According to the attack vectors, you cannot specify which product is failing or which product is working well because there's no agent. The best case scenario is to add an agent solution where an agent would have the ability to actually detect which programs aren't working. For the attack software, you put a peer on the cloud site, and you have another peer internal network. There is IPS, firewall, WAF, and DBS amongst these peers. The cloud's peer is trying to send the attack file to the internal network. Maybe the firewall is blocking it, maybe the IP, maybe the WAF, but you cannot see the details. You can say, "Yes, my security product is blocking that attack scenario," or, "I cannot block this attack."