Skip to content
← All services

Service

Security

Palo Alto, FortiGate — policies, NAT, VPN.

Scope of work

  1. 01Security posture assessment and risk identification
  2. 02Zone-based security architecture design
  3. 03Next-Generation Firewall (NGFW) deployment and configuration
  4. 04Application- and user-based policy (App-ID, User-ID)
  5. 05Remote-access VPN (GlobalProtect, SSL-VPN) and IPSec site-to-site VPN
  6. 06Firewall High Availability configuration
  7. 07Logging, monitoring and security alerting
  8. 08Periodic policy review, optimisation and documentation

Some organisations have bought an expensive firewall, but when I open the configuration I find an "allow any any" rule created on deployment day to get things working and never narrowed down. Good hardware, almost no security. Network security is not something added after the system is running. It must be designed from the start alongside the network architecture, and then operated continuously.

What is an enterprise firewall — and how does it differ from Windows Firewall or a router?

A traditional firewall filters by IP address and port: allow or block. Today most applications run over the same few ports such as 80 and 443, so that approach is no longer enough. A Next-Generation Firewall (NGFW) looks deeper: which application is running, which user is using it, and whether the behaviour is abnormal.

App-ID and User-ID — writing policy in the language of the business

With Palo Alto Networks, App-ID identifies the real application regardless of the port it uses, and User-ID ties traffic to specific users from Active Directory. Policy can then be written the way a manager thinks: "finance may use the accounting application, guests may only browse the web, nobody may use anonymous file-sharing apps". Policy follows people and applications, not IP addresses that change constantly.

Threat Prevention — blocking threats in real time

An NGFW runs layers of protection in parallel: detecting and blocking exploits, malware and connections to malicious command-and-control servers, plus URL filtering by category. These layers rely on threat intelligence updated continuously by the vendor, so keeping subscriptions active is part of security, not an optional extra.

Zone-based design — the foundational principle

A firewall can only protect what passes through it. The network therefore needs clear zones: users, servers, DMZ, management, cameras and IoT, guests. Traffic between zones is controlled on a "deny by default, allow only what is needed" basis. This limits damage: if one zone is compromised, an attacker cannot move freely into the rest.

VPN — remote access and site-to-site connectivity

Remote-access VPN lets staff outside the office connect securely to internal systems. Palo Alto GlobalProtect and FortiGate SSL-VPN both support device posture checks before granting a connection. IPSec site-to-site VPN joins branches to one internal network over an encrypted tunnel, replacing expensive private circuits. What I always recommend is two-factor authentication on every remote-access VPN, because a password alone is the weakest link.

FortiGate — an integrated solution

Fortinet FortiGate combines several functions in one appliance: firewall, IPS, antivirus, web filtering, application control and VPN, kept current through FortiGuard. For small and medium organisations or multi-branch networks, FortiGate balances capability against investment cost.

High Availability — when the firewall cannot be allowed to fail

The firewall is the choke point for all inbound and outbound traffic, so if it fails the whole system loses connectivity. High Availability uses two units running together: if the primary fails, the secondary takes over almost instantly and sessions are preserved. For 24/7 buildings such as hotels and office towers, this should be part of the design from day one.

Logging, monitoring and continuous operation

The mistakes I see most are not in the hardware but in operations: overly broad rules never narrowed, no logging so anomalies go unnoticed, firmware patches not applied, no policy review process. Real security is a process, not a one-time configuration.

When does network security need reassessing?

When a firewall has run for years without a rule review, when remote work becomes necessary, when branches are being connected, when cameras and IoT share the internal network, or after any incident suspected to be a breach.

Need advice on this solution?

Get in touch