High Availability with VSF and VSX
VSF and VSX high availability
VSF stacks ArubaOS-Switch devices into one logical switch with a commander and standby, using VSF links. It simplifies management and provides link-level redundancy, but requires compatible models/firmware and a split-detection method. VSX pairs two ArubaOS-CX switches in an active-active pair. Each switch runs its own control plane but shares a common data-path view through the ISL; VSX LAGs span both peers. A separate keepalive link detects peer liveness and helps split-brain handling. VSX uses primary/secondary roles for control, but both peers forward traffic. VSX requires matching firmware, a working ISL (often a LAG), management reachability, and consistent LAG configuration.
Traps: running the VSX keepalive over the ISL; mixing incompatible VSF members; ignoring split-brain scenarios; and assuming VSX makes STP unnecessary everywhere. Also verify orphan ports and that the ISL has enough bandwidth for failover traffic.
Worked example/rule of thumb: If the VSX ISL fails but the keepalive path remains up, the pair can detect that the peer is alive and take split-brain action rather than both forwarding independently. Rule: dedicate a separate physical path for keepalive, test ISL failure, and confirm VSX LAGs stay active-active only while synchronization is healthy.