Cloudanix VulnDB

CVE-2026-98484

None

CVSS

—

EPSS

0.2%

Published

Oct 18, 2026

CISA due date

—

Cloudanix analysis

Next patch cycle SSVC: Track

A bug in Linux's built-in firewall can cause certain rules that are supposed to expire to keep interfering with new rules, potentially letting traffic slip through or be blocked longer than intended.

Exploitation
Local Privilege
Attacker needs
Low Privilege
Complexity
Moderate
Impact
Other
Component
OS kernel
Class
Race Condition
Remediation
Upgrade
Effort
Medium
Patch risk
Low

What it actually does

In nf_tables catchall set elements, insert and delete paths use only nft_set_elem_active() while lookup and dump paths also skip expired elements, allowing a timed-out catchall element to block re-insertion (EEXIST or silent no-op) and to be incorrectly selected for deletion until the GC interval runs.

Why it matters to the business

Firewall or packet-filtering rules that rely on catchall elements with timeouts can silently stop functioning for an arbitrary period after expiration, leading to unexpected traffic handling or policy bypass until the next garbage-collection cycle.

How an attack unfolds

An attacker creates a catchall set element with a short timeout, waits for it to expire, then attempts to re-insert a replacement element; the stale entry still blocks the operation, causing either an error or a no-op that leaves the catchall rule inactive until GC removes it.

How to fix it

Upgrade to a Linux kernel containing the referenced fix that aligns catchall insert/delete walks with the active-and-not-expired predicate used by lookup and dump paths; apply the update as soon as a distribution package is available.

How to confirm the fix worked

After patching, confirm the kernel version contains the nf_tables catchall predicate alignment change, then exercise insert/delete operations on timed-out catchall elements and verify that expired entries no longer block new insertions or cause erroneous deletions.

Typical attack path

  1. Create nftables set with catchall element that has a timeout
  2. Allow the element to expire while remaining in the set list
  3. Issue NLM_F_EXCL or non-exclusive insert for a replacement element
  4. Observe EEXIST failure or silent insertion of nothing
  5. Trigger traffic that should match the now-inactive catchall rule

How to detect it

Examine system logs for repeated nftables netlink errors (EEXIST on insert or ENOENT on delete of catchall elements) around the expected expiration time of set elements; correlate with nft monitor or audit records showing set modification attempts that produce no visible change in dumps.

What your security team should do

  • Prioritize kernel updates for systems that use nftables sets with timeout-bearing catchall elements
  • Monitor upstream distribution advisories for the specific backported fix
  • Review nftables configurations that depend on catchall rules and consider shortening GC intervals where feasible until patched

The record

In the Linux kernel, the following vulnerability has been resolved: netfilter: nf_tables: skip expired catchall elements on insert and delete nft_setelem_catchall_insert() looks up duplicates with nft_set_elem_active() only, while nft_set_catchall_lookup() and the dump path additionally skip expired elements. Once a catchall element with a timeout expires, this predicate drift makes it invisible to userspace dumps, yet it still blocks re-insertion: with NLM_F_EXCL the request fails with -EEXIST, and without it the request reports success but silently inserts nothing. The stale entry only goes away when the (user-tunable) gc interval elapses, so the catchall rule may silently stop matching for an arbitrarily long time after its first expiration. The delete path shows the same drift: nft_setelem_catchall_deactivate() picks the first active-next entry in the catchall list, so with an expired entry still pending GC it retires the stale entry instead of the fresh one, and it deactivates an element that userspace no longer sees instead of failing with -ENOENT. Align both walks with the lookup and dump predicates: only an element that is active and not expired counts as a duplicate or delete candidate, using the per-netns timestamp taken at transaction start, in line with the set backend .insert/.deactivate and catchall GC sync paths.

Affected packages

Ecosystem Package Introduced Fixed