Re: Only one ndb node will run at a time. Starting two causes them both to shutdown

[Mikael Ronström]

Steve Ferguson wrote:

Hi,
Don't see any errors but the high number on the node type indicates that there
might be computers with different endians in the cluster. This won't work unfortunately.

Rgrds Mikael

> Thanks in advance for any help you can provide.
>
> I am trying to setup a relatively simple cluster
> with mysql. I have 3 computers.
>
> 1.1.1.1 is the first storage server. it will also
> be a sql server
> 1.1.1.2 is the second storage server. it will also
> be a sql server
>
> 1.1.1.3 is the management server.
>
> If i setup my configuration for the management
> server to include only one of these storage
> node/mysql combinations, the one node cluster
> works correctly. However, the moment I try to
> introduce the redundant node into the
> configuration, upon starting the second ndbd
> process, both ndb processes shutdown! Here is my
> config.ini
>
>
> NoOfReplicas=2
> DataMemory=80M
> IndexMemory=52M
>
>
> hostname=1.1.1.3
> datadir=/var/lib/mysql-cluster
>
>
> hostname=1.1.1.1
> datadir=/var/lib/mysql/teof
>
>
> hostname=1.1.1.2
> datadir=/var/lib/mysql/teof
>
>
>
>
> Here is my my.cnf from both the .1 and .2
> machines
>
>
> ndbcluster
> ndb-connectstring=1.1.1.3
>
>
> ndb-connectstring=1.1.1.3
>
>
>
> upon starting the second ndbd on 1.1.1.2 I recieve
> the following error message :
>
> 2005-05-02 11:31:45 INFO -- Configuration
> fetched at 1.1.1.3 port 1186
> Error handler shutting down system (unknown type:
> 1868919855)
> Error handler shutdown completed - exiting
>
>
> the following is the output in the management
> server log :
>
>
>
> 2005-05-02 11:28:06 INFO -- Mgmt server
> state: nodeid 2 reserved for ip 1.1.1.1,
> m_reserved_nodes 0000000000000006.
> 2005-05-02 11:28:06 INFO -- Node 1: Node 2
> Connected
> 2005-05-02 11:28:09 INFO -- Mgmt server
> state: nodeid 3 reserved for ip 1.1.1.2,
> m_reserved_nodes 000000000000000e.
> 2005-05-02 11:28:09 INFO -- Node 1: Node 3
> Connected
> 2005-05-02 11:28:09 INFO -- Node 2: Node 3
> Connected
> 2005-05-02 11:28:09 INFO -- Node 3: Node 2
> Connected
> 2005-05-02 11:28:09 INFO -- Node 2: Start
> phase 1 completed
> 2005-05-02 11:28:10 ALERT -- Node 1: Node 2
> Disconnected
> 2005-05-02 11:28:11 INFO -- Mgmt server
> state: nodeid 2 freed, m_reserved_nodes
> 000000000000000a
> .
> 2005-05-02 11:28:11 ALERT -- Node 1: Node 3
> Disconnected
> 2005-05-02 11:28:12 INFO -- Mgmt server
> state: nodeid 3 freed, m_reserved_nodes
> 0000000000000002
>
>
>
> These are running 4.1.11 on 3 Suse Linux
> Enterprise 9 vmware virtual machines.
>
>
> Thanks so much for any help with this.