2026-09-29 17:43CVE-2026-102716eclipse
PUBLISHED5.2CWE-401

An unauthenticated client can drain the RTSP server's packet pool with a couple of dozen requests

that carry a Session header the parser cannot convert.

The Session branch returns the raw NetX error code instead of an RTSP status code:

```c

/* addons/rtsp/nx_rtsp_server.c:2754 */

status = _nx_utility_string_to_uint(field_value_ptr, field_value_length, &session_id);

if (status)

{

return(status); /* NX_INVALID_PARAMETERS / NX_SIZE_ERROR / NX_OVERFLOW */

}

```

Every other branch of the same function maps its failure to an RTSP status first. The CSeq branch

eighteen lines earlier does exactly that (line 2736 returns NX_RTSP_STATUS_CODE_BAD_REQUEST). The

raw code then reaches `_nx_rtsp_server_error_response_send` (nx_rtsp_server.c:1234), which does not

recognise it, takes a path that returns without releasing the response packet it already allocated,

and the block never goes back to the pool.

Six requests with an empty Session header against a 22 packet pool:

```

valid requests: after request 6: pool available = 21, AFTER = 22 / 22

malformed requests: after request 6: pool available = 16, AFTER = 17 / 22

```

One block per request, not returned when the client disconnects. Twenty six requests take the pool

to zero and the server starts failing allocations, after which it serves nobody. If the pool is

shared with the rest of the application, as it is in the shipped sample, the rest of the stack

stops with it.

Convert the `_nx_utility_string_to_uint` failure in the Session branch into

NX_RTSP_STATUS_CODE_BAD_REQUEST the way the CSeq branch does, and release the response packet on

every exit path of `_nx_rtsp_server_error_response_send`.

Problem type

Affected products

Eclipse Foundation

eclipse-threadx/netxduo

<= 6.5.1 - AFFECTED

References

GitHub Security Advisories

GHSA-4637-2x4v-2wq5

An unauthenticated client can drain the RTSP server's packet pool with a couple of dozen requests...

https://github.com/advisories/GHSA-4637-2x4v-2wq5

An unauthenticated client can drain the RTSP server's packet pool with a couple of dozen requests

that carry a Session header the parser cannot convert.

The Session branch returns the raw NetX error code instead of an RTSP status code:




/* addons/rtsp/nx_rtsp_server.c:2754 */



status = _nx_utility_string_to_uint(field_value_ptr, field_value_length, &session_id);



if (status)



{

    return(status);      /* NX_INVALID_PARAMETERS / NX_SIZE_ERROR / NX_OVERFLOW */


}


Every other branch of the same function maps its failure to an RTSP status first. The CSeq branch

eighteen lines earlier does exactly that (line 2736 returns NX_RTSP_STATUS_CODE_BAD_REQUEST). The

raw code then reaches _nx_rtsp_server_error_response_send (nx_rtsp_server.c:1234), which does not

recognise it, takes a path that returns without releasing the response packet it already allocated,

and the block never goes back to the pool.

Six requests with an empty Session header against a 22 packet pool:




valid requests:      after request 6: pool available = 21,  AFTER = 22 / 22



malformed requests:  after request 6: pool available = 16,  AFTER = 17 / 22


One block per request, not returned when the client disconnects. Twenty six requests take the pool

to zero and the server starts failing allocations, after which it serves nobody. If the pool is

shared with the rest of the application, as it is in the shipped sample, the rest of the stack

stops with it.

Convert the _nx_utility_string_to_uint failure in the Session branch into

NX_RTSP_STATUS_CODE_BAD_REQUEST the way the CSeq branch does, and release the response packet on

every exit path of _nx_rtsp_server_error_response_send.

JSON source

https://cveawg.mitre.org/api/cve/CVE-2026-102716
Click to expand
{
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "cveMetadata": {
    "cveId": "CVE-2026-102716",
    "assignerOrgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
    "assignerShortName": "eclipse",
    "dateUpdated": "2026-09-29T17:43:55.870Z",
    "dateReserved": "2026-09-29T16:15:12.001Z",
    "datePublished": "2026-09-29T17:43:55.870Z",
    "state": "PUBLISHED"
  },
  "containers": {
    "cna": {
      "providerMetadata": {
        "orgId": "e51fbebd-6053-4e49-959f-1b94eeb69a2c",
        "shortName": "eclipse",
        "dateUpdated": "2026-09-29T17:43:55.870Z"
      },
      "descriptions": [
        {
          "lang": "en",
          "value": "An unauthenticated client can drain the RTSP server's packet pool with a couple of dozen requests\n\n\n\nthat carry a Session header the parser cannot convert.\n\n\n\nThe Session branch returns the raw NetX error code instead of an RTSP status code:\n\n\n\n```c\n\n\n\n/* addons/rtsp/nx_rtsp_server.c:2754 */\n\n\n\nstatus = _nx_utility_string_to_uint(field_value_ptr, field_value_length, &session_id);\n\n\n\nif (status)\n\n\n\n{\n\n    return(status);      /* NX_INVALID_PARAMETERS / NX_SIZE_ERROR / NX_OVERFLOW */\n\n\n}\n\n\n\n```\n\n\n\nEvery other branch of the same function maps its failure to an RTSP status first. The CSeq branch\n\n\n\neighteen lines earlier does exactly that (line 2736 returns NX_RTSP_STATUS_CODE_BAD_REQUEST). The\n\n\n\nraw code then reaches `_nx_rtsp_server_error_response_send` (nx_rtsp_server.c:1234), which does not\n\n\n\nrecognise it, takes a path that returns without releasing the response packet it already allocated,\n\n\n\nand the block never goes back to the pool.\n\n\n\nSix requests with an empty Session header against a 22 packet pool:\n\n\n\n```\n\n\n\nvalid requests:      after request 6: pool available = 21,  AFTER = 22 / 22\n\n\n\nmalformed requests:  after request 6: pool available = 16,  AFTER = 17 / 22\n\n\n\n```\n\n\n\nOne block per request, not returned when the client disconnects. Twenty six requests take the pool\n\n\n\nto zero and the server starts failing allocations, after which it serves nobody. If the pool is\n\n\n\nshared with the rest of the application, as it is in the shipped sample, the rest of the stack\n\n\n\nstops with it.\n\n\n\nConvert the `_nx_utility_string_to_uint` failure in the Session branch into\n\n\n\nNX_RTSP_STATUS_CODE_BAD_REQUEST the way the CSeq branch does, and release the response packet on\n\n\n\nevery exit path of `_nx_rtsp_server_error_response_send`.",
          "supportingMedia": [
            {
              "type": "text/html",
              "base64": false,
              "value": "<p>An unauthenticated client can drain the RTSP server's packet pool with a couple of dozen requests</p><p>that carry a Session header the parser cannot convert.</p><p>The Session branch returns the raw NetX error code instead of an RTSP status code:</p><p>```c</p><p>/* addons/rtsp/nx_rtsp_server.c:2754 */</p><p>status = _nx_utility_string_to_uint(field_value_ptr, field_value_length, &amp;session_id);</p><p>if (status)</p><p>{</p><code>    return(status);      /* NX_INVALID_PARAMETERS / NX_SIZE_ERROR / NX_OVERFLOW */</code><br><p>}</p><p>```</p><p>Every other branch of the same function maps its failure to an RTSP status first. The CSeq branch</p><p>eighteen lines earlier does exactly that (line 2736 returns NX_RTSP_STATUS_CODE_BAD_REQUEST). The</p><p>raw code then reaches `_nx_rtsp_server_error_response_send` (nx_rtsp_server.c:1234), which does not</p><p>recognise it, takes a path that returns without releasing the response packet it already allocated,</p><p>and the block never goes back to the pool.</p><p>Six requests with an empty Session header against a 22 packet pool:</p><p>```</p><p>valid requests:      after request 6: pool available = 21,  AFTER = 22 / 22</p><p>malformed requests:  after request 6: pool available = 16,  AFTER = 17 / 22</p><p>```</p><p>One block per request, not returned when the client disconnects. Twenty six requests take the pool</p><p>to zero and the server starts failing allocations, after which it serves nobody. If the pool is</p><p>shared with the rest of the application, as it is in the shipped sample, the rest of the stack</p><p>stops with it.</p><p>Convert the `_nx_utility_string_to_uint` failure in the Session branch into</p><p>NX_RTSP_STATUS_CODE_BAD_REQUEST the way the CSeq branch does, and release the response packet on</p><p>every exit path of `_nx_rtsp_server_error_response_send`.</p>"
            }
          ]
        }
      ],
      "affected": [
        {
          "vendor": "Eclipse Foundation",
          "product": "eclipse-threadx/netxduo",
          "defaultStatus": "unaffected",
          "versions": [
            {
              "version": "0",
              "status": "affected",
              "versionType": "semver",
              "lessThanOrEqual": "6.5.1"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "lang": "en",
              "description": "CWE-401 Missing Release of Memory after Effective Lifetime",
              "cweId": "CWE-401",
              "type": "CWE"
            }
          ]
        }
      ],
      "references": [
        {
          "url": "https://github.com/eclipse-threadx/netxduo/security/advisories/GHSA-q462-246c-x3jc"
        }
      ],
      "metrics": [
        {
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "value": "L0stHeart",
          "type": "reporter"
        }
      ]
    }
  }
}