Hi,

For better understend and stategic use of RPT 10 dataset, Can you explain how work the new mecanism of read, reponsing questions below?

1. In a execution fired by RPT Workbench, is the dataset stored in a single point for reading all the agents involved in the test?

2. If so, where does RPT store the dataset for reading three agents for example? On the desktop that fires the test?

3. At each reading of a new dataset record, does the agent search for the data at this single point?

4. In the old datapool, is the mechanism different? Was the data sent to the agents?

5. When using RTAS, is the dataset read directly from the RTAS server?

Regards,

Thiago.

Hi, the new dataset implementation is different the old datapool implementation.

With datapools the datapool file was physically copied to each agent's deployment root directory

With datasets each agent makes a call back to the RPT workbench on the agent listening port + 1, so if the default agent port is 7080 then the call from the agent to get a new dataset value will be on port 7081

We've had some tickets where the firewall from agent machine to RPT workbench for port 7080 was open but not port 7081, so to use datasets we need two ports open, the agent port then agent port +1, otherwise the test will fail because it can't get the new dataset value.

Here's the link to the Knowledge Center on dataset overview

https://www.ibm.com/support/knowledgecenter/en/SSBLQQ_10.0.1/com.ibm.rational.test.lt.common.doc/topics/cdatapool.html

The open modes are slightly different for datasets vs datapools, this technote explains the various open modes for datasets:

https://www.ibm.com/support/pages/understanding-various-open-modes-datasets-rational-performance-tester