Plugin for NetBox to document Data Flows between systems and applications.

This plugin is considered Work in Progress (Beta). Please use caution if using this plugin for production!

• Document data flows between IP addresses, IP ranges and prefixes
• Document the application that requires these data flows
• WIP: Prepare the list of data flows to be injested as firewall rules

Full reference: Using Plugins - NetBox Documentation

• NetBox (>=3.6.0)
• Python 3.8 or higher

Note: the plugin uses some classes that are not explicitely exported in NetBox's plugin API, such as MPTT Tree-based models. Upward compatiblity is not guaranteed.

Install the Python package:

source /opt/netbox/venv/bin/activate
pip install netbox-data-flows

Add the plugin in NetBox configuration

# Add in: /opt/netbox/netbox/netbox/configuration.py

PLUGINS = [
  'netbox_data_flows',
]

Create the database migrations:

source /opt/netbox/venv/bin/activate
/opt/netbox/netbox/manage.py migrate netbox_data_flows

The plugin will be removed at the next NetBox update.

Add the Python package to local_requirements:

echo netbox-data-flows >> /opt/netbox/local_requirements.txt 

Add the plugin in NetBox configuration

# Add in: /opt/netbox/netbox/netbox/configuration.py

PLUGINS = [
  'netbox_data_flows',
]

Run the upgrade.sh script:

/opt/netbox/upgrade.sh

There is no PLUGIN_CONFIG configuration for this plugin. However, several other aspects can be configured.

The name of Data Flows, Data Flow Groups and Object Aliases is not constrained. You may wish to enforce your own validation rules in your configuration, e.g.:

# Add in: /opt/netbox/netbox/netbox/configuration.py

CUSTOM_VALIDATORS = {
    "netbox_data_flows.objectalias": [
        {
            "name": {
                "regex": "(host|net)_[a-z_]+"
            },
        }
    ]
}

Similar settings can be applied to:

• Applications: netbox_data_flows.application
• Application Roles: netbox_data_flows.applicationrole
• Data Flows: netbox_data_flows.dataflow
• Data Flow Groups: netbox_data_flows.dataflowgroup
• Object Aliases: netbox_data_flows.objectalias

Full reference: CUSTOM_VALIDATORS - NetBox Documentation

You can edit the list of available protocols when creating a data flow.

# Add in: /opt/netbox/netbox/netbox/configuration.py

FIELD_CHOICES = {
    'netbox_data_flows.DataFlow.protocol+': (
        ('igmp', "IGMP"),
    )
}

Full reference: FIELD_CHOICES - NetBox Documentation

Applications are logical grouping of data flows and can be business applications or infrastructure. Examples of applications:

• Active Directory
• MySuperBusinessApp
• Network management
• ...

Application Role is a label to help you categorize your applications. Each Application may have one Application Role. Examples of roles:

• Infrastructure
• Business Division 1
• ...

Data Flows modelize a network connection between two objects. They may be assigned to an Application

Data Flows should have a source, a destination, a protocol, source ports and destination ports. Only the protocol is mandatory.

Data Flow Groups form a forest of groups. A tree can be assigned to a single Application. Data Flow Groups can be enabled and disabled and inherit the status of their parent. Disabled Data Flow Groups disable all the Data Flows contained within.

Object Aliases are a group of references to other NetBox objects. Object Aliases are used as sources and destinations of Data Flows and corresponds to the groups or aliases used in firewall configuration. Internally, Object Aliases contain Object Alias Targets, because Django cannot create ManyToMany relationships to generic objects. Object Alias Targets are not exposed in the interface and should be transparent for the user.

Object Aliases currently supports:

• IP Addresses (ipam.ipaddress)
• IP Ranges (ipam.iprange)
• Prefixes (ipam.prefix) If an IP Address is assigned to a device or virtual machine, that device is also displayed.

Contributions are welcomed. This plugin is developped on the free time of its author, so do not expect regular releases.

• REST API and GraphQL API are not tested

See also the list of issues

• Include data's type/nature and link to data flows or at rest on devices