Security Advisory

CVE-2023-38469: Avahi Reachable Assertion (Local DoS)

Last updated:
  • CVE-2023-38469

CVE-2023-38469 is a reachable assertion (CWE-617) in avahi_dns_packet_append_record: a local user who publishes a service with an oversized TXT record over D-Bus makes avahi-daemon before 0.9 abort, and service discovery stops. NVD scores it 5.5; commit a337a1b in v0.9-rc1 fixes it, and disable-user-service-publishing=yes mitigates it.

Vulnerability summary

CVSS base score

5.5

Medium severity

EPSS score (FIRST)

0.3%

Probability of exploitation in the wild in the next 30 days

CISA KEV catalog

No

Not listed

Published

Vulnerability ID: CVE-2023-38469 · Source: NVD

Affected components

ComponentCPEVersion range
Avahi (avahi-daemon)cpe:2.3:a:avahi:avahi:*:*:*:*:*:*:*:*before 0.9 (fixed upstream in v0.9-rc1, commit a337a1b; backported in Buildroot avahi 0.8 patch 0004 and Debian 0.8-5+deb11u3)

What does this mean for affected devices?

CVE-2023-38469 is one of five reachable assertions in Avahi published together in November 2023 (CVE-2023-38469 to CVE-2023-38473). NVD describes it as "a reachable assertion" in avahi_dns_packet_append_record, classifies it as CWE-617 and scores it CVSS v3.1 5.5 (AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H); Red Hat, the CNA, scores it 6.2 with no privileges required. NVD's configuration covers Avahi before 0.9.

The upstream report, Avahi issue 455, shows how it is triggered: an unprivileged local user runs avahi-publish with a TXT record far larger than a DNS packet can hold, and the daemon aborts on the assertion size <= AVAHI_DNS_RDATA_MAX. The report notes that this works "unless disable-user-service-publishing is set to yes explicitly". The fix, commit a337a1b of 2023-10-25 ("core: reject overly long TXT resource records"), is part of the v0.9-rc1 tag.

For device makers the important source is the build system. Buildroot's avahi package builds Avahi 0.8 and applies the fix as 0004-core-reject-overly-long-txt-resource-records.patch, followed by patches 0005 to 0010 for CVE-2023-38470 to CVE-2023-38473, and lists all five CVEs as ignored for its CVE scan. An image built before those patches were added reports the same Avahi 0.8 and is affected. Our guide to Yocto and Buildroot SBOMs shows where the patch list is recorded, and firmware diffing and patch analysis shows how to check the binary.

The Debian security tracker lists bullseye fixed in 0.8-5+deb11u3 (DLA-3990-1) and bookworm in 0.8-10+deb12u1, and records for buster: "Minor issue; can be mitigated by setting disable-user-service-publishing to yes".

That mitigation is the practical answer for firmware that cannot be rebuilt quickly. Most devices publish their services from avahi-daemon's own service files and never need local applications to publish over D-Bus, so setting disable-user-service-publishing=yes in avahi-daemon.conf closes the path. On a printer, camera, NAS or gateway the impact is a crashed daemon and the device dropping out of discovery until a restart. As of 2026-11-16 the CVE is not in the CISA KEV catalog (version 2026.11.13), and FIRST EPSS gives it 0.0032 (25th percentile) in its 2026-11-15 dataset. Our Avahi and mDNS/DNS-SD vulnerabilities guide covers the full hardening configuration; the related local flaw CVE-2021-3468 and the network-reachable CVE-2026-24401 need the same patch-list check.

How to fix this class of issueNo dedicated SoftWSP guide covers this weakness class; the fix is usually the vendor patch or firmware update referenced above. Our related guide explains how to plan and verify it: NIST SP 800-40 Rev. 4: enterprise patch management planning, explained.

Evidence

  • Reachable assertion in avahi_dns_packet_append_record, CWE-617; NVD CVSS v3.1 5.5 (Red Hat CNA 6.2); Avahi before 0.9. — NVD
  • Triggered by avahi-publish with an oversized TXT record unless disable-user-service-publishing is set to yes. — Avahi issue 455
  • Buildroot builds Avahi 0.8 with patch 0004 (upstream a337a1b) and patches 0005 to 0010 for CVE-2023-38470 to 38473. — Buildroot package/avahi/avahi.mk
  • Debian: bullseye 0.8-5+deb11u3 (DLA-3990-1), bookworm 0.8-10+deb12u1; fix commit a337a1b (v0.9-rc1); mitigation disable-user-service-publishing=yes. — Debian Security Tracker
  • Not in the CISA KEV catalog version 2026.11.13 (checked 2026-11-16). — CISA KEV
  • EPSS 0.0032, 25th percentile, FIRST dataset of 2026-11-15. — FIRST EPSS

Related advisories

Sources

  1. NVD: CVE-2023-38469
  2. Avahi issue 455 (reachable assertion in avahi_dns_packet_append_record)
  3. Avahi commit a337a1b (fix)
  4. Buildroot: package/avahi/avahi.mk
  5. Debian Security Tracker: CVE-2023-38469
  6. CISA KEV catalog
  7. FIRST EPSS

Frequently asked questions

Who can trigger CVE-2023-38469?

A local user or process that can publish a service through avahi-daemon's D-Bus interface, for example with avahi-publish. It is not reachable from the network. The upstream report says it works unless disable-user-service-publishing is set to yes (per Avahi issue 455 and NVD).

What is the quickest mitigation on a device?

Set disable-user-service-publishing=yes in avahi-daemon.conf if no local application needs to publish services over D-Bus. Debian records this as the mitigation for releases it did not patch (per the Debian security tracker).

Which fix should a Buildroot or Yocto image carry?

Upstream commit a337a1b, part of v0.9-rc1. Buildroot ships it as patch 0004 on Avahi 0.8, together with patches for CVE-2023-38470 to CVE-2023-38473, which share the same root cause class (per Buildroot avahi.mk).

How severe is CVE-2023-38469?

NVD scores it CVSS v3.1 5.5 (medium), local with availability impact only; Red Hat as CNA scores 6.2. It is not in the CISA KEV catalog (version 2026.11.13), and its EPSS is 0.0032 in the 2026-11-15 dataset (per NVD, CISA KEV and FIRST EPSS).