MySQL Bugs: #119458: server cannot restart after crash when relay_log_recovery=1 and have both mgr and normal slave

Bug #119458	server cannot restart after crash when relay_log_recovery=1 and have both mgr and normal slave
Submitted:	25 Nov 9:23	Modified:	25 Nov 9:31
Reporter:	phoenix Zhang (OCA)	Email Updates:
Status:	Open	Impact on me:	None
Category:	MySQL Server: Group Replication	Severity:	S2 (Serious)
Version:	8.0.32, 8.4.4	OS:	Any
Assigned to:		CPU Architecture:	Any

Description:
1. the server configure in group_replication
2. the server still act as a normal slave from another mysql server
3. for group_replication channel, group_replication_start_on_boot=ON
4. for normal slave, relay_log_recovery=1

when this server crash and restart, it will hang and cannot work.

How to repeat:
run new add testcase:
mtr gr_file_pos_replica_recovery_bug --nocheck-testcases

after server1 restart, we can see below log, and server hang:
2025-11-25T09:04:58.819387Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 575.
2025-11-25T09:04:58.819416Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 853.
2025-11-25T09:04:58.819445Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 1131.
2025-11-25T09:04:58.819473Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 1409.
2025-11-25T09:04:58.819501Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 1687.
2025-11-25T09:04:58.819529Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 1965.
2025-11-25T09:04:58.819557Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 2243.
2025-11-25T09:04:58.819588Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 2521.
2025-11-25T09:04:58.819617Z 0 [Note] [MY-010574] [Repl] Slave: MTS group recovery relay log info group_master_log_name server-binary-log.000001, event_master_log_pos 2799.
2025-11-25T09:04:58.822483Z 0 [Note] [MY-010541] [Repl] MTS recovery: starting coordinator thread to fill MTS gaps.

put the file in mysql-test/suite/group_replication/t

Attachment: gr_file_pos_replica_recovery_bug.test (application/octet-stream, text), 3.35 KiB.

also put this file in mysql-test/suite/group_replication/t

Attachment: gr_file_pos_replica_recovery_bug.cnf (application/octet-stream, text), 313 bytes.

and i find an similar bug, https://bugs.mysql.com/bug.php?id=108278. 

from bug#108278, it says dup with Bug#34233264, this is in-correctly. The Bug#34233264 is an innodb bug, which detail is:
InnoDB: After upgrading to MySQL 8.0.29, a failure occurred when attempting to access a table with an instantly added column. (Bug #34233264)

so, this is a mistake, and bug#108278 not fix yet, and still happen in 8.0.32, as this bug way. And it also can happen in 8.4.4.