Add system communication interfaces for IPS engines

Normal Interfaces are used for communication between the IPS engine and the Management Server.

Physical Interfaces correspond to network ports on the IPS engine. In an IPS Cluster, each physical interface definition represents a network interface on all nodes of the cluster. By default, the numbering of the Physical Interfaces in the Management Client corresponds to the operating system interface numbering on the engine. For example, Interface ID 0 is mapped to eth0, and Interface ID 1 is mapped to eth1. However, the mapping is not fixed and you can change it through the engine command line. This mapping can be done differently from node to node as long as you take care that the same interface on each node is correctly cabled to the same network.

In a Single IPS, Normal Interfaces are used for:
  • Communication between the IPS engine and the Management Server.
  • Sending event information and traffic recordings to Log Servers.
  • As the Reset Interface for sending TCP Reset responses.
In an IPS Cluster, Normal Interfaces handle all traffic for which the end-point of the communication is a node itself. Normal Interfaces are used for:
  • Heartbeat communication between the nodes.
  • Communication between each individual node and the Management Server.
  • Sending event information and traffic recordings to Log Servers.
  • For any other traffic between the node itself and some other host.
Normal Interfaces in an IPS Cluster are also used as the Reset Interface for sending TCP Reset responses.

Each Single IPS engine or node in an IPS Cluster needs at least one Normal interface for communicating with other SMC components. You can define more than one system communication interface if it is necessary in your network environment. It is recommended to create a separate Normal Interface that is used for communication with the Management Server rather than using the same Normal Interface for sending event information and traffic recordings to Log Servers, and for communication with the Management Server.

  For more details about the product and how to configure features, click Help or press F1.

Steps

  1. Right-click a Single IPS or IPS Cluster and select Edit <element type>.
    The Engine Editor opens.
  2. In the navigation pane on the left, select Interfaces.
    The Interfaces pane opens on the right.
  3. Right-click the empty space and select New Physical Interface.
  4. Define the Physical Interface properties.
    Select Normal Interface as the Type.
    CAUTION:
    Make sure that you set the interface speed correctly. When the bandwidth is set, the IPS engine always scales the total amount of traffic on this interface to the bandwidth you defined. This scaling happens even if there are no bandwidth limits or guarantees defined for any traffic.
  5. Click OK. The Physical Interface is added to the interface list.
  6. Continue the configuration in one of the following ways:
    • Add VLAN interfaces to the Physical Interface.
    • Add IP addresses to the system communication interfaces.

Physical Interface Properties dialog box (IPS engine)

Use this dialog box to define the Physical Interface properties for a Single IPS engine, IPS Cluster, Virtual IPS engine, or Master NGFW Engine in the IPS role.

Note: The available options can vary depending on the type of IPS engine.
Option Definition
General tab
Interface ID The Interface ID automatically maps to a physical network port of the same number during the initial configuration of the engine. The mapping can be changed as necessary through the engine’s command line interface.
Note: Changes to the Master NGFW Engine interface mapping do not affect the Interface IDs that are defined for Virtual NGFW Engines in Virtual Resource elements.
Type

Normal Interface — Corresponds to a single network interface on the engine. Not supported on Virtual IPS engines.

Capture Interface — Network traffic is monitored only. Traffic cannot be blocked.

Inline Interface — The interface is directly on the traffic path so that traffic must always pass through the engine to reach its destination. Only traffic that attempts to pass through Inline Interfaces can be actively filtered.

Zone

(Optional)

Select the network zone to which the interface belongs. Click Select to select an element, or click New to create an element.
MTU

(Optional, not supported on Virtual NGFW Engines)

The maximum transmission unit (MTU) size on the connected link. Either enter a value between 400–65535 or select a common MTU value from the list.

If the interface is a Physical Interface, the same MTU is automatically applied to any VLANs created under it.

The default value (also the maximum standard MTU in Ethernet) is 1500. Do not set a value larger than the standard MTU, unless you know that all devices along the communication path support it.

Note: To set the MTU for a Virtual NGFW Engine, you must configure the MTU for the interface on the Master NGFW Engine that hosts the Virtual NGFW Engine, then refresh the policy on the Master NGFW Engine and the Virtual NGFW Engine.
Comment

(Optional)

 A comment for your own reference.
Option Definition
General tab, Inline Interface Settings or Capture Interface Settings section
Reset Interface

(Capture Interfaces only)

Select the Reset Interface to specify the interface through which TCP connection resets are sent when Reset responses are used in your IPS policy.
Logical Interface Select the Logical Interface.
Note: You cannot use the same Logical Interface element for both Inline and Capture Interfaces on the same engine.
Failure Mode

(Inline Interfaces only)

Select how traffic to the Inline Interface is handled if the engine goes offline.

  • Bypass — traffic is allowed through the Inline Interface without inspection.
  • Normal — traffic is not allowed through the Inline Interface.
Note: If there are VLAN Interfaces under the Inline Interface, you must select Bypass.
Note: For Virtual IPS engines, this option is set on the Master NGFW Engine.
CAUTION:
Using bypass mode requires a fail-open network interface card. If the ports that represent the interfaces cannot fail open, policy installation fails on the NGFW Engine. Bypass mode is not compatible with VLAN retagging. In network environments where VLAN retagging is used, normal mode is automatically enforced.
Inspect unspecified VLANs

(Capture or Inline Interfaces only)

Select this option to make the engine inspect traffic from VLANs that are not included in the engine’s interface configuration. We recommend that you keep this option selected if you do not have a specific reason to deselect it.
Bypass unspecified VLANs

(Master NGFW Engines only)

(Inline Interfaces only)

When this option is selected, traffic from VLANs that are not allocated to any Virtual NGFW Engine is bypassed without inspection. Deselect this option to make the Master NGFW Engine block traffic from VLANs that are not allocated to any Virtual NGFW Engine. We recommend that you keep this option selected if you do not have a specific reason to deselect it.
Option Definition
General tab, Second Interface section

(Inline Interfaces only)

ID Select a Second Interface ID. The Interface ID is mapped to a physical network port of the same number during the initial configuration of the engine.
Zone

(Optional)

Select the network zone to which the interface belongs. Click Select to select an element, or click New to create an element.
Second Interface ID

(Inline Interfaces on Master NGFW Engines only)

Select a Second Interface ID. The Interface ID is mapped to a physical network port of the same number during the initial configuration of the engine.
Option Definition
General tab, Quality of Service and Bandwidth Management section
QoS Mode

(Optional)

Defines how QoS is applied to the link on this interface.

If Full QoS or DSCP Handling and Throttling is selected, a QoS policy must also be selected. If Full QoS is selected, the throughput must also be defined.

If the interface is a Physical Interface, the same QoS mode is automatically applied to any VLANs created under it.

QoS Policy

(DSCP Handling and Throttling and Full QoS modes only)

The QoS policy for the link on this interface.

If the interface is a Physical Interface, the same QoS policy is automatically selected for any VLANs created under it.

Note: If a Virtual Resource has a throughput limit defined, the interfaces on the Virtual NGFW Engine that use a QoS policy all use the same policy. The policy used in the first interface is used for all the interfaces.
Interface Throughput Limit

(Full QoS mode only)

Enter the throughput for the link on this interface as megabits per second.

If the interface is a Physical Interface, the same throughput is automatically applied to any VLANs created under it.

The throughput is for uplink speed (outgoing traffic) and typically must correspond to the speed of an Internet link (such as an ADSL line), or the combined speeds of several such links when connected to a single interface.

CAUTION:
Make sure that you set the interface speed correctly. When the bandwidth is set, the NGFW Engine always scales the total amount of traffic on this interface to the bandwidth you defined. This scaling happens even if there are no bandwidth limits or guarantees defined for any traffic.
CAUTION:
The throughput for a Physical Interface for a Virtual NGFW Engine must not be higher than the throughput for the Master NGFW Engine interface that hosts the Virtual NGFW Engine. Contact the administrator of the Master NGFW Engine before changing this setting.
Option Definition
General tab, Virtual Resource section

(Master NGFW Engines only)

Virtual Resource

The Virtual Resource associated with the interface. Select the same Virtual Resource in the properties of the Virtual IPS element to add the Virtual NGFW Engine to the Master NGFW Engine.

Only one Virtual Resource can be selected for each Physical Interface. If you want to add multiple Virtual Resources, add VLAN Interfaces to the Physical Interface, then select the Virtual Resource in the VLAN Interface properties.

Allow VLAN Definition in Virtual Engine When selected, allows VLAN Interfaces to be added to the automatically created Physical Interfaces in the Virtual NGFW Engine that is associated with this interface.
Virtual Engine Interface ID Specifies the Interface ID of the Physical Interface in the Virtual NGFW Engine that is associated with this interface.
Option Definition
Advanced tab

(All optional settings)

Override Engine's Default Settings When selected, the default settings of the engine are overridden.
Note: These options cannot be selected if using Capture Interfaces.
SYN Rate Limits
  • Default — The interface uses the SYN rate limits defined for the engine on the Advanced Settings branch of the Engine Editor.
  • None — Disables SYN rate limits on the interface.
  • Automatic — This is the recommended mode if you want to override the general SYN rate limits defined for the engine on the Advanced Settings branch of the Engine Editor. The engine calculates the number of allowed SYN packets per second and the burst size (the number of allowed SYNs before the engine starts limiting the SYN rate) based on the engine’s capacity and memory size.
  • Custom — Enter the values for Allowed SYNs per Second and Burst Size.
Allowed SYNs per Second Defines the number of allowed SYN packets per second.
Burst Size The number of allowed SYNs before the engine starts limiting the SYN rate.
Tip:

We recommend that you set the burst size to be at least one tenth of the Allowed SYNs per Second value. If the burst size is too small, SYN rate limits do not work.

For example, if the value for Allowed SYNs per Second is 10000, set the value for Burst Size to at least 1000.

Enable Log Compression

Allows you to define the maximum number of separately logged entries.

For each event type, Antispoofing or Discard, you can define:

  • Log Rate (Entries/s) — The maximum number of entries per second. The default value for antispoofing entries is 100 entries/s. By default, Discard log entries are not compressed.
  • Burst Size (Entries) — The maximum number of matching entries in a single burst. The default value for antispoofing entries is 1000 entries. By default, Discard log entries are not compressed.
Set to Default Returns all changes to the log compression settings to the default settings.