GHSA-J72P-V7RM-CWM3
Unbounded Content-Encoding decompression chain allows denial of service
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N
Fuente: GitHub
Ciclo de vida de la vulnerabilidad
| Fecha | Fase | Evento | Detalle |
|---|---|---|---|
| Divulgación | Publicada | GHSA-J72P-V7RM-CWM3 divulgada · CVSS 8.2 (GitHub) | |
| Remediación | Parche 2.20.0 | Scrapy · 0 d desde divulgación | |
| Monitoreo | Detectada | Ingresó al índice | |
| Monitoreo | Verificada | Última verificación contra las fuentes |
Impact
Since version 2.12.0, `HttpCompressionMiddleware` https://docs.scrapy.org/en/latest/topics/downloader-middleware.html#scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware decodes every content coding listed in the `Content-Encoding` https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Headers/Content-Encoding header of a response, as allowed by RFC 9110 https://www.rfc-editor.org/rfc/rfc9110#name-content-encoding, with no limit on how many.
A malicious server could send a response with a `Content-Encoding` header that chains thousands of codings (e.g. `gzip, gzip, gzip, …`), over a body compressed that many times, to make Scrapy run as many decompression passes, causing high CPU and memory usage.
Similar issues were fixed in curl https://curl.se/docs/CVE-2022-32206.html, urllib3 https://github.com/urllib3/urllib3/security/advisories/GHSA-gm62-xv2j-4w53 and undici https://github.com/nodejs/undici/security/advisories/GHSA-g9mf-h72j-4rw9.
Patches
Upgrade to Scrapy 2.20.0+.
Scrapy 2.20.0 decodes at most 1 content coding per response by default, and drops responses that apply more with `IgnoreRequest` https://docs.scrapy.org/en/latest/topics/exceptions.html#scrapy.exceptions.IgnoreRequest. To allow more, use the new `COMPRESSION_MAX_ENCODINGS` https://docs.scrapy.org/en/latest/topics/downloader-middleware.html#compression-max-encodings setting.
Workarounds
If you cannot upgrade to Scrapy 2.20.0, consider the following mitigations.
- Disable the middleware: If you do not need Scrapy to decompress responses, set `COMPRESSION_ENABLED`
https://docs.scrapy.org/en/latest/topics/downloader-middleware.html#compression-enabledto `False`. - Drop responses with chained codings: Add a downloader middleware
https://docs.scrapy.org/en/latest/topics/downloader-middleware.htmlthat runs before `HttpCompressionMiddleware` on responses, i.e. with a higher order than its 590, and ignores responses that chain more than 1 content coding. For example:
from scrapy.exceptions import IgnoreRequest
class MaxEncodingsMiddleware:
def process_response(self, request, response, spider):
codings = [
coding.strip()
for header in response.headers.getlist("Content-Encoding")
for coding in header.split(b",")
if coding.strip().lower() not in (b"", b"identity")
]
if len(codings) > 1:
raise IgnoreRequest(f"Too many content codings in {response}")
return responseDOWNLOADER_MIDDLEWARES = {
"myproject.middlewares.MaxEncodingsMiddleware": 591,
}Desglose de métricas
Rangos afectados
Estado del parche por paquete
| Paquete / producto | Estado | Corregida en | Desde divulgación | Última versión |
|---|---|---|---|---|
pypiScrapy |
Corregida | 2.20.0 | 0 d | 2.20.0 no afectada |
2.20.0Arsenal público
Sin exploits, módulos ni PoC públicos conocidos a la fecha de verificación.
https://github.com/scrapy/scrapy/security/advisories/GHSA-j72p-v7rm-cwm3Las referencias externas se muestran para copiar: cve.llama.ec no lo redirige a otros sitios.
