Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
Tools/GitHubGitHub/soroushneshati/cve-2026-98432-ubuntu-mitigation
Defensive ToolsVulnerability AnalysisScripting & AutomationConfiguration AuditingNetwork SecurityPenetration TestingMisconfigurationIncident ResponseLog Analysis
Share
GitHubsoroushneshati/cve-2026-98432-ubuntu-mitigation

cve-2026-98432-ubuntu-mitigation

Check an Ubuntu host for CVE-2026-98432 and, after you confirm, apply the interim Red Hat RHSB-2026-016 mitigation. Ubuntu 20.04 to 26.10.

View Repository
717 days agoNot yet reviewed

CVE-2026-98432 mitigation for Ubuntu

License: MIT Ubuntu shellcheck tests

A script for Ubuntu servers that estimates the running kernel's patch status for CVE-2026-98432 from package metadata and vendor data and, after you confirm, applies the interim mitigation from Red Hat bulletin RHSB-2026-016. It can also count matching packets without dropping them, so you can measure before you enforce.

This is not a fix, and it does not guarantee protection. The input filtering rule is based on Red Hat's bulletin; forwarding coverage and operational features are additions by this project. This is not an official Red Hat or Canonical tool. It drops matching IPv4 TCP packets with IP options on the configured hooks to reduce exposure. Red Hat says it "cannot fully prevent" the issue, and only a fixed vendor kernel solves it. Install that kernel and reboot as soon as it is released, then remove this.

The vulnerability

CVE-2026-98432 is a use-after-free in the Linux kernel's IPv4 TCP handshake. tcp_v4_syn_recv_sock() hands the IP options saved from a SYN to the new socket without copying them, so another CPU that handles a retransmitted SYN can read that memory after it was freed. A similar race exists in request socket migration (net.ipv4.tcp_migrate_req=1, or a BPF program). The upstream fix defers the free until an RCU grace period has passed.

AffectedLinux 4.4 and later, until the fix
ScoreCVSS 9.8 from the kernel CNA, Red Hat (preliminary) and Ubuntu's record; 8.1 from SUSE, Oracle and Amazon. Ubuntu's priority is High
Upstream fixe3f33b0a1d89 in 7.3-rc7; stable 5.10.272, 5.15.223, 6.1.190, 6.6.159, 6.12.113, 6.18.56, 7.2.10
NeedsA reachable TCP listener over IPv4, a SYN that carries IP options, and winning a race
IPv6Not affected by the main issue

What is known about exploitation, as of 30 October 2026. Anthropic reports a remote root exploit in its own testing against a server with a Multipath TCP (MPTCP) listener, and has not published the code. The Go security team notes that Go 1.24 and later open MPTCP listeners by default, and that the bug is not specific to MPTCP, so servers without such a listener may be vulnerable too. The CVE is not in the CISA KEV catalog (2026-10-30 edition).

Ubuntu status, from the Ubuntu tracker on 30 October 2026. Nothing is marked as released yet.

Releaselinux package
26.10Pending (7.3.0-10.10)
26.04 LTSPending (7.0.0-43.43)
24.04 LTSPending (6.8.0-149.149)
22.04 LTSPending (5.15.0-203.213; HWE 6.8.0-149.149~22.04.1)
20.04 LTSNeeded, no version listed

"Pending" means a fix is on its way and is not released. Check the tracker for the current state.

Quick start

Read a script before you run it as root.

curl -fsSLO https://raw.githubusercontent.com/soroushneshati/cve-2026-98432-ubuntu-mitigation/main/cve-2026-98432-mitigation.sh
less cve-2026-98432-mitigation.sh
sudo bash cve-2026-98432-mitigation.sh check

That only reads. This release is ready for lab validation, not yet for production deployment. On a non-production host, measure first, then enforce:

sudo bash cve-2026-98432-mitigation.sh --monitor-only
sudo bash cve-2026-98432-mitigation.sh

Each run shows the assessment and a plan, asks, applies, verifies the loaded rule and runs a self-test. Add --dry-run to see the plan and the rules without changing anything. docs/EXAMPLES.md has real output for each state.

Patch status and mitigation status are separate

The tool answers these questions independently, and one never stands in for another.

QuestionField in check and JSONValues
Does the running kernel have the fix?Patch statusPATCHED, VULNERABLE, UNKNOWN, NOT_AFFECTED
Is the rule enforcing in the kernel right now?Mitigation enforcementACTIVE_VERIFIED, PARTIAL, MONITORING_ONLY, MISCONFIGURED, MISSING, UNKNOWN
Will it survive a reboot and a reload?PersistenceOK, PARTIAL, INTERRUPTED, NOT_INSTALLED
Did the probes behave as expected?Runtime verificationPASS, FAIL, INCONCLUSIVE, SKIPPED, NOT_RUN, STALE
Is the status reporting working?ReportingOK, STALE, TIMER_NOT_ACTIVE, NO_REPORT_YET, NOT_INSTALLED

The last line combines them, and sets the exit code of the default run, check and status:

ResultExitMeaning
PATCHED, NOT_AFFECTED0Package metadata or vendor data indicates a fix or no impact. Confirm the running kernel and exact flavour before removing protection
MITIGATED1Not known to be fixed, and the rule is verified in the kernel. Lower exposure, not a fix
VULNERABLE2Vendor data says the running kernel is unfixed, and the rule is not enforcing
UNKNOWN3Not enough evidence or conflicting evidence, and the rule is not enforcing. Treat as vulnerable
4The tool itself failed, or refused to change something

With automation, count 0 and 1 as success. MITIGATED needs ACTIVE_VERIFIED: an empty table, a changed rule, a wrong priority, a counting-only rule or a host you cannot inspect never gives it. Monitor-only mode is not a security control, so it reports the kernel's own result.

How the patch status is decided

Only the running kernel counts, and a missing CVE mention is never evidence. The order is:

  1. The changelog of the running kernel's package (linux-modules-<release>) names the CVE or the upstream commit subject: PATCHED. A revert of that fix gives UNKNOWN.
  2. Vendor data for the kernel's exact source package, read with dpkg-query, so -aws, -azure, HWE and the others are compared with their own versions. The script embeds a dated snapshot of the Ubuntu tracker (116 rows on 2026-10-30). A released version that is met gives PATCHED; one that is not met, or a pending version above the running one, gives VULNERABLE.
  3. Otherwise UNKNOWN, with a reason code.

Upstream version numbers are not compared with vendor kernels, because vendors backport. For a kernel no package explains, they only appear as a hint, and the result stays UNKNOWN. Livepatch is reported separately (Livepatch:) and never changes the status, because an active livepatch does not say that it covers this CVE.

The snapshot ages. After 14 days the tool adds VENDOR_DATA_STALE, and a "needed" row without a version becomes UNKNOWN. Refresh it without a new release of the tool:

curl -fsSL https://ubuntu.com/security/cves/CVE-2026-98432.json -o tracker.json
python3 tools/update_vendor_data.py tracker.json --output vendor.txt
sudo bash cve-2026-98432-mitigation.sh check --vendor-data vendor.txt

Why UNKNOWN happens, and what to do

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    696
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    499
  • 3

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    396
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    340
  • 5

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    258
  • 6

    antifox

    antifox-ai/antifox

    AI multi-agent autonomous penetration testing platform that runs recon, attack-surface discovery, exploitation validation and evidence-backed report delivery.

    257

Explore all tools

Browse our collection of tools

View all tools →