解决办法：每台主机上执行此脚本 if [ $# -ne 1 ]; then echo \ echo \ hostsFile\ exit 1 fi HOSTS_FILE=$1 LOCALDATE=`date +%Y%m%d` for ip in `cat ${HOSTS_FILE}` do ssh $ip echo never > /sys/kernel/mm/redhat_transparent_hugepage/defrag ssh $ip echo never > /sys/kernel/mm/redhat_transparent_hugepage/enabled ssh $ip echo never > /sys/kernel/mm/transparent_hugepage/enabled ssh $ip echo never > /sys/kernel/mm/transparent_hugepage/defrag ssh $ip python /usr/lib/python2.6/site-packages/ambari_agent/HostCleanup.py --silent --skip=users done
6.4. 选择安装服务
Installed Service Version Description Yes HDFS 2.6.0.2.2.0.0 Apache Hadoop Distributed File System Yes YARN + MapReduce2 2.6.0.2.2.0.0 Apache Hadoop NextGen MapReduce (YARN) Yes Tez 0.5.2.2.2.0.0 Tez is the next generation Hadoop Query Processing framework written on top of YARN. Yes Nagios 3.5.0 Nagios Monitoring and Alerting system Yes Ganglia 3.5.0 Ganglia Metrics Collection system (RRDTool will be installed too) Yes Hive 0.14.0.2.2.0.0 Data warehouse system for ad-hoc queries & analysis of large datasets and table & storage management service Yes HBase 0.98.4.2.2.0.0 Non-relational distributed database and centralized service for configuration management & synchronization Yes Pig 0.14.0.2.2.0.0 Scripting platform for analyzing large datasets Yes Sqoop 1.4.5.2.2.0.0 Tool for transferring bulk data between Apache Hadoop and structured data stores such as relational databases Yes Oozie 4.1.0.2.2.0.0 System for workflow coordination and execution of Apache Hadoop jobs. This also includes the installation of the optional Oozie Web Console which relies on and will install the ExtJS Library. Yes ZooKeeper 3.4.6.2.2.0.0 Centralized service which provides highly reliable distributed coordination Yes Falcon 0.6.0.2.2.0.0 Data management and processing platform Yes Storm 0.9.3.2.2.0.0 Apache Hadoop Stream processing framework Yes Flume 1.5.2.2.2.0.0 A distributed service for collecting, aggregating, and moving large amounts of streaming data into HDFS Yes Slider 0.60.0.2.2.0.0 A framework for deploying, managing and monitoring existing distributed applications on YARN. Yes Knox 0.5.0.2.2.0.0 Provides a single point of authentication and access for Apache Hadoop services in a cluster Yes Kafka 0.8.1.2.2.0.0 A high-throughput distributed messaging system 根据需要选择，可以根据硬件资源的情况选择服务，平衡服务在各个节点的分布
分配节点和客户端 修复错误
[root@hadoop-db jdk1.7.0_71]# ambari-server setup --jdbc-db=oracle --jdbc-driver=/usr/share/java/ojdbc6-11.2.0.3.jar Using python /usr/bin/python Setup ambari-server Copying /usr/share/java/ojdbc6-11.2.0.3.jar to /var/lib/ambari-server/resources If you are updating existing jdbc driver jar for oracle with ojdbc6-11.2.0.3.jar. Please remove the old driver jar, from all hosts. Restarting services that need the driver, will automatically copy the new jar to the hosts. JDBC driver was successfully initialized. Ambari Server 'setup' completed successfully.
Summary Admin Name : admin Cluster Name : xxxxx_hadoop_sit Total Hosts : 9 (9 new) Repositories: ? redhat6 (HDP-2.6): http://172.29.0.25/HDP/HDP-2.6.0/HDP/centos6/ redhat6 (HDP-UTILS-1.1.0.21): http://172.29.0.25/HDP/HDP-UTILS/ ? Services: ? HDFS o DataNode : 7 hosts o NameNode : hadoop-db o NFSGateway : 7 hosts o SNameNode : hadoop01 ?
? ?
?
? ? ?
? ?
? ? ?
? ? ?
?
?
YARN + MapReduce2
o App Timeline Server : hadoop01 o NodeManager : 7 hosts
o ResourceManager : hadoop01 Tez
o Clients : 9 hosts Hive
o Metastore : hadoop01 o HiveServer2 : hadoop-db
o WebHCat Server : hadoop-db
o Database : Existing Oracle Database HBase
o Master : hadoop01 o RegionServer : 3 hosts
o Phoenix Query Server : 3 hosts Pig
o Clients : 9 hosts Sqoop
o Clients : 9 hosts Oozie
o Server : hadoop01
o Database : Existing Oracle Database ZooKeeper
o Server : 3 hosts Storm
o DRPC Server : hadoop01 o Nimbus : hadoop01 o UI Server : hadoop01 o Supervisor : 4 hosts Flume
o Flume : 3 hosts Ambari Infra
o Infra Solr Instance : hadoop-db Ambari Metrics
o Metrics Collector : hadoop01 o Grafana : hadoop-db Kafka
o Broker : 4 hosts Knox
o Gateway : hadoop-db SmartSense
o Activity Analyzer : hadoop-db o Activity Explorer : hadoop-db o HST Server : hadoop-db Spark2
o Livy for Spark2 Server : 7 hosts o History Server : hadoop01 o Thrift Server : 1 host Slider
o Clients : 9 hosts