Regarding Grid Configuration File issues

moalik · Post by **moalik** » Wed Mar 12, 2014 5:11 am

Hi All,

We have created the Grid Configuration on Datastage Engine. I have configured the Master Configuration file as well as project-level configuration file.

[abc@XXXXX etc]# cat master_config.apt

Code: Select all

{
        // Conductor node entry
        node "CONDUCTOR"
        {
                fastname "ABC Server"
                pools "conductor"
                resource disk "/opt/IBM/InformationServer/Server/Datasets" {pools ""}
                resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {pools ""}
        }

        node "COMPUTE_default"
        {
                fastname "ABC Server"
                pools "default"
                resource disk "/opt/IBM/InformationServer/Server/Datasets" {pools ""}
                resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {pools ""}
        }

        // Compute node entry
        node "node01"
        {
                fastname "ABC Server"
                pools "node01"
                resource disk "/opt/IBM/InformationServer/Server/Datasets" {pools ""}
                resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {pools ""}
        }
        node "node02"
        {
                fastname "GHI Server"
                pools "node02"
                resource disk "/opt/IBM/InformationServer/Server/Datasets" {pools ""}
                resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {pools ""}

}

My Project-level Default Configuration file is :

Code: Select all

{
	node "COMPUTE_01"
	{
		fastname "ABC Server"
		pools "node01"
		resource disk "/opt/IBM/InformationServer/Server/Datasets" {}
		resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {}
	}
	node "COMPUTE_02"
	{
		fastname "GHI Server"
		pools "node02"
		resource disk "/opt/IBM/InformationServer/Server/Datasets" {}
		resource scratchdisk "/opt/IBM/InformationServer/Server/Scratch" {}
	}
}

When i am trying to run a parallel job,i am getting a compilation error with below details:

The grid resource node pools are no longer available for this project, and will be removed: """
Please go to the designer job properties grid tab to check the current resources for this job.

I did all possible configurations but still getting the above error. Could you please help me in resolving the issue.

Thanks,

chulett · Post by **chulett** » Wed Mar 12, 2014 7:25 am

You seem to be missing the closing } for 'node02'.

PaulVL · Post by **PaulVL** » Wed Mar 12, 2014 9:00 am

Also, you mean CLUSTER and not GRID I believe. Nothing in your APT file indicates that it is dynamically generated (GRID deployment). A Cluster is a static setup across multiple servers.

moalik · Post by **moalik** » Mon Mar 24, 2014 2:44 am

Hi All,

Thanks a lot for the information.

Actually, we have our linux servers in cluster. So we just disabled the Grid tab and added the second server node details in configuration details and worked fine.

Thanks,
Mohsin Khan

PaulVL · Post by **PaulVL** » Mon Mar 24, 2014 7:12 am

BTW: I recommend changing your resource and scratch disk locations.

Your scratch is mounted on an NFS or CFS mount, that will affect your job performance quite a bit. You will want something local to each compute node/server.

Your datasets are being saved to the same mount that your engine binaries are on. You run the risk of filling up that mount and crashing the entire datastage engine.