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


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


CloudEngine 12800, 12800E, 8800, 7800, 6800, and 5800 Series Switches VXLAN Best Practices

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).
Reliability Design

Reliability Design

Network Redundancy

Faults caused by various factors are inevitable, so fault recovery technologies are important. Device-level redundancy is an important method to improve network reliability.

On the hardware distributed VXLAN network using the gateway/spine/leaf three-layer architecture shown in Figure 1, device-level redundancy is used to resolve network faults.

  1. Server link fault: When one server is dual homed to two leaf nodes, the NICs on the server work in load balancing or active/standby mode. If one server link fails, services are switched to the redundant link or standby link.
  2. Leaf/Border leaf node fault: When M-LAG (recommended), stack, or SVF is configured for the leaf or border leaf nodes, services are switched to the other leaf or border leaf node if one node fails.
  3. Leaf uplink fault: When spine and leaf nodes are connected through multiple links to implement ECMP, services are switched to the redundant link if one uplink fails.
  4. Spine node fault: If one spine node fails, traffic is forwarded by the other spine node.
  5. LB and firewall fault: When LBs and firewalls are configured to work in active/standby mode, traffic is switched to the standby LB and firewall if the active LB and firewall fail.
  6. Stack link or peer-link fault: If a stack link or peer-link fails, DAD triggers all interfaces on the standby device except the management network interface, peer-link interface, and stack interface to enter the Error-Down state. This prevents the problem of dual master devices and improves the network reliability.
Figure 2-43 Device redundancy to resolve network faults

Network Loop Prevention

If loops occur on a network, a large number of link bandwidth resources are consumed, leading to service interruption. Loop prevention measures need to be deployed on the network to improve reliability.

Devices (including vSwitches on servers) with the Layer 2 forwarding function are deployed on the network shown in Figure 2. Loops may occur if the configurations are incorrect or devices are incorrectly connected.

The recommended prevention measures are as follows:

  • Configure interface-based traffic suppression. The recommended suppression rates for broadcast packets, multicast packets, unknown unicast packets, and ARP packets are 1000 pps, 200 pps, 1%, and 200 pps, respectively.
  • Send alarms when the number of MAC address flappings detected in a BD reaches the specified value. MAC address flapping detection is enabled on devices by default.
Figure 2-44 Network loops

Updated: 2018-07-02

Document ID: EDOC1100004176

Views: 17699

Downloads: 550

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