oozie4.3.0+sqoop1.4.6实现mysql到hive的增量抽取

1.准备数据源

mysql中表bigdata,数据如下：

2. 准备目标表

目标表存放hive中数据库dw_stg表bigdata

保存路径为 hdfs://localhost:9000/user/hive/warehouse/dw_stg.db/bigdata

hive中建表语句如下：

create external table bigdata(
class_id string comment '课程id',
class_name string comment '课程名称',
class_month int comment '课程周期',
teacher string comment '课程老师',
update_time string comment '更新日期'
) 
partitioned by(dt string comment '年月日')
row format delimited
fields terminated by '01'
lines terminated by '
'
stored as textfile;

注意点： 字段分隔符使用01，行分隔符使用 ，增加表分区dt格式为yyyMMdd

在hive中创建上面表bigdata.

3. 编写oozie脚本文件

3.1 配置job.properties

# 公共变量
timezone=Asia/Shanghai
jobTracker=dwtest-name1:8032
nameNode=hdfs://dwtest-name1:9000
queueName=default
warehouse=/user/hive/warehouse
dw_stg=${warehouse}/dw_stg.db
dw_mdl=${warehouse}/dw_mdl.db
dw_dm=${warehouse}/dw_dm.db
app_home=/user/oozie/app
oozie.use.system.libpath=true

# coordinator
oozie.coord.application.path=${nameNode}${app_home}/bigdata/coordinator.xml
workflow=${nameNode}${app_home}/bigdata

# source
connection=jdbc:mysql://192.168.1.100:3306/test
username=test
password=test
source_table=bigdata

# target
target_path=${dw_stg}/bigdata

# 脚本启动时间，结束时间
start=2018-01-24T10:00+0800
end=2199-01-01T01:00+0800

3.2  配置coordinator.xml

<coordinator-app name="coord_bigdata" frequency="${coord:days(1)}" start="${start}" end="${end}" timezone="${timezone}" xmlns="uri:oozie:coordinator:0.5">
    <action>
        <workflow>
            <app-path>${workflow}</app-path>
            <configuration>
                <property>
                    <name>startTime</name>
                    <value>${coord:formatTime(coord:dateOffset(coord:nominalTime(), -1, 'DAY'), 'yyyy-MM-dd 00:00:00')}</value>
                </property>
                <property>
                    <name>endTime</name>
                    <value>${coord:formatTime(coord:dateOffset(coord:nominalTime(), 0, 'DAY'), 'yyyy-MM-dd 00:00:00')}</value>
                </property>
                <property>
                    <name>outputPath</name>
                    <value>${target_path}/dt=${coord:formatTime(coord:dateOffset(coord:nominalTime(), 0, 'DAY'), 'yyyyMMdd')}/</value>
                </property>
            </configuration>
        </workflow>

    </action>
</coordinator-app>

注意点：

增量的开始时间startTime获取:   当前时间的前一天 输出值为 2018-01-23 00:00:00

${coord:formatTime(coord:dateOffset(coord:nominalTime(), -1, 'DAY'), 'yyyy-MM-dd 00:00:00')}

增量的结束时间endTime获取:  输出值为 2018-01-24 00:00:00

${coord:formatTime(coord:dateOffset(coord:nominalTime(), 0, 'DAY'), 'yyyy-MM-dd 00:00:00')}

输出路径需要带上分区字段dt:  输出值 /user/hive/warehouse/dw_stg.db/bigdata/dt=20180124/

${target_path}/dt=${coord:formatTime(coord:dateOffset(coord:nominalTime(), 0, 'DAY'), 'yyyyMMdd')}/

3.3 配置workflow.xml

<?xml version="1.0" encoding="UTF-8"?>
<!--
  Licensed to the Apache Software Foundation (ASF) under one
  or more contributor license agreements.  See the NOTICE file
  distributed with this work for additional information
  regarding copyright ownership.  The ASF licenses this file
  to you under the Apache License, Version 2.0 (the
  "License"); you may not use this file except in compliance
  with the License.  You may obtain a copy of the License at

       http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License.
-->
<workflow-app xmlns="uri:oozie:workflow:0.4" name="wf_bigdata">
    <start to="sqoop-node"/>

    <action name="sqoop-node">
        <sqoop xmlns="uri:oozie:sqoop-action:0.2">
            <job-tracker>${jobTracker}</job-tracker>
            <name-node>${nameNode}</name-node>
            <prepare>
                <delete path="${nameNode}${outputPath}"/>
            </prepare>
            <configuration>
                <property>
                    <name>mapred.job.queue.name</name>
                    <value>${queueName}</value>
                </property>
            </configuration>
            <arg>import</arg>
            <arg>--connect</arg>
            <arg>${connection}</arg>
            <arg>--username</arg>
            <arg>${username}</arg>
            <arg>--password</arg>
            <arg>${password}</arg>
            <arg>--verbose</arg>
            <arg>--query</arg>
            <arg>select class_id,class_name,class_month,teacher,update_time from ${source_table} where $CONDITIONS and update_time &gt;= '${startTime}' and update_time &lt; '${endTime}'</arg>
            <arg>--fields-terminated-by</arg>
            <arg>01</arg>
            <arg>--target-dir</arg>
            <arg>${outputPath}</arg>
            <arg>-m</arg>
            <arg>1</arg>
        </sqoop>
        <ok to="end"/>
        <error to="fail"/>
    </action>

    <kill name="fail">
        <message>Sqoop free form failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
    </kill>
    <end name="end"/>
</workflow-app>

4. 上传脚本

将以上3个文件上传到hdfs的oozie目录app下如下：

5. 执行job

oozie job -config job.properties -run

6. 查看job状态

7. 查询hive中表

使用 msck repair table bigdata 自动修复分区，然后查询结果，测试没用问题。

 8. 开发中遇到的坑如下：

 8.1 workflow.xml中字段分隔符不能带单引号。正确的是<arg>01</arg> ，错误的是<arg>'01'</arg>

 8.2 由于sqoop的脚本配置在xml中，所以在判断条件时使用小于号"<"会报错,xml文件校验不通过。

      解决方法使用 &lt; 代替 "<"  ,所以使用大于号时最好也使用 &gt;代替 ">"