An administrator running the Instana agent as a Docker container reports that no host processes appear in the Instana UI even though the container is running and all other infrastructure metrics are visible. Which Docker run flag must be added to enable host process visibility?
Select an answer to reveal the explanation.
Short Explanation and Infographic
The --pid=host flag shares the host machine PID namespace with the Docker container, which allows the Instana agent to enumerate and monitor all processes running on the host system. Without this flag, the agent is confined to its own container PID namespace and cannot report on host-level processes to the Instana backend. While --privileged grants broad privileges, the targeted solution for process visibility specifically is PID namespace sharing via --pid=host.
Full explanation below image
Full Explanation
The --pid=host flag shares the host machine PID namespace with the Docker container, which allows the Instana agent to enumerate and monitor all processes running on the host system. Without this flag, the agent is confined to its own container PID namespace and cannot report on host-level processes to the Instana backend. While --privileged grants broad privileges, the targeted solution for process visibility specifically is PID namespace sharing via --pid=host. The correct answer is '--pid=host'. The incorrect options — '--network=host', '--privileged', '--cap-add=SYS_PTRACE' — are wrong because they do not align with IBM Instana's architecture or recommended practices for this scenario. Understanding this concept is essential for the Domain 3: Installation domain of the IBM Instana Observability certification.