Common Weakness Enumeration

CWE-400

Weakness Discouraged

Uncontrolled Resource Consumption

Abstraction: Class · Status: Draft

The product does not properly control the allocation and maintenance of a limited resource.

Dimensions

  • Behavior-Property: Uncontrolled
  • Behavior-Type: Consumption
  • Resource-Type: Resource
Mapping guidance

CWE-400 is intended for incorrect behaviors in which the product is expected to track and restrict how many resources it consumes, but CWE-400 is often misused because it is conflated with the "technical impact" of vulnerabilities in which resource consumption occurs. It is sometimes used for low-information vulnerability reports. It is a level-1 Class (i.e., a child of a Pillar).

Closely analyze the specific mistake that is causing resource consumption, and perform a CWE mapping for that mistake. Consider children/descendants.

Relationships and members
ChildOf: CWE-664 · View CWE-1000
Modes of introduction
Operation

The product could be operated in a system or environment with lower resource limits than expected, which might make it easier for attackers to consume all available resources.

System Configuration

The product could be configured with lower resource limits than expected, which might make it easier for attackers to consume all available resources.

Architecture and Design

The designer might not consider how to handle and throttle excessive resource requests, which typically requires careful planning to handle more gracefully than a crash or exit.

Implementation

There are at least three distinct scenarios that can commonly lead to resource exhaustion:

Lack of throttling for the number of allocated resources.

Losing all references to a resource before reaching the shutdown stage.

Not closing/returning a resource after processing.

Resource exhaustion problems often occur due to an incorrect implementation of the following situations:

Error conditions and other exceptional circumstances.

Confusion over which part of the program is responsible for releasing the resource.

Detection methods

Automated Static Analysis

Automated static analysis typically has limited utility in recognizing resource exhaustion problems, except for program-independent system resources such as files, sockets, and processes. For system resources, automated static analysis may be able to detect circumstances in which resources are not released after they have expired. Automated analysis of configuration files may be able to detect settings that do not specify a maximum value.

Automated static analysis tools will not be appropriate for detecting exhaustion of custom resources, such as an intended security policy in which a bulletin board user is only allowed to make a limited number of posts per day.

Effectiveness: Limited

Automated Dynamic Analysis

Certain automated dynamic analysis techniques may be effective in spotting resource exhaustion problems, especially with resources such as processes, memory, and connections. The technique may involve generating a large number of requests to the product within a short time frame.

Effectiveness: Moderate

Fuzzing

While fuzzing is typically geared toward finding low-level implementation bugs, it can inadvertently find resource exhaustion problems. This can occur when the fuzzer generates a large number of test cases but does not restart the targeted product in between test cases. If an individual test case produces a crash, but it does not do so reliably, then an inability to handle resource exhaustion may be the cause.

Effectiveness: Opportunistic

Notes

Theoretical

Vulnerability theory is largely about how behaviors and resources interact. "Resource exhaustion" can be regarded as either a consequence or an attack, depending on the perspective. This entry is an attempt to reflect the underlying weaknesses that enable these attacks (or consequences) to take place.

Other

Database queries that take a long time to process are good DoS targets. An attacker would have to write a few lines of code to generate enough traffic to exceed the site's ability to keep up. This would effectively prevent authorized users from using the site at all. Resources can be exploited simply by ensuring that the target machine must do much more work and consume more resources in order to service a request than the attacker must do to initiate a request.

A prime example of this can be found in old switches that were vulnerable to "macof" attacks (so named for a tool developed by Dugsong). These attacks flooded a switch with random IP and MAC address combinations, therefore exhausting the switch's cache, which held the information of which port corresponded to which MAC addresses. Once this cache was exhausted, the switch would fail in an insecure way and would begin to act simply as a hub, broadcasting all traffic on all ports and allowing for basic sniffing attacks.

Limited resources include memory, file system storage, database connection pool entries, CPU, and others.

Content history
Submission: CLASP · 2006-07-19 · CWE Draft 3
Contribution: "Mapping CWE to 62443" Sub-Working Group · CWE-CAPEC ICS/OT SIG · 2023-01-24 · CWE 4.10

Suggested mappings to ISA/IEC 62443.

Contribution: "Mapping CWE to 62443" Sub-Working Group · CWE-CAPEC ICS/OT SIG · 2023-04-25 · CWE 4.11

Suggested mappings to ISA/IEC 62443.

Contribution: Abhi Balakrishnan · 2025-02-25 · CWE 4.17

Provided diagram to improve CWE usability.

Modification: Eric Dalci · Cigital · 2008-07-01 · CWE 1.0

updated Time_of_Introduction

Modification: · Veracode · 2008-08-15 · CWE 1.0

Suggested OWASP Top Ten 2004 mapping

Modification: CWE Content Team · MITRE · 2008-09-08 · CWE 1.0

updated Common_Consequences, Relationships, Other_Notes, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2008-10-14 · CWE 1.0.1

updated Description, Name, Relationships

Modification: CWE Content Team · MITRE · 2009-01-12 · CWE 1.2

updated Description

Modification: CWE Content Team · MITRE · 2009-05-27 · CWE 1.4

updated Name, Relationships

Modification: CWE Content Team · MITRE · 2009-07-27 · CWE 1.5

updated Description, Relationships

Modification: CWE Content Team · MITRE · 2009-10-29 · CWE 1.6

updated Relationships

Modification: CWE Content Team · MITRE · 2009-12-28 · CWE 1.7

updated Common_Consequences, Demonstrative_Examples, Detection_Factors, Likelihood_of_Exploit, Observed_Examples, Other_Notes, Potential_Mitigations, References

Modification: CWE Content Team · MITRE · 2010-02-16 · CWE 1.8

updated Detection_Factors, Potential_Mitigations, References, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2010-04-05 · CWE 1.8.1

updated Related_Attack_Patterns

Modification: CWE Content Team · MITRE · 2010-06-21 · CWE 1.9

updated Description

Modification: CWE Content Team · MITRE · 2010-09-27 · CWE 1.10

updated Demonstrative_Examples

Modification: CWE Content Team · MITRE · 2011-06-01 · CWE 1.13

updated Common_Consequences, Relationships, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2012-05-11 · CWE 2.2

updated Demonstrative_Examples, Related_Attack_Patterns, Relationships, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2013-07-17 · CWE 2.5

updated Relationships

Modification: CWE Content Team · MITRE · 2014-07-30 · CWE 2.8

updated Relationships, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2015-12-07 · CWE 2.9

updated Related_Attack_Patterns, Relationships

Modification: CWE Content Team · MITRE · 2017-01-19 · CWE 2.10

updated Relationships

Modification: CWE Content Team · MITRE · 2017-11-08 · CWE 3.0

updated Applicable_Platforms, Demonstrative_Examples, Likelihood_of_Exploit, Potential_Mitigations, References, Relationships

Modification: CWE Content Team · MITRE · 2018-03-27 · CWE 3.1

updated References, Type

Modification: CWE Content Team · MITRE · 2019-01-03 · CWE 3.2

updated Alternate_Terms, Description, Name, Relationships, Taxonomy_Mappings, Theoretical_Notes

Modification: CWE Content Team · MITRE · 2019-06-20 · CWE 3.3

updated Related_Attack_Patterns, Relationships

Modification: CWE Content Team · MITRE · 2019-09-19 · CWE 3.4

updated Description, Relationships

Modification: CWE Content Team · MITRE · 2020-02-24 · CWE 4.0

updated Description, References, Related_Attack_Patterns, Relationships

Modification: CWE Content Team · MITRE · 2020-06-25 · CWE 4.1

updated Description, Maintenance_Notes

Modification: CWE Content Team · MITRE · 2020-08-20 · CWE 4.2

updated Relationships

Modification: CWE Content Team · MITRE · 2022-04-28 · CWE 4.7

updated Related_Attack_Patterns

Modification: CWE Content Team · MITRE · 2022-06-28 · CWE 4.8

updated Observed_Examples, Relationships

Modification: CWE Content Team · MITRE · 2022-10-13 · CWE 4.9

updated Observed_Examples, Relationships

Modification: CWE Content Team · MITRE · 2023-01-31 · CWE 4.10

updated Common_Consequences, Description, Detection_Factors, Maintenance_Notes, Related_Attack_Patterns, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2023-04-27 · CWE 4.11

updated Demonstrative_Examples, Relationships, Taxonomy_Mappings

Modification: CWE Content Team · MITRE · 2023-06-29 · CWE 4.12

updated Mapping_Notes, Relationships

Modification: CWE Content Team · MITRE · 2024-11-19 · CWE 4.16

updated Relationships

Modification: CWE Content Team · MITRE · 2025-04-03 · CWE 4.17

updated Common_Consequences, Description, Diagram, Modes_of_Introduction, Other_Notes, Time_of_Introduction

Modification: CWE Content Team · MITRE · 2025-09-09 · CWE 4.18

updated Observed_Examples, References

Modification: CWE Content Team · MITRE · 2025-12-11 · CWE 4.19

updated Applicable_Platforms, Maintenance_Notes, Weakness_Ordinalities

Modification: CWE Content Team · MITRE · 2026-04-30 · CWE 4.20

updated Applicable_Platforms, Maintenance_Notes, Mapping_Notes, Other_Notes

Modification: CWE Content Team · MITRE · 2026-10-28 · CWE 5.0

updated Applicable_Platforms, Dimensions, References, Relationships, Taxonomy_Mappings

7207 vulnerabilities reference this CWE, most recent first.

GHSA-86WM-68PQ-5JWQ

Vulnerability from github – Published: 2026-10-21 13:51 – Updated: 2026-10-21 13:51
VLAI
Summary
ApostropheCMS: Unauthenticated resource exhaustion via pre-auth aposBigUpload processing in @apostrophecms/import-export
Details

Summary

A potential unauthenticated denial-of-service vulnerability exists in the chunked import upload path used by @apostrophecms/import-export.

The affected import API routes invoke:

self.apos.http.bigUploadMiddleware()

before the actual import handler performs its authentication check.

Because no authorize callback is supplied to bigUploadMiddleware(), unauthenticated requests can reach the aposBigUpload start, chunk, and end operations before the later req.user check rejects the import.

As a result, an unauthenticated remote attacker can create server-side upload state, write chunk data to upload storage, trigger temporary-file processing, and cause attacker-controlled cleanup operations before authentication is enforced.

Affected Component

Package:

@apostrophecms/import-export

Confirmed affected version:

3.6.2

Potential earlier affected versions:

The issue may affect earlier 3.x releases in which the current big-upload-based import implementation is present. I have not yet fully verified the exact introduction version and recommend that the maintainers confirm the affected version range.

Vulnerability Type

Primary:

CWE-400 — Uncontrolled Resource Consumption

Related:

CWE-306 — Missing Authentication for Critical Function

CWE-770 — Allocation of Resources Without Limits or Throttling

Technical Details

The page import route registers the following middleware chain:

packages/import-export/modules/@apostrophecms/import-export-page/index.js

importExportImport: [
  self.apos.http.bigUploadMiddleware(),
  async (req) => {
    return self.apos.modules['@apostrophecms/import-export']
      .import(req, self.__meta.name);
  }
]

No authorize callback is passed to bigUploadMiddleware().

The bigUploadMiddleware() implementation explicitly supports an authorize callback and performs that authorization before processing the aposBigUpload protocol when supplied.

Without that callback, requests containing:

aposBigUpload[type]=start

aposBigUpload[type]=chunk

or

aposBigUpload[type]=end

are processed immediately.

The actual authentication check occurs later in:

packages/import-export/lib/methods/import.js

async import(req, moduleName) {
  if (!req.user) {
    throw self.apos.error('forbidden');
  }

This means authentication occurs only after the chunked-upload middleware has already performed server-side work.

Attack Flow

1. Start

An unauthenticated attacker submits an aposBigUpload[type]=start request.

The server generates an upload ID and stores attacker-supplied upload metadata, including the declared chunks count.

2. Chunk

The attacker can upload chunk data.

bigUploadChunk() copies attacker-controlled data into:

/big-uploads/<id>-<n>-<chunk>

3. End

The attacker sends an aposBigUpload[type]=end request.

bigUploadEnd() begins processing the upload and creates temporary files before invoking the downstream import handler.

Only after this processing completes does execution reach the later authentication check in import().

4. Cleanup amplification

The attacker-controlled chunks value is also used by bigUploadCleanupOne():

for (const { chunks } of Object.values(bigUpload.files)) {
  for (let i = 0; (i < chunks); i++) {
    await ufs.remove(...);
  }
}

No small upper bound on the declared chunk count appears to be enforced at upload initialization.

This allows an unauthenticated attacker to increase the number of filesystem cleanup operations performed for an upload.

Proof of Concept

The following reproduction was performed only against an isolated test environment.

Start an upload without authentication

curl -s -X POST \
  'http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=start' \
  -H 'Content-Type: application/json' \
  --data '{"files":{"file":{"name":"evil.tar.gz","size":4,"type":"application/gzip","chunks":1}}}'

The server returns an upload ID despite the request having no authenticated session.

Upload a chunk

printf 'PWN!' >/tmp/chunk.bin

curl -s -X POST \
  'http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=chunk&aposBigUpload[id]=<ID>&aposBigUpload[n]=0&aposBigUpload[chunk]=0' \
  -F 'chunk=@/tmp/chunk.bin;type=application/octet-stream'

The request succeeds and causes server-side upload storage to be used.

Finalize the upload

curl -i -X POST \
  'http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=end&aposBigUpload[id]=<ID>'

Observed behavior:

  • start returns HTTP 200 and an upload ID without authentication.
  • chunk returns HTTP 200 and results in server-side chunk storage.
  • end eventually reaches the downstream authentication check and is rejected.
  • However, the server-side big-upload processing has already occurred before that authentication rejection.

In my isolated harness:

  • chunks=1 resulted in one cleanup removal attempt.
  • chunks=50 resulted in 50 cleanup removal attempts.

This demonstrates that the attacker-controlled chunk count directly influences server-side cleanup work.

Security Impact

The vulnerability allows an unauthenticated remote attacker to cause server-side storage and filesystem operations that should only be reachable by authorized import users.

Possible consequences include:

  • Consumption of upload and temporary storage.
  • Creation of server-side upload state by unauthenticated users.
  • Repeated filesystem I/O.
  • Cleanup amplification based on attacker-controlled chunk counts.
  • Potential service degradation or denial of service when the attack is repeated or sufficiently amplified.

Root Cause

Authentication is enforced in the downstream import handler rather than before the chunked-upload middleware.

The middleware itself already supports an authorize callback specifically for preventing unauthorized use of the big-upload functionality.

Suggested Remediation

Require authentication before any aposBigUpload operation is processed.

For example:

  • Pass an appropriate authorize callback to bigUploadMiddleware() on all import routes.
  • Reject unauthenticated users before start, chunk, or end.
  • Enforce reasonable limits on declared chunk counts.
  • Validate declared size against actual uploaded data.
  • Introduce per-user and global storage/rate limits.
  • Ensure temporary files are removed when downstream processing fails.
  • Add regression tests confirming that anonymous start, chunk, and end requests are rejected before upload state or files are created.

Relevant reviewed commit:

d3ff8c9bb001be1acf7fdc3f035190250ac64b45

Researcher Credit

Kai Zhi

GitHub:

https://github.com/kaizhi888

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "npm",
        "name": "apostrophe"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "4.33.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "npm",
        "name": "@apostrophecms/import-export"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "3.7.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-113711"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400",
      "CWE-770"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-10-21T13:51:21Z",
    "nvd_published_at": "2026-10-20T18:17:21Z",
    "severity": "HIGH"
  },
  "details": "## Summary\n\nA potential unauthenticated denial-of-service vulnerability exists in the chunked import upload path used by `@apostrophecms/import-export`.\n\nThe affected import API routes invoke:\n\n`self.apos.http.bigUploadMiddleware()`\n\nbefore the actual import handler performs its authentication check.\n\nBecause no `authorize` callback is supplied to `bigUploadMiddleware()`, unauthenticated requests can reach the `aposBigUpload` `start`, `chunk`, and `end` operations before the later `req.user` check rejects the import.\n\nAs a result, an unauthenticated remote attacker can create server-side upload state, write chunk data to upload storage, trigger temporary-file processing, and cause attacker-controlled cleanup operations before authentication is enforced.\n\n## Affected Component\n\nPackage:\n\n`@apostrophecms/import-export`\n\nConfirmed affected version:\n\n`3.6.2`\n\nPotential earlier affected versions:\n\nThe issue may affect earlier 3.x releases in which the current big-upload-based import implementation is present. I have not yet fully verified the exact introduction version and recommend that the maintainers confirm the affected version range.\n\n## Vulnerability Type\n\nPrimary:\n\n**CWE-400 \u2014 Uncontrolled Resource Consumption**\n\nRelated:\n\n**CWE-306 \u2014 Missing Authentication for Critical Function**\n\n**CWE-770 \u2014 Allocation of Resources Without Limits or Throttling**\n\n## Technical Details\n\nThe page import route registers the following middleware chain:\n\n`packages/import-export/modules/@apostrophecms/import-export-page/index.js`\n\n```javascript\nimportExportImport: [\n  self.apos.http.bigUploadMiddleware(),\n  async (req) =\u003e {\n    return self.apos.modules[\u0027@apostrophecms/import-export\u0027]\n      .import(req, self.__meta.name);\n  }\n]\n```\n\nNo `authorize` callback is passed to `bigUploadMiddleware()`.\n\nThe `bigUploadMiddleware()` implementation explicitly supports an `authorize` callback and performs that authorization before processing the `aposBigUpload` protocol when supplied.\n\nWithout that callback, requests containing:\n\n`aposBigUpload[type]=start`\n\n`aposBigUpload[type]=chunk`\n\nor\n\n`aposBigUpload[type]=end`\n\nare processed immediately.\n\nThe actual authentication check occurs later in:\n\n`packages/import-export/lib/methods/import.js`\n\n```javascript\nasync import(req, moduleName) {\n  if (!req.user) {\n    throw self.apos.error(\u0027forbidden\u0027);\n  }\n```\n\nThis means authentication occurs only after the chunked-upload middleware has already performed server-side work.\n\n## Attack Flow\n\n### 1. Start\n\nAn unauthenticated attacker submits an `aposBigUpload[type]=start` request.\n\nThe server generates an upload ID and stores attacker-supplied upload metadata, including the declared `chunks` count.\n\n### 2. Chunk\n\nThe attacker can upload chunk data.\n\n`bigUploadChunk()` copies attacker-controlled data into:\n\n`/big-uploads/\u003cid\u003e-\u003cn\u003e-\u003cchunk\u003e`\n\n### 3. End\n\nThe attacker sends an `aposBigUpload[type]=end` request.\n\n`bigUploadEnd()` begins processing the upload and creates temporary files before invoking the downstream import handler.\n\nOnly after this processing completes does execution reach the later authentication check in `import()`.\n\n### 4. Cleanup amplification\n\nThe attacker-controlled `chunks` value is also used by `bigUploadCleanupOne()`:\n\n```javascript\nfor (const { chunks } of Object.values(bigUpload.files)) {\n  for (let i = 0; (i \u003c chunks); i++) {\n    await ufs.remove(...);\n  }\n}\n```\n\nNo small upper bound on the declared chunk count appears to be enforced at upload initialization.\n\nThis allows an unauthenticated attacker to increase the number of filesystem cleanup operations performed for an upload.\n\n## Proof of Concept\n\nThe following reproduction was performed only against an isolated test environment.\n\n### Start an upload without authentication\n\n```bash\ncurl -s -X POST \\\n  \u0027http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=start\u0027 \\\n  -H \u0027Content-Type: application/json\u0027 \\\n  --data \u0027{\"files\":{\"file\":{\"name\":\"evil.tar.gz\",\"size\":4,\"type\":\"application/gzip\",\"chunks\":1}}}\u0027\n```\n\nThe server returns an upload ID despite the request having no authenticated session.\n\n### Upload a chunk\n\n```bash\nprintf \u0027PWN!\u0027 \u003e/tmp/chunk.bin\n\ncurl -s -X POST \\\n  \u0027http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=chunk\u0026aposBigUpload[id]=\u003cID\u003e\u0026aposBigUpload[n]=0\u0026aposBigUpload[chunk]=0\u0027 \\\n  -F \u0027chunk=@/tmp/chunk.bin;type=application/octet-stream\u0027\n```\n\nThe request succeeds and causes server-side upload storage to be used.\n\n### Finalize the upload\n\n```bash\ncurl -i -X POST \\\n  \u0027http://localhost:3000/api/v1/@apostrophecms/page/importExportImport?aposBigUpload[type]=end\u0026aposBigUpload[id]=\u003cID\u003e\u0027\n```\n\nObserved behavior:\n\n* `start` returns HTTP 200 and an upload ID without authentication.\n* `chunk` returns HTTP 200 and results in server-side chunk storage.\n* `end` eventually reaches the downstream authentication check and is rejected.\n* However, the server-side big-upload processing has already occurred before that authentication rejection.\n\nIn my isolated harness:\n\n* `chunks=1` resulted in one cleanup removal attempt.\n* `chunks=50` resulted in 50 cleanup removal attempts.\n\nThis demonstrates that the attacker-controlled chunk count directly influences server-side cleanup work.\n\n## Security Impact\n\nThe vulnerability allows an unauthenticated remote attacker to cause server-side storage and filesystem operations that should only be reachable by authorized import users.\n\nPossible consequences include:\n\n* Consumption of upload and temporary storage.\n* Creation of server-side upload state by unauthenticated users.\n* Repeated filesystem I/O.\n* Cleanup amplification based on attacker-controlled chunk counts.\n* Potential service degradation or denial of service when the attack is repeated or sufficiently amplified.\n\n## Root Cause\n\nAuthentication is enforced in the downstream import handler rather than before the chunked-upload middleware.\n\nThe middleware itself already supports an `authorize` callback specifically for preventing unauthorized use of the big-upload functionality.\n\n## Suggested Remediation\n\nRequire authentication before any `aposBigUpload` operation is processed.\n\nFor example:\n\n* Pass an appropriate `authorize` callback to `bigUploadMiddleware()` on all import routes.\n* Reject unauthenticated users before `start`, `chunk`, or `end`.\n* Enforce reasonable limits on declared chunk counts.\n* Validate declared size against actual uploaded data.\n* Introduce per-user and global storage/rate limits.\n* Ensure temporary files are removed when downstream processing fails.\n* Add regression tests confirming that anonymous `start`, `chunk`, and `end` requests are rejected before upload state or files are created.\n\nRelevant reviewed commit:\n\n`d3ff8c9bb001be1acf7fdc3f035190250ac64b45`\n\n## Researcher Credit\n\n**Kai Zhi**\n\nGitHub:\n\n**https://github.com/kaizhi888**",
  "id": "GHSA-86wm-68pq-5jwq",
  "modified": "2026-10-21T13:51:21Z",
  "published": "2026-10-21T13:51:21Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/apostrophecms/apostrophe/security/advisories/GHSA-86wm-68pq-5jwq"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-113711"
    },
    {
      "type": "WEB",
      "url": "https://github.com/apostrophecms/apostrophe/commit/9500b8b0e1d630d5e07fbb5fd2c25ed0943845ce"
    },
    {
      "type": "WEB",
      "url": "https://github.com/apostrophecms/apostrophe/commit/f84fd41cbf6c22eb8d0008fc9c25df1f7130b350"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/apostrophecms/apostrophe"
    },
    {
      "type": "WEB",
      "url": "https://github.com/apostrophecms/apostrophe/releases/tag/apostrophe@4.33.0"
    }
  ],
  "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": "ApostropheCMS: Unauthenticated resource exhaustion via pre-auth aposBigUpload processing in @apostrophecms/import-export"
}

GHSA-8736-X895-X6JP

Vulnerability from github – Published: 2022-11-29 06:30 – Updated: 2022-12-02 18:30
VLAI
Details

LINE client for iOS before 12.17.0 might be crashed by sharing an invalid shared key of e2ee in group chat.

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2022-41568"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2022-11-29T05:15:00Z",
    "severity": "HIGH"
  },
  "details": "LINE client for iOS before 12.17.0 might be crashed by sharing an invalid shared key of e2ee in group chat.",
  "id": "GHSA-8736-x895-x6jp",
  "modified": "2022-12-02T18:30:29Z",
  "published": "2022-11-29T06:30:19Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2022-41568"
    },
    {
      "type": "WEB",
      "url": "https://hackerone.com/reports/1701642"
    }
  ],
  "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"
    }
  ]
}

GHSA-8737-QX52-HJFF

Vulnerability from github – Published: 2026-09-04 21:32 – Updated: 2026-09-04 21:32
VLAI
Summary
vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds
Details

Summary

The /v1/completions/derender and /v1/chat/completions/derender endpoints accept caller-supplied GenerateResponse objects and postprocess every nested choices[*].token_ids list directly. Unlike the normal render/generate path, derender does not enforce model context length, resolved max_tokens, max_num_seqs, choice-count, or response-size bounds before detokenizing and returning the supplied token IDs. An authenticated API client can therefore make the CPU-only render frontend, or any server exposing these /v1 derender routes, spend CPU and memory proportional to attacker-chosen generated-output-shaped JSON rather than to a bounded generation result.

Technical Details

The render router registers /v1/chat/completions/derender and /v1/completions/derender in vllm/entrypoints/serve/render/api_router.py, and the OpenAI API server attaches this router whenever "generate" or "render" is in supported_tasks (vllm/entrypoints/openai/api_server.py). The routes are under /v1, so they are part of the OpenAI-compatible HTTP API surface and are protected by the API-key middleware when --api-key is configured.

The request types trust generated-output-shaped data from the client. In vllm/entrypoints/serve/disagg/protocol.py, GenerateResponseChoice accepts token_ids: list[int] | None = None, GenerateResponse accepts choices: list[GenerateResponseChoice], and DerenderCompletionRequest accepts generate_responses: list[GenerateResponse]. These fields have no max length, max item count, or relationship to a prior GenerateRequest.

The sink is OnlineDerenderer. derender_completion() iterates every supplied generate_responses entry and every nested choice, calls tokenizer.decode(choice.token_ids, skip_special_tokens=True), appends the decoded text to the response choices, and increments total_completion_tokens from the same supplied list length. derender_chat() has the same shape for a single supplied generate_response, and can also feed the decoded text into tool/reasoning parsers when a parser and chat_request are present. ServingRender.derender_completion_response() calls online_derenderer.derender_completion(request.generate_responses, request.prompt_tokens) before applying any completion-level validation beyond the model check.

Normal render and generation paths derive output limits from max_model_len, the rendered prompt length, request max_tokens / max_completion_tokens, and scheduler limits. Derender bypasses that invariant because it accepts the already-generated output shape directly from the HTTP caller. The missing invariant is: derender should only postprocess bounded generated output, and client-supplied derender payloads must be rejected if their nested generated token/logprob structures exceed the same limits that generation would have enforced.

PoV

The following bounded PoV can be run from a current vLLM checkout containing PR #43606. It asserts the current source facts for the derender routes, unchecked request fields, and decode sink, then simulates the same derender loop with a counting tokenizer. The negative control is a one-choice, 32-token response. The amplified payload keeps the test bounded but demonstrates that all decoded work and returned text scale directly with caller-supplied GenerateResponse contents.

#!/usr/bin/env python3
import subprocess
from dataclasses import dataclass
from pathlib import Path

SOURCE = Path(".")

def require_source_fact(path: str, needles: list[str]) -> None:
    text = (SOURCE / path).read_text()
    missing = [needle for needle in needles if needle not in text]
    if missing:
        raise AssertionError(f"{path} missing expected facts: {missing}")

def source_head() -> str:
    return subprocess.check_output(["git", "rev-parse", "HEAD"], cwd=SOURCE, text=True).strip()

@dataclass
class Choice:
    index: int
    token_ids: list[int]

@dataclass
class GenerateResponse:
    request_id: str
    choices: list[Choice]

class CountingTokenizer:
    def __init__(self) -> None:
        self.decode_calls = 0
        self.decoded_ids = 0
    def decode(self, token_ids: list[int], *, skip_special_tokens: bool = True) -> str:
        self.decode_calls += 1
        self.decoded_ids += len(token_ids)
        return "x" * len(token_ids)

def derender_completion_like_current_head(generate_responses: list[GenerateResponse], tokenizer: CountingTokenizer) -> tuple[int, int, int]:
    output_chars = 0
    choices = 0
    total_completion_tokens = 0
    for gen in generate_responses:
        for choice in gen.choices:
            if not choice.token_ids:
                raise ValueError("choice has empty or null token_ids")
            decoded_text = tokenizer.decode(choice.token_ids, skip_special_tokens=True)
            output_chars += len(decoded_text)
            total_completion_tokens += len(choice.token_ids)
            choices += 1
    return choices, total_completion_tokens, output_chars

def make_payload(responses: int, choices_per_response: int, tokens_per_choice: int) -> list[GenerateResponse]:
    token_ids = [42] * tokens_per_choice
    return [GenerateResponse(request_id=f"gen-{r}", choices=[Choice(index=c, token_ids=list(token_ids)) for c in range(choices_per_response)]) for r in range(responses)]

def run_case(name: str, payload: list[GenerateResponse]) -> None:
    tokenizer = CountingTokenizer()
    choices, completion_tokens, output_chars = derender_completion_like_current_head(payload, tokenizer)
    print(f"{name}: responses={len(payload)} choices={choices} decode_calls={tokenizer.decode_calls} decoded_token_ids={tokenizer.decoded_ids} completion_tokens={completion_tokens} output_chars={output_chars}")

require_source_fact("vllm/entrypoints/serve/render/api_router.py", ['"/v1/completions/derender"', '"/v1/chat/completions/derender"', "app.include_router(router)"])
require_source_fact("vllm/entrypoints/serve/disagg/protocol.py", ["class GenerateResponseChoice(BaseModel):", "token_ids: list[int] | None = None", "class GenerateResponse(BaseModel):", "choices: list[GenerateResponseChoice]", "class DerenderCompletionRequest(BaseModel):", "generate_responses: list[GenerateResponse]"])
require_source_fact("vllm/renderers/online_derenderer.py", ["async def derender_completion(", "for gen, pt in zip(generate_responses, prompt_tokens_list):", "for choice in gen.choices:", "decoded_text = tokenizer.decode(", "total_completion_tokens += len(choice.token_ids)"])
print("source_checks=ok")
print(f"source_head={source_head()}")
run_case("negative_control", make_payload(responses=1, choices_per_response=1, tokens_per_choice=32))
run_case("amplified_payload", make_payload(responses=16, choices_per_response=4, tokens_per_choice=8192))
print("observation=derender decodes every caller-supplied token id before any max_model_len, max_tokens, max_num_seqs, or response-size check")

Impact

An attacker with access to the /v1 API can send derender requests that consume CPU and memory in the frontend/postprocessing process and can cause large responses unrelated to any bounded generation. In disaggregated deployments, this affects the CPU-only render frontend; in servers where the render router is attached alongside generation, it affects the same OpenAI-compatible server process that handles normal client traffic. This can degrade availability for other clients sharing the process.

Likely CWE: CWE-400 (Uncontrolled Resource Consumption) / CWE-770 (Allocation of Resources Without Limits or Throttling). Conservative CVSS v3.1: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L (4.3). This is not Low severity because a regular network API client can induce availability impact in a shared service without local access, invalid model artifacts, or special runtime privileges. If the server is deployed without API-key enforcement for /v1, the privileges component becomes PR:N.

Suggested Fix

Validate derender payloads before any detokenization or parser invocation. Apply bounded limits to generate_response(s), choices, token_ids, prompt_logprobs, logprobs.content, top_logprobs, and routed_experts that are at least as strict as the corresponding generation-side limits. For completions, reject generate_responses counts above the number of prompts that /v1/completions/render would have produced, and reject total nested choice counts above max_num_seqs / n limits. For each choice, reject token_ids longer than the resolved output-token budget, or require derender callers to submit the original bounded GenerateRequest / sampling metadata and validate the GenerateResponse against it before decoding.

Add regression tests for both derender endpoints. The tests should show that a normal bounded derender payload succeeds, while oversized generate_responses, oversized choices, oversized token_ids, and oversized logprob/top-logprob structures are rejected before tokenizer.decode() or parser execution.

Affected Package/Versions

Confirmed affected: current main at ddd3855a28a561a5bb54d380c6e6b8b1e883cc4a and downstream/nightly builds that include the derender endpoints introduced by PR #43606. The derender router, request models, decode sink, render serving bridge, and OpenAI API router attachment have no relevant diff from 00e045b7c7b82599f626779e111233abd4d0a64e to ddd3855a28a561a5bb54d380c6e6b8b1e883cc4a.

Latest release checked: v0.23.0, published on 2026-06-15. Its vllm/entrypoints/serve/render/api_router.py does not expose /v1/completions/derender or /v1/chat/completions/derender, so v0.23.0 was not confirmed affected.

Advisory History

PR #43606 ("[Render] Add /derender endpoints for disaggregated postprocessing") introduced the derender endpoints on main. PR #44285 later refactored the render serving code, and current head still contains the unchecked derender flow.

Public issue search for derender GenerateResponse token_ids returned no reports. Public search for "/v1/completions/derender" returned the derender feature RFC #42729 and unrelated bugs, but no size-bound, DoS, or generated-output postprocessing issue.

Related public request-fanout and resource-bound advisories are distinct:

  • GHSA-3mwp-wvh9-7528 covers an unbounded n parameter on the normal OpenAI completion/chat generation routes. Its root cause is missing upper-bound validation for generated sequence count, its sink is request fanout and request-object copying into the async engine path before scheduling, its precondition is a caller-controlled n, and its fix surface is a cap on generated sequence count. This report reaches /v1/completions/derender and /v1/chat/completions/derender, not the normal generate routes; its root cause is unchecked caller-supplied GenerateResponse / choices / token_ids structures, its sink is OnlineDerenderer detokenization and response construction after generation, its precondition is access to the derender API with generated-output-shaped JSON, and its fix surface is derender payload validation before decode.
  • PR #45390 includes the GHSA-83mh-6mwq-3hg9 batch-message fanout fix class: it bounds the outer BatchChatCompletionRequest.messages conversation list to prevent one request from creating many conversation/request objects before normal generation. This report has no batch conversation list and does not rely on n; one derender request can instead supply oversized nested GenerateResponse choices and token IDs that are detokenized and returned directly. A batch-message max_length limit would not bound derender generate_response(s) or per-choice token/logprob structures.

The completed local report titled "Explicit truncation_side disables tokenizer-level prompt truncation" is also distinct. That report used /v1/completions and /v1/chat/completions with ordinary prompt text plus truncate_prompt_tokens and explicit truncation_side; its root cause was the renderer omitting tokenizer-level max_length and the pre-tokenization character guard before post-token slicing; its sink was prompt tokenization; and its fix surface was preserving tokenizer-level truncation or rejecting over-budget prompts before tokenization. This derender report uses /v1 derender routes, has no prompt text tokenization or truncation-side control, starts from caller-supplied generated-output token IDs, and needs aggregate bounds on derender generate_response(s), choices, token IDs, logprobs, parser inputs, and response construction before detokenization.

Other adjacent vLLM advisories for Rust/gRPC token-id and logprob bounds, structured-output grammar amplification, repetition-detection windows, and pooling/rerank batch fanout are distinct. Those issues affect Rust/gRPC request conversion, grammar compilation, scheduler loops, or engine fanout. This issue affects /v1 derender postprocessing of caller-supplied generated-output objects and requires derender-specific request validation before detokenization.

Resources

  • vllm/entrypoints/serve/render/api_router.py
  • vllm/entrypoints/serve/disagg/protocol.py
  • vllm/renderers/online_derenderer.py
  • vllm/entrypoints/serve/render/serving.py
  • vllm/entrypoints/openai/api_server.py
  • PR #43606: https://github.com/vllm-project/vllm/pull/43606
  • PR #44285: https://github.com/vllm-project/vllm/pull/44285
  • GHSA-3mwp-wvh9-7528: https://github.com/vllm-project/vllm/security/advisories/GHSA-3mwp-wvh9-7528
  • PR #45390: https://github.com/vllm-project/vllm/pull/45390
  • Release v0.23.0: https://github.com/vllm-project/vllm/releases/tag/v0.23.0
Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "vllm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.26.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-71486"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400",
      "CWE-770"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-09-04T21:32:07Z",
    "nvd_published_at": "2026-08-17T20:16:45Z",
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nThe `/v1/completions/derender` and `/v1/chat/completions/derender` endpoints accept caller-supplied `GenerateResponse` objects and postprocess every nested `choices[*].token_ids` list directly. Unlike the normal render/generate path, derender does not enforce model context length, resolved `max_tokens`, `max_num_seqs`, choice-count, or response-size bounds before detokenizing and returning the supplied token IDs. An authenticated API client can therefore make the CPU-only render frontend, or any server exposing these `/v1` derender routes, spend CPU and memory proportional to attacker-chosen generated-output-shaped JSON rather than to a bounded generation result.\n\n## Technical Details\n\nThe render router registers `/v1/chat/completions/derender` and `/v1/completions/derender` in `vllm/entrypoints/serve/render/api_router.py`, and the OpenAI API server attaches this router whenever `\"generate\"` or `\"render\"` is in `supported_tasks` (`vllm/entrypoints/openai/api_server.py`). The routes are under `/v1`, so they are part of the OpenAI-compatible HTTP API surface and are protected by the API-key middleware when `--api-key` is configured.\n\nThe request types trust generated-output-shaped data from the client. In `vllm/entrypoints/serve/disagg/protocol.py`, `GenerateResponseChoice` accepts `token_ids: list[int] | None = None`, `GenerateResponse` accepts `choices: list[GenerateResponseChoice]`, and `DerenderCompletionRequest` accepts `generate_responses: list[GenerateResponse]`. These fields have no max length, max item count, or relationship to a prior `GenerateRequest`.\n\nThe sink is `OnlineDerenderer`. `derender_completion()` iterates every supplied `generate_responses` entry and every nested choice, calls `tokenizer.decode(choice.token_ids, skip_special_tokens=True)`, appends the decoded text to the response choices, and increments `total_completion_tokens` from the same supplied list length. `derender_chat()` has the same shape for a single supplied `generate_response`, and can also feed the decoded text into tool/reasoning parsers when a parser and `chat_request` are present. `ServingRender.derender_completion_response()` calls `online_derenderer.derender_completion(request.generate_responses, request.prompt_tokens)` before applying any completion-level validation beyond the model check.\n\nNormal render and generation paths derive output limits from `max_model_len`, the rendered prompt length, request `max_tokens` / `max_completion_tokens`, and scheduler limits. Derender bypasses that invariant because it accepts the already-generated output shape directly from the HTTP caller. The missing invariant is: derender should only postprocess bounded generated output, and client-supplied derender payloads must be rejected if their nested generated token/logprob structures exceed the same limits that generation would have enforced.\n\n## PoV\n\nThe following bounded PoV can be run from a current vLLM checkout containing PR `#43606`. It asserts the current source facts for the derender routes, unchecked request fields, and decode sink, then simulates the same derender loop with a counting tokenizer. The negative control is a one-choice, 32-token response. The amplified payload keeps the test bounded but demonstrates that all decoded work and returned text scale directly with caller-supplied `GenerateResponse` contents.\n\n```python\n#!/usr/bin/env python3\nimport subprocess\nfrom dataclasses import dataclass\nfrom pathlib import Path\n\nSOURCE = Path(\".\")\n\ndef require_source_fact(path: str, needles: list[str]) -\u003e None:\n    text = (SOURCE / path).read_text()\n    missing = [needle for needle in needles if needle not in text]\n    if missing:\n        raise AssertionError(f\"{path} missing expected facts: {missing}\")\n\ndef source_head() -\u003e str:\n    return subprocess.check_output([\"git\", \"rev-parse\", \"HEAD\"], cwd=SOURCE, text=True).strip()\n\n@dataclass\nclass Choice:\n    index: int\n    token_ids: list[int]\n\n@dataclass\nclass GenerateResponse:\n    request_id: str\n    choices: list[Choice]\n\nclass CountingTokenizer:\n    def __init__(self) -\u003e None:\n        self.decode_calls = 0\n        self.decoded_ids = 0\n    def decode(self, token_ids: list[int], *, skip_special_tokens: bool = True) -\u003e str:\n        self.decode_calls += 1\n        self.decoded_ids += len(token_ids)\n        return \"x\" * len(token_ids)\n\ndef derender_completion_like_current_head(generate_responses: list[GenerateResponse], tokenizer: CountingTokenizer) -\u003e tuple[int, int, int]:\n    output_chars = 0\n    choices = 0\n    total_completion_tokens = 0\n    for gen in generate_responses:\n        for choice in gen.choices:\n            if not choice.token_ids:\n                raise ValueError(\"choice has empty or null token_ids\")\n            decoded_text = tokenizer.decode(choice.token_ids, skip_special_tokens=True)\n            output_chars += len(decoded_text)\n            total_completion_tokens += len(choice.token_ids)\n            choices += 1\n    return choices, total_completion_tokens, output_chars\n\ndef make_payload(responses: int, choices_per_response: int, tokens_per_choice: int) -\u003e list[GenerateResponse]:\n    token_ids = [42] * tokens_per_choice\n    return [GenerateResponse(request_id=f\"gen-{r}\", choices=[Choice(index=c, token_ids=list(token_ids)) for c in range(choices_per_response)]) for r in range(responses)]\n\ndef run_case(name: str, payload: list[GenerateResponse]) -\u003e None:\n    tokenizer = CountingTokenizer()\n    choices, completion_tokens, output_chars = derender_completion_like_current_head(payload, tokenizer)\n    print(f\"{name}: responses={len(payload)} choices={choices} decode_calls={tokenizer.decode_calls} decoded_token_ids={tokenizer.decoded_ids} completion_tokens={completion_tokens} output_chars={output_chars}\")\n\nrequire_source_fact(\"vllm/entrypoints/serve/render/api_router.py\", [\u0027\"/v1/completions/derender\"\u0027, \u0027\"/v1/chat/completions/derender\"\u0027, \"app.include_router(router)\"])\nrequire_source_fact(\"vllm/entrypoints/serve/disagg/protocol.py\", [\"class GenerateResponseChoice(BaseModel):\", \"token_ids: list[int] | None = None\", \"class GenerateResponse(BaseModel):\", \"choices: list[GenerateResponseChoice]\", \"class DerenderCompletionRequest(BaseModel):\", \"generate_responses: list[GenerateResponse]\"])\nrequire_source_fact(\"vllm/renderers/online_derenderer.py\", [\"async def derender_completion(\", \"for gen, pt in zip(generate_responses, prompt_tokens_list):\", \"for choice in gen.choices:\", \"decoded_text = tokenizer.decode(\", \"total_completion_tokens += len(choice.token_ids)\"])\nprint(\"source_checks=ok\")\nprint(f\"source_head={source_head()}\")\nrun_case(\"negative_control\", make_payload(responses=1, choices_per_response=1, tokens_per_choice=32))\nrun_case(\"amplified_payload\", make_payload(responses=16, choices_per_response=4, tokens_per_choice=8192))\nprint(\"observation=derender decodes every caller-supplied token id before any max_model_len, max_tokens, max_num_seqs, or response-size check\")\n```\n\n\n## Impact\n\nAn attacker with access to the `/v1` API can send derender requests that consume CPU and memory in the frontend/postprocessing process and can cause large responses unrelated to any bounded generation. In disaggregated deployments, this affects the CPU-only render frontend; in servers where the render router is attached alongside generation, it affects the same OpenAI-compatible server process that handles normal client traffic. This can degrade availability for other clients sharing the process.\n\nLikely CWE: CWE-400 (Uncontrolled Resource Consumption) / CWE-770 (Allocation of Resources Without Limits or Throttling). Conservative CVSS v3.1: `CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L` (4.3). This is not Low severity because a regular network API client can induce availability impact in a shared service without local access, invalid model artifacts, or special runtime privileges. If the server is deployed without API-key enforcement for `/v1`, the privileges component becomes `PR:N`.\n\n## Suggested Fix\n\nValidate derender payloads before any detokenization or parser invocation. Apply bounded limits to `generate_response(s)`, `choices`, `token_ids`, `prompt_logprobs`, `logprobs.content`, `top_logprobs`, and `routed_experts` that are at least as strict as the corresponding generation-side limits. For completions, reject `generate_responses` counts above the number of prompts that `/v1/completions/render` would have produced, and reject total nested choice counts above `max_num_seqs` / `n` limits. For each choice, reject `token_ids` longer than the resolved output-token budget, or require derender callers to submit the original bounded `GenerateRequest` / sampling metadata and validate the `GenerateResponse` against it before decoding.\n\nAdd regression tests for both derender endpoints. The tests should show that a normal bounded derender payload succeeds, while oversized `generate_responses`, oversized `choices`, oversized `token_ids`, and oversized logprob/top-logprob structures are rejected before `tokenizer.decode()` or parser execution.\n\n## Affected Package/Versions\n\nConfirmed affected: current main at `ddd3855a28a561a5bb54d380c6e6b8b1e883cc4a` and downstream/nightly builds that include the derender endpoints introduced by PR `#43606`. The derender router, request models, decode sink, render serving bridge, and OpenAI API router attachment have no relevant diff from `00e045b7c7b82599f626779e111233abd4d0a64e` to `ddd3855a28a561a5bb54d380c6e6b8b1e883cc4a`.\n\nLatest release checked: `v0.23.0`, published on `2026-06-15`. Its `vllm/entrypoints/serve/render/api_router.py` does not expose `/v1/completions/derender` or `/v1/chat/completions/derender`, so `v0.23.0` was not confirmed affected.\n\n## Advisory History\n\nPR `#43606` (\"[Render] Add `/derender` endpoints for disaggregated postprocessing\") introduced the derender endpoints on main. PR `#44285` later refactored the render serving code, and current head still contains the unchecked derender flow.\n\nPublic issue search for `derender GenerateResponse token_ids` returned no reports. Public search for `\"/v1/completions/derender\"` returned the derender feature RFC `#42729` and unrelated bugs, but no size-bound, DoS, or generated-output postprocessing issue.\n\nRelated public request-fanout and resource-bound advisories are distinct:\n\n- `GHSA-3mwp-wvh9-7528` covers an unbounded `n` parameter on the normal OpenAI completion/chat generation routes. Its root cause is missing upper-bound validation for generated sequence count, its sink is request fanout and request-object copying into the async engine path before scheduling, its precondition is a caller-controlled `n`, and its fix surface is a cap on generated sequence count. This report reaches `/v1/completions/derender` and `/v1/chat/completions/derender`, not the normal generate routes; its root cause is unchecked caller-supplied `GenerateResponse` / `choices` / `token_ids` structures, its sink is `OnlineDerenderer` detokenization and response construction after generation, its precondition is access to the derender API with generated-output-shaped JSON, and its fix surface is derender payload validation before decode.\n- PR `#45390` includes the `GHSA-83mh-6mwq-3hg9` batch-message fanout fix class: it bounds the outer `BatchChatCompletionRequest.messages` conversation list to prevent one request from creating many conversation/request objects before normal generation. This report has no batch conversation list and does not rely on `n`; one derender request can instead supply oversized nested `GenerateResponse` choices and token IDs that are detokenized and returned directly. A batch-message `max_length` limit would not bound derender `generate_response(s)` or per-choice token/logprob structures.\n\nThe completed local report titled \"Explicit truncation_side disables tokenizer-level prompt truncation\" is also distinct. That report used `/v1/completions` and `/v1/chat/completions` with ordinary prompt text plus `truncate_prompt_tokens` and explicit `truncation_side`; its root cause was the renderer omitting tokenizer-level `max_length` and the pre-tokenization character guard before post-token slicing; its sink was prompt tokenization; and its fix surface was preserving tokenizer-level truncation or rejecting over-budget prompts before tokenization. This derender report uses `/v1` derender routes, has no prompt text tokenization or truncation-side control, starts from caller-supplied generated-output token IDs, and needs aggregate bounds on derender `generate_response(s)`, choices, token IDs, logprobs, parser inputs, and response construction before detokenization.\n\nOther adjacent vLLM advisories for Rust/gRPC token-id and logprob bounds, structured-output grammar amplification, repetition-detection windows, and pooling/rerank batch fanout are distinct. Those issues affect Rust/gRPC request conversion, grammar compilation, scheduler loops, or engine fanout. This issue affects `/v1` derender postprocessing of caller-supplied generated-output objects and requires derender-specific request validation before detokenization.\n\n## Resources\n\n- `vllm/entrypoints/serve/render/api_router.py`\n- `vllm/entrypoints/serve/disagg/protocol.py`\n- `vllm/renderers/online_derenderer.py`\n- `vllm/entrypoints/serve/render/serving.py`\n- `vllm/entrypoints/openai/api_server.py`\n- PR `#43606`: `https://github.com/vllm-project/vllm/pull/43606`\n- PR `#44285`: `https://github.com/vllm-project/vllm/pull/44285`\n- `GHSA-3mwp-wvh9-7528`: `https://github.com/vllm-project/vllm/security/advisories/GHSA-3mwp-wvh9-7528`\n- PR `#45390`: `https://github.com/vllm-project/vllm/pull/45390`\n- Release `v0.23.0`: `https://github.com/vllm-project/vllm/releases/tag/v0.23.0`",
  "id": "GHSA-8737-qx52-hjff",
  "modified": "2026-09-04T21:32:08Z",
  "published": "2026-09-04T21:32:07Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-8737-qx52-hjff"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-71486"
    },
    {
      "type": "WEB",
      "url": "https://github.com/vllm-project/vllm/pull/47260"
    },
    {
      "type": "WEB",
      "url": "https://github.com/vllm-project/vllm/commit/8e61b646e2d157f9b93451fa048f9c8530c8a67b"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/vllm-project/vllm"
    },
    {
      "type": "WEB",
      "url": "https://github.com/vllm-project/vllm/releases/tag/v0.26.0"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L",
      "type": "CVSS_V3"
    }
  ],
  "summary": "vLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output bounds"
}

GHSA-8754-HM6G-JHW5

Vulnerability from github – Published: 2022-05-17 02:51 – Updated: 2025-04-20 03:34
VLAI
Details

A vulnerability in the network stack of MikroTik Version 6.38.5 released 2017-03-09 could allow an unauthenticated remote attacker to exhaust all available CPU via a flood of TCP RST packets, preventing the affected router from accepting new TCP connections.

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2017-7285"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2017-03-29T14:59:00Z",
    "severity": "HIGH"
  },
  "details": "A vulnerability in the network stack of MikroTik Version 6.38.5 released 2017-03-09 could allow an unauthenticated remote attacker to exhaust all available CPU via a flood of TCP RST packets, preventing the affected router from accepting new TCP connections.",
  "id": "GHSA-8754-hm6g-jhw5",
  "modified": "2025-04-20T03:34:58Z",
  "published": "2022-05-17T02:51:03Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2017-7285"
    },
    {
      "type": "WEB",
      "url": "https://cxsecurity.com/issue/WLB-2017030242"
    },
    {
      "type": "WEB",
      "url": "https://www.exploit-db.com/exploits/41752"
    },
    {
      "type": "WEB",
      "url": "http://www.securityfocus.com/bid/97266"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-8788-3QJ3-84J7

Vulnerability from github – Published: 2023-01-18 15:30 – Updated: 2023-01-26 18:30
VLAI
Details

Plex media server 1.21 and before is vulnerable to ddos reflection attack via plex service.

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2021-33959"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-346",
      "CWE-400"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2023-01-18T14:15:00Z",
    "severity": "HIGH"
  },
  "details": "Plex media server 1.21 and before is vulnerable to ddos reflection attack via plex service.",
  "id": "GHSA-8788-3qj3-84j7",
  "modified": "2023-01-26T18:30:48Z",
  "published": "2023-01-18T15:30:31Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2021-33959"
    },
    {
      "type": "WEB",
      "url": "https://github.com/lixiang957/CVE-2021-33959"
    },
    {
      "type": "WEB",
      "url": "https://www.freebuf.com/articles/web/260338.html"
    }
  ],
  "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"
    }
  ]
}

GHSA-87HM-5J5G-7HM2

Vulnerability from github – Published: 2026-07-22 00:32 – Updated: 2026-07-22 00:32
VLAI
Details

Vulnerability in the Oracle Solaris product of Oracle Systems (component: Filesystems). The supported version that is affected is 11.4. Easily exploitable vulnerability allows low privileged attacker with logon to the infrastructure where Oracle Solaris executes to compromise Oracle Solaris. Successful attacks of this vulnerability can result in unauthorized ability to cause a hang or frequently repeatable crash (complete DOS) of Oracle Solaris. CVSS 3.1 Base Score 5.5 (Availability impacts). CVSS Vector: (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H).

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2026-61052"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-07-21T22:18:36Z",
    "severity": "MODERATE"
  },
  "details": "Vulnerability in the Oracle Solaris product of Oracle Systems (component: Filesystems).   The supported version that is affected is 11.4. Easily exploitable vulnerability allows low privileged attacker with logon to the infrastructure where Oracle Solaris executes to compromise Oracle Solaris.  Successful attacks of this vulnerability can result in unauthorized ability to cause a hang or frequently repeatable crash (complete DOS) of Oracle Solaris. CVSS 3.1 Base Score 5.5 (Availability impacts).  CVSS Vector: (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H).",
  "id": "GHSA-87hm-5j5g-7hm2",
  "modified": "2026-07-22T00:32:12Z",
  "published": "2026-07-22T00:32:12Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-61052"
    },
    {
      "type": "WEB",
      "url": "https://www.oracle.com/security-alerts/cpujul2026.html"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-87M3-6QJ3-P3XH

Vulnerability from github – Published: 2024-02-07 15:30 – Updated: 2024-10-02 18:38
VLAI
Summary
Liferay Portal denial of service (memory consumption)
Details

The Document and Media widget In Liferay Portal 7.2.0 through 7.3.6, and older unsupported versions, and Liferay DXP 7.3 before service pack 3, 7.2 before fix pack 13, and older unsupported versions, does not limit resource consumption when generating a preview image, which allows remote authenticated users to cause a denial of service (memory consumption) via crafted PNG images.

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "Maven",
        "name": "com.liferay.portal:release.portal.bom"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "7.2.0"
            },
            {
              "fixed": "7.3.7"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2024-25143"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400",
      "CWE-770"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2024-02-07T19:32:22Z",
    "nvd_published_at": "2024-02-07T15:15:08Z",
    "severity": "HIGH"
  },
  "details": "The Document and Media widget In Liferay Portal 7.2.0 through 7.3.6, and older unsupported versions, and Liferay DXP 7.3 before service pack 3, 7.2 before fix pack 13, and older unsupported versions, does not limit resource consumption when generating a preview image, which allows remote authenticated users to cause a denial of service (memory consumption) via crafted PNG images.",
  "id": "GHSA-87m3-6qj3-p3xh",
  "modified": "2024-10-02T18:38:12Z",
  "published": "2024-02-07T15:30:50Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2024-25143"
    },
    {
      "type": "WEB",
      "url": "https://github.com/liferay/liferay-portal/commit/29b73b9b896c7d44fb5d1800a402698c303d1cf6"
    },
    {
      "type": "WEB",
      "url": "https://github.com/liferay/liferay-portal/commit/4381c10ad0722b3b00c3e3567b68538ab0994145"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/liferay/liferay-portal"
    },
    {
      "type": "WEB",
      "url": "https://github.com/liferay/liferay-portal/releases/tag/7.3.7-ga8"
    },
    {
      "type": "WEB",
      "url": "https://liferay.dev/portal/security/known-vulnerabilities/-/asset_publisher/jekt/content/cve-2024-25143"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    },
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "Liferay Portal denial of service (memory consumption)"
}

GHSA-87M9-RV8P-RGMG

Vulnerability from github – Published: 2024-06-10 18:36 – Updated: 2024-06-17 15:27
VLAI
Summary
go-grpc-compression has a zstd decompression bombing vulnerability
Details

Impact

A malicious user could cause a denial of service (DoS) when using a specially crafted gRPC request. The decompression mechanism for zstd did not respect the limits imposed by gRPC, allowing rapid memory usage increases.

Versions v1.1.4 through to v1.2.2 made use of the Decoder.DecodeAll function in github.com/klauspost/compress/zstd to decompress data provided by the peer. The vulnerability is exploitable only by attackers who can send gRPC payloads to users of github.com/mostynb/go-grpc-compression/zstd or github.com/mostynb/go-grpc-compression/nonclobbering/zstd.

Patches

Version v1.2.3 of github.com/mostynb/go-grpc-compression avoids the issue by not using the Decoder.DecodeAll function in github.com/klauspost/compress/zstd.

All users of github.com/mostynb/go-grpc-compression/zstd or github.com/mostynb/go-grpc-compression/nonclobbering/zstd in the affected versions should update to v1.2.3.

Workarounds

Other compression formats were not affected, users may consider switching from zstd to another format without upgrading to a newer release.

References

This issue was uncovered during a security audit performed by Miroslav Stampar of 7ASecurity, facilitated by OSTIF, for the OpenTelemetry project.

https://opentelemetry.io/blog/2024/cve-2024-36129 https://github.com/open-telemetry/opentelemetry-collector/security/advisories/GHSA-c74f-6mfw-mm4v

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "Go",
        "name": "github.com/mostynb/go-grpc-compression"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.1.4"
            },
            {
              "fixed": "1.2.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2024-06-10T18:36:23Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "### Impact\n\nA malicious user could cause a denial of service (DoS) when using a specially crafted gRPC request. The decompression mechanism for zstd did not respect the limits imposed by gRPC, allowing rapid memory usage increases.\n\nVersions v1.1.4 through to v1.2.2 made use of the Decoder.DecodeAll function in github.com/klauspost/compress/zstd to decompress data provided by the peer. The vulnerability is exploitable only by attackers who can send gRPC payloads to users of github.com/mostynb/go-grpc-compression/zstd or github.com/mostynb/go-grpc-compression/nonclobbering/zstd.\n\n### Patches\n\nVersion v1.2.3  of github.com/mostynb/go-grpc-compression avoids the issue by not using the Decoder.DecodeAll function in github.com/klauspost/compress/zstd.\n\nAll users of github.com/mostynb/go-grpc-compression/zstd or github.com/mostynb/go-grpc-compression/nonclobbering/zstd in the affected versions should update to v1.2.3.\n\n### Workarounds\n\nOther compression formats were not affected, users may consider switching from zstd to another format without upgrading to a newer release.\n\n### References\n\nThis issue was uncovered during a security audit performed by [Miroslav Stampar](https://github.com/stamparm/) of [7ASecurity](https://7asecurity.com/), facilitated by [OSTIF](https://ostif.org/), for the OpenTelemetry project.\n\nhttps://opentelemetry.io/blog/2024/cve-2024-36129\nhttps://github.com/open-telemetry/opentelemetry-collector/security/advisories/GHSA-c74f-6mfw-mm4v",
  "id": "GHSA-87m9-rv8p-rgmg",
  "modified": "2024-06-17T15:27:43Z",
  "published": "2024-06-10T18:36:23Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/mostynb/go-grpc-compression/security/advisories/GHSA-87m9-rv8p-rgmg"
    },
    {
      "type": "WEB",
      "url": "https://github.com/mostynb/go-grpc-compression/commit/629c44d3acb9624993cc7de629f47d72109e2ce5"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/mostynb/go-grpc-compression"
    },
    {
      "type": "WEB",
      "url": "https://pkg.go.dev/vuln/GO-2024-2911"
    }
  ],
  "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": "go-grpc-compression has a zstd decompression bombing vulnerability"
}

GHSA-87MC-M7H4-WXP9

Vulnerability from github – Published: 2026-10-02 06:30 – Updated: 2026-10-02 12:31
VLAI
Details

The OMGF | GDPR/DSGVO Compliant, Faster Google Fonts. Easy. WordPress plugin before 6.3.11 does not require authentication or a valid nonce on an action that issues a slow server-side loopback request, allowing unauthenticated attackers to exhaust the site's PHP worker pool and make the entire site unavailable.

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2026-91828"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-10-02T06:16:42Z",
    "severity": "HIGH"
  },
  "details": "The OMGF | GDPR/DSGVO Compliant, Faster Google Fonts. Easy. WordPress plugin before 6.3.11 does not require authentication or a valid nonce on an action that issues a slow server-side loopback request, allowing unauthenticated attackers to exhaust the site\u0027s PHP worker pool and make the entire site unavailable.",
  "id": "GHSA-87mc-m7h4-wxp9",
  "modified": "2026-10-02T12:31:09Z",
  "published": "2026-10-02T06:30:55Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-91828"
    },
    {
      "type": "WEB",
      "url": "https://wpscan.com/vulnerability/be25f7b5-5790-4db7-9056-b43b538a7183"
    }
  ],
  "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"
    }
  ]
}

GHSA-87Q4-V2F4-JH82

Vulnerability from github – Published: 2026-02-11 15:30 – Updated: 2026-02-12 15:32
VLAI
Details

An uncontrolled resource consumption vulnerability has been reported to affect Qsync Central. If a local attacker gains a user account, they can then exploit the vulnerability to launch a denial-of-service (DoS) attack.

We have already fixed the vulnerability in the following version: Qsync Central 5.0.0.4 ( 2026/01/20 ) and later

Show details on source website

{
  "affected": [],
  "aliases": [
    "CVE-2025-54151"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-400",
      "CWE-770"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-02-11T13:15:54Z",
    "severity": "MODERATE"
  },
  "details": "An uncontrolled resource consumption vulnerability has been reported to affect Qsync Central. If a local attacker gains a user account, they can then exploit the vulnerability to launch a denial-of-service (DoS) attack.\n\nWe have already fixed the vulnerability in the following version:\nQsync Central 5.0.0.4 ( 2026/01/20 ) and later",
  "id": "GHSA-87q4-v2f4-jh82",
  "modified": "2026-02-12T15:32:42Z",
  "published": "2026-02-11T15:30:25Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-54151"
    },
    {
      "type": "WEB",
      "url": "https://www.qnap.com/en/security-advisory/qsa-26-02"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    },
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
      "type": "CVSS_V4"
    }
  ]
}

Mitigation
Architecture and Design

Design throttling mechanisms into the system architecture. The best protection is to limit the amount of resources that an unauthorized user can cause to be expended. A strong authentication and access control model will help prevent such attacks from occurring in the first place. The login application should be protected against DoS attacks as much as possible. Limiting the database access, perhaps by caching result sets, can help minimize the resources expended. To further limit the potential for a DoS attack, consider tracking the rate of requests received from users and blocking requests that exceed a defined rate threshold.

Mitigation
Architecture and Design
  • Mitigation of resource exhaustion attacks requires that the target system either:
  • recognizes the attack and denies that user further access for a given amount of time, or
  • uniformly throttles all requests in order to make it more difficult to consume resources more quickly than they can again be freed.
  • The first of these solutions is an issue in itself though, since it may allow attackers to prevent the use of the system by a particular valid user. If the attacker impersonates the valid user, they may be able to prevent the user from accessing the server in question.
  • The second solution is simply difficult to effectively institute -- and even when properly done, it does not provide a full solution. It simply makes the attack require more resources on the part of the attacker.
Mitigation
Architecture and Design

Ensure that protocols have specific limits of scale placed on them.

Mitigation
Implementation

Ensure that all failures in resource allocation place the system into a safe posture.

CAPEC-147: XML Ping of the Death

An attacker initiates a resource depletion attack where a large number of small XML messages are delivered at a sufficiently rapid rate to cause a denial of service or crash of the target. Transactions such as repetitive SOAP transactions can deplete resources faster than a simple flooding attack because of the additional resources used by the SOAP protocol and the resources necessary to process SOAP messages. The transactions used are immaterial as long as they cause resource utilization on the target. In other words, this is a normal flooding attack augmented by using messages that will require extra processing on the target.

CAPEC-227: Sustained Client Engagement

An adversary attempts to deny legitimate users access to a resource by continually engaging a specific resource in an attempt to keep the resource tied up as long as possible. The adversary's primary goal is not to crash or flood the target, which would alert defenders; rather it is to repeatedly perform actions or abuse algorithmic flaws such that a given resource is tied up and not available to a legitimate user. By carefully crafting a requests that keep the resource engaged through what is seemingly benign requests, legitimate users are limited or completely denied access to the resource.

CAPEC-492: Regular Expression Exponential Blowup

An adversary may execute an attack on a program that uses a poor Regular Expression(Regex) implementation by choosing input that results in an extreme situation for the Regex. A typical extreme situation operates at exponential time compared to the input size. This is due to most implementations using a Nondeterministic Finite Automaton(NFA) state machine to be built by the Regex algorithm since NFA allows backtracking and thus more complex regular expressions.