SSL Version 2 and 3 Protocol Detection

Tenable plugin 20007 fires when a remote service will still negotiate SSLv2 and/or SSLv3. This is not a single-product bug with a neat fixed build; it is a protocol-hardening finding that can appear on web servers, mail gateways, load balancers, management planes, appliances, and anything else exposing TLS. In practice, the affected 'version range' is simply any service configuration that still permits SSLv2 or SSLv3, including cases where the application is current but the OS crypto policy, reverse proxy, or appliance listener still allows the legacy protocol.

Tenable's Critical / 9.8 label overshoots reality for most enterprises. The real exploit paths are usually man-in-the-middle downgrade abuse for SSLv3 or cross-protocol decryption conditions for SSLv2 such as RSA key exchange and sometimes shared private keys, not unauthenticated one-shot remote takeover. That's a real security problem and a valid remediation target, but for a patch queue managing 10,000 hosts it is usually medium priority crypto hygiene unless the endpoint is internet-facing, carries credentials, or fronts sensitive admin traffic.

Why this verdict

• Vendor score assumes worst-case impact, not real attack friction: Tenable's 9.8 treats deprecated protocol support like straight network compromise, but the common paths are POODLE-style downgrade/MITM or DROWN-style decryption conditions.
• Attacker position is the major brake: SSLv3 exploitation typically requires a network-on-path position plus a victim session, which implies the attacker is already controlling or intercepting part of the communication path.
• Exposure population is narrower than the headline suggests: many findings are on internal appliances, legacy middleware, or non-browser services, which means the attacker already has internal access or the traffic value is lower.
• Modern tooling blocks parts of the chain: browser hardening, HSTS, ECDHE/PFS adoption, SSLv2 removal from libraries, and better TLS defaults all reduce exploitability compared with 2014-2016 assumptions.
• Still not LOW: if this sits on a public login portal, admin console, or shared-key legacy farm, the confidentiality risk is real and the blast radius can justify elevating those specific assets above the default estate-wide score.

Why not higher?

There is no one-packet remote takeover here. The exploit path usually depends on downgradeability, traffic interception, or RSA/shared-key conditions that are absent in a large share of modern deployments. That combination is too much friction for a default HIGH or CRITICAL rating across an enterprise estate.

Why not lower?

This is not harmless compliance trivia. The protocols are formally deprecated by the IETF, exploitation techniques are well documented, and the finding often lands on exactly the systems that carry credentials and session cookies. On internet-facing auth or admin surfaces, it can quickly graduate from hygiene issue to material confidentiality exposure.

1. Disable SSLv2 and SSLv3 at the edge — Change the TLS policy on the actual termination point first, whether that is the load balancer, reverse proxy, WAF, VPN concentrator, or appliance listener. For a MEDIUM verdict there is no mitigation SLA; fold this into the next routine crypto-hardening change and complete full remediation within 365 days.
2. Prioritize public auth and admin surfaces — Don't burn cycles treating every internal hit the same. Clear internet-facing login portals, mail/web access gateways, and management consoles first because they are the small subset where the downgrade/decryption path has the best payoff; for the overall MEDIUM bucket there is no mitigation SLA, but those exposed assets should go in the earliest change window.
3. Break shared-key legacy islands — For SSLv2 specifically, shared certificates or private keys across legacy services can amplify DROWN-style risk. Reissue or segregate keys where needed as part of normal remediation; there is no noisgate mitigation SLA for MEDIUM, but finish the hard fix inside the 365-day remediation window.
4. Enforce modern cipher and protocol policy centrally — Use OS crypto policy, reverse-proxy templates, or appliance baselines so teams cannot quietly re-enable SSLv3 for 'compatibility'. This prevents reintroduction and is the cleanest way to burn down a broad plugin class within the 365-day remediation window.

Run this from an auditor workstation or scanner node that can reach the target service; it does not need to run on the endpoint itself. Invoke it as bash check_ssl_legacy.sh example.com 443; it requires nmap with the ssl-enum-ciphers NSE script and does not require root for standard TCP scanning.

#!/usr/bin/env bash
# check_ssl_legacy.sh
# Detect whether a remote service still negotiates SSLv2 or SSLv3.
# Usage: bash check_ssl_legacy.sh <host> <port>
# Exit codes:
#   0 = VULNERABLE
#   1 = PATCHED
#   2 = UNKNOWN / error

set -u

HOST="${1:-}"
PORT="${2:-}"

if [[ -z "$HOST" || -z "$PORT" ]]; then
  echo "UNKNOWN - usage: bash check_ssl_legacy.sh <host> <port>"
  exit 2
fi

if ! command -v nmap >/dev/null 2>&1; then
  echo "UNKNOWN - nmap is not installed"
  exit 2
fi

TMP_OUT="$(mktemp 2>/dev/null || echo /tmp/check_ssl_legacy.$$)"
trap 'rm -f "$TMP_OUT"' EXIT

# -Pn avoids host discovery issues on filtered networks.
# --script-timeout keeps the check from hanging indefinitely.
if ! nmap -Pn -p "$PORT" --script ssl-enum-ciphers --script-timeout 60s "$HOST" >"$TMP_OUT" 2>&1; then
  echo "UNKNOWN - nmap scan failed"
  cat "$TMP_OUT"
  exit 2
fi

OUTPUT="$(cat "$TMP_OUT")"

# If the script didn't return TLS details, we can't conclude much.
if ! grep -Eq 'ssl-enum-ciphers|TLSv|SSLv' "$TMP_OUT"; then
  echo "UNKNOWN - no SSL/TLS enumeration data returned"
  cat "$TMP_OUT"
  exit 2
fi

if grep -Eq 'SSLv2|SSLv3' "$TMP_OUT"; then
  echo "VULNERABLE - legacy protocol support detected on ${HOST}:${PORT}"
  grep -E 'SSLv2|SSLv3' "$TMP_OUT" | sed 's/^/  /'
  exit 0
fi

if grep -Eq 'TLSv1\.2|TLSv1\.3|TLSv1\.1|TLSv1\.0' "$TMP_OUT"; then
  echo "PATCHED - no SSLv2/SSLv3 support detected on ${HOST}:${PORT}"
  exit 1
fi

echo "UNKNOWN - scan completed but protocol state was inconclusive"
cat "$TMP_OUT"
exit 2

Sources

1. Tenable Nessus Plugin 20007
2. Tenable severity change discussion for plugin 20007
3. RFC 6176: Prohibiting SSL 2.0
4. RFC 7568: Deprecating SSL 3.0
5. OpenSSL user's guide to DROWN
6. DROWN technical paper (USENIX Security 2016)
7. FIRST CVSS v3.1 example: SSLv3 POODLE
8. DROWN project site