# Cannot start second server in cluster

**URL:** <https://forums.percona.com/t/cannot-start-second-server-in-cluster/3822>\
**Category:** Percona XtraDB Cluster 5.x\
**Created:** [October 15, 2014, 10:50am UTC](https://forums.percona.com/t/cannot-start-second-server-in-cluster/3822 "2014-10-15T10:50:03Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![llowry](https://avatars.discourse-cdn.com/v4/letter/l/b19c9b/32.png) [@llowry](https://forums.percona.com/u/llowry)\
**Post date:** [October 15, 2014, 10:50am UTC](https://forums.percona.com/t/cannot-start-second-server-in-cluster/3822/1 "2014-10-15T10:50:03Z")

</div>

Using Ubuntu server 14.04.1 and Percona XtraDB Cluster 5.6.xxx.  
I have the mysql service running on each server.  
I moved the datadir to a second SSD Drive (more space and faster)  
Checked that the servie works, Ok.

The first node bootstraps fine. But the second and third servers fail to start.  
It appears to be the WSREP failing to get the initial data transfer from the first server.

Go to first server and run (from the manual)  
innobackupex --user=username --password=password /tmp/

This fails with the following:  
09:03:24 innobackupex: Connecting to MySQL server with DSN ‘dbi:mysql:;mysql\_read\_default\_group=xtrabackup’ as ‘root’ (using password: YES).  
innobackupex: Error: Failed to connect to MySQL server: DBI connect(‘;mysql\_read\_default\_group=xtrabackup’,‘username’,…) failed: Access denied for user ‘root’@‘localhost’ (using password: YES) at /usr/bin/innobackupex line 2949.

I have a feeling it is because I moved the datadir, but it is correct in the my.cnf.  
Anyone with an idea?

Thanks.

Larry

[mysqld]  
datadir=/data/mysql/mysql/  
user=mysql

log\_error = /data/mysql

#Skip reverse DNS lookup of clients  
skip-host-cache  
skip-name-resolve

wsrep\_provider=/usr/lib/libgalera\_smm.so  
wsrep\_cluster\_address=gcomm://  
wsrep\_cluster\_address=gcomm://192.168.90.140,192.168.90.141,192.168.90.142

max\_connections=1000  
max\_connect\_errors=100  
max\_allowed\_packet= 32M  
max-heap-table-size = 32M  
tmp-table-size = 32M  
thread\_cache\_size = 2000  
open-files-limit = 65535

log\_bin\_use\_v1\_row\_events=1  
gtid\_mode=0  
binlog\_checksum=NONE  
binlog\_format=ROW  
default\_storage\_engine=InnoDB  
innodb\_locks\_unsafe\_for\_binlog=1  
innodb\_autoinc\_lock\_mode=2  
innodb\_buffer\_pool\_size=5G  
innodb\_log\_file\_size=256M  
#innodb\_log\_buffer\_size=50M

wsrep\_node\_address=192.168.90.140  
wsrep\_sst\_method=xtrabackup  
wsrep\_cluster\_name=prodrr  
wsrep\_sst\_auth=“username:password”

slow\_query\_log = ‘ON’  
long\_query\_time = 2  
log-queries-not-using-index = YES

---

<div class="post-metadata">

**Author:** ![tkauppin](https://avatars.discourse-cdn.com/v4/letter/t/a9adbd/32.png) [@tkauppin](https://forums.percona.com/u/tkauppin)\
**Post date:** [October 16, 2014, 2:34am UTC](https://forums.percona.com/t/cannot-start-second-server-in-cluster/3822/2 "2014-10-16T02:34:49Z")

</div>

‘Access denied’ is likely because incorrect password. Maybe the easiest way is to run password update to the user ‘root’@‘localhost’ and retry.

---

<div class="post-metadata">

**Author:** ![llowry](https://avatars.discourse-cdn.com/v4/letter/l/b19c9b/32.png) [@llowry](https://forums.percona.com/u/llowry)\
**Post date:** [October 17, 2014, 3:14pm UTC](https://forums.percona.com/t/cannot-start-second-server-in-cluster/3822/3 "2014-10-17T15:14:29Z")

</div>

Finally figured it out. The username for the replication evidently has to be username@localhost.

We had it set for username@%.

Larry