Next-Gen Firewall · START

The Forcepoint model: SMC, engines, elements and Install policy

How the Security Management Center and the engines work together, what elements and policies are, first contact and the Install policy workflow, and your first two labs.

36 min read7 chapters2 labs15 quiz7 scenarios15 interview Q&A

This first module is free: read the lesson and take the quiz. Labs need an account: sign up free to run up to 3 hands-on labs.

Log inStart free trial
This reading material cannot be printed. Read it on config.nwkings.com.
01

Firewall basics recap and what you will learn

What you will learn in this module. You will learn how a Forcepoint-style Next Generation Firewall (NGFW) is managed: what the Security Management Center (SMC) holds, what the engine does, how elements and policies are built, how an engine first makes contact with the SMC, and how the Install Policy step turns your edits into real protection. The tagline of this track is: click-ops, but you understand what each click does. You will also do two hands-on labs in the CONFIG simulator: add the engine and make first contact, and create the element set.

Prerequisites. CCNA-level networking (IP addresses, subnets, default gateway, VLAN basics, TCP and UDP ports) and the firewall basics chapter in the sec-start module (Firewalls from zero) or the firewall start module of your choice. We recap the key ideas below, but we do not teach them from zero again.

This is independent training based on published documentation. It does not claim to cover any certification. If you plan to take an exam, check Forcepoint's current training and certification pages. Wherever behaviour depends on the product release, check the documentation for your NGFW version. The CONFIG console is an original generic console, not Forcepoint software.

Start with an analogy: the security desk and the rule book

Imagine a company with one head office and many branch offices. Every building has a security desk. The guard at the desk checks each visitor against a rule book: who is coming, where they want to go, what they want to do, and whether the answer is yes or no. If every branch guard wrote their own rule book, the company would have chaos: different answers at different doors. So the head office writes one rule book and sends a copy to every guard. The guards do the checking at the door, but the head office controls the rules.

That is the Forcepoint idea in one sentence: rules are written centrally in the SMC, and enforced locally by the engines.

Recap: what a firewall rule says

A firewall rule answers four questions:

  • Source: who starts the connection (a host, a network, a range).
  • Destination: where it wants to go.
  • Service: which protocol and port (for example HTTPS is TCP 443).
  • Action: what to do. Allow lets it through. Discard drops it silently. Refuse drops it and tells the sender.

Rules are read from top to bottom and the first match wins. If nothing matches, the connection is dropped. This is the implicit deny at the end.

Recap: stateful connections

A stateful firewall remembers each allowed connection. You write a rule for the first packet going out, for example Staff LAN to the internet on HTTPS. The reply packets coming back are recognised as part of that same connection and are allowed automatically. You do not write a second rule for the return traffic. Because of this memory, the firewall can also drop packets that claim to belong to a connection but do not.

Why manage firewalls centrally

One firewall is easy. Twenty branch firewalls are not. Logging into each one to add a rule is slow, and mistakes creep in: one branch gets the rule, another forgets. Central management gives you four things: consistency (one definition of "Staff LAN" used everywhere), speed (change once, install on many), visibility (one place to see status and logs) and audit (you know what was installed and when).

SMC (NK-SMC)10.0.0.10elements + policies + logs Engine (NK-PUNE-FW)control IP 10.0.0.1processes the traffic Install policy logs and status come back Staff LAN Internet Admin edits here (not in the traffic path)

The SMC is the head office rule book. The engine is the guard at the door. User traffic goes through the engine only.

The lab company: Network Kings, Pune branch

All examples use an invented company, Network Kings, with a Pune branch. The SMC is NK-SMC at 10.0.0.10/24 on the management network. The engine element is NK-PUNE-FW with four interfaces: eth0 control 10.0.0.1/24, eth1 WAN 203.0.113.2/30 (the ISP gateway is .1), eth2 Staff LAN 10.1.1.1/24 and eth3 DMZ 172.16.20.1/24. A PC STAFF-PC (10.1.1.10) sits on the Staff LAN and a web server WEB1 (172.16.20.20) sits in the DMZ.

Worked example. Network Kings wants Staff to browse the web. In plain words: source Staff LAN (10.1.1.0/24), destination any, service HTTPS, action Allow. On a single firewall you type that rule on the box. With central management you create the network element net-staff once, write the rule in a policy in the SMC, and press Install policy. The engine receives a compiled copy. Tomorrow, when a second branch needs the same rule, you reuse the same elements.

A first look at the console

The CONFIG simulator console has these pages: Engines (the firewall devices you manage), Elements (hosts, networks, services), Policies (rules), Inspection, VPN, Logs, Monitoring and alerts, and Tools. A bar at the bottom or top, the Install policy bar, is where changes become real. In this module you use Engines, Elements, Policies and the Install policy bar.

Common mistake. Thinking that saving a rule in the SMC protects the network. Saving only changes the SMC database. The engine keeps enforcing its old policy until you install. You will meet this again in chapter 6.

Exam trap. The SMC is the management plane, not the traffic path. If the SMC is switched off, the engines keep processing traffic with the last installed policy. Only management, new installs and central logging are affected.

Three branches, three different answers

A retail company ran three branch firewalls, each configured by a different engineer. A new web application was allowed at Pune and Delhi, but Mumbai users got blocked. The engineer found that Mumbai's rule used a typed subnet with a wrong mask. After the company moved to central management, "Staff LAN" became one named element and one rule, installed on all three engines. The error could not repeat.

Lesson: central management removes copy-paste drift, but only if you reuse named elements instead of typing values each time.

"Why would a company manage firewalls centrally, and does the management server sit in the traffic path?"

Strong answer: for consistency, speed, visibility and audit. The management server holds the configuration and logs, but traffic is processed by the engines, so if the management server is down, traffic still flows under the last installed policy. Mention that you would still protect and back up the management server because without it you cannot change policy.

Key takeaways

  • A rule is source, destination, service, action; first match wins; no match means discard.
  • Stateful firewalls allow reply traffic of an allowed connection automatically.
  • Central management gives consistency, speed, visibility and audit.
  • Rules live in the SMC; traffic is processed by the engine; Install policy connects the two.
  • Lab company: Network Kings, SMC NK-SMC 10.0.0.10, engine NK-PUNE-FW.

Upgrade to keep reading

The free plan lets you read chapter 1 of this module. Upgrade to open the remaining chapters.

  1. The architecture: SMC, engines and the Management Client
  2. Elements: the named building blocks
  3. Policies, templates and sub-policies
  4. How an engine contacts the SMC: initial contact and trust
  5. The Install Policy workflow
  6. Management Client vs engine command line, and the module summary

Join the community