Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Incomplete
-
10.4.7
-
Galera Cluster with three 10.4.7-MariaDB, Linux version 3.10 centos
Description
Galera Cluster with 3 nodes failed, respectively all nodes was in Read-Only state. The count of cluster members was still 3.
At the same time there were a lot of messages in Server-log like
"mysqld: 2019-10-16 3:15:20 0 [Note] WSREP: Created page /var/lib/mysql/gcache.page.000988 of size 134217728 bytes"
At the time of the problem we load about 40 GB of data (Insert Statemenst).
Our running galera config is
[galera]
binlog_format=ROW
default-storage-engine=innodb
innodb_autoinc_lock_mode=2
bind-address=0.0.0.0
wsrep_on=ON
wsrep_provider=/usr/lib64/galera-4/libgalera_smm.so
wsrep_provider_options = "gcache.size=2G;gcs.fc_factor=0.8"
wsrep_slave_threads=8
wsrep_sst_method=rsync
The problem could only be solved by
Node 3: stop
Node 2: stop
Node 1: stop
Node 1: galera_new_cluster
Node 2: rm -rf /var/lib/mysql/*
Node 2: start
Node 3: start