Interface Management Configuration
Introduction
Section titled “Introduction”Ethernet interfaces are generally used as service interfaces for service transmission and are sometimes also called ports.
Numbering sequence
Section titled “Numbering sequence”The Ethernet interface numbering method of this series of products is: EthernetX . X represents the interface number, which increases from 1. It should be noted that the naming intervals of different series of interfaces are different. For enterprise-level products with 100G interfaces, the naming interval of 100G interfaces is 4. Taking CX206P-48S-MH as an example, the first 100G interface numbered 49 is named: Ethernet49, the second 100G interface is named Ethernet53, and so on.
Basic interface configuration
Section titled “Basic interface configuration”Table 1 Basic interface configuration
| Operation | Command | Description |
|---|---|---|
| Enter the system configuration view | configure | |
| Enter the interface view | interface Ethernet ID | |
| Set interface description | description String | |
| Set interface speed | speed {1000|10000|25000} | Different types of interfaces support different configuration parameters. Please execute the speed ? command in the relevant interface view to view specific options. |
| Enable interface | no shutdown | By default, the Ethernet interface is in the enabled state. Execute the shutdown command to disable the interface service function and set the interface physical state to down . Note: The power supply of the POE power supply interface is not affected by this command. |
| Modify interface FEC mode | fec {rs|fc} | By default, the FEC mode is disabled. The FEC modes at both ends of the link must be consistent. Otherwise, the interface will not be Up. |
| Enable the POE power supply function of the interface | poe enable | By default, the POE power supply function of the interface is disabled. The poe enable command disables the PoE power supply on the interface. |
| Configuring Interface Loopback | loopback | The physical interface loops back, and the port becomes up after the loopback. The traffic output from the loopback port will re-enter the loopback port. |
| Tunnel stripping | decape {gre|vxlan|erspan|cfp|gtp_u|pppoe|mpls|ipinip} | Strip the tunnel of the traffic coming from the port and select the type of tunnel to strip. |
| Inner parser | inner-parser {gre\|gtp_u} | Perform inner-layer parsing on the tunnel traffic entering from the port; currently supports only GRE and GTP packets. |
| Port delayed start | startup-delay (1..600) | Set the delay in seconds before the port starts |
| Port Splitting | breakout {4x25G|4x10G|4x1G} | Split a port into four 25G/10G/1G channels |
| Enable the PTP function on the port | ptp enable | Enable port PTP |
Maintenance interface
Section titled “Maintenance interface”Table 2 Maintenance interface
| Operation | Command | Description |
|---|---|---|
| View interface description | show interface description | |
| View interface POE supply information | show interface poe [detail] | |
| View interface status | show interface status | |
| View interface module information | **show interface transceiver eeprom dom ** [ethernet ID] | |
| View interface statistics: | **show counters interface ** [ethernet ID] | |
| Clear interface statistics | clear counters interface |
Tunnel stripping
Section titled “Tunnel stripping”Tunnel stripping means that the traffic received from the ingress will be stripped before being forwarded to the egress.
Precautions
Section titled “Precautions”When tunnel stripping and ACL filtering are used at the same time, stripping is performed first and then ACL filtering. For example, if the original message carries a GRE tunnel, after GRE stripping is enabled, the inner five-element information is matched (the inner VLAN does not support filtering). For the Ethernet layer that is completed after stripping the tunnel , only the source and destination MAC 00:00:00:00:00:00 can be used as the filtering condition.
Tunnel stripping basic configuration
Section titled “Tunnel stripping basic configuration”Tunnel stripping takes effect on the input port of the traffic
Table 3 Tunnel stripping basic configuration
| Operation | Command | Description |
|---|---|---|
| Enter system configuration view | configure | |
| Enter interface view | interface Ethernet ID | |
| Tunnel stripping | decape {gre|vxlan|erspan|cfp|gtp_u|pppoe|mpls|ipinip} | Strip the tunnel of the traffic coming from the port and select the type of tunnel to strip |
Tunnel inner parsering
Section titled “Tunnel inner parsering”Inner-layer tunnel parsing refers to matching and hashing based on the inner layer of tunnel traffic received at the ingress port.
Precautions
Section titled “Precautions”Currently, only the inner layer parsing of GRE/GTP packets is supported, and the packet structure must meet the requirements:
1. GRE does not have extended fields such as seq and key.
2. GTP does not have a seq extension field.
After enabling the inner-layer parsing, the tunnel packets that meet the requirements support the following matching fields. Tunnel packets that do not meet the requirements or non-tunnel packets still support L2/EX/L3/L3V6 matching of the outer-layer fields.
Table 4 Supported rule matching fields for GRE/GTP packets
| Packet type | Supported rule matching fields | Description |
|---|---|---|
| GRE+ETH+IP, GRE +IP | EX(inner five-tuple), L2 default forward | |
| GRE+ETH+IPV6, GRE +IPV6 | CX102S-M：L3/L3V6 default forward Other device：EX(inner five-tuple), L2 default forward, not support L3/L3V6 | |
| GTP+IP | EX(inner five-tuple), L2 default forward | |
| GTP+IPV6 | CX102S-M：L3/L3V6 default forward Other device：EX(inner five-tuple), L2 default forward, not support L3/L3V6 |
After enabling inner-layer parsing, the tunnel packets that meet the requirements support hashing based on the inner five-tuple, but not based on smac or dmac. It does not affect the hashing based on the outer fields for other packets.
After enabling inner-layer parsing, this ethernet does not support configuring tunnel stripping, does not support configuring the four types of custom source and destination symmetric hashing(ipv4-symmetric-src-ip, ipv4-symmetric-dst-ip, ipv6-symmetric-src-ip, ipv6-symmetric-dst-ip), and it is not recommended to enable the global MPLS parsing.
Tunnel inner parsering basic configuration
Section titled “Tunnel inner parsering basic configuration”Tunnel inner parsering takes effect on the input port of the traffic
Table 5 Tunnel inner parsering basic configuration
| Operation | Command | Description |
|---|---|---|
| Enter system configuration view | configure | |
| Enter interface view | interface Ethernet ID | |
| Inner parser | inner-parser {gre\|gtp_u} | Perform inner-layer parsing on the tunnel traffic entering from the port; currently supports only GRE and GTP packets |
Load balancing hash algorithm configuration
Section titled “Load balancing hash algorithm configuration”Custom hash key basic configuration
Section titled “Custom hash key basic configuration”The custom hash takes effect on the input port of the traffic, and the custom hash key is used when the traffic from this port is forwarded to the output LAG. 16 custom hash algorithms can be configured.
Table 6 Creating hash group
| Operation | Command | Description |
|---|---|---|
| Enter system configuration view | configure | |
| Creating a custom hash | hash mode alias NAME | Create a custom hash group and name it |
| Add a hash key to a custom hash group | hash key {src-ip|dst-ip|src-mac|dst-mac|src-port|dst-port|mac-symmetric|ipv4-symmetric|ipv6-symmetric|l4- symmetric|ipv4-symmetric-src-ip|ipv4-symmetric-dst-ip|ipv6-symmetric-src-ip|ipv6-symmetric-dst-ip}* |
hash key description:
Select a key as the load balancing condition. Symmetric represents symmetric hashing and does not care about the source and destination. Mac-symmetric indicates the symmetric hash of the source MAC and the destination MAC, ipv4-symmetric applies symmetric hashing to the source and destination IPv4 addresses, ipv6-symmetric applies symmetric hashing to the source and destination IPv4 addresses and l4-symmetric performs a symmetric hash over the source and destination transport-layer (Layer 4) ports. ipv4-symmetric-src-ip is usually used with ipv4-symmetric-dst-ip on different ports, as long as the ports fetch the same ip address, it will be loaded to the same output port, and the key is not used with any other key on the same port. Ipv6-symmetric-src-ip is usually used in conjunction with ipv6-symmetric-dst-ip on different ports, as long as the ports fetch the same ip address, it will be loaded onto the same output port, and the key is not used with any other key on the same port.
Table 7 Binding hash groups
| Operation | Command | Description |
|---|---|---|
| Enter system configuration view | configure | |
| Enter interface view | interface Ethernet ID | Enter the entry to be bound |
| Applying a custom hash group | hash mode alias NAME | Bind the hash mode to the input port |
QinQ message analysis
Section titled “QinQ message analysis”QinQ message parsing is not enabled during traffic forwarding , the message will only be parsed up to QinQ . After QinQ message parsing is enabled, the traffic will continue to be parsed and can match the five-tuple information of the message.
Table 8 QinQ message analysis
| Operation | Command | Description |
|---|---|---|
| Enter system configuration view | configure | |
| Enter interface view | interface Ethernet ID | Enter the traffic input port where the QinQ message is to be parsed |
| Configure the protocol number of QinQ packets | qinq protocol ether_type | ether_type eg : 0x88a8 0x9100 |
| Enable qinq message parsing | qinq enable | Enable qinq message parsing on the port |