I have a problem with my third node. It's a new node and when I start it all the replication is performed and the mysqld stay ready, but at any given time the mysqld is killed.
[mysqld]
explicit_defaults_for_timestamp = true
sql_mode = NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
datadir = /data/database
socket = /var/lib/mysql/mysql.sock
user = mysql
tmpdir = /var/tmp
secure_file_priv = ""
# Cache
thread-cache-size = 4096
table-open-cache = 4096
table-definition-cache = 2048
query-cache-size = 128M
query-cache-limit = 8M
query_cache_type = 1
# Per-thread Buffers
sort-buffer-size = 8M
read-rnd-buffer-size = 8M
join-buffer-size = 4M
read_buffer_size = 8M
# Temp Tables
tmp-table-size = 1G
max-heap-table-size = 2G
# Networking
skip-name-resolve
back-log = 100
max-connections = 400
max-connect-errors = 1000
max-allowed-packet = 48M
interactive-timeout = 600
wait-timeout = 7200
connect_timeout = 120
innodb-log-buffer-size = 8M
innodb_log_file_size = 384M
innodb-file-per-table = 1
innodb-open-files = 65535
innodb_lock_wait_timeout = 60
transaction-isolation=READ-COMMITTED
# Performance schema
performance_schema
performance_schema_events_waits_history_size=20
performance_schema_events_waits_history_long_size=15000
# Logging
log-warnings = 2
long-query-time = 5
slow-query-log = ON
slow-query-log-file = /var/log/mysql/mysqld.slow
log-slow-slave-statements
log-slow-admin-statements
log-queries-not-using-indexes
binlog_format = ROW
bind-address = 0.0.0.0
default_storage_engine = innodb
innodb_autoinc_lock_mode = 2
innodb_flush_log_at_trx_commit = 0
# InnoDB speed hack
innodb_buffer_pool_instances = 16
innodb_buffer_pool_size = 16G
# Disk Tunning
innodb_flush_method = O_DIRECT
innodb_io_capacity = 3000
wsrep_provider=/usr/lib64/galera-3/libgalera_smm.so
wsrep_provider_options="gcache.size=1G;gcache.page_size=300M;socket.ssl_key=/etc/my.cnf.d/server-key.pem;socket.ssl_cert=/etc/my.cnf.d/server-cert.pem;socket.ssl_ca=/etc/my.cnf.d/ca-cert.pem;evs.keepalive_period=PT3S;evs.suspect_timeout=PT1M;evs.inactive_timeout=PT1M;evs.install_timeout=PT1M;evs.join_retrans_period=PT10.5S"
wsrep_cluster_name="zabbix_data_cluster"
wsrep_cluster_address="gcomm://
172.16.187.1,172.16.187.2,172.16.187.3"
wsrep_node_address=172.16.187.3
wsrep_sst_method=xtrabackup-v2
wsrep_sst_auth=xtrabackup:**********
wsrep_notify_cmd=/usr/bin/galeranotify.py
wsrep_debug=OFF
wsrep_retry_autocommit=4
[sst]
encrypt = 3
tca = /etc/my.cnf.d/ca-cert.pem
tkey = /etc/my.cnf.d/server-key.pem
tcert = /etc/my.cnf.d/server-cert.pem
streamfmt = xbstream
transferfmt = socat
[mysqld_safe]
log-error=/var/log/mysql/mysqld.log
open-files-limit = 65535
malloc_lib=/usr/lib64/libtcmalloc_minimal.so.4
[mysqldump]
quick
[mysqlhotcopy]
interactive-timeout = 600