From the Scanning and Rating Components lesson, the Study Guide explicitly states:
"FortiSandbox allows you to modify the number of CPUs and memory assigned to a custom VM. This feature is supported on hardware models and private cloud VMs." Hardware models = Device-based (Option C) Private cloud VMs = Private cloud (Option A) Azure non-nested mode and FortiSandbox Cloud do not support custom VM creation as per the Study Guide.

The best answer is B. The Study Guide explains that under VM settings, "FortiSandbox has a Browser selection that allows you to choose which internet browser the VM instance will use. This helps to customize the test using an internet browser that more closely resembles the user's environment or just monitor if the test delivers different results." It also states that the default browser choices are Internet Explorer, Firefox, Chrome, and Edge. In addition, the guide says that "The VM images provided by Fortinet might not suit your needs... You can generate a custom VM that fits your organization's needs and upload it to FortiSandbox." Because only endpoints using Opera are affected, the clean verdict likely occurred because the sandbox environment does not accurately reproduce the exploited browser environment. The most effective fix is to make the sandbox environment match the real target environment more closely by using a custom VM with the same browser behavior as the affected endpoints. The other answers do not address the root cause. STIX/TAXII is unrelated, changing the scan profile file type does not solve a browser-specific exploit path, and job queue priority affects order, not analysis fidelity. Therefore, the required action is to configure a custom VM to use the same browser as the exploited end stations.

From the High Availability and Management lesson, the Study Guide states:
"You must configure the HA group name, password, and the cluster virtual IP. The worker nodes provide load balancing. The primary node distributes scan jobs to the worker nodes."
"You must configure the HA group name, password, and the virtual IP only on the primary node... Devices will interact with the cluster using this virtual IP." From the exhibit topology:
Cluster Virtual IP address = 10.25.1.50
Primary Node port1 = 10.25.1.30
Secondary Node port1 = 10.25.1.40
Worker Node port1 = 10.75.1.10
The worker node must be configured to point to the Cluster Virtual IP (10.25.1.50), not the individual primary node IP. This is because worker nodes join the cluster by connecting to the cluster virtual IP address.
Therefore the correct command is: hc-worker -a -sI0.25.1.50 -p<password>

From the Attack Methodologies lesson, the Study Guide explicitly states:
"During the lateral movement stage, the attacker is trying to compromise and infect other computers in the network. If these computers are protected with FortiClient, FortiClient can send any file that the computer downloads, to FortiSandbox for analysis."
"FortiDeceptor creates a network of decoys, to lure attackers and monitor their activities on the network. When attackers attack a decoy, an alert is generated. FortiDeceptor engages FortiSandBox to get a verdict on the suspected malware."
"If you deploy FortiGate as an ISFW firewall, FortiGate can analyze the traffic moving across subnets and send any files to FortiSandbox for analysis to prevent propagation." Both FortiDeceptor (Option B) and FortiGate (Option D) are specifically identified as protecting against the lateral movement stage through their FortiSandbox integration.