You can as the others have noted - but usually it is easier to just stack the switches together so they become fault tolerant for each other with no tricky config.
In a warm spare MS configuration, VRRP will be leveraged to exchange heartbeats and determine if the warm spare should take over when the primary fails. But if you just create a physical stack, and stagger your devices (especially any dual-homed devices) across multiple switches in a physical stack, that tends to meet the majority of use-cases. You can also do cross-stack LACP so if you're uplinking/downlinking to other switches or switch stacks, that works too. Double check this doc https://documentation.meraki.com/MS/Layer_3_Switching/MS_Warm_Spare_(VRRP)_Overview to see if it's really what your requirements dictate as opposed to physical stacking, and let us know if something's not clear.
Well one big difference is that with warm spare the devices can be further away from each other. Stacking is limited to 3m due to the need for stacking cables.
That's true on models which have dedicated stack ports. For models that have flexible stacking, technically a valid optics configuration can work as well.
Aside from this caveat, physical stacking provides a better redundancy and more flexibility than a warm spare (VRRP) configuration. In addition to opening the door to cross-stack LAGs (which can eliminate STP blocked links), a switch stack can also use dynamic routing and multicast routing, which is not supported in a warm spare configuration.
Get notified when there are additional replies to this discussion.
