Case Study

How a Global Oil & Gas Company Brought Governed Orchestration to a Segmented OT Network With Itential

A network engineering team replaced manual, device-by-device changes with governed, auditable orchestration built to operate inside strict OT segmentation and compliance requirements, turning IOS upgrades, VM provisioning, and BGP changes into repeatable workflows without opening new security exposure.

Challenge

Manual, error-prone network changes across a segmented OT environment, with no orchestration layer built to operate inside the team’s security and compliance boundaries.

Solution

Itential orchestrated IOS upgrades, VM provisioning, and BGP configuration management into governed, auditable workflows built to run inside the team’s existing segmentation and change control requirements.

icon showing a checkmark on lines of text or code

Impact

Manual, multi-hour network changes now run in minutes, with every action logged and governed, cutting risk exposure while freeing the team to focus on higher-value work.

The Challenge

Manual Work in a Network Built to Stay Closed

The network engineering team at a major oil and gas company runs a complex infrastructure spanning multiple regions and thousands of devices supporting critical business operations. The tools to manage it well were already in place: NetBox as the source of truth, SolarWinds for monitoring, ServiceNow for change management, GitLab for version control, vSphere for virtualization, but nothing orchestrated across them. Every iOS upgrade, every VM request, every BGP change meant an engineer stitching those systems together by hand.

The environment made the stakes higher than most. As a critical infrastructure operator, the team’s OT networks run under strict segmentation requirements: no persistent external connections, tightly scoped access, every integration justified on its own merits.

That posture isn’t caution for its own sake. It’s how critical infrastructure networks have to be built. A compromised or mismanaged connection into a control network doesn’t just risk downtime; it risks a safety event, a regulatory finding, or a production outage with consequences well beyond the network team. Any platform brought into this environment had to operate inside the same segmentation and change control standards, industry frameworks like IEC 62443 and the team’s own management-of-change process, that already governed everything else on the network. Fully governed and fully auditable wasn’t a nice-to-have. It was the baseline requirement.

Secrets management was where that gap showed up most concretely, and where the risk was hardest to ignore. Credentials were managed out of local instances of a password vault kept on shared jump boxes, workable when the team had a handful of devices to touch, but never built for a fleet in the thousands, and not the kind of setup a security team wants sitting in front of OT infrastructure.

A shared jump box with locally stored credentials is exactly the sort of soft target that becomes a pivot point into more sensitive parts of the network if it’s ever compromised. “It’s not an automation-friendly way of doing it,” as one team member put it, and the same manual pattern ran through everything else that touched the network: upgrade pre-checks run by hand against multiple systems before every change, and virtual machine requests that moved through GitLab, Terraform, and a change board with a person in the loop at every handoff.

You’d be surprised the tolerance of pain we’ll go to in order to keep these OT environments segmented, with no persistent connections external.

OT Security Lead

Global Oil & Gas Company

Why Itential

Why They Chose Itential

The team’s technology stack was already in place. What it lacked was a platform to orchestrate across it without asking the security team for new exceptions.

Any orchestration layer here had to clear two bars at once: connect the systems the team already ran, and do it without loosening the segmentation and change control the environment demands. Itential cleared both.

An Orchestration Layer Built for a Regulated, Segmented Environment

Explore the Platform

Orchestrates the Stack You Already Run

Itential connected directly into NetBox, SolarWinds, ServiceNow, GitLab, Cisco, and vSphere and orchestrated across them, rather than asking the team to consolidate onto a new set of tools.

Built for Regulated, Segmented Environments

Every workflow was designed to operate inside the team’s existing OT segmentation and change control requirements, including standards like IEC 62443 and formal management-of-change, rather than asking for new exceptions to reach the network.

Governed, Repeatable Workflows

Every workflow runs the same way every time: pre-checks, validation, and post-checks built into the sequence, replacing tribal knowledge with an orchestrated, auditable process.

Audit-Ready by Default

Every orchestrated action is logged and traceable: who touched a device, when, and under what approval, turning routine changes into evidence a compliance review or audit can actually use, rather than something the team has to reconstruct after the fact.

Self-Service Through ServiceNow

Orchestrated workflows are surfaced directly in the team’s ServiceNow instance, so requesters can kick off an IOS upgrade, a VM request, or a BGP change from the same catalog they already use for change management. No new portal, no new place to go.

High-Availability by Design

The platform runs on a full HA architecture, a three-node etcd cluster and three-node MongoDB configuration, built to support production-scale orchestration, not a lightweight scripting layer.

Built to Scale to the Full Fleet

With three workflows orchestrated, the team has a foundation to extend the same model across its full device infrastructure and take on new use cases as they come up.

We weren’t looking for another tool to manage. We already had NetBox, SolarWinds, ServiceNow: what we didn’t have was anything orchestrating across them. Itential didn’t ask us to rip any of that out. It just finally connected it.

Director of Network Infrastructure

Global Oil & Gas Company

The Orchestration

Three Core Workflows, Orchestrated Inside the Existing Security Model

The team built its orchestration layer around three high-friction workflows, tying its existing systems together into governed, repeatable processes rather than one-off scripts. Every workflow was designed to run inside the boundaries already in place: no new external connections, no changes to how OT segmentation worked, and credentials pulled through the same governed access model instead of a new one.

Command templates handle CLI interaction, a transformation engine handles data manipulation, and configuration management runs automated backups and diffs underneath all three: the connective tissue that turns individual automations into an orchestrated operating model, without asking the security team to trust anything new.

IOS Upgrade Orchestration

Pre-checks on CPU, memory, and disk space, device health pulled from SolarWinds, software versions retrieved from NetBox, the upgrade itself with built-in validation, and post-upgrade verification are now sequenced and orchestrated as a single workflow, not a per-device checklist an engineer runs by hand.

VM Vending Orchestration

Virtual machine provisioning is orchestrated end-to-end across GitLab and Terraform, moving a request from submission through change approval to deployment with retry logic and error handling built directly into the workflow.

BGP Configuration Orchestration

Configuration backup, Jinja2 template-based rendering, deployment, and validation of both BGP and routing tables on data center routers are now coordinated as one orchestrated sequence instead of a manual, router-by-router change.

The pre-checks and validation built into the IOS upgrade workflow weren’t a nice-to-have. They replaced a real, remembered risk. Manual changes to core infrastructure carried consequences the team had seen firsthand.

That kind of change live could go sideways very easily, very quickly. We bricked the main router and everything was gone.

Senior Network Engineer

Global Oil & Gas Company

The Outcome

Results & Value Delivered

Orchestrating its highest-friction workflows changed more than the time on the clock. It changed how the team operates, and how much risk it carries while doing it:

Operational Efficiency: Manual, repetitive tasks that once consumed significant engineering time are gone from the workflow entirely. iOS upgrades, VM provisioning, and BGP changes that took hours of hands-on effort across multiple systems now run in minutes through orchestrated execution.
Enhanced Reliability: Automated pre-checks and validation, orchestrated into every workflow, catch problems before they become outages. CPU, memory, disk space, and software versions are all verified automatically before an iOS upgrade proceeds, replacing the kind of live, manual change that used to put core infrastructure, and the safety and compliance obligations tied to it, at risk.
Improved Visibility: Orchestrated integration with SolarWinds and NetBox gives the team real-time visibility into device health and configuration state, replacing point-in-time manual checks with continuous, data-driven insight that also strengthens the team’s audit and compliance posture.
Team Enablement: Building the orchestration layer alongside Itential gave the network engineering team direct, hands-on fluency with the platform: command templates, the transformation engine, workflow design, so they can extend and maintain these workflows themselves rather than depending on outside support for every change.
A Foundation to Scale: With its orchestration layer proven across three workflows, the team has a foundation to extend the same model across its full device infrastructure and take on new use cases as they come up, not a one-off project, but infrastructure for the next one.

Together, these gains reframe network operations from a set of individual scripts into a governed, repeatable orchestration model: the difference between fixing today’s bottleneck and building the platform that absorbs tomorrow’s, without ever loosening the security posture the environment demands.

95%

Reduction in manual effort per network change

1,200+

Engineering hours reclaimed annually

60%

Fewer change-related incidents

2 FTEs

Worth of capacity redirected to strategic engineering work

A Foundation to Scale

With its orchestration layer proven across three workflows, the team has a foundation to extend the same model across its full device infrastructure and take on new use cases as they come up, not a one-off project, but infrastructure for the next one.

The platform paid for itself the first time we didn’t have to babysit a router through an upgrade at 2 a.m.

Senior Network Engineer

Global Oil & Gas Company

Looking Ahead

From Orchestrated Automation to Agentic Operations, Without Loosening Governance

The team’s next chapter isn’t just more workflows. It’s a different way of running them, without loosening the governance the environment requires. With iOS upgrades, VM vending, and BGP management proven out as orchestrated, governed processes, the team is now exploring how FlowAgents and Itential’s FlowAI capabilities could sit on top of that same orchestration layer: letting engineers describe an intent in natural language and have it reasoned through and executed against the workflows, integrations, and guardrails already running in production, rather than building each new automation by hand.

It’s a natural next step rather than a new platform decision. The orchestration layer already in place, the same command templates, transformation logic, and validated integrations across NetBox, SolarWinds, GitLab, vSphere, and ServiceNow, becomes the trusted execution layer an agent reasons over, with every action still governed, auditable, and scoped inside the same OT segmentation and compliance boundaries the team already requires for production changes. Agentic doesn’t mean less governed. In an environment like this, it can’t.

Keep Learning