Interfaces Configuration

Configuration Hierarchy

The diagram illustrates the interface configuration hierarchy.

Interface Configuration Hierarchy

Configuration Syntax and Commands

The following sections describe the interface configuration syntax and commands.

MTU Profile Configuration

This section describes how to configure MTU profiles.

Syntax:

set forwarding-options mtu-profile <attribute> <value>

Attribute Description

mtu-profile <mtu-profile-name>

MTU profile name

mtu-size <mtu-size>

MTU size. Range: 64 to 9216 bytes

mtu-type <mtu-type>

Specify the MTU type:

  • physical: Port based MTU profile

  • pppoe: subscriber IFL-based MTU profile for L2TP and PPPoE. This MTU profile is used by PPPoE subscribers to set the default MTU size of 1492. A configured size of 1492 bytes limits the size of the IPv4 or IPv6 header plus payload.

  • ipv4: MTU profile of type IPv4. Only IPv4 traffic on the logical interface will be impacted.

  • ipv6: MTU profile of type IPv6. Only IPv6 traffic on the logical interface will be impacted.

  • ip: MTU profile of type IP. Both IPv4 and IPv6 traffic on the logical interface will be impacted.

action <mtu-action>

Specify the MTU action. The following options are supported:
drop: This indicates that when the MTU check fails, the action "drop" is performed.
redirect-to-cpu: This is an action of redirecting packets to the CPU in a traffic behavior. A redirect-to-cpu action must be configured for fragmentation to occur.

Example 1: Configuration of the MTU Profile for the Physical Port

{
  "ietf-restconf:data": {
    "rtbrick-config:forwarding-options": {
      "mtu-profile": [
        {
          "mtu-profile-name": "portMtu",
          "size": 5000,
          "type": "physical",
          "action": "redirect-to-cpu"
        }
      ]
    }
  }
}

Example 2: MTU Profile Configuration of Type IPv4

{
  "ietf-restconf:data": {
    "rtbrick-config:forwarding-options": {
      "mtu-profile": [
        {
          "mtu-profile-name": "ipv4Mtu",
          "size": 1300,
          "type": "ipv4",
          "action": "redirect-to-cpu"
        }
      ]
    }
  }
}

Example 3: MTU Profile Configuration of Type IPv6

{
  "ietf-restconf:data": {
    "rtbrick-config:forwarding-options": {
      "mtu-profile": [
        {
          "mtu-profile-name": "ipv6Mtu",
          "size": 1400,
          "type": "ipv6",
          "action": "redirect-to-cpu"
        }
      ]
    }
  }
}

Example 4: Configuration of the MTU Profile for PPPoE

{
  "ietf-restconf:data": {
    "rtbrick-config:forwarding-options": {
      "mtu-profile": [
        {
          "mtu-profile-name": "pppoeMtu",
          "size": 1492,
          "type": "pppoe",
          "action": "redirect-to-cpu"
        }
      ]
    }
  }
}

Enabling Hostpath Fragmentation

This section describes how to enable or disable fragmentation by CPU. It is necessary to configure MTU profile action "redirect-to-cpu" so that fragmentation takes place. By default, fragmentation is disabled.

Syntax:

set forwarding-options fragmentation ipv4 state <value>

Attribute Description

disabled | cpu

Enables fragmentation of IPv4 packets.
There are two options:
disabled—Fragmentation is disabled. It is the default setting.
cpu—Fragmentation is performed by CPU.

Example: Configuration of Hostpath Fragmentation

{
  "ietf-restconf:data": {
    "rtbrick-config:forwarding-options": {
      "fragmentation": {
        "ipv4": {
          "state": "cpu"
        }
      }
    }
  }
}

Physical Interface Configuration

This section describes configuration options at the physical interface (IFP) level.

Syntax:

set interface <interface-name> <attribute> <value>

Attribute Description

<interface-name>

Name of the interface. Example: ifp-0/0/1.

admin-status <down|up>

Administrative state of the interface.

auto-negotiation true

Enable auto-negotiation. Note: To disable auto-negotiation, use the delete form of the command. NOTE: Port speed configuration and auto-negotiation are mutually exclusive.

holddown <disable|enable>

Enable or disable the holddown timer feature. By default, the feature is enabled.

holddown-down-delay <…​>

Configure the down delay time in milliseconds (ms).

holddown-up-delay <…​>

Configure the down delay time in milliseconds (ms).

class-of-service <profile-name>

Apply class-of-service profile name.

description

Configure physical interface description.

host-if <container-interface>

Configure a host interface. For example, if the container interface eth1 connects to the host interface vethXYZ123, use this command option to bound hostif-0/0/1 to eth1. Please note the Linux virtual ethernet (veth) interface needs to be created separately. It cannot be created via RBFS configuration.

forward-error-correction <fec-type>

Configure Forward Error Correction (FEC) on the physical interface. FEC allows you to send the required information to correct errors through the link along with the payload data. A benefit of "forward" in FEC is that errors detected at the receiver do not need to be retransmitted. Currently, the supported FEC types are: base-r, rsfec, none.

NOTE: rsfec is the only FEC supported for 100G on the QAX platform.

link-training true

Enable link training. Note: To disable link training, use the delete form of the command.

master <true|false>

Memif role, master or slave, applicable only to memif interface. One end needs to be configured as master, and the other one as slave.

memif-id <id>

Configure memif ID, applicable only to memif interface. Needs to match on both ends.

mtu-profile <mtu-profile-name>

Attach MTU profile to a physical interface. This is a mandatory attribute.

mru <size>

Maximum receive unit size on the physical interface.

speed <speed>

Configure speed mode for the interface. Port speed refers to the maximum amount of data transmitted. The speed value is specified in Gigabits per second (Gbps).

Currently, RBFS supports 10G and 100G ports, and you can make the following changes:

  • 100G port speed can be changed to 40G

  • 10G port speed can be changed to 1G

Example 1: Physical Interface Configuration

{
    "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "description": "Link to leaf1",
        "speed": "10G",
        "mtu-profile": "portMtu",
        "mru": 5000
      }
    ]
}

Example 2: Memory Interface Configuration

A End:

{
    "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "description": "Master",
        "memif-id": 11,
        "master": "true",
      }
    ]
}

B End:

{
    "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "description": "Slave",
        "memif-id": 11,
        "master": "false",
      }
    ]
  }

Example 3: Host Interface Configuration

{
    "rtbrick-config:interface": [
        "name": "ifp-0/0/1",
        "description": "Represents eth1 as ifp-0/0/1 in RBFS",
        "host-if": "eth1",
    ]
}

Example 4: MRU Configuration for Physical Interface

{
    "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/7",
        "mru": 5000
      }
    ]
  }

Example 5: FEC Configuration for Physical Interface

{
    "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/40",
        "forward-error-correction": "base-r"
      }
    ]
  }

Logical Interface Configuration

This section describes configuration options at the logical interface (IFL) level.

Syntax:

set interface <interface-name> unit <unit-id> <attribute> <value>

Attribute Description

unit <unit-id>

Create a logical interface (also referred to as a sub-interface) under the physical interface.

admin-status <down|up>

Administrative state of the logical interface.

class-of-service <profile-name>

Apply class-of-service profile name.

description <description>

Description of the logical interface.

inner-vlan <inner-vlan-id>

Inner VLAN ID.

instance <instance>

Assign the logical interface to an instance.

ipv4-admin-status <down|up>

Enable or disable IPv4.

ipv4-mtu-profile <ipv4-mtu-profile>

Attach IPv4 MTU profile to an L3 interface.

ipv6-admin-status <down|up>

Enable or disable IPv6.

ipv6-mtu-profile <ipv6-mtu-profile>

Attach IPv6 MTU profile to an L3 interface.

ip-mtu-profile <ip-mtu-profile>

Attach IP MTU profile to an L3 interface.

mpls-admin-status <down | up>

Enable or disable MPLS.

mpls-mtu <mpls-mtu-size>

MPLS maximum transmission unit size.

neighbor <ipv4 | ipv6> <ip-address> mac <mac-address>

Configure a static IPv4 or IPv6 neighbor.

unnumbered interface <loopback-interface-name>

Configure an un-numbered interface.

vlan <outer-vlan-id>

Outer VLAN ID.

vlan-encapsulation <802.1ad | 802.1q>

Specify the outer VLAN tag.

Example 1: Logical Interface Configuration with IPv4 MTU Profile

 {
   "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "unit": [
          {
            "unit-id": 1,
            "description": "VLAN 101",
            "instance": "default",
            "ipv4-mtu-profile": "ipv4Mtu"

          }
        ]
      }
   ]
 }

Example 2: Logical Interface Configuration with IPv6 MTU Profile

 {
   "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "unit": [
          {
            "unit-id": 1,
            "description": "VLAN 101",
            "instance": "default",
            "ipv6-mtu-profile": "ipv6Mtu"

          }
        ]
      }
   ]
}

Example 3: Logical Interface Configuration with IP MTU Profile

 {
   "rtbrick-config:interface": [
      {
        "name": "ifp-0/0/1",
        "unit": [
          {
            "unit-id": 1,
            "description": "VLAN 101",
            "instance": "default",
            "ip-mtu-profile": "ipMtu"

          }
        ]
      }
   ]
}

Interface Address Configuration

This section describes how to configure interface IP addresses.

Syntax:

set interface <interface-name> unit <unit-id> address <afi> <attribute> <value>

Attribute Description

<afi>

Address family identifier (AFI). Supported values: ipv4 and ipv6

<prefix4|prefix6>

Assign IPv4 or IPv6 address to the interface unit.

community <community-value>

Configure list of communities associated with the address.

extended-community <community-value>

Configure list of extended communities associated with the address.

label <label-value>

Configure label associated with the address.
Supported MPLS label values are 0 - 1048575. The reserved MPLS label range is 0 - 15. In RBFS, BGP uses the label range 20000 - 100000. It is recommended to assign label values outside of these reserved ranges to avoid conflicts.

secondary true

Enable a secondary IPv4 address. Note: To disable the secondary IP configuration, use the delete form of the command.
Broadcast and network IP addresses cannot be configured on logical (IFL) interfaces.

Example: Interface Address Configuration

{
    "rtbrick-config:interface": [
      {
        "name": "lo-0/0/1",
        "unit": [
          {
            "unit-id": 1,
            "address": {
              "ipv4": [
                {
                  "prefix4": "198.51.100.103/24",
                  "label": 12346
                }
              ]
            }
          }
        ]
      }
    ]
  }

Global Interface Configuration

This section describes a configuration option applied globally to all interfaces.

Syntax:

set global interface all <attribute> <value>

Attribute Description

admin-status <up|down>

Configure state of the interface.

  • The interface level enable/disable command has higher precedence than the global interface enable/disable command.

  • You can disable all unused physical interfaces.

  • Before executing the global interface disable all command ensure that all physical interfaces are in the link Up state.

Example: Enabling or Disabling all Interfaces

{
  "ietf-restconf:data": {
    "rtbrick-config:global": {
      "interface": {
        "all": {
          "admin-status": "down"
        }
      }
    }
  }
}

Interface Holddown Configuration

By default, the Interface Hold-down feature is enabled. You can disable the feature if you want. It is allowed to configure timer value for options such as 'holddown down delay' and 'holddown up delay'.

Syntax:

set interface <interface-name> <attribute> <value>

Attribute Description

<interface-name>

Name of the interface. Example: ifp-0/0/1.

holddown <disable|enable>

Enable or disable the holddown timer feature. By default, the feature is enabled.

holddown-down-delay

Configure the down delay time in milliseconds (ms). The allowed values include 0, 100, 200, 500, 1000, 2000, 5000, and 10000. It overrides the default delay of '0' milliseconds.

holddown-up-delay

Configure the up delay time milliseconds (ms). The allowed values include 0, 100, 200, 500, 1000, 2000, 5000, and 10000.

Enabling Gratuitous ARP on Interface

You can enable or disable the sending of Gratuitous ARP (GARP) on an interface. This section describes enabling or disabling GARP on an interface.

Syntax:

set interface <interface-name> unit <unit-id> gratitous-arp <enable|disable>

Attribute Description

interface <interface-name>

Name of the interface. Examples: ifp-0/0/1.

gratitous-arp

Specify enable or disable.

Example: Enabling Gratitous ARP

supervisor@rtbrick.net: cfg> set interface hostif-0/0/0 unit 100 gratitous-arp enable|disable

Breakout Interfaces

Using the breakout configuration, the high-speed interfaces(100G) on the UfiSpace S9600-32X platform can be divided into multiple lower-speed interfaces(10G or 25G), allowing them to connect with lower rate ports(10G or 25G) in other network devices.

RBFS supports the following breakout configuration options:

Rate Technology Breakout Capable Electric Lanes Optical Lanes

100G

QSFP28

Yes

4 x 25G

4 x 25G

100G

QSFP28

Yes

4 x 10G

4 x 10G

For example, to set up the individual peer-side (25G/10G) interfaces, we need to configure the breakout on the 100G port to 4x10G or 4x25G based on the peer-side speed.

When a breakout is configured on a high-speed interface (for example, ifp-0/1/12), it is divided into four lower-speed interfaces: ifp-0/1/12#0, ifp-0/1/12#1, ifp-0/1/12#2, and ifp-0/1/12#3. As a result, the original high-speed interface will no longer be available.

The following diagram shows the breakout for the interface ifp-0/1/12 on the high-speed 100G to 4x25G ports (#0, #1, #2, #3).

breakout

Guidelines and Limitations

Currently, the following limitations exist.

  • Autonegotiation, speed configuration, and traffic sampling are not supported for breakout interfaces.

  • A breakout interface cannot be a member of a LAG.

  • Using breakout interfaces with routing protocols is not supported.

Configuring Breakout

To configure for the interface ifp-0/1/12 on the high-speed 100G to 4x25G ports., enter the following command:

Syntax:

set interface <interface-name> breakout map < 4x10G | 4x25G >

Example:

set interface ifp-0/1/12 breakout map 4x25G

Viewing Breakout Configuration Details

To view the breakout configuration details, enter the following command:

show interface <interface-name> detail

Example of Breakout Configuration:

supervisor@rtbrick>SPINE01: cfg> show interface ifp-0/1/12#0 detail
Interface: ifp-0/1/12#0
  Admin/Link/Operational status: Up/Up/Up
  Damping Status: holddown
  Breakout: 4x25G
  Speed configured: 25G
  Speed maximum: 25G
  Duplex: Full
  Encapsulation mode: ieee
  MRU: 10000
  MTU: 9000
  Maximum frame size: 10000
  Interface type: ethernet
  Interface index: 100609
  MAC: e8:c5:7a:8f:56:c6
  Uptime:  Mon Nov 25 05:37:10 GMT +0000 2024, Holddown up/down delay: 10000ms/0ms
  Flap count:  1
  Holddown transitions:  1
  Description: Physical interface #12 from node 0, chip 1
  LED Status: Off, LED programmed value: 82
  Packet statistics:
    Rx packets: 3          Tx packets: 3
    Rx bytes: 381          Tx bytes: 387
supervisor@rtbrick>SPINE01: cfg> show interface ifp-0/1
/12#1 detail
Interface: ifp-0/1/12#1
  Admin/Link/Operational status: Up/Up/Up
  Damping Status: stable
  Breakout: 4x25G
  Speed configured: 25G
  Speed maximum: 25G
  Duplex: Full
  Encapsulation mode: ieee
  MRU: 10000
  MTU: 9000
  Maximum frame size: 10000
  Interface type: ethernet
  Interface index: 102657
  MAC: e8:c5:7a:8f:56:c7
  Uptime:  Mon Nov 25 05:37:10 GMT +0000 2024, Holddown up/down delay: 10000ms/0ms
  Flap count:  1
  Holddown transitions:  1
  Description: Physical interface #12 from node 0, chip 1
  LED Status: Off, LED programmed value: 4
  Packet statistics:
    Rx packets: 1          Tx packets: 1
    Rx bytes: 127          Tx bytes: 129
supervisor@rtbrick>SPINE01: cfg> show interface ifp-0/1
/12#2 detail
Interface: ifp-0/1/12#2
  Admin/Link/Operational status: Up/Up/Up
  Damping Status: stable
  Breakout: 4x25G
  Speed configured: 25G
  Speed maximum: 25G
  Duplex: Full
  Encapsulation mode: ieee
  MRU: 10000
  MTU: 9000
  Maximum frame size: 10000
  Interface type: ethernet
  Interface index: 104705
  MAC: e8:c5:7a:8f:56:c8
  Uptime:  Mon Nov 25 05:37:10 GMT +0000 2024, Holddown up/down delay: 10000ms/0ms
  Flap count:  1
  Holddown transitions:  1
  Description: Physical interface #12 from node 0, chip 1
  LED Status: Off, LED programmed value: 0
  Packet statistics:
    Rx packets: 1          Tx packets: 1
    Rx bytes: 127          Tx bytes: 129
supervisor@rtbrick>SPINE01: cfg> show interface ifp-0/1
/12#3 detail
Interface: ifp-0/1/12#3
  Admin/Link/Operational status: Up/Up/Up
  Damping Status: stable
  Breakout: 4x25G
  Speed configured: 25G
  Speed maximum: 25G
  Duplex: Full
  Encapsulation mode: ieee
  MRU: 10000
  MTU: 9000
  Maximum frame size: 10000
  Interface type: ethernet
  Interface index: 106753
  MAC: e8:c5:7a:8f:56:c9
  Uptime:  Mon Nov 25 05:37:10 GMT +0000 2024, Holddown up/down delay: 10000ms/0ms
  Flap count:  1
  Holddown transitions:  1
  Description: Physical interface #12 from node 0, chip 1
  LED Status: Off, LED programmed value: 0
  Packet statistics:
    Rx packets: 1          Tx packets: 1
    Rx bytes: 127          Tx bytes: 129