We have 15+ networks with 1500+ nodes, which will grow to 5000+ over the next 12 months. In managing large networks, we’ve found a few improvements to the dashboard that would be helpful at scale.
Top level Networks Page:
- Allow networks to be grouped/nested by usages/type/etc. I have networks for specific uses or systems. For example, I have a couple development networks, 6 networks for OSPF areas, public networks, etc. It would be great if I could group or nest them. Then I could name the group, OSPF Areas for example, then better utilize the network name. Right now, I have to group them by adding a use to the name. Which works, but my number of networks grows regularly, so at scale, this will get messy. You only display a small amount of characters of the network name. We often exceed this, so it gets cumbersome to manage from a display perspective.
Individual network page > Members list:
- Allow for me to search members by physical IP or managed IP. This always comes up while troubleshooting hardware. We track IPs in our infrastructure management as thats whats used by our tooling and automation. Not node ID. So, it can be troublesome to find a node if we only know an IP, the device is offline, so we can’t get a node ID, and I’ve got a list of 1200+ members to search through. We’ve started adding IP to description so it’s searchable, but that takes up character space that we could use for actual descriptions. Names can vary, so name isn’t always a bulletproof search term.
- Allow me to sort by last seen or version. We regularly audit our connections and remove connections that haven’t been used in X timeframe to maintain healthy license counts. Sorting by last seen would make it very easy to find stale connections. Sorting by version would also be helpful to develop a list for an update push across 1000s of devices that aren’t touched for months.
- Allow me to export the member list, full or filtered, to csv. This will allow us to better manage 1000s of hosts for updates, or reporting on certain groups of devices. We sort of do this today with the API, but it’s a pain, when the dashboard already has all the information as we need it.
- Allow me to import a csv of hosts and node IDs to add and authorize many members at once. We get hardware manufactured in large batches. Zerotier is installed as part of the image and the node IDs, serials #s, names, etc are captured during the imaging process at our manufacturer. They provide a CSV with all necessary info. We then have to run this through a script that uses the api to add and authorize each node. This works, but is another few steps in the process. Having the ability to quickly format a ZT import CSV from that data, and clicking an import button would avoid a few additional steps to create an import function on our side.