Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Industroyer and Triton are the two malware families that proved purpose-built industrial attack tooling exists: one spoke grid protocols to cut electricity in Kyiv, the other reached into a plant’s safety system.

What Are Industroyer and Triton?
Only a handful of malware families have been written to manipulate industrial equipment directly rather than simply to disrupt the computers around it. Two of them appeared within a year of each other, and between them they changed how the sector thinks about the problem. Industroyer attacked electricity transmission by speaking the grid’s own protocols. Triton attacked a petrochemical plant’s safety instrumented system, the last line of defence between a process and a catastrophe.
On 17 December 2016, a transmission substation north of Kyiv opened its breakers and roughly a fifth of the city lost power for about an hour. Six months later, researchers at ESET published the analysis of the malware responsible, naming it Industroyer; a parallel analysis from Dragos called it CrashOverride.
What made it notable was its architecture. Rather than scripting an operator workstation, it carried four payload modules implementing industrial protocols natively: IEC 60870-5-101 for serial links, IEC 60870-5-104 for its network equivalent, IEC 61850 for substation devices, and OPC Data Access. Each module could enumerate the devices it found and issue control commands in the correct syntax, meaning the malware understood the grid rather than merely running on a computer connected to it. A data wiper destroyed system state afterwards to slow recovery, and a denial of service module exploited a known vulnerability in a family of protection relays to make them unresponsive.
Industroyer was attributed to the Sandworm group, associated with Russian military intelligence. On 8 April 2022, Ukraine’s computer emergency response team and ESET disclosed Industroyer2, a stripped-down successor with IEC 60870-5-104 configuration hardcoded rather than read from the environment, aimed at a Ukrainian energy provider and scheduled to execute at a specific time. It was found and blocked before it operated. In 2024, a further Sandworm operation against a Ukrainian heating utility used different tooling again, confirming that the capability is maintained rather than historical.
In August 2017, a petrochemical plant in Saudi Arabia shut down unexpectedly. Twice. The engineering investigation found that the plant’s safety instrumented system, a Schneider Electric Triconex controller, had tripped, and tracing why led to malware that was published in December 2017 as Triton, also called Trisis and HatMan.
The attack chain is instructive. The intruders had been inside the plant’s networks for a long period, moved from the business network to the process control network, and compromised an engineering workstation with access to the safety system. From there they used the proprietary TriStation protocol to communicate with the safety controller, exploited a previously unknown flaw to escalate privilege, and wrote an implant into the controller’s firmware memory that would let them change the safety logic at will.
The plant shut down because the attackers’ code failed a validation check inside the controller, which did exactly what a safety system should do and tripped the process. The incident was discovered because the attack was imperfect, which is an uncomfortable thing to build a defence on. The US Treasury attributed the operation in October 2020 to a Russian government research institute and imposed sanctions; the same actor was subsequently found at another critical infrastructure facility, and the FBI has warned of continued interest in safety systems.
For UK operators, these cases sit behind the expectations of the NCSC Cyber Assessment Framework and the NIS Regulations 2018, which ask for demonstrable detection and response capability in the systems supporting an essential function rather than in the corporate estate alone. The engineering counter-measures belong in the zone and conduit model of IEC 62443, and the broader context is covered under critical infrastructure security.
At Privia Security, we test control and safety system boundaries through our OT-SCADA Penetration Test.
No. The December 2015 attack on regional distribution companies used BlackEnergy malware and hands-on-keyboard operation of the utilities’ own control systems, cutting supply to roughly 225,000 customers. Industroyer was used in the separate December 2016 attack on a transmission substation near Kyiv, and differed by implementing grid protocols natively rather than driving operator workstations.
Because it targeted a safety instrumented system, the independent layer that takes a process to a safe state when something goes wrong. Manipulating safety logic while the process runs normally creates the possibility of physical harm rather than only disruption. It was the first publicly documented malware designed to interfere with a safety system in a live plant.
By accident, in effect. The attackers’ code failed a validation check inside the Triconex controller, which responded by tripping the process and shutting the plant down. The engineering investigation into the unexplained trips led to the implant. Had the code been written correctly, it could have remained in place indefinitely, which is the uncomfortable part of the case.
Protocol-aware passive monitoring on the control network. Both operations produced traffic that no legitimate process would generate: enumeration and control commands in industrial protocols from unexpected sources, and engineering-mode communication with a safety controller. Endpoint tooling on Windows machines would not have seen it; a sensor parsing the industrial protocols would have.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.