CVE-2026-54874

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires. Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service. CWE: CWE-405: Asymmetric Resource Consumption (Amplification) Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up. Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network. An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity. FIPS impact: no No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary. OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue. OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22. Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell. -- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: In a server or client configuration with RFC7250 Raw Public Keys (RPKs) enabled, and only the private key (with no associated certificate) configured locally, a NULL pointer dereference may occur when the remote peer solicits raw public keys and also sends the typically omitted "signature_algorithms_cert" TLS extension.

Impact summary: The impact is limited to a possible Denial of Service as a result of an application abort, no data disclosure or remote command execution are possible.

CWE: CWE-476: NULL Pointer Dereference

Description: While a passing comment in sample code in the documentation suggests that key-only RPK configurations are supported, the best-practice RPK configuration is to always configure a corresponding certificate (possibly self-signed or signed by any convenient CA).

When the private key is configured along with a matching certificate, the "signature_algorithms_cert" extension is handled reliably even without the fix, and peer clients or servers that don't support raw public keys may be able to complete a TLS connection by pinning or verifying the corresponding certificate or its public key.

Deployments that prefer to configure just a private key with no certificate need to upgrade to an updated release as noted below.

FIPS impact: no

No FIPS modules are affected by this issue, as the SSL protocol implementation is outside the OpenSSL FIPS module boundary.(CVE-2026-14457)

Issue summary: QUIC server may double free QRX (QUIC record layer RX) object when channel creation fails for initial packet.

Impact summary: Double free leads to heap corruption, which typically results in termination of QUIC server process, leading to Denial of Service. There is so far no evidence that this double free is exploitable for remote code execution, thus it is considered highly improbable.

CWE: CWE-415: Double Free

Description: In order to validate initial packet, OpenSSL QUIC stack default packet handler (port_default_packet_handler()) creates a so-called QRX object. If the initial packet validates successfully with QRX object, the default packet handler proceeds to channel (connection object) creation. The QRX object used for packet validation is passed to port_bind_channel(), so it becomes part of the newly created connection. If port_bind_channel() fails, then it also frees the QRX object. Once port_bind_channel() returns, the port_default_packet_handler() detects the failure and proceeds to the error branch, where the same QRX object is freed for the second time.

The failure in port_bind_channel() function can be induced with a relatively low effort by a malformed (non RFC 9000 compliant) INITIAL packet. If the packet carries DCID (destination connection ID) which is shorter than 8 bytes, then port_bind_channel() jumps to the error path after ossl_quic_lcidm_enrol_odcid() detects that the DCID has invalid length.

FIPS impact: no The FIPS module is not affected, as the QUIC implementation is outside of the OpenSSL FIPS module boundary.(CVE-2026-18798)

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early…

OpenSSL is a robust, commercial-grade, and full-featured toolkit for the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols.

Security Fix(es):

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell(CVE-2026-54874)

Issue summary: OpenSSL CMS decryption sizes the key-unwrap output buffer based on querying the unwrapped key size, but the AES-WRAP-PAD unwrap primitive can write and cleanse more bytes than that query reports, causing an 8-byte out-of-bounds heap write.

Impact summary: An attacker who supplies a crafted CMS message can trigger a deterministic 8-byte out-of-bounds heap write when the victim decrypts it with CMS_decrypt(), corrupting the heap and typically resulting in a Denial of Service.

CWE: CWE-787: Out-of-bounds Write

Description: The key-wrap OID is potentially attacker-controlled on the wire. CMS unwrapping allows both id-aesNNN-wrap-pad and id-aesNNN-wrap ciphers. An attacker can take a legitimate message and change a single OID byte to select the padded variant while leaving the message otherwise valid. Since the unwrap key is derived from the recipient's private operation (ECDH key agreement or ML-KEM decapsulation), the RFC 5649 integrity check cannot pass, and the decryption fails with integrity failure.

The write is a fixed-size (8-byte), fixed-value (zero) heap overflow immediately past the allocation, requires no special configuration,…

OpenSSL is a toolkit that implements the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols, as well as a full-strength general-purpose cryptography library.

Security Fix(es):

• openssl: OpenSSL: Denial of Service via unbounded memory growth in QUIC server (CVE-2026-14456)

• openssl: QUIC server may trigger double free when processing INITIAL packet (CVE-2026-18798)

• openssl: heap buffer overflow in CMS key unwrapping (CVE-2026-63072)

• openssl: invalid pointer dereference in CMP server via crafted protectionAlg (CVE-2026-63076)

• openssl: RPK server signature algorithm selection can dereference a missing certificate (CVE-2026-14457)

• openssl: excessive memory use buffering DTLS records for a future epoch (CVE-2026-54874)

• openssl: untrusted sender DN used as format string in CMP response validation (CVE-2026-63073)

• openssl: CMP indefinite cache growth of ExtraCerts (CVE-2026-63074)

• openssl: QUIC ACK-only packet retention can cause memory exhaustion (CVE-2026-63075)

Bug Fix(es) and Enhancement(s):

• openssl: HollowByte remote memory-exhaustion DoS fix may be missing [Rocky Linux 10.2.z] (JIRA:Rocky Linux-212362)

For more details about the security issue(s), including the impact, a CVSS score, acknowledgments, and other related information, refer to the CVE page(s) listed in the References section.

OpenSSL is a toolkit that implements the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols, as well as a full-strength general-purpose cryptography library.

Security Fix(es):

• openssl: OpenSSL: Denial of Service via unbounded memory growth in QUIC server (CVE-2026-14456)

• openssl: QUIC server may trigger double free when processing INITIAL packet (CVE-2026-18798)

• openssl: heap buffer overflow in CMS key unwrapping (CVE-2026-63072)

• openssl: invalid pointer dereference in CMP server via crafted protectionAlg (CVE-2026-63076)

• openssl: RPK server signature algorithm selection can dereference a missing certificate (CVE-2026-14457)

• openssl: excessive memory use buffering DTLS records for a future epoch (CVE-2026-54874)

• openssl: untrusted sender DN used as format string in CMP response validation (CVE-2026-63073)

• openssl: CMP indefinite cache growth of ExtraCerts (CVE-2026-63074)

• openssl: QUIC ACK-only packet retention can cause memory exhaustion (CVE-2026-63075)

Bug Fix(es) and Enhancement(s):

• openssl: HollowByte remote memory-exhaustion DoS fix may be missing [Rocky Linux 9.8.z] (JIRA:Rocky Linux-232828)

For more details about the security issue(s), including the impact, a CVSS score, acknowledgments, and other related information, refer to the CVE page(s) listed in the References section.

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires. Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service. CWE: CWE-405: Asymmetric Resource Consumption (Amplification) Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up. Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network. An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity. FIPS impact: no No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary. OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue. OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22. Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell. -- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell

OpenSSL is a toolkit that implements the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols, as well as a full-strength general-purpose cryptography library.

Security Fix(es):

• openssl: OpenSSL: Denial of Service via unbounded memory growth in QUIC server (CVE-2026-14456)
• openssl: QUIC server may trigger double free when processing INITIAL packet (CVE-2026-18798)
• openssl: heap buffer overflow in CMS key unwrapping (CVE-2026-63072)
• openssl: invalid pointer dereference in CMP server via crafted protectionAlg (CVE-2026-63076)
• openssl: RPK server signature algorithm selection can dereference a missing certificate (CVE-2026-14457)
• openssl: excessive memory use buffering DTLS records for a future epoch (CVE-2026-54874)
• openssl: untrusted sender DN used as format string in CMP response validation (CVE-2026-63073)
• openssl: CMP indefinite cache growth of ExtraCerts (CVE-2026-63074)
• openssl: QUIC ACK-only packet retention can cause memory exhaustion (CVE-2026-63075)

Bug Fix(es) and Enhancement(s):

• openssl: HollowByte remote memory-exhaustion DoS fix may be missing [AlmaLinux 10.2.z] (JIRA:AlmaLinux-212362)

For more details about the security issue(s), including the impact, a CVSS score, acknowledgments, and other related information, refer to the CVE page(s) listed in the References section.

OpenSSL is a toolkit that implements the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols, as well as a full-strength general-purpose cryptography library.

Security Fix(es):

• openssl: OpenSSL: Denial of Service via unbounded memory growth in QUIC server (CVE-2026-14456)
• openssl: QUIC server may trigger double free when processing INITIAL packet (CVE-2026-18798)
• openssl: heap buffer overflow in CMS key unwrapping (CVE-2026-63072)
• openssl: invalid pointer dereference in CMP server via crafted protectionAlg (CVE-2026-63076)
• openssl: RPK server signature algorithm selection can dereference a missing certificate (CVE-2026-14457)
• openssl: excessive memory use buffering DTLS records for a future epoch (CVE-2026-54874)
• openssl: untrusted sender DN used as format string in CMP response validation (CVE-2026-63073)
• openssl: CMP indefinite cache growth of ExtraCerts (CVE-2026-63074)
• openssl: QUIC ACK-only packet retention can cause memory exhaustion (CVE-2026-63075)

Bug Fix(es) and Enhancement(s):

• openssl: HollowByte remote memory-exhaustion DoS fix may be missing [AlmaLinux 9.8.z] (JIRA:AlmaLinux-232828)

For more details about the security issue(s), including the impact, a CVSS score, acknowledgments, and other related information, refer to the CVE page(s) listed in the References section.

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell

Issue summary: Receiving a DTLS record for a future epoch while a handshake is in progress causes OpenSSL to buffer far more memory than the record itself requires.

Impact summary: A peer can use a small amount of network traffic to make an OpenSSL DTLS endpoint retain a disproportionately large amount of memory, which may lead to a Denial of Service.

CWE: CWE-405: Asymmetric Resource Consumption (Amplification)

Description: While a DTLS handshake is in progress, a peer may legitimately have already moved on to the next epoch (for example, having sent its ChangeCipherSpec and Finished messages) before the local endpoint has processed the same transition, typically because of reordering on the underlying UDP transport. OpenSSL buffers such early records so that they can be processed once the local endpoint catches up.

Buffering a record currently retains the entire read buffer it arrived in, which is sized to hold the largest possible DTLS record (around 16 kilobytes), rather than just the bytes that make up the record itself. Up to 100 such records may be buffered per connection. As a result, a peer that sends a stream of small forged records claiming to belong to the next epoch can cause an OpenSSL DTLS endpoint to retain around 1.7 megabytes of memory, despite sending only a small fraction of that amount of data over the network.

An attacker therefore gains a memory amplification factor of around 1200, and can multiply the effect across as many associations as it is able to open, making this a remote memory exhaustion Denial of Service risk for DTLS servers. Since the memory retained per connection remains bounded, and any limit an application already places on the number of concurrent associations also bounds the total exposure, this issue has been assessed as Low severity.

FIPS impact: no

No FIPS modules are affected by this issue as the affected code is outside the OpenSSL FIPS module boundary.

OpenSSL 4.0, 3.6, 3.5, 3.4, 3.0, 1.1.1 and 1.0.2 are vulnerable to this issue.

OpenSSL 4.0 users should upgrade to OpenSSL 4.0.2. OpenSSL 3.6 users should upgrade to OpenSSL 3.6.4. OpenSSL 3.5 users should upgrade to OpenSSL 3.5.8. OpenSSL 3.4 users should upgrade to OpenSSL 3.4.7. OpenSSL 3.0 users should upgrade to OpenSSL 3.0.22.

Premium support customers only: OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1zi OpenSSL 1.0.2 users should upgrade to OpenSSL 1.0.2zr

This issue was reported on 18 May 2026 by Amazon Web Services. The fix has been developed by Matt Caswell.

-- cut (non-publishing metadata for internal use) -- Reported by: Amazon Web Services Fixed by: Matt Caswell

• CVE-2026-54874: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch, capping per-connection memory amplification

• CVE-2026-54874: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch, capping per-connection memory amplification

• CVE-2026-54874: fix excessive memory use when buffering DTLS records for a future epoch by copying only the record's own bytes instead of the whole read buffer, and lowering the unprocessed-records queue limit to 16
• CVE-2026-63072: fix 8-byte out-of-bounds heap write in CMS key unwrapping by sizing the AES-WRAP-PAD unwrap output buffer for the worst case

• SECURITY UPDATE: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch (ssl/record/rec_layer_d1.c)
  • debian/patches/CVE-2026-54874.patch: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch (ssl/record/rec_layer_d1.c)
  • CVE-2026-54874
• SECURITY UPDATE: fix 8-byte out-of-bounds heap write in CMS AES-WRAP-PAD key unwrapping (crypto/cms/cms_kari.c)
  • debian/patches/CVE-2026-63072.patch: fix 8-byte out-of-bounds heap write in CMS AES-WRAP-PAD key unwrapping (crypto/cms/cms_kari.c)
  • debian/patches/CVE-2026-63072-test.patch: add the upstream regression test for the AES-WRAP-PAD unwrap overflow (test/cmsapitest.c)
  • CVE-2026-63072

• CVE-2026-54874: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch, capping per-connection memory amplification

• SECURITY UPDATE: Memory amplification denial of service in the DTLS record layer: buffering a record that arrives early for the next epoch retained the entire read buffer, sized for the largest possible DTLS record (~16.7KB), rather than only the record's own bytes, so a peer sending a stream of small forged next-epoch records could make an endpoint retain around 1.7MB per connection.
  • debian/patches/CVE-2026-54874.patch: copy only the record's on-wire bytes into the queue entry and rebase the record pointers on retrieval in ssl/record/rec_layer_d1.c, ssl/record/record.h, ssl/record/record_local.h and ssl/record/ssl3_record.c, and lower the unprocessed_rcds queue limit to 16.
  • CVE-2026-54874
• SECURITY UPDATE: Out-of-bounds heap write in CMS decryption: the key-unwrap output buffer was sized from the cipher's length query, but the AES-WRAP-PAD unwrap primitive writes and cleanses up to inlen bytes on its integrity-failure paths, so a crafted CMS message naming a padded key-wrap OID caused an 8-byte write past the allocation.
  • debian/patches/CVE-2026-63072.patch: size the unwrap output buffer for the worst case in cms_kek_cipher() in crypto/cms/cms_kari.c.
  • CVE-2026-63072

• CVE-2026-54874: fix excessive memory use when buffering DTLS records for a future epoch by copying only the record's own bytes instead of the whole read buffer, and lowering the unprocessed-records queue limit to 16
• CVE-2026-63072: fix 8-byte out-of-bounds heap write in CMS key unwrapping by sizing the AES-WRAP-PAD unwrap output buffer for the worst case

• CVE-2026-54874: fix excessive memory use when buffering DTLS records for a future epoch by copying only the record's own bytes instead of the whole read buffer, and lowering the unprocessed-records queue limit to 16
• CVE-2026-63072: fix 8-byte out-of-bounds heap write in CMS key unwrapping by sizing the AES-WRAP-PAD unwrap output buffer for the worst case

• CVE-2026-54874: avoid allocating a full read buffer for each buffered DTLS next-epoch record and lower the unprocessed_rcds queue limit; prevents excessive memory use from small bogus next-epoch records
• CVE-2026-63072: size the CMS key-unwrap output buffer for the worst case written by AES-WRAP-PAD; prevents an 8-byte out-of-bounds heap write in cms_kek_cipher (crypto/cms/cms_kari.c)

• SECURITY UPDATE: Unbounded memory retention in the DTLS record layer. dtls1_buffer_record() adopted the entire live read buffer for every queued record and allocated a fresh one, so a peer could pin roughly 16.7 KB of heap per queued record regardless of how small the record actually was, up to the 100-entry queue limit, exhausting memory on the endpoint.
  • debian/patches/CVE-2026-54874.patch: copy only the received record into a right-sized allocation and restore it on retrieval, leaving the live read buffer in place, in ssl/record/rec_layer_d1.c, ssl/record/record.h and ssl/record/ssl3_record.c.
  • CVE-2026-54874
• SECURITY UPDATE: Heap buffer overflow in CMS key-agreement recipient processing. For AES key wrap with padding the output size query returns inlen-8 and drives the allocation, while the key-unwrap failure paths then cleanse the full inlen, writing eight bytes past the allocation.
  • debian/patches/CVE-2026-63072.patch: allocate at least the input length in cms_kek_cipher() in crypto/cms/cms_kari.c.
  • debian/patches/CVE-2026-63072-test.patch: add the upstream regression test to test/cmsapitest.c and test/recipes/80-test_cmsapi.t.
  • CVE-2026-63072

• SECURITY UPDATE: Unbounded memory retention in the DTLS record layer. dtls1_buffer_record() adopted the entire live read buffer for every queued record and allocated a fresh one, so a peer could pin roughly 17.7 KB of heap per queued record regardless of how small the record actually was, up to the 100-entry queue limit, exhausting memory on the endpoint.
  • debian/patches/CVE-2026-54874.patch: copy only the received record into a right-sized allocation and restore it on retrieval, leaving the live read buffer in place, in ssl/d1_pkt.c, ssl/d1_lib.c and ssl/dtls1.h.
  • CVE-2026-54874

• CVE-2026-54874: buffer only a DTLS record's own bytes instead of the whole read buffer, and lower the unprocessed records queue limit, to avoid remote memory exhaustion
• CVE-2026-63072: size the CMS key-unwrap output buffer for the AES-WRAP-PAD worst case to prevent an 8-byte out-of-bounds heap write in CMS_decrypt()

• CVE-2026-54874: copy only a DTLS record's own on-wire bytes when buffering it for a future epoch, instead of taking ownership of the whole ~16KB read buffer and allocating a fresh one

• CVE-2026-54874: copy only a DTLS record's own on-wire bytes when buffering it for a future epoch, instead of taking ownership of the whole ~16KB read buffer and allocating a fresh one

• CVE-2026-54874: copy only a DTLS record's own on-wire bytes when buffering it for a future epoch, instead of taking ownership of the whole ~16KB read buffer and allocating a fresh one

• CVE-2026-54874: avoid buffering the whole DTLS read buffer for a record arriving early for the next epoch, capping per-connection memory amplification

Unbounded Memory Growth in QUIC Server Incoming Channel Queue. (CVE-2026-14456) QUIC Server May Trigger Double Free When Processing INITIAL Packet. (CVE-2026-18798) Heap Buffer Overflow in CMS Key Unwrapping. (CVE-2026-63072) Invalid Pointer Dereference in CMP Server via Crafted protectionAlg. (CVE-2026-63076) RPK Server Signature Algorithm Selection Can Dereference a Missing Certificate. (CVE-2026-14457) Excessive Memory Use Buffering DTLS Records for a Future Epoch. (CVE-2026-54874) Untrusted Sender DN Used as Format String in CMP Response Validation. (CVE-2026-63073) CMP Indefinite Cache Growth of ExtraCerts. (CVE-2026-63074) QUIC ACK-only Packet Retention Can Cause Memory Exhaustion. (CVE-2026-63075)

These are all security issues fixed in the libopenssl-3-devel-3.5.3-8.1 package on the GA media of openSUSE Tumbleweed.

This update for openssl-3 fixes the following issues:

August 2026 release.

• CVE-2026-14456: unbounded memory growth in QUIC server incoming channel queue (bsc#1274791).
• CVE-2026-14457: RPK server signature algorithm selection can dereference a missing certificate (bsc#1274792).
• CVE-2026-18798: QUIC server may trigger double free when processing INITIAL packet (bsc#1274777).
• CVE-2026-34181: PKCS#12 files with PBMAC1 are accepted with short HMAC keys (bsc#1266343).
• CVE-2026-54874: excessive memory use buffering DTLS records for a future epoch (bsc#1274795).
• CVE-2026-63072: heap buffer overflow in CMS key unwrapping (bsc#1274788).
• CVE-2026-63073: untrusted sender DN used as format string in CMP response validation (bsc#1274796).
• CVE-2026-63074: CMP indefinite cache growth of ExtraCerts (bsc#1274797).
• CVE-2026-63075: QUIC ACK-only packet retention can cause memory exhaustion (bsc#1274798).
• CVE-2026-63076: invalid pointer dereference in CMP server via crafted protectionAlg (bsc#1274790).
• CVE-2026-75803: AEAD forgeries with empty ciphertext when using EVP_Cipher() (bsc#1275837).

Root has patched CVE-2026-54874 in the openssl package for Root:Alpine:3.15. Multiple fixed versions available.

Root has patched CVE-2026-54874 in the openssl package for Root:Alpine:3.17. Multiple fixed versions available.

Root has patched CVE-2026-54874 in the openssl package for Root:Debian:11. Multiple fixed versions available.

Root has patched CVE-2026-54874 in the openssl package for Root:Debian:12. Multiple fixed versions available.

This update for openssl-3 fixes the following issues:

August 2026 release.

• CVE-2026-14456: unbounded memory growth in QUIC server incoming channel queue (bsc#1274791).
• CVE-2026-14457: RPK server signature algorithm selection can dereference a missing certificate (bsc#1274792).
• CVE-2026-18798: QUIC server may trigger double free when processing INITIAL packet (bsc#1274777).
• CVE-2026-34181: PKCS#12 files with PBMAC1 are accepted with short HMAC keys (bsc#1266343).
• CVE-2026-54874: excessive memory use buffering DTLS records for a future epoch (bsc#1274795).
• CVE-2026-63072: heap buffer overflow in CMS key unwrapping (bsc#1274788).
• CVE-2026-63073: untrusted sender DN used as format string in CMP response validation (bsc#1274796).
• CVE-2026-63074: CMP indefinite cache growth of ExtraCerts (bsc#1274797).
• CVE-2026-63075: QUIC ACK-only packet retention can cause memory exhaustion (bsc#1274798).
• CVE-2026-63076: invalid pointer dereference in CMP server via crafted protectionAlg (bsc#1274790).
• CVE-2026-75803: AEAD forgeries with empty ciphertext when using EVP_Cipher() (bsc#1275837).

This update for openssl-3 fixes the following issues:

August 2026 release.

• CVE-2026-75803: AEAD forgeries with empty ciphertext when using EVP_Cipher() (bsc#1275837).
• CVE-2026-54874: excessive memory use when buffering DTLS records for a future epoch (bsc#1274795).
• CVE-2026-63072: heap buffer overflow in CMS key unwrapping (bsc#1274788).
• CVE-2026-63074: unbounded growth of extraCerts cache in the CMP server (bsc#1274797).
• CVE-2026-63076: invalid pointer dereference in the CMP server via crafted protectionAlg (bsc#1274790).

This update for openssl-3 fixes the following issues:

August 2026 release.

• CVE-2026-75803: AEAD forgeries with empty ciphertext when using EVP_Cipher() (bsc#1275837).
• CVE-2026-54874: excessive memory use when buffering DTLS records for a future epoch (bsc#1274795).
• CVE-2026-63072: heap buffer overflow in CMS key unwrapping (bsc#1274788).
• CVE-2026-63074: unbounded growth of extraCerts cache in the CMP server (bsc#1274797).
• CVE-2026-63076: invalid pointer dereference in the CMP server via crafted protectionAlg (bsc#1274790).

This update for openssl-3 fixes the following issues:

August 2026 release.

• CVE-2026-14456: unbounded memory growth in QUIC server incoming channel queue (bsc#1274791).
• CVE-2026-14457: RPK server signature algorithm selection can dereference a missing certificate (bsc#1274792).
• CVE-2026-18798: QUIC server may trigger double free when processing INITIAL packet (bsc#1274777).
• CVE-2026-34181: PKCS#12 files with PBMAC1 are accepted with short HMAC keys (bsc#1266343).
• CVE-2026-54874: excessive memory use buffering DTLS records for a future epoch (bsc#1274795).
• CVE-2026-63072: heap buffer overflow in CMS key unwrapping (bsc#1274788).
• CVE-2026-63073: untrusted sender DN used as format string in CMP response validation (bsc#1274796).
• CVE-2026-63074: CMP indefinite cache growth of ExtraCerts (bsc#1274797).
• CVE-2026-63075: QUIC ACK-only packet retention can cause memory exhaustion (bsc#1274798).
• CVE-2026-63076: invalid pointer dereference in CMP server via crafted protectionAlg (bsc#1274790).
• CVE-2026-75803: AEAD forgeries with empty ciphertext when using EVP_Cipher() (bsc#1275837).