def process_db_vhosts(workspace, simulation, target_list=None,dont_scan_ips=None,config_file=None):
workspace_mode = lib.db.get_workspace_mode(workspace)[0][0]
all_commands = []
output_base_dir = lib.db.get_output_dir_for_workspace(workspace)[0][0]
try:
os.stat(output_base_dir)
except:
print("[+] Output directory does not exist. Creating " + output_base_dir)
os.makedirs(output_base_dir)
unique_unscanned_vhosts = db.get_inscope_unsubmitted_vhosts(workspace)
unique_unscanned_vhosts_list = []
[unique_unscanned_vhosts_list.append(i[0]) for i in unique_unscanned_vhosts] #converts list of tuples that contains the IPs, to a list of IPs
if target_list:
for vhost in target_list:
#TODO: try this: if vhost in unique_unscanned_vhosts_list:
for unscanned_vhost in unique_unscanned_vhosts_list:
if str(vhost) == str(unscanned_vhost):
vhost_explicitly_out_of_scope = lib.db.is_vhost_explicitly_out_of_scope(str(vhost), workspace)
if not vhost_explicitly_out_of_scope:
try:
IPAddress(vhost)
if not dont_scan_ips:
command_list = populate_comamnds(vhost, workspace, simulation, output_base_dir,config_file=config_file)
except:
command_list = populate_commands_vhost_http_https_only(vhost, workspace, simulation,output_base_dir,config_file=config_file)
if len(command_list) > 0:
print("Submitted [{1}] tasks for {0}".format(unscanned_vhost, len(command_list)))
all_commands = all_commands + command_list
else:
for vhost in unique_unscanned_vhosts_list:
command_list = []
#print(vhost)
vhost_explicitly_out_of_scope = lib.db.is_vhost_explicitly_out_of_scope(vhost, workspace)
if not vhost_explicitly_out_of_scope:
try:
IPAddress(vhost)
if not dont_scan_ips:
command_list = populate_comamnds(vhost, workspace, simulation, output_base_dir,config_file=config_file)
if len(command_list) > 0:
print("Submitted [{1}] tasks for {0}".format(vhost, len(command_list)))
except:
command_list = populate_commands_vhost_http_https_only(vhost, workspace, simulation, output_base_dir,config_file=config_file)
if len(command_list) > 0:
print("Submitted [{1}] tasks for {0}".format(vhost, len(command_list)))
all_commands = all_commands + command_list
shuffle(all_commands)
for populated_command_tuple in all_commands:
#print populated_command_tuple
send_commands_to_celery(populated_command_tuple,output_base_dir,simulation)
total_tasks_num = len(all_commands)
if total_tasks_num > 0:
print("\n\n[+] Summary:\tSubmitted {0} tasks to the [{1}] workspace.".format(total_tasks_num,workspace))
print("[+]\t\tThere might be additional tasks added to the queue during post processing\n[+]")
print("[+]\t\tTo keep an eye on things, run one of these commands: \n[+]")
print("[+]\t\tcelerystalk query [watch]")
print("[+]\t\tcelerystalk query brief [watch]")
print("[+]\t\tcelerystalk query summary [watch]\n")
def process_db_services(output_base_dir,
simulation,
workspace,
target=None,
host=None,
config_file=None):
celery_path = sys.path[0]
config, supported_services = config_parser.read_config_ini(config_file)
task_id_list = []
total_tasks_num = 0
if host:
target = db.get_vhost_ip(host)
try:
os.stat(output_base_dir)
except:
print("[+] Output directory does not exist. Creating " +
output_base_dir)
os.makedirs(output_base_dir)
#unique_hosts = db.get_unique_hosts(workspace)
unique_unscanned_vhosts = db.get_inscope_unsubmitted_vhosts(workspace)
for row in unique_unscanned_vhosts:
vhost = row[0]
#print("in proccess_db_services - vhost:" + vhost)
vhost_ip = db.get_vhost_ip(vhost, workspace)[0]
#print(target)
#print(vhost_ip)
#print(str(vhost_ip))
if (IPAddress(vhost_ip[0]) == target) or (target is None):
host_dir = output_base_dir + vhost
host_data_dir = host_dir + "/celerystalkOutput/"
# Creates something like /pentest/10.0.0.1, /pentest/10.0.0.2, etc.
utils.create_dir_structure(vhost, host_dir)
#Next two lines create the file that will contain each command that was executed. This is not the audit log,
#but a log of commands that can easily be copy/pasted if you need to run them again.
summary_file_name = host_data_dir + "ScanSummary.log"
summary_file = open(summary_file_name, 'a')
#THIS is just a work around until i have a real solution. Really, UDP scans should be done
#For every host in the scanned host list, launch a quick UDP scan (top 100 ports)
scan_output_base_host_filename = host_data_dir + vhost
for (cmd_name, cmd) in config.items("nmap-commands"):
if cmd_name == "udp_scan":
#print(cmd_name,cmd)
outfile = scan_output_base_host_filename + "_" + cmd_name
populated_command = cmd.replace("[TARGET]", vhost).replace(
"[OUTPUT]", outfile)
populated_command = replace_user_config_options(
config_file, populated_command)
#print(cmd)
#cmd_name = "udp-top100"
#populated_command = 'nmap -sV -sC -Pn -sU --top-ports 100 -oN {0}_nmap_UDP_service_scan.txt -oX {0}_nmap_UDP_service_scan.xml {1}'.format(
# scan_output_base_host_filename, vhost)
if simulation:
populated_command = "#" + populated_command
#outfile = scan_output_base_host_filename + "_nmap_UDP_service_scan.txt"
task_id = uuid()
utils.create_task(cmd_name, populated_command, vhost,
outfile + ".txt", workspace, task_id)
result = chain(
# insert a row into the database to mark the task as submitted. a subtask does not get tracked
# in celery the same way a task does, for instance, you can't find it in flower
#tasks.cel_create_task.subtask(args=(cmd_name, populated_command, vhost, outfile + ".txt", workspace, task_id)),
# run the command. run_task takes care of marking the task as started and then completed.
# The si tells run_cmd to ignore the data returned from a previous task
tasks.run_cmd.si(cmd_name, populated_command,
celery_path,
task_id).set(task_id=task_id
), )() # .apply_async()
task_id_list.append(result.task_id)
if not simulation:
db.update_vhosts_submitted(vhost, vhost, workspace, 1)
#print "IP Address: {0}".format(vhost)
db_services = db.get_all_services_for_ip(vhost_ip[0], workspace)
for db_service in db_services:
(ip, scanned_service_port, scanned_service_protocol,
scanned_service_name, product, version, extra_info,
workspace) = db_service
scan_output_base_file_name = host_data_dir + vhost + "_" + str(
scanned_service_port
) + "_" + scanned_service_protocol + "_"
#If the service name is not in the supported service list, give the user notice so they can add the service
# and add some commands to the service. This is a major GAP right now. If the service is not in the config,
# the script completely ignores it, which is not good!
if scanned_service_name not in supported_services:
print(
"[!] Nmap reports {0}:{1} is running: [{2}]. There are no commands to run against {2} in config.ini."
.format(vhost, scanned_service_port,
scanned_service_name))
summary_file.write(
"[!] Nmap reports {0}:{1} is running: [{2}]. There are no commands to run against {2} in config.ini\n"
.format(vhost, scanned_service_port,
scanned_service_name))
#updated_port_scan = utils.nmap_follow_up_scan(vhost, scanned_service_port)
#scanned_service_name = updated_port_scan.hosts[0]._services[0].service
cmd_name = "nmap_service_scan"
populated_command = 'nmap -sV -sC -Pn -p {0} -oN {1}_nmap_service_scan.txt {2}'.format(
scanned_service_port, scan_output_base_file_name,
vhost)
populated_command = replace_user_config_options(
config_file, populated_command)
if simulation:
populated_command = "#" + populated_command
outfile = scan_output_base_file_name + "_nmap_service_scan.txt"
task_id = uuid()
utils.create_task(cmd_name, populated_command, vhost,
outfile, workspace, task_id)
result = chain(
# insert a row into the database to mark the task as submitted. a subtask does not get tracked
# in celery the same way a task does, for instance, you can't find it in flower
#tasks.cel_create_task.subtask(args=(cmd_name, populated_command, vhost, outfile , workspace, task_id)),
# run the command. run_task takes care of marking the task as started and then completed.
# The si tells run_cmd to ignore the data returned from a previous task
tasks.run_cmd.si(cmd_name, populated_command,
celery_path,
task_id).set(task_id=task_id
), )() # .apply_async()
task_id_list.append(result.task_id)
else:
parse_config_and_send_commands_to_celery(
scanned_service_name, scanned_service_port,
scan_output_base_file_name, config, simulation,
output_base_dir, host_dir, workspace, task_id_list,
vhost, scanned_service_protocol)
#task_id_list = task_id_list + new_tasks_list
summary_file.close()
print("[+] Submitted {0} tasks to the queue.".format(
len(task_id_list)))
total_tasks_num = total_tasks_num + len(task_id_list)
task_id_list = []
print(
"\n\n[+] Summary:\tSubmitted {0} tasks to the [{1}] workspace.".format(
total_tasks_num, workspace))
print(
"[+]\t\tThere might be additional tasks added to the queue during post processing\n[+]"
)
print("[+]\t\tTo keep an eye on things, run one of these commands: \n[+]")
print("[+]\t\tcelerystalk query [watch]")
print("[+]\t\tcelerystalk query brief [watch]")
print("[+]\t\tcelerystalk query summary [watch]\n")
def post_process_domains_bb(vhosts, command_name, populated_command, output_base_dir, workspace, simulation,
celery_path,out_of_scope_hosts):
config, supported_services = config_parser.read_config_ini()
vhosts = vhosts.splitlines()
# from https://stackoverflow.com/questions/14693701/how-can-i-remove-the-ansi-escape-sequences-from-a-string-in-python
ansi_escape = re.compile(r'\x1B\[[0-?]*[ -/]*[@-~]')
for vhost in vhosts:
# print("raw:\t" + vhost)
vhost = ansi_escape.sub('', vhost)
# print("escaped:\t" + vhost)
if re.match(r'\w', vhost):
try:
ip = socket.gethostbyname(vhost)
if vhost not in out_of_scope_hosts:
print("Found subdomain (in scope):\t" + vhost)
db_vhost = (ip, vhost, 1, 0, workspace)
db.create_vhost(db_vhost)
else:
print("Found subdomain (out of scope):\t" + vhost)
db_vhost = (ip, vhost, 0, 0, workspace)
db.create_vhost(db_vhost)
except:
print("1There was an issue running the nmap scan against {0}.").format(vhost)
ip = ""
db_vhost = (ip, vhost, 0, 0, workspace) # not in scope if no IP
print(db_vhost)
db.create_vhost(db_vhost)
# pull all in scope vhosts that have not been submitted
inscope_vhosts = db.get_inscope_unsubmitted_vhosts(workspace)
for scannable_vhost in inscope_vhosts:
scannable_vhost = scannable_vhost[0]
ip = db.get_vhost_ip(scannable_vhost, workspace)
ip = ip[0][0]
print("I'm going to scan: " + scannable_vhost + ":" + ip)
db_scanned_services = db.get_all_services_for_ip(ip, workspace)
for (
id, ip, scanned_service_port, scanned_service_protocol, scanned_service_name, workspace) in db_scanned_services:
# run chain on each one and then update db as submitted
scan_output_base_file_name = output_base_dir + "/" + ip + "/celerystalkOutput/" + scannable_vhost + "_" + str(
scanned_service_port) + "_" + scanned_service_protocol + "_"
host_dir = output_base_dir + "/" + ip
# TODO: This def might introduce a bug - same code as parse config submit jobs to celery. need to just call that function here
for section in config.sections():
if (section == "http") or (section == "https"):
if section == scanned_service_name:
for (cmd_name, cmd) in config.items(section):
outfile = scan_output_base_file_name + cmd_name
populated_command = cmd.replace("[TARGET]", scannable_vhost)\
.replace("[PORT]",str(scanned_service_port))\
.replace("[OUTPUT]", outfile) \
.replace("[PATH]", "")
if simulation:
# debug - sends jobs to celery, but with a # in front of every one.
populated_command = "#" + populated_command
# Grab a UUID from celery.utils so that i can assign it to my task at init, which is amazing because
# that allows me to pass it to all of the tasks in the chain.
task_id = uuid()
result = chain(
cel_create_task.subtask(args=(cmd_name,populated_command, scannable_vhost, outfile + ".txt",workspace, task_id)),
run_cmd.si(cmd_name, populated_command, celery_path, task_id).set(task_id=task_id),
post_process.si(cmd_name, populated_command, output_base_dir, workspace, scannable_vhost,
host_dir,
simulation,
scanned_service_port, scanned_service_name,
scanned_service_protocol, celery_path),
)()
host_audit_log = host_dir + "/" + "{0}_executed_commands.txt".format(ip)
f = open(host_audit_log, 'a')
f.write(populated_command + "\n\n")
f.close()
db.update_vhosts_submitted(ip, scannable_vhost, workspace, 1)
def post_process_domains(vhosts,command_name,populated_command,output_base_dir,workspace,domain,simulation,celery_path,scan_mode):
config,supported_services = config_parser.read_config_ini()
vhosts = vhosts.splitlines()
# from https://stackoverflow.com/questions/14693701/how-can-i-remove-the-ansi-escape-sequences-from-a-string-in-python
ansi_escape = re.compile(r'\x1B\[[0-?]*[ -/]*[@-~]')
for vhost in vhosts:
#print("raw:\t" + vhost)
vhost = ansi_escape.sub('', vhost)
#print("escaped:\t" + vhost)
if re.match(r'\w', vhost):
in_scope,ip = utils.domain_scope_checker(vhost,workspace)
if in_scope == 1:
print("Found subdomain (in scope):\t" + vhost)
db_vhost = (ip,vhost,1, 0,workspace)
db.create_vhost(db_vhost)
else:
print("Found subdomain (out of scope):\t" + vhost)
db_vhost = (ip, vhost, 0, 0, workspace)
db.create_vhost(db_vhost)
# elif scan_mode == "BB":
#
# cmd_name, cmd = config['nmap-bug-bounty_mode']
#
# utils.
#
# db_vhost = ("", vhost, 1, 0, workspace)
# db.create_vhost(db_vhost)
#pull all in scope vhosts that have not been submitted
inscope_vhosts = db.get_inscope_unsubmitted_vhosts(workspace)
for scannable_vhost in inscope_vhosts:
scannable_vhost = scannable_vhost[0]
ip = db.get_vhost_ip(scannable_vhost,workspace)
ip = ip[0][0]
db_scanned_services = db.get_all_services_for_ip(ip, workspace)
for (id,ip,scanned_service_port,scanned_service_protocol,scanned_service_name,workspace) in db_scanned_services:
#run chain on each one and then update db as submitted
scan_output_base_file_name = output_base_dir + "/" + ip + "/celerystalkOutput/" + scannable_vhost + "_" + str(scanned_service_port) + "_" + scanned_service_protocol + "_"
host_dir = output_base_dir + "/" + ip
#TODO: This def might introduce a bug - same code as parse config submit jobs to celery. need to just call that function here
for section in config.sections():
if (section == "http") or (section == "https"):
if section == scanned_service_name:
for (cmd_name, cmd) in config.items(section):
outfile = scan_output_base_file_name + cmd_name
populated_command = cmd.replace("[TARGET]", scannable_vhost).replace("[PORT]",
str(scanned_service_port)).replace("[OUTPUT]", outfile).replace("[PATH]", "")
if simulation:
# debug - sends jobs to celery, but with a # in front of every one.
populated_command = "#" + populated_command
# Grab a UUID from celery.utils so that i can assign it to my task at init, which is amazing because
# that allows me to pass it to all of the tasks in the chain.
task_id = uuid()
result = chain(
# insert a row into the database to mark the task as submitted. a subtask does not get tracked
# in celery the same way a task does, for instance, you can't find it in flower
cel_create_task.subtask(args=(cmd_name,populated_command, scannable_vhost, outfile + ".txt", workspace, task_id)),
# run the command. run_task takes care of marking the task as started and then completed.
# The si tells run_cmd to ignore the data returned from a previous task
run_cmd.si(cmd_name, populated_command, celery_path, task_id).set(task_id=task_id),
# right now, every executed command gets sent to a generic post_process task that can do
# additinoal stuff based on the command that just ran.
post_process.si(cmd_name, populated_command, output_base_dir, workspace, scannable_vhost, host_dir,
simulation,
scanned_service_port, scanned_service_name,
scanned_service_protocol,celery_path),
)() # .apply_async()
#task_id_list.append(result.task_id)
host_audit_log = host_dir + "/" + "{0}_executed_commands.txt".format(ip)
f = open(host_audit_log, 'a')
f.write(populated_command + "\n\n")
f.close()
db.update_vhosts_submitted(ip,scannable_vhost,workspace,1)