When Ransomware Knocked on the Factory Floor — And Nobody Answered

    A European automotive parts manufacturer operating 4 production plants with over 200 PLCs and HMIs on the shop floor

    48 views

    Challenge

    After a peer company suffered a ransomware attack that halted production for 11 days, leadership fast-tracked OT security. Legacy Siemens S7 PLCs and Rockwell controllers were running firmware from 2014 with no patching path.

    Solution

    FlintX deployed non-intrusively via SPAN ports — protocol-aware monitoring for S7comm and EtherNet/IP, automated behavioral baselining, and integration with the plant's existing IT SIEM.

    Results

    3 ransomware propagation attempts blocked at the IT/OT boundary within the first quarter, firmware vulnerability inventory completed across all 4 plants, 99.97% production uptime maintained.

    The phone call came on a Tuesday morning. Not to this manufacturer — to their competitor. A ransomware attack had encrypted servers across the competitor's production network. Assembly lines stopped. Robotic welding cells went dark. The ERP system was unreachable. It took 11 days to restore operations. The estimated cost, including lost production, incident response, and customer penalties for missed deliveries, exceeded €15 million.

    By Wednesday, this manufacturer's board had scheduled an emergency meeting. By Friday, OT security had moved from "next year's budget" to "do it now."

    The Reality of a Legacy Shop Floor

    The manufacturer operated four production plants across Central Europe, producing precision components for major automotive OEMs. The shop floors ran on a mix of Siemens S7-300 and S7-400 PLCs, Rockwell Allen-Bradley ControlLogix systems, and a variety of HMIs from different vendors and different decades.

    Some of these PLCs had been in continuous operation since 2010. Their firmware had not been updated because the vendor no longer supported the specific version, or because the production process they controlled had been validated against that firmware version, and any change would require re-validation — a process that could take weeks and cost hundreds of thousands of euros.

    In other words, patching was not an option. These devices would run their current firmware until they were physically replaced. And replacement was years away.

    The IT security team had done what they could. Firewalls separated the corporate network from the production network. Antivirus ran on the Windows-based HMI stations (those that could support it — several were still running Windows 7). But there was no monitoring of the OT protocols themselves. Nobody could see the S7comm traffic between the engineering stations and the PLCs. Nobody knew what "normal" PLC communication looked like, which meant nobody would notice if it changed.

    Starting with What You Cannot Patch

    FlintX was deployed across all four plants over a three-week period. The deployment was entirely passive — SPAN ports on the production network switches feeding traffic to FlintX sensors. No inline devices. No agents on PLCs (which would not have been possible regardless). No changes to the production control systems.

    The first priority was not threat detection — it was asset inventory. The manufacturer needed to know exactly what was on each plant's production network, what firmware each device was running, and what known vulnerabilities existed.

    The results were sobering. Across the four plants, FlintX identified:

    • 217 PLCs (the plant maintenance system listed 189)
    • 43 HMI stations, 12 of which were running Windows 7 with no endpoint protection
    • 28 unmanaged Ethernet switches on the shop floor
    • 9 engineering workstations with connections to both the production network and the corporate LAN
    • 4 devices running FTP servers with anonymous access enabled

    The firmware inventory was equally revealing. 61% of the Siemens PLCs were running firmware versions with known vulnerabilities — vulnerabilities that had been publicly disclosed but could not be patched due to the operational constraints described above.

    If You Cannot Patch, You Monitor

    This is the fundamental reality of OT security in manufacturing: you often cannot fix the vulnerability. The PLC with the known exploit in its firmware is also the PLC controlling a €3 million stamping press that runs 24/7. You cannot take it offline, you cannot update its firmware, and you cannot put an antivirus agent on it.

    What you can do is watch it. Know who is talking to it. Know what commands it is receiving. Know the instant something deviates from the established baseline.

    FlintX built behavioral profiles for every PLC on the network. It learned the S7comm read/write patterns from the SCADA server. It learned the EtherNet/IP implicit messaging intervals from the Rockwell controllers. It learned which engineering workstations connected to which PLCs, and when.

    Any deviation from those baselines generated an alert — contextualized with the asset's criticality, the type of protocol anomaly, and recommended response actions.

    Three Attempts. Three Blocks.

    Within the first quarter of operation, FlintX detected three separate incidents where ransomware attempted to propagate from the corporate network into the production environment.

    The first was a variant of a known ransomware family that had entered the corporate network through a phishing email. It spread laterally through SMB shares and reached a dual-homed engineering workstation — one of the nine that FlintX had identified during the initial asset inventory. FlintX detected the anomalous SMB traffic on the production side of the workstation and alerted within seconds. The workstation was isolated before the ransomware could reach any production systems.

    The second incident involved a compromised vendor laptop connected to the plant network for a scheduled maintenance visit. The laptop was infected with a worm that began scanning for open ports on the production network. FlintX detected the port scan immediately — port scans are not normal behavior on a network where PLCs communicate in predictable patterns — and the laptop was disconnected.

    The third was the most concerning. An attacker had gained access to the corporate VPN using stolen credentials and was methodically exploring the network. When they reached a segment that carried S7comm traffic, FlintX detected the reconnaissance activity: connection attempts to PLC ports from an IP address that had never communicated with those devices before. The SOC team was alerted, the VPN session was terminated, and the credentials were revoked.

    In all three cases, the gap between detection and response was measured in minutes. In all three cases, production was unaffected.

    What the Numbers Do Not Show

    The headline metric — 99.97% production uptime — is impressive, but it obscures the more important story. The manufacturer did not achieve that uptime by preventing every possible threat. They achieved it by seeing threats early enough to respond before they reached anything that mattered.

    The nine dual-homed workstations were reconfigured. The four FTP servers were shut down. The 12 unprotected HMI stations were placed behind network micro-segmentation rules. The firmware vulnerability inventory became a living document that informed the long-term PLC replacement roadmap.

    None of these changes required the production lines to stop. None required the PLCs to be updated. None required the OT engineers to change how they operated their equipment.

    The factory floor kept running. The presses kept stamping. The robots kept welding. The only difference was that now, someone was watching the network between them — and would notice the moment something tried to come in uninvited.

    "Our competitor lost 11 days. We have not lost 11 minutes. The difference is not luck — it is visibility." — Plant Operations Director

    Share: