Completely new Meraki setup

[Kenny_M8]

Hello team,

As someone who only has experience with Meraki WiFi and classic Cisco solutions, I need some advice regarding new Meraki setup. We plan to implement completely new network and because the building is still in renovation phase we would like to onboard the devices, form the stacks, do the upgrade if necessary and so on and that way prepare as much as possible in advance for installation. Perform some kind of staging.

My plan was to power on the devices connect them to the uplink (they will obtain IP from DHCP) and get them online.

Is it recommended while devices are still in this staging area to configure them over the dashboard? They will get all the information like hostnames, VLANs, subnets and so on, but will also need to have different static mgmt addresses. When you change mgmt VLAN and address and apply it to the device I suppose that device will lose connectivity with dashboard?

Or is it better option to wait and when devices are installed in the racks and everything is connected, then to do initial configuration over the local status page and set IP addresses and VLAN tagging manually, because we don't have any device with DHCP on site, only P2P link to ISP. After device is online in the dashboard apply configuration as planed. 

I suppose that with second option you have smaller chances for error and potential tshoot on the day of installation.

All advices are welcome 😁

[TyShawn]

I've seen two methods with Meraki and for the most part, both can work depending on your customer and business practices.

1. Drop-Ship

This method is when you preconfigured the devices in the dashboard before the units arrive onsite. This is great when you want to get everything configured and ready to go and trust your field techs to be able to connect the devices and be able to handle any one-off issue that could occur in the field. I use this method if I am shipping small deployments like Z3 / Z4 units to my sites or remote users.

2. Pre-Configure

I use this method when deploying larger installations to ensure all devices are working before arriving on-site. This allows me to do the extra things like labeling, true testing, firmware updates, and a small burn-in. This method helps me the most as I can focus on working with my techs on delivery vs configuring on the fly while trying to turn over a site.

At the end of the day, the process method you choose to develop is the one that best fits your company and customer's use case. 

Pro-tip

If you have multiple sites that will follow a repeatable process I would recommend looking at templates to save even more time. 

[cmr]

@Kenny_M8 why do they need static management addresses? Old style syslog server or authorisation servers?

If you can set them up as a set couldn't you have all the addressing with only the MX WAN being 'wrong' at this point?