Same node group when there are 2 datanodes and RF is 1 in mysql cluster

Question

I followed this tutorial and it works fine. This is the what I get when I showed in ndb_mgm client.

ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)] 2 node(s)
id=2    @1.1.1.10  (mysql-5.6.11 ndb-7.3.2, Nodegroup: 0, Master)
id=3    @1.1.1.8  (mysql-5.6.11 ndb-7.3.2, Nodegroup: 0)

[ndb_mgmd(MGM)] 1 node(s)
id=1    @1.1.1.5  (mysql-5.6.11 ndb-7.3.2)

[mysqld(API)]   2 node(s)
id=4    @1.1.1.9  (mysql-5.6.11 ndb-7.3.2)
id=5    @1.1.1.11  (mysql-5.6.11 ndb-7.3.2)

That's fine.

Then I changed NoOfReplicas=1 in config.ini of managemant node, and restarted all nodes.

Now I expect 2 nodegroups according to this.

[number_of_node_groups] = number_of_data_nodes / NoOfReplicas

But I still get the same output for show command in ndb_mgm client.

Am I doing anything wrong? Please advice.

Thanks in advance.

Answer 1

I'm answering my own question since I found what has been gone wrong. When we change config.ini in management node we have to do 2 things.

1) Restart Management node with --reload option

2) Restart data nodes with --initial option

Now my SHOW output is as follows.

ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)] 2 node(s)
id=2    @1.1.1.10  (mysql-5.6.11 ndb-7.3.2, Nodegroup: 0, Master)
id=3    @1.1.1.8  (mysql-5.6.11 ndb-7.3.2, Nodegroup: 1)

[ndb_mgmd(MGM)] 1 node(s)
id=1    @1.1.1.5  (mysql-5.6.11 ndb-7.3.2)

[mysqld(API)]   2 node(s)
id=4    @1.1.1.9  (mysql-5.6.11 ndb-7.3.2)
id=5    @1.1.1.11  (mysql-5.6.11 ndb-7.3.2) 

Answer 2

Looks a lot like mysql cluster bug#30831 - in that datanodes are still 'starting' and nodegroup(s) > 0 are reported as 0: http://bugs.mysql.com/bug.php?id=30831

You don't say, but I'm assuming the correct nodegroups are reported eventually.