Supported Platforms
Related Documentation
- SRX Series
- Understanding Chassis Cluster Redundancy Group Interface Monitoring
- Understanding Chassis Cluster Redundancy Group Failover
- Understanding Chassis Cluster Redundant Ethernet Interfaces
- Example: Configuring Chassis Cluster Minimum Links
- Additional Information
- Chassis Cluster Feature Guide for Security Devices
Understanding Chassis Cluster Redundant Ethernet Interface LAG Failover
To control failover of redundant Ethernet (reth) interfaces, it is important to configure the weights of interface monitoring according to the minimum-links setting. This configuration requires that the weights be equally distributed among the monitored links such that when the number of active physical interface links falls below the minimum-links setting, the computed weight for that redundancy group falls to zero or below zero. This triggers a failover of the redundant Ethernet interfaces link aggregation group (LAG) once the number of physical links falls below the minimum-links value.
Consider a reth0 interface LAG with four underlying physical links and the minimum-links value set as 2. In this case, a failover is triggered only when the number of active physical links is less than 2.
 | Note:
|
Configure the underlying interface attached to the redundant Ethernet LAG.
Specify the minimum number of links for the redundant Ethernet interface as 2.
Set up interface monitoring to monitor the health of the interfaces and trigger redundancy group failover.
The following scenarios provide examples of how to monitor redundant Ethernet LAG failover:
Scenario 1: Monitored Interface Weight Is 255
Specify the monitored interface weight as 255 for each underlying interface.
In this case, although there are three active physical links and the redundant Ethernet LAG could have handled the traffic because of minimum-links configured, one physical link is still down, which triggers a failover based on the computed weight.
Scenario 2: Monitored Interface Weight Is 75
Specify the monitored interface weight as 75 for each underlying interface.
In this case, when three physical links are down, the redundant Ethernet interface will go down due to minimum-links configured. However, the failover will not happen, which in turn will result in traffic outage.
Scenario 3: Monitored Interface Weight Is 100
Specify the monitored interface weight as 100 for each underlying interface.
In this case, when the three physical links are down, the redundant Ethernet interface will go down because of the minimum-links value. However, at the same time a failover would be triggered because of interface monitoring computed weights, ensuring that there is no traffic disruption.
Of all the three scenarios, scenario 3 illustrates the most ideal way to manage redundant Ethernet LAG failover.
Related Documentation
- SRX Series
- Understanding Chassis Cluster Redundancy Group Interface Monitoring
- Understanding Chassis Cluster Redundancy Group Failover
- Understanding Chassis Cluster Redundant Ethernet Interfaces
- Example: Configuring Chassis Cluster Minimum Links
- Additional Information
- Chassis Cluster Feature Guide for Security Devices
Published: 2015-02-27
Supported Platforms
Related Documentation
- SRX Series
- Understanding Chassis Cluster Redundancy Group Interface Monitoring
- Understanding Chassis Cluster Redundancy Group Failover
- Understanding Chassis Cluster Redundant Ethernet Interfaces
- Example: Configuring Chassis Cluster Minimum Links
- Additional Information
- Chassis Cluster Feature Guide for Security Devices
