Cisco 350XG series Administration Manual page 115

Administration: Stack Management
Unit Failure in Stack
NOTE
Cisco 350XG & 550XG Series 10G Stackable Managed Switches
• Reconnecting the Original Master Unit After Failover
Failure of Master Unit
If the Master fails, the backup unit takes over the master role and continues to
operate the stack normally.
For the backup to be able to take the place of the master, both units maintain a
warm standby at all times. In warm standby, the master and its backup units are
synchronized with the static configuration (contained in both the Startup and
Running configuration files). Backup configuration files are not synchronized. The
backup configuration file remains on the previous master.
Dynamic process-state information, such as the STP state table, dynamically-
learned MAC addresses, dynamically-learned Smartport types, MAC Multicast
tables, LACP, and GVRP are not synchronized.
When a master is being configured, it synchronizes the backup immediately.
Synchronization is performed as soon as a command is executed. This is
transparent.
If a unit is inserted into a running stack, and is selected as a backup unit, the master
synchronizes it so that it has an up-to date configuration, and then generates a
SYNC COMPLETE SYSLOG message. This is a unique SYSLOG message that
appears only when backup is converging with the master unit, and looks like this:
%DSYNCH-I-SYNCH_SUCCEEDED: Synchronization with unit 2 is finished
successfully.
Master/Backup Switchover
When a master fails on the backup unit, a switchover occurs.
The backup unit becomes the master, and all of its processes and protocol stacks
are initialized to take responsibility for the entire stack. As a result, there is
temporarily no traffic forwarding in this unit, but slave units remain active.
When STP is used and the ports are in link up, the STP port's state is temporarily
Blocking, and it cannot forward traffic or learn MAC addresses. This is to prevent
spanning tree loops between active units.
7
110