H3C S5120-EI Series Configuration Manual

Also see for S5120-EI Series: Configuration guide Installation manual Configuration guide Operation manual Configuration guide

H3C S5120-EI Series Configuration Manual Manual pdf 14 page image

6Configuration synchronization mechanismIRF uses a strict running-configuration synchronization mechanism so all chassis in an IRF fabric can workas a single node, and after the master fails, other members can operate normally.In an IRF fabric, all chassis get and run the running configuration of the master. Any configuration youhave made is propagated to all members.When you execute the save [ safely ] [ backup | main ] [ force ] command or the save file-url allcommand, the system saves the running configuration, as follows:• If the configuration auto-update function (the slave auto-update config command) is enabled, savesthe configuration as the startup configuration on all member switches for the next startup.• If the configuration auto-update function is disabled, saves the configuration as the startupconfiguration on the master for the next startup.By default, configuration auto-update is enabled.For more information about configuration management, see Fundamentals Configuration Guide.Master electionMaster election is held each time the IRF fabric topology changes, for example, when the IRF fabric isestablished, a new member device is plugged in, the master device fails or is removed, the IRF fabricsplits, or IRF fabrics merge.Master election uses the following rules in descending order:1. Current master, even if a new member has higher priority.When an IRF fabric is being formed, all member switches consider themselves as the master, andthis rule is skipped2. Member with higher priority.3. Member with the longest system uptime.4. Member with the lowest bridge MAC address.The IRF fabric is formed on election of the master.During an IRF merge, the switches of the IRF fabric that fails the master election must reboot to re-join theIRF fabric that wins the election.After a master election, all subordinate switches reboot with the configuration on the master. Theirconfiguration files do not take effect.IRF multi-active detectionAn IRF link failure causes an IRF fabric to split in two IRF fabrics operating with the same Layer 3configurations, including the same IP address. To avoid IP address collision and network problems, IRFuses multi-active detection (MAD) mechanisms to detect the presence of multiple identical IRF fabrics,handle collisions, and recover from faults.Multi-active handling procedureThe multi-active handling procedure includes detection, collision handling and failure recovery.