CVE-2026-26264 is an out-of-bounds read (CWE-125) in bacnet-stack, the open-source BACnet C library for embedded systems: a truncated WriteProperty request underflows a length check in wp_decode_service_request(), and the decoder reads past the buffer and crashes the BACnet service. NVD scores it 8.1. It is fixed in 1.4.3 and 1.5.0.
Vulnerability summary
CVSS base score
8.1
High severity
EPSS score (FIRST)
0.4%
Probability of exploitation in the wild in the next 30 days
CISA KEV catalog
No
Not listed
Published
Vulnerability ID: CVE-2026-26264 · Source: NVD
Affected components
| Component | CPE | Version range |
|---|---|---|
| bacnet-stack (WriteProperty service decoder, wp_decode_service_request) | cpe:2.3:a:bacnetstack:bacnet_stack:*:*:*:*:*:*:*:* | 1.4.0 to before 1.4.3, 1.4.3 rc1 and 1.5.0 rc1 to rc3 (NVD); fixed in 1.4.3rc2 and 1.5.0rc4 (NVD), releases 1.4.3 and 1.5.0 (GHSA-phjh-v45p-gmjj) |
| Embedded BACnet controllers and gateways whose firmware is built on bacnet-stack | — | per vendor firmware; affected when the bundled bacnet-stack predates the fix and handles WriteProperty (GHSA-phjh-v45p-gmjj) |
What does this mean for affected devices?
CVE-2026-26264 is in the WriteProperty decoder of bacnet-stack. NVD's description: "When decoding the optional priority context tag, the code passes apdu_len - apdu_size to bacnet_unsigned_context_decode without validating that apdu_size <= apdu_len. If a truncated APDU reaches this path, apdu_len - apdu_size underflows, resulting in a large size being used for decoding and an out-of-bounds read." The result is "a crash (DoS)". NVD's Primary CVSS v3.1 score is 8.1 (high), vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:H, with CWE-125 (NVD). GitHub, the CNA, scores it 7.8 in CVSS v4.0 with no privileges required and E:P, proof-of-concept code public.
What does the project advisory show?
A call path that any WriteProperty handler takes. GHSA-phjh-v45p-gmjj traces it as apdu_handler → handler_write_property → wp_decode_service_request, tested on bacnet-stack-1.4.2-92, and lists the impact as "process crash or instability" for "any BACnet service handling WriteProperty requests". WriteProperty is the service that changes setpoints and outputs, so it is enabled on almost every BACnet server. The fix, commit 4cc8067 from PR #1231, also secured bacnet_action_command_decode(), which had the same issue. CISA's SSVC entry in the CVE record, dated 2026-02-13, lists exploitation as "poc" and the attack as automatable.
Which versions and devices are affected?
NVD's CPE configuration covers 1.4.0 to before 1.4.3, 1.4.3 rc1, and 1.5.0 rc1 to rc3. NVD names 1.4.3rc2 and 1.5.0rc4 as fixed; the GitHub advisory and the project's SECURITY.md list the releases 1.4.3 and 1.5.0, and the 1.4.3 CHANGELOG, dated 2026-04-16, records the fix under Security.
No device vendor or CISA ICS advisory naming a product affected by this CVE was found (CISA's CSAF advisory set checked 2026-11-17). The library is built into controller and gateway firmware, so exposure depends on the version each maker bundled. SECURITY.md marks 1.5.x unsupported, and supports 1.4.x and 1.6.x and later.
How does it relate to other bacnet-stack bugs?
It is one of several decoder bugs fixed in 1.4.3. Our BACnet protocol security guide lists it with CVE-2026-41475, the WritePropertyMultiple over-read fixed in the same release, and the older CVE-2019-12480. A firmware SBOM records which stack version each product carries. For the weakness class, see our memory corruption fix guide.
CVE-2026-26264 is not in the CISA KEV catalog as of 2026-11-17 (catalog version 2026.11.13). FIRST EPSS gives it 0.00437, the 38th percentile, in the 2026-11-16 dataset (CISA KEV, FIRST EPSS).
To fix it, rebuild firmware on bacnet-stack 1.4.3 or later, or on the newest release of a supported branch. Device owners should ask the maker which stack version their firmware carries and keep BACnet/IP (commonly UDP 47808) reachable only from the building management network until patched firmware is installed.
Evidence
- wp_decode_service_request passes apdu_len - apdu_size without validation; truncated APDU underflows and causes an out-of-bounds read and crash; fixed in 1.5.0rc4 and 1.4.3rc2; NVD Primary CVSS v3.1 8.1 (PR:L); CWE-125; GitHub CVSS v4.0 7.8 with E:P; CISA SSVC 2026-02-13 exploitation poc, automatable yes. — NVD
- Call path apdu_handler → handler_write_property → wp_decode_service_request; tested bacnet-stack-1.4.2-92; impact process crash or instability for any service handling WriteProperty; resolution also secured bacnet_action_command_decode() (#1231); patched 1.5.0, 1.4.3. — GHSA-phjh-v45p-gmjj
- Fix commit 4cc8067 (PR #1231, merged 2026-02-13). — bacnet-stack GitHub
- 1.4.3 (2026-04-16) Security: secured decoding length underflow in wp_decode_service_request() and bacnet_action_command_decode() (#1231). — bacnet-stack CHANGELOG 1.4.3
- Not in the CISA KEV catalog as of 2026-11-17 (catalog version 2026.11.13). — CISA KEV
- EPSS 0.00437 (38th percentile), dataset 2026-11-16, read 2026-11-17. — FIRST EPSS

Related advisories
Sources
Frequently asked questions
What happens when CVE-2026-26264 is triggered?
The WriteProperty decoder reads out of bounds and the BACnet process crashes, a denial of service. The project advisory lists process crash or instability as the result (per NVD and GHSA-phjh-v45p-gmjj).
Why does NVD use PR:L when GitHub uses PR:N?
NVD's analysts scored low privileges required, giving 8.1 in CVSS v3.1; GitHub's CVSS v4.0 score of 7.8 assumes none. Plain BACnet/IP WriteProperty requests carry no credential (per NVD and our BACnet guide).
Which bacnet-stack release fixes CVE-2026-26264?
1.4.3 and 1.5.0, released 2026-04-16; NVD names release candidates 1.4.3rc2 and 1.5.0rc4 as the first fixed builds (per NVD, GHSA-phjh-v45p-gmjj and the bacnet-stack CHANGELOG).
Is CVE-2026-26264 in the CISA KEV catalog?
No, not as of 2026-11-17 (catalog version 2026.11.13). FIRST EPSS gives it 0.00437, the 38th percentile, in the 2026-11-16 dataset (per CISA KEV and FIRST EPSS).
Questions about this page?