CVEReports
CVEReports

Automated vulnerability intelligence platform. Comprehensive reports for high-severity CVEs generated by AI.

Product

  • Home
  • Sitemap
  • RSS Feed

Company

  • About
  • Contact
  • Privacy Policy
  • Terms of Service

© 2026 CVEReports. All rights reserved.

Made with love by Amit Schendel & Alon Barad



CVE-2026-48480

CVE-2026-48480: Undetected Stream Truncation in netty-incubator-codec-ohttp

Alon Barad
Alon Barad
Software Engineer

Jun 24, 2026·7 min read·21 visits

Executive Summary (TL;DR)

An on-path adversary can cleanly truncate a chunked-OHTTP stream at a non-final chunk boundary, bypassing integrity checks without triggering decryption errors or application exceptions.

The Netty incubator codec for Oblivious HTTP (OHTTP) fails to verify that a cryptographically signed final chunk is received before the outer HTTP body terminates. This missing validation allows an on-path adversary to truncate chunked-OHTTP messages cleanly at a non-final chunk boundary, leading to undetected data truncation and compromising message integrity. The vulnerability affects multiple versions of the maven package io.netty.incubator:netty-incubator-codec-ohttp prior to 0.0.22.Final.

Vulnerability Overview

Chunked Oblivious HTTP (OHTTP) specifies a mechanism for routing encrypted HTTP messages through intermediate relays while preserving sender privacy. In this architecture, messages are segmented into a sequence of cryptographically encapsulated chunks. The protocol relies on a dedicated end-of-stream flag inside the final chunk to assert message completeness and protect against unauthorized stream truncation by intermediate transport layers.

Prior to version 0.0.22.Final, the netty-incubator-codec-ohttp component failed to verify the presence of this final chunk flag before declaring a stream successfully processed. This vulnerability is classified as CWE-325: Missing Cryptographic Step. It allows an adversary to manipulate the boundaries of OHTTP communications without triggering parser-level exceptions.

Because the OHTTP layer did not cross-reference the transport layer's termination state with the cryptographic payload's internal structure, the receiving application would process incomplete payloads as if they were fully transmitted. This compromises message integrity and enables silent denial-of-service or information omission attacks. The vulnerability is especially critical in secure proxies or privacy-centric routing services.

Root Cause Analysis

The vulnerability stems from a lack of state enforcement within the chunk parsing loop of OHttpVersionChunkDraft.java. The parse method is designed to iterate through incoming bytes, reconstructing individual chunk frames via parseNextChunk and decoding them sequentially. However, the parser lacked tracking logic to monitor whether any of the parsed chunk frames possessed the isFinal attribute set to true.

When the underlying transport layer completes transmission, it signals the parser by setting the completeBodyReceived flag to true. In vulnerable versions, receiving this signal caused the parser to conclude processing immediately upon consuming the current buffer, regardless of whether a terminal cryptographic chunk had been resolved. The parser assumed that any clean termination of the transport envelope equated to a complete and correct transmission of the cryptographic stream.

An on-path adversary can exploit this design flaw by terminating the TCP connection or ending the HTTP/2 frame sequence early. If the truncation is performed cleanly at a non-final chunk boundary, the parser handles the truncated stream successfully. The absence of a post-decryption validation step prevents the engine from realizing that subsequent chunks were omitted by the sender or dropped in transit. This omission bypasses the designed cryptographic integrity verification.

Code Analysis

An analysis of the patch committed to the Netty project repository reveals the introduction of explicit state tracking for the final-chunk marker. The revised method signature and decoding loop maintain a local boolean state across the parsing execution. If the transport layer reports complete body reception but the final-chunk state is false, the parser now raises a CorruptedFrameException.

Below is the comparison of the code paths within the parse method of OHttpVersionChunkDraft.java before and after remediation:

// Vulnerable Implementation
public void parse(ByteBufAllocator alloc, ByteBuf in, boolean completeBodyReceived, Decoder decoder, List<Object> out) {
    // ... 
    while (in.isReadable()) {
        ChunkInfo chunkInfo = parseNextChunk(in, completeBodyReceived, maxChunkSize);
        if (chunkInfo == null) {
            break;
        }
        decoder.decodeChunk(alloc, in, chunkInfo.length, chunkInfo.isFinal, out);
    }
    // No validation of chunkInfo.isFinal against completeBodyReceived exists here
}
// Patched Implementation
public void parse(ByteBufAllocator alloc, ByteBuf in, boolean completeBodyReceived, Decoder decoder, List<Object> out) {
    // ...
    boolean finalChunk = false; // Added to track final chunk presence
    while (in.isReadable()) {
        ChunkInfo chunkInfo = parseNextChunk(in, completeBodyReceived, maxChunkSize);
        if (chunkInfo == null) {
            break;
        }
        finalChunk |= chunkInfo.isFinal; // Accumulate final chunk flag using bitwise OR
        decoder.decodeChunk(alloc, in, chunkInfo.length, chunkInfo.isFinal, out);
    }
    // Enforce that complete transport body corresponds to a complete cryptographic body
    if (completeBodyReceived && !finalChunk) {
        throw new CorruptedFrameException("OHTTP stream ended without a final chunk");
    }
}

While this fix prevents truncation when a transport stream finishes, security teams should evaluate caller behavior. Because finalChunk is maintained as a local variable within a single execution of the parse method, state preservation depends on the upstream handler correctly accumulating buffers across fragmented TCP packets before calling the parser with completeBodyReceived set to true. If a caller calls parse with completeBodyReceived set to true on an empty buffer invocation, the local finalChunk evaluates to false and triggers an exception. Developers must ensure correct usage of the parser interface in customized pipeline configurations.

Exploitation Methodology

To successfully execute this attack, the adversary must reside on-path between the client and the gateway, such as functioning as the OHTTP Relay or intercepting transport-layer communications as a machine-in-the-middle (MITM). No prior authentication credentials or specialized configuration parameters are required.

The attack begins when a target client initiates a multi-chunk OHTTP transaction, sending a sequence of encapsulated payloads. The attacker intercepts the transmission and identifies the target boundary to truncate, deciding which chunks to forward and which to discard. The adversary forwards only the initial chunks and drops all subsequent chunks, including the cryptographically signed final chunk.

Immediately after transmitting the truncated prefix, the attacker cleanly terminates the outer HTTP transport layer, sending a proper TCP FIN segment or an HTTP/2 END_STREAM flag. The receiving OHTTP gateway receives the clean connection closure, flags completeBodyReceived as true, and processes the partial chunks successfully. The application processes the incomplete data payload without throwing a cryptographic decryption exception or signaling an integrity failure. This enables clean manipulation of message content at the boundary of a chunk.

Impact Assessment

The security impact of CVE-2026-48480 centers on the loss of message integrity within secure communication systems utilizing OHTTP. Because OHTTP is typically deployed in privacy-sensitive environments to proxy DNS queries, telemetry data, or secure API payloads, the silent truncation of these streams can have significant operational consequences.

For example, an attacker could truncate an administrative telemetry upload to omit critical security alerts or system status events. In application protocols where the receiver assumes the completeness of a message based on parser completion, the truncated stream might be parsed as a shorter, valid message, leading to execution state mismatches or logic bypasses.

The vulnerability is assigned a CVSS v4.0 base score of 6.6, indicating medium severity. The attack vector is Network (AV:N), and the attack complexity is Low (AC:L). Although confidentiality is unaffected, the integrity impact is rated as High (VI:H) because the application layer has no cryptographically sound mechanism to detect data loss in vulnerable versions. This can cause applications to operate on incomplete and potentially misleading information.

Remediation & Mitigation

The primary remediation path is upgrading the io.netty.incubator:netty-incubator-codec-ohttp dependency to version 0.0.22.Final or later. This release introduces the finalChunk check within the OHTTP frame parser, which successfully aborts processing with a CorruptedFrameException if truncation is detected.

In scenarios where immediate patching of the runtime binaries is not feasible, organizations should implement deep packet inspection or application-layer length validations where possible. If the application payload contains an internal length field or structured schema (such as JSON or Protobuf), the receiver should explicitly validate that the deserialized object matches the expected structure and length before execution.

Additionally, security teams should monitor application-level exception logs. The appearance of io.netty.handler.codec.CorruptedFrameException: OHTTP stream ended without a final chunk should be aggregated and alerted on, as it indicates either a misconfigured transport layer or an active attempt to truncate secure OHTTP message streams. System administrators should configure perimeter filters to drop connections from relays showing high rates of abrupt closures.

Fix Analysis (1)

Technical Appendix

CVSS Score
6.6/ 10
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N/E:U

Affected Systems

io.netty.incubator:netty-incubator-codec-ohttp

Affected Versions Detail

Product
Affected Versions
Fixed Version
netty-incubator-codec-ohttp
Netty
< 0.0.22.Final0.0.22.Final
AttributeDetail
CWE IDCWE-325
Vulnerability TypeMissing Cryptographic Step
Attack VectorNetwork (AV:N)
CVSS v4.06.6 (Medium)
EPSS Score0.00167
Exploit StatusPoC / Test-Only
CISA KEV StatusNot Listed

MITRE ATT&CK Mapping

T1496Data Destruction / Manipulation
Impact
CWE-325
Missing Cryptographic Step

The application employs a process that should involve multiple cryptographic steps, but it omits one of those steps.

Attack Flow Diagram

Press enter or space to select a node. You can then use the arrow keys to move the node around. Press delete to remove it and escape to cancel.
Press enter or space to select an edge. You can then press delete to remove it or escape to cancel.

More Reports

•about 1 hour ago•CVE-2026-71850
4.8

CVE-2026-71850: Server-Side Rendering Data Exposure in Hono JSX Memoization

A session data exposure vulnerability in the Hono web application framework (hono/jsx module) allows consecutive users to receive cached HTML outputs containing private data. When JSX components wrapped in `memo()` are rendered on the server, the caching mechanism utilizes a module-level closure that persists across independent HTTP requests. When subsequent requests occur with matching props, the components are not re-evaluated, and cached HTML is served. If these components read request-scoped or session-specific data via ambient APIs, the data of the first user is exposed to subsequent users.

Amit Schendel
Amit Schendel
1 views•6 min read
•about 2 hours ago•CVE-2026-71851
9.0

CVE-2026-71851: Use of Cryptographically Weak PRNG in crypto-js (Ill Bloom)

A severe, twelve-year-old cryptographic weakness in crypto-js (versions < 4.0.0) generated pseudorandom numbers using a custom Multiply-With-Carry (MWC) algorithm seeded from the non-secure Math.random(). This reduces 128-bit and 256-bit key spaces to just 2^39 and 2^47 possibilities, allowing offline brute-force attacks.

Alon Barad
Alon Barad
2 views•7 min read
•about 3 hours ago•CVE-2026-71870
4.8

CVE-2026-71870: Uncontrolled Resource Consumption (DoS) in pypdf ToUnicode CMap Parsing

An uncontrolled resource consumption vulnerability (CWE-400) exists in pypdf prior to version 6.15.0. When extracting text from a specially crafted PDF document, the parser fails to restrict token lengths within /ToUnicode CMap streams, causing unbounded memory allocation and process termination via Out-of-Memory (OOM) crashes.

Amit Schendel
Amit Schendel
5 views•7 min read
•about 4 hours ago•CVE-2026-71852
4.8

CVE-2026-71852: Denial of Service via Excessive Iteration and Memory Exhaustion in pypdf CID Font Parsing

A Denial of Service (DoS) vulnerability exists in pypdf prior to version 6.15.0. When parsing maliciously crafted PDF files containing excessively large CID font width ranges, the library suffers from CPU starvation and memory exhaustion due to unconstrained loop expansion.

Amit Schendel
Amit Schendel
6 views•7 min read
•about 5 hours ago•CVE-2026-56818
6.5

CVE-2026-56818: Denial of Service via Memory Pinning in Netty Redis Array Aggregator

A vulnerability in Netty's Redis codec allows remote unauthenticated attackers to cause a memory-pinning Denial of Service (DoS) due to the failure to release partial aggregate state when specific error conditions occur in RedisArrayAggregator. When processing Redis Serialization Protocol (RESP) messages, the aggregator fails to clear internal queues and release retained direct byte buffers on exception paths triggered by exceeded maxElements or invalid length properties. If the pipeline does not explicitly tear down the connection upon detecting a decoder error, subsequent elements continue utilizing the stale context, allowing memory blocks to remain indefinitely pinned.

Amit Schendel
Amit Schendel
6 views•5 min read
•about 6 hours ago•CVE-2026-54164
6.5

CVE-2026-54164: Missing IRI Type Validation in API Platform Core Enables Resource Type Confusion

CVE-2026-54164 is a class/type confusion vulnerability (CWE-843) in API Platform Core. When processing relationships via Internationalized Resource Identifiers (IRIs) in write requests, the framework's normalizer fails to verify if the resolved resource matches the expected type. For PHP applications utilizing untyped properties, the mismatched object is silently assigned, breaking domain logic and data integrity.

Alon Barad
Alon Barad
4 views•6 min read