Cloud management and user traffic have different roles
Meraki Dashboard provides cloud-based configuration, visibility and administration for network components. This does not mean that all user traffic passes through Dashboard. When designing distributed offices or retail branches, assess management connectivity, local access and inter-site data paths separately. Decide which teams can administer each network, which settings are centrally controlled and what local support owns. The diagram therefore does not add Dashboard as an inline data-plane device. A useful cloud-management model does more than provide a common panel: it also makes ownership of configuration changes and the method of observing service behavior understandable across the participating teams.
Read hub and spoke roles in MX Auto VPN
Auto VPN manages VPN connectivity between supported WAN appliances in the same Dashboard organization. Hubs connect to other hubs and to spokes that select them; spokes connect to their configured hubs. This does not mean every appliance creates a direct tunnel to every other appliance. The official Standard Hub & Spoke example below shows the data center, MPLS and internet transports, and RS-A and RS-B. Dashed paths describe overlay relationships, while solid lines distinguish physical connectivity. No direct spoke-to-spoke tunnel absent from the source has been added. Choose a topology after identifying which central or cloud resources applications need to reach.
Meraki Auto VPN: Standard Hub & Spoke
Select nodes to read their roles; zoom or enter full screen to follow the connections.
- Physical link
- MPLS virtual path
- Internet virtual path
Select a node to read its role. Drag to pan or use the buttons to zoom.
DC, MPLS, internet, RS-A and RS-B follow the official Standard Hub & Spoke example. Solid lines distinguish physical connections from dashed MPLS/internet overlay paths. Dashboard is not added as an inline user-traffic hop.
Choose subnet participation and tunneling behavior
Explicitly select the local networks that participate in VPN. Split tunneling sends traffic for participating VPN subnets through the overlay, while other traffic follows the relevant local routing decision. Full tunneling is a separate choice that directs applicable traffic through a hub. Rather than choosing it based only on a “more secure” label, evaluate internet access, central capacity and application dependencies together. Do not assume that management traffic follows the same rules as client traffic. Acceptance should separately observe branch-to-center access, internet egress and prohibited subnet communication. This makes the effect of the routing choice on user experience and security scope visible.
Use a common panel without merging design problems
Meraki MS switching and MR/CW wireless options can share an administration experience, but wired access and wireless coverage are different design problems. Evaluate VLAN and uplink arrangements, PoE needs, RF conditions, client density and authentication together. Branch templates can simplify operations without eliminating physical differences between sites. A design that works in one store should not be assumed to produce identical results in a building with different construction or user density. Site surveys and pilot measurements remain important. Visibility from a common panel becomes a more useful operational record when it is checked against actual application transactions and the conditions at the site.
Evaluate MV and MT against the observation need
The catalog also includes Meraki MV cameras and MT sensors within the portfolio. Rather than treating them simply as additional network-device counts, determine which area needs observation and for what purpose. Assign ownership, access permissions and retention requirements for camera or sensor data through the organization's relevant operational decisions. Check connectivity, power and device-administration conditions during technical design. Adding all of these components to the MX Auto VPN topology would misrepresent its source, which does not show them. When a separate site requirement arises, evaluate supported devices and placement options within that scope and define acceptance criteria for the intended observation.
Check model, release and mode in Catalyst convergence
Supported Catalyst switching platforms have Cloud Management with IOS XE options. Do not assume every Catalyst device or all existing CLI features can move unchanged into cloud management. Compare the hardware model, current software, licensing and target administration mode with official compatibility and migration documentation. Examine how management addressing, SVIs and uplinks behave in the target mode. Firmware changes can introduce additional conditions for reverting. The conversion plan should therefore include configuration records, management-access checks and recovery steps. Evaluating the Catalyst–Meraki shared experience with these boundaries in view creates a realistic expectation of what the transition will deliver.
Test redundancy through bidirectional application access
Two uplinks or multiple hubs at a branch do not demonstrate that all failures are handled identically. Evaluate hub order, advertised subnets, uplink behavior and return paths together. A one-way branch-to-center reachability check is insufficient; inspect application responses and, where relevant, communication initiated from the center. Include real operating conditions such as a transport failure, appliance restart or loss of management connectivity in the test scope. Measure behavior for the chosen architecture rather than publishing a fixed takeover time without evidence. Recording configuration and load conditions alongside measurements makes the result reproducible and useful when the environment changes.
Build rollout and support around the actual sites
For distributed locations, installation order and coordination with local teams matter alongside appliance readiness. Keep the site inventory, device assignment, uplink details, target template and acceptance scenarios in one rollout plan. When pilot findings are applied elsewhere, recheck physical differences between sites. Handover should explain Dashboard permissions, change procedures, alarm ownership and support escalation. Trustnet's catalog describes assessment, design, transition and operations as the project approach for this preparation. A successful result extends beyond every appliance appearing online: users need access to required applications, and support teams need a clear way to investigate problems using evidence from the relevant site.



