hadoop 安装总结

系统基础设置

系统的请关闭ipv6

linux检测ipv6 关闭方式

cat /proc/sys/net/ipv6/conf/all/disable_ipv6

如果是0 的话就是关闭是1就是开启

防火墙关闭

Ubuntu sudo ufw disable

Linux sudo service iptable

修改/etc/hosts 文件

你要加入的主机的ip地址名称加入其中

并在其他主机做同样操作

修改主机名称

Vi /etc/hostsname

安装ssh 并设定无验证连接

Ubuntu apt-get install ssh

Linux yum install openssh*

生成秘钥

Ssh-keygen –t rsa –P “”

在你所在权限下会出现

复制 id_rsa.pub 到其他的datanode 上

Scp authorized_keys Hadoop@datanode1:~/.ssh

安装Java

用alien 转rpm 到deb

设置/etc/profile

export JAVA_HOME=/usr/java/jdk1.8.0_25

export CLASSPATH=.:$JAVA_HOME/lib:$HADOOP_HOME/lib:$CLASSPATH

export PATH=$JAVA_HOME/bin

Hadoop 应用于64位安装

安装 svn

下载 Hadoop 源代

Yum install autofaunto make libtool cmake

Yum install ncurses-devel

Yum install opensll-ssl

Yum intall gcc*

安装maven

安装 ant

安装 protobuf

mvn install -DskipTests

mvn package -Pdist,native -DskipTests -Dtar -Dmaven.javadoc.skip=true

hbase 选择，选择 hbase-x.x.x-hadoop2

完全进程启动

hbase regionserver 安装系统时间要同步

ntpdate -u time.nist.gov

并使用 date 查看所有节点的时间是否正常

配置好安装好系统，看到所有的进程都已经起来了，那么请查看hbase下的logs 是否会出现什么问题，

例如:

2014-11-28 09:51:41,249 WARN [regionserver60020-SendThread(hd3.hthorizon.com:2181)] zookeeper.ClientCnxn: Session 0x149f41a1cfb0000 for server null, unexpected error, closing socket connection and attempting reconnect

java.net.ConnectException: Connection refused

at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)

at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:716)

at org.apache.zookeeper.ClientCnxnSocketNIO.doTransport(ClientCnxnSocketNIO.java:361)

at org.apache.zookeeper.ClientCnxn$SendThread.run(ClientCnxn.java:1081)

这是说明 zookeeper 没有起来，需要去zookeeper 的bin 下查看zookeeper.out, zookeeper 出了什么问题

eclipse 连接HBASE

配置好hbase 之后

从windows 系统上连接服务器上的hadoop，我只是使用了网上的代码，应用全部hbase 目录下lib 中所有的jar 包，然后开始测试，会出现以下问题

系统不显示log 进程所以需要配置log4j.properties，如果日志不出现查看日志文件把所有的log4j.appender.myLogFile.threshold=info

log4j.appender.myConsole.threshold=info

windows 下系统连接，调试hdaoop 系统那 windows系统需要下载安装

winutils.exe 并在系统环境下配置HADOOP_HOME,winutils.exe 在此路径的bin 文件下，如果没有 bin 文件夹，要建立一个.

如果发现和hadoop 系统连接上了，但是操作hbase 的数据库命令运转超时，那么他们会报各种错误比如连接超时等，估计是找不到host文件的主机名，请完全复制到windows 上， Hadoop 服务器上的主机配置在windows 下设定好，hosts 文件在 C:\WINDOWS\system32\drivers\etc

错误：

找不到ssh 服务

sed: -e expression #1, char 6: unknown option to `s'

-c: Unknown cipher type 'cd'

master: namenode running as process 4191. Stop it first.

guard.: ssh: Could not resolve hostname guard.: Name or service not known

now.: ssh: Could not resolve hostname now.: Name or service not known

noexecstack'.: ssh: Could not resolve hostname noexecstack'.: Name or service not known

try: ssh: Could not resolve hostname try: Name or service not known

fix: ssh: Could not resolve hostname fix: Name or service not known

HotSpot(TM): ssh: Could not resolve hostname HotSpot(TM): Name or service not known

The authenticity of host 'java (127.0.53.53)' can't be established.

ECDSA key fingerprint is bf:cb:c6:3a:cc:92:df:85:0f:d3:0f:f0:2d:f5:b4:9d.

Are you sure you want to continue connecting (yes/no)? <libfile>',: ssh: Could not resolve hostname <libfile>',: Name or service not known

it: ssh: Could not resolve hostname it: Name or service not known

have: ssh: Could not resolve hostname have: Name or service not known

loaded: ssh: Could not resolve hostname loaded: Name or service not known

highly: ssh: Could not resolve hostname highly: Name or service not known

VM: ssh: Could not resolve hostname VM: Name or service not known

It's: ssh: Could not resolve hostname It's: Name or service not known

have: ssh: Could not resolve hostname have: Name or service not known

the: ssh: Could not resolve hostname the: Name or service not known

recommended: ssh: Could not resolve hostname recommended: Name or service not known

'-z: ssh: Could not resolve hostname '-z: Name or service not known

fix: ssh: Could not resolve hostname fix: Name or service not known

library: ssh: Could not resolve hostname library: Name or service not known

which: ssh: Could not resolve hostname which: Name or service not known

Client: ssh: Could not resolve hostname Client: Name or service not known

or: ssh: Could not resolve hostname or: Name or service not known

the: ssh: Could not resolve hostname the: Name or service not known

will: ssh: Could not resolve hostname will: Name or service not known

that: ssh: Could not resolve hostname that: Name or service not known

to: ssh: connect to host to port 22: Connection refused

with: ssh: Could not resolve hostname with: Name or service not known

'execstack: ssh: Could not resolve hostname 'execstack: Name or service not known

You: ssh: Could not resolve hostname You: Name or service not known

disabled: ssh: Could not resolve hostname disabled: Name or service not known

stack: ssh: Could not resolve hostname stack: Name or service not known

you: ssh: Could not resolve hostname you: Name or service not known

might: ssh: Could not resolve hostname might: Name or service not known

with: ssh: Could not resolve hostname with: Name or service not known

The: ssh: Could not resolve hostname The: Name or service not known

VM: ssh: Could not resolve hostname VM: Name or service not known

stack: ssh: Could not resolve hostname stack: Name or service not known

warning:: ssh: Could not resolve hostname warning:: Name or service not known

guard: ssh: Could not resolve hostname guard: Name or service not known

library: ssh: Could not resolve hostname library: Name or service not known

link: ssh: Could not resolve hostname link: Name or service not known

解决方法：

在hadoop-evn.xml加入

export HADOOP_OPTS="$HADOOP_OPTS -Djava.net.preferIPv4Stack=true - Djava.library.path=$HADOOP_PREFIX/lib"

测试命令

shirdrn@master:~/cloud/hadoop2/hadoop-2.0.4-alpha$ bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.0.4-alpha.jar randomwriter out

hdfs-size 参数表

参数 取值范例
详细说明

dfs.default.chunk.view.size 32768
namenode的http访问页面中针对每个文件的内容显示大小，通常无需设置。

dfs.datanode.du.reserved 1073741824
每块磁盘所保留的空间大小，需要设置一些，主要是给非hdfs文件使用，默认是不保留，0字节

dfs.name.dir /opt/data1/hdfs/name,

/opt/data2/hdfs/name,

/nfs/data/hdfs/name NN所使用的元数据保存，一般建议在nfs上保留一份，作为1.0的HA方案使用，也可以在一台服务器的多块硬盘上使用

dfs.web.ugi nobody,nobody
NN,JT等所使用的web tracker页面服务器所使用的用户和组

dfs.permissions true | false
dfs权限是否打开，我一般设置false，通过开发工具培训别人界面操作避免误操作，设置为true有时候会遇到数据因为权限访问不了。

dfs.permissions.supergroup Supergroup
设置hdfs超级权限的组，默认是supergroup，启动hadoop所使用的用户通常是superuser。

dfs.data.dir

/opt/data1/hdfs/data,

/opt/data2/hdfs/data,

/opt/data3/hdfs/data, 真正的datanode数据保存路径，可以写多块硬盘，逗号分隔

dfs.datanode.data.dir.perm 755
datanode所使用的本地文件夹的路径权限，默认755

dfs.replication 3
hdfs数据块的复制份数，默认3，理论上份数越多跑数速度越快，但是需要的存储空间也更多。有钱人可以调5或者6

dfs.replication.max 512
有时dn临时故障恢复后会导致数据超过默认备份数。复制份数的最多数，通常没什么用，不用写配置文件里。

dfs.replication.min 1
最小份数，作用同上。

dfs.block.size 134217728
每个文件块的大小，我们用128M，默认是64M。这个计算需要128*1024^2，我碰上过有人直接写128000000，十分浪漫。

dfs.df.interval 60000
磁盘用量统计自动刷新时间，单位是毫秒。

dfs.client.block.write.retries 3
数据块写入的最多重试次数，在此次数之前不会捕获失败。

dfs.heartbeat.interval 3
DN的心跳检测时间间隔。秒

dfs.namenode.handler.count 10
NN启动后展开的线程数。

dfs.balance.bandwidthPerSec 1048576
做balance时所使用的每秒最大带宽，使用字节作为单位，而不是bit

dfs.hosts /opt/hadoop/conf/hosts.allow
一个主机名列表文件，这里的主机是允许连接NN的，必须写绝对路径，文件内容为空则认为全都可以。

dfs.hosts.exclude /opt/hadoop/conf/hosts.deny
基本原理同上，只不过这里放的是禁止访问NN的主机名称列表。这在从集群中摘除DN会比较有用。

dfs.max.objects 0
dfs最大并发对象数，HDFS中的文件，目录块都会被认为是一个对象。0表示不限制

dfs.replication.interval 3
NN计算复制块的内部间隔时间，通常不需写入配置文件。默认就好

dfs.support.append true | false
新的hadoop支持了文件的APPEND操作，这个就是控制是否允许文件APPEND的，但是默认是false，理由是追加还有bug。

dfs.datanode.failed.volumes.tolerated 0
能够导致DN挂掉的坏硬盘最大数，默认0就是只要有1个硬盘坏了，DN就会shutdown。

dfs.secondary.http.address 0.0.0.0:50090
SNN的tracker页面监听地址和端口

dfs.datanode.address 0.0.0.0:50010
DN的服务监听端口，端口为0的话会随机监听端口，通过心跳通知NN

dfs.datanode.http.address 0.0.0.0:50075
DN的tracker页面监听地址和端口

dfs.datanode.ipc.address 0.0.0.0:50020
DN的IPC监听端口，写0的话监听在随机端口通过心跳传输给NN

dfs.datanode.handler.count 3
DN启动的服务线程数

dfs.http.address 0.0.0.0:50070
NN的tracker页面监听地址和端口

dfs.https.enable true | false
NN的tracker是否监听在HTTPS协议，默认false

dfs.datanode.https.address 0.0.0.0:5047
DN的HTTPS的tracker页面监听地址和端口

dfs.https.address 0.0.0.0:50470
NN的HTTPS的tracker页面监听地址和端口

参考文章
http://www.cnblogs.com/huligong1234/p/4136331.html
Habse 相关网站是：
http://hbase.apache.org/book.html#basic.prerequisites