two mgmd node add two data node failed (no replies)

hi, all
I have 2 mgm+ 2 data node + 4 sql node cluster running in production ,
it worked fine,

and I want to add 2 data node ,and the create new group,
I followed the http://dev.mysql.com/doc/mysql-cluster-excerpt/5.5/en/mysql-cluster-online-add-node-example.html
and the example is with 0ne mgm node

here is my step:
=============================
1 , stop master mgm node
2 , change the config.ini
added :
[ndbd]
hostname=10.1.1.5
datadir=/opt/mysqlcluster/ndbcluster2
NodeId=13
LockExecuteThreadToCPU=1
LockMaintThreadsToCPU=0

[ndbd]
hostname=10.1.1.7
datadir=/opt/mysqlcluster/ndbcluster2
NodeId=14
LockExecuteThreadToCPU=1
LockMaintThreadsToCPU=0

3, stop the second mgm node
4, start the master mgm node with --reload
5, start the second mgm node with --reload
6, login to any mgm node :
and execute show command:
only see the same:
ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)] 4 node(s)
id=10 (not connected, accepting connect from 10.1.1.155)
id=12 (not connected, accepting connect from 10.1.1.154)
id=23 (not connected, accepting connect from 10.1.1.154)
id=24 (not connected, accepting connect from 10.1.1.155)

[ndb_mgmd(MGM)] 2 node(s)
id=1 @10.1.1.155 (mysql-5.5.22 ndb-7.2.6)
id=2 @10.1.1.154 (mysql-5.5.22 ndb-7.2.6)

[mysqld(API)] 4 node(s)
id=25 (not connected, accepting connect from 10.1.1.154)
id=26 (not connected, accepting connect from 10.1.1.155)
id=27 (not connected, accepting connect from any host)
id=28 (not connected, accepting connect from any host)

the cluster is still running, and can do CUID opreations

HELP

or the step is wrong? Any help will be gratefully appreciated