def get_fingerprinting_wordlist(wordlist):
"""
Load the wordlist of fingerprints and prepare the info in a dict.
It using as a keys the name of the server family and, as value, an
iterable with the keywords related with this web server.
:return: The results of load of webservers keywords info and related webservers.
:rtype: tuple(WEBSERVER_KEYWORDS, RELATED_SERVES) <=> (dict(SERVERNAME: set(str(KEYWORDS))), dict(SERVER_NAME, set(str(RELATED_SERVERS)))
"""
# Load the wordlist
m_w = WordListLoader.get_wordlist_as_dict(wordlist, separator=";", smart_load=True)
# Load references.
#
# References in the wordlist are specified by # prefix.
#
already_parsed = set()
related = defaultdict(set)
m_webservers_keys = extend_items(m_w, already_parsed, related)
return (m_webservers_keys, related)
def http_analyzers(main_url, update_status_func, number_of_entries=4):
"""
Analyze HTTP headers for detect the web server. Return a list with most possible web servers.
:param main_url: Base url to test.
:type main_url: str
:param update_status_func: function used to update the status of the process
:type update_status_func: function
:param number_of_entries: number of resutls tu return for most probable web servers detected.
:type number_of_entries: int
:return: Web server family, Web server version, Web server complete description, related web servers (as a dict('SERVER_RELATED' : set(RELATED_NAMES))), others web server with their probabilities as a dict(CONCRETE_WEB_SERVER, PROBABILITY)
"""
# Load wordlist directly related with a HTTP fields.
# { HTTP_HEADER_FIELD : [wordlists] }
m_wordlists_HTTP_fields = {
"Accept-Ranges" : "accept-range",
"Server" : "banner",
"Cache-Control" : "cache-control",
"Connection" : "connection",
"Content-Type" : "content-type",
"WWW-Authenticate" : "htaccess-realm",
"Pragma" : "pragma",
"X-Powered-By" : "x-powered-by"
}
m_actions = {
'GET' : { 'wordlist' : 'Wordlist_get' , 'weight' : 1 , 'protocol' : 'HTTP/1.1', 'method' : 'GET' , 'payload': '/' },
'LONG_GET' : { 'wordlist' : 'Wordlist_get_long' , 'weight' : 1 , 'protocol' : 'HTTP/1.1', 'method' : 'GET' , 'payload': '/%s' % ('a' * 200) },
'NOT_FOUND' : { 'wordlist' : 'Wordlist_get_notfound' , 'weight' : 2 , 'protocol' : 'HTTP/1.1', 'method' : 'GET' , 'payload': '/404_NOFOUND__X02KAS' },
'HEAD' : { 'wordlist' : 'Wordlist_head' , 'weight' : 3 , 'protocol' : 'HTTP/1.1', 'method' : 'HEAD' , 'payload': '/' },
'OPTIONS' : { 'wordlist' : 'Wordlist_options' , 'weight' : 2 , 'protocol' : 'HTTP/1.1', 'method' : 'OPTIONS' , 'payload': '/' },
'DELETE' : { 'wordlist' : 'Wordlist_delete' , 'weight' : 5 , 'protocol' : 'HTTP/1.1', 'method' : 'DELETE' , 'payload': '/' },
'TEST' : { 'wordlist' : 'Wordlist_attack' , 'weight' : 5 , 'protocol' : 'HTTP/1.1', 'method' : 'TEST' , 'payload': '/' },
'INVALID' : { 'wordlist' : 'Wordlist_wrong_method' , 'weight' : 5 , 'protocol' : 'HTTP/9.8', 'method' : 'GET' , 'payload': '/' },
'ATTACK' : { 'wordlist' : 'Wordlist_wrong_version' , 'weight' : 2 , 'protocol' : 'HTTP/1.1', 'method' : 'GET' , 'payload': "/etc/passwd?format=%%%%&xss=\x22><script>alert('xss');</script>&traversal=../../&sql='%20OR%201;"}
}
# Store results for others HTTP params
m_d = ParsedURL(main_url)
m_hostname = m_d.hostname
m_port = m_d.port
m_debug = False # Only for develop
# Counter of banners. Used when others methods fails.
m_banners_counter = Counter()
# Score counter
m_counters = HTTPAnalyzer(debug=m_debug)
# Var used to update the status
m_data_len = len(m_actions)
i = 1 # element in process
for l_action, v in m_actions.iteritems():
if m_debug:
print "###########"
l_method = v["method"]
l_payload = v["payload"]
l_proto = v["protocol"]
l_wordlist = v["wordlist"]
# Each type of probe hast different weight.
#
# Weights go from 0 - 5
#
l_weight = v["weight"]
# Make the URL
l_url = urljoin(main_url, l_payload)
# Make the raw request
#l_raw_request = "%(method)s %(payload)s %(protocol)s\r\nHost: %(host)s:%(port)s\r\nConnection: Close\r\n\r\n" % (
l_raw_request = "%(method)s %(payload)s %(protocol)s\r\nHost: %(host)s\r\n\r\n" % (
{
"method" : l_method,
"payload" : l_payload,
"protocol" : l_proto,
"host" : m_hostname,
"port" : m_port
}
)
if m_debug:
print "REQUEST"
print l_raw_request
# Do the connection
l_response = None
try:
m_raw_request = HTTP_Raw_Request(l_raw_request)
discard_data(m_raw_request)
l_response = HTTP.make_raw_request(
host = m_hostname,
port = m_port,
raw_request = m_raw_request,
callback = check_raw_response)
if l_response:
discard_data(l_response)
except NetworkException,e:
Logger.log_error_more_verbose("Server-Fingerprint plugin: No response for URL (%s) '%s'. Message: %s" % (l_method, l_url, str(e)))
continue
if not l_response:
Logger.log_error_more_verbose("No response for URL '%s'." % l_url)
continue
if m_debug:
print "RESPONSE"
print l_response.raw_headers
# Update the status
update_status_func((float(i) * 100.0) / float(m_data_len))
Logger.log_more_verbose("Making '%s' test." % (l_wordlist))
i += 1
# Analyze for each wordlist
#
# Store the server banner
try:
m_banners_counter[l_response.headers["Server"]] += l_weight
except KeyError:
pass
#
# =====================
# HTTP directly related
# =====================
#
#
for l_http_header_name, l_header_wordlist in m_wordlists_HTTP_fields.iteritems():
# Check if HTTP header field is in response
if l_http_header_name not in l_response.headers:
continue
l_curr_header_value = l_response.headers[l_http_header_name]
# Generate concrete wordlist name
l_wordlist_path = Config.plugin_extra_config[l_wordlist][l_header_wordlist]
# Load words for the wordlist
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(l_wordlist_path)
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_curr_header_value)
m_counters.inc(l_matches, l_action, l_weight, l_http_header_name, message="HTTP field: " + l_curr_header_value)
#
# =======================
# HTTP INdirectly related
# =======================
#
#
#
# Status code
# ===========
#
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["statuscode"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_response.status)
m_counters.inc(l_matches, l_action, l_weight, "statuscode", message="Status code: " + l_response.status)
#
# Status text
# ===========
#
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["statustext"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_response.reason)
m_counters.inc(l_matches, l_action, l_weight, "statustext", message="Status text: " + l_response.reason)
#
# Header space
# ============
#
# Count the number of spaces between HTTP field name and their value, for example:
# -> Server: Apache 1
# The number of spaces are: 1
#
# -> Server:Apache 1
# The number of spaces are: 0
#
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["header-space"])
# Looking for matches
try:
l_http_value = l_response.headers[0] # get the value of first HTTP field
l_spaces_num = str(abs(len(l_http_value) - len(l_http_value.lstrip())))
l_matches = l_wordlist_instance.matches_by_value(l_spaces_num)
m_counters.inc(l_matches, l_action, l_weight, "header-space", message="Header space: " + l_spaces_num)
except IndexError:
print "index error header space"
pass
#
# Header capitalafterdash
# =======================
#
# Look for non capitalized first letter of field name, for example:
# -> Content-type: ....
# Instead of:
# -> Content-Type: ....
#
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["header-capitalafterdash"])
# Looking for matches
l_valid_fields = [x for x in l_response.headers.iterkeys() if "-" in x]
if l_valid_fields:
l_h = l_valid_fields[0]
l_value = l_h.split("-")[1] # Get the second value: Content-type => type
l_dush = None
if l_value[0].isupper(): # Check first letter is lower
l_dush = 1
else:
l_dush = 0
l_matches = l_wordlist_instance.matches_by_value(l_dush)
m_counters.inc(l_matches, l_action, l_weight, "header-capitalizedafterdush", message="Capital after dash: %s" % str(l_dush))
#
# Header order
# ============
#
l_header_order = ','.join(l_response.headers.iterkeys())
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["header-order"])
l_matches = l_wordlist_instance.matches_by_value(l_header_order)
m_counters.inc(l_matches, l_action, l_weight, "header-order", message="Header order: " + l_header_order)
#
# Protocol name
# ============
#
# For a response like:
# -> HTTP/1.0 200 OK
# ....
#
# Get the 'HTTP' value.
#
try:
l_proto = l_response.protocol # Get the 'HTTP' text from response, if available
if l_proto:
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["protocol-name"])
l_matches = l_wordlist_instance.matches_by_value(l_proto)
m_counters.inc(l_matches, l_action, l_weight, "proto-name", message="Proto name: " + l_proto)
except IndexError:
print "index error protocol name"
pass
#
# Protocol version
# ================
#
# For a response like:
# -> HTTP/1.0 200 OK
# ....
#
# Get the '1.0' value.
#
try:
l_version = l_response.version # Get the '1.0' text from response, if available
if l_version:
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["protocol-version"])
l_matches = l_wordlist_instance.matches_by_value(l_version)
m_counters.inc(l_matches, l_action, l_weight, "proto-version", message="Proto version: " + l_version)
except IndexError:
print "index error protocol version"
pass
if "ETag" in l_response.headers:
l_etag_header = l_response.headers["ETag"]
#
# ETag length
# ================
#
l_etag_len = len(l_etag_header)
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["etag-legth"])
l_matches = l_wordlist_instance.matches_by_value(l_etag_len)
m_counters.inc(l_matches, l_action, l_weight, "etag-length", message="ETag length: " + str(l_etag_len))
#
# ETag Quotes
# ================
#
l_etag_striped = l_etag_header.strip()
if l_etag_striped.startswith("\"") or l_etag_striped.startswith("'"):
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["etag-quotes"])
l_matches = l_wordlist_instance.matches_by_value(l_etag_striped[0])
m_counters.inc(l_matches, l_action, l_weight, "etag-quotes", message="Etag quotes: " + l_etag_striped[0])
if "Vary" in l_response.headers:
l_vary_header = l_response.headers["Vary"]
#
# Vary delimiter
# ================
#
# Checks if Vary header delimiter is something like this:
# -> Vary: Accept-Encoding,User-Agent
# Or this:
# -> Vary: Accept-Encoding, User-Agent
#
l_var_delimiter = ", " if l_vary_header.find(", ") else ","
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["vary-delimiter"])
l_matches = l_wordlist_instance.matches_by_value(l_var_delimiter)
m_counters.inc(l_matches, l_action, l_weight, "vary-delimiter", message="Vary delimiter: " + l_var_delimiter)
#
# Vary capitalizer
# ================
#
# Checks if Vary header delimiter is something like this:
# -> Vary: Accept-Encoding,user-Agent
# Or this:
# -> Vary: accept-encoding,user-agent
#
l_vary_capitalizer = str(0 if l_vary_header == l_vary_header.lower() else 1)
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["vary-capitalize"])
l_matches = l_wordlist_instance.matches_by_value(l_vary_capitalizer)
m_counters.inc(l_matches, l_action, l_weight, "vary-capitalize", message="Vary capitalizer: " + l_vary_capitalizer)
#
# Vary order
# ================
#
# Checks order between vary values:
# -> Vary: Accept-Encoding,user-Agent
# Or this:
# -> Vary: User-Agent,Accept-Encoding
#
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["vary-order"])
l_matches = l_wordlist_instance.matches_by_value(l_vary_header)
m_counters.inc(l_matches, l_action, l_weight, "vary-order", message="Vary order: " + l_vary_header)
#
# =====================
# HTTP specific options
# =====================
#
#
if l_action == "HEAD":
#
# HEAD Options
# ============
#
l_option = l_response.headers.get("Allow")
if l_option:
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["options-public"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_option)
m_counters.inc(l_matches, l_action, l_weight, "options-allow", message="HEAD option: " + l_option)
if l_action == "OPTIONS" or l_action == "INVALID" or l_action == "DELETE":
if "Allow" in l_response.headers:
#
# Options allow
# =============
#
l_option = l_response.headers.get("Allow")
if l_option:
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["options-public"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_option)
m_counters.inc(l_matches, l_action, l_weight, "options-allow", message="OPTIONS allow: " + l_action + " # " + l_option)
#
# Allow delimiter
# ===============
#
l_option = l_response.headers.get("Allow")
if l_option:
l_var_delimiter = ", " if l_option.find(", ") else ","
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["options-delimited"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_var_delimiter)
m_counters.inc(l_matches, l_action, l_weight, "options-delimiter", message="OPTION allow delimiter " + l_action + " # " + l_option)
if "Public" in l_response.headers:
#
# Public response
# ===============
#
l_option = l_response.headers.get("Public")
if l_option:
l_wordlist_instance = WordListLoader.get_wordlist_as_dict(Config.plugin_extra_config[l_wordlist]["options-public"])
# Looking for matches
l_matches = l_wordlist_instance.matches_by_value(l_option)
m_counters.inc(l_matches, l_action, l_weight, "options-public", message="Public response: " + l_action + " # " + l_option)
