[MDEV-20362] Failed to add the second node into MariaDB galera-cluster, error info: [ERROR] WSREP: Application received wrong state: Created: 2019-08-16  Updated: 2020-12-08  Resolved: 2019-08-27

Status: Closed
Project: MariaDB Server
Component/s: Galera
Affects Version/s: 10.3.7
Fix Version/s: 10.3.17

Type: Bug Priority: Major
Reporter: willhan Assignee: Jan Lindström (Inactive)
Resolution: Fixed Votes: 0
Labels: galera
Environment:

centos6 centos7
1. libgalera_smm.so, compile from https://github.com/MariaDB/galera.git from 3.x
2. libgalera_smm.so, galear 3, download rpm from https://galeracluster.com/


Attachments: PNG File 1.png     PNG File 2.png     File node1.err     File node2.err    

 Description   

I get strange error when I new a MariaDB-Galera cluster. My environment is Mariadb-10.3.14 and galera3, I build 3 MariaDB instance on one machine.
Next is the node1 and node2 config (mainly about wsrep)

wsrep_cluster_address="gcomm://9.128.x.x:23021,9.128.x.x:23022,9.128.x.x:23023"
#wsrep_cluster_address="gcomm://"
wsrep_node_address=9.128.x.x:23021
wsrep_sst_receive_address=9.128.x.x:9021
innodb_doublewrite=1
query_cache_size=0
query_cache_type=0
wsrep_node_name= node1
wsrep_sst_auth="willtest:will123"
binlog_format=row
default_storage_engine=InnoDB
innodb_autoinc_lock_mode=2
wsrep_slave_threads=1
innodb_flush_log_at_trx_commit=0
wsrep_sst_method=rsync
#wsrep_sst_method=xtrabackup
#wsrep_causal_reads=ON
wsrep_on=ON

*This is my operation step: *

for node1: cd /usr/local/mysql/ && ./scripts/mysql_install_db --defaults-file=/etc/my.cnf.25000 --user=mysql
for node2: cd /usr/local/mysql/ && ./scripts/mysql_install_db --defaults-file=/etc/my.cnf.25000 --user=mysql
for node1: ./bin/mysqld_safe --defaults-file=/etc/my.cnf.25000  --wsrep-new-cluster  --user=mysql &  
for node2: ./bin/mysqld_safe --defaults-file=/etc/my.cnf.25001 --user=mysql &  

Node1's status is all right, see annex 1.png
But when I add node2 and some errors happend, see annex 2.png
The annex node1.err/node2.err is the error log from instance 25000/25001

The grastate.dat of node2:

# GALERA saved state
version: 2.1
uuid:    00000000-0000-0000-0000-000000000000
seqno:   0
safe_to_bootstrap: 0



 Comments   
Comment by Geoff Montee (Inactive) [ 2019-08-17 ]

I wonder if this could be another symptom of MDEV-17458. MDEV-17458 has been fixed in MariaDB 10.3.16 and later. You are currently using MariaDB 10.3.7, which is very old. I would suggest upgrading to the latest MariaDB 10.3 release to determine if upgrading fixes the issue.

Comment by willhan [ 2019-08-20 ]

I will try

Comment by willhan [ 2019-08-21 ]

mariadb-10.3.17 is OK

Generated at Thu Feb 08 08:58:53 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.