OT Incident Response: What to Do During an ICS Cyber Incident
An industrial cyber incident is different from an ordinary IT security event.
In an office environment, disconnecting a compromised computer may be a straightforward response. In an industrial environment, disconnecting the wrong system can affect production, safety, monitoring or physical processes.
This is why OT incident response must consider both cybersecurity and the industrial process.
The objective is not simply to remove malware or isolate a compromised device. The objective is to understand what happened, limit further harm, preserve useful evidence and restore safe and reliable operations.
What Is OT Incident Response?
OT incident response is the structured process used to prepare for, detect, analyze, contain, investigate and recover from cybersecurity incidents affecting operational technology environments.
OT environments can include industrial control systems, SCADA, distributed control systems, PLCs, engineering workstations, industrial networks and other systems that monitor or control physical processes.
NIST describes OT as technology that interacts with the physical environment and emphasizes that OT security must account for performance, reliability and safety requirements.
That requirement should remain central throughout incident response.
Why IT Incident Response Cannot Simply Be Copied Into OT
Traditional incident response often assumes that systems can be isolated, powered down, reimaged or restored with relatively limited operational consequences.
That assumption may not hold in an industrial environment.
- A PLC may control an active process.
- An HMI may provide critical operator visibility.
- A historian may be required for operational analysis.
- An engineering workstation may contain important configuration information.
- A network device may support communications between critical systems.
- A safety related system may have separate operational requirements.
Before taking containment action, responders should understand the potential operational consequences.
The First Question Is Not “How Do We Disconnect It?”
During a suspected OT incident, the natural reaction may be to disconnect the affected device immediately.
Sometimes that is the correct action.
Sometimes it can create a larger operational problem.
The better question is:
“What is the safest action that limits the incident while preserving the industrial process?”
NIST’s OT guidance emphasizes that response decisions should consider the service being provided and the operational consequences of available response options.
Prepare Before an Incident Happens
The most effective incident response begins before the first alert.
An organization should know which systems are important, who owns them, how the network is structured and who has authority to make emergency decisions.
- OT asset inventory
- Network architecture diagrams
- Critical process information
- System ownership
- Vendor contacts
- Remote access inventory
- Backup information
- Configuration records
- PLC logic backups
- Engineering documentation
- Incident escalation contacts
- Emergency communication procedures
Without this information, responders may spend valuable time trying to understand the environment while the incident is developing.
Know What Normal Looks Like
Incident detection depends on understanding normal behavior.
Unexpected network traffic, new accounts, unusual remote connections or changes to engineering systems can be difficult to recognize when there is no baseline for comparison.
Useful baseline information can include:
- Normal network communication
- Expected remote access
- Normal engineering workstation activity
- Known administrative accounts
- Expected system services
- Normal controller communication
- Typical maintenance windows
Recognize Possible OT Incident Indicators
An OT incident may not begin with a clear ransomware message or obvious system failure.
Possible indicators can include:
- Unexpected network traffic
- Unknown remote connections
- Unexpected account creation
- Unusual administrator activity
- Unexpected configuration changes
- Unexpected engineering workstation behavior
- Missing or altered logs
- Unusual CPU or disk activity
- Loss of operator visibility
- Unexpected controller behavior
- Unexpected changes in process communication
These indicators do not automatically prove compromise. They are signals that require investigation.
Detection and Analysis Come Before Major Containment Decisions
The first phase of an investigation should establish what is known and what remains uncertain.
Important questions include:
- What was observed?
- When did it begin?
- Which system was involved?
- Is the behavior still occurring?
- Is the affected system connected to critical OT systems?
- Is there evidence of lateral movement?
- Is there evidence of unauthorized remote access?
- Could the event affect physical operations?
The objective is to establish enough situational awareness to make a safe response decision.
Preserve Evidence Before It Disappears
Industrial incidents can produce evidence across many different sources.
- Firewall logs
- Network traffic captures
- VPN logs
- Authentication records
- Windows or Linux logs
- Engineering workstation records
- SCADA logs
- Historian information
- System configuration
- Application logs
- Security monitoring data
Some evidence may be overwritten quickly. Some may disappear when a system is restarted or reconfigured.
Evidence preservation should therefore be part of the response plan rather than something considered after the investigation has already changed the environment.
OT Digital Forensics Requires Additional Care
NIST has published a dedicated Digital Forensics and Incident Response framework for OT because conventional digital forensics does not always address the unique characteristics of operational environments.
OT forensic work may need to consider system availability, process safety, specialized devices, proprietary technologies and the potential effect of investigative actions on operations.
This makes forensic planning particularly important when investigating industrial systems.
Build a Timeline
A useful incident investigation should establish a timeline.
| Time | Evidence | Question |
|---|---|---|
| Initial observation | Alert or anomaly | What was first noticed? |
| Prior activity | Authentication or network logs | Was there unusual activity beforehand? |
| First confirmed event | System or application evidence | When did the incident become apparent? |
| Response action | Firewall or access change | What containment action was taken? |
| Post containment | Monitoring evidence | Did suspicious activity stop? |
A timeline helps separate facts from assumptions and can reveal gaps in the investigation.
Containment Must Consider the Industrial Process
Containment means limiting the incident so that it cannot continue to cause harm or spread further.
In OT, containment can take different forms depending on the architecture.
- Restricting network communication
- Disabling an unauthorized remote access path
- Blocking specific communication at a firewall
- Separating a compromised workstation
- Disabling a compromised user account
- Removing unnecessary external connectivity
- Increasing monitoring around affected systems
The appropriate response depends on the affected system and the operational consequences.
Do Not Automatically Power Off an OT System
Powering down a compromised device can remove an attacker’s access, but it can also interrupt a physical process and destroy volatile evidence.
Whether shutdown is appropriate should therefore be determined by qualified personnel who understand the process and the security situation.
In some cases, network containment may provide a safer first step. In other cases, immediate shutdown may be necessary.
There is no universal response action that should be applied to every industrial system.
Separate Cybersecurity Decisions From Process Safety Decisions
Cybersecurity responders should not make process control decisions without understanding the operational environment.
Similarly, plant personnel should not be expected to make complex forensic decisions without appropriate cybersecurity support.
A mature response capability brings these functions together.
- OT engineering
- Operations
- Cybersecurity
- IT
- Safety
- Management
- Relevant vendors
The exact participants depend on the organization and incident.
Investigate Lateral Movement
Finding one compromised system does not necessarily mean that the incident is limited to that system.
Investigators should determine whether the affected system communicated with other systems before or during the suspected incident.
Useful evidence can include:
- Firewall logs
- Network flow records
- Packet captures
- Authentication events
- Remote access logs
- Endpoint telemetry
- DNS activity
- System logs
This is where network visibility becomes critical.
Use Network Evidence Carefully
Network evidence can help establish relationships between systems without immediately interacting with the industrial devices themselves.
Packet captures can reveal communication patterns, unexpected connections and protocol activity. Firewall and flow data can provide additional evidence about source and destination relationships.
The purpose is to reconstruct what happened while minimizing unnecessary interaction with the affected process.
Determine Whether Credentials Were Compromised
Unauthorized account activity can provide an important clue during an investigation.
Review authentication evidence for:
- Unexpected successful logins
- Repeated authentication failures
- New accounts
- Unexpected privileged access
- Remote logins outside normal hours
- Accounts used from unusual locations
- Disabled or locked accounts
Credentials should not be assumed compromised solely because an unusual login occurred. The event should be correlated with other evidence.
Eradication Should Follow Understanding
Removing malware or deleting a suspicious account may be necessary, but eradication should not begin before the investigation has established enough information to understand the incident.
Otherwise, responders may remove evidence while leaving the underlying access path intact.
Before major eradication actions, consider:
- What caused the incident?
- How did access occur?
- Which credentials were involved?
- Which systems were affected?
- Is lateral movement suspected?
- Is persistence suspected?
- Are there additional exposed systems?
Recovery Is More Than Restoring a Backup
Recovery means returning the affected environment to a trustworthy and operational state.
Depending on the incident, this may involve:
- Restoring systems
- Restoring configurations
- Validating PLC logic
- Checking engineering workstations
- Reviewing firewall rules
- Rotating credentials
- Removing unauthorized accounts
- Validating network communication
- Monitoring for recurring suspicious activity
The system should not be considered fully recovered simply because it has restarted.
Verify PLC Logic and Engineering Configuration
Industrial incidents can involve engineering systems and controller configurations.
Where appropriate, recovery should include verification that important configurations, programs and engineering data match trusted versions.
Reliable backups are therefore an important part of OT incident preparedness.
Keep Backups Separate and Tested
A backup that has never been tested may not provide reliable recovery capability.
OT recovery information can include:
- System configurations
- PLC programs
- Engineering project files
- Network configurations
- Server backups
- Application configurations
- Relevant documentation
Backups should be protected from the same incident that could compromise the production systems and should be periodically tested according to the organization’s recovery requirements.
Validate Before Returning Systems to Normal Operation
Recovery should include validation.
- Are unauthorized accounts removed?
- Are suspicious connections gone?
- Are firewall rules correct?
- Are remote access controls functioning?
- Are trusted configurations restored?
- Is monitoring active?
- Is the industrial process operating normally?
The validation process should involve the people responsible for the relevant industrial systems.
Document What Happened
A technically successful recovery is not the end of the incident.
The organization should document the timeline, evidence, decisions, containment actions, recovery actions and remaining risks.
This information can reveal weaknesses that should be addressed before the next incident.
Perform a Post Incident Review
The final question should not be only:
“How did we recover?”
It should also be:
“Why was this possible, and what should change?”
- Was the asset unnecessarily exposed?
- Was remote access too broad?
- Was segmentation insufficient?
- Were credentials poorly controlled?
- Was monitoring inadequate?
- Were backups available?
- Was the incident response plan practical?
- Did engineering and cybersecurity teams communicate effectively?
Test the Incident Response Plan Before a Real Incident
An incident response document sitting in a folder does not create a response capability by itself.
Organizations should exercise their response procedures using realistic scenarios appropriate to their environment.
Examples can include:
- Compromised engineering workstation
- Unauthorized vendor remote access
- Loss of SCADA visibility
- Ransomware affecting supporting OT systems
- Unexpected PLC configuration change
- Compromised remote access credentials
- Suspicious network communication
CISA recommends that organizations maintain and regularly drill IT and OT incident response plans, including scenarios relevant to the organization’s environment.
Common OT Incident Response Mistakes
- Disconnecting systems without understanding operational consequences.
- Powering down equipment before preserving important evidence.
- Treating OT exactly like enterprise IT.
- Ignoring the process owner during containment.
- Failing to identify the complete affected scope.
- Destroying evidence during eradication.
- Restoring systems without validating configurations.
- Failing to investigate compromised credentials.
- Ignoring remote access pathways.
- Failing to review network traffic.
- Using untested backups.
- Never exercising the incident response plan.
What a Professional OT Incident Response Assessment Should Deliver
A practical OT incident response review should determine whether the organization can detect, investigate, contain and recover from an industrial cyber incident without unnecessarily increasing operational risk.
- OT incident response plan review
- Asset and network dependency review
- Incident escalation assessment
- Evidence source identification
- Logging and monitoring review
- Remote access investigation capability
- Containment decision review
- Backup and recovery review
- PLC and engineering configuration recovery review
- Vendor escalation process
- Incident response exercise
- Risk prioritized recommendations
OT Incident Response Decision Framework
| Question | Decision |
|---|---|
| Is there evidence of an active incident? | Start formal investigation and escalation. |
| Could the affected system influence a physical process? | Include qualified operations and engineering personnel. |
| Is the attack path known? | Identify and restrict the relevant communication. |
| Could containment affect production or safety? | Evaluate operational consequences before acting. |
| Is important evidence volatile? | Preserve appropriate evidence before disruptive actions. |
| Has the incident been contained? | Validate that suspicious activity has stopped. |
| Can trusted configurations be restored? | Begin controlled recovery. |
| Has the root cause been addressed? | Complete remediation and post incident review. |
Final Perspective
OT incident response is not simply an emergency version of IT incident response.
The responder must understand both the cyber event and the industrial process it may affect.
The strongest response capability begins with preparation: accurate asset information, network visibility, trusted backups, clear escalation paths, known evidence sources and tested procedures.
When an incident occurs, the priority should be to establish situational awareness, preserve useful evidence, contain the threat in a controlled manner and protect the safety and reliability of the industrial process.
Recovery should then restore trusted configurations, validate the environment and address the conditions that allowed the incident to occur.
The best incident response plan is not the one with the longest document. It is the one that engineers, operators, cybersecurity personnel and management can actually use when the plant is under pressure.
Need an Independent OT Incident Response Readiness Review?
Cautivra provides focused OT incident response readiness reviews and investigation support for organizations that need to understand their ability to detect, investigate, contain and recover from industrial cybersecurity incidents.
The review can examine OT assets, network architecture, remote access, logging, evidence sources, backup capability, containment procedures and recovery processes.
Request an OT Incident Response Review
Sources and Further Reading
- NIST SP 800-82 Rev. 3, Guide to Operational Technology Security
- NIST SP 800-61 Rev. 3, Incident Response Recommendations and Considerations for Cybersecurity Risk Management
- NISTIR 8428, Digital Forensics and Incident Response Framework for Operational Technology
- CISA, Developing an Industrial Control Systems Cybersecurity Incident Response Capability
- CISA, Cybersecurity Incident and Vulnerability Response Playbooks

Hi, this is a comment.
To get started with moderating, editing, and deleting comments, please visit the Comments screen in the dashboard.
Commenter avatars come from Gravatar.