node 2 cann't join cluster

wsrep_cluster_address has a bad value in your config. Use two slashes after 'gcomm' protocol selector in the URI:

wsrep_cluster_address="gcomm://10.224.150.206:4567"

Thanks. But after changing the wsrep.cnf,still have some issues.

[root@mysqlha02 conf.d]# more wsrep.cnf
[mysqld]
wsrep_cluster_address="gcomm://10.224.150.206:4567"
wsrep_provider=/usr/lib64/galera/libgalera_smm.so
wsrep_sst_auth=sst:pass

----------------------------------------------------------------------

131030 7:26:41 [Note] WSREP: Assign initial position for certification: -1, protocol version: -1
131030 7:26:41 [Note] WSREP: wsrep_sst_grab()
131030 7:26:41 [Note] WSREP: Start replication
131030 7:26:41 [Note] WSREP: Setting initial position to 00000000-0000-0000-0000-000000000000:-1
131030 7:26:41 [Note] WSREP: protonet asio version 0
131030 7:26:41 [Note] WSREP: backend: asio
131030 7:26:41 [Note] WSREP: GMCast version 0
131030 7:26:41 [Note] WSREP: (9f3b20be-4134-11e3-b4a6-3bcf5b2e9ace, 'tcp://0.0.0.0:4567') listening at tcp://0.0.0.0:4567
131030 7:26:41 [Note] WSREP: (9f3b20be-4134-11e3-b4a6-3bcf5b2e9ace, 'tcp://0.0.0.0:4567') multicast: , ttl: 1
131030 7:26:41 [Note] WSREP: EVS version 0
131030 7:26:41 [Note] WSREP: PC version 0
131030 7:26:41 [Note] WSREP: gcomm: connecting to group 'my_wsrep_cluster', peer '10.224.150.206:4567'
131030 7:26:41 [Warning] WSREP: (9f3b20be-4134-11e3-b4a6-3bcf5b2e9ace, 'tcp://0.0.0.0:4567') address 'tcp://10.224.150.206:4567' points to own listening address, blacklisting
131030 7:26:44 [Warning] WSREP: no nodes coming from prim view, prim not possible
131030 7:26:44 [Note] WSREP: view(view_id(NON_PRIM,9f3b20be-4134-11e3-b4a6-3bcf5b2e9ace,1) memb {
        9f3b20be-4134-11e3-b4a6-3bcf5b2e9ace,
} joined {
} left {
} partitioned {
})
131030 7:26:45 [Warning] WSREP: last inactive check more than PT1.5S ago (PT3.50137S), skipping check
131030 7:27:14 [Note] WSREP: view((empty))
131030 7:27:14 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out)
         at gcomm/src/pc.cpp:connect():141
131030 7:27:14 [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():196: Failed to open backend connection: -110 (Connection timed out)
131030 7:27:14 [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1290: Failed to open channel 'my_wsrep_cluster' at 'gcomm://10.224.150.206:4567': -110 (Connection timed out)
131030 7:27:14 [ERROR] WSREP: gcs connect failed: Connection timed out
131030 7:27:14 [ERROR] WSREP: wsrep::connect() failed: 7
131030 7:27:14 [ERROR] Aborting

131030 7:27:14 [Note] WSREP: Service disconnected.
131030 7:27:15 [Note] WSREP: Some threads may fail to exit.
131030 7:27:15 [Note] /usr/sbin/mysqld: Shutdown complete
-------------------------------------------------------------------------------------------------------------------

wsrep_cluster_address tells the node how to initialize a connection to the cluster. It should contain an IP address of one node in the cluster, or preferably a list of all nodes present in the cluster.

Now it looks like 10.224.150.206 is actually the IP address of node2. When you start node2 with this configuration, he tries to connect to himself, and bails out.

=> write in wsrep_cluster_address the IP address on node1 and you should be able to initialize the connection between these two nodes.

The next step in node joining process will be that node2 will ask for database state transfer (SST). If you configure:

wsrep_sst_method=rsync

..then node1 will try to send mysql datadirectory over rsync to node2. This is likely the most simple way to kick start a Galera Cluster for experimentation.

You can read more about Galera configuration e.g. here: http://www.codership.com/wiki/doku.php?id=mysql_galera_configuration

Thanks a lot,!it works now~~