Hi folks,

We are facing issues in setting up three node mysql cluster for high availability.
We have set management server on one machine and the same machine has two 
nodes running locally on them. The connection seems to be working fine after following
the documentations in the following link . Everything works like a charm.

But when we attach a third node which is another machine with a different IP address. 
We get into lot of issues. The commands that we ran to start management server, 
ndb_mgm, ndbd and then mysqld can be seen below :

status of localhost can be seen below :

[root@test bin]# ./ndb_mgm -e show
Connected to Management Server at: localhost:1186
Cluster Configuration
---------------------
[ndbd(NDB)] 2 node(s)
id=3 @127.0.0.1  (mysql-5.5.29 ndb-7.2.10, Nodegroup: 0, Master)
id=4 @127.0.0.1  (mysql-5.5.29 ndb-7.2.10, Nodegroup: 0)

[ndb_mgmd(MGM)] 1 node(s)
id=1 @127.0.0.1  (mysql-5.5.29 ndb-7.2.10)

[mysqld(API)] 1 node(s)
id=50 @192.168.105.40  (mysql-5.5.29 ndb-7.2.10)

Status of localhost after connecting third node can be seen below :

[root@test bin]# ./ndb_mgm -e show
Connected to Management Server at: localhost:1186
Cluster Configuration
---------------------
[ndbd(NDB)] 3 node(s)
id=3 (not connected, accepting connect from 192.168.105.40)
id=4 (not connected, accepting connect from 192.168.105.40)
id=11 (not connected, accepting connect from 192.168.105.63)

[ndb_mgmd(MGM)] 1 node(s)
id=1 @192.168.105.40  (mysql-5.5.29 ndb-7.2.10)

[mysqld(API)] 3 node(s)
id=50 (not connected, accepting connect from any host)
id=51 (not connected, accepting connect from 192.168.105.63)
id=52 (not connected, accepting connect from any host)

Error while starting mysqld service on management server :

bash-4.1$ /devops/mysqlc/bin/mysqld --defaults-file=/devops/my_cluster/conf/my.cnf &
[1] 22598
bash-4.1$ 130227 17:27:47 [Note] Plugin 'FEDERATED' is disabled.
130227 17:27:47 InnoDB: The InnoDB memory heap is disabled
130227 17:27:47 InnoDB: Mutexes and rw_locks use GCC atomic builtins
130227 17:27:47 InnoDB: Compressed tables use zlib 1.2.3
130227 17:27:47 InnoDB: Using Linux native AIO
130227 17:27:47 InnoDB: Initializing buffer pool, size = 128.0M
130227 17:27:47 InnoDB: Completed initialization of buffer pool
130227 17:27:47 InnoDB: highest supported file format is Barracuda.
130227 17:27:47  InnoDB: Waiting for the background threads to start
130227 17:27:48 InnoDB: 1.1.8 started; log sequence number 1595675
130227 17:28:23 [Warning] NDB: server id set to zero - changes logged to bin log with server id zero will be logged with another server id by slave mysqlds
130227 17:28:23 [Note] Starting Cluster Binlog Thread
130227 17:28:23 [Note] Server hostname (bind-address): '0.0.0.0'; port: 5000
130227 17:28:23 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
130227 17:28:23 [Note] Server socket created on IP: '0.0.0.0'.
130227 17:28:23 [Note] Event Scheduler: Loaded 0 events
130227 17:28:23 [Note] /devops/mysqlc/bin/mysqld: ready for connections.
Version: '5.5.29-ndb-7.2.10-cluster-gpl'  socket: '/tmp/mysql.sock'  port: 5000  MySQL Cluster Community Server (GPL)
130227 17:28:53 [Warning] NDB : Tables not available after 30 seconds.  Consider increasing --ndb-wait-setup value

I am also attaching config.ini and my.cnf files of management server and data node respectively.

config_management_server.ini : management server config.ini file
my_cnf_data_node.cnf : my.cnf on third data node

Please let us know where are we going wrong.

Thank you,

-- 
Ashutosh Narayan

http://ashutoshn.wordpress.com/