MySQL Bugs: #103808: group repl MY-013328: Certification information is too large for transmission

Bug #103808	group repl MY-013328: Certification information is too large for transmission
Submitted:	26 May 2021 4:09	Modified:	30 Jun 2021 13:14
Reporter:	beon feng	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Server: Group Replication	Severity:	S3 (Non-critical)
Version:	8.0.22	OS:	CentOS (CentOS Linux release 7.8.2003 (Core))
Assigned to:	MySQL Verification Team	CPU Architecture:	x86

Description:
when running sysbench on a MGR cluster, the mysql instance (on which sysbench is running to) exit group, and the errors is below:

2021-05-25T16:20:15.877971+08:00 15487 [ERROR] [MY-013328] [Repl] Plugin group_replication reported: 'The certification information could not be set in this server: 'Certification information is too large for transmission.''
2021-05-25T16:20:15.878027+08:00 15487 [ERROR] [MY-011624] [Repl] Plugin group_replication reported: 'Error when processing certification information in the incremental recovery process'
2021-05-25T16:20:15.878043+08:00 15487 [ERROR] [MY-011620] [Repl] Plugin group_replication reported: 'Fatal error during the incremental recovery process of Group Replication. The server will leave the group.' 

How to repeat:
1. build a mgr cluster
2. running sysbench on 130.46 which is also a node of mgr, the cmd as follows:
sysbench /tool/sysbench-1.0.20/src/lua/oltp_read_write.lua --mysql-host=172.16.130.47 --mysql-port=3306 --mysql-user=root --mysql-password=root --mysql-db=sbtest --tables=30 --table_size=5000000 --report-interval=2 --threads=200 --db-driver=mysql --skip-trx=on --db-ps-mode=disable --create-secondary=off  --time=600 --simple-ranges=0 --sum-ranges=0 --order-ranges=0 --distinct-ranges=0 run
3. mgr instance of 130.47 exit with following message in error.log

2021-05-25T16:20:15.877971+08:00 15487 [ERROR] [MY-013328] [Repl] Plugin group_replication reported: 'The certification information could not be set in this server: 'Certification information is too large for transmission.''
2021-05-25T16:20:15.878027+08:00 15487 [ERROR] [MY-011624] [Repl] Plugin group_replication reported: 'Error when processing certification information in the incremental recovery process'
2021-05-25T16:20:15.878043+08:00 15487 [ERROR] [MY-011620] [Repl] Plugin group_replication reported: 'Fatal error during the incremental recovery process of Group Replication. The server will leave the group.' 

Suggested fix:
1. self-adapte the parameters such as slave_parallel_workers to make cluster optimal;
2. optimize the framework and treatment

Hi,

This looks like duplicate of https://bugs.mysql.com/bug.php?id=99778

can you retry this with 8.0.25

Thanks
Bogdan

ok, thanks for suggestion.
I have read the bug-99778, and I have a question:
The reason for bug-99778 is that the certification information is too big after cloneing procedure, which happens on group_replication_recovery .
Then considering the common points of this two, is the cause of my bug-103880 poor disk capacity of the server during performance test？in other words, if flashback speed is slow or takes a long time, then "Certification information is too large for
transmission." happens. whether group_replication_recovery or group_replication_applier ?

Hi,

It was happening when disk/network performance was not good or the database is too big anyhow, it happened when cloning took a considerable amount of time. This should be fixed in .22 and even more tuned in .25 that's why I asked if you can reproduce this with 8.0.25, as with 8.0.25 you should not have this problem anymore. If you can also reproduce it with 8.0.25 please let us know.

ok, thanks, 
I will retry that with 8.0.25.

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".