Security analysts working at monitoring consoles in a server room
Security policy

Palo Alto Security Policy and Firewall Rule Review

Every security and NAT rule, read and scored: any-any rules, internet-facing services, App-ID use, hit counts, shadowed rules and logging.

IT director holding a laptop with firewall audit results
Exposure

One NAT rule can publish remote desktop to the whole internet

We list every service your firewalls publish, who can reach it and what inspects it.

Two network engineers reviewing a Palo Alto firewall audit on a laptop at a rack
Clean-up

Rules nobody uses are rules nobody reviews

Hit counts show which rules have matched no traffic for 90 days or more, so you can retire them with confidence.

25+ yearsIT & cybersecurity experience
Certified expertsCISSP · C|CISO · Microsoft · Cisco
Southern CaliforniaLocal and on-site, based in Irvine
NationwideRemote audits & advisory
Security policy

What the security policy rule review covers

Rule bases grow. A vendor needs access, a project needs a port, a test rule stays in place. After a few years most Palo Alto firewalls carry rules that are broader than anyone intended, rules that never match, and NAT rules that publish internal services to the internet.

We read every security and NAT rule on each firewall and in Panorama device groups, with its hit count, App-ID use, security profiles and logging, and score 27 controls. You also get the full rule base in Excel, with each rule flagged.

Two IT engineers with laptops in a data center
The controls

27 Security policy, NAT & object hygiene controls we evaluate

Each control ends as Pass, Fail, Warning or Manual review, with the evidence behind it. Some checks raise or lower the default severity to match what was found.

Security Policy · 22 controls

IDControlDefault severity
SP-01No rule allows any source to any destination for any application and serviceCritical
SP-02Allow rules name the applications they permit (application is not “any”)High
SP-03Allow rules use application-default ports instead of “any” serviceMedium
SP-04Allow rules have security profiles attachedHigh
SP-05Inbound internet rules limit the permitted source addressesMedium
SP-06Remote administration and database services are not reachable from the internetCritical
SP-07Anonymizers and peer-to-peer applications are not allowedHigh
SP-08Remote-access tools are restricted to approved users or systemsMedium
SP-09Rules do not rely on negated source or destinationLow
SP-10Disabled rules are removedLow
SP-11No rule is shadowed (made unreachable) by an earlier ruleHigh
SP-12No rule duplicates another ruleLow
SP-13Rules are documented (description, owner, purpose)Low
SP-14Temporary rules are removed on timeMedium
SP-15An explicit, logged “deny all” rule closes the rulebaseLow
SP-16The rulebase is of a manageable sizeLow
SP-17Policy uses user identity (User-ID) for internal accessLow
SP-18Every enabled rule has been usedMedium
SP-19Rules unused for 180 days are reviewedLow
SP-20Rule usage counters cover a long enough periodLow
SP-21Deny rules block known-malicious IP addresses (threat intelligence feeds)Medium
SP-22Allow rules use application-default instead of custom port listsLow

Object Hygiene · 5 controls

IDControlDefault severity
OH-01Unused address objects and groups are removedLow
OH-02Unused service objects are removedLow
OH-03No address group is emptyLow
OH-04Duplicate address objects are consolidatedLow
OH-05Unused security profiles and profile groups are removedLow
Why it matters

What we look for and why

01

Overly broad rules

Rules with any source, destination, application and service turn a next-generation firewall into a router. We find them, along with port-based rules that ignore App-ID.

02

What the internet can reach

Inbound NAT and security rules decide what is published. We list each published service, flag remote desktop, file sharing, SSH and database ports, and check what inspects the traffic.

03

Dead weight and blind spots

Rules with no hits in 90 days, rules shadowed by a broader rule above them and rules that do not log at session end. Each one is listed with its last hit.

Sample

A finding from the sample report

SAMPLE · Veltrana Logistics (fictional company) · illustrative data
Area
Security policy
Finding
Two enabled rules allow all traffic
Severity
Critical
Evidence
Any-Out-Temp (trust to untrust) and Any-Any-Allow (any zone to any zone) allow every source, destination, application and service. Any-Any-Allow has no security profiles and more than 7 million hits.
Fix
Replace each rule with specific rules for the traffic that is actually needed, using the rule usage counters and traffic logs to identify it, then remove it.

Every finding in the real report has this shape. See the full sample report

How it works

A read-only audit in five steps

STEP 1

Scoping call

We confirm the firewalls, Panorama, sites, GlobalProtect and who on your side should be told before the run. 30 minutes.

STEP 2

Read-only access

Your administrator creates one Superuser (read-only) account on each firewall, or on Panorama if it manages them. That account is the only change, and your team makes it.

STEP 3

Read-only collection

The collector reads the configuration, rule hit counts and system status over HTTPS through the PAN-OS XML API, from a computer inside your network.

STEP 4

Review and scoring

141 controls in 13 areas are scored with evidence, including a review of the last 7 days of firewall logs. Ali reviews every finding, groups them into business risks and maps them to NIST CSF 2.0 and ISO/IEC 27001:2022.

STEP 5

Report and roadmap

Executive and technical reports, a rulebase review, a log review, a risk register, a roadmap and an Excel workbook, walked through on a call with your owner and your firewall administrator.

Read-only means read-only. The audit never commits, changes or deletes a setting, never installs software or content updates and never restarts a firewall. It is designed so your end users notice nothing.
Related

Where to go next

Security profiles & threat prevention

Which allow rules inspect traffic, and which only let it through.

Zone protection & exposed services

Flood and scan protection, Authentication Portal and User-ID on internet-facing zones.

This series

Page 4 of 13: Security policy & NAT rule review

Ali Hassani, CISO, OC Security Audit
CISO-led audit

Guided by Ali Hassani, CISO

Certified CISO (C|CISO) and CISSP with 25+ years in IT, cybersecurity, compliance and infrastructure, and dozens of Microsoft and Cisco certifications, including MCSE, MCITP, MCSA Security, CCNP and CCNA. Every Palo Alto audit is read-only, evidence-based and written so an owner and a firewall administrator can both act on it.

FAQ

Questions IT leaders ask

Do you change or clean up rules for us?

No. The audit is read-only. The rule-base workbook marks each rule to remove, tighten or keep, and IT Perfection, a business also owned by Ali Hassani, can make the changes if you want hands-on help.

How do you know a rule is unused?

PAN-OS keeps a hit count and a last-hit time for each rule. We read those counters and note when they were last reset. A rule with no hits in 90 days is flagged for review, not deleted.

Do you review Panorama rules?

Yes. Pre-rules and post-rules from Panorama device groups are read together with local rules, in the order the firewall evaluates them.

What do you need to read hit counts?

Only the read-only audit account. Hit counts are read with an operational show command; no counter is reset.

Will anyone notice the audit?

End users should not notice anything. Your firewall administrators and your SIEM will see logins by the audit account in the System log. We tell your IT or security contact the date, time, account and computer in advance.

Optional implementation help

OC Security Audit only audits, read-only. If you want hands-on help, IT Perfection, a managed IT business also owned by Ali Hassani, can implement the fixes and support ongoing firewall and VPN management. Using it is optional.

IT Perfection managed firewall & VPN
Engineer reviewing firewall logs at a data center desk

Know what your firewall really allows, before an attacker tests it

Schedule a consultation. We scope the Palo Alto firewall audit in 30 minutes and quote one fixed price.

Read-only audit. No changes to your firewalls. Irvine, Orange County and Southern California on-site; remote nationwide.