一，spark streaming load写入失败

Reason: column count mismatch, expect=25 real=1. src line: [demo-masking	1	2018-02-17	1	B35C300672_1	B600FA855B	72d2b68c1f79fd73d8910ff388685ab0	9cac8970477ec95540ec64a1b2450ee3	2	0.0070	3	0.0076	1.0000	0	0	0.0000	0.0000		0.00	0.00	0	0	0	2018-02-21 21:38:05	2018-02-21 21:38:05]; 
Reason: column count mismatch, expect=25 real=1. src line: [demo-masking	1	2018-02-19	1	B35C300672_1	B600FA855B	72d2b68c1f79fd73d8910ff388685ab0	9cac8970477ec95540ec64a1b2450ee3	2	0.0038	2	0.0030	1.0000	0	0	0.0000	0.0000		0.00	0.00	0	0	0	2018-02-21 21:36:17	2018-02-21 21:36:17]; 

原因是分隔符设置为"\t"，没有加转义字符，改为"\\t"即正常。

二，SQL 错误 [1064] [42000]: Key columns should be a ordered prefix of the schema.

建表时UNIQUE KEY必须放在所有字段前面声明且保证有序

三，Reason: column(type)) value is null while columns is not nullable. src line: [[‘finance’, ‘15’, ‘100001’, ‘152’, ‘0’, ‘2021-11-03T11:55:57’, ‘2021-11-03T11:55:57’, ‘0’]];

写入StarRocks时报错，但对应的数据并不为空。

Reason: column(type)) value is null while columns is not nullable. src line: [['finance', '16', '100001', '153', '0', '2021-11-03T11:55:57', '2021-11-03T11:55:57', '0']]; 
Reason: column(type)) value is null while columns is not nullable. src line: [['finance', '3', '1', '152', '0', '2021-10-20T18:03:12', '2021-10-20T18:03:12', '0']]; 

原因是type对应的类型在StarRocks建表是是tinyint，导致int类型的值插不进去。

解决办法：将type的类型改为int