def load_execute_host_plugins():
""" Import and run host plugins """
textutils.output_info('Executing ' + str(len(host.__all__)) + ' host plugins')
for plugin_name in host.__all__:
plugin = __import__ ("plugins.host." + plugin_name, fromlist=[plugin_name])
if hasattr(plugin , 'execute'):
plugin.execute()
def sample_root_404():
""" Get the root 404, this has to be done as soon as possible since plugins could use this information. """
manager = ThreadManager()
textutils.output_info('Benchmarking root 404')
for ext in conf.crafted_404_extensions:
random_file = str(uuid.uuid4())
path = conf.path_template.copy()
if path['url'] != '/':
path['url'] = '/' + random_file + ext
else:
path['url'] = random_file + ext
# Were not using the fetch cache for 404 sampling
database.fetch_queue.put(path)
# Forced bogus path check
random_file = str(uuid.uuid4())
path = conf.path_template.copy()
path['url'] = '/' + random_file + '/'
# Were not using the fetch cache for 404 sampling
database.fetch_queue.put(path)
workers = manager.spawn_workers(len(conf.crafted_404_extensions),
FetchCrafted404Worker)
manager.wait_for_idle(workers, database.fetch_queue)
def output_stats(workers=None):
elapsed_time = datetime.now() - database.scan_start_time
if not elapsed_time.seconds:
request_per_seconds = 0
else:
request_per_seconds = database.successful_fetch_count / elapsed_time.seconds
if request_per_seconds:
remaining_seconds = int(database.fetch_queue.qsize() / request_per_seconds)
remaining_timedelta = timedelta(seconds=remaining_seconds)
else:
remaining_seconds = 0
remaining_timedelta = timedelta(seconds=remaining_seconds)
request_per_seconds = "%.2f" % round(request_per_seconds,2)
stats_string = ''.join([
str(request_per_seconds), ' reqs/sec',
', Done: ', str(database.successful_fetch_count),
', Queued: ', str(database.fetch_queue.qsize()),
', Timeouts: ', str(database.total_timeouts), ' (~', str(database.latest_successful_request_time), 's)',
', Remaining: ', str(remaining_timedelta),
', Workers: ', str(len(workers)),
' (hit ctrl+c again to exit)'
])
textutils.output_info(stats_string)
def load_execute_file_plugins():
""" Import and run path plugins """
textutils.output_info('Executing ' + str(len(file.__all__)) + ' file plugins')
for plugin_name in file.__all__:
plugin = __import__ ("plugins.file." + plugin_name, fromlist=[plugin_name])
if hasattr(plugin , 'execute'):
plugin.execute()
def load_execute_file_plugins():
""" Import and run path plugins """
textutils.output_info('Executing ' + str(len(file.__all__)) + ' file plugins')
for plugin_name in file.__all__:
plugin = __import__ ("plugins.file." + plugin_name, fromlist=[plugin_name])
if hasattr(plugin , 'execute'):
plugin.execute()
def execute():
""" Fetch /.svn/entries and parse for target paths """
textutils.output_info(" - Svn Plugin: Searching for /.svn/entries")
target_url = conf.target_base_path + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False
)
if response_code is 200 or response_code is 302:
if conf.allow_download:
textutils.output_info(" - Svn Plugin: /.svn/entries found! crawling... (will download files to output/)")
else:
textutils.output_info(
" - Svn Plugin: /.svn/entries found! crawling... (use -a to download files instead of printing)"
)
# parse entries
parse_svn_entries(conf.target_base_path)
# Clean up display
if conf.allow_download:
textutils.output_info("")
else:
textutils.output_info(" - Svn Plugin: no /.svn/entries found")
def execute():
""" Generate common simple paths (a-z, 0-9) """
path_added = 0
file_added = 0
for char in range(ord('a'), ord('z')):
add_generated_dir(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
for char in range(ord('0'), ord('9')):
add_generated_dir(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
for year in range(1990, date.today().year + 5):
add_generated_dir(str(year))
path_added += 1
textutils.output_info(' - PathGenerator Plugin: added ' + str(path_added) +
' computer generated path.')
textutils.output_info(' - PathGenerator Plugin: added ' + str(file_added) +
' computer generated files.')
def execute():
""" Fetch /.svn/entries and parse for target paths """
textutils.output_info(' - Svn Plugin: Searching for /.svn/entries')
target_url = conf.target_base_path + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
if response_code is 200 or response_code is 302:
if conf.allow_download:
textutils.output_info(
' - Svn Plugin: /.svn/entries found! crawling... (will download files to output/)'
)
else:
textutils.output_info(
' - Svn Plugin: /.svn/entries found! crawling... (use -a to download files instead of printing)'
)
# parse entries
parse_svn_entries(conf.target_base_path)
# Clean up display
if conf.allow_download:
textutils.output_info('')
else:
textutils.output_info(' - Svn Plugin: no /.svn/entries found')
def execute():
""" Generate common simple paths (a-z, 0-9) """
path_added = 0
file_added = 0
if "skipAlpha" not in plugin_settings:
for char in range(ord('a'), ord('z') + 1):
add_generated_path(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
if "skipNumeric" not in plugin_settings:
for char in range(ord('0'), ord('9') + 1):
add_generated_path(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
if "skipYear" not in plugin_settings:
for year in range(1990, date.today().year + 5):
add_generated_path(str(year))
path_added += 1
textutils.output_info(' - PathGenerator Plugin: added ' + str(path_added) +
' computer generated path.')
textutils.output_info(' - PathGenerator Plugin: added ' + str(file_added) +
' computer generated files.')
def sample_root_404():
""" Get the root 404, this has to be done as soon as possible since plugins could use this information. """
manager = ThreadManager()
textutils.output_info('Benchmarking root 404')
for ext in conf.crafted_404_extensions:
random_file = str(uuid.uuid4())
path = conf.path_template.copy()
if path['url'] != '/':
path['url'] = '/' + random_file + ext
else:
path['url'] = random_file + ext
# Were not using the fetch cache for 404 sampling
database.fetch_queue.put(path)
# Forced bogus path check
random_file = str(uuid.uuid4())
path = conf.path_template.copy()
path['url'] = '/' + random_file + '/'
# Were not using the fetch cache for 404 sampling
database.fetch_queue.put(path)
workers = manager.spawn_workers(len(conf.crafted_404_extensions), FetchCrafted404Worker)
manager.wait_for_idle(workers, database.fetch_queue)
def load_execute_host_plugins():
""" Import and run host plugins """
textutils.output_info('Executing ' + str(len(host.__all__)) + ' host plugins')
for plugin_name in host.__all__:
plugin = __import__ ("plugins.host." + plugin_name, fromlist=[plugin_name])
if hasattr(plugin , 'execute'):
plugin.execute()
def execute():
""" Generate common simple paths (a-z, 0-9) """
path_added = 0
file_added = 0
if "skipAlpha" not in plugin_settings:
for char in range(ord('a'), ord('z')+1):
add_generated_path(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
if "skipNumeric" not in plugin_settings:
for char in range(ord('0'), ord('9')+1):
add_generated_path(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
if "skipYear" not in plugin_settings:
for year in range(1990, date.today().year + 5):
add_generated_path(str(year))
path_added += 1
textutils.output_info(' - PathGenerator Plugin: added ' + str(path_added) + ' computer generated path.')
textutils.output_info(' - PathGenerator Plugin: added ' + str(file_added) + ' computer generated files.')
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not self.kill_received and not queue.empty():
try:
sleep(0.1)
except KeyboardInterrupt:
try:
stats.output_stats()
sleep(1)
except KeyboardInterrupt:
textutils.output_info('Keyboard Interrupt Received, cleaning up threads')
# Clean reference to sockets
database.connection_pool = None
self.kill_received = True
# Kill remaining workers but don't join the queue (we want to abort:))
for worker in workers:
worker.kill_received = True
if worker is not None and worker.isAlive():
worker.join(1)
# Kill the soft
sys.exit()
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " + str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
for worker in workers:
worker.kill_received = True
worker.join()
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not database.kill_received and not queue.empty():
try:
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " +
str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
except KeyboardInterrupt:
try:
stats.output_stats(workers)
sleep(
1
) # The time you have to re-press ctrl+c to kill the app.
except KeyboardInterrupt:
textutils.output_info(
'Keyboard Interrupt Received, waiting for blocking threads to exit'
)
# Clean reference to sockets
database.connection_pool = None
database.kill_received = True
self.kill_workers(workers)
sys.exit(0)
# Make sure we get all the worker's results before continuing the next step
self.kill_workers(workers)
def output_stats():
elapsed_time = datetime.now() - database.scan_start_time
if not elapsed_time.seconds:
request_per_seconds = 0
else:
request_per_seconds = database.successful_fetch_count / elapsed_time.seconds
if request_per_seconds:
remaining_seconds = int(database.fetch_queue.qsize() /
request_per_seconds)
remaining_timedelta = timedelta(seconds=remaining_seconds)
else:
remaining_seconds = 0
remaining_timedelta = timedelta(seconds=remaining_seconds)
stats_string = ''.join([
str(request_per_seconds), ' reqs/sec', ', Done: ',
str(database.successful_fetch_count), ', Queued: ',
str(database.fetch_queue.qsize()), ', Timeouts: ',
str(database.total_timeouts), ' (~',
str(database.latest_successful_request_time), 's)', ', remaining: ',
str(remaining_timedelta), ' (hit ctrl+c again to exit)'
])
textutils.output_info(stats_string)
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not self.kill_received and not queue.empty():
try:
sleep(0.1)
except KeyboardInterrupt:
try:
stats.output_stats()
sleep(1)
except KeyboardInterrupt:
textutils.output_info(
'Keyboard Interrupt Received, cleaning up threads')
self.kill_received = True
# Kill remaining workers but don't join the queue (we want to abort:))
for worker in workers:
worker.kill_received = True
if worker is not None and worker.isAlive():
worker.join(1)
# Kill the soft
sys.exit()
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " +
str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
for worker in workers:
worker.kill_received = True
worker.join()
def test_paths_exists():
"""
Test for path existence using http codes and computed 404
Spawn workers and turn off output for now, it would be irrelevant at this point.
"""
manager = ThreadManager()
# Fill work queue with fetch list
for path in database.paths:
dbutils.add_path_to_fetch_queue(path)
# Consider some file target as potential path
for file in database.files:
if not file.get('no_suffix'):
file_as_path = file.copy()
file_as_path['url'] = '/' + file_as_path['url']
dbutils.add_path_to_fetch_queue(file_as_path)
done_paths = []
recursion_depth = 0
textutils.output_debug('Cached: ' + str(database.path_cache))
while database.fetch_queue.qsize() > 0:
textutils.output_info('Probing ' + str(database.fetch_queue.qsize()) +
' paths')
# Wait for initial valid path lookup
workers = manager.spawn_workers(conf.thread_count,
TestPathExistsWorker)
manager.wait_for_idle(workers, database.fetch_queue)
recursion_depth += 1
if not conf.recursive:
break
if recursion_depth >= conf.recursive_depth_limit:
break
for validpath in database.valid_paths:
if validpath['url'] == '/' or validpath['url'] in done_paths:
continue
done_paths.append(validpath['url'])
for path in database.paths:
if path['url'] in ('/', ''):
continue
path = path.copy()
path['url'] = validpath['url'] + path['url']
dbutils.add_path_to_fetch_queue(path)
textutils.output_info('Found ' + str(len(database.valid_paths)) +
' valid paths')
def parse_svn_entries(url):
description_file = 'SVN entries file at'
description_dir = "SVN entries Dir at"
target_url = url + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url,
conf.user_agent,
conf.fetch_timeout_secs,
limit_len=False,
add_headers=base_headers)
if response_code in conf.expected_file_responses and content:
tokens = content.decode().split('\n')
if 'dir' in tokens:
for pos, token in enumerate(tokens):
if token == 'dir':
# Fetch more entries recursively
if tokens[pos - 1] != '':
textutils.output_debug(' - Svn Plugin: Found dir: ' +
url + '/' + tokens[pos - 1])
if conf.allow_download:
textutils.output_info(
' - Svn Plugin: Downloading: ' + url + '/' +
tokens[pos - 1] + '\r')
else:
textutils.output_found(description_dir + ' at: ' +
url + '/' + tokens[pos - 1])
# Parse next
parse_svn_entries(url + "/" + tokens[pos - 1])
elif token == 'file':
textutils.output_debug(' - Svn Plugin: Found file: ' +
url + '/' + tokens[pos - 1])
if conf.allow_download:
textutils.output_info(' - Svn Plugin: Downloading: ' +
url + '/' + tokens[pos - 1] +
'\r')
# Fetch text-base file
path = url + "/.svn/text-base" + '/' + tokens[
pos - 1] + ".svn-base"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
path,
conf.user_agent,
conf.fetch_timeout_secs,
limit_len=False)
save_file(url + '/' + tokens[pos - 1], content)
else:
textutils.output_found(description_file + ' at: ' +
url + '/' + tokens[pos - 1])
def test_paths_exists():
"""
Test for path existence using http codes and computed 404
Spawn workers and turn off output for now, it would be irrelevant at this point.
"""
manager = ThreadManager()
# Fill work queue with fetch list
for path in database.paths:
dbutils.add_path_to_fetch_queue(path)
# Consider some file target as potential path
for file in database.files:
if not file.get('no_suffix'):
file_as_path = file.copy()
file_as_path['url'] = '/' + file_as_path['url']
dbutils.add_path_to_fetch_queue(file_as_path)
done_paths = []
recursion_depth = 0
textutils.output_debug('Cached: ' + str(database.path_cache))
while database.fetch_queue.qsize() > 0:
textutils.output_info('Probing ' + str(database.fetch_queue.qsize()) + ' paths')
# Wait for initial valid path lookup
workers = manager.spawn_workers(conf.thread_count, TestPathExistsWorker)
manager.wait_for_idle(workers, database.fetch_queue)
recursion_depth += 1
if not conf.recursive:
break
if recursion_depth >= conf.recursive_depth_limit:
break
for validpath in database.valid_paths:
if validpath['url'] == '/' or validpath['url'] in done_paths:
continue
done_paths.append(validpath['url'])
for path in database.paths:
if path['url'] in ('/', ''):
continue
path = path.copy()
path['url'] = validpath['url'] + path['url']
dbutils.add_path_to_fetch_queue(path)
textutils.output_info('Found ' + str(len(database.valid_paths)) + ' valid paths')
def get_session_cookies():
""" Fetch initial session cookies """
textutils.output_info('Fetching session cookie')
path = conf.path_template.copy()
path['url'] = '/'
# Were not using the fetch cache for session cookie sampling
fetcher = Fetcher()
code, content, headers = fetcher.fetch_url('/', conf.user_agent, 10)
if code is 200:
cookies = headers.get('Set-Cookie')
if cookies:
database.session_cookie = cookies
def execute():
""" Fetch /.svn/entries and parse for target paths """
current_template = dict(conf.path_template)
current_template['description'] = '/.svn/entries found directory'
target_url = urljoin(conf.target_base_path, "/.svn/entries")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
if response_code is 200 or response_code is 302 and content:
added = 0
try:
tree = ElementTree.fromstring(content)
entry_tags = tree.iter()
if entry_tags:
for entry in entry_tags:
kind = entry.attrib.get("kind")
if kind and kind == "dir":
current_template = current_template.copy()
current_template['url'] = '/' + entry.attrib["name"]
database.paths.append(current_template)
added += 1
except Exception:
textutils.output_info(' - Svn Plugin: no usable entries in /.svn/entries')
else:
if added > 0:
textutils.output_info(' - Svn Plugin: added ' + str(added) + ' base paths using /.svn/entries')
else :
textutils.output_info(' - Svn Plugin: no usable entries in /.svn/entries')
else:
textutils.output_info(' - Svn Plugin: no /.svn/entries found')
def get_session_cookies():
""" Fetch initial session cookies """
textutils.output_info('Fetching session cookie')
path = conf.path_template.copy()
path['url'] = '/'
# Were not using the fetch cache for session cookie sampling
fetcher = Fetcher()
code, content, headers = fetcher.fetch_url('/', conf.user_agent, 10)
if code is 200:
cookies = headers.get('Set-Cookie')
if cookies:
database.session_cookie = cookies
def execute():
""" Fetch /.svn/entries and parse for target paths """
textutils.output_info(' - Svn Plugin: Searching for /.svn/entries')
target_url = conf.target_base_path + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
svn_legacy = True
if response_code in conf.expected_file_responses and content:
if conf.allow_download:
textutils.output_info(
' - Svn Plugin: /.svn/entries found! crawling... (will download files to output/)'
)
else:
textutils.output_info(
' - Svn Plugin: /.svn/entries found! crawling... (use -a to download files instead of printing)'
)
# test for version 1.7+
target_url = conf.target_base_path + "/.svn/wc.db"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url,
conf.user_agent,
conf.fetch_timeout_secs,
limit_len=False)
#if response_code in conf.expected_file_responses and content:
# textutils.output_info(' - Svn Plugin: SVN 1.7+ detected, parsing wc.db')
# svn_legacy = False
# save_file(conf.target_base_path + '/wc.db', content)
# Process index
if svn_legacy:
# parse entries
parse_svn_entries(conf.target_base_path)
else:
parse_svn_17_db(conf.target_base_path + '/wc.db')
# Clean up display
if conf.allow_download:
textutils.output_info('')
else:
textutils.output_info(' - Svn Plugin: no /.svn/entries found')
def load_cookie_file(afile):
"""
Loads the supplied cookie file
"""
if not afile:
return None
try:
with open(afile, 'r') as cookie_file:
content = cookie_file.read()
content = content.replace('Cookie: ', '')
content = content.replace('\n', '')
return content
except IOError:
textutils.output_info('Supplied cookie file not found, will use server provided cookies')
return None
def output_stats():
lock = Lock()
lock.acquire()
average_timeouts = database.timeouts / database.item_count
estimated_future_timeouts = average_timeouts * database.fetch_queue.qsize()
estimated_total_remaining = int(estimated_future_timeouts + database.fetch_queue.qsize())
total_requests = database.item_count + database.timeouts
elapsed_time = datetime.now() - database.scan_start_time
request_per_seconds = elapsed_time / total_requests
remaining = request_per_seconds * estimated_total_remaining
textutils.output_info(str(total_requests / elapsed_time.seconds) + ' reqs/sec' + ', Done: ' + str(database.item_count) + ', Queued: ' + str(database.fetch_queue.qsize()) + ', Timeouts: ' +
str(database.timeouts) + ', throttle: ' + str(database.throttle_delay) + "s, remaining: " + str(remaining)[:-7] + " (press ctrl+c again to exit)")
lock.release()
def load_cookie_file(afile):
"""
Loads the supplied cookie file
"""
if not afile:
return None
try:
with open(afile, 'r') as cookie_file:
content = cookie_file.read()
content = content.replace('Cookie: ', '')
content = content.replace('\n', '')
return content
except IOError:
textutils.output_info(
'Supplied cookie file not found, will use server provided cookies')
return None
def execute():
""" Fetch /.svn/entries and parse for target paths """
textutils.output_info(' - Svn Plugin: Searching for /.svn/entries')
target_url = conf.target_base_path + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
svn_legacy = True
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
if response_code in conf.expected_file_responses and content:
if conf.allow_download:
textutils.output_info(' - Svn Plugin: /.svn/entries found! crawling... (will download files to output/)')
else:
textutils.output_info(' - Svn Plugin: /.svn/entries found! crawling... (use -a to download files instead of printing)')
# test for version 1.7+
target_url = conf.target_base_path + "/.svn/wc.db"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
#if response_code in conf.expected_file_responses and content:
# textutils.output_info(' - Svn Plugin: SVN 1.7+ detected, parsing wc.db')
# svn_legacy = False
# save_file(conf.target_base_path + '/wc.db', content)
# Process index
if svn_legacy:
# parse entries
parse_svn_entries(conf.target_base_path)
#else:
# parse_svn_17_db(conf.target_base_path + '/wc.db')
# Clean up display
if conf.allow_download:
textutils.output_info('')
else:
textutils.output_info(' - Svn Plugin: no /.svn/entries found')
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not database.kill_received and not queue.empty():
try:
sleep(0.1)
except KeyboardInterrupt:
try:
stats.output_stats()
sleep(1)
except KeyboardInterrupt:
textutils.output_info(
'Keyboard Interrupt Received, cleaning up threads')
# Clean reference to sockets
database.connection_pool = None
database.kill_received = True
# Kill remaining workers but don't join the queue (we want to abort:))
for worker in workers:
if worker is not None and worker.isAlive():
worker.kill_received = True
worker.join(0)
# Set leftover done in cas of a kill.
while not queue.empty():
queue.get()
queue.task_done()
break
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " +
str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
# Make sure we get all the worker's results before continuing the next step
for worker in workers:
if worker is not None and worker.isAlive():
worker.kill_received = True
worker.join()
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not database.kill_received and not queue.empty():
try:
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " + str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
except KeyboardInterrupt:
try:
stats.output_stats(workers)
sleep(1) # The time you have to re-press ctrl+c to kill the app.
except KeyboardInterrupt:
textutils.output_info('Keyboard Interrupt Received, waiting for blocking threads to exit')
# Clean reference to sockets
database.connection_pool = None
database.kill_received = True
self.kill_workers(workers)
sys.exit(0)
# Make sure we get all the worker's results before continuing the next step
self.kill_workers(workers)
def parse_svn_entries(url):
description_file = "SVN entries file at"
description_dir = "SVN entries Dir at"
target_url = url + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False
)
if response_code is 200 or response_code is 302 and content:
tokens = content.split("\n")
if "dir" in tokens:
for pos, token in enumerate(tokens):
if token == "dir":
# Fetch more entries recursively
if tokens[pos - 1] != "":
textutils.output_debug(" - Svn Plugin: Found dir: " + url + "/" + tokens[pos - 1])
if conf.allow_download:
textutils.output_info(" - Svn Plugin: Downloading: " + url + "/" + tokens[pos - 1] + "\r")
else:
textutils.output_found(description_dir + " at: " + url + "/" + tokens[pos - 1])
# Parse next
parse_svn_entries(url + "/" + tokens[pos - 1])
elif token == "file":
textutils.output_debug(" - Svn Plugin: Found file: " + url + "/" + tokens[pos - 1])
if conf.allow_download:
textutils.output_info(" - Svn Plugin: Downloading: " + url + "/" + tokens[pos - 1] + "\r")
# Fetch text-base file
path = url + "/.svn/text-base" + "/" + tokens[pos - 1] + ".svn-base"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
path, conf.user_agent, conf.fetch_timeout_secs, limit_len=False
)
save_file(url + "/" + tokens[pos - 1], content)
else:
textutils.output_found(description_file + " at: " + url + "/" + tokens[pos - 1])
def parse_svn_entries(url):
description_file = 'SVN entries file at'
description_dir = "SVN entries Dir at"
target_url = url + "/.svn/entries"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False, add_headers=base_headers)
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
if response_code in conf.expected_file_responses and content:
tokens = content.split('\n')
if 'dir' in tokens:
for pos, token in enumerate(tokens):
if token == 'dir':
# Fetch more entries recursively
if tokens[pos-1] != '':
textutils.output_debug(' - Svn Plugin: Found dir: ' + url + '/' + tokens[pos-1])
if conf.allow_download:
textutils.output_info(' - Svn Plugin: Downloading: ' + url + '/' + tokens[pos-1] + '\r')
else:
textutils.output_found(description_dir + ' at: ' + url + '/' + tokens[pos-1])
# Parse next
parse_svn_entries(url + "/" + tokens[pos-1])
elif token == 'file':
textutils.output_debug(' - Svn Plugin: Found file: ' + url + '/' + tokens[pos-1])
if conf.allow_download:
textutils.output_info(' - Svn Plugin: Downloading: ' + url + '/' + tokens[pos-1] + '\r')
# Fetch text-base file
path = url + "/.svn/text-base" + '/' + tokens[pos-1] + ".svn-base"
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(path, conf.user_agent,
conf.fetch_timeout_secs, limit_len=False)
save_file(url + '/' + tokens[pos-1], content)
else:
textutils.output_found(description_file + ' at: ' + url + '/' + tokens[pos-1])
def output_stats():
lock = Lock()
lock.acquire()
average_timeouts = database.timeouts / database.item_count
estimated_future_timeouts = average_timeouts * database.fetch_queue.qsize()
estimated_total_remaining = int(estimated_future_timeouts +
database.fetch_queue.qsize())
total_requests = database.item_count + database.timeouts
elapsed_time = datetime.now() - database.scan_start_time
request_per_seconds = elapsed_time / total_requests
remaining = request_per_seconds * estimated_total_remaining
textutils.output_info(
str(total_requests / elapsed_time.seconds) + ' reqs/sec' + ', Done: ' +
str(database.item_count) + ', Queued: ' +
str(database.fetch_queue.qsize()) + ', Timeouts: ' +
str(database.timeouts) + ', throttle: ' +
str(database.throttle_delay) + "s, remaining: " + str(remaining)[:-7] +
" (press ctrl+c again to exit)")
lock.release()
def wait_for_idle(self, workers, queue):
""" Wait until fetch queue is empty and handle user interrupt """
while not database.kill_received and not queue.empty():
try:
sleep(0.1)
except KeyboardInterrupt:
try:
stats.output_stats()
sleep(1)
except KeyboardInterrupt:
textutils.output_info('Keyboard Interrupt Received, cleaning up threads')
# Clean reference to sockets
database.connection_pool = None
database.kill_received = True
# Kill remaining workers but don't join the queue (we want to abort:))
for worker in workers:
if worker is not None and worker.isAlive():
worker.kill_received = True
worker.join(0)
# Set leftover done in cas of a kill.
while not queue.empty():
queue.get()
queue.task_done()
break
# Make sure everything is done before sending control back to application
textutils.output_debug("Threads: joining queue of size: " + str(queue.qsize()))
queue.join()
textutils.output_debug("Threads: join done")
# Make sure we get all the worker's results before continuing the next step
for worker in workers:
if worker is not None and worker.isAlive():
worker.kill_received = True
worker.join()
def execute():
""" Generate common simple paths (a-z, 0-9) """
path_added = 0
file_added = 0
for char in range(ord('a'), ord('z')):
add_generated_dir(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
for char in range(ord('0'), ord('9')):
add_generated_dir(chr(char))
path_added += 1
add_generated_file(chr(char))
file_added += 1
for year in range(1990, date.today().year + 5):
add_generated_dir(str(year))
path_added += 1
textutils.output_info(' - PathGenerator Plugin: added ' + str(path_added) + ' computer generated path.')
textutils.output_info(' - PathGenerator Plugin: added ' + str(file_added) + ' computer generated files.')
def execute():
""" Fetch sitemap.xml and add each entry as a target """
current_template = dict(conf.path_template)
current_template['description'] = 'sitemap.xml entry'
target_url = urljoin(conf.target_base_path, "/sitemap.xml")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url,
conf.user_agent,
conf.fetch_timeout_secs,
limit_len=False,
add_headers={})
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
if response_code is 200 or response_code is 302 and content:
regexp = re.compile('(?im).*<url>\s*<loc>(.*)</loc>\s*</url>.*')
matches = re.findall(regexp, content)
textutils.output_debug("SitemapXML plugin")
added = 0
for match in matches:
if not isinstance(match, str):
match = match.decode('utf-8', 'ignore')
parsed = urlparse(match)
if parsed.path:
new_path = parsed.path
else:
continue
# Remove trailing /
if new_path.endswith('/'):
new_path = new_path[:-1]
if add_path(new_path):
added += 1
textutils.output_debug(" - Added: %s from /sitemap.xml" % new_path)
if added > 0:
textutils.output_info(' - SitemapXML Plugin: added %d base paths '
'using /sitemap.xml' % added)
else:
textutils.output_info(' - SitemapXML Plugin: no usable entries '
'in /sitemap.xml')
else:
textutils.output_info(
' - SitemapXML Plugin: /sitemap.xml not found on '
'target site')
def execute():
""" Fetch /robots.txt and add the disallowed paths as target """
current_template = dict(conf.path_template)
current_template['description'] = 'Robots.txt entry'
target_url = urljoin(conf.target_base_path, "/robots.txt")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
if isinstance(content, str):
content = content.encode('utf-8')
if response_code is 200 or response_code is 302 and content:
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
matches = re.findall(r'Disallow:\s*/[a-zA-Z0-9-/\r]+\n', content)
textutils.output_debug(content)
added = 0
for match in matches:
# Filter out some characters
match = filter(lambda c: c not in ' *?.\n\r\t', match)
if match:
match = ''.join(match)
# Split on ':'
splitted = match.split(':')
if splitted[1]:
target_path = splitted[1]
textutils.output_debug(target_path)
# Remove trailing /
if target_path.endswith('/'):
target_path = target_path[:-1]
current_template = current_template.copy()
current_template['url'] = target_path
database.paths.append(current_template)
textutils.output_debug(' - Robots Plugin Added: ' +
str(target_path) + ' from robots.txt')
added += 1
if added > 0:
textutils.output_info(' - Robots Plugin: added ' + str(added) +
' base paths using /robots.txt')
else:
textutils.output_info(
' - Robots Plugin: no usable entries in /robots.txt')
else:
textutils.output_info(
' - Robots Plugin: /robots.txt not found on target site')
def execute():
""" Fetch sitemap.xml and add each entry as a target """
current_template = dict(conf.path_template)
current_template['description'] = 'sitemap.xml entry'
target_url = urljoin(conf.target_base_path, "/sitemap.xml")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url,
conf.user_agent,
conf.fetch_timeout_secs,
limit_len=False,
add_headers={}
)
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
if response_code is 200 or response_code is 302 and content:
regexp = re.compile('(?im).*<url>\s*<loc>(.*)</loc>\s*</url>.*')
matches = re.findall(regexp, content)
textutils.output_debug("SitemapXML plugin")
added = 0
for match in matches:
new_path = match.decode().split(conf.target_host)[1]
# Remove trailing /
if new_path.endswith('/'):
new_path = new_path[:-1]
add_path(new_path)
add_file(new_path)
textutils.output_debug(" - Added: %s from /sitemap.xml" % new_path)
added += 1
if added > 0:
textutils.output_info(' - SitemapXML Plugin: added %d base paths '
'using /sitemap.xml' % added)
else :
textutils.output_info(' - SitemapXML Plugin: no usable entries '
'in /sitemap.xml')
else:
textutils.output_info(' - SitemapXML Plugin: /sitemap.xml not found on '
'target site')
def execute():
""" Fetch /robots.txt and add the disallowed paths as target """
current_template = dict(conf.path_template)
current_template['description'] = 'Robots.txt entry'
target_url = urljoin(conf.target_base_path, "/robots.txt")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
if isinstance(content, str):
content = content.encode('utf-8')
if response_code is 200 or response_code is 302 and content:
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
matches = re.findall(r'Disallow:\s*/[a-zA-Z0-9-/\r]+\n', content)
textutils.output_debug(content)
added = 0
for match in matches:
# Filter out some characters
match = filter(lambda c: c not in ' *?.\n\r\t', match)
if match:
match = ''.join(match)
# Split on ':'
splitted = match.split(':')
if splitted[1]:
target_path = splitted[1]
textutils.output_debug(target_path)
# Remove trailing /
if target_path.endswith('/'):
target_path = target_path[:-1]
current_template = current_template.copy()
current_template['url'] = target_path
database.paths.append(current_template)
textutils.output_debug(' - Robots Plugin Added: ' + str(target_path) + ' from robots.txt')
added += 1
if added > 0:
textutils.output_info(' - Robots Plugin: added ' + str(added) + ' base paths using /robots.txt')
else :
textutils.output_info(' - Robots Plugin: no usable entries in /robots.txt')
else:
textutils.output_info(' - Robots Plugin: /robots.txt not found on target site')
def execute():
""" Fetch /.svn/entries and parse for target paths """
current_template = dict(conf.path_template)
current_template['description'] = '/.svn/entries found directory'
target_url = urljoin(conf.target_base_path, "/.svn/entries")
fetcher = Fetcher()
response_code, content, headers = fetcher.fetch_url(
target_url, conf.user_agent, conf.fetch_timeout_secs, limit_len=False)
if response_code is 200 or response_code is 302 and content:
added = 0
try:
tree = ElementTree.fromstring(content)
entry_tags = tree.iter()
if entry_tags:
for entry in entry_tags:
kind = entry.attrib.get("kind")
if kind and kind == "dir":
current_template = current_template.copy()
current_template['url'] = '/' + entry.attrib["name"]
database.paths.append(current_template)
added += 1
except Exception:
textutils.output_info(
' - Svn Plugin: no usable entries in /.svn/entries')
else:
if added > 0:
textutils.output_info(' - Svn Plugin: added ' + str(added) +
' base paths using /.svn/entries')
else:
textutils.output_info(
' - Svn Plugin: no usable entries in /.svn/entries')
else:
textutils.output_info(' - Svn Plugin: no /.svn/entries found')
def load_target_paths():
""" Load the target paths in the database """
textutils.output_info('Loading target paths')
# Add files
database.paths += loaders.load_targets('data/path.lst')
conf.target_base_path = parsed_path
conf.is_ssl = is_ssl
textutils.output_debug('Version: ' + str(conf.version))
textutils.output_debug('Max timeouts per url: ' +
str(conf.max_timeout_count))
textutils.output_debug('Worker threads: ' + str(conf.thread_count))
textutils.output_debug('Target Host: ' + str(conf.target_host))
textutils.output_debug('Using Tor: ' + str(conf.use_tor))
textutils.output_debug('Eval-able output: ' + str(conf.eval_output))
textutils.output_debug('Using User-Agent: ' + str(conf.user_agent))
textutils.output_debug('Search only for files: ' + str(conf.files_only))
textutils.output_debug('Search only for subdirs: ' +
str(conf.directories_only))
textutils.output_info('Starting Discovery on ' + conf.target_host)
if conf.use_tor:
textutils.output_info('Using Tor, be patient it WILL be slow!')
textutils.output_info(
'Max timeout count and url fetch timeout doubled for the occasion ;)'
)
conf.max_timeout_count *= 2
conf.fetch_timeout_secs *= 2
# Handle keyboard exit before multi-thread operations
try:
# Resolve target host to avoid multiple dns lookups
resolved, port = dnscache.get_host_ip(parsed_host, parsed_port)
# Benchmark target host
def run(self):
while not self.kill_received:
try:
queued = database.fetch_queue.get(block=False)
url = conf.target_base_path + queued.get('url')
description = queued.get('description')
textutils.output_debug("Testing directory: " + url + " " + str(queued))
stats.update_stats(url)
# Add trailing / for paths
if not url.endswith('/') and url != '/':
url += '/'
# Fetch directory
start_time = datetime.now()
response_code, content, headers = self.fetcher.fetch_url(url, conf.user_agent, database.latest_successful_request_time, limit_len=False)
end_time = datetime.now()
# Fetch '/' but don't submit it to more logging/existance tests
if queued.get('url') == '/':
if queued not in database.valid_paths:
database.valid_paths.append(queued)
database.fetch_queue.task_done()
continue
if response_code == 500:
textutils.output_debug("HIT 500 on: " + str(queued))
# handle timeout
if response_code in conf.timeout_codes:
handle_timeout(queued, url, self.thread_id, output=self.output)
elif response_code == 404 and detect_tomcat_fake_404(content):
database.valid_paths.append(queued)
textutils.output_found('Tomcat redirect, ' + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"special": "tomcat-redirect",
"severity": queued.get('severity'),
})
elif response_code in conf.expected_path_responses:
# Compare content with generated 404 samples
is_valid_result = test_valid_result(content)
if is_valid_result:
# Test if behavior is ok.
normal_behavior = test_behavior(content)
else:
# We don't compute behavior on invalid results
normal_behavior = True
if normal_behavior and database.behavior_error:
textutils.output_info('Normal behavior seems to be restored.')
database.behavior_error = False
if is_valid_result and not normal_behavior:
# We don't declare a behavior change until the current hit has exceeded the maximum
# chances it can get.
if not database.behavior_error and queued.get('behavior_chances', 0) >= conf.max_behavior_tries:
textutils.output_info('Behavior change detected! Results may '
'be incomplete or tachyon may never exit.')
textutils.output_debug('Chances taken: ' + str(queued.get('behavior_chances', 0)))
textutils.output_debug(queued.get('url'))
database.behavior_error = True
# If we find a valid result but the behavior buffer is not full, we give a chance to the
# url and increase it's chances count. We consider this a false behavior test.
# We do this since an incomplete behavior buffer could give false positives
# Additionally, if the fetch queue is empty and we're still not in global behavior error, we
# consider all the remaining hits as valid, as they are hits that were given a chance.
if is_valid_result and len(database.behavioral_buffer) < conf.behavior_queue_size \
and not database.behavior_error and database.fetch_queue.qsize() != 0:
if not queued.get('behavior_chances'):
queued['behavior_chances'] = 1
else:
queued['behavior_chances'] += 1
if queued['behavior_chances'] < conf.max_behavior_tries:
textutils.output_debug('Time for a chance')
textutils.output_debug('Chance left to target ' + queued.get('url') + ', re-queuing ' +
' qsize: ' + str(database.fetch_queue.qsize()) +
' chances: ' + str(queued.get('behavior_chances')))
database.fetch_queue.put(queued)
database.fetch_queue.task_done()
continue
else:
textutils.output_debug('Chances count busted! ' + queued.get('url') +
' qsize: ' + str(database.fetch_queue.qsize()))
elif response_code == 401:
# Output result, but don't keep the url since we can't poke in protected folder
textutils.output_found('Password Protected - ' + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
# At this point, we have a valid result and the behavioral buffer is full.
# The behavior of the hit has been taken in account and the app is not in global behavior error
elif is_valid_result:
# Add path to valid_path for future actions
database.valid_paths.append(queued)
# If we reach this point, all edge-cases should be handled and all subsequent requests
# should be benchmarked against this new behavior
reset_behavior_database()
if response_code == 500:
textutils.output_found('ISE, ' + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code == 403:
textutils.output_found('*Forbidden* ' + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
else:
textutils.output_found(description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code in conf.redirect_codes:
if queued.get('handle_redirect', True):
location = headers.get('location')
if location:
handle_redirects(queued, location)
# Stats
if response_code not in conf.timeout_codes:
stats.update_processed_items()
compute_request_time(start_time, end_time)
# Mark item as processed
database.fetch_queue.task_done()
except Empty:
continue
textutils.output_debug('Version: ' + str(conf.version))
textutils.output_debug('Max timeouts per url: ' + str(conf.max_timeout_count))
textutils.output_debug('Worker threads: ' + str(conf.thread_count))
textutils.output_debug('Target Host: ' + str(conf.target_host))
textutils.output_debug('Using Tor: ' + str(conf.use_tor))
textutils.output_debug('Eval-able output: ' + str(conf.eval_output))
textutils.output_debug('JSON output: ' + str(conf.json_output))
textutils.output_debug('Using User-Agent: ' + str(conf.user_agent))
textutils.output_debug('Search only for files: ' + str(conf.files_only))
textutils.output_debug('Search only for subdirs: ' + str(conf.directories_only))
if conf.proxy_url:
textutils.output_debug('Using proxy: ' + str(conf.proxy_url))
textutils.output_info('Starting Discovery on ' + conf.target_host)
if conf.use_tor:
textutils.output_info('Using Tor, be patient it WILL be slow!')
textutils.output_info('Max timeout count and url fetch timeout doubled for the occasion ;)')
conf.max_timeout_count *= 2
conf.fetch_timeout_secs *= 2
# Handle keyboard exit before multi-thread operations
print_results_worker = None
try:
# Resolve target host to avoid multiple dns lookups
if not conf.proxy_url:
resolved, port = dnscache.get_host_ip(conf.target_host, conf.target_port)
# disable urllib'3 SSL warning (globally)
def run(self):
while not self.kill_received:
try:
# Non-Blocking get since we use the queue as a ringbuffer
queued = database.fetch_queue.get(block=False)
url = conf.target_base_path + queued.get('url')
description = queued.get('description')
match_string = queued.get('match_string')
textutils.output_debug("Testing: " + url + " " + str(queued))
stats.update_stats(url)
# Fetch the target url
start_time = datetime.now()
if match_string:
response_code, content, headers = self.fetcher.fetch_url(url, conf.user_agent, database.latest_successful_request_time, limit_len=False)
# Make sure we always match string against a string content
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
else:
response_code, content, headers = self.fetcher.fetch_url(url, conf.user_agent, database.latest_successful_request_time)
end_time = datetime.now()
# handle timeout
if response_code in conf.timeout_codes:
handle_timeout(queued, url, self.thread_id, output=self.output)
elif response_code == 500:
textutils.output_found('ISE, ' + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code in conf.expected_file_responses:
# Test if result is valid
is_valid_result = test_valid_result(content, is_file=True)
if is_valid_result:
# Test if behavior is ok.
normal_behavior = test_behavior(content)
textutils.output_debug('Normal behavior ' + str(normal_behavior) + ' ' + str(response_code))
else:
normal_behavior = True
# Reset behavior chance when we detect a new state
if normal_behavior and database.behavior_error:
textutils.output_info('Normal behavior seems to be restored.')
database.behavior_error = False
if is_valid_result and not normal_behavior:
# Looks like the new behavior is now the norm. It's a false positive.
# Additionally, we report a behavior change to the user at this point.
if not database.behavior_error:
textutils.output_info('Behavior change detected! Results may '
'be incomplete or tachyon may never exit.')
textutils.output_debug('Chances taken: ' + str(queued.get('behavior_chances', 0)))
textutils.output_debug(queued.get('url'))
database.behavior_error = True
# If we find a valid result but the behavior buffer is not full, we give a chance to the
# url and increase it's chances count. We consider this a false behavior test.
# We do this since an incomplete behavior buffer could give false positives
# Additionally, if the fetch queue is empty and we're still not in global behavior error, we
# consider all the remaining hits as valid, as they are hits that were given a chance.
elif is_valid_result and len(database.behavioral_buffer) < conf.behavior_queue_size \
and not database.behavior_error and database.fetch_queue.qsize() != 0:
if not queued.get('behavior_chances'):
queued['behavior_chances'] = 1
else:
queued['behavior_chances'] += 1
if queued['behavior_chances'] < conf.max_behavior_tries:
textutils.output_debug('Chance left to target, re-queuing')
database.fetch_queue.put(queued)
elif is_valid_result:
# Make sure we base our next analysis on that positive hit
reset_behavior_database()
if len(content) == 0:
textutils.output_found('Empty ' + description + ' at: ' + conf.base_url + url, {
"description": "Empty " + description,
"url": conf.base_url + url,
"code": response_code,
"severity": 'info',
})
else:
textutils.output_found(description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif match_string and re.search(re.escape(match_string), content, re.I):
textutils.output_found("String-Matched " + description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"string": match_string,
"severity": queued.get('severity'),
})
elif response_code in conf.redirect_codes:
if queued.get('handle_redirect', True):
location = headers.get('location')
if location:
handle_redirects(queued, location)
# Stats
if response_code not in conf.timeout_codes:
stats.update_processed_items()
compute_request_time(start_time, end_time)
# Mark item as processed
database.fetch_queue.task_done()
except Empty:
continue
def execute():
""" This plugin process the hostname to generate host and filenames relatives to it """
target = conf.target_host
# Remove char to figure out the human-likely expressed domain name
# host.host.host.com = hosthosthost.com. host.com hostcom, host, /host.ext
# We don't test for domain.dom/domain since "cp * ./sitename" is unlikely to happen (questionable)
added = 0
# http://oksala.org -> oksala.org
target = target.replace('http://', '')
target = target.replace('https://', '')
# Remove subpath
first_slash = target.find('/')
if first_slash > 0:
target = target[0:first_slash]
target = target.replace('/', '')
new_target = conf.path_template.copy()
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# www.oksala.org -> oksala.org
target = target.replace('www.', '')
new_target = conf.path_template.copy()
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# oksala.org -> oksala
dom_pos = target.rfind('.')
nodom_target = target[0:dom_pos]
new_target = conf.path_template.copy()
new_target['url'] = nodom_target
new_target['description'] = "HostProcessor generated filename"
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# shortdom (blabla.ok.ok.test.com -> test)
new_target = conf.path_template.copy()
dom_pos = target.rfind('.')
if dom_pos > 0:
nodom_target = target[0:dom_pos]
start_pos = nodom_target.rfind('.')
if start_pos > 0:
short_dom = nodom_target[start_pos + 1:]
else:
short_dom = nodom_target
new_target['url'] = short_dom
new_target['description'] = "HostProcessor generated filename"
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " +
str(new_target))
added += 1
new_target = new_target.copy()
new_target['url'] = short_dom + 'admin'
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " +
str(new_target))
added += 1
new_target = new_target.copy()
new_target['url'] = short_dom + '-admin'
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " +
str(new_target))
added += 1
# flatten subdomains
target = target.replace('.', '')
new_target = conf.path_template.copy()
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
textutils.output_info(" - HostProcessor Plugin: added " + str(added) +
" new filenames")
def run(self):
while not self.kill_received:
try:
queued = database.fetch_queue.get(block=False)
url = conf.target_base_path + queued.get('url')
description = queued.get('description')
textutils.output_debug("Testing directory: " + url + " " +
str(queued))
stats.update_stats(url)
# Add trailing / for paths
if not url.endswith('/') and url != '/':
url += '/'
# Fetch directory
start_time = datetime.now()
response_code, content, headers = self.fetcher.fetch_url(
url,
conf.user_agent,
database.latest_successful_request_time,
limit_len=False)
end_time = datetime.now()
# Fetch '/' but don't submit it to more logging/existance tests
if queued.get('url') == '/':
if queued not in database.valid_paths:
database.valid_paths.append(queued)
database.fetch_queue.task_done()
continue
if response_code == 500:
textutils.output_debug("HIT 500 on: " + str(queued))
# handle timeout
if response_code in conf.timeout_codes:
handle_timeout(queued,
url,
self.thread_id,
output=self.output)
elif response_code == 404 and detect_tomcat_fake_404(content):
database.valid_paths.append(queued)
textutils.output_found(
'Tomcat redirect, ' + description + ' at: ' +
conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"special": "tomcat-redirect",
"severity": queued.get('severity'),
})
elif response_code in conf.expected_path_responses:
# Compare content with generated 404 samples
is_valid_result = test_valid_result(content)
if is_valid_result:
# Test if behavior is ok.
normal_behavior = test_behavior(content)
else:
# We don't compute behavior on invalid results
normal_behavior = True
if normal_behavior and database.behavior_error:
textutils.output_info(
'Normal behavior seems to be restored.')
database.behavior_error = False
if is_valid_result and not normal_behavior:
# We don't declare a behavior change until the current hit has exceeded the maximum
# chances it can get.
if not database.behavior_error and queued.get(
'behavior_chances',
0) >= conf.max_behavior_tries:
textutils.output_info(
'Behavior change detected! Results may '
'be incomplete or tachyon may never exit.')
textutils.output_debug(
'Chances taken: ' +
str(queued.get('behavior_chances', 0)))
textutils.output_debug(queued.get('url'))
database.behavior_error = True
# If we find a valid result but the behavior buffer is not full, we give a chance to the
# url and increase it's chances count. We consider this a false behavior test.
# We do this since an incomplete behavior buffer could give false positives
# Additionally, if the fetch queue is empty and we're still not in global behavior error, we
# consider all the remaining hits as valid, as they are hits that were given a chance.
if is_valid_result and len(database.behavioral_buffer) < conf.behavior_queue_size \
and not database.behavior_error and database.fetch_queue.qsize() != 0:
if not queued.get('behavior_chances'):
queued['behavior_chances'] = 1
else:
queued['behavior_chances'] += 1
if queued['behavior_chances'] < conf.max_behavior_tries:
textutils.output_debug('Time for a chance')
textutils.output_debug(
'Chance left to target ' + queued.get('url') +
', re-queuing ' + ' qsize: ' +
str(database.fetch_queue.qsize()) +
' chances: ' +
str(queued.get('behavior_chances')))
database.fetch_queue.put(queued)
database.fetch_queue.task_done()
continue
else:
textutils.output_debug(
'Chances count busted! ' + queued.get('url') +
' qsize: ' + str(database.fetch_queue.qsize()))
elif response_code == 401:
# Output result, but don't keep the url since we can't poke in protected folder
textutils.output_found(
'Password Protected - ' + description + ' at: ' +
conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
# At this point, we have a valid result and the behavioral buffer is full.
# The behavior of the hit has been taken in account and the app is not in global behavior error
elif is_valid_result:
# Add path to valid_path for future actions
database.valid_paths.append(queued)
# If we reach this point, all edge-cases should be handled and all subsequent requests
# should be benchmarked against this new behavior
reset_behavior_database()
if response_code == 500:
textutils.output_found(
'ISE, ' + description + ' at: ' +
conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code == 403:
textutils.output_found(
'*Forbidden* ' + description + ' at: ' +
conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
else:
textutils.output_found(
description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code in conf.redirect_codes:
if queued.get('handle_redirect', True):
location = headers.get('location')
if location:
handle_redirects(queued, location)
# Stats
if response_code not in conf.timeout_codes:
stats.update_processed_items()
compute_request_time(start_time, end_time)
# Mark item as processed
database.fetch_queue.task_done()
except Empty:
continue
def execute():
""" This plugin process the hostname to generate host and filenames relatives to it """
target = conf.target_host
# Remove char to figure out the human-likely expressed domain name
# host.host.host.com = hosthosthost.com. host.com hostcom, host, /host.ext
# We don't test for domain.dom/domain since "cp * ./sitename" is unlikely to happen (questionable)
added = 0
# http://oksala.org -> oksala.org
target = target.replace('http://', '')
target = target.replace('https://', '')
# Remove subpath
first_slash = target.find('/')
if first_slash > 0:
target = target[0:first_slash]
target = target.replace('/', '')
new_target = dict(conf.path_template)
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# www.oksala.org -> oksala.org
target = target.replace('www.', '')
new_target = dict(conf.path_template)
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# oksala.org -> oksala
dom_pos = target.rfind('.')
nodom_target = target[0:dom_pos]
new_target = dict(conf.path_template)
new_target['url'] = nodom_target
new_target['description'] = "HostProcessor generated filename"
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin added: " + str(new_target))
added += 1
# shortdom (blabla.ok.ok.test.com -> test)
new_target = dict(conf.path_template)
dom_pos = target.rfind('.')
if dom_pos > 0:
nodom_target = target[0:dom_pos]
start_pos = nodom_target.rfind('.')
if start_pos > 0:
short_dom = nodom_target[start_pos+1:]
else:
short_dom = nodom_target
new_target['url'] = short_dom
new_target['description'] = "HostProcessor generated filename"
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin: added " + str(new_target))
added += 1
new_target = dict(new_target)
new_target['url'] = short_dom + 'admin'
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin: added " + str(new_target))
added += 1
new_target = dict(new_target)
new_target['url'] = short_dom + '-admin'
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin: added " + str(new_target))
added += 1
# flatten subdomains
target = target.replace('.', '')
new_target = dict(conf.path_template)
new_target['url'] = target
new_target['description'] = "HostProcessor generated filename"
if new_target not in database.files:
database.files.append(new_target)
textutils.output_debug(" - HostProcessor Plugin: added " + str(new_target))
added += 1
textutils.output_info(" - HostProcessor Plugin: added " + str(added) + " new filenames")
def load_target_paths():
""" Load the target paths in the database """
textutils.output_info('Loading target paths')
database.paths += loaders.load_targets('data/path.lst')
def load_target_files():
""" Load the target files in the database """
textutils.output_info('Loading target files')
database.files += loaders.load_targets('data/file.lst')
def load_target_paths(running_path):
""" Load the target paths in the database """
textutils.output_info('Loading target paths')
database.paths += loaders.load_targets(running_path + '/data/path.lst')
def load_target_files(running_path):
""" Load the target files in the database """
textutils.output_info('Loading target files')
database.files += loaders.load_targets(running_path + '/data/file.lst')
def run(self):
while not self.kill_received:
try:
# Non-Blocking get since we use the queue as a ringbuffer
queued = database.fetch_queue.get(block=False)
url = conf.target_base_path + queued.get('url')
description = queued.get('description')
match_string = queued.get('match_string')
textutils.output_debug("Testing: " + url + " " + str(queued))
stats.update_stats(url)
# Fetch the target url
start_time = datetime.now()
if match_string:
response_code, content, headers = self.fetcher.fetch_url(
url,
conf.user_agent,
database.latest_successful_request_time,
limit_len=False)
# Make sure we always match string against a string content
if not isinstance(content, str):
content = content.decode('utf-8', 'ignore')
else:
response_code, content, headers = self.fetcher.fetch_url(
url, conf.user_agent,
database.latest_successful_request_time)
end_time = datetime.now()
# handle timeout
if response_code in conf.timeout_codes:
handle_timeout(queued,
url,
self.thread_id,
output=self.output)
elif response_code == 500:
textutils.output_found(
'ISE, ' + description + ' at: ' + conf.base_url + url,
{
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif response_code in conf.expected_file_responses:
# Test if result is valid
is_valid_result = test_valid_result(content, is_file=True)
if is_valid_result:
# Test if behavior is ok.
normal_behavior = test_behavior(content)
textutils.output_debug('Normal behavior ' +
str(normal_behavior) + ' ' +
str(response_code))
else:
normal_behavior = True
# Reset behavior chance when we detect a new state
if normal_behavior and database.behavior_error:
textutils.output_info(
'Normal behavior seems to be restored.')
database.behavior_error = False
if is_valid_result and not normal_behavior:
# Looks like the new behavior is now the norm. It's a false positive.
# Additionally, we report a behavior change to the user at this point.
if not database.behavior_error:
textutils.output_info(
'Behavior change detected! Results may '
'be incomplete or tachyon may never exit.')
textutils.output_debug(
'Chances taken: ' +
str(queued.get('behavior_chances', 0)))
textutils.output_debug(queued.get('url'))
database.behavior_error = True
# If we find a valid result but the behavior buffer is not full, we give a chance to the
# url and increase it's chances count. We consider this a false behavior test.
# We do this since an incomplete behavior buffer could give false positives
# Additionally, if the fetch queue is empty and we're still not in global behavior error, we
# consider all the remaining hits as valid, as they are hits that were given a chance.
elif is_valid_result and len(database.behavioral_buffer) < conf.behavior_queue_size \
and not database.behavior_error and database.fetch_queue.qsize() != 0:
if not queued.get('behavior_chances'):
queued['behavior_chances'] = 1
else:
queued['behavior_chances'] += 1
if queued['behavior_chances'] < conf.max_behavior_tries:
textutils.output_debug(
'Chance left to target, re-queuing')
database.fetch_queue.put(queued)
elif is_valid_result:
# Make sure we base our next analysis on that positive hit
reset_behavior_database()
if len(content) == 0:
textutils.output_found(
'Empty ' + description + ' at: ' +
conf.base_url + url, {
"description": "Empty " + description,
"url": conf.base_url + url,
"code": response_code,
"severity": 'info',
})
else:
textutils.output_found(
description + ' at: ' + conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"severity": queued.get('severity'),
})
elif match_string and re.search(re.escape(match_string),
content, re.I):
textutils.output_found(
"String-Matched " + description + ' at: ' +
conf.base_url + url, {
"description": description,
"url": conf.base_url + url,
"code": response_code,
"string": match_string,
"severity": queued.get('severity'),
})
elif response_code in conf.redirect_codes:
if queued.get('handle_redirect', True):
location = headers.get('location')
if location:
handle_redirects(queued, location)
# Stats
if response_code not in conf.timeout_codes:
stats.update_processed_items()
compute_request_time(start_time, end_time)
# Mark item as processed
database.fetch_queue.task_done()
except Empty:
continue
