
Cloud Vulnerability DB
コミュニティ主導の脆弱性データベース
CVE-2026-59980 is a Denial of Service vulnerability in the hpack HTTP/2 Header Encoding library for Python (python-hyper/hpack). The flaw stems from unbounded variable integer decoding that produces O(n²) runtime complexity when processing malformed input, effectively blocking further processing with sufficiently large unsanitized input. All versions up to and including 4.1.0 are affected; the vulnerability was published on June 23, 2026, and disclosed publicly on September 23, 2026. It carries a CVSS v3.1 base score of 7.5 (High) and a CVSS v4.0 base score of 6.3 (Medium) (GitHub Advisory, Red Hat Bugzilla).
The root cause is uncontrolled resource consumption (CWE-400) and allocation of resources without limits or throttling (CWE-770) in the decode_integer() function within src/hpack/hpack.py. The HPACK RFC 7541 Section 5.1 does not define a maximum size for variable-length integer encoding, and the original implementation had no upper bound on the number of continuation octets it would process. An attacker can craft HTTP/2 HEADERS frames containing long sequences of 0xFF bytes in the variable-length integer fields; each additional 0xFF byte causes the decoder to perform progressively more work, resulting in O(n²) CPU consumption. The fix (commit 8cfb02c) introduces a VARINT_MAX_LENGTH constant of 5 octets (sufficient to encode a uint32 prefix) and raises an HPACKDecodingError if the representation exceeds this limit (GitHub Advisory, Patch Commit).
Successful exploitation causes excessive CPU consumption on the server processing HTTP/2 connections, rendering the application unresponsive and constituting a Denial of Service condition. There is no impact on confidentiality or integrity — only availability is affected. Because hpack is commonly used as a downstream dependency of the python-hyper/h2 HTTP/2 server and client library, any Python application relying on h2 for HTTP/2 support may be indirectly exposed (GitHub Advisory, Red Hat Bugzilla).
No public proof-of-concept exploit code has been identified, and there is no evidence of in-the-wild exploitation at this time (GitHub Advisory). The vulnerability is remotely exploitable with no authentication or user interaction required, lowering the barrier for abuse. The EPSS score is 0.0, reflecting the current absence of observed exploitation activity. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog.
hpack or h2 libraries (e.g., via HTTP/2 fingerprinting tools or banner analysis on target endpoints).0xFF continuation bytes (e.g., \xff\xff\xff\xff\xff\xff... repeated hundreds or thousands of times).decode_integer() function will enter O(n²) computation as it processes each 0xFF byte without bound.0xFF sequence, the server's CPU is saturated processing the single malformed frame, blocking all further HTTP/2 request handling and effectively taking the service offline (GitHub Advisory, Patch Commit).HPACKDecodingError exceptions (post-patch); absence of completed responses for connections that remain open and CPU-intensive.Upgrade hpack to version 4.2.0 or later, which restricts variable integer decoding to a maximum of 5 octets (uint32 range) and raises an HPACKDecodingError for oversized representations (GitHub Advisory, Patch Commit). As a temporary workaround where immediate upgrade is not possible, sanitize input to the hpack decoder by rejecting or truncating long sequences of 0xFF bytes before they reach the decode_integer() function. Organizations using python-hyper/h2 or other libraries that depend on hpack should audit their dependency trees and update accordingly.
Red Hat tracked the issue via Bugzilla (Bug 2539827) with a low severity/priority rating, and the OpenSUSE security announce list also referenced the advisory (Red Hat Bugzilla). The vulnerability was credited to researcher Hiroki Nishino (GitHub: tawAsh1), who reported the issue to the python-hyper project (Patch Commit). No significant broader media coverage or notable social media discussion has been observed beyond standard vulnerability aggregator postings.
主要なLinuxディストリビューションおよびそのリリースにおける修正の提供状況。
ソース: このレポートは AI を使用して生成されました
無料の脆弱性評価
9つのセキュリティドメインにわたるクラウドセキュリティプラクティスを評価して、リスクレベルをベンチマークし、防御のギャップを特定します。
パーソナライズされたデモを見る
"私が今まで見た中で最高のユーザーエクスペリエンスは、クラウドワークロードを完全に可視化します。"
"Wiz を使えば、クラウド環境で何が起こっているかを 1 つの画面で確認することができます"
"Wizが何かを重要視した場合、それは実際に重要であることを私たちは知っています。"