Two MySQL tables with different codes (utf8, utf8mb4) are joined, resulting in index failure

1. Problem description

1. log_risk_front ( Risk control incoming log )
    There are fields session_id( code utf8),  And there's an index index_session_Id
   
2. risk_result_info ( Risk control incoming parts )
    There are fields session_id( code utf8mb4),  And there's an index index_session_Id

 Two tables ( One to one relationship ,  adopt session_id relation )

--  Found as follows SQL Can't use index  !
SELECT
	t1.session_id,
	t2.session_id
FROM
	log_risk_front t1
	INNER JOIN risk_result_info t2 ON t1.session_id = t2.session_id
WHERE
	t2.session_id = 'OT742714925312245760'

2. Solution

2.1 Permanent solution , Unified ( library , surface , Field ) Character encoding of , If utf8mb4

ALTER TABLE `log_risk_front` 
MODIFY COLUMN `session_id` varchar(64) CHARACTER SET utf8mb4 COLLATE utf8mb4_general_ci NULL DEFAULT NULL COMMENT ' Risk control system session ID';

2.2 A temporary plan 1, use utf8 surface , drive utf8mb4 surface

SELECT
	t1.session_id,
	t2.session_id
FROM
	log_risk_front t1
	INNER JOIN risk_result_info t2 ON t1.session_id = t2.session_id
WHERE
	t1.session_id = 'OT742714925312245760'

2.3 A temporary plan 2, use CONVERT(expr USING transcoding_name)

SELECT
	t1.session_id,
	t2.session_id
FROM
	log_risk_front t1
	INNER JOIN risk_result_info t2 ON t1.session_id = CONVERT(t2.session_id USING UTF8)
WHERE
	t2.session_id = 'OT742714925312245760'

3. Relevant reference

• Function operation on index field , It may get half the result with twice the effort ( Code farmer network )
• MySQL Index invalidation caused by different character sets of table fields ( A pot of tea is not enough )