099_0001_m_000000_1, Status : FAILED
Error: org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z
13/09/06 05:11:13 INFO mapreduce.Job: Task Id : attempt_1378415309099_0001_m_000000_2, Status : FAILED
Error: org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z
13/09/06 05:11:19 INFO mapreduce.Job:  map 100% reduce 0%
13/09/06 05:11:19 INFO mapreduce.Job: Job job_1378415309099_0001 failed with state FAILED due to: Task failed task_1378415309099_0001_m_000000
Job failed as tasks failed. failedMaps:1 failedReduces:0

13/09/06 05:11:19 INFO mapreduce.Job: Counters: 6
    Job Counters 
        Failed map tasks=4
        Launched map tasks=4
        Other local map tasks=3
        Data-local map tasks=1
        Total time spent by all maps in occupied slots (ms)=42989
        Total time spent by all reduces in occupied slots (ms)=0

I searched google about the error message "Error: org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z", haven't find the solution to this problem. So I want to know how can I enable snappy compression in Hadoop 2.x? Thanks.

Answer 1

good server:

$ hadoop checknative -a | grep snappy
15/06/18 14:51:05 INFO bzip2.Bzip2Factory: Successfully loaded & initialized native-bzip2 library system-native
15/06/18 14:51:05 INFO zlib.ZlibFactory: Successfully loaded & initialized native-zlib library
snappy:  true /opt/cloudera/parcels/CDH-5.4.2-1.cdh5.4.2.p0.2/lib/hadoop/lib/native/libsnappy.so.1

not good server:

$ hadoop checknative -a | grep snappy
15/06/18 14:50:31 INFO bzip2.Bzip2Factory: Successfully loaded & initialized native-bzip2 library system-native
15/06/18 14:50:31 INFO zlib.ZlibFactory: Successfully loaded & initialized native-zlib library
snappy:  true /usr/lib64/libsnappy.so.1

Notice in one case "hadoop checknative" command returns system's libsnappy, in another case - one from a Hadoop distrubution. If you have hadoop on some hosts report that they use system's libsnappy, errors like ".hadoop.mapred.YarnChild: Error running child : java.lang.UnsatisfiedLinkError: org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z at org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy(Native Method) " are quite possible.

Answer 2

From the output you’ve presented ( WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... etc.. ) it looks like your native library isn't loading. I think it would be worth trying to build and install the native library and then adding the Snappy library inside it at <HADOOP_HOME>/lib/native/ . I think that might resolve the issue. Hope that helps!

Answer 3

It looks like your code is not able to load snappy native library.There could be two possible explanation for this. Either you are using a version of snappy which is not compatible with the version of hadoop you have installed. OR You have not added the library in your classpath.

Answer 4

The second parameter on that guide is wrong.

Use mapreduce.map.output.compress.codec instead.