转载自:http://blog.csdn.net/viproc/article/details/7570866
情景描述:
按照hadoop手册,搭建Single Node.
启动hadoop时 #./start-all.sh
报错:
#######################################################
localhost: Unrecognized option: -jvm
localhost: Could not create the Java virtual machine.
######################################################
环境:
服务器:RHEL4.7_x86_64 内存:512M
hadoop版本:hadoop-0.20.203.0rc1.tar.gz
java版本:jdk-6u25-linux-x64.bin
解决:
可以修改的地方有两个
第一个(次要的):/usr/local/hadoop/conf/hadoop-env.sh
修改参数: export HADOOP_HEAPSIZE=256 #默认值为1000M,为Java虚拟机占用的内存的大小
第二个(主要的):
查看/usr/local/hadoop/bin/hadoop源码:
####################################################################
if [[ $EUID -eq 0 ]]; then
HADOOP_OPTS="$HADOOP_OPTS -jvm server $HADOOP_DATANODE_OPTS"
else
HADOOP_OPTS="$HADOOP_OPTS -server $HADOOP_DATANODE_OPTS"
fi
####################################################################
如果以root身份启动,则hadoop启动java虚拟机时,会带上-jvm参数。
不同的java版本,启动参数可能会不一致,由此导致以上问题。
其他解释
(详细见链接地址:
http://mail-archives.apache.org/mod_mbox/hadoop-hdfs-user/201105.mbox
/%3CBANLkTinm3q4EAo=GkeMjG67ZnXXRLRV+yw@mail.gmail.com%3E):
####################################################################
> I think, that code is there for jsvc. If you run hadoop in secure mode, it
> executes this path. There is nothing wrong in -jvm option since it meant for
> jsvc.
>
> In case of secure mode, jsvc process runs as root and invokes the datanode
> process as hdfs user (or any other configured user).
> But it seems there is a bug in the code, if you run as root in non-secured
> mode this issue can trigger.
>
####################################################################
因此,
新建一个hadoop用户,
重新初始化操作:#./hadoop namenode -format
再次启动hadoop: #./start-all.sh
正常。
补充: 修改JAVA_HEAP_MAX的大小:
# vi bin/hadoop
JAVA_HEAP_MAX=-Xmx1000m
输出结果:
##############################################################################################################
[hadoop@localhost bin]$ ./start-all.sh
starting namenode, logging to /usr/local/hadoop/bin/../logs/hadoop-hadoop-namenode-localhost.out
localhost: starting datanode, logging to /usr/local/hadoop/bin/../logs/hadoop-hadoop-datanode-localhost.out
localhost: starting secondarynamenode, logging to /usr/local/hadoop/bin/../logs/hadoop-hadoop-secondarynamenode-localhost.out
starting jobtracker, logging to /usr/local/hadoop/bin/../logs/hadoop-hadoop-jobtracker-localhost.out
localhost: starting tasktracker, logging to /usr/local/hadoop/bin/../logs/hadoop-hadoop-tasktracker-localhost.out
####################################################################################