本文介绍在E-MapReduce集群中HDFS服务集成Kerberos。
master节点执行:
sudo yum install krb5-server krb5-devel krb5-workstation -yslave节点执行:
sudo yum install krb5-devel krb5-workstation -y备注： 配置中emr-header-1.cluster-xxxx替换成自己集群的hostname
[logging] default = FILE:/var/log/krb5libs.log kdc = FILE:/var/log/krb5kdc.log admin_server = FILE:/var/log/kadmind.log [libdefaults] default_realm = EMR.COM dns_lookup_realm = false dns_lookup_kdc = false ticket_lifetime = 24h renew_lifetime = 7d forwardable = true [realms] EMR.COM = { kdc = emr-header-1.cluster-xxxx admin_server = emr-header-1.cluster-xxxx } [domain_realm] .emr.com = EMR.COM emr.com = EMR.COMb) /var/kerberos/krb5kdc/kdc.conf
[kdcdefaults] kdc_ports = 88 kdc_tcp_ports = 88 [realms] EMR.COM = { #master_key_type = aes256-cts acl_file = /var/kerberos/krb5kdc/kadm5.acl dict_file = /usr/share/dict/words admin_keytab = /var/kerberos/krb5kdc/kadm5.keytab supported_enctypes = aes256-cts:normal aes128-cts:normal des3-hmac-sha1:normal arcfour-hmac:normal des-hmac-sha1:normal des-cbc-md5:normal des-cbc-crc:normal }c) /var/kerberos/krb5kdc/kadm5.acl
*/admin@EMR.COM * slave节点修改配置 只需将上面master节点修改过的/etc/krb5.conf文件拷贝到slave节点对应文件夹即可。在master节点上面执行：
sudo kdb5_util create -r EMR.COM -s备注：若出现Loading random data卡住(需要等一会)，可以另外开一个终端执行一些耗费cpu的操作，增加随机数采集
在master节点执行:
sudo service krb5kdc start sudo service kadmin start在master节点root账号上面执行
$kadmin.local #进入kadmin后继续执行: $addprinc root/admin #输入密码，记住后面执行kadmin时需要输入后续可以在所有集群所有节点上使用kadmin命令来管理Kerberos的一些数据库操作(如添加principal等)
备注:kadmin.local只能在kadmin server所在的机器(即master节点)且拥有root权限情况下才能执行，其它情况使用kadmin
在集群的每个节点上面创建对应的keytab文件，用于HDFS服务各个Daemon(如NameNode/DataNode等)之间的身份认证，防止非法的节点加入集群。
E-MapReduce集群中的HDFS的所有Daemon都是在hdfs账号下启动，所以各个Daemon使用共用相同的keytab配置。
接下来分别在集群的每台机器上面分别执行下面命令：以master节点为例，其它节点按照同样的方式操作
$sudo su hdfs $hostname emr-header-1.cluster-xxxx #后面需要使用hostname $sudo kadmin #输入密码，进入kadmin后执行 # principal使用了上面的hostname即emr-header-1.cluster-xxxx $kadmin: addprinc -randkey hdfs/emr-header-1.cluster-xxxx@EMR.COM $kadmin: addprinc -randkey HTTP/emr-header-1.cluster-xxxx@EMR.COM $kadmin: xst -k hdfs-unmerged.keytab hdfs/emr-header-1.cluster-xxxx@EMR.COM $kadmin: xst -k http.keytab HTTP/emr-header-1.cluster-xxxx@EMR.COM $kadmin: exit #合并http.keytab和hdfs-unmerged.keytab $sudo ktutil #进入ktutil后执行: $ktutil: rkt hdfs-unmerged.keytab $ktutil: rkt http.keytab $ktutil: wkt hdfs.keytab $ktutil: exit #将hdfs.keytab拷贝到/etc/emr/hadoop-conf $sudo cp hdfs.keytab /etc/emr/hadoop-conf $sudo chown hdfs:hadoop /etc/emr/hadoop-conf/hdfs.keytab $sudo chmod 400 /etc/emr/hadoop-conf/hdfs.keytabHDFS服务集成Kerberos需要修改core-site.xml和hdfs-site.xml,如下：
备注: 集群所有节点都需要修改
a) core-site.xml路径: /etc/emr/hadoop-conf/core-site.xml使用hadoop账号来操作sudo su hadoop
添加如下配置项:
<property> <name>hadoop.security.authentication</name> <value>kerberos</value> <!-- A value of "simple" would disable security. --> </property> <property> <name>hadoop.security.authorization</name> <value>true</value> </property>修改如下配置项:将value值master_host_name换成自己集群的master的hostname(如emr-header-1.cluster-xxx)
<property> <name>master_hostname</name> <value>master_host_name</value> </property>b) hdfs-site.xml路径: /etc/emr/hadoop-conf/hdfs-site.xml使用hadoop账号来操作sudo su hadoop
添加如下配置项:
<!-- General HDFS security config --> <property> <name>dfs.block.access.token.enable</name> <value>true</value> </property> <!-- NameNode security config --> <property> <name>dfs.namenode.keytab.file</name> <value>/etc/emr/hadoop-conf/hdfs.keytab</value> <!-- path to the HDFS keytab --> </property> <property> <name>dfs.namenode.kerberos.principal</name> <value>hdfs/_HOST@EMR.COM</value> </property> <property> <name>dfs.namenode.kerberos.internal.spnego.principal</name> <value>HTTP/_HOST@EMR.COM</value> </property> <!-- Secondary NameNode security config --> <property> <name>dfs.secondary.namenode.keytab.file</name> <value>/etc/emr/hadoop-conf/hdfs.keytab</value> <!-- path to the HDFS keytab --> </property> <property> <name>dfs.secondary.namenode.kerberos.principal</name> <value>hdfs/_HOST@EMR.COM</value> </property> <property> <name>dfs.secondary.namenode.kerberos.internal.spnego.principal</name> <value>HTTP/_HOST@EMR.COM</value> </property> <!-- DataNode security config --> <property> <name>dfs.datanode.data.dir.perm</name> <value>700</value> </property> <property> <name>dfs.datanode.keytab.file</name> <value>/etc/emr/hadoop-conf/hdfs.keytab</value> <!-- path to the HDFS keytab --> </property> <property> <name>dfs.datanode.kerberos.principal</name> <value>hdfs/_HOST@EMR.COM</value> </property> <!-- datanode SASL配置 --> <property> <name>dfs.http.policy</name> <value>HTTPS_ONLY</value> </property> <property> <name>dfs.data.transfer.protection</name> <value>integrity</value> </property> <property> <name>dfs.web.authentication.kerberos.principal</name> <value>HTTP/_HOST@EMR.COM</value> </property> <property> <name>dfs.web.authentication.kerberos.keytab</name> <value>/etc/emr/hadoop-conf/hdfs.keytab</value> <!-- path to the HTTP keytab --> </property>HDFS中使用HTTPS来传输数据，需要有keystore相关配置。
在master节点上面执行：
$sudo su hadoop #生成了ca相关文件 $openssl req -new -x509 -keyout ca-key -out ca-cert -days 1000继续在master节点上重复按照如下命令，分别为集群所有节点生成keystore/truststore文件
备注: 每次为新节点重复执行，需要更换命令中的一些文件名称(防止被覆盖)，下面以尖括号(<>)标出
# 以为master节点生成keystore/truststore为例 $keytool -keystore <keystore> -alias localhost -validity 1000 -genkey 输入密钥库口令: 再次输入新口令: 您的名字与姓氏是什么? [Unknown]: emr-header-1 #备注: 不同节点不一样,如emr-worker-1 您的组织单位名称是什么? [Unknown]: EMR 您的组织名称是什么? [Unknown]: EMR 您所在的城市或区域名称是什么? [Unknown]: EMR 您所在的省/市/自治区名称是什么? [Unknown]: EMR 该单位的双字母国家/地区代码是什么? [Unknown]: EMR CN=emr-worker-2, OU=EMR, O=EMR, L=EMR, ST=EMR, C=EMR是否正确? [否]: 是 输入 <localhost> 的密钥口令 (如果和密钥库口令相同, 按回车): $keytool -keystore <truststore> -alias CARoot -import -file ca-cert $keytool -keystore <keystore> -alias localhost -certreq -file <cert-file> #下面命令中your_pwd替换成自己的 $openssl x509 -req -CA ca-cert -CAkey ca-key -in <cert-file> -out <cert-signed> -days 1000 -CAcreateserial -passin pass:your_pwd $keytool -keystore <keystore> -alias CARoot -import -file ca-cert $keytool -keystore <keystore> -alias localhost -import -file <cert-signed>执行完上述命令后，将在当前文件夹下会生成新文件<keystore>和<truststore>拷贝scp到对应机器的/etc/emr/hadoop-conf/目录下
#master节点不需要scp，直接cp过去 $cp keystore /etc/emr/hadoop-conf $cp keystore /etc/emr/hadoop-conf在master节点上面执行
$sudo su hadoop $cp /etc/emr/hadoop-conf/ssl-server.xml.example /etc/emr/hadoop-conf/ssl-server.xml修改，不是覆盖ssl-server.xml文件中相关配置项对应的key备注： 配置中密码需要替换成自己的上面生成keystore/truststore时的密码
<property> <name>ssl.server.truststore.location</name> <value>/etc/emr/hadoop-conf/truststore</value> <description>Truststore to be used by NN and DN. Must be specified. </description> </property> <property> <name>ssl.server.truststore.password</name> <value>YOUR_TRUSTSTORE_PASSWD</value> <description>Optional. Default value is "". </description> </property> <property> <name>ssl.server.keystore.location</name> <value>/etc/emr/hadoop-conf/keystore</value> <description>Keystore to be used by NN and DN. Must be specified. </description> </property> <property> <name>ssl.server.keystore.password</name> <value>YOUR_KEYSTORE_PASSWD</value> <description>Must be specified. </description> </property> <property> <name>ssl.server.keystore.keypassword</name> <value>YOUR_KEYSTORE_PASSWD</value> <description>Must be specified. </description> </property>最后，将master节点的这个ssl-server.xml文件 scp 到其它所有节点/etc/emr/hadoop-conf目录下面。
在master节点上面执行:
$sudo su hdfs #停止集群HDFS服务 $/usr/lib/hadoop-current/sbin/stop-dfs.sh #停止SecondaryNameNode $/usr/lib/hadoop-current/sbin/hadoop-daemon.sh stop secondarynamenode #启动NameNode $/usr/lib/hadoop-current/sbin/hadoop-daemon.sh start namenode #启动SecondaryNameNode $/usr/lib/hadoop-current/sbin/hadoop-daemon.sh start secondarynamenode在slave节点上面执行:
#启动DataNode $sudo su hdfs $/usr/lib/hadoop-current/sbin/hadoop-daemon.sh start datanode在master节点上面执行：
$useradd testkb $sudo su testkb $hadoop fs -ls / 17/05/09 12:04:19 WARN ipc.Client: Exception encountered while connecting to the server : javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] ls: Failed on local exception: 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)]; Host Details : local host is: "emr-header-1.cluster-xxxx/10.26.6.62"; destination host is: "emr-header-1.cluster-xxxx":9000;出现上面错误，说明HDFS服务的Kerberos认证生效了，接着执行：
#从testkb账号退出到root账号执行 # 添加testkb的principal $kadmin.local $kadmin.local: addprinc testkb重新进入testkb账号
$sudo su testkb $hadoop fs -ls / 17/05/09 12:04:19 WARN ipc.Client: Exception encountered while connecting to the server : javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Failed to find any Kerberos tgt)] ls: Failed on local exception: 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)]; Host Details : local host is: "emr-header-1.cluster-xxxx/10.26.6.62"; destination host is: "emr-header-1.cluster-xxxx":9000; #获取testkb的TGT $kinit testkb #验证成功 $hadoop fs -ls / drwxr-xr-x - hadoop hadoop 0 2017-05-09 10:12 /apps drwxr-xr-x - hadoop hadoop 0 2017-05-09 11:57 /spark-history drwxrwxrwx - hadoop hadoop 0 2017-05-09 10:12 /tmp drwxr-xr-x - hadoop hadoop 0 2017-05-09 10:14 /usr