VIGIL MESH

Documentation

Verify connectivity

An enrolled machine is only truly ready when it is reachable. This page describes four simple checks to run in order: confirm the assigned address, reach a peer, verify MagicDNS resolution and watch the real-time topology in the console. Each one isolates a different layer, so if something goes wrong you know where to look.

Verify the assigned address

First check: did the machine receive an address on the mesh network? At enrollment, each device is assigned a stable address in the 100.64.0.0/10 range. Stable means it does not change at every reboot: you can rely on it in your configurations.

  • On the machine, list the network interfaces: the VIGIL-MESH interface must carry an address starting with 100.64.
  • In the console, the device appears with active status and the same address next to its name.
  • If no address appears, the enrollment did not complete: go back to the corresponding step.

Reach a peer

An address is not enough: two machines have to actually talk to each other. Pick another active device on the same network and test its reachability from the machine you just enrolled.

  1. 1
    Spot a peerIn the console, note the address (100.64.x) or the MagicDNS name of another active device on the same network.
  2. 2
    Test reachabilityFrom the new machine, send a ping to the peer's address. A reply confirms the path is established.
  3. 3
    Test a real serviceBeyond ping, try to reach a service exposed by the peer (for example SSH, a share or an internal web page) to validate a concrete use.

MagicDNS resolution

Working with numeric addresses is tedious. MagicDNS gives each device a readable name and resolves it automatically inside your workspace: you reach your machines by name, without maintaining a hosts file or a separate DNS server.

  • From the new machine, resolve the MagicDNS name of a peer: it must return its 100.64.x address.
  • Reach that peer by its name (ping or service) rather than by its address, to confirm the resolution is properly in effect.
  • If the address is reachable but the name is not, the problem is on the name resolution side, not connectivity: that helps target the diagnosis.

See the real-time topology in the console

The console gives an overview that local tests do not: the device inventory, their status and the network topology. It is the place to check at a glance that the new machine is properly seen by the rest of the workspace.

  • The device shows up in the inventory, with active status, with its address and its MagicDNS name.
  • The topology shows its links to the peers of the same network.
  • The status indicators tell you whether a peer is online and since when.

What to do if it does not work

If one of the checks fails, proceed by elimination, starting again from the lowest layer. The first check that fails tells you where to look.

  • No 100.64.x address: the enrollment is not finalized, or the device is suspended, revoked or expired. Check its status in the console and re-enroll if needed.
  • Address present but peer unreachable: check that the peer is active and that the access policies do allow this link; retry after a few seconds, giving the direct path time to establish.
  • Address reachable but name not resolved: the issue is on the MagicDNS side, not connectivity; check the machine's name resolution configuration.
  • Nothing shows up in the console: wait one refresh cycle (about 30 seconds) before drawing conclusions.
Read nextBuild and understand your mesh network