FKIE_CVE-2026-95690
Vulnerability from fkie_nvd - Published: 2026-11-18 12:17 - Updated: 2026-11-20 12:18
Severity
Summary
Issue summary: Receiving crafted out-of-sequence DTLS handshake fragments may
cause OpenSSL to allocate a full message-length reassembly buffer for each
buffered message, so a small amount of wire data consumes a large amount of
memory.
Impact summary: A remote attacker can send a few small crafted DTLS handshake
fragments to make a DTLS server or client allocate well over a megabyte of
memory per connection, which may lead to memory exhaustion and Denial of
Service if many connections are made.
CWE: CWE-770: Allocation of Resources Without Limits or Throttling
Description: DTLS handshake messages can arrive out of order and fragmented,
so OpenSSL buffers fragments of up to ten out-of-sequence messages until
reassembly is possible. When the first fragment of a new message arrives,
OpenSSL allocates a buffer sized to the full message length declared in the
fragment header, plus a reassembly bitmap, instead of storing only the bytes
actually received. The declared length is capped by the maximum certificate
list size, 128KB by default, and that cap applies to every handshake message
type rather than only to Certificate messages.
Only DTLS endpoints reach this code. TLS, QUIC and applications that do not
use DTLS are unaffected. An attacker sends fragments that declare a large
message length but carry almost no data, using up to ten distinct message
sequence numbers.
A single connection can then hold roughly 1.4MB of buffers from about 156
bytes on the wire. Because the number of buffered messages is capped, memory
exhaustion requires many connections or many peers, which is why the severity
is Low.
FIPS impact: no
The affected code is outside the FIPS module boundary.
References
Impacted products
| Vendor | Product | Version |
|---|
{
"affected": [
{
"affectedData": [
{
"defaultStatus": "unaffected",
"product": "OpenSSL",
"vendor": "OpenSSL",
"versions": [
{
"lessThan": "4.1.1",
"status": "affected",
"version": "4.1.0",
"versionType": "semver"
},
{
"lessThan": "4.0.4",
"status": "affected",
"version": "4.0.0",
"versionType": "semver"
},
{
"lessThan": "3.5.10",
"status": "affected",
"version": "3.5.0",
"versionType": "semver"
},
{
"lessThan": "3.0.24",
"status": "affected",
"version": "3.0.0",
"versionType": "semver"
},
{
"lessThan": "1.1.1zk",
"status": "affected",
"version": "1.1.1",
"versionType": "custom"
},
{
"lessThan": "1.0.2zt",
"status": "affected",
"version": "1.0.2",
"versionType": "custom"
}
]
}
],
"source": "openssl-security@openssl.org"
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "Issue summary: Receiving crafted out-of-sequence DTLS handshake fragments may\ncause OpenSSL to allocate a full message-length reassembly buffer for each\nbuffered message, so a small amount of wire data consumes a large amount of\nmemory.\n\nImpact summary: A remote attacker can send a few small crafted DTLS handshake\nfragments to make a DTLS server or client allocate well over a megabyte of\nmemory per connection, which may lead to memory exhaustion and Denial of\nService if many connections are made.\n\nCWE: CWE-770: Allocation of Resources Without Limits or Throttling\n\nDescription: DTLS handshake messages can arrive out of order and fragmented,\nso OpenSSL buffers fragments of up to ten out-of-sequence messages until\nreassembly is possible. When the first fragment of a new message arrives,\nOpenSSL allocates a buffer sized to the full message length declared in the\nfragment header, plus a reassembly bitmap, instead of storing only the bytes\nactually received. The declared length is capped by the maximum certificate\nlist size, 128KB by default, and that cap applies to every handshake message\ntype rather than only to Certificate messages.\n\nOnly DTLS endpoints reach this code. TLS, QUIC and applications that do not\nuse DTLS are unaffected. An attacker sends fragments that declare a large\nmessage length but carry almost no data, using up to ten distinct message\nsequence numbers.\n\nA single connection can then hold roughly 1.4MB of buffers from about 156\nbytes on the wire. Because the number of buffered messages is capped, memory\nexhaustion requires many connections or many peers, which is why the severity\nis Low.\n\nFIPS impact: no\nThe affected code is outside the FIPS module boundary."
}
],
"id": "CVE-2026-95690",
"lastModified": "2026-11-20T12:18:25.127",
"metrics": {
"cvssMetricV31": [
{
"cvssData": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "LOW",
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L",
"version": "3.1"
},
"exploitabilityScore": 3.9,
"impactScore": 1.4,
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"type": "Secondary"
}
],
"ssvcV203": [
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"id": "CVE-2026-95690",
"options": [
{
"exploitation": "none"
},
{
"automatable": "yes"
},
{
"technicalImpact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-11-18T13:50:26.611679Z",
"version": "2.0.3"
}
}
]
},
"published": "2026-11-18T12:17:24.037",
"references": [
{
"source": "openssl-security@openssl.org",
"url": "https://github.com/openssl/openssl/pull/33620"
},
{
"source": "openssl-security@openssl.org",
"url": "https://github.com/openssl/openssl/pull/33729"
}
],
"sourceIdentifier": "openssl-security@openssl.org",
"vulnStatus": "Awaiting Analysis",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "CWE-770"
}
],
"source": "openssl-security@openssl.org",
"type": "Secondary"
},
{
"description": [
{
"lang": "en",
"value": "CWE-400"
}
],
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"type": "Secondary"
}
]
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Loading…
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Loading…
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.
Loading…