MySQL Bugs: #100103: Slave SQL for channel 'group_replication_recovery': Could not execute Write

Bug #100103	Slave SQL for channel 'group_replication_recovery': Could not execute Write_rows
Submitted:	3 Jul 2020 13:17	Modified:	7 Aug 2020 0:39
Reporter:	Sankar Narayanan	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Server: Group Replication	Severity:	S1 (Critical)
Version:	8.0.11	OS:	Linux (CentOS)
Assigned to:	MySQL Verification Team	CPU Architecture:	Any
Tags:	MySQL Group Replication

Description:
 [ERROR] [MY-010584] [Repl] Slave SQL for channel 'group_replication_recovery': Could not execute Write_rows event on table **.***; Duplicate entry '2373215' for key 'PRIMARY', Error_code: 1062; handler error HA_ERR_FOUND_DUPP_KEY; the event's master log binlog.000017, end_log_pos 975155589, Error_code: MY-001062

We have been used 5 nodes enabling Group Replication.

Due to the above issue, the entire cluster goes down.

Please suggest to fix the problem.

How to repeat:
1. Create a group with 5 members (node1,node2,node3,node4,node5), using multi-primary mode

2. Create multiple tables using all nodes. Create table script will be executed in any one node for a table.

3. Load data into these tables in parallel using all nodes

4. GR failed with the below error after loading millions of records

[ERROR] [MY-010584] [Repl] Slave SQL for channel 'group_replication_applier': Error executing row event: 'Table 'dbname.tablename' doesn't exist', Error_code: MY-001146

Hi Sankar,

I created a 5 node group replication with 8.0.20 and I cannot reproduce this.

Can you reproduce this issue with 8.0.20 too ?

kind regards
Bogdan

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".