[mod] botdetection: trusted proxies

Replaces `x_for` functionality with `trusted_proxies`. This allows defining which IP / ranges to trust extracting the client IP address from X-Forwarded-For and X-Real-IP headers.

We don't know if the proxy chain will give us the proper client address, so we rely on reading the headers of the proxy before SearXNG (if there is one, in that case it must be added to trusted_proxies) hoping it has done the proper checks. In case a proxy in the chain does not check the client address correctly, integrity is compromised and this should be fixed by whoever manages the proxy, not us.

I had to move the get_cnf func to another file (config.py) to prevent cyclic imports since we need to read the list inside _helpers.py

Closes https://github.com/searxng/searxng/issues/4907
Closes https://github.com/searxng/searxng/issues/3632
Closes https://github.com/searxng/searxng/issues/3191
Closes https://github.com/searxng/searxng/issues/1237

Related https://github.com/searxng/searxng-docker/issues/386
Related https://github.com/inetol-infrastructure/searxng-container/issues/81
This commit is contained in:
Ivan Gabaldon 2025-06-12 14:16:09 +02:00
parent 4b9644eb27
commit eb9f20a823
No known key found for this signature in database
GPG key ID: 075587C93FA67582
10 changed files with 147 additions and 102 deletions

View file

@ -95,8 +95,6 @@ Implementation
from __future__ import annotations
import sys
from pathlib import Path
from ipaddress import ip_address
import flask
import werkzeug
@ -124,34 +122,15 @@ from searx.botdetection import (
# coherency, the logger is "limiter"
logger = logger.getChild('limiter')
CFG: config.Config = None # type: ignore
_INSTALLED = False
LIMITER_CFG_SCHEMA = Path(__file__).parent / "limiter.toml"
"""Base configuration (schema) of the botdetection."""
CFG_DEPRECATED = {
# "dummy.old.foo": "config 'dummy.old.foo' exists only for tests. Don't use it in your real project config."
}
def get_cfg() -> config.Config:
global CFG # pylint: disable=global-statement
if CFG is None:
from . import settings_loader # pylint: disable=import-outside-toplevel
cfg_file = (settings_loader.get_user_cfg_folder() or Path("/etc/searxng")) / "limiter.toml"
CFG = config.Config.from_toml(LIMITER_CFG_SCHEMA, cfg_file, CFG_DEPRECATED)
return CFG
def filter_request(request: SXNG_Request) -> werkzeug.Response | None:
# pylint: disable=too-many-return-statements
cfg = get_cfg()
real_ip = ip_address(get_real_ip(request))
network = get_network(real_ip, cfg)
cfg = config.get_cfg()
real_ip = get_real_ip(request)
network = get_network(real_ip)
if request.path == '/healthz':
return None
@ -228,7 +207,7 @@ def initialize(app: flask.Flask, settings):
# even if the limiter is not activated, the botdetection must be activated
# (e.g. the self_info plugin uses the botdetection to get client IP)
cfg = get_cfg()
cfg = config.get_cfg()
valkey_client = valkeydb.client()
botdetection.init(cfg, valkey_client)