need help for version 8.0.15 group replicaton

[winsun Hall]

hi,dear all.

I use mysql-8.015 to setup a Group Replication cluster.It is a Single Master group with 3 nodes.I follow the advice here (https://mysqlhighavailability.com/setting-up-mysql-group-replication-with-mysql-docker-images/). After setup the cluster run up normally. node1 is the master ,node2 and node3 are the slaves;


I want to test some GR fault tolerance scenarios.I just use "docker kill node3" and then insert some data on node1.


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 | 28e9a66d-26b1-11e9-9f6b-0242ac130002 | node1 | 3306 | ONLINE | PRIMARY | 8.0.15 |
| group_replication_applier | 28f0fb75-26b1-11e9-9f22-0242ac130003 | node2 | 3306 | ONLINE | SECONDARY | 8.0.15 |
| group_replication_applier | 28f26da5-26b1-11e9-a080-0242ac130004 | node3 | 3306 | ONLINE | SECONDARY | 8.0.15 |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
3 rows in set (0.00 sec)

mysql> insert into test.t1 values(2,'hws2');

then the node1 has no response. I change to node2 and check the cluster status.


mysql> 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 | 28e9a66d-26b1-11e9-9f6b-0242ac130002 | node1 | 3306 | UNREACHABLE | PRIMARY | 8.0.15 |
| group_replication_applier | 28f0fb75-26b1-11e9-9f22-0242ac130003 | node2 | 3306 | ONLINE | SECONDARY | 8.0.15 |
| group_replication_applier | 28f26da5-26b1-11e9-a080-0242ac130004 | node3 | 3306 | UNREACHABLE | SECONDARY | 8.0.15 |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
3 rows in set (0.00 sec)

mysql>

I double check the steps and get the same result!

Because I am a newbie , maybe I miss something Important . So I download the version 8.0.13 and repeat the steps just as what I do with 8.0.15. The 8.0.13 cluster just work fine !

mysql> SELECT * FROM performance_schema.replication_group_members;
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | RSION |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| group_replication_applier | 174c7349-269d-11e9-9ca0-0242ac130002 | node1 | 3306 | ONLINE | SECONDARY | 8.0.13 |
| group_replication_applier | 17531092-269d-11e9-a16f-0242ac130003 | node2 | 3306 | ONLINE | PRIMARY | 8.0.13 |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
2 rows in set (0.00 sec)



Is something change in 8.0.15 and I miss in setup steps? Looking forward to your reply,thanks!




The init docker command is just like this:

for N in 1 2 3
do docker run -d --name=node$N --net=groupnet --hostname=node$N \
-e MYSQL_ROOT_PASSWORD=mypass \
mysql/mysql-server:8.0.15 \
--server-id=$N \
--log-bin=binlog \
--relay-log=node$N-relay-bin \
--binlog-format=ROW \
--enforce-gtid-consistency='ON' \
--log-slave-updates='ON' \
--gtid-mode='ON' \
--transaction-write-set-extraction='XXHASH64' \
--binlog-checksum='NONE' \
--master-info-repository='TABLE' \
--relay-log-info-repository='TABLE' \
--plugin-load='group_replication.so' \
--relay-log-recovery='ON' \
--group-replication-start-on-boot='OFF' \
--group-replication-group-name='aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee' \
--group-replication-local-address="node$N:33061" \
--group-replication-group-seeds='node1:33061,node2:33061,node3:33061'
done