AWS remote run crashes if monkey is already running

VakarisZ commented 2 years ago

Describe the bug

To Reproduce

Steps to reproduce the behavior:

Run monkey on remote AWS machine
Run monkey there again, when it's already running

Get white screen, island output:

2021-11-16 15:18:16,382 - app.py:1455 - log_exception() - ERROR - Exception on /api/remote-monkey [POST]
Traceback (most recent call last):
File "/tmp/.mount_Infectn0XuRq/opt/python3.7/lib/python3.7/site-packages/flask/app.py", line 1513, in full_dispatch_request
rv = self.dispatch_request()
File "/tmp/.mount_Infectn0XuRq/opt/python3.7/lib/python3.7/site-packages/flask/app.py", line 1499, in dispatch_request
return self.ensure_sync(self.view_functions[rule.endpoint])(**req.view_args)
File "/tmp/.mount_Infectn0XuRq/opt/python3.7/lib/python3.7/site-packages/flask_restful/__init__.py", line 467, in wrapper
resp = resource(*args, **kwargs)
File "/tmp/.mount_Infectn0XuRq/opt/python3.7/lib/python3.7/site-packages/flask/views.py", line 83, in view
return self.dispatch_request(*args, **kwargs)
File "/tmp/.mount_Infectn0XuRq/opt/python3.7/lib/python3.7/site-packages/flask_restful/__init__.py", line 582, in dispatch_request
resp = meth(*args, **kwargs)
File "/tmp/.mount_Infectn0XuRq/usr/src/monkey_island/cc/resources/auth/auth.py", line 63, in wrapper
return fn(*args, **kwargs)
File "/tmp/.mount_Infectn0XuRq/usr/src/monkey_island/cc/resources/remote_run.py", line 56, in post
result = self.run_aws_monkeys(body)
File "/tmp/.mount_Infectn0XuRq/usr/src/monkey_island/cc/resources/remote_run.py", line 29, in run_aws_monkeys
return RemoteRunAwsService.run_aws_monkeys(instances, island_ip)
File "/tmp/.mount_Infectn0XuRq/usr/src/monkey_island/cc/services/remote_run_aws.py", line 53, in run_aws_monkeys
lambda _, result: result.is_success,
File "/tmp/.mount_Infectn0XuRq/usr/src/common/cmd/cmd_runner.py", line 54, in run_multiple_commands
command_result_pairs = CmdRunner.wait_commands(list(command_instance_dict.keys()))
File "/tmp/.mount_Infectn0XuRq/usr/src/common/cmd/cmd_runner.py", line 99, in wait_commands
str(command[1]),
TypeError: 'Cmd' object is not subscriptable
2021-11-16 15:18:16,384 - pywsgi.py:1226 -      write() - INFO - 88.222.20.171 - - [2021-11-16 15:18:16] "POST /api/remote-monkey HTTP/1.1" 500 163 6.524820

Expected behavior

Get the same pop-up as you would get on the "run on island" option.

Screenshots

If applicable, add screenshots to help explain your problem.

Machine version (please complete the following information):

AWS

Tasks

[x] Fix it (0d) @ilija-lazoroski

ilija-lazoroski commented 1 year ago

This issue is outdated. We have changed the way we run agents on AWS instances. I have run an agent and I got that the singleton is locked which means that an agent is already running.

The running of agents is done on different threads so the issue above is fixed. Check this code on who we do it.

Agent log while another agent is running: 2022-11-24T16.55.38.800Z-ip-172-31-24-149.log

ilija-lazoroski commented 1 year ago

Resolved in https://github.com/guardicore/monkey/issues/1928.

guardicore / monkey