Hi, You should compile Hadoop source code with -Drequire.snappy, such as: mvn -DskipTests -Pnative,dist -Dtar,require.snappy package
but you also need to install snappy before your compiling.. On Wed, Mar 25, 2015 at 9:39 AM, 王鹏飞 <[email protected]> wrote: > How to recompiling library?Cause my hadoop was built from tar ball and in > /lib/native there exist files like *.a and *.so.Does I need build from > source code to recompiling library? > > On Tue, Mar 24, 2015 at 6:35 PM, Tsuyoshi Ozawa <[email protected]> wrote: > >> Sometimes compiled native libraries included in tar ball doesn't work >> correctly - how about recompiling library on your environment? >> >> Thanks, >> - Tsuyoshi >> >> On Tue, Mar 24, 2015 at 6:18 PM, 王鹏飞 <[email protected]> wrote: >> > I noticed a map-reduce job encountered an >> > >> Exception:java.lang.UnsatisfiedLinkError:org.apache.hadoop.util.NativeCodeLoader.buildSupportsSnappy()Z >> > I googled it and released that it was lack of snappy tool.I >> > Installed the snappy and copied libsnappy.so* to $HADOOP_HOME/lib/native >> > In .bashrc added: >> > export LD_LIBRARY_PATH="$HADOOP_HOME/lib/native/" >> > export JAVA_LIBRARY_PATH="$HADOOP_HOME/lib/native/" >> > export >> > HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib/native" >> > In core-site.xml added: >> > <property> >> > <name>io.compression.codecs</name> >> > <value> >> > >> > org.apache.hadoop.io.compress.GzipCodec, >> > >> > org.apache.hadoop.io.compress.DefaultCodec, >> > >> > org.apache.hadoop.io.compress.BZip2Codec, >> > >> > org.apache.hadoop.io.compress.SnappyCodec >> > </value> >> > </property> >> > In mapred-site.xml added: >> > <property> >> > >> <name>mapreduce.map.output.compress</name> >> > <value>true</value> >> > </property> >> > >> > <property> >> > >> <name>mapred.map.output.compress.codec</name> >> > >> > <value>org.apache.hadoop.io.compress.SnappyCodec</value> >> > </property> >> > >> > <property> >> > <name>mapreduce.admin.user.env</name> >> > >> > >> <value>LD_LIBRARY_PATH=/home/hadoop/software/hadoop-2.4.1/lib/native</value> >> > </property> >> > In yarn-site.xml added: >> > <property> >> > <name>yarn.app.mapreduce.am.env</name> >> > >> > >> <value>LD_LIBRARY_PATH=/home/hadoop/software/hadoop-2.4.1/lib/native</value> >> > </property> >> > Finally make the same setting in the datanode ,but it still >> didn't >> > work 。But in a Pseudo-Distributed cluster, I only copy the >> libsnappy.so* to >> > $HADOOP_HOME/lib/native without changing the configuration ,it works >> well >> > .Why in this cluster it appears so difficult? >> > >> > >> > >> > >> > >> > >> > >
