Recovery of MySQL 8 Cluster instance after server crash and corrupted data in log event

There is a MySQL 8 Cluster InnoDB of three servers and one of the server crashed with a bad RAM. The same setup is described here – Install and deploy MySQL 8 InnoDB Cluster with 3 nodes under CentOS 8 and MySQL Router for HA. The failed server got restarted without clean shutdown and after booting up the MySQL Cluster node tried to recover automatically, but the recover process failed and the node left the group of the three server:

2022-05-31T04:00:00.322469Z 24 [ERROR] [MY-011620] [Repl] Plugin group_replication reported: 'Fatal error during the incremental recovery process of Group Replication. The server will leave the group.'
2022-05-31T04:00:00.322489Z 24 [Warning] [MY-011645] [Repl] Plugin group_replication reported: 'Skipping leave operation: concurrent attempt to leave the group is on-going.'
2022-05-31T04:00:00.322500Z 24 [ERROR] [MY-011712] [Repl] Plugin group_replication reported: 'The server was automatically set into read only mode after an error was detected.'
2022-05-31T04:00:03.448475Z 0 [System] [MY-011504] [Repl] Plugin group_replication reported: 'Group membership changed: This member has left the group.'

The recovery process proposed here follows these steps

  1. Connect with mysqlsh (MySQL Shell) to a MySQL instance, which is currently a part of the cluster group. The member, which left the group is not part any more, though the MySQL Cluster status shows it is part of the cluster topology, but with error.
  2. Remove the bad instance from the MySQL Cluster with removeInstance
  3. Add the instance with addInstance and the recovery process will kick in. The type of the recovery process will be chosen by the setup if not specified. In this case, the setup chooses the Incremental state recovery over (full) clone mode.
  4. Initiate the cluster rescan operation to recovery the group replication and the MySQL Cluster.

mysql

Summery of the recovery process

  • The recovery process was successful.
  • The distributed recovery with Incremental state recovery has finished for 24 hours for 200Mbyte database, which is really strange and the speed was really bad. The instance uses ordinary disks, not SSDs and a 1Gbps network.
  • No need to change or manage the MySQL Router in any of the steps or the recovery stages. It handled the situation from the very beginning by removing the bad instance and then adding it again only after the recovery process had finished successfully.
  • MySQL Shell should be connected to an healthy instance currently a part of the Cluster.

In the console output logs all commands and important lines are highlighted.

STEP 1) Remove the bad instance from the cluster.

The status of the cluster with the bad instance.

[root@db-cluster-3 ~]# mysqlsh
MySQL Shell 8.0.28

Copyright (c) 2016, 2022, Oracle and/or its affiliates.
Oracle is a registered trademark of Oracle Corporation and/or its affiliates.
Other names may be trademarks of their respective owners.

Type '\help' or '\?' for help; '\quit' to exit.
 MySQL  JS > \connect clusteradmin@db-cluster-1
Creating a session to 'clusteradmin@db-cluster-1'
Fetching schema names for autocompletion... Press ^C to stop.
Closing old connection...
Your MySQL connection id is 39806649 (X protocol)
Server version: 8.0.28 MySQL Community Server - GPL
No default schema selected; type \use <schema> to set one.
 MySQL  db-cluster-1:33060+ ssl  JS > var cluster = dba.getCluster()
 MySQL  db-cluster-1:33060+ ssl  JS > cluster.status()
{
    "clusterName": "mycluster1", 
    "defaultReplicaSet": {
        "name": "default", 
        "primary": "db-cluster-1:3306", 
        "ssl": "REQUIRED", 
        "status": "OK_NO_TOLERANCE", 
        "statusText": "Cluster is NOT tolerant to any failures. 1 member is not active.", 
        "topology": {
            "db-cluster-1:3306": {
                "address": "db-cluster-1:3306", 
                "memberRole": "PRIMARY", 
                "mode": "R/W", 
                "readReplicas": {}, 
                "replicationLag": null, 
                "role": "HA", 
                "status": "ONLINE", 
                "version": "8.0.28"
            }, 
            "db-cluster-2:3306": {
                "address": "db-cluster-2:3306", 
                "memberRole": "SECONDARY", 
                "mode": "R/O", 
                "readReplicas": {}, 
                "replicationLag": null, 
                "role": "HA", 
                "status": "ONLINE", 
                "version": "8.0.28"
            }, 
            "db-cluster-3:3306": {
                "address": "db-cluster-3:3306", 
                "instanceErrors": [
                    "ERROR: group_replication has stopped with an error."
                ], 
                "memberRole": "SECONDARY", 
                "memberState": "ERROR", 
                "mode": "R/O", 
                "readReplicas": {}, 
                "role": "HA", 
                "status": "(MISSING)", 
                "version": "8.0.28"
            }
        }, 
        "topologyMode": "Single-Primary"
    }, 
    "groupInformationSourceMember": "db-cluster-1:3306"
}

Keep on reading!