GHSA-WCQJ-WX85-C75P

Vulnerability from github – Published: 2026-11-20 14:54 – Updated: 2026-11-20 14:54
VLAI
Summary
aMQTT: Denial of Service (DoS) via Memory Leak in Unmanaged QoS 2 Session Handshakes
Details

Impact

amqtt is vulnerable to a remote denial of service caused by incomplete MQTT QoS 2 transactions remaining allocated after the associated client disconnects.

MQTT QoS 2 uses a four-step handshake to provide exactly-once delivery. If a client begins a QoS 2 transaction but disconnects before the handshake completes, amqtt retains the associated transaction state and packet identifier without an expiration mechanism to remove it.

QoS 1 transactions are protected by a timeout mechanism, which defaults to 5 seconds, allowing abandoned state to be released. Equivalent timeout handling was not applied to incomplete QoS 2 transactions.

An attacker can repeatedly connect to the broker, initiate QoS 2 transactions, and disconnect before completing the handshake. Each incomplete transaction leaves state allocated in memory. Repeating this process causes broker memory consumption to grow continuously.

Successful exploitation can cause:

  • Unbounded growth in broker memory usage
  • Accumulation of incomplete QoS 2 transaction state
  • Increasing memory pressure on the host system
  • Termination of the broker process when available memory is exhausted
  • Operating-system Out-Of-Memory (OOM) intervention
  • Complete loss of availability for the affected broker process

Exploitation does not require maintaining long-lived connections. An attacker can create incomplete QoS 2 transactions and disconnect, leaving the broker to retain the associated state after the network connection has closed.

Under the default configuration, anonymous clients can exploit the vulnerability without credentials. When authentication is enabled, any authenticated client permitted to initiate QoS 2 message flows may still be able to exploit it. Authentication alone is therefore not a sufficient mitigation.

  • Ecosystem: pip
  • Package: amqtt
  • Affected versions: >= 0.11.0, <= 0.12.0

The vulnerable code is present in the v0.11.0 release commit, 352678c87e50e501a8a5bb91fab73c9400fa046c, and remained present on the main branch as of 5ebbf366f6b79ab12a06f426ffd18ff28954e4ea. Versions before 0.11.0 were not evaluated.

This is a denial-of-service vulnerability only. No confidentiality or integrity impact has been identified. Successful exploitation causes degradation and eventual loss of availability through memory exhaustion.

Patches

The issue will be been patched in the v0.12.1 released version; all users should upgrade.

The completed fix:

  • Apply transaction expiration handling consistently to incomplete QoS 1 and QoS 2 message flows
  • Expire abandoned QoS 2 state when the associated transaction does not complete within the configured timeout
  • Release the associated message state and packet identifiers when the transaction expires
  • Prevent disconnected clients from leaving incomplete QoS 2 transactions resident in broker memory indefinitely

The fixing commit is 2755fd801a6f95e9b73e11116948b57a95348466

Workarounds

Administrators who cannot immediately upgrade should limit the ability of untrusted clients to create large numbers of incomplete QoS 2 transactions.

Where application requirements permit, preventing untrusted clients from using QoS 2 provides the strongest temporary mitigation.

Connection and request rate limiting at a reverse proxy, load balancer, firewall, or other network layer can reduce the rate at which an attacker creates abandoned transactions, but does not eliminate the underlying vulnerability. An attacker able to continue creating incomplete QoS 2 transactions can still cause memory usage to increase over time.

Administrators should also monitor broker memory consumption and configure appropriate process or container memory limits to prevent exhaustion of host resources. Restarting the broker may reclaim abandoned state and restore availability, but the broker remains vulnerable until the patched version is installed.

MQTT Keep Alive limits are not a sufficient mitigation because the vulnerable state persists after the client connection has already terminated.

Show details on source website

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.12.0"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "amqtt"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0.11.0"
            },
            {
              "fixed": "0.12.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-130176"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-401",
      "CWE-772"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-11-20T14:54:50Z",
    "nvd_published_at": "2026-11-19T20:17:28Z",
    "severity": "HIGH"
  },
  "details": "### Impact\n\n`amqtt` is vulnerable to a remote denial of service caused by incomplete MQTT QoS 2 transactions remaining allocated after the associated client disconnects.\n\nMQTT QoS 2 uses a four-step handshake to provide exactly-once delivery. If a client begins a QoS 2 transaction but disconnects before the handshake completes, `amqtt` retains the associated transaction state and packet identifier without an expiration mechanism to remove it.\n\nQoS 1 transactions are protected by a timeout mechanism, which defaults to 5 seconds, allowing abandoned state to be released. Equivalent timeout handling was not applied to incomplete QoS 2 transactions.\n\nAn attacker can repeatedly connect to the broker, initiate QoS 2 transactions, and disconnect before completing the handshake. Each incomplete transaction leaves state allocated in memory. Repeating this process causes broker memory consumption to grow continuously.\n\nSuccessful exploitation can cause:\n\n* Unbounded growth in broker memory usage\n* Accumulation of incomplete QoS 2 transaction state\n* Increasing memory pressure on the host system\n* Termination of the broker process when available memory is exhausted\n* Operating-system Out-Of-Memory (OOM) intervention\n* Complete loss of availability for the affected broker process\n\nExploitation does not require maintaining long-lived connections. An attacker can create incomplete QoS 2 transactions and disconnect, leaving the broker to retain the associated state after the network connection has closed.\n\nUnder the default configuration, anonymous clients can exploit the vulnerability without credentials. When authentication is enabled, any authenticated client permitted to initiate QoS 2 message flows may still be able to exploit it. Authentication alone is therefore not a sufficient mitigation.\n\n* Ecosystem: `pip`\n* Package: `amqtt`\n* Affected versions: `\u003e= 0.11.0, \u003c= 0.12.0`\n\nThe vulnerable code is present in the v0.11.0 release commit, 352678c87e50e501a8a5bb91fab73c9400fa046c, and remained present on the main branch as of 5ebbf366f6b79ab12a06f426ffd18ff28954e4ea. Versions before 0.11.0 were not evaluated.\n\nThis is a denial-of-service vulnerability only. No confidentiality or integrity impact has been identified. Successful exploitation causes degradation and eventual loss of availability through memory exhaustion.\n\n### Patches\n\nThe issue will be been patched in the `v0.12.1` released version; all users should upgrade.\n\nThe completed fix:\n\n* Apply transaction expiration handling consistently to incomplete QoS 1 and QoS 2 message flows\n* Expire abandoned QoS 2 state when the associated transaction does not complete within the configured timeout\n* Release the associated message state and packet identifiers when the transaction expires\n* Prevent disconnected clients from leaving incomplete QoS 2 transactions resident in broker memory indefinitely\n\nThe fixing commit is [`2755fd801a6f95e9b73e11116948b57a95348466`](https://github.com/Yakifo/amqtt/commit/2755fd801a6f95e9b73e11116948b57a95348466)\n\n### Workarounds\n\nAdministrators who cannot immediately upgrade should limit the ability of untrusted clients to create large numbers of incomplete QoS 2 transactions.\n\nWhere application requirements permit, preventing untrusted clients from using QoS 2 provides the strongest temporary mitigation.\n\nConnection and request rate limiting at a reverse proxy, load balancer, firewall, or other network layer can reduce the rate at which an attacker creates abandoned transactions, but does not eliminate the underlying vulnerability. An attacker able to continue creating incomplete QoS 2 transactions can still cause memory usage to increase over time.\n\nAdministrators should also monitor broker memory consumption and configure appropriate process or container memory limits to prevent exhaustion of host resources. Restarting the broker may reclaim abandoned state and restore availability, but the broker remains vulnerable until the patched version is installed.\n\nMQTT Keep Alive limits are not a sufficient mitigation because the vulnerable state persists after the client connection has already terminated.",
  "id": "GHSA-wcqj-wx85-c75p",
  "modified": "2026-11-20T14:54:51Z",
  "published": "2026-11-20T14:54:50Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/Yakifo/amqtt/security/advisories/GHSA-wcqj-wx85-c75p"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-130176"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Yakifo/amqtt/pull/404"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Yakifo/amqtt/commit/2755fd801a6f95e9b73e11116948b57a95348466"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Yakifo/amqtt/commit/64a6f8c3ebc47640f2e092885c1d504c8f34f1af"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/Yakifo/amqtt"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Yakifo/amqtt/releases/tag/v0.12.1"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "aMQTT: Denial of Service (DoS) via Memory Leak in Unmanaged QoS 2 Session Handshakes"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.

Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

JSON
How durations are calculated

Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.

Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.

Events included in the timeline

Loading timeline…


    Loading…

    Loading…

    Loading…

    Related by attack behaviour

    Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.