ICS / OT Testing

How one exposed network segment led to a simulated train-control outage

Client
Passenger rail operator
Service
ICS and network penetration testing
Findings
35+, including 4 Critical
Verified
Fixes re-assessed by ioSENTRIX
The challenge

What did the client need?

A passenger rail operator asked ioSENTRIX to test the security of its communication-based train control system. A successful attack could stop operations for days and put passengers at risk.

Engagement details

Client
Passenger rail operator
System
Communication-based train control (CBTC) in service for several years
Scope
Segmented networks, proprietary equipment and several ICS systems
Environment
Lab replica of production, without live trains
Engagement
A multi-week engagement
Verified
Fixes re-assessed by ioSENTRIX
Tested by
ioSENTRIX ICS and network testers
Our approach

How did ioSENTRIX approach it?

There are no open testing standards for much proprietary rail and ICS software, so there was nothing for a scanner to check against. Finding this took learning how the equipment works and testing it by hand in a production-like lab.

Severity was rated against NIST standards. The report gave the operator a risk assessment of the whole system and recommendations it could act on without stopping a live service.

What we found

What did ioSENTRIX find?

From an exposed segment to the nodes that control train communications.

Theme · NetworkFinding 1

Unpatched bridging nodes let testers cross segments

The nodes that connected several segments ran unpatched Windows, left unpatched because the network was air-gapped. From an exposed segment, our testers compromised those nodes, then pivoted from them into other segmented networks.

Business impact. Segmentation slowed an attacker down but did not stop one from reaching operational systems.

Theme · ICSFinding 2

Train communication nodes could be reached and shut down

Our testers gained access to sensitive nodes that control train system communication. In the lab replica, simulated shutdowns of several trains produced an outage across different operating regions.

Business impact. An attack of this kind could halt service, cost days of downtime and put passenger safety at risk.

One chain from an exposed segment to a simulated multi-region outage
One chain · segment to train control
  1. 1Exposed segmentinitial foothold
  2. 2Pivot nodessegments connected
  3. 3Train commscontrol nodes reached
Outcome: Simulated outage (trains stopped across regions)
The impact

What changed as a result?

Starting from one exposed network segment, our testers pivoted across segmented networks to the nodes that control train communications, and simulated shutdowns that caused an outage across operating regions in the lab. The client fixed the findings and ioSENTRIX re-assessed them.

1
exposed network segment was enough to start
35+
findings, 4 of them Critical
0
live trains touched: every test ran in a lab replica

How many findings, and how severe?

35+ findings in total, including 4 Critical and 10 High.

  • Critical 4
  • High 10
  • Medium
  • Low

What changed after the test?

What changed in the train control environment
AreaBefore the testAfter remediation
SegmentationUnpatched Windows on bridging nodes allowed pivoting between segmentsFixes implemented by the operator’s technical staff
Train communicationsControl nodes were reachable; simulated shutdowns caused a multi-region outage in the labFixes implemented, then re-assessed by ioSENTRIX
Incident responseThe plan did not account for these attack pathsThe new threats were added to the incident response plan

✓ Verification: Fixes re-assessed by ioSENTRIX.

Key insight

Built to fail safe, the system had not been built to keep an attacker out.

Lessons

Lessons for operators of industrial systems

01

Air-gapped is not patched

The nodes bridging segments ran unpatched Windows because the network was air-gapped. Isolation delayed patching; it did not remove the risk.

Finding 1 · NIST SP 800-82 Rev. 3, Guide to Operational Technology (OT) Security
02

Test hard, in a faithful lab

A replica of production let our testers push all the way to a simulated outage without touching a live train.

Approach
03

Fail-safe is not cyber-safe

Safety engineering assumes faults, not attackers. Add attack scenarios like these to incident response, as this operator did.

Finding 2
Next step

Running operational technology? See a sample ICS test report or book a strategy session.

Service
ICS, IoT and IIoT penetration testing
Explore the service

Details in this case study are anonymised and generalised. Methods that could harm other products are withheld.

FAQ

Frequently asked questions

What did ioSENTRIX test in this case study?

A passenger rail operator asked ioSENTRIX to test the security of its communication-based train control system. A successful attack could stop operations for days and put passengers at risk.

What was the most serious issue?

Unpatched bridging nodes let testers cross segments. Segmentation slowed an attacker down but did not stop one from reaching operational systems.

How many findings were there, and how severe?

35+ findings in total, including 4 Critical and 10 High.

Were the fixes verified?

Fixes re-assessed by ioSENTRIX.

Which ioSENTRIX service does this case study cover?

ICS, IoT and IIoT penetration testing, delivered by the ioSENTRIX ICS and network testers.

Which standards does this align with?

NIST SP 800-82 Rev. 3, Guide to Operational Technology (OT) Security. These are cited as alignment guidance, not a compliance claim.

How to get started

Ready to strengthen your security? Fill out our quick form, and a cybersecurity expert will reach out to discuss your needs and next steps.
DecorativeDecorative