Security Advisory

CVE-2021-38425: Fast DDS RTPS Discovery Amplification

Last updated:
  • CVE-2021-38425

CVE-2021-38425 is a network amplification flaw (CWE-406) in eProsima Fast DDS: one crafted RTPS discovery packet makes a Fast DDS participant flood a target with unwanted traffic. NVD scores it 9.1 and CISA 7.5. NVD lists versions before 2.4.0, but the fix it references, PR #2269, first shipped in Fast DDS 2.6.1.

Vulnerability summary

CVSS base score

9.1

Critical severity

EPSS score (FIRST)

5.0%

Probability of exploitation in the wild in the next 30 days

CISA KEV catalog

No

Not listed

Published

Vulnerability ID: CVE-2021-38425 · Source: NVD

Affected components

ComponentCPEVersion range
eProsima Fast DDScpe:2.3:a:eprosima:fast_dds:*:*:*:*:*:*:*:*NVD and CISA: before 2.4.0; the referenced PR #2269 was merged on 2022-04-01 and first released in 2.6.1 (2022-06-09)

What does this mean for affected devices?

NVD's description reads: "eProsima Fast DDS versions prior to 2.4.0 (#2269) are susceptible to exploitation when an attacker sends a specially crafted packet to flood a target device with unwanted traffic, which may result in a denial-of-service condition and information exposure." NVD's Primary score is 9.1, vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H; CISA's own score is 7.5 with availability impact only. CISA classes it CWE-406, network amplification (NVD).

The version boundary and the fix reference do not line up. PR #2269, "Authentication exponential backoff", was merged into the master branch on 2022-04-01, months after Fast DDS 2.4.0 was released on 2021-09-21. The first release whose notes list #2269 is Fast DDS 2.6.1, released on 2022-06-09; the notes for 2.4.1, 2.5.0 to 2.5.2 (the last of which came out on 2022-07-21) and 2.6.0 do not list it. A scanner that follows NVD's "before 2.4.0" CPE range will pass 2.4.0 to 2.6.0 builds that lack the change, so treat 2.6.1 or a later supported release as the floor.

CISA's ICS advisory ICSA-21-315-02, first published on 2021-11-11 and updated on 2022-02-01, covers 13 CVEs in six DDS implementations and marks critical infrastructure sectors as "Multiple" and deployment as "Worldwide". Its remediation for Fast DDS is short: "eProsima recommends users apply the latest Fast DDS patches." The researchers' write-up explains the class. RTPS discovery is "chatty", expects a reply from every contacted participant and runs over spoofable UDP, which makes it a reflection and amplification channel. In their test, Fast DDS reached a bandwidth amplification factor of 9.875 and kept sending for the whole 139-second run; they marked it only partially mitigated on the master branch and wrote that a full fix needs changes to the RTPS specification itself. The scope they list is "Fast-DDS, ROS 2".

CVE-2021-38425 is not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13). FIRST EPSS gives it 0.0502, the 92.4th percentile, in the 2026-11-15 dataset (CISA KEV, FIRST EPSS).

For a robot, controller or gateway that ships Fast DDS, read the exact version from the firmware SBOM, move to 2.6.1 or a later supported release, and keep the DDS domain off networks where an attacker can spoof source addresses. Our DDS security guide explains the RTPS discovery defaults that make this possible, and the input-handling fix guide covers resource-exhaustion weaknesses. The same disclosure produced the same flaw in RTI Connext (CVE-2021-38487) and CoreDX DDS (CVE-2021-43547).

How to fix this class of issueThis advisory's weakness class is Resource exhaustion. Our practical guide covers how to detect, fix, verify and disclose it: How to fix SQL injection, XSS, resource exhaustion and deserialization flaws in embedded firmware.

Evidence

  • Fast DDS prior to 2.4.0 (#2269); crafted packet floods target; DoS and information exposure; NVD Primary 9.1 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H); CISA 7.5; CISA CWE-406. — NVD
  • PR #2269 'Authentication exponential backoff' merged into master 2022-04-01; fixes issue #2267. — eProsima Fast-DDS PR #2269
  • Fast DDS 2.4.0 released 2021-09-21; its notes do not list #2269. — Fast DDS v2.4.0 release
  • Fast DDS 2.6.1 released 2022-06-09 lists #2269; 2.4.1, 2.5.0, 2.5.1, 2.5.2 and 2.6.0 release notes do not; merge commit 01550cf is contained in v2.6.1 and not in v2.6.0 (GitHub compare). — Fast DDS v2.6.1 release
  • Remediation: eProsima recommends users apply the latest Fast DDS patches; advisory published 2021-11-11, updated 2022-02-01. — CISA ICS Advisory ICSA-21-315-02
  • RTPS discovery is chatty and spoofable over UDP; Fast DDS BAF 9.875, 100% of 139 s attack duration, partially mitigated on master branch; full mitigation needs RTPS spec changes; scope Fast-DDS, ROS 2. — Trend Micro Research, 2022-04-19
  • Not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13). — CISA KEV
  • EPSS 0.05022 (92.4th percentile), dataset 2026-11-15, read 2026-11-16. — FIRST EPSS

Related advisories

Sources

  1. NVD
  2. CISA ICS Advisory ICSA-21-315-02: Multiple Data Distribution Service (DDS) Implementations
  3. CISA CSAF record for ICSA-21-315-02 (revision 2, 2022-02-01)
  4. eProsima Fast DDS PR #2269: Authentication exponential backoff
  5. eProsima Fast DDS v2.6.1 release notes
  6. eProsima Fast DDS v2.4.0 release notes
  7. Trend Micro Research: Critically Underrated: Studying the Data Distribution Service (DDS) Protocol
  8. CISA KEV
  9. FIRST EPSS

Frequently asked questions

Which Fast DDS version fixes CVE-2021-38425?

NVD and CISA say versions before 2.4.0 are affected, but the change NVD references, PR #2269, was merged on 2022-04-01 and first listed in the 2.6.1 release notes of 2022-06-09. Use 2.6.1 or a later supported release (per NVD and the Fast DDS release notes).

How large is the amplification in CVE-2021-38425?

Trend Micro's researchers measured a bandwidth amplification factor of 9.875 for Fast DDS, and the reflected traffic lasted the whole 139-second test (per Trend Micro Research).

Why do NVD and CISA score CVE-2021-38425 differently?

NVD rates confidentiality impact high as well as availability, giving 9.1; CISA rates availability only, giving 7.5. Both use a network vector with no privileges (per NVD).

Is CVE-2021-38425 exploited in the wild?

It is not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13). FIRST EPSS gives it 0.0502, the 92.4th percentile, in the 2026-11-15 dataset (per CISA KEV and FIRST EPSS).