MySQL Bugs: #95389: slave_rows_search_algorithms='INDEX_SCAN,HASH

Bug #95389	slave_rows_search_algorithms='INDEX_SCAN,HASH_SCAN' lead to 'record not foun'
Submitted:	16 May 2019 9:04	Modified:	5 Jul 2019 20:41
Reporter:	chengqing hu	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Server: Row Based Replication ( RBR )	Severity:	S3 (Non-critical)
Version:	5.7.21	OS:	Linux
Assigned to:	MySQL Verification Team	CPU Architecture:	x86

Description:
In the case where the master-slave data is consistent, the slave sql thread reports that the record cannot be found.

the error info:
Worker 1 failed executing transaction '96b91fd7-bd68-11e8-b2c0-b083fecd29fa:1797272' at master log mysql-bin.000146, end_log_pos 141967282; Could not execute Update_rows event on table tt.t1; Can't find record in 't1', Error_code: 1032; Can't find record in 't1', Error_code: 1032; handler error HA_ERR_KEY_NOT_FOUND; the event's master log mysql-bin.000146, end_log_pos 141967282

How to repeat:
On slave:
set global slave_rows_search_algorithms='INDEX_SCAN,HASH_SCAN';

On master:

1. CREATE TABLE t1 ( 
  A INT UNIQUE KEY, 
  B INT 
); 
insert into t1 values (1,2); 

2. replace into t1 values (1,3),(1,4); 

3. Then there will be an error on the slave

4. On slave:
set global slave_rows_search_algorithms='INDEX_SCAN,HASH_SCAN';
start slave;

5. The error will be resolved

Suggested fix:
The explanation on the reference manual is:

Do not use the combination TABLE_SCAN,HASH_SCAN. This setting forces hashing for all searches. It has no advantage over INDEX_SCAN,HASH_SCAN, and it can lead to “record not found” errors or duplicate key errors in the case of a single event containing multiple updates to the same row, or updates that are order-dependent.

I think it's a bug.

edit synopsis

Hi,

Can you elaborate why you think this is a bug as the quoted documentation is warning that you will have exactly the issue you are having?!

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".

I believe the reporter is referring to hash_scan bug 
that might be related to bug support@oracle team is referring. 
https://support.oracle.com/knowledge/Oracle%20Database%20Products/2473168_1.html

For this support.oracle link, Is there a separate bug that talks about Hash_scan bug?