Tell me more ×
Database Administrators Stack Exchange is a question and answer site for database professionals who wish to improve their database skills and learn from others in the community. It's 100% free, no registration required.

After successfully setting up a test MySQL Cluster on Ubuntu I am now attempting it on Centos and have successfully started the management node, however I cannot get the data nodes to connect to the manager. My config files are set the same (except for the hostnames that were changed) as the ones I used in my working Ubuntu mysql cluster, so I dont feel the problem lies there. What is different though is how I installed the software, which was via RPM files. When I unpacked them they basically installed themselves to places unbeknownst to me and so I feel that is probably where the issue stems from. But then again why would the the software install itself to the wrong location? And why would the management node work and not the data nodes? (Also I have made sure iptables is disabled on each host.)

As I said the manager connects perfectly but the data nodes cannot connect to the manager. The following is the error I receive when trying to connect the data nodes to the management node:

Unable to connect with connect string: nodeid=0, 10.0.14.31:1186
Retrying every 5 seconds. Attempts left: 12 11 10 ... 2 1, failed.
[ndbd ERROR    -- Could not connect to management server, error: ''

One final note, I notice that in the error it says unable to connect with connect string: nodeid=0. In my config.ini file the manager node is specified with a nodeid of 1 (nodeid=1) not 0. If this is the problem how can I force the data node to recognize the manager as nodeid=1?

Edit: With the following I attempted to force the data node to connect to nodeid=1 and yet it still did not connect and resulted in the same error as shown above although with nodeid=1 instead of nodeid=0.

ndbd --initial --ndb-connectstring="nodeid=1;10.0.14.31:1186"

Because of this I am now thinking my initial assumption is wrong and that the problem is in fact in the configuration file. I may specified the wrong base directory for data nodes in my config.ini.

    [ndbd default]
    # Options affecting ndbd processes on all data nodes:
    NoOfReplicas=2    # Number of replicas

    [tcp default]
    # TCP/IP options:

    [ndb_mgmd]
    # Management process options:
    hostname=10.0.14.31         # Hostname or IP address of MGM node
    datadir=/var/lib/mysql-cluster  # Directory for MGM node log files
    NodeId=1

    [ndbd]
    # Options for data node-1:
                                    # (one [ndbd] section per data node)
    hostname=10.0.14.32            # Hostname or IP address
    datadir=/usr/local/mysql/data   # Directory for this data node's data files
    NodeId=2

    [ndbd]
    # Options for data node-2:
    hostname=10.0.14.33             # Hostname or IP address
    datadir=/usr/local/mysql/data   # Directory for this data node's data files
    NodeId=3

    #one [mysqld] per storage node
    [mysqld]
    [mysqld]

share|improve this question
have you checked the firewall settings etc.? – johannes Aug 28 '12 at 2:43
Hi, thanks for the reply. I have disabled iptables on each host. There is a firewall elsewhere on the network though I don't believe it is between any of these hosts. Out of curiosity, what settings on a firewall would interfere here given the scenario? – sswahn Aug 28 '12 at 13:31
Any setting that prevents binding to that port ... a simple test you can do is something like telnet 10.0.14.31 1186 from the machine running the data nodes and see whether that can connect on TCP level at least. – johannes Aug 28 '12 at 14:02
That was it. It turned out iptables was apparently 'disabled' incorrectly on the management node. Too many cooks in the kitchen here. Thanks for your help with this! – sswahn Aug 28 '12 at 14:30
Thanks for asking here and commenting with your response—I'm now closing the question because it isn't likely to help others as it turned out :) – Jack Douglas Oct 28 '12 at 9:02

closed as too localized by Jack Douglas Oct 28 '12 at 9:02

This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet. For help making this question more broadly applicable, see the FAQ.