AWS IoT Device Defender Agent SDK (Python)

Example implementation of an AWS IoT Device Defender metrics collection agent, and other Device Defender Python samples.

On starting up for the first time - the DD agent publishes the metric values read from the network stats to DD, without computing any metric values delta. It does this because when it starts up it does not have any information of the previously collected metric values. The side-effect of this is the device's metrics will indicate a large spike each time the device restarts or the agent is restarted which can cause false-positives. Now, we have updated the agent to not send any metrics if it cannot compute the delta.

The provided sample agent can be used as a basis to implement a custom metrics collection agent.

Prerequisites

Minimum System Requirements

The Following requirements are shared with the AWS IoT Device SDK for Python

• Python 3.5+ for X.509 certificate-based mutual authentication via port 8883 and MQTT over WebSocket protocol with AWS Signature Version 4 authentication
• Python 3.5+ for X.509 certificate-based mutual authentication via port 443
• OpenSSL version 1.0.1+ (TLS version 1.2) compiled with the Python executable for X.509 certificate-based mutual authentication

Connect your Device to AWS IoT

If you have never connected your device to AWS IoT before, please follow the Getting Started with AWS IoT Guide. Make sure you note the location of your certificates, you will need to provide the location of these to the Device Defender Sample Agent.

Notes on the sample agent implementation

client id: The sample agent requires a client id that will also be used as the "Thing Name". This only for the sake of making the sample easy to get started with. To customize this behavior, you can modify the way the agent generates the MQTT topic for publishing metrics reports, to use a value other than client id as the thing name portion of the topic.

metric selection: The sample agent attempts to gather all supported Device Defender metrics. Depending on your platform requirements and use case, you may wish to customize your agent to a subset of the metrics.

Quickstart

Installation

1. Clone the repository

git clone https://github.com/aws-samples/aws-iot-device-defender-agent-sdk-python.git

1. Install Using pip

Pip is the easiest way to install the sample agent, it will take care of installing dependencies

pip install /path/to/sample/package

Running the Sample Agent

python agent.py --endpoint <your.custom.endpoint.amazonaws.com>  --rootCA </path/to/rootca>  --cert </path/to/cert> --key <path/to/key> --format json -i 300 -id <ThingName>

Command line options

To see a summary of all commandline options:

python agent.py --help

Test Metrics Collection Locally

python collector.py -n 1 -s 1

Custom Metric Integration

The sample agent has a flag allowing it to publish custom metrics

python agent.py --include-custom-metrics --endpoint <your.custom.endpoint.amazonaws.com>  --rootCA </path/to/rootca>  --cert </path/to/cert> --key <path/to/key> --format json -i 300 -id <ThingName>

This flag will tell the agent to publish the custom metric cpu_usage, a number float representing the current cpu usage as a percent. How this looks in the generated report can be seen in the sample report below.

We can run the command seen below to create the custom_metric for cpu_usage.

aws iot create-custom-metric --metric-name "cpu_usage" --metric-type "number" --client-request-token "access-test" --region us-east-1

After creating this custom_metric you will be able to create security profiles that use it.

aws iot create-security-profile \
--security-profile-name CpuUsageIssue \
--security-profile-description "High-Cpu-Usage"  \
--behaviors "[{\"name\":\"great-than-75\",\"metric\":\"cpu_usage\",\"criteria\":{\"comparisonOperator\":\"greater-than\",\"value\":{\"count\":75},\"consecutiveDatapointsToAlarm\":5,\"consecutiveDatapointsToClear\":1}}]" \
--region us-east-1

AWS IoT Greengrass Integration

Overview

AWS IoT Device Defender can be used in conjunction with AWS Greengrass. Integration follows the standard Greengrass lambda deployment model, making it easy to add AWS IoT Device Defender security to your Greengrass Core devices.

Prerequisites

1. Greengrass environment setup
2. Greengrass core configured and running
3. Ensure you can successfully deploy and run a lambda on your core

Using Device Defender with Greengrass Core devices

You can deploy a Device Defender to your Greengrass core in two ways:

1. Using the pre-built Greengrass Device Defender Connector (recommended)
2. Create a lambda package manually

Using Greengrass Connector

The Device Defender Greengrass Connector provides the most streamlined and automated means of deploy the Device Defender agent to your Greengrass core, and is the recommended method of using Device Defender with Greengrass.

For detailed information about using Greengrass Connectors see Getting Started with Greengrass Connectors For information about configuring the Device Defender Connector see Device Defender Connector Details

1. Create a local resource to allow your lambda to collect metrics from the Greengrass Core host
   • Follow the instructions here
   • Use the following parameters:
     • Resource Name: Core_Proc
     • Type: Volume
     • Source Path: /proc
     • Destination Path: /host_proc (make sure the same value is configured for the PROCFS_PATH environment variable above)
     • Group owner file access permission: "Automatically add OS group permissions of the Linux group that owns the resource"
     • Associate the resource with your metrics lambda
2. From the detail page of your Greengrass Group, click "Connectors" in the left-hand menu
3. Click the "Add a Connector" button
4. In the "Select a connector" screen, select the "Device Defender" connector from the list, click "Next"
5. On the "Configure parameters" screen, select the resource you created in Step 1, in the "Resource for /proc" box
6. In the "Metrics reporting interval" box, enter 300, or larger if you wish to use a longer reporting interval
7. Click the "add" button
8. Deploy your connector to your Greengrass Group

Create Your Lambda Package Manually

For this portion will be following the general process outlined here

Note: Due to platform-specific binary extensions in the psutil package, this process should be performed on the platform where you plan to deploy your lambda.

1. Clone the AWS IoT Device Defender Python Samples Repository

   git clone https://github.com/aws-samples/aws-iot-device-defender-agent-sdk-python.git

2. Create, and activate a virtual environment (optional, recommended)

   pip install virtualenv
   virtualenv metrics_lambda_environment
   source metrics_lambda_environment/bin/activate

3. Install the AWS IoT Device Defender sample agent in the virtual environment Install from PyPi

   pip install AWSIoTDeviceDefenderAgentSDK

   Install from downloaded source

   cd aws-iot-device-defender-agent-sdk-python
   #This must be run from the same directory as setup.py
   pip install .

4. Create an empty directory to assemble your lambda, we will refer to this as your "lambda directory"

   mkdir metrics_lambda
   cd metrics_lambda

5. Complete steps 1-4 from this guide

6. Unzip the Greengrass python sdk into your lambda directory

   unzip ../aws_greengrass_core_sdk/sdk/python_sdk_1_1_0.zip
   cp -R ../aws_greengrass_core_sdk/examples/HelloWorld/greengrass_common .
   cp -R ../aws_greengrass_core_sdk/examples/HelloWorld/greengrasssdk .
   cp -R ../aws_greengrass_core_sdk/examples/HelloWorld/greengrass_ipc_python_sdk .

7. Copy the AWSIoTDeviceDefenderAgentSDK module to the root level of your lambda

   cp -R ../aws-iot-device-defender-agent-sdk-python/AWSIoTDeviceDefenderAgentSDK .

8. Copy the Greengrass agent to the root level of your lambda directory

   cp ../aws-iot-device-defender-agent-sdk-python/samples/greengrass/greengrass_core_metrics_agent/greengrass_defender_agent.py .

9. Copy the dependencies from your virtual environment or your system, into the the root level of your lambda

   cp -R ../metrics_lambda_environment/lib/python2.7/site-packages/psutil .
   cp -R ../metrics_lambda_environment/lib/python2.7/site-packages/cbor .

10. Create your lambda zipfile Note: you should perform this command in the root level of your lambda directory

    rm *.zip
    zip -r greengrass_defender_metrics_lambda.zip *

Configure and deploy your Greengrass Lambda

1. Upload your lambda zip file
2. Select the Python 2.7 runtime, and enter greengrass_defender_agent.function_handler in the Handler field
3. Configure your lambda as a long-lived lambda
4. Configure the following environment variables:
   • SAMPLE_INTERVAL_SECONDS: The metrics generation interval. The default is 300 seconds. Note: 5 minutes (300 seconds) is the shortest reporting interval supported by AWS IoT Device Defender
   • PROCFS_PATH: The destination path that you will configure for your /proc resource as shown below.
5. Configure a subscription from your lambda to the AWS IoT Cloud Note: For AWS IoT Device Defender, a subscription from AWS IoT Cloud to your lambda is not required
6. Create a local resource to allow your lambda to collect metrics from the Greengrass Core host
   • Follow the instructions here
   • Use the following parameters:
     • Resource Name: Core_Proc
     • Type: Volume
     • Source Path: /proc
     • Destination Path: /host_proc (make sure the same value is configured for the PROCFS_PATH environment variable above)
     • Group owner file access permission: "Automatically add OS group permissions of the Linux group that owns the resource"
     • Associate the resource with your metrics lambda
7. Deploy your connector to your Greengrass Group

Troubleshooting

Reviewing AWS IoT Device Defender device metrics using AWS IoT Console

1. Temporarily modify your publish topic in your Greengrass lambda to something such as metrics/test
2. Deploy the lambda
3. Add a subscription to the temporary topic in the "Test" section of the iot console, shortly you should the metrics your Greengrass Core is emitting

Metrics Report Details

Overall Structure

Long Name	Short Name	Required	Type	Constraints	Notes
header	hed	Y	Object		Complete block required for well-formed report
metrics	met	Y	Object		Complete block required for well-formed report
custom_metrics	cmet	N	Object		Complete block required for well-formed report

Header Block

Long Name	Short Name	Requi red	Type	Constr aints	Notes
report _id	rid	Y	Inte ger		Monotonically increasing value, epoch timestamp recommended
versio n	v	Y	Stri ng	Major. Minor	Minor increments with addition of field, major increments if metrics removed

Metrics Block

TCP Connections

Long Name	Short Name	Parent Element	Required	Type	Constraints	Notes
tcp_connections	tc	metrics	N	Object
established_connections	ec	tcp_connections	N	List		ESTABLISHED TCP State
connections	cs	established_connections	N	List
remote_addr	rad	connections	Y	Number	ip:port	ip can be ipv6 or ipv4
local_port	lp	connections	N	Number	>0
local_interface	li	connections	N	String		interface name
total	t	established_connections	N	Number	>= 0	Number established connections

Listening TCP Ports

Long Name	Short Name	Parent Element	Required	Type	Constraints	Notes
listening_tcp_ports	tp	metrics	N	Object
ports	pts	listening_tcp_ports	N	List	> 0
port	pt	ports	N	Number	>= 0	ports should be numbers > 0
interface	if	ports	N	String		Interface Name
total	t	listening_tcp_ports	N	Number	>= 0

Listening UDP Ports

Long Name	Short Name	Parent Element	Required	Type	Constraints	Notes
listening_udp_ports	up	metrics	N	Object
ports	pts	listening_udp_ports	N	List	> 0
port	pt	ports	N	Number	> 0	ports should be numbers > 0
interface	if	ports	N	String		Interface Name
total	t	listening_udp_ports	N	Number	>= 0

Network Stats

Long Name	Short Name	Parent Element	Required	Type	Constraints	Notes
network_stats	ns	metrics	N	Object
bytes_in	bi	network_stats	N	Number	Delta Metric, >= 0
bytes_out	bo	network_stats	N	Number	Delta Metric, >= 0
packets_in	pi	network_stats	N	Number	Delta Metric, >= 0
packets_out	po	network_stats	N	Number	Delta Metric, >= 0

Custom Metrics

Long Name	Short Name	Parent Element	Required	Type	Constraints	Notes
cpu_usage	cpu	custom_metrics	N	Number

Sample Metrics Reports

Long Field Names

{
    "header": {
        "report_id": 1529963534,
        "version": "1.0"
    },
    "metrics": {
        "listening_tcp_ports": {
            "ports": [
                {
                    "interface": "eth0",
                    "port": 24800
                },
                {
                    "interface": "eth0",
                    "port": 22
                },
                {
                    "interface": "eth0",
                    "port": 53
                }
            ],
            "total": 3
        },
        "listening_udp_ports": {
            "ports": [
                {
                    "interface": "eth0",
                    "port": 5353
                },
                {
                    "interface": "eth0",
                    "port": 67
                }
            ],
            "total": 2
        },
        "network_stats": {
            "bytes_in": 1157864729406,
            "bytes_out": 1170821865,
            "packets_in": 693092175031,
            "packets_out": 738917180
        },
        "tcp_connections": {
            "established_connections":{
                "connections": [
                    {
                    "local_interface": "eth0",
                    "local_port": 80,
                    "remote_addr": "192.168.0.1:8000"
                    },
                    {
                    "local_interface": "eth0",
                    "local_port": 80,
                    "remote_addr": "192.168.0.1:8000"
                    }
                ],
                "total": 2
            }
        }
    },
    "custom_metrics": {
        "cpu_usage": [
            {
                "number": 26.1
            }
        ]
    }
}

Short Field Names

{
    "hed": {
        "rid": 1529963534,
        "v": "1.0"
    },
    "met": {
        "tp": {
            "pts": [
                {
                    "if": "eth0",
                    "pt": 24800
                },
                {
                    "if": "eth0",
                    "pt": 22
                },
                {
                    "if": "eth0",
                    "pt": 53
                }
            ],
            "t": 3
        },
        "up": {
            "pts": [
                {
                    "if": "eth0",
                    "pt": 5353
                },
                {
                    "if": "eth0",
                    "pt": 67
                }
            ],
            "t": 2
        },
        "ns": {
            "bi": 1157864729406,
            "bo": 1170821865,
            "pi": 693092175031,
            "po": 738917180
        },
        "tc": {
            "ec":{
                "cs": [
                    {
                    "li": "eth0",
                    "lp": 80,
                    "rad": "192.168.0.1:8000"
                    },
                    {
                    "li": "eth0",
                    "lp": 80,
                    "rad": "192.168.0.1:8000"
                    }
                ],
                "t": 2
            }
        }
    },
    "cmet": {
        "cpu": [
            {
                "number": 26.1
            }
        ]
    }
}

API Documentation

You can find the API documentation here

References

• AWS Lambda: Creating a Deployment Package (Python)
• Monitoring with AWS Greengrass Logs
• Troubleshooting AWS Greengrass Applications
• Access Local Resources with Lambda Functions

License

This library is licensed under the Apache 2.0 License.

Support

If you have technical questions about the AWS IoT Device SDK, use the AWS IoT Forum. For any other questions about AWS IoT, contact AWS Support.