YARN & HDFS2 安装和配置Kerberos

今天尝试在Hadoop 2.x开发集群上配置Kerberos，遇到一些问题，记录一下

设置hadoop security

core-site.xml

        <property>

           <name>hadoop.security.authentication</name>

           <value>kerberos</value>

        </property>

        <property>

           <name>hadoop.security.authorization</name>

           <value>true</value>

        </property>

hadoop.security.authentication默认是simple方式，也就是基于文件系统的验证方式，这里我们改为kerberos

设置hdfs security

hdfs-site.xml

        <property>

                <name>dfs.block.access.token.enable</name>

                <value>true</value>

        </property>

        <property>

                <name>dfs.https.enable</name>

                <value>false</value>

        </property>

        <property>

                <name>dfs.namenode.https-address</name>

                <value>dev80.hadoop:50470</value>

        </property>

        <property>

                <name>dfs.https.port</name>

                <value>50470</value>

        </property>

        <property>

                <name>dfs.namenode.keytab.file</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>dfs.namenode.kerberos.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.namenode.kerberos.https.principal</name>

                <value>host/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.namenode.secondary.http-address</name>

                <value>dev80.hadoop:50090</value>

        </property>

        <property>

                <name>dfs.namenode.secondary.https-port</name>

                <value>50470</value>

        </property>

        <property>

                <name>dfs.namenode.secondary.keytab.file</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>dfs.namenode.secondary.kerberos.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.namenode.secondary.kerberos.https.principal</name>

                <value>host/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.datanode.data.dir.perm</name>

                <value>700</value>

        </property>

        <property>

                <name>dfs.datanode.address</name>

                <value>0.0.0.0:1003</value>

        </property>

        <property>

                <name>dfs.datanode.http.address</name>

                <value>0.0.0.0:1007</value>

        </property>

        <property>

                <name>dfs.datanode.https.address</name>

                <value>0.0.0.0:1005</value>

        </property>

        <property>

                <name>dfs.datanode.keytab.file</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>dfs.datanode.kerberos.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.datanode.kerberos.https.principal</name>

                <value>host/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.datanode.data.dir.perm</name>

                <value>700</value>

        </property>

        <property>

                <name>dfs.datanode.address</name>

                <value>0.0.0.0:1003</value>

        </property>

        <property>

                <name>dfs.datanode.http.address</name>

                <value>0.0.0.0:1007</value>

        </property>

        <property>

                <name>dfs.datanode.https.address</name>

                <value>0.0.0.0:1005</value>

        </property>

        <property>

                <name>dfs.datanode.keytab.file</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>dfs.datanode.kerberos.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.datanode.kerberos.https.principal</name>

                <value>host/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>dfs.web.authentication.kerberos.principal</name>

                <value>HTTP/_HOST@DIANPING.COM</value>

        </property>

        <property>

              <name>dfs.web.authentication.kerberos.keytab</name>

              <value>/etc/hadoop.keytab</value>

              <description>

                            The Kerberos keytab file with the credentials for the

                                  HTTP Kerberos principal used by Hadoop-Auth in the HTTP endpoint.

              </description>

        </property>

dfs.datanode.address表示data transceiver RPC server所绑定的hostname或IP地址，如果开启security，端口号必须小于1024，否则的话启动datanode时候会报“Cannot start secure cluster without privileged resources”错误

namenode和secondary namenode都是以hadoop用户身份启动

datanode需要以root用户身份用jsvc来启动，而Hadoop 2.x自身带的jsvc是32位版本的，需要去jsvc官网上重新下载编译

1. wget http://mirror.esocc.com/apache//commons/daemon/binaries/commons-daemon-1.0.15-bin.tar.gz

2. cd src/native/unix; configure; make

生成jsvc 64位executable，把它拷贝到$HADOOP_HOME/libexec

[hadoop@dev80 unix]$ file jsvc
jsvc: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped

3. mvn package

编译commons-daemon-1.0.15.jar，拷贝到$HADOOP_HOME/share/hadoop/hdfs/lib下，同时删除自带版本的commons-daemon jar包

hadoop-env.sh中修改

# The jsvc implementation to use. Jsvc is required to run secure datanodes.

export JSVC_HOME=/usr/local/hadoop/hadoop-2.1.0-beta/libexec

# On secure datanodes, user to run the datanode as after dropping privileges

export HADOOP_SECURE_DN_USER=hadoop

# The directory where pid files are stored. /tmp by default

export HADOOP_SECURE_DN_PID_DIR=/usr/local/hadoop

# Where log files are stored in the secure data environment.

export HADOOP_SECURE_DN_LOG_DIR=/data/logs

分发配置和jar到整个集群

用hadoop帐号启动namenode，然后切换到root，再启动datanode，发现namenode web页面上有显示"

Security is
ON

设置yarn security

yarn-site.xml

        <property>

                <name>yarn.resourcemanager.keytab</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>yarn.resourcemanager.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>yarn.nodemanager.keytab</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>yarn.nodemanager.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

        <property>

                <name>yarn.nodemanager.container-executor.class</name>

                <value>org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor</value>

        </property>

        <property>

                <name>yarn.nodemanager.linux-container-executor.group</name>

                <value>hadoop</value>

        </property>

container-executor默认是DefaultContainerExecutor，是以起Nodemanager的用户身份启动container的，切换为LinuxContainerExecutor会以提交application的用户身份来启动，它使用一个setuid可执行文件来启动和销毁container

这个可执行文件在bin/container-executor，不过Hadoop默认带的还是32位版本，所以需要重新编译

下载Hadoop 2.x source code

mvn package -Pdist,native -DskipTests -Dtar -Dcontainer-executor.conf.dir=/etc

注：container-executor.conf.dir必须显示注明，它表示setuid可执行文件依赖的配置文件路径，默认会在$HADOOP_HOME/etc/hadoop下，不过由于该文件需要父目录和以上的目录的owner都为root，要不然会有以下报错，所以为了方便我们设置为/etc

Caused by: org.apache.hadoop.util.Shell$ExitCodeException: File /usr/local/hadoop/hadoop-2.1.0-beta/etc/hadoop must be owned by root, but is owned by 500

        at org.apache.hadoop.util.Shell.runCommand(Shell.java:458)

        at org.apache.hadoop.util.Shell.run(Shell.java:373)

        at org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:578)

        at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.init(LinuxContainerExecutor.java:147)

默认的寻找configuration路径

[root@dev80 bin]# strings container-executor | grep etc

../etc/hadoop/container-executor.cfg

看出来是默认加载$HADOOP_HOME/etc/hadoop
/container-executor.cfg

加上container-executor.conf.dir=/etc 再编译后

[hadoop@dev80 bin]$ strings container-executor | grep etc

/etc/container-executor.cfg

container-executor.cfg中设置

yarn.nodemanager.linux-container-executor.group=hadoop

min.user.id=499

将container-executor拷贝到$HADOOP_HOME/bin

chown root:hadoop container-executor /etc/container-executor.cfg

chmod 4750 container-executor

chmod 400 /etc/container-executor.cfg

同步配置文件到整个集群，用hadoop帐号启动ResourceManager和Nodemanager

设置jobhistory server security

mapred-site.xml

        <property>

                <name>mapreduce.jobhistory.keytab</name>

                <value>/etc/hadoop.keytab</value>

        </property>

        <property>

                <name>mapreduce.jobhistory.principal</name>

                <value>hadoop/_HOST@DIANPING.COM</value>

        </property>

启动JobHistoryServer

sbin/mr-jobhistory-daemon.sh start historyserver

执行命令kinit,获得一张tgt(ticket granting ticket)

[hadoop@dev80 hadoop]$ kinit -r 24l -k -t /home/hadoop/.keytab hadoop

[hadoop@dev80 hadoop]$ klist

Ticket cache: FILE:/tmp/krb5cc_500

Default principal: hadoop@DIANPING.COM

Valid starting     Expires            Service principal

09/11/13 15:25:34  09/12/13 15:25:34  krbtgt/DIANPING.COM@DIANPING.COM

 renew until 09/12/13 15:25:34

其中/tmp/krb5cc_500就是ticket cache file, 500表示hadoop帐号的uid，默认会读取

用户也可以通过设置export KRB5CCNAME=/tmp/krb5cc_500来指定ticket cache路径

用完之后可以kdestroy销毁掉该ticket cache

如果本地没有ticket cache，会报如下错误

13/09/11 16:21:35 ERROR security.UserGroupInformation: PriviledgedActionException as:hadoop (auth:KERBEROS) cause:java.io.IOException: javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)]

附上keytab中的principal

[hadoop@dev80 hadoop]$ klist -k -t /etc/hadoop.keytab

Keytab name: WRFILE:/etc/hadoop.keytab

KVNO Timestamp         Principal

---- ----------------- --------------------------------------------------------

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 hadoop/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 host/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

   1 06/17/12 22:01:24 HTTP/dev80.hadoop@DIANPING.COM

本文链接http://blog.csdn.net/lalaguozhe/article/details/11570009
，转载请注明

YARN & HDFS2 安装和配置Kerberos的更多相关文章

（转）RedHat/CentOS安装和配置kerberos
RedHat/CentOS安装和配置kerberos 需要在kerberos server和客户端都先安装ntp (Internet时间协议,保证服务器和客户机时间同步 ) 1 kerberos 服 ...
CentOS6安装各种大数据软件第九章：Hue大数据可视化工具安装和配置
相关文章链接 CentOS6安装各种大数据软件第一章:各个软件版本介绍 CentOS6安装各种大数据软件第二章:Linux各个软件启动命令 CentOS6安装各种大数据软件第三章:Linux基础 ...
Hadoop-2.6.0 集群的安装与配置
1. 配置节点bonnie1 hadoop环境 (1) 下载hadoop- 2.6.0 并解压缩 [root@bonnie1 ~]# wget http://apache.fayea.com/had ...
Spark(三): 安装与配置
参见 HDP2.4安装(五):集群及组件安装 ,安装配置的spark版本为1.6, 在已安装HBase.hadoop集群的基础上通过 ambari 自动安装Spark集群,基于hadoop yarn ...
在虚拟机VM中安装的Ubuntu上安装和配置Hadoop
一.系统环境: 我使用的Ubuntu版本是:ubuntu-12.04-desktop-i386.iso jdk版本:jdk1.7.0_67 hadoop版本:hadoop-2.5.0 二.下载jdk和 ...
完全分布式Hadoop2.3安装与配置
一.Hadoop基本介绍 Hadoop优点 1.高可靠性:Hadoop按位存储和处理数据 2.高扩展性:Hadoop是在计算机集群中完成计算任务,这个集群可以方便的扩展到几千台 3.高效性:Hadoo ...
Mysql多实例安装以及配置
MySQL多实例 1.什么是MySQL多实例简单地说,Mysql多实例就是在一台服务器上同时开启多个不同的服务端口(3306.3307),同时运行多个Mysql服务进程,这些服务进程通过不同的soc ...
在Linux上怎么安装和配置Apache Samza
samza是一个分布式的流式数据处理框架(streaming processing),它是基于Kafka消息队列来实现类实时的流式数据处理的.(准确的说,samza是通过模块化的形式来使用kafka的 ...
浅谈 zookeeper 原理,安装和配置
当前云计算流行, 单一机器额的处理能力已经不能满足我们的需求,不得不采用大量的服务集群.服务集群对外提供服务的过程中,有很多的配置需要随时更新,服务间需要协调工作,那么这些信息如何推送到各个节点?并且 ...

随机推荐

IM与工作信息流整合
IM与工作信息流整合,希望减轻用户“信息”负担从36氪此前的<“明道”现在推出个人免费版本,是怎样的逻辑?>一文中,我们可以了解到,国内现在的协作产品设计思路主要有两种:一种是像t ...
Linux相关问题-CentOS6.5 x64版本号下Tomcat无法自启动的解决的方法
前段时间使用阿里云server.使用的是Linux CentOS6.5系统,在搭建完Tomcat后发现,Tomcat无法自启动. 将启动tomcat的命令为tomcat_home/bin/startu ...
MessageDigest简单介绍
本文博客原文參考文章:http://blog.sina.com.cn/s/blog_4f36423201000c1e.html 一.概述 java.security.MessageDigest类用于 ...
理解SQL SERVER中的分区表
转自:http://www.cnblogs.com/sienpower/archive/2011/12/31/2308741.html 简介分区表是在SQL SERVER2005之后的版本引入的特性 ...
Java学习之位运算和逻辑运算符
今天看了一下HashMap类的源码,在HashMap的源码中定义了初始空间的大小 static final int DEFAULT_INITIAL_CAPACITY = 1 << 4; 当 ...
NOIP2015酱油记
day0 坐动车到广州..下午就在酒店颓... day1 早上6:30起床...大概8:00到六中..ZSJZ众貌似很晚才到..毕竟他们酒店就在学校门口(真的就刚刚好是门口...),大概8:15进去机 ...
BZOJ 1620: [Usaco2008 Nov]Time Management 时间管理( 二分答案 )
二分一下答案就好了... --------------------------------------------------------------------------------------- ...
ELK 之四：搭建集群处理日PV 四亿次超大访问量优化方法
最近公司的网站访问量越来越大,采用4台高配置服务器做后端Server,前端使用一个负载,日志从后端4台服务器收集到ELK统计,但是最近Logstash经常出问题,每次启动运行三四个小时就挂了,分析是由 ...
通过代码...CGRectmake 创建的控件如何自适应屏幕。
[[UIScreen mainScreen] bounds].size.width [[UIScreen mainScreen] bounds].size.height 得到屏幕的宽高,然后将变量写入 ...
django最佳实践
导入的时候使用绝对导入或者清晰的相对导入相对导入用法: from __future__ import absolute_import from .models import what_u_need ...

YARN & HDFS2 安装和配置Kerberos

YARN & HDFS2 安装和配置Kerberos的更多相关文章

随机推荐

热门专题