MySQL Bugs: #108797: MGR Sencondary node don`t appling relay logs

Bug #108797	MGR Sencondary node don`t appling relay logs
Submitted:	17 Oct 2022 9:00	Modified:	18 Oct 2022 1:38
Reporter:	YUANHENG WANG	Email Updates:
Status:	Can't repeat	Impact on me:	None
Category:	MySQL Server: Group Replication	Severity:	S3 (Non-critical)
Version:	8.0.26	OS:	Linux (CentOS Linux release 7.6.1810)
Assigned to:	MySQL Verification Team	CPU Architecture:	Any
Tags:	group_replication, replication

Description:
In the MGR cluster of Sencondary node don`t appling relay logs,The COUNT_TRANSACTIONS_REMOTE_IN_APPLIER_QUEUE has been growing,but there are no exceptions in the error log.

How to repeat:
We don't have the ability to repeat this scene.

Other:
mysql> SELECT * FROM performance_schema.replication_group_members;
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| CHANNEL_NAME              | MEMBER_ID                            | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | MEMBER_VERSION |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| group_replication_applier | 49ca1288-4c49-11ed-bc98-98f18159dd6e | 10.165.21.3 |        3306 | ONLINE       | SECONDARY   | 8.0.26         |
| group_replication_applier | 763459bc-5f0f-11ec-84bc-5ca72143194d | 10.165.21.2 |        3306 | ONLINE       | SECONDARY   | 8.0.26         |
| group_replication_applier | a9afbb5a-5f11-11ec-b918-642fc7bc5ef9 | 10.165.21.1 |        3306 | ONLINE       | PRIMARY     | 8.0.26         |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
3 rows in set (0.00 sec)

By command “show master status;”,The Executed_Gtid_Set result is unchanged ；
By Executed_Gtid_Set viewing the corresponding records in the relay_log, we find that the card is stuck in a simple update statement。

The other secondary node is fine, and relay_log application is not stuck。

Hi,

8.0.26 is rather old, have you upgraded in the meantime (8.0.31 available currently)?

I am unable to reproduce this with 8.0.31, did you maybe fetch the show slave status from the secondary with problem?

Thanks

We can also consider upgrading, but there is no judgment on whether the upgrade has an effect

Thank you for your reply.

Hi,

I cannot reproduce the issue with 8.0.31 so unless you do there's not much we can do about it.