 |
For an offline or printed copy of this document, simply choose ⋮ Options > Printer Friendly Page. You may then Print, Print to PDF or copy and paste to any other document format you like. |
Jonothan Eaves
NOTE: This function has been in early trials with Cisco DNA Center version 2.1.1
General availability is with Cisco DNA Center version 2.1.2
Introduction
About Group-Based Policy Analytics
Group-Based Policy Analytics is an application on Cisco DNA Center which helps administrators visualize group to group interactions, and discover ports and protocols being used between groups. These ports and protocols can subsequently be used within the Group-Based Access Control functions of Cisco DNA Center to enforce traffic as appropriate.
The 'Define' section below covers the challenges customers face today and the solution provided by Group-Based Policy Analytics. Also covered are the delivery phases and components of the solution.
About This Guide
This guide is intended to provide technical guidance for deploying Group-Based Policy Analytics. The guide covers design topics, deployment best practices and how to get the most out of the technology operation.
Figure 1: Guide workflow
As highlighted in figure 1 above, there are four major sections in this document. The initial, define section presents a high-level overview of the challenges customers face today and the solution that Group-Based Policy Analytics provides. Next, in the design section, we will see how to design the environment ready for deploying Group-Based Policy Analytics. Third, in the deploy part, the various configuration and best practice guidance will be provided for key components such as Cisco Identity Services Engine (ISE), Cisco DNA Center, Switches and Wireless LAN Controllers. Lastly, in the operate section, we will learn how best to operate the application within Cisco DNA Center.
Define
Organizations have challenges today. High profile attacks, we all either experience or read on the news, are driving customers towards an internal segmentation strategy, it's no longer effective purely securing the perimeter. However, the internal network is often largely unknown and it's difficult to understand the network behavior of people and things. Customers are asking for help in creating a network segmentation policy that is effective in today's world.
As a solution to this challenge, Cisco is providing an Application on Cisco DNA Center which provides 3 things
1) Discovery of group interaction and visibility based on behavior - Basically this is discovering what ports and policies would be required within an associated security policy.
2) Modeling and simulation of new groups and policies **
3) helping with the orchestration or authoring of these policies on the network **
** Note: The first phase of this App will cover Discovery only. Modelling and authoring will be delivered in future phases.
Figure 2: Logical Diagram
The basis of this Cisco DNA Center tool is to visualize network behavior based on groups.
As can be seen from the figure above, group entities are learned from a number of sources today.
One is ISE. Once there is connectivity to ISE the App can learn of scalable groups (SGTs), deployed on the network; plus Profile groups which indicate different types of endpoints connected.
Another way we can learn of groups is from Endpoint Analytics. With its Machine learning capabilities and multi factor classifications, unknowns can be reduced in the network and more accurate profile groups provided.
Also, Stealthwatch is optionally integrated in order to learn of the Host Groups provided by that platform.
Along with this group information, Group-Based Policy Analytics also receives Netflow from the network devices and stitches all this context together to produce graphs and tables to help the administrators visualize network behavior based on groups.
The idea is to provide an application to aid in network discovery and visualization and ultimately deliver mechanisms to help analyze security policy requirements and create effective security policy designs.
Design
Group-Based Policy Analytics is an Application on Cisco DNA Center but has been designed to discover group to group interactions whether the groups are assigned inside or outside an SD-Access fabric.
Hardware and Software Requirements
The following information covers the hardware and software requirements for Group-Based Policy Analytics:
Figure 3: Cisco DNA Center Requirements and Scale
Group-Based Policy Analytics version 1.0
For Cisco DNA Center GA version 2.1.2, ISE versions recommended: 2.4 P13 / 2.6 P7 / 2.7 P2 (or later)
For Cisco DNA Center EFT version 2.1.1, ISE versions recommended: 2.4 P7 / 2.6 P1 / 2.7 (or later)
Cisco DNA Advantage License
Stealthwatch version 7.x (or later)
For Group-Based Policy Analytics to display data, the following prerequisites are required:
1) Identity Services Engine needs to be connected to Cisco DNA Center
2) Netflow for endpoints in question needs to be sent to Cisco DNA Center
3) The Network Access Devices hosting the endpoints need to be discovered in Cisco DNA Center
4) If data from Stealthwatch Host Groups are required then Stealthwatch Management Console needs to be connected to Cisco DNA Center
5) Cisco DNA Advantage license is required
Note:
ISE sends the scalable groups and ISE Profiles, along with the associated network access device, to Cisco DNA Center over pxGrid. Cisco DNA Center will only accept this information if those network access devices have been discovered and are visible within the Inventory (point number 3 above). This session filtering mechanism has been incorporated to cater for multiple Cisco DNA Center platforms connecting to a single ISE.
Network Device Compatibility Information
The following is the network device compatibility information:
Figure 4: Compatibility
Accessing the Graphical User Interface
Navigate to Group-Based Policy Analytics via the top left icon
Figure 5: Cisco DNA Center Menu Access
and then select Policy > Group-Based Access Control:
Figure 6: Cisco DNA Center Menu
When in the Group-Based Access Control screen, navigate to Group-Based Policy Analytics by selecting the 'Analytics' menu:
Figure 7: Policy Analytics Menu
If ISE is not yet connected when navigating to Group-Based Policy Analytics, then a wizard will be displayed helping the administrator to connect ISE, Steathwatch (optional) and provide help for setting up Netflow.
Once ISE is connected to Cisco DNA Center, Group-Based Policy Analytics needs to execute the first data aggregation process before flow data is displayed. Data Aggregation operation then occurs every hour after which the flow data is updated.
Deploy
Zero Day Workflow
When Cisco DNA Center release 2.1.1 or later is installed then Group-Based Policy Analytics is available with Cisco DNA Advantage license.
As mentioned previously, if ISE has not previously been connected then the navigation to Policy > Group-Based Access Control > Analytics will instigate a Zero Day workflow.
Figure 8: Zero Day Workflow, Get Started
Click 'Get Started'.
Figure 9: Zero Day Workflow, Let's Do It
Click 'Let's Do It'. The Data Connectors can then be configured for ISE and optionally Stealthwatch:
Figure 10: Configure Group Data Connectors
Click on the 'Configure' link for ISE, a new browser tab will open to allow a new authentication and policy server to be added with your ISE details as follows:
Server IP Address: The IP Address of the ISE Primary PAN node.
Shared Secret: The ISE shared secret used between ISE and network devices
Username: ISE username
Password: ISE password (GUI and CLI password must be the same)
FQDN: ISE Primary PAN URL
Figure 11: Add ISE Details
The configuration status will take a few minutes to activate and you may have to accept the pxGrid approval request in ISE depending on the settings. The connection process is complete when an ACTIVE status is seen for the entry:
Figure 12: ISE Status
Once Active, close that browser tab and return to the tab displaying the Group Data Connectors. Optionally, click on the 'Configure' link for Stealthwatch, a new browser tab will open to help with connectivity to Stealthwatch:
Figure 13: Stealthwatch Configuration Options
Note: The Stealthwatch Analytics App (option 1 above) is just required to be installed to be able to retrieve the Host Groups. It does not need to be set up because network devices do not need to send NetFlow data to Stealthwatch. While that may be a particular requirement for the deployment, it is not a requirement purely for Group-Based Policy Analytics. Click “check here” to see if the Stealthwatch Analytics App is installed.
Click “settings” for option 2 above (that is, to set up the Stealthwatch Management Console) and complete the connectivity information:
Figure 14: Connecting to Stealthwatch
Once the connectors are set up, the workflow provides a Summary:
Figure 15: Connectivity Summary
Click on Edit for the Communication Connectors to be taken to a page describing different ways that network devices can be configured to send NetFlow data to Cisco DNA Center
Figure 16: Options for setting up Netflow
As can be seen from the options, the Template Editor can be used to send Netflow configuration to the network devices. However, the use of the Telemetry function makes this configuration much easier (but note the device types supported in the GUI text displayed above and the prerequisite 'lan' keyword requirement).
[Note: release 2.1.2 supports a new methodology of enabling Netflow on device interfaces as well as continuing to support 'lan' in the description. See this link for the criteria comparison of the two methods].
Clicking on 'Telemetry in Network Settings' navigates to the Telemetry section in the Design menu. Scroll down and enable 'Use Cisco DNA Center as NetFlow Collector server' as shown below and save the change:
Figure 17: Setting Cisco DNA Center as a Netflow Collector Server
Then, to push Netflow configuration to the network devices, navigate to
Provision >
Inventory, select the network device, and under
Actions, select
Telemetry >
Enable Application Telemetry:
Figure 18: Enable Application Telemetry (push Netflow config to network devices)
Notes:
1) If Stealthwatch Security Analytics (SSA) has already been used to deploy Netflow config to network devices then there may be a configuration conflict experienced. If Netflow is required to be sent to both Stealthwatch and Cisco DNA Center then the devices could be configured to send Netflow to a platform like Cisco UDP Director which could then send the Netflow to multiple destinations.
2) Netflow will be enabled on network device interfaces via Telemetry that have the letters 'lan' in the interface description (for wired) or in the WLAN profile name (for wireless). This is not automated by Cisco DNA Center. If configuring the description manually on the network devices then ensure a resync (
Provision >
Inventory, select device,
Actions>
Inventory >
Resync Device) is carried out before enabling the Application Telemetry to push the Netflow configuration. In release 2.1.2, there is a new method of enabling Netflow on device interfaces without needing 'lan' in the interface description, see
this link for a comparison of the criteria.
Wired Netflow Configuration
The wired Netflow configuration that is pushed to network devices is shown below. This will help in Template creation if required. The 'description lan' on the interface is not pushed and adding this manually is one method that Cisco DNA Center detects on which interfaces to enable the function:
|
flow record dnacrecord
match ipv4 version
match ipv4 protocol
match application name
match connection client ipv4 address
match connection server ipv4 address
match connection server transport port
match flow observation point
collect timestamp absolute first
collect timestamp absolute last
collect flow direction
collect connection initiator
collect connection client counter packets long
collect connection client counter bytes network long
collect connection server counter packets long
collect connection server counter bytes network long
collect connection new-connections
|
flow exporter dnacexporter
destination <Cisco DNA Center IP Address>
source Vlan <vlan>
transport udp 6007
export-protocol ipfix
option interface-table timeout 10
option vrf-table timeout 10
option sampler-table
option application-table timeout 10
option application-attributes timeout 10
|
|
flow monitor dnacmonitor
exporter dnacexporter
cache timeout inactive 10
cache timeout active 60
record dnacrecord
|
interface GigabitEthernet1/0/1
description lan
switchport mode access
ip flow monitor dnacmonitor input
ip flow monitor dnacmonitor output
|
Wireless Netflow Configuration
The wireless Netflow configuration that is pushed to network devices is shown below. This will help in Template creation if required. The term 'lan' in the Profile Name is not pushed and adding this manually is one method that Cisco DNA Center detects on which profiles to enable the function:
Creating Exporter and Monitor
config flow create exporter dnacexporter <exporterIp> port <exporterPort>
config flow create monitor dnacmonitor
config flow add monitor dnacmonitor exporter dnacexporter
config flow add monitor dnacmonitor record ipv4_client_src_dst_flow_record
Applying monitor to WLAN
config wlan disable <wlan_id>
config wlan flow <wlan_id> monitor dnacmonitor enable
config wlan enable <wlan_id>
Figure 19: WLC Profile Name
Once ISE and Netflow is setup (and optionally Stealthwatch) and data is being received, navigating to the Policy Analytics main page will show counts in the group boxes:
Figure 20: Policy Analytics Main Page
If the setup screens need to be accessed again once deployed, navigate to the Settings > Configuration menu as shown below:
Figure 21: Settings > Configuration
Operate
The main page (navigation is Policy > Group-Based Access Control > Analytics) consists of a search bar and counts within tiles. If connectivity to Stealthwatch has been made then three tiles will be shown, otherwise just two will be present:
Figure 22: Main Page
From left to right, the tiles show the number of scalable groups, ISE Profiles and optionally Stealthwatch Host Groups that contain members that are communicating with members of other groups. The tiles will show a count of 0 if there are no members of containing groups communicating with other groups.
Scalable Groups
Click on the number in the left tile (number 9 in the example above) to display communications between Scalable Groups:
Figure 23: Scalable Group to Scalable Group
Note the group entities shown on the left are source scalable groups and on the right are destination scalable groups. If members within those groups are communicating then a flow interaction will be displayed between the groups.
At the top right of the screen, there is a toggle button to toggle between graph/chart view and table view. When toggled to table view this is the result:
Figure 24: Table View of Scalable Group to Scalable Group
There is also a date/time selection near the top right. This allows the displayed data to be shown for any 1, 12 or 24 hour period within the last 14 days:
Figure 25: Date/Time Period Selection
Near the top left of the screen is displayed a filter which by default shows 'Communicating With: Scalable Groups'. This can be changed by clicking on the blue filter icon to show 'Communicating With: ISE Profiles' or 'Communicating With: Stealthwatch Host Groups':
Figure 26: Changing 'Communicating With'
As an example, if 'Communicating With: ISE Profiles' is selected, the data shows group to group communications from Scalable Groups to ISE Profiles:
Figure 27: Communicating With ISE Profiles
Hover over the flow interactions to show the number of flows detected between those groups:
Figure 28: Display Flow Count
Hover over a source scalable group to show which destination groups members are communicating with:
Figure 29: Source Scalable Group interactions
Click on a source group to navigate to a screen showing the destination groups that members of the source group are interacting with:
Figure 30: Source Scalable Group to Multiple Destination Scalable Groups
Screen explanations follow:
Figure 31: GUI Layout Explanation
1) Search for Scalable Groups.
2) Click the blue filter icon to change the destination group type (Communicating With)
3) Breadcrumbs (Click each item to change the display.)
4) Click individual flow to show details of that flow.
5) Show more destination groups if more exist.
6) Toggle between chart view and table view.
7) Set date and time range for data display.
Choose between inbound or outbound flows for the displayed source group.
9) Number of unique traffic flows detected
10) Number of destination groups
11) Create a report (Found at Report > Generated Reports).
12) Download a previously created report.
13) Click destination group to navigate to the detail page.
14) Pagination control
If 'Create Report' is selected (number 11 above), a report is automatically created with the details as displayed on the current chart page. Clicking navigates to Report > Generated Reports.
The blue icon under “Last Run” shows the data is being collated, below is an example:
Figure 32: Report being created
When the data collection has completed, the icon turns green, and a download icon is displayed:
Figure 33: Report completed and ready for download
If the report is downloaded the format is as per this example:
Figure 34: Example Downloaded Report
From the main chart or the chart showing a single source group, the flows or groups highlighted below can be selected:
Figure 35: Click to Display Applications, Ports and Protocols
When selected, a chart showing a one-to-one group interaction is displayed along with applications, ports and protocols discovered between those groups. It is this information which will allow access control policies to be built with confidence:
Figure 36: Discovered Applications, Ports and Protocols Between Groups
Screen explanations follow:
Figure 37: GUI Layout Explanation
1) Breadcrumbs (Click each item to navigate back to previous screens).
2) Toggle between chart view and table view.
3) Set the date and time range for the data display.
4) Change the data displayed depending on the direction of flow.
5) Filter the results based on Service name, Protocol, and/or Port.
6) Create a report of the displayed data
7) View the contract for the source and destination groups displayed (see below)
Shows the direction of flow
9) Shows the Application/Service name
10) Shows the protocol
11) Shows the port number
If 'View Contract' is selected (number 7 above), then not only will the discovered ports and protocols be displayed between those two groups, but the configured contract between the groups will also be shown for an easy comparison of discovered traffic vs configured policy:
Figure 38: From Policy Analytics: Contract and Discovered Activity Side-by-Side
When navigated to the Group-Based Access Control screen and assigning a contract within a policy, a 'View Contract Activity' option is shown:
Figure 39: From Group-Based Access Control: View Contract Activity
When selected, the discovered activity is shown beside the configured contract:
Figure 40: From Group-Based Access Control: Contract and Discovered Activity Side-by-Side
ISE Profile Groups
Back at the main Policy Analytics page, click on the number in the middle tile (number 10 in this example):
Figure 41: Policy Analytics Main Page
This displays communications from ISE Profiles to Scalable Groups:
Figure 42: Communications from ISE Profiles to Scalable Groups
Clicking the source group and flow interactions and navigating down to show applications, ports and protocols works the same as explained in the previous section. Currently, you cannot display ISE Profile to ISE Profile interactions, one end has to be Scalable Groups.
Stealthwatch Host Groups
Back at the main Policy Analytics page, click on the number in the right tile (number 5 in this example):
Figure 43: Policy Analytics Main Page
This displays communications from Stealthwatch Host Groups to Scalable Groups:
Figure 44: Communications from Stealthwatch Host Groups to Scalable Groups
Clicking the source group and flow interactions and navigating down to show applications, ports and protocols works the same as explained in the previous sections. Currently, you cannot display Stealthwatch Host Groups to Stealthwatch Host Groups interactions, one end has to be Scalable Groups.
Search Function
IP addresses, group names, and/or MAC addresses can be searched in the main page search bar:
Figure 45: Main Page Search Bar
Possible result categories are displayed as you type the characters. In the following example, entering 10.4 provides the possible categories of Source IP Address or Destination IP Address:
Figure 46: Search Categories Automatically Displayed
Click on one of the available categories and a search is executed for the entered characters:
Figure 47: Search result
A more detailed search can be carried out by selecting the filter icon at the top right of the table. The figure below shows that filter, and further search criteria entered for scalable group and destination IP address. As can be seen, AND/OR functions can be used within the columns and it is an AND operation across columns:
Figure 48: Detailed Search Criteria
Further columns can be added by clicking on the three dots on the right of the headings:
Figure 49: Selecting Search Columns
The search criteria can be saved by clicking the ribbon icon near the top right corner and selecting 'Save Current Search':
Figure 50: Save Search Criteria
Deployment Guide Summary
Group-Based Policy Analytics provides an understanding of group to group communication patterns and visualization into the ports and protocols needed in access control, or Group-Based policies.
This guide is an aid to deploying the Group-Based Policy Analytics application as well as the components necessary to provide the solution. The guide also covers design aspects and operation of the system.
This is just phase one of the journey for Policy Analytics covering discovery. This guide will be updated over time when further features are implemented like policy creation/authoring and policy simulation/modelling.
Appendix
List of Acronyms
AAA Authentication, Authorization and Accounting
DNA (Cisco) Digital Network Architecture
FQDN Fully Qualified Domain Name
ISE Identity Services Engine
LAN Local Area Network
MFC Multi Factor Classification
SDA Software Defined Access
SGT Scalable Group Tag
SSA Stealthwatch Security Analytics
UDP User Datagram Protocol
WLAN Wireless Local Area Network
WLC Wireless LAN Controller
