sqoop文档:http://sqoop.apache.org/docs/1.4.7/SqoopUserGuide.html#_prerequisites

在hive创建表和导入数据时必须添加分隔符,否则数据导出时会报错

1.下载安装

[root@node1 ~]# wget http://mirrors.shu.edu.cn/apache/sqoop/1.4.7/sqoop-1.4.7.bin__hadoop-2.6.0.tar.gz

[root@node1 ~]# tar xvf sqoop-1.4.7.bin__hadoop-2.6.0.tar.gz -C /opt/

[root@node1 ~]# cd /opt/

[root@node1 opt]# mv sqoop-1.4.7.bin__hadoop-2.6.0/ sqoop-1.4.7

[root@node1 opt]# vim /etc/profile

export SQOOP_HOME=/opt/sqoop-1.4.7

export HADOOP_HOME=/opt/hadoop-2.8.5

export HADOOP_CLASSPATH=/opt/hive-2.3.4/lib/*

export HCAT_HOME=/opt/sqoop-1.4.7/testdata/hcatalog

export ACCUMULO_HOME=/opt/sqoop-1.4.7/src/java/org/apache/sqoop/accumulo

export PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_HOME/bin:$SQOOP_HOME/bin

[root@node1 opt]# source  /etc/profile

[root@node1 opt]# sqoop help             --帮助信息

[root@node1 opt]# sqoop import --help    --参数帮助信息

2.修改yarn配置文件

[root@node1 ~]# vim /opt/hadoop-2.8.5/etc/hadoop/yarn-site.xml

yarn.nodemanager.resource.memory-mb

2048

yarn.nodemanager.resource.cpu-vcores

2

[root@node1 ~]# scp /opt/hadoop-2.8.5/etc/hadoop/yarn-site.xml node2:/opt/hadoop-2.8.5/etc/hadoop/     --将配置文件复制到各节点

yarn-site.xml                   100% 1414   804.3KB/s   00:00

[root@node1 ~]# scp /opt/hive-2.3.4/conf/hive-site.xml /opt/sqoop-1.4.7/conf/     --hive的配置文件也要放在sqoop下面,因为sqoop要调用hive

[root@node1 ~]# stop-all.sh

[root@node1 ~]# start-all.sh

3.将mysql数据导入到HDFS

参数解释:

--append          追加数据

--as-textfile     导入后形成文本文件

--columns         只导入哪些字段

--delete-target-dir    --如果导入的目录存在先删除再导入

--fetch-size        --每次读多少数据

-m                     --起多少任务

-e                      --查询语句(select)

--table    --表名

--target-dir dir              --指定HDFS目录

--warehouse-dir dir      --导入的表将在此目录之下(表名与目录名一至)

--where where clause --where条件

-z                        --数据压缩

--direct               --绕过mysql数据库,直接导入(忧化参数)

[root@node1 ~]# sqoop import --connect jdbc:mysql://172.16.9.100/hive --username hive --password system --table TBL_PRIVS  --target-dir /user/sqoop --direct -m 1 --fields-terminated-by '\t'

[root@node1 ~]# hdfs dfs -ls /user/sqoop       --查看导入的目录

Found 2 items

-rw-r--r--   3 root supergroup          0 2019-03-19 12:43 /user/sqoop/_SUCCESS

-rw-r--r--   3 root supergroup        176 2019-03-19 12:43 /user/sqoop/part-m-00000

[root@node1 ~]# hdfs dfs -cat /user/sqoop/part-m-00000       --查看导入的数据

6,1552878877,1,root,USER,root,USER,INSERT,6

7,1552878877,1,root,USER,root,USER,SELECT,6

8,1552878877,1,root,USER,root,USER,UPDATE,6

9,1552878877,1,root,USER,root,USER,DELETE,6

[root@node1 ~]#

4.将mysql数据导入到hive中

参数详解:

--hive-home dir           指定hive目录

--hive-import               导入到hive

--hive-database           导入指定的库

--hive-overwrite           覆盖到hive

--create-hive-table      在hive中创建表

--hive-table table-name         指定hive表名

--hive-partition-value  v         hive分区

[root@node1 ~]# sqoop import --connect jdbc:mysql://172.16.9.100/hive --username hive --password system --table TBL_PRIVS --target-dir /user/tmp --hive-import --hive-table tt -m 1 --create-hive-table --delete-target-dir --direct --fields-terminated-by '\t'

[root@node1 conf]# hive

Logging initialized using configuration in jar:file:/opt/hive-2.3.4/lib/hive-common-2.3.4.jar!/hive-log4j2.properties Async: true

Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.

hive> show tables;

OK

tt

Time taken: 11.464 seconds, Fetched: 1 row(s)

hive> select * from tt;

OK

615528788771rootUSERrootUSERINSERT6

715528788771rootUSERrootUSERSELECT6

815528788771rootUSERrootUSERUPDATE6

915528788771rootUSERrootUSERDELETE6

Time taken: 3.978 seconds, Fetched: 4 row(s)

hive>

5.将mysql数据导入到hive指定的库中

[root@node1 ~]# sqoop import --connect jdbc:mysql://172.16.9.100/hive --username hive --password system --table TABLE_PARAMS --hive-import --hive-table tt1 -m 1 --create-hive-table --hive-database tong --direct --fields-terminated-by '\t'

[root@node1 conf]# hive

Logging initialized using configuration in jar:file:/opt/hive-2.3.4/lib/hive-common-2.3.4.jar!/hive-log4j2.properties Async: true

Hive-on-MR is deprecated in Hive 2 and may not be available in the future versions. Consider using a different execution engine (i.e. spark, tez) or using Hive 1.X releases.

hive> use tong;

OK

Time taken: 14.34 seconds

hive> show tables;

OK

tt1

Time taken: 0.374 seconds, Fetched: 1 row(s)

hive> select * from tt1;

OK

6numFiles1

6numRows0

6rawDataSize0

6totalSize8

6transient_lastDdlTime1552878901

11commentImported by sqoop on 2019/03/19 15:36:21

11numFiles1

11numRows0

11rawDataSize0

11totalSize176

11transient_lastDdlTime1552981011

16commentImported by sqoop on 2019/03/19 16:04:22

16numFiles1

16numRows0

16rawDataSize0

16totalSize239

16transient_lastDdlTime1552982688

Time taken: 3.004 seconds, Fetched: 17 row(s)

hive>

6.将HDFS的数据导入到mysql中

[root@node1 ~]# hdfs dfs -cat /user/tmp/part-m-00000

12

34

56

[root@node1 ~]# sqoop export --connect jdbc:mysql://172.16.9.100/tong --username tong --password system --export-dir /user/tmp/part-m-00000 --table t1 --direct --fields-terminated-by '\t'

[root@node1 ~]# mysql -u root -psystem

Welcome to the MariaDB monitor.  Commands end with ; or \g.

Your MySQL connection id is 1006876

Server version: 5.6.35 MySQL Community Server (GPL)

Copyright (c) 2000, 2017, Oracle, MariaDB Corporation Ab and others.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

MySQL [(none)]> use tong

MySQL [tong]> select * from t1;

+------+------+

| a    | b    |

+------+------+

|    3 |    4 |

|    5 |    6 |

|    1 |    2 |

+------+------+

3 rows in set (0.00 sec)

MySQL [tong]>

报错信息:(卡在Running job不动,不向下执行)

19/03/19 11:20:09 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1552965562217_0001

19/03/19 11:20:10 INFO impl.YarnClientImpl: Submitted application application_1552965562217_0001

19/03/19 11:20:10 INFO mapreduce.Job: The url to track the job: http://node1:8088/proxy/application_1552965562217_0001/

19/03/19 11:20:10 INFO mapreduce.Job: Running job: job_1552965562217_0001

解决方法:

[root@node1 ~]# vim /opt/hadoop-2.8.5/etc/hadoop/yarn-site.xml   --限制内存,cpu的资源,并将配置文件同步到其它node,重启hadoop服务

yarn.nodemanager.resource.memory-mb

2048

yarn.nodemanager.resource.cpu-vcores

2

[root@node1 ~]#

报错信息:(mysql导入到hive中)

19/03/19 14:34:25 INFO hive.HiveImport: Loading uploaded data into Hive

19/03/19 14:34:25 ERROR hive.HiveConfig: Could not load org.apache.hadoop.hive.conf.HiveConf. Make sure HIVE_CONF_DIR is set correctly.

19/03/19 14:34:25 ERROR tool.ImportTool: Import failed: java.io.IOException: java.lang.ClassNotFoundException: org.apache.hadoop.hive.conf.HiveConf

at org.apache.sqoop.hive.HiveConfig.getHiveConf(HiveConfig.java:50)

at org.apache.sqoop.hive.HiveImport.getHiveArgs(HiveImport.java:392)

at org.apache.sqoop.hive.HiveImport.executeExternalHiveScript(HiveImport.java:379)

解决方法:

[root@node1 ~]# vim /etc/profile    --添加lib变量

export HADOOP_CLASSPATH=/opt/hive-2.3.4/lib/*

[root@node1 ~]# source /etc/profile

报错信息:(是因为sqoop和hive的jackson包冲突)

19/03/19 15:32:11 INFO ql.Driver: Concurrency mode is disabled, not creating a lock manager

19/03/19 15:32:11 INFO ql.Driver: Executing command(queryId=root_20190319153153_63feddd9-a2c8-4217-97d4-23dd9840a54b): CREATE TABLE `tt` ( `TBL_GRANT_ID` BIGINT, `CREATE_TIME` INT,

`GRANT_OPTION` INT, `GRANTOR` STRING, `GRANTOR_TYPE` STRING, `PRINCIPAL_NAME` STRING, `PRINCIPAL_TYPE` STRING, `TBL_PRIV` STRING, `TBL_ID` BIGINT) COMMENT 'Imported by sqoop on 2019/03/19

15:31:49' ROW FORMAT DELIMITED FIELDS TERMINATED BY '\001' LINES TERMINATED BY '\012' STORED AS TEXTFILE

19/03/19 15:32:11 INFO ql.Driver: Starting task [Stage-0:DDL] in serial mode

19/03/19 15:32:12 ERROR exec.DDLTask: java.lang.NoSuchMethodError: com.fasterxml.jackson.databind.ObjectMapper.readerFor(Ljava/lang/Class;)Lcom/fasterxml/jackson/databind/ObjectReader;

at org.apache.hadoop.hive.common.StatsSetupConst$ColumnStatsAccurate.(StatsSetupConst.java:165)

at org.apache.hadoop.hive.common.StatsSetupConst.parseStatsAcc(StatsSetupConst.java:297)

at org.apache.hadoop.hive.common.StatsSetupConst.setBasicStatsState(StatsSetupConst.java:230)

at org.apache.hadoop.hive.common.StatsSetupConst.setBasicStatsStateForCreateTable(StatsSetupConst.java:292)

解决方法:

[root@node1 ~]# mv /opt/sqoop-1.4.7/lib/jackson-* /home/

[root@node1 ~]# cp -a /opt/hive-2.3.4/lib/jackson-* /opt/sqoop-1.4.7/lib/

报错信息:

19/03/19 18:38:40 INFO metastore.HiveMetaStore: 0: Done cleaning up thread local RawStore

19/03/19 18:38:40 INFO HiveMetaStore.audit: ugi=rootip=unknown-ip-addrcmd=Done cleaning up thread local RawStore

19/03/19 18:38:40 ERROR tool.ImportTool: Import failed: java.io.IOException: Hive CliDriver exited with status=1

at org.apache.sqoop.hive.HiveImport.executeScript(HiveImport.java:355)

at org.apache.sqoop.hive.HiveImport.importTable(HiveImport.java:241)

at org.apache.sqoop.tool.ImportTool.importTable(ImportTool.java:537)

at org.apache.sqoop.tool.ImportTool.run(ImportTool.java:628)

解决方法:

create table t1(a int,b int) row format delimited fields terminated by '\t';      --创建表时必须加分隔符

sqoop import --connect jdbc:mysql://172.16.9.100/hive --username hive --password system --table TBL_PRIVS  --target-dir /user/sqoop --direct -m 1 --fields-terminated-by '\t'