Intelligent Application Bypass

The following topics describe how to configure access control polices to use Intelligent Application Bypass (IAB)

Introduction to IAB

IAB identifies applications that you trust to traverse your network without further inspection if performance and flow thresholds are exceeded. For example, if a nightly backup significantly impacts system performance, you can configure thresholds that, if exceeded, trust traffic generated by your backup application.

The system implements IAB on traffic allowed by access control rules or the access control policy's default action, before the traffic is subject to deep inspection. A test mode allows you to determine whether thresholds are exceeded and, if so, to identify the application flows that would have been bypassed if you had actually enabled IAB (called bypass mode).

The following graphic illustrates the IAB decision-making process:



IAB Options

State

Enables or disables IAB.

Performance Sample Interval

Specifies the time in seconds between IAB performance sampling scans, during which the system collects system performance metrics for comparison to IAB performance thresholds. A value of 0 disables IAB.

Bypassable Applications and Filters

Provides an editor where you can specify bypassable applications and sets of applications (filters). See Application Conditions (Application Control).

Performance and Flow Thresholds

You must configure at least one inspection performance threshold and one flow bypass threshold. When a performance threshold is exceeded, the system examines flow thresholds and, if one threshold is exceeded, trusts the specified traffic.If you enable more than one of either, only one of each must be exceeded.

Inspection performance thresholds provide intrusion inspection performance limits that, if exceeded, trigger the inspection of flow thresholds. IAB does not use inspection performance thresholds set to 0. You can configure one or more of the following inspection performance thresholds:

Drop Percentage

Average packets dropped as a percentage of total packets, when packets are dropped because of performance overloads caused by expensive intrusion rules, file policies, decompression, and so on. This does not refer to packets dropped by normal configurations such as intrusion rules. Note that specifying an integer greater than 1 activates IAB when the specified percentage of packets is dropped. When you specify 1, any percentage from 0 through 1 activates IAB. This allows a small number of packets to activate IAB.

Processor Utilization Percentage

Average percentage of processor resources used.

Package Latency

Average packet latency in microseconds.

Flow Rate

The rate at which the system processes flows, measured as the number of flows per second. Note that this option configures IAB to measure flow rate, not flow count.

Flow bypass thresholds provide flow limits that, if exceeded, trigger IAB to trust bypassable application traffic in bypass mode or allow application traffic subject to further inspection in test mode. IAB does not use flow bypass thresholds set to 0. You can configure one or more of the following flow bypass thresholds:

Bytes per Flow

The maximum number of kilobytes a flow can include.

Packets per Flow

The maximum number of packets a flow can include.

Flow Duration

The maximum number of seconds a flow can remain open.

Flow Velocity

The maximum transfer rate in kilobytes per second.

Requirements and Prerequisites for Intelligent Application Bypass

Model Support

Any

Supported Domains

Any

User Roles

  • Admin

  • Access Admin

  • Network Admin

Configuring Intelligent Application Bypass


Caution


Not all deployments require IAB, and those that do might use it in a limited fashion. Do not enable IAB unless you have expert knowledge of your network traffic, especially application traffic, and system performance, including the causes of predictable performance issues. Before you run IAB in bypass mode, make sure that trusting the specified traffic does not expose you to risk.


Before you begin

For Classic devices, you must have the Control license.

Procedure


Step 1

In the access control policy editor, click Advanced, then click Edit (edit icon) next to Intelligent Application Bypass Settings.

If View (view button) appears instead, settings are inherited from an ancestor policy, or you do not have permission to modify the settings. If the configuration is unlocked, uncheck Inherit from base policy to enable editing.

Step 2

Configure IAB options:

  • State—Turn IAB Off or On, or enable IAB in Test mode.
  • Performance Sample Interval—Enter the time in seconds between IAB performance-sampling scans. If you enable IAB, even in test mode, enter a non-zero value. Entering 0 disables IAB.
  • Bypassable Applications and Filters—Click the number of bypassed applications and filters and specify the applications whose traffic you want to bypass; see Configuring Application Conditions and Filters.
  • Inspection Performance Thresholds—Click Configure and enter at least one threshold value.
  • Flow Bypass Thresholds—Click Configure and enter at least one threshold value.

You must specify at least one inspection performance threshold and one flow bypass threshold; both must be exceeded for IAB to trust traffic. If you enter more than one threshold of each type, only one of each type must be exceeded. For detailed information, see IAB Options.

Step 3

Click OK to save IAB settings.

Step 4

Click Save to save the policy.


What to do next

IAB Logging and Analysis

IAB forces an end-of-connection event that logs bypassed flows and flows that would have been bypassed, regardless of whether you have enabled connection logging. Connection events indicate flows that are bypassed in bypass mode or that would have been bypassed in test mode. Custom dashboard widgets and reports based on connection events can display long-term statistics for bypassed and would-have-bypassed flows.

IAB Connection Events

Action

When Reason includes Intelligent App Bypass:

Allow -

indicates that the applied IAB configuration was in test mode and traffic for the application specified by Application Protocol remains available for inspection.

Trust -

indicates that the applied IAB configuration was in bypass mode and traffic for the application specified by Application Protocol has been trusted to traverse the network without further inspection.

Reason

Intelligent App Bypass indicates that IAB triggered the event in bypass or test mode.

Application Protocol

This field displays the application protocol that triggered the event.

Example

In the following truncated graphic, some fields are omitted. The graphic shows the Action, Reason, and Application Protocol fields for two connection events resulting from different IAB settings in two separate access control policies.

For the first event, the Trust action indicates that IAB was enabled in bypass mode and Bonjour protocol traffic was trusted to pass without further inspection.

For the second event, the Allow action indicates that IAB was enabled in test mode, so Ubuntu Update Manager traffic was subject to further inspection but would have been bypassed if IAB had been in bypass mode.


A diagram showing three columns and two rows in a correlation events view. In the row for the first event, the columns read as follows: Action: Trust; Reason: Intelligent App Bypass; Application Protocol: Bonjour. In the row for the second event, the columns read as follows: Action: Allow; Reason: Intelligent App Bypass; Application Protocol: Ubuntu Update Manager.

Example

In the following truncated graphic, some fields are omitted. The flow in the second event was both bypassed (Action: Trust; Reason: Intelligent App Bypass) and inspected by an intrusion rule (Reason: Intrusion Monitor). The Intrusion Monitor reason indicates that an intrusion rule set to Generate Events detected but did not block an exploit during the connection. In the example, this happened before the application was detected. After the application was detected, IAB recognized the application as bypassable and trusted the flow.


A diagram showing four columns and two rows in a correlation events view. The row for the first event is irrelevant to the example. The columns for the second event read as follows: Last Packet: a time stamp is displayed; Action: Trust; Reason: The seemingly illogical combination of the two reasons Intelligent App Bypass and Intrusion Monitor; Application Protocol: HTTP.

IAB Custom Dashboard Widgets

You can create a Custom Analysis dashboard widget to display long-term IAB statistics based on connection events. Specify the following when creating the widget:

  • Preset: None

  • Table: Application Statistics

  • Field: any

  • Aggregate: either of:

    • IAB Bypassed Connections

    • IAB Would Bypass Connections

  • Filter: any

Examples

In the following Custom Analysis dashboard widget examples:

  • The Bypassed example shows statistics for application traffic bypassed because the applications were specified as bypassable and IAB was enabled in bypass mode in the deployed access control policy.

  • The Would Have Bypassed example shows statistics for application traffic that would have been bypassed because the applications were specified as bypassable and IAB was enabled in test mode in the deployed access control policy. .


Two essentially identical Custom Analysis widgets except that the selection for the Aggregate field is "IAB Bypassed Connections" for the widget configured to display bypassed traffic information, whereas the selection for the Aggregate field is "IAB Would Bypass Connections" for the widget configured for test mode

IAB Custom Reports

You can create a custom report to display long-term IAB statistics based on connection events. Specify the following when creating the report:

  • Table: Application Statistics

  • Preset: None

  • Filter: any

  • X-Axis: any

  • Y-Axis: either of:

    • IAB Bypassed Connections

    • IAB Would Bypass Connections

Examples

The following graphic shows two abbreviated report examples:

  • The Bypassed example shows statistics for application traffic bypassed because the applications were specified as bypassable and IAB was enabled in bypass mode in the deployed access control policy.

  • The Would Have Bypassed example shows statistics for application traffic that would have been bypassed because the applications were specified as bypassable and IAB was enabled in test mode in the deployed access control policy.


Two essentially identical reports where, for both, the Table field selection is "Application Statistics" and the selection for the X-Axis field is "Application." However, the selection for the Y-Axis field is "IAB Bypassed Connections" for the report configured to display bypassed traffic information, whereas the selection for the Y-Axis field is "IAB Would Bypass Connections" for the widget configured for test mode.