I feel like there’s some room for improvement in the communications pipeline from the VyOS team when impactful CVEs are made public.
Obviously, patching a CVE is not as simple as just pushing a kernel update. Testing is required, and sometimes manual backports are necessary, particularly with LTS running older modified kernels. This means patched releases can lag behind CVE announcements, which is understandable. However, communication and remediation don’t necessarily need to follow the same timeline.
We saw some community frustration surrounding the recent CVE-2026-98432, the IPv4/TCP use-after-free vulnerability. The vulnerability was published on October 15, an advisory email was sent by VyOS on November 4 with mitigation steps, and patched releases followed on November 9. Mitigation steps were known well ahead of the November 4 advisory.
I think the VyOS blog would be a good place for these communications, ideally alongside email notifications.
Now we have NoisyTCP (CVE-2026-99288), which potentially affects fairly common VyOS deployments with TCP services exposed to the internet. I’ve seen numerous posts over the years from users running BGP sessions with their providers without firewall filtering. Some users also expose SSH directly to the internet.
The published mitigations for NoisyTCP aren’t necessarily attractive for a router. Disabling SYN cookies introduces additional exposure to SYN floods, while reducing RX processing to a single queue can negatively impact throughput.
This is where VyOS-specific security guidance could be particularly valuable. General Linux CVE advisories don’t necessarily account for how a system is deployed, and the published mitigations for NoisyTCP don’t discuss firewall filtering. For a router running VyOS, restricting access to exposed TCP services through firewall rules may be a more practical approach to reducing exposure without sacrificing performance while users await a patched release.
This is exactly the kind of guidance I’d like to see from the VyOS team. Even a brief advisory acknowledging the vulnerability, identifying affected releases, and recommending mitigations appropriate for VyOS deployments would help administrators make informed decisions while waiting for a patched release.
Rolling generally receives timely kernel updates when vulnerabilities are addressed upstream, but that only benefits users who update frequently or are already aware of a vulnerability and know to update. Users running LTS may have to wait longer for patched releases. In either case, timely communication from the VyOS team is valuable, as administrators shouldn’t have to independently monitor Linux kernel CVEs to determine whether their routers are affected.
Ideally, a security advisory would communicate three things:
- Whether the vulnerability affects VyOS and which releases are impacted.
- Any recommended temporary mitigations, preferably using native VyOS configuration where possible.
- When a patched release becomes available.
With AI-assisted vulnerability discovery accelerating, having a predictable communication process is only going to become more important.