Data Source Onboarding and Configuration
Onboarding sequence
Onboarding is a chain: define the data source (type, version, connection), install the S-TAP or configure the inspection engine, associate the S-TAP with a collector and the data source, then verify status and traffic. GIM deploys and manages S-TAP; manual installation is the fallback when GIM cannot reach the host.
Key parameters
- ktap/tap and database home or instance parameters (for example oracle_home, db2_instance) must match the host.
- The S-TAP must reach the collector on the required ports; firewalls are a common failure.
- Data source groups scope policies and reports; put related instances in one group.
- A sniffer/SPAN feed can capture traffic without an agent but lacks local context for some events.
Worked example
An Oracle RAC database has two nodes. Create one data source for the cluster, install an S-TAP on each node, and point both to the same collector. Verify each S-TAP shows connected and that sessions from both nodes appear under the data source. Add the data source to the PCI group so the policy applies to every node, not just the first.
Traps
Registering an S-TAP but never assigning it to a data source; wrong ORACLE_HOME so the tap loads but does not parse; opening only the GIM port and not the S-TAP port; placing the S-TAP on a collector in another segment with no route.