< Home

(Optional) Advanced Configurations

This section describes advanced cluster configurations.

Procedure

  • Enable cluster traffic flow aggregation in the system view.

    cluster traffic-aggregation enable

    By default, the cluster flow aggregation function is disabled.

    By default, this command is executed on the management master device. After running the cluster standby config enable command on the master device, you can also run the command on the management slave device.

    For TCP proxy, SSL proxy, mail proxy, and content security services, the forward and reverse paths must be consistent. Otherwise, the services may be interrupted. When using these functions in the cluster, you can use the following methods to ensure consistent forward and return paths:
    • (Recommended) Deploy proper networking.
    • Enable the cluster traffic aggregation function.
    Enabling the traffic aggregation function affects device performance. Therefore, ensure that the forward and return paths of traffic are consistent and the traffic aggregation function is disabled.

    After enabling the cluster traffic aggregation function, run the cluster session fast-sync enable command to enable the fast session synchronization function.

  • Configure the cluster member monitoring function.

    Monitoring affects member devices' health ratings. If a monitored device goes Down, the device health rating decreases, triggering the master/backup switchover in the business group. The recommended monitored objects are channel links of cluster members and upstream and downstream service interfaces.

    Configure the following commands on each member device:

    • Run the cluster track ip-link ip-link-name [ weight weight-value ] command to monitor an IP-link.
    • Run the cluster track interface interface-type interface-number [ weight weight-value ] command to monitor an interface.
    • Run the cluster track trunk-member enable command to monitor Eth-Trunk member interfaces.

      Before running the cluster track trunk-member enable command, run the cluster track interface eth-trunk command first.

    • Run the cluster track bfd-session local-discr-value [ weight weight-value ] command to track a BFD session.

  • Run the cluster switch master to node-id command in the system view to specify the management master device of the cluster.

    This command takes effect only once after being executed. For example, device A is specified as the management master device. After device A restarts and is added to the cluster again, device A performs cluster negotiation according to the normal process. Generally, device A functions as the management slave device.

    By default, this command is executed on the management master device. After running the cluster standby config enable command on the master device, you can also run the command on the management slave device.

Copyright © Huawei Technologies Co., Ltd.
Copyright © Huawei Technologies Co., Ltd.
< Previous topic