文章开始之前先说明环境情况，这里kylin消费的kafka数据是从Oracle 数据库用Ogg For Bigdata以json格式将数据投递到kafka topic的，投递的时候，关于insert和update 之前的数据投递到名为 ZTVOUCHER_INS 的topic，而delete和update之后的数据投递到名为 ZTVOUCHER_DEL 的topic中，这里主要介绍kylin如何消费数据创建流式cube。
一、源端做DML操作
1.源端表ztvoucher目前没有数据，现在做insert，并查询：
insert into ztvoucher (MANDT, GJAHR, BUKRS, BELNR, BUZEI, MONAT, BUDAT, HKONT, DMBTR, ZZ0014)
values ('666', '2222', '3432', '2200001414', '001', '01', '20190101', '9101000000', 100.00, '101');
1 row created.
SQL> commit;
Commit complete.
SQL> alter system switch logfile;
System altered.
SQL> select * from ztvoucher;
MANDT GJAHR BUKRS BELNR BUZEI MONAT BUDAT HKONT DMBTR ZZ0014
666 2222 3432 2200001414 001 01 20190101 9101000000 100.00 1012.去kafka查看：
[root@hadoop kafka]# ./console.sh
input topic:ZTVOUCHER_INS
Using the ConsoleConsumer with old consumer is deprecated and will be removed in a future major release. Consider using the new consumer by passing [bootstrap-server] instead of [zookeeper]
.{"table":"SCOTT.ZTVOUCHER_INS","op_type":"I","op_ts":"2019-05-22 16:18:58.353767","current_ts":"2019-05-22T16:19:11.352000","pos":"00000000080000012086","tokens":{"TKN-OP-TYPE":"INSERT"},"a
fter":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":100.00,"ZZ0014":"101"}}发现源端做的insert已经在 topic:ZTVOUCHER_INS有了。
3.源端做update操作：
update ztvoucher set dmbtr=50 where mandt='666';
commit;
alter system switch logfile;4.去kafka查看：
先看ZTOVOUCHER_INS 内容：
[root@hadoop kafka]# ./console.sh
input topic:ZTVOUCHER_INS
Using the ConsoleConsumer with old consumer is deprecated and will be removed in a future major release. Consider using the new consumer by passing [bootstrap-server] instead of [zookeeper]
.{"table":"SCOTT.ZTVOUCHER_INS","op_type":"I","op_ts":"2019-05-22 16:18:58.353767","current_ts":"2019-05-22T16:19:11.352000","pos":"00000000080000012086","tokens":{"TKN-OP-TYPE":"INSERT"},"a
fter":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":100.00,"ZZ0014":"101"}}{"table":"SCOTT.ZTVOUCHER_INS","op_type":"I","op_ts":"2019-05-22 16:22:48.354189","current_ts":"2019-05-22T16:23:33.799000","pos":"00000000080000012613","tokens":{"TKN-OP-TYPE":"SQL COMPUPD
ATE"},"after":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":50.00,"ZZ0014":"101"}}发现除了之前的insert操作，现在update之后的数据也进来了。
再看ZTVOUCHER_DEL：
[root@hadoop kafka]# ./console.sh
input topic:ZTVOUCHER_DEL
Using the ConsoleConsumer with old consumer is deprecated and will be removed in a future major release. Consider using the new consumer by passing [bootstrap-server] instead of [zookeeper]
.{"table":"SCOTT.ZTVOUCHER_DEL","op_type":"I","op_ts":"2019-05-22 16:22:48.354189","current_ts":"2019-05-22T16:23:23.781000","pos":"00000000080000012345","tokens":{"TKN-OP-TYPE":"SQL COMPUPD
ATE"},"after":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":100.00,"ZZ0014":"101"}}发现DEL的topic中也存入了update之前的数据。
5.源端做delete操作：
delete from ztvoucher where mandt='666';
commit;
alter system switch logfile;6.去kafka查看：
[root@hadoop kafka]# ./console.sh
input topic:ZTVOUCHER_DEL
Using the ConsoleConsumer with old consumer is deprecated and will be removed in a future major release. Consider using the new consumer by passing [bootstrap-server] instead of [zookeeper]
.{"table":"SCOTT.ZTVOUCHER_DEL","op_type":"I","op_ts":"2019-05-22 16:22:48.354189","current_ts":"2019-05-22T16:23:23.781000","pos":"00000000080000012345","tokens":{"TKN-OP-TYPE":"SQL COMPUPD
ATE"},"after":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":100.00,"ZZ0014":"101"}}{"table":"SCOTT.ZTVOUCHER_DEL","op_type":"I","op_ts":"2019-05-22 16:26:26.353705","current_ts":"2019-05-22T16:27:15.049000","pos":"00000000080000012857","tokens":{"TKN-OP-TYPE":"DELETE"},"a
fter":{"MANDT":"666","GJAHR":"2222","BUKRS":"3432","BELNR":"2200001414","BUZEI":"001","MONAT":"01","BUDAT":"20190101","HKONT":"9101000000","DMBTR":50.00,"ZZ0014":"101"}}发现除了上面update之前的数据以外，还写入了刚做的delete操作的数据。
好了，现在数据都组织好了，现在去流式创建cube。
二、流式构建cube
流式构建cube官方连接(本人用的2.4版本)：
http://kylin.apache.org/cn/docs24/tutorial/cube_streaming.html
流式构建cube需要一个类型为timestamp的时间列字段用来标识消息的时间，从前面两个topic中的json数据可以看到，op_ts字段满足这个要求。
1、用j'son数据定义一张表
先来构建
