tcp: prevent collapsing skbs across boundary in rtx queue
Assessment
- Severity
- Important
- CVSS base score
- 7.4CVSS:3.1/
AV:N/ AC:H/ PR:N/ UI:N/ S:U/ C:H/ I:H/ A:N - Impact
- Not stated
- EPSS, next 30 days
- Not scored yet
- Public exploit code
- No Nuclei template lists it
- Exploitability
- Not assessed
- Publicly disclosed
- No
- Customer action
- Required: apply the update
- Component
- Mariner
- Weakness
- —
- Issued by
- Linux
- Published
- 20 Oct 2026 · October 2026
- Last revised
- 21 Oct 2026Information published.
- Attack vector
- NetworkAV:N
- Attack complexity
- HighAC:H
- Privileges required
- NonePR:N
- User interaction
- NoneUI:N
- Scope
- UnchangedS:U
- Confidentiality
- HighC:H
- Integrity
- HighI:H
- Availability
- NoneA:N
Updates that fix it
0 KBsNo KB listed
MSRC lists no downloadable update, which usually means a service-side fix or a release-notes update. Check the MSRC advisory.
Affected products
1 affected product
- azl3 kernel 6.6.157.1-1 on Azure Linux 3.0
What changed
First observed 21 Oct 2026- 23 Oct 2026 · 09:33:49 UTC Public exploit codeUnknown: exploit-code evidence is incomplete or retained → No Nuclei template listed
- 23 Oct 2026 · 09:33:49 UTC CISA KEVUnknown: catalog evidence is incomplete or retained → Not listed in the collected catalog
- 23 Oct 2026 · 09:33:49 UTC Source coverage: public-exploit-codeIncomplete → Current
- 23 Oct 2026 · 09:33:49 UTC Source coverage: first-epssIncomplete → Current
- 23 Oct 2026 · 09:33:49 UTC Source coverage: cisa-kevIncomplete → Current
- 23 Oct 2026 · 09:33:49 UTC Source coverage: msrc-cvrfIncomplete → Current
- 23 Oct 2026 · 08:31:50 UTC Public exploit codeNo Nuclei template listed → Unknown: exploit-code evidence is incomplete or retained
- 23 Oct 2026 · 08:31:50 UTC CISA KEVNot listed in the collected catalog → Unknown: catalog evidence is incomplete or retained
- 23 Oct 2026 · 08:31:50 UTC Source coverage: public-exploit-codeCurrent → Incomplete
- 23 Oct 2026 · 08:31:50 UTC Source coverage: first-epssCurrent → Incomplete
- 23 Oct 2026 · 08:31:50 UTC Source coverage: cisa-kevCurrent → Incomplete
- 23 Oct 2026 · 08:31:50 UTC Source coverage: msrc-cvrfCurrent → Incomplete
Show 1 more recent changes
- 21 Oct 2026 · 20:45:25 UTCLast revised 20 Oct 2026 14:43 UTC → 21 Oct 2026 18:39 UTC
The most recent 100 field changes are listed. Complete recorded revisions are kept separately.
Cite this recorded revision to keep the same title, summary, facts and links when the page changes.
A day selects the latest observation by the end of that UTC day; today's selection can change until midnight. History starts when this installation records it.
Explore global source statements, affected packages and advisory history for CVE-2026-98549. The catalog has a separate collection scope from Microsoft release records.
Collected exploitation and prediction evidence
Risk source coverageCISA KEV is observed exploitation evidence. FIRST EPSS is a dated prediction of exploitation in the next 30 days. Missing scores are unknown. This collector requests predictions only for explicit CVE aliases in the public NuGet archive; Microsoft release views retain their existing prediction scope.
No collected risk observation for these identifiers. This does not establish absence of exploitation or a zero probability.
cve-program · CVE-2026-98549
Publishedtcp: prevent collapsing skbs across boundary in rtx queue
In the Linux kernel, the following vulnerability has been resolved: tcp: prevent collapsing skbs across boundary in rtx queue tcp_write_collapse_fence() sets TCP_SKB_CB(skb)->eor = 1 on tcp_write_queue_tail(sk) to prevent skbs queued after a switch to device encryption from being collapsed into earlier skbs. The fence is a no-op if all earlier data has already been transmitted when the switch happens: sk->sk_write_queue is empty. The not yet acknowledged earlier skbs wait in sk->tcp_rtx_queue with eor 0. On a subsequent retransmit or SACK shift, tcp_retrans_try_collapse() or tcp_shift_skb_data() can then merge an skb queued after the switch into one queued before it. Both users of the fence are affected: - psp: devices only encrypt skbs with skb->decrypted set. The merged skb keeps decrypted = 0 from the earlier skb, so merged data sent after psp_sock_assoc_set_tx() is retransmitted in cleartext. - tls device offload: the merged skb straddles the start marker set in tls_set_device_offload(). The software fallback (fill_sg_in() returns -EINVAL) and the mlx5, nfp and funeth drivers cannot handle such an skb and drop it. Every retransmit rebuilds the same skb, so the connection stalls. Fix this in two places, for defense in depth: 1. Fall back to tcp_rtx_queue_tail(sk) in tcp_write_collapse_fence() when tcp_write_queue_tail(sk) is NULL. 2. Check !skb_cmp_decrypted(to, from) in tcp_skb_can_collapse(), as tcp_skb_can_collapse_rx() does on receive. skb_shift(), which both collapse paths call, already has a DEBUG_NET_WARN_ON_ONCE() for this condition.
- Source published
- 18 Oct 2026 15:18:17 UTC
- Source modified
- 24 Oct 2026 14:09:22 UTC
- Observed here
- 24 Oct 2026 16:17:04 UTC
- Source terms
- CVE Program Terms of Use
Original source record · Cite this exact observation · Complete original publisher JSON
This view projects the publisher's source fields for reading; the original JSON retains additional fields and CVE enrichment containers. Attribution: cve-program · CVE Program Terms of Use. Source text is retained; field grouping and change labels are provided by ChangeIntel.
CVSS 3.1 · 9.1 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:H