Cyber security is one of the topics found in the news daily. There are a lot of different attacks that can happen, so the consequences they cause can also vary from negligible to devastating. How to make a secure system is a never-ending question that has to be addressed from multiple perspectives.
Most security mechanisms in use today try to detect a threat in real time and block its actions. The most common representatives of this group are firewalls and antiviruses. Another kind of security mechanism, which is getting more and more adopted and is very common in operational networks, is the intrusion detection system or IDS. They are not yet completely accurate, so normally they are not allowed to make any changes to live systems, but they are exceptionally good at detecting anomalies, which are then reported to the SOC as alerts.
Another method for making our system secure, which I emphasised in this thesis, are honeypots. Through using them we want to find out what the newest techniques of attacks are by simulating a system and luring hackers into thinking they found the real thing and attacking it. By logging every interaction we can later analyse them and hope to find unknown attacks and take action to prevent them in the future. While exploring this theme, I set up a honeypot that tracks protocols specific to industrial networks. My goal was not to identify new attack methods, but to paint a picture of how often attackers search for this kind of network, and to see what kind of commands they use to identify them.
The design and construction of operational networks require taking into consideration different things than those of conventional networks. Redundancy and long expected lifetime of the hardware are of extreme importance, as outages have much greater consequences. Usually such networks transport a very small amount of data, for which low latency is a priority. This is allowed by special protocols, most of which use predefined function codes. The traffic in operational networks is usually not encrypted, as that would increase transmission time. This represents only one of many vulnerabilities that are present in operational networks and solved in informational networks.
|