In Strategy 10.x 2 nodes clustering environment, the best practice to configure Enterprise Manager to monitor one Intelligence Server node only?
What Enterprise manger service really monitored is the metadata, not Intelligence Server, thus monitoring one node is enough.
When configuring the cluster environment, each node pointing to the same metadata, and in the last step, each node pointing to the same Enterprise Manager warehouse. Each Intelligence server would log the statistics to the same Enterprise Manager warehouse, and each Intelligence server would write the changed objects to the same metadata. After data load, all things would write into Enterprise Manager warehouse. At the backend, Intelligence server is not distinguished. So, for the cluster environment, monitor one server is sufficient to monitor the whole metadata.
Where to get Progress code for Data Load properties?
If the data load is running, below command is used to check how much percentage the current data load have completed.
“LIST PROPERTIES FOR CURRENT DATA LOAD …”
For example, ‘Progress = 3.69’ means the data load is running and it have completed 3.69%.
If the data load already finished, below command is used to check last data load is successful or failed.
“LIST PROPERTIES FOR LAST DATA LOAD …”
A value of -2 for the property ‘Progress’ indicates a successful data load. A value of -1 indicates a failed data load.
For cluster environment, is it a mandatory to add both nodes when CREATE DATA LOAD?
The below one is DATA Load with ENVIRONMENT of node A then is it necessary to add node B as well?
There is no need to add both nodes to create data load, and there’s no impact to specify which node to do the data load. It is only necessary to tell Enterprise Manager Service which project to monitor.
The project must be loaded on the node, or the CREATE command would be failed. If only one node has loaded the project, the DSS Migration process would only run on this node. If both nodes have loaded the project, Load balancing is triggered to decide which node to run the data load, no matter what node is used in the create data load command.
