【正文】 sessionid = 0x136f1fa35960000, negotiated timeout = 30000WATCHER::WatchedEvent state:SyncConnected type:None path:nullHive安裝（數(shù)據庫）Cd /media/shared/sudo tar xzvf ./ C /home/app/vim ~/.bashrcexport HIVE_HOME={{pwd}}修改環(huán)境變量export PATH=${JAVA_HOME}bin:${HIVE_HOME}bin:$PATHvim export HADOOP_HOME=/home/app/cp 設置主要環(huán)境變量（手動）export HADOOP_HOME=/home/app/(你自己的hadoop安裝路徑) 在hdfs建立目錄用來保存hive數(shù)據$ $HADOOP_HOME/bin/hadoop fs mkdir /tmp $ $HADOOP_HOME/bin/hadoop fs mkdir /user/hive/warehouse $ $HADOOP_HOME/bin/hadoop fs chmod g+w /tmp $ $HADOOP_HOME/bin/hadoop fs chmod g+w /user/hive/warehouse 運行運行 bin/hive 即可hive SET =localhost:50030。hive show tables。(此表創(chuàng)建時沒有指定數(shù)據分隔方式和行分隔方式,下面將刪除它重新建)OKTime taken: secondshive drop table log_stat。\t39。\n39。(指定數(shù)據分隔方式’\t’,行分隔’\n’,以文本存儲)hive load data local inpath 39。 overwrite into table log_stat。Found 1 itemsdrwxrxrx root supergroup 0 20120621 10:19 /user/hive/warehouse/log_stathive dfs ls /user/hive/warehouse/log_stat。/media/shared/reg_339。(將表數(shù)據導入本地文件)hive load data inpath 39。 overwrite into table tt。hdfs://localhost:9000/user/root/outlog/click_mini/2012062839。hive set =true。外部表創(chuàng)建桶不會拆分目錄hive select * from ad_3rd tablesample(bucket 3 out of 3 on rand())。這句執(zhí)行有問題select count(*) from ad_3rd tablesample(bucket 1 out of 3 on access_date)。hive39。12345639。hive39。%39。賦權mysql select user()。sudo aptget install mysqlserver使用上面的命令下載的版本較低移除mysqlsudo aptget autoremove purge sudo aptget remove mysqlserversudo aptget autoremove mysqlserversudo aptget remove mysqlmon (非常重要)sudo tar zxvf ./ C/home/apphive連接mysqlconf/property name/name !valuejdbc:derby:。create=true/value valuejdbc:mysql://localhost:3306/hive?createDatabaseIfNotExist=true/value descriptionJDBC connect string for a JDBC metastore/description/propertyproperty name/name !value/value value/value descriptionDriver class name for a JDBC metastore/description/propertyproperty name/name !valueAPP/value valuehive/value descriptionusername to use against metastore database/description/propertyproperty name/name !valuemine/value value123456/value descriptionpassword to use against metastore database/description/property復制mysqljdbc驅動至hive rootzhuweiVirtualBox:/home/app/ bin/hivehive show tables。Reading table information for pletion of table and column namesYou can turn off this feature to get a quicker startup with ADatabase changedmysql select * from TBLS。 hadoop安裝目錄/contrib/eclipseplugin/ 到 eclipse安裝目錄/plugins/ 下。 如果安裝插件成功，打開WindowPreferens，你會發(fā)現(xiàn)Hadoop Map/Reduce選項，在這個選項里你需要配置Hadoop installation directory。 。 在Map/Reduce Locations中新建一個Hadoop Location。在彈出的對話框中你需要配置Location name，如myubuntu，還有Map/Reduce Master和DFS Master。如： Map/Reduce Master Java代碼 1. localhost 2. 9000localhost9000配置完后退出。 第三步，新建項目。 復制 hadoop安裝目錄/src/example/org/apache/hadoop/example/。WorldWorld)rootzhuweiVirtualBox:/home/app/(HelloGoodbye~/.bashrc16