Simple definition
MLAG (Multi-Chassis Link Aggregation) lets a device build one link aggregation across two separate physical switches that operate as a logical pair.
Technical definition
The two MLAG switches synchronize the state required for the bundle, typically through a peer link and a control or keepalive channel. The attached device uses LACP as it would with a normal link aggregation and usually needs no vendor-specific multi-chassis protocol.
How it works / role
MLAG provides both chassis redundancy and active use of multiple links. The peers must coordinate MAC, LACP, and often VLAN state to avoid duplicate forwarding, traffic blackholes, or split-brain behavior if the peer link fails.
What is it used for?
Remove the single-switch failure point while keeping aggregated bandwidth and fast failover for connected servers or network devices.
Practical example
A server has two LACP interfaces, one connected to switch A and the other to switch B, with both switches forming the same MLAG pair.
Common issues
- Peer link or keepalive failure
- Split-brain between the peer switches
- Inconsistent VLAN or LACP configuration
- Orphan ports handled incorrectly during a failure
Key takeaway: MLAG spans multiple chassis, but it still depends on reliable state synchronization between the two peer switches.