Full Firewall Audit¶
Difficulty: Advanced | Time: 20-45 minutes | Risk: Read-only
Problem Statement¶
You need a comprehensive security review of all OPNsense firewall rules across every interface. This could be triggered by a compliance review, a suspected breach, or routine security hygiene after accumulating rules over months of operation.
Prerequisites¶
- OPNsense plugin installed and authenticated
- Read access to firewall, security, and interfaces tools
- Netex umbrella plugin recommended for cross-vendor audit via NetworkSecurityAgent
Workflow¶
Step 1: List All Rules by Interface¶
The plugin calls opnsense__firewall__list_rules for each interface. The output includes rule count, any/any rules, and disabled rules that may be re-enabled accidentally.
Step 2: Identify Overly Broad Rules¶
The plugin scans for rules where source, destination, port, and protocol are all set to any. These rules effectively bypass network segmentation.
Look for:
- pass any from any to any on non-WAN interfaces
- Rules with source any and destination any on VLAN interfaces
- Allow rules that predate more specific deny rules (order matters)
Step 3: Check Rule Ordering¶
OPNsense evaluates rules top-to-bottom, first match wins. A broad allow rule above a specific deny rule shadows the deny, making it ineffective.
Step 4: Verify Inter-VLAN Isolation¶
For each untrusted VLAN, verify: - Explicit deny rule blocking traffic to management subnets - Deny rule positioned before any broader allow rules - No NAT rules that bypass the firewall intent
Step 5: Review NAT Rules¶
The plugin calls opnsense__firewall__list_nat_rules. Look for:
- Port forwards from WAN to internal hosts (attack surface)
- 1:1 NAT mappings that expose entire hosts
- Outbound NAT rules that break intended isolation
Step 6: Check Alias Definitions¶
Aliases that reference stale IPs or overly broad networks weaken rules that depend on them.
Step 7: Cross-Vendor Audit (if netex installed)¶
The NetworkSecurityAgent performs automated analysis across all seven finding categories, producing a severity-ranked report.
Expected Output¶
A severity-tiered report:
- CRITICAL: Management plane exposed to untrusted segments
- HIGH: Any/any rules, missing inter-VLAN isolation, stale port forwards
- MEDIUM: Rule ordering risks, aliases with stale values
- LOW: Disabled rules, informational notes
Working Safely¶
This workflow is entirely read-only. The NetworkSecurityAgent never makes changes -- it only reports findings. To remediate findings, use the specific write commands with --apply.
Required safety notice: Network changes can result in outages that disconnect you from your ability to correct them. Never make changes to a network you cannot reach through an out-of-band path (serial console, IPMI/iDRAC, a separate management VLAN on a different physical interface, or physical access). Netex will assess this risk for you, but it cannot guarantee your recovery path -- only you can verify that.