GCVE Workshop - 22 September 2026 (14:00-18:00), Luxembourg Before The Vulnopticon Conference - Registration

GHSA-F3WC-3VXV-XMVR

Vulnerability from github – Published: 2023-05-24 17:28 – Updated: 2025-02-13 18:54
VLAI
Summary
Synapse Outgoing federation to specific hosts can be disabled by sending malicious invites
Details

Impact

A malicious user on a Synapse homeserver X with permission to create certain state events can disable outbound federation from X to an arbitrary homeserver Y.

Synapse instances with federation disabled are not affected.

Details

The Matrix protocol allows homeservers to provide an invite_room_state field on a room invite containing a summary of room state. In versions of Synapse up to and including v1.73.0, Synapse did not limit the size of invite_room_state, meaning that it was possible to create an arbitrarily large invite event.

An attacker with an account on a vulnerable Synapse homeserver X could exploit this by having X create an over-sized invite event in a room with a user from another homeserver Y. Once acknowledged by the invitee's homeserver, the invite event would be sent in a batch of events to Y. If the malicious invite is so large that the entire batch is rejected as too large, X's outgoing traffic to Y would become "stuck", meaning that messages and state events created by X would remain unseen by Y.

Patches

Synapse 1.74 refuses to create oversized invite_room_state fields. Server operators should upgrade to Synapse 1.74 or newer urgently.

Workarounds

There are no robust workarounds.

This attack needs an account on Synapse homeserver X to deny federation from X to another homeserver Y. As a partial mitigation, Synapse operators can disable open registration to limit the ability of attackers to create new accounts on homeserver X.

If homeserver X has been attacked in this way, restarting it will resume outgoing federation by entering "catchup mode". For catchup mode to ignore the oversized invites, every attacked room must have a correctly-sized event sent by X which is newer than any oversized invite. This is difficult to arrange, and does not prevent the attacker from repeating their attack.

References

  • https://github.com/matrix-org/synapse/issues/14492 was caused by this issue.
  • https://github.com/matrix-org/synapse/issues/14642 includes the patch described above.

For more information

If you have any questions or comments about this advisory, e-mail us at security@matrix.org.

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "matrix-synapse"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.74.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2023-32323"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-20"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2023-05-24T17:28:26Z",
    "nvd_published_at": "2023-05-26T14:15:10Z",
    "severity": "MODERATE"
  },
  "details": "### Impact\n\nA malicious user on a Synapse homeserver X with permission to create certain state events can disable outbound federation from X to an arbitrary homeserver Y.\n\nSynapse instances with federation disabled are not affected.\n\n#### Details\n\nThe Matrix protocol allows homeservers to provide an `invite_room_state` field on a [room invite](https://spec.matrix.org/v1.5/client-server-api/#mroommember) containing a [summary of room state](https://spec.matrix.org/v1.5/client-server-api/#stripped-state).  In versions of Synapse up to and including v1.73.0, Synapse did not limit the size of `invite_room_state`, meaning that it was possible to create an arbitrarily large invite event.\n\nAn attacker with an account on a vulnerable Synapse homeserver X could exploit this by having X create an over-sized invite event in a room with a user from another homeserver Y. Once acknowledged by the invitee\u0027s homeserver, the invite event would be sent in a batch of events to Y. If the malicious invite is so large that the entire batch is rejected as too large, X\u0027s outgoing traffic to Y would become \"stuck\", meaning that messages and state events created by X would remain unseen by Y.\n\n### Patches\n\nSynapse 1.74 refuses to create oversized `invite_room_state` fields. Server operators should upgrade to Synapse 1.74 or newer urgently.\n\n### Workarounds\n\nThere are no robust workarounds.\n\nThis attack needs an account on Synapse homeserver X to deny federation from X to another homeserver Y. As a partial mitigation, Synapse operators can disable open registration to limit the ability of attackers to create new accounts on homeserver X.\n\nIf homeserver X has been attacked in this way, restarting it will resume outgoing federation by entering \"catchup mode\". For catchup mode to ignore the oversized invites, every attacked room must have a correctly-sized event sent by X which is newer than any oversized invite. This is difficult to arrange, and does not prevent the attacker from repeating their attack.\n\n### References\n\n- https://github.com/matrix-org/synapse/issues/14492 was caused by this issue.\n- https://github.com/matrix-org/synapse/issues/14642  includes the patch described above.\n\n### For more information\n\nIf you have any questions or comments about this advisory, e-mail us at [security@matrix.org](mailto:security@matrix.org).",
  "id": "GHSA-f3wc-3vxv-xmvr",
  "modified": "2025-02-13T18:54:39Z",
  "published": "2023-05-24T17:28:26Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/matrix-org/synapse/security/advisories/GHSA-f3wc-3vxv-xmvr"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2023-32323"
    },
    {
      "type": "WEB",
      "url": "https://github.com/matrix-org/synapse/issues/14492"
    },
    {
      "type": "WEB",
      "url": "https://github.com/matrix-org/synapse/pull/14642"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/matrix-org/synapse"
    },
    {
      "type": "WEB",
      "url": "https://github.com/pypa/advisory-database/tree/main/vulns/matrix-synapse/PYSEC-2023-67.yaml"
    },
    {
      "type": "WEB",
      "url": "https://lists.fedoraproject.org/archives/list/package-announce@lists.fedoraproject.org/message/UJIJRP5ZH6B3KGFLHCAKR2IX2Y4Z25QD"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:N/I:N/A:L",
      "type": "CVSS_V3"
    },
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:N/SC:N/SI:N/SA:L",
      "type": "CVSS_V4"
    }
  ],
  "summary": "Synapse Outgoing federation to specific hosts can be disabled by sending malicious invites"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

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.

Loading…

Detection rules are retrieved from Rulezet.

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.


Loading…