© Copyright Lenovo 2016 Chapter 18: Stacking 299If multiple stack links or stack Member switches fail, thereby separating the Masterand Backup into separate sub‐stacks, the Backup automatically becomes an activeMaster for the partial stack in which it resides. Later, if the topology failures arecorrected, the partial stacks will merge, and the two active Masters will come intocontact.In this scenario, if both the (original) Master and the Backup (acting as Master) arein operation when the merger occurs, the original Master will reassert its role asactive Master for the entire stack. If any configuration elements were changed andapplied on the Backup during the time it acted as Master (and forwarded to itsconnected Members), the Backup and its affected Members will reboot and will bereconfigured by the returning Master before resuming their regular roles.However, if the original Master switch is disrupted (powered down or in theprocess of rebooting) when it is reconnected with the active stack, the Backup(acting as Master) will retain its acting Master status to avoid disruption to thefunctioning stack. The deferring Master will temporarily assume a role as Backup.If both the Master and Backup are rebooted, all member switches in the stack willalso reboot. When the switches resume operation, they will assume their originallyconfigured roles.If, while the stack is still split, the Backup (acting as Master) is explicitlyreconfigured to become a regular Master, then when the split stacks are finallymerged, the Master with the lowest MAC address will become the new activeMaster for the entire stack.Merging Independent StacksIf switches from different stacks are linked together in a stack topology withoutfirst reconfiguring their roles as recommended, it is possible that more than oneswitch in the stack might be configured as a Master.Although all switches which are configured for stacking and joined by stackinglinks are recognized as potential stack participants by any operational Masterswitches, they are not brought into operation within the stack until explicitlyassigned (or “bound”) to a specific Master switch.Consider two independent stacks, Stack A and Stack B, which are merged into onestacking topology. The stacks will behave independently until the switches inStack B are bound to Master A (or vice versa). In this example, once the Stack Bswitches are bound to Master A, Master A will automatically reconfigure them tooperate as Stack A Members, regardless of their original status within Stack B.However, for purposes of future Backup selection, reconfigured Masters retaintheir identity as configured Masters, even though they otherwise act as Members.In case the configured Master goes down and the Backup takes over as the newMaster, these reconfigured Masters become the new Backup. When the originalconfigured Master of the stack boots up again, it acts as a Member. This is one wayto have multiple backups in a stack.
