Greenbone / OpenVAS → Tenable Nessus

Greenbone / OpenVAS ↔ Tenable Nessus: integration to migration path.

OpenVAS deploys alongside Nessus first, proves it matches your real findings against three clean scan cycles, and only then takes over scope one asset class at a time. No flag day, no re-credentialing your estate — and every phase rolls back in minutes.

The honest end-state is partial retirement, not zero. Nessus stays for the PCI ASV external scans OpenVAS cannot certify for — we say so up front, because the rest of this only matters if you can trust it.

The idea

Scan in parallel first. Retire Nessus last — and only from internal scope.

The topology that makes this zero-downtime is two scan estates feeding one unified plane. OpenVAS and Nessus run against the same targets on offset schedules, both exporting into DefectDojo where findings dedup on CVE plus CPE plus hostname. DefectDojo becomes the durable system of record — tickets, exceptions and SLAs live there, not in either scanner — so cutover is a config flag, not a rebuild. OpenVAS earns each asset class by matching Nessus on real data before it takes over, while the ASV external scope never moves at all.

The phases

Seven steps. Each one reversible.

0

Baseline & inventory

We read your Nessus estate without touching it: scanner footprint, every scan policy and schedule, the top 500 plugins by detect-count, and which IPs sit inside PCI scope. Each asset is classified CDE / PCI-connected / out-of-scope.

Users see: No user impact.

Rollback: N/A — read-only.

1

OpenVAS stands up beside Nessus

Greenbone Community Edition and DefectDojo deploy in your network. Nessus keeps scanning production untouched; OpenVAS only runs against a non-prod range to prove the wiring. DefectDojo starts ingesting Nessus exports.

Users see: None — Nessus still owns every production scan.

Rollback: Decommission OpenVAS and DefectDojo. Nessus untouched.

2

Parallel-scan window

Both scanners run the same representative asset classes on offset schedules at least an hour apart. Findings dedup in DefectDojo on CVE plus CPE plus hostname, and a weekly delta report shows where OpenVAS matches, misses, or over-reports against Nessus.

Users see: Doubled scan load on the target subset only — watched and backed off if production feels it.

Rollback: Disable the OpenVAS schedules. Nessus and DefectDojo continue.

3

Per-asset-class cutover

Once three clean cycles show at least 85% overlap on critical findings, OpenVAS becomes the system of record for that class — non-prod first, then non-PCI Linux, Windows, network gear and databases. Nessus drops to a monthly verification cadence for 30 days before its schedule is cut.

Users see: Ticket source changes from Nessus-tagged to OpenVAS-tagged in DefectDojo — cosmetic if asset-group rollups are used.

Rollback: Re-enable the Nessus schedule. Under 15 minutes — it is left disabled, not deleted.

4

Exception migration

Every active Nessus exception and risk-acceptance is exported and recreated in DefectDojo keyed on the dedup hash, with its original approver, reason and expiry preserved. Where the hash differs between scanners, the security owner re-approves once.

Users see: None for finding consumers; the security team runs one re-approval pass for hash-changed exceptions.

Rollback: Reactivate Nessus exception state — DefectDojo risk-acceptances are additive, not destructive.

5

Decommission Nessus internal

With all internal classes on OpenVAS for 30 days, internal Nessus scan policies are disabled, internal credentials revoked, and the Nessus Agent fleet uninstalled (or kept where it covers roaming assets). The Tenable licence downsizes to an ASV-external footprint at renewal.

Users see: None at the finding-consumer layer; agent-hosted endpoints stop reporting once agents are removed — confirmed against OpenVAS coverage first.

Rollback: Re-enable internal Nessus policies and reinstall agents from MDM. Hours-to-day depending on fleet size.

6

Final state: partial retirement

Nessus is not retired. It stays active for ASV external scanning under PCI DSS v4.0 Req 11.3.2 — OpenVAS cannot certify as an ASV. OpenVAS owns all internal scope, and DefectDojo dedups across both as the unified plane.

Users see: None — steady state.

Rollback: N/A — this is the operational end-state.

Feature parity

Where OpenVAS matches Nessus — and where it honestly does not.

CapabilityGreenbone / OpenVASTenable NessusParity
Network scanning (unauth) openvas-scanner + Community Feed NVTs (Full and fast) Nessus plugins (Basic Network Scan) At parity
Authenticated scanning NVT LSC families with SSH/SMB/SNMP creds — Credentialed Patch Audit equivalent Nessus Credentialed Patch Audit plugins Partial
Agent-based scanning None first-party; osquery bolt-on only Nessus Agent for roaming, ephemeral and air-gapped endpoints SaaS only
Vuln content / feed cadence Community Feed (daily sync); lags Tenable days to weeks on high-pri/KEV CVEs; Enterprise Feed paid Tenable same-day plugin shipping Partial
Prioritisation scoring CVSS v3.1 only; EPSS + CISA KEV + criticality join in DefectDojo Tenable VPR (ML-driven 28-day predictive rating) Partial
PCI ASV certification Not ASV-certified; cannot certify the binary Tenable ASV service offering on the PCI SSC list SaaS only
Asset inventory Per-scan; no cross-scan auto-dedup Nessus Manager / Tenable.io asset registry Partial
Remediation workflow / ticketing DefectDojo Engagements to Jira/ServiceNow Nessus findings to console / ticketing integration At parity
NVT/plugin source inspection Community Feed NVTs are NASL source — readable, forkable Nessus plugins are precompiled .nbin binaries OSS only
Custom check authoring NASL — author NVTs, drop into the feed directory compliance audit-files + Custom Audit policies; bespoke plugins are contract-engagement Partial
API surface GMP (XML over TLS) + gvm-tools + python-gvm Nessus REST API (Tenable.io richer) At parity
Compliance scanning (CIS/STIG) Limited CE NVTs; Enterprise Feed adds policy compliance Nessus CIS Benchmark + DISA STIG audit-file content Partial
Deployment & HA Self-hosted GVM + Postgres, air-gap deployable, you own HA + ops Nessus Pro / Manager on-prem; Tenable.io cloud-only Partial
Cost model Self-hosted compute only; no per-asset licence Per-host / per-asset licence Partial

What we're honest about

The caveats most vendors leave out.

PCI ASV scans cannot move to OpenVAS

ASV certification attaches to the vendor org running the scans, not to the scanner binary — OpenVAS cannot be ASV-certified, ever. If any system in PCI scope is reachable from the internet, the four external scans a year under Req 11.3.2 must stay on Nessus (or another ASV). The honest end-state is partial retirement, and we brief your QSA before the migration, not after.

No first-party agent for roaming or ephemeral assets

This is the single largest functional gap. Nessus Agent catches off-VPN laptops and short-lived cloud workloads that an OpenVAS network scan misses entirely. We either keep Nessus Agents on those scopes as a documented exception, bolt on an osquery-plus-CVE-match pipeline, or accept the gap with explicit sign-off — your call, named up front.

Community Feed cadence lags Tenable

Tenable ships plugins same-day; the OpenVAS Community Feed can trail by days to weeks on high-priority and CISA-KEV CVEs. We close that with an EPSS-plus-KEV prioritisation overlay in DefectDojo and a nuclei template for KEV adds, or budget for the Greenbone Enterprise Feed — but we document the window as accepted risk.

No VPR, weaker compliance content, and you own the ops

Tenable's ML-driven VPR has no OSS equal; we rebuild ranking from EPSS, KEV and asset-criticality in DefectDojo, which approximates it without the proprietary threat intel. Community Feed CIS and STIG coverage is thin, and self-hosting means you own patching, backups and DR for GVM and DefectDojo — managed, not just installed.

Why this beats a flag day

Reversible in minutes, retired only after a long soak.

No phase forces an outage. Each per-asset-class cutover rolls back in under 15 minutes because the Nessus schedule is left disabled, not deleted — and a class only counts as migrated after at least 30 days of OpenVAS-primary scanning within plus-or-minus 15% of the Nessus baseline. Nessus is never cancelled outright: its internal scope retires after the soak, while the ASV external contract stays on indefinitely under PCI DSS v4.0 Req 11.3.2.

See whether your internal scope migrates cleanly.

A call with a senior vulnerability-management engineer. We map your Nessus policies and plugins to OpenVAS NVTs, size your PCI ASV and agent gaps honestly, and tell you exactly how much of Nessus you can retire — and how much you should keep.

Map my migration →