No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

CX11x, CX31x, CX710 (Earlier Than V6.03), and CX91x Series Switch Modules V100R001C10 Configuration Guide 12

The documents describe the configuration of various services supported by the CX11x&CX31x&CX91x series switch modules The description covers configuration examples and function configurations.
Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Principles of Smart Link

Principles of Smart Link

This section describes the principles of Smart Link.

Basic Concepts

Smart Link improves reliability by implementing backup between two interfaces. The two interfaces constitute a Smart Link group. One interface is called master interface and the other is called slave interface. Smart Link also uses Flush packet, Smart Link instance, and control VLAN to implement fast switchover and load balancing.

Figure 11-70 Networking of Smart Link
Smart Link Group

A Smart Link group consists of two interfaces: master interface and slave interface. Normally, the master interface is in active state, and the slave interface is in inactive state.

As shown in Figure 11-70, Interface1 and Interface2 on SwitchD form a Smart Link group.

Master and Slave Interfaces

As shown in Figure 11-70, Interface1 is the master interface and Interface2 is the slave interface in a Smart Link group.

When the two interfaces in a Smart Link group are both in Up state, the master interface is in active state and the slave interface is in inactive state. When the link of the master interface becomes faulty, the slave interface becomes active and the master interface becomes inactive.

Flush Packet

When a switchover occurs between active and standby links of a Smart Link group, existing forwarding entries no longer apply to the new topology. All the MAC address entries and ARP entries on the network need to be updated. Therefore, the Smart Link group needs to send Flush packets to notify other devices so that the other devices can update their MAC address entries and ARP entries. In Figure 11-70, when a switchover occurs, SwitchD sends multicast Flush packets to request SwitchA, SwitchB, and SwitchC to update their MAC address entries and ARP entries.

Control VLAN
  • Control VLAN for sending Flush packets

    A Smart Link group uses this control VLAN to broadcast Flush packets. As shown in Figure 11-70, if SwitchD is enabled to send Flush packets, it broadcasts Flush packets in the control VLAN when a switchover occurs.

  • Control VLAN for receiving Flush packets

    The upstream devices use this control VLAN to receive and process Flush packets. As shown in Figure 11-70, upstream devices (SwitchA, SwitchB, and SwitchC) are able to identify Flush packets and enabled to receive Flush packets. When traffic is switched between links, the upstream devices process the received Flush packets and then update the MAC address entries and ARP entries.

Load Balancing

Smart Link supports load balancing based on multiple instances matching different VLANs. When the active and standby links of a Smart Link group work normally, Smart Link allows these two links to forward different data traffic. In load balancing mode, both the two interfaces are active. The slave interface forwards traffic from the VLANs associated with configured load balancing instances and the master interface forwards traffic from other VLANs. When a link fails, the Smart Link group automatically switches all traffic to the other link.

Smart Link Instance

The standby link of a Smart Link group is bound to different instances to implement load balancing. Smart Link can reference Multiple Spanning Tree Protocol instances (MSTIs). Each MSTI is bound to multiple VLANs, and MSTIs are bound to different VLANs.

Implementation

This section uses the network shown in Figure 11-71 as an example to describe how Smart Link works when both links are working properly, a link fails, and the link fault is rectified.

Figure 11-71 Networking of Smart Link
Both Links Are Working Properly

Interface1 (master interface) and Interface2 (slave interface) on SwitchD form a Smart Link group. When both uplinks work normally, the master interface is in active state, and the link connected to the master interface is the active link. The slave interface is in inactive state, and the link connected to the slave interface is the standby link. As shown in Figure 11-72, data is transmitted along the active link. No loop exists on the network, so broadcast storms will not occur on the network.

Figure 11-72 Links are working properly
The Active Link Fails

As shown in Figure 11-73, when the active link on SwitchD is faulty, the master interface (Interface1) transits to inactive state, and the slave interface (Interface2) transits to active state. Existing MAC address entries and ARP entries on some devices no longer apply to the network. A mechanism is required to update the MAC address entries and ARP entries. Two mechanisms are available: through Flush packets and through traffic-triggered automatic update.

Figure 11-73 Active link fails

Sending Flush Packets to Request Upstream Devices to Update Entries

This mechanism applies to scenarios where upstream devices (for example, SwitchA, SwitchB, and SwitchC in Figure 11-73) support the Smart Link function. To implement fast switchover, SwitchD is enabled to send Flush packets and all interfaces on the upstream devices are enabled to receive Flush packets.

  1. After a switchover is complete on SwitchD, SwitchD sends a Flush packet through the new active link (Interface2).
  2. When an upstream device receives a Flush packet, it checks whether the control VLAN carried by this Flush packet is allowed by the interface. If the VLAN is not allowed by the interface, the device directly forwards the Flush packet. If the VLAN is allowed by the interface, the device processes the Flush packet and updates the MAC address entries and ARP entries.

When SwitchA receives a data packet destined for SwitchD, SwitchA forwards this packet according to the updated MAC address entries or ARP entries.

Triggering Entry Update through Traffic

This mechanism is applicable to scenarios where upstream devices (including non-Huawei devices) do not support the Smart Link function. These devices update MAC address entries and ARP entries when they receive uplink traffic.

In Figure 11-74, if SwitchD does not transmit traffic to trigger the update of MAC address entries and ARP entries on SwitchA, SwitchA forwards data packets to SwitchD through Interface3. However, the packets cannot reach SwitchD, causing traffic loss. Traffic can be forwarded to SwitchD correctly until the original MAC address entries or ARP entries on SwitchA are aged out.

Figure 11-74 Active link fails

In Figure 11-75, after the link becomes faulty, MAC address entries and ARP entries learned by Interface1 are deleted. If SwitchD needs to send traffic, it must broadcast ARP packets first. When SwitchA receives traffic through Interface4, it updates its MAC address entries and ARP entries. When SwitchA receives a data packet destined for SwitchD, it forwards this packet through Interface4. The packet is then forwarded to SwitchD by SwitchC.

Figure 11-75 Active link fails

This mechanism improves entry update efficiency because upstream devices can update their MAC address entries and ARP entries before the entries are aged out. Generally, traffic is switched to the standby link in milliseconds, which minimizes traffic loss.

The Active Link Recovers

Interface1 remains blocked after the original active link recovers. To switch traffic to the original active link, use either of the following methods:

  • Enable revertive switching on SwitchD. When the original active link recovers, Smart Link waits until the WTR timer expires and then automatically switches the traffic back to the original active link.
  • Use a command to forcibly switch the traffic back to the original active link.

In Figure 11-71, revertive switching is enabled on SwitchD. When the link of Interface1 recovers, traffic is switched back to Interface1 after the WTR time is reached. When you use a command to forcibly switch the traffic back to the original active link, Interface2 is blocked and changes to inactive state immediately, while interface1 changes to active state.

Load Balancing

In dual-homing networking, only one uplink is in forwarding state and the other link does not forward traffic. Therefore, the link utilization is only 50%. Smart Link supports load balancing instances. The standby link transmits traffic from VLANs associated with specified load balancing instances. In this way, traffic from different VLANs is load balanced between the two links.

Figure 11-76 Networking of Smart Link load balancing

As shown in Figure 11-76, when no load balancing instance is configured, SwitchD forwards all packets through the active link connected to Interface1. To improve link use efficiency, configure a load balancing instance on SwitchD and associate VLAN 300 to VLAN 400 with the instance. Then data packets sent from these VLANs are forwarded through Interface2. VLAN 100 to VLAN 200 are not associated with the load balancing instance, so packets sent from these VLANs are forwarded through Interface1. In this way, traffic from different VLANs is load balanced between the two links.

Smart Link NSR

A master/slave switchover triggered by a system fault or performed by a network administrator when upgrading software or maintaining the system interrupts routing and causes traffic loss. Smart Link non-stop routing (NSR) can address this problem.

NSR ensures uninterrupted traffic transmission if a fault occurs on the control plane and a backup control plane is available to take over the traffic. During the switchover, the fault is transparent to the control plane of a neighbor.

Smart Link NSR uses the following approaches to synchronize data between switch modules in real time:
  1. Smart Link backs up the link status and instance forwarding status of interfaces but does not back up interface names or interface roles on the slave switch module. Interface names and interface roles can be restored on the slave switch module using the source data.
  2. If a master/slave switch module switchover is performed due to a fault on the device, the new switch module delivers the instance forwarding status of interfaces, ensuring that traffic is not interrupted.
Translation
Download
Updated: 2019-08-09

Document ID: EDOC1000041694

Views: 58407

Downloads: 3621

Average rating:
This Document Applies to these Products
Related Version
Related Documents
Share
Previous Next