Like issuing a recall for horse-drawn carriages — the affected vehicles left the road decades ago
CVE-1999-0001 is a denial-of-service flaw in ip_input.c, the IP packet ingestion path of BSD-derived TCP/IP stacks. Sending crafted IP packets to an affected host can crash or hang the kernel. Affected versions are FreeBSD 3.0, OpenBSD 2.3 and 2.4, and BSDI BSD/OS 3.1 — all released in the late 1990s and long since end-of-lifed. The bug was patched in OpenBSD errata for 2.3 and subsequent FreeBSD/BSDI releases.
The vendor CVSS v2 score of 5.0 MEDIUM was reasonable *in 1999* — unauthenticated remote DoS with low complexity is a real concern on a production box. But in 2026, the severity is effectively zero. There is no supported operating system on earth still running these versions. Even niche embedded BSD appliances have long since moved past these codebases. The vendor score is a historical artifact, not an operational risk.
3 steps from start to impact.
Identify a host running FreeBSD 3.0, OpenBSD 2.3/2.4, or BSDI 3.1
- Target host runs an unpatched BSD OS from 1998–1999
- These OS versions are extinct in enterprise environments
- No vendor support, no hardware compatibility with modern servers
- Even retro-computing hobbyists typically run newer BSD releases
Send crafted IP packets to trigger ip_input.c crash
ip_input.c, causing a crash or hang.- Network reachability to the target host on any IP-bearing interface
- Any modern firewall or IPS would likely normalize or drop the malformed packets
- The specific packet construction details target a 1990s-era parsing bug that has been patched everywhere
Host crashes or hangs — DoS achieved
- Successful delivery of crafted packets to an unpatched host
- Impact ceiling is a single-host DoS — no persistence, no escalation
- Reboot restores service
The supporting signals.
| In-the-wild exploitation | None known. No campaigns, no threat actor usage documented in 27 years. |
|---|---|
| KEV status | Not listed. CISA KEV does not include this CVE. |
| Proof of concept | GitHub repos referencing this CVE exist but are CVE indexing projects (e.g., trickest/cve), not functional exploit code. No weaponized PoC is meaningful against modern systems. |
| EPSS | 0.03351 (≈3.4%) — low probability of exploitation, and even this is inflated given no viable target population exists. |
| CVSS v2 vector | AV:N/AC:L/Au:N/C:N/I:N/A:P — Network/Low-complexity/No-auth, but impact is Availability-Partial only. No confidentiality or integrity impact. |
| Affected versions | FreeBSD 3.0, OpenBSD 2.3, OpenBSD 2.4, BSDI BSD/OS 3.1. All EOL for 20+ years. |
| Fixed versions | OpenBSD 2.3 errata (#tcpfix), FreeBSD 3.1+, OpenBSD 2.5+, BSDI 4.0+. Every supported BSD release since ~2000 is patched. |
| Exposure data | Effectively zero. Shodan/Censys show no public-facing hosts running FreeBSD 3.0 or OpenBSD 2.3/2.4. These OS versions predate modern internet-facing infrastructure. |
| Disclosure date | 1999-12-30 — over 26 years ago. |
| Reporter | Original BSD security community; specific researcher not attributed in NVD/MITRE records. |
noisgate verdict.
The single most decisive factor is zero surviving target population: the affected OS versions (FreeBSD 3.0, OpenBSD 2.3/2.4, BSDI 3.1) have been end-of-life for over two decades and do not exist in any enterprise asset inventory. A vulnerability with no reachable targets carries no operational risk regardless of its theoretical exploitability.
Why this verdict
- Dead target population: FreeBSD 3.0, OpenBSD 2.3/2.4, and BSDI 3.1 are not present in any modern enterprise. Hardware that could run these kernels is itself largely decommissioned.
- DoS-only impact ceiling: Even if a target existed, the blast radius is a single-host availability loss with no persistence, no code execution, and no lateral movement. A reboot restores service.
- Role multiplier: There is no high-value-role analysis to perform because the affected component (late-1990s BSD kernels) does not occupy *any* role in modern infrastructure — not as a hypervisor, DC, CI/CD server, edge appliance, or anything else. The installed base in 2026 is effectively 0%.
- Patched for 26+ years: Every maintained BSD derivative has carried the fix since at least 2000. Even unmaintained forks inherited the patch long ago.
Why not higher?
There is no scenario in which this vulnerability warrants any severity rating above IGNORE. The affected software does not exist in production anywhere. Assigning even LOW would imply backlog hygiene work is needed, but there is nothing to patch — no host runs these versions.
Why not lower?
IGNORE is already the lowest possible rating. No further downgrade is possible.
What to do — in priority order.
- Confirm no affected BSD versions in your CMDB — Run a single asset-inventory query filtering for FreeBSD < 3.1, OpenBSD < 2.5, or BSDI < 4.0. If zero results (expected), document the result and close the ticket. This takes minutes, not days.
- Patching — there is nothing to patch in a modern environment; the affected OS versions are not present
- Network segmentation — irrelevant when no vulnerable host exists
- IPS signatures — no vendor maintains signatures for 1999-era BSD DoS vectors, nor should they
Crowdsourced verification payload.
Run on any Linux/macOS/BSD auditor workstation with SSH access to your fleet, or against your CMDB export. No special privileges needed. Example: bash check_cve_1999_0001.sh (reads from stdin or a host list file).
#!/usr/bin/env bash
# check_cve_1999_0001.sh — CVE-1999-0001 applicability check
# Checks whether any host in the environment runs an affected BSD version.
# Usage: echo 'hostlist.txt' | bash check_cve_1999_0001.sh
# or: bash check_cve_1999_0001.sh < hostlist.txt
# Output: VULNERABLE / PATCHED / UNKNOWN per host, summary at end.
# Requires: ssh access to target hosts (key-based recommended).
VULN_COUNT=0
SAFE_COUNT=0
UNKNOWN_COUNT=0
while IFS= read -r host; do
[ -z "$host" ] && continue
os_info=$(ssh -o ConnectTimeout=5 -o StrictHostKeyChecking=no "$host" 'uname -srm' 2>/dev/null)
if [ $? -ne 0 ]; then
echo "$host: UNKNOWN (unreachable)"
((UNKNOWN_COUNT++))
continue
fi
# Check for affected versions
if echo "$os_info" | grep -qiE '(FreeBSD 3\.0|OpenBSD 2\.[34]|BSD/OS 3\.1)'; then
echo "$host: VULNERABLE ($os_info)"
((VULN_COUNT++))
else
echo "$host: PATCHED ($os_info)"
((SAFE_COUNT++))
fi
done
echo ""
echo "=== Summary ==="
echo "VULNERABLE: $VULN_COUNT"
echo "PATCHED: $SAFE_COUNT"
echo "UNKNOWN: $UNKNOWN_COUNT"
if [ "$VULN_COUNT" -gt 0 ]; then
echo "RESULT: VULNERABLE"
exit 1
elif [ "$UNKNOWN_COUNT" -gt 0 ] && [ "$SAFE_COUNT" -eq 0 ]; then
echo "RESULT: UNKNOWN"
exit 2
else
echo "RESULT: PATCHED"
exit 0
fiIf you remember one thing.
Sources
What defenders are saying.
Crowdsourced verification outputs.
Results submitted by users who ran the verification payload against their environment.