Skip to main content
Vulnerability Database/CVE-2026-107580

CVE-2026-107580: hMailServer DOS Vulnerability

CVE-2026-107580 is a denial of service vulnerability in Progressive Robot hMailServer caused by inefficient algorithmic complexity in message header decoding. This flaw allows attackers to overwhelm IMAP, SMTP, and POP3 services. This article covers technical details, affected versions, impact assessment, and mitigation strategies.

Published:

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

CVE-2026-107580 Overview

CVE-2026-107580 is an algorithmic complexity vulnerability [CWE-407] in Progressive Robot hMailServer versions 6.0.0 through 6.3.5. A remote unauthenticated attacker can send a single crafted email message to make the IMAP, SMTP, POP3, and webmail services unavailable. The flaw lives in the server's decoding of RFC 2047 encoded-word header fields such as Subject. The server unfolds decoded header values with quadratic work relative to the number of line breaks, and an encoded word may expand to arbitrarily many line breaks. Processing such a message occupies a worker thread for minutes or longer, and IMAP worker threads are shared with SMTP and POP3.

Critical Impact

A single unauthenticated email with a crafted encoded-word header can exhaust shared worker threads and render IMAP, SMTP, POP3, and webmail unresponsive.

Affected Products

  • Progressive Robot hMailServer 6.0.0 through 6.3.5
  • IMAP, SMTP, and POP3 services provided by the above versions
  • The hMailServer webmail component on affected versions

Discovery Timeline

  • 2026-10-08 - CVE-2026-107580 published to the National Vulnerability Database (NVD)
  • 2026-10-08 - Last updated in NVD database

Technical Details for CVE-2026-107580

Vulnerability Analysis

The defect is an algorithmic complexity attack [CWE-407] in hMailServer's MIME header decoder. When the server decodes an RFC 2047 encoded word, it then unfolds the decoded value by scanning from the end for each line break and removing it, shifting the remaining bytes each time. Work scales with the square of the number of line breaks in the decoded value.

Because an encoded word is a short ASCII token that can decode to arbitrary content, an attacker can submit a tiny Subject header that expands to a value containing a very large number of line breaks. The decoder then spends minutes or longer per read.

The header is re-read in multiple paths: when an IMAP client searches, sorts, or threads a folder by a header field, when the webmail lists the folder, when a server rule evaluates a header, when a spam tag is appended to the Subject, and when an abuse report is parsed during delivery. A second defect compounds the issue: when reading a message header from disk, the server rescans everything read so far after every 4000 bytes, costing seconds for headers tens of megabytes in size.

Root Cause

The root cause is an O(n²) unfolding routine applied to attacker-controlled decoded header content, combined with a header file reader that performs repeated full scans across accumulated bytes. Neither path bounds the size or line-break count of a decoded header before performing the quadratic work.

Attack Vector

The attack vector is network-based and requires no authentication. An attacker sends a single email with a crafted Subject or other header containing an RFC 2047 encoded word that decodes to many line breaks. Delivery alone can trigger processing through rules and spam tagging. Any subsequent IMAP search, sort, thread operation, webmail folder listing, or abuse report parse re-triggers the slow decode. Because IMAP worker threads are shared with SMTP and POP3, a small number of such messages block all three services from responding.

Detection Methods for CVE-2026-107580

Indicators of Compromise

  • Inbound messages whose headers contain unusually long RFC 2047 encoded words (=?charset?B?...?= or =?charset?Q?...?=) in the Subject or similar fields.
  • hMailServer worker threads pinned at 100% CPU for minutes while processing a single message.
  • IMAP, SMTP, and POP3 clients timing out or receiving no response while the service process remains running.
  • Message store files whose header section is in the multi-megabyte range.

Detection Strategies

  • Parse inbound SMTP traffic for encoded-word tokens whose base64 or quoted-printable payload exceeds a reasonable size threshold.
  • Alert on hMailServer worker threads exceeding sustained CPU thresholds without corresponding client throughput.
  • Correlate IMAP, SMTP, and POP3 connection failures with a specific message delivery event in the hMailServer logs.

Monitoring Recommendations

  • Monitor the hMailServer service process for thread starvation and queue depth growth across IMAP, SMTP, and POP3 listeners.
  • Track per-message processing time in delivery and rule evaluation logs; investigate any message exceeding tens of seconds.
  • Baseline average Subject header length and alert on statistical outliers reaching the mail store.

How to Mitigate CVE-2026-107580

Immediate Actions Required

  • Upgrade hMailServer to version 6.3.6 or later, which addresses the quadratic decoding behavior.
  • Quarantine or delete already-delivered messages that contain oversized encoded-word headers before IMAP clients open the affected folders.
  • Restart the hMailServer service to free worker threads currently blocked on malicious messages.

Patch Information

Progressive Robot published the fix in hMailServer 6.3.6. See the GitLab hMailServer Release 6.3.6 notes and the tracking issue in GitLab hMailServer Work Item #69 for technical context on the decoder and header-reader changes.

Workarounds

  • Place a filtering mail gateway in front of hMailServer that rejects or strips headers containing encoded words above a bounded decoded length.
  • Disable server-side rules that evaluate header fields and disable Subject spam tagging until the patch is applied.
  • Restrict IMAP, SMTP, and POP3 exposure to trusted networks where feasible to reduce the attack surface for unauthenticated submission.
bash
# Configuration example
# No vendor-provided configuration workaround is published for CVE-2026-107580.
# Upgrade to hMailServer 6.3.6 or later: https://gitlab.com/hmailserver/hmailserver/-/releases/v6.3.6

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Sources: National Vulnerability Database, CISA KEV, Exploit-DB

Experience the Most Advanced Cybersecurity Platform

See how the world’s most intelligent, autonomous cybersecurity platform can protect your organization today and into the future.