sqoop的表的导入

#------ hdfs -> mysql ------

create table sqp_order(
create_date date,
user_name varchar(20),
total_volume decimal(10,2)
);

sqoop export \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_order \
-m 1 \
--export-dir /kb12/hive/orderinfo \
--fields-terminated-by '\t'

#若发现导入表有问题 NOTFINDCALSS

[root@singlelucky ~]# find / -name 'sqp_order.class'
/tmp/sqoop-root/compile/e24fd4ec30c24475569ee1569d0873a7/sqp_order.class
/tmp/sqoop-root/compile/04c79abd0daca46e7a2a116716d2eb1f/sqp_order.class
[root@singlelucky ~]# cd /tmp/sqoop-root/compile/
[root@singlelucky compile]# ls
04c79abd0daca46e7a2a116716d2eb1f  e24fd4ec30c24475569ee1569d0873a7
[root@singlelucky compile]# vim /etc/profile.d/myenv.sh
[root@singlelucky compile]# cd 04c79abd0daca46e7a2a116716d2eb1f
[root@singlelucky 04c79abd0daca46e7a2a116716d2eb1f]# ls
sqp_order.class  sqp_order.jar  sqp_order.java
#在jdk中CLASSPATH追加sqp_order.jar
export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JAVA_HOME/lib/sqp_order.jar

#------ mysql -> hdfs ------
#–全量导入

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_order \
-m 1 \
--delete-target-dir \
--target-dir /kb12/sqoop/m2h_all \
--fields-terminated-by '\t'
--lines-terminated-by '\n'

#列裁剪

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_order \
--columns user_name,total_volume \
-m 1 \
--delete-target-dir \
--target-dir /kb12/sqoop/m2h_colcut \
--fields-terminated-by '\t' \
--lines-terminated-by '\n'

#行列裁剪+多个reducer

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_order \
--columns user_name,total_volume \
--where "total_volume>=200" \
-m 2 \
--split-by user_name \
--delete-target-dir \
--target-dir /kb12/sqoop/m2h_rowcut \
--fields-terminated-by ',' \
--lines-terminated-by '\n'


sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--query "select user_name,total_volume from sqp_order where total_volume>=300 and \$CONDITIONS" \
-m 2 \
--split-by user_name \
--delete-target-dir \
--target-dir /kb12/sqoop/m2h_mgt2 \
--fields-terminated-by ',' \
--lines-terminated-by '\n'

#增量导入append|merge

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--query "select * from studentinfo where \$CONDITIONS" \
-m 1 \
--target-dir /kb12/sqoop/m2h_incr_append \
--fields-terminated-by ',' \
--lines-terminated-by '\n' \
--check-column stuId \
--incremental append \
--last-value 0

insert into studentinfo values
(49,'蔡1',32,'男','14568758132',25201,6),
(50,'焦1',28,'男','15314381033',23489,7),
(51,'庞1',23,'男','13892411574',25578,2),
(52,'吴1',27,'男','13063638045',22617,4),
(53,'孟1',32,'男','13483741056',26284,2);

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--query "select * from studentinfo where \$CONDITIONS" \
-m 1 \
--target-dir /kb12/sqoop/m2h_incr_append \
--fields-terminated-by ',' \
--lines-terminated-by '\n' \
--check-column stuId \
--incremental append \
--last-value 48

#增量导入lastmodified

create table sqp_incr_time(
incrName varchar(20),
incrTime timestamp
);

insert into sqp_incr_time(incrName) values('henry'),('pola'),('ariel'),('john'),('mike'),('jerry'),('mary');

insert into sqp_incr_time(incrName,incrTime) values
('jack','2021-06-29 13:21:40.0'),
('rose','2021-06-29 14:21:40.0'),
('xiaoming','2021-06-29 15:21:40.0'),
('anglea','2021-06-29 16:21:40.0'),
('licheng','2021-06-29 17:21:40.0');


sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--query "select * from sqp_incr_time where \$CONDITIONS" \
-m 1 \
--target-dir /kb12/sqoop/m2h_incr_lastmodified \
--fields-terminated-by ',' \
--lines-terminated-by '\n' \
--check-column incrTime \
--incremental lastmodified \
--merge-key incrTime \
--last-value '2021-06-29 13:21:40.0'

--last-value '0000-00-00 00:00:00'

#分区表单分区导入

#开启动态分区：一次性将某张表中的数据写入另一张分区表的多个分区中
create table sqp_partition(
id int,
name varchar(20),
dotime datetime
);

insert into sqp_partition(id,name,dotime) values
(1,'henry','2021-06-01 12:13:14'),
(2,'pola','2021-06-01 12:55:32'),
(3,'ariel','2021-06-01 13:02:55'),
(4,'rose','2021-06-01 13:22:46'),
(5,'jack','2021-06-01 14:15:12');

insert into sqp_partition(id,name,dotime) values
(6,'henry','2021-06-29 12:13:14'),
(7,'pola','2021-06-29 12:55:32'),
(8,'ariel','2021-06-29 13:02:55'),
(9,'rose','2021-06-29 13:22:46'),
(10,'jack','2021-06-29 14:15:12');

create table sqp_partition(
id int,
name string,
dotime timestamp
)
partitioned by (dodate date)
row format delimited
fields terminated by ','
lines terminated by '\n'
stored as textfile;

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_partition \
--where "cast(dotime as date)='2021-06-01'" \
-m 1 \
--delete-target-dir \
--target-dir /user/hive/warehouse/kb12.db/sqp_partition/dodate=2021-06-01 \
--fields-terminated-by ',' \
--lines-terminated-by '\n'

alter table sqp_partition add partition(dodate='2021-06-01');

#xshell脚本

#!/bin/bash

DATE=`date -d '-1 day' +%F`

sqoop import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--table sqp_partition \
--where "cast(dotime as date)='$DATE'" \
-m 1 \
--delete-target-dir \
--target-dir /user/hive/warehouse/kb12.db/sqp_partition/dodate=$DATE \
--fields-terminated-by ',' \
--lines-terminated-by '\n'

hive -e "alter table test.sqp_partition add partition(dodate='$DATE')"
=====================================================================
[root@singlelucky sqoop_job]# vim sq_m2hv_par.sh
[root@singlelucky sqoop_job]# chmod u+x sq_m2hv_par.sh
[root@singlelucky sqoop_job]# ./sq_m2hv_par.sh
=====================================================================

#sqoop job

#vim sqoop-site.xml 将以下注释打开
  <property>
    <name>sqoop.metastore.client.record.password</name>
    <value>true</value>
    <description>If true, allow saved passwords in the metastore.
    </description>
  </property>

#查看job列表

sqoop job --list

#删除job

sqoop job --delete jobname

#查看Job定义

sqoop job --show jobname

#创建job

sqoop job \
--create job_m2hv_par \
-- import \
--connect jdbc:mysql://singlelucky:3306/test \
--username root \
--password kb12kb12 \
--query "select * from sqp_incr_time where \$CONDITIONS" \
-m 1 \
--target-dir /kb12/sqoop/m2h_incr_lastmodified \
--fields-terminated-by ',' \
--lines-terminated-by '\n' \
--check-column incrTime \
--incremental lastmodified \
--append \
--last-value '2021-06-29 13:21:40.0'

#执行job

sqoop job --exec job_m2hv_par

#------ mysql -> hive --------

sqoop import \
--connect jdbc:mysql://192.168.19.130:3306/test \
--username root \
--password kb12kb12 \
--table user_info \
-m 1 \
--hive-import \
--hive-table kb12.user_info \
--create-hive-table 


sqoop import \
--connect jdbc:mysql://192.168.19.130:3306/test \
--username root \
--password kb12kb12 \
--table studentinfo \
--where 'stuId>=50'
-m 1 \
--hive-import \
--hive-table kb12.studentinfo 

#尝试分区导入

create table sqp_user_par(
user_id int,
user_name string,
user_account string,
user_pass string,
user_phone string,
user_gender string,
user_pid string,
user_province string,
user_city string,    
user_district string,
user_address string, 
user_balance decimal(10,2) 
)
partitioned by (id_rang string)
row format delimited
fields terminated by ','
lines terminated by '\n'
stored as textfile;



sqoop import \
--connect jdbc:mysql://192.168.19.130:3306/test \
--username root \
--password kb12kb12 \
--table user_info \
--where "user_id between 1 and 19" \
-m 1 \
--hive-import \
--hive-table kb12.sqp_user_par \
--hive-partition-key id_rang \
--hive-partition-value '1-19' \
--fields-terminated-by ',' \
--lines-terminated-by '\n'

#!/bin/bash

B=$1
E=$2

sqoop import \
--connect jdbc:mysql://192.168.19.130:3306/test \
--username root \
--password kb12kb12 \
--table user_info \
--where "user_id between $B and $E" \
-m 1 \
--hive-import \
--hive-table kb12.sqp_user_par \
--hive-partition-key id_rang \
--hive-partition-value "$B-$E" \
--fields-terminated-by ',' \
--lines-terminated-by '\n'

#------ mysql -> hbase -------

sqoop import \
--connect jdbc:mysql://192.168.19.130:3306/test \
--username root \
--password kb12kb12 \
--table studentinfo \
--where "stuid between 20 and 38" \
--hbase-table test:studentinfo \
--column-family base \
--hbase-create-table \
--hbase-row-key stuid

mysql+jdbc -> hbase-client ->hbase

java代码java2hbase

java -jar prolog-1.0-jar-with-dependencies.jar /root/data/flume/ logconf/logger.properties
java -jar java2hbase-1.0-jar-with-dependencies.jar mysqlToHbaseConfig/datasources.properties

nohup java -jar java2hbase-1.0-jar-with-dependencies.jar mysqlToHbaseConfig/datasources.properties>/dev/null 2>&1 &