Description

Issue summary: The DTLS retransmission logic does not correctly handle a handshake message write that is suspended part-way through. The retransmitted message can be read past the message buffer and the retransmission overwrites the internal state the suspended write needs to resume correctly. Impact summary: The retransmitted message can disclose a heap memory to the peer as plaintext handshake data or cause a crash and a Denial of Service when the read reaches an unmapped memory region. CWE: CWE-125: Out-of-bounds Read Description: DTLS handshake messages can be written out in multiple fragments, and a write can suspend mid-message (returning WANT_WRITE) if the underlying transport temporarily cannot accept more data. While such a write is suspended, the DTLS retransmission timer may independently fire and ask the retransmission logic to resend an earlier, already-acknowledged-as-sent message from its retransmit queue. The retransmission logic reused the same internal buffer and position tracking as the message that was still being written, without resetting the position back to the start of the message being retransmitted. As a result the retransmission was read starting from wherever the suspended write had left off, producing a mislabelled message whose body was leftover bytes from the other, larger message still in flight - content that was never meant to be sent at that point, and which could run past the end of the allocated buffer. Separately, even when the retransmission is positioned correctly, allowing it to run to completion while another write is suspended overwrites the same shared bookkeeping that the suspended write depends on to resume. When the application later resumes the suspended write (via a subsequent SSL_read(), SSL_write(), SSL_accept(), or SSL_connect() call), it finds that bookkeeping in a state inconsistent with the message and aborts the process in a debugging build. The fix resets the retransmission's read position to the start of the message before resending, and skips retransmission entirely whenever a handshake write is still suspended, deferring to the next call that resumes it instead. FIPS impact: no The affected code is outside the FIPS module boundary.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.2",
  "pubDate": "2026-09-29T16:17:12.500Z",
  "pubdate": "2026-09-29T16:17:12.500Z",
  "executiveSummary": "This vulnerability is an out-of-bounds read (CWE-125) occurring within the DTLS handshake retransmission logic. It affects systems utilizing DTLS implementations where handshake message writing can be suspended due to transport layer congestion (WANT_WRITE).\nThe flaw allows for a heap-based information disclosure or a Denial of Service (DoS) via process abortion. An attacker can exploit this race condition during the handshake process when a DTLS retransmission timer fires while a previous handshake message write is partially suspended.\nThe risk is significant as it involves memory corruption and leakage of sensitive handshake state or heap data. Successful exploitation requires an attacker to interact with the handshake process over a network, potentially forcing the transport into a state that triggers a suspended write. Authentication is not required for the initial handshake, making this accessible to unauthenticated remote attackers. The impact on availability is high, as the inconsistent state leads to mandatory process termination.\nThis issue is localized to the DTLS retransmission handling logic and does not impact FIPS-validated modules.",
  "technicalDetails": "The root cause of this vulnerability lies in the improper sharing and management of internal buffers and state bookkeeping between active handshake message writes and the retransmission logic.\nIn DTLS, handshake messages may be fragmented. If the transport cannot accept more data, the write operation is suspended, returning WANT_WRITE, and leaving the internal buffer pointer and state markers at the current offset of the suspended message.\nThe vulnerability occurs when the retransmission logic triggers while a write is suspended. The retransmission function inappropriately reuses the same shared buffers and tracking state variables without resetting the read position to the beginning of the retransmission queue. Consequently, the retransmission logic reads from the buffer at the exact offset where the suspended write stopped. This leads to two critical failure modes:\n1. Information Disclosure: The retransmission sends out the remaining bytes of the previously suspended message appended to the retransmission request. This results in the leakage of heap memory content that was never intended for that specific protocol flight, effectively disclosing internal memory to the peer.\n2. Memory Corruption and DoS: Because the retransmission logic continues reading past the intended end of the retransmission buffer, it can access unmapped memory, causing a crash. Furthermore, by overwriting the shared state bookkeeping (used to track current progress of the message), the code invalidates the progress of the original suspended write. When the application resumes the suspended operation via calls such as SSL_read() or SSL_write(), the state is found to be corrupted or inconsistent, causing an intentional abort/crash in debug builds or undefined behavior in production builds.\nThe attack flow follows these steps: 1) The client/server initiates a DTLS handshake. 2) A write operation is initiated but suspended (WANT_WRITE) due to transport flow control. 3) The DTLS retransmission timer triggers before the write is resumed. 4) The retransmission logic uses the partially-used buffer and stale position metadata. 5) The peer receives an malformed or OOB packet causing memory leakage or a subsequent crash when the original write is resumed."
}