The stupidest way to solve the error "Java.lang.UnsatisfiedLinkError:no Snappyjava in Java.library.path" using snappy compression

Previously wrote an article: http://blog.csdn.net/stark_summer/article/details/47361603, that time the Linux environment Spark use snappy way compression is not good, And today my colleague Hive on Hadoop uses snappy compression method also reported this mistake, now I, feel this problem must solve

I think, can only use the most stupid way to solve this problem, put the libsnappyjava.so file under the $JAVA _home/jre/lib/amd64/

The operation process is as follows:

First copy the $HADOOP _home/share/hadoop/common/lib/snappy-java-1.1.1.7.jar to the temp directory, and then unzip Snappy-java-1.1.1.7.jar

The following directories will be available after decompression:

4096 15:58 meta-inf
4096 Apr 16:05 org

Enter the directory where libsnappyjava.so is located:

$CD org/xerial/snappy/native/linux/x86_64/

You can see the following files:

Libsnappyjava.so

Copy to $JAVA _home/jre/lib/amd64/

The test procedure is as follows:

Import org.xerial.snappy.snappy;/** *  * Created by Stark_summer on 15/8/8. *   */public class Testsnappy {    Publ IC static void Main (string[] args) throws exception{        String input = "Hello snappy-java! Snappy-java is a jni-based wrapper of "                +" Snappy, a fast compresser/decompresser. ";        byte[] Compressed = snappy.compress (Input.getbytes ("UTF-8"));        byte[] uncompressed = snappy.uncompress (compressed);        string result = new String (Uncompressed, "UTF-8");        SYSTEM.OUT.PRINTLN (result);}    }

Compilation: Javac-classpath./snappy-java-1.1.1.7.jar Testsnappy.java

Execution: Java testsnappy

Execution Result:
Hello snappy-java! Snappy-java is a jni-based wrapper of Snappy, a fast compresser/decompresser.

Copy the "libsnappyjava.so" file to the Hadoop cluster: $JAVA _home/jre/lib/amd64/, Hive on Hadoop, using snappy mode compression is not an error, it's done.

Will spark_home/conf/spark-defaults.conf

Spark.io.compression.codec LZF Note or remove this line, spark will use the default snappy compression method, and no longer an error.

Ps:

It's the dumbest way to fix it now, but it's still going to be a matter of further troubleshooting.

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

The stupidest way to solve the error "Java.lang.UnsatisfiedLinkError:no Snappyjava in Java.library.path" using snappy compression