[FEAT] traceroute support at autoresponder

[denlenin]

Hi Yeraze!
Is it possible to implement this feature in an autoresponder or some other way?

The Meshmonitor receives a command in a channel or private message like "!trace me (or node name)" -> do traceroute to this node -> sends the result in the same way as the request

[tylerpieper]

This would be neat, along with other node actions. But I don't know how feasible it is because traceroutes can take a while (or not complete at all) and the script engine may time out before it's finished.

[denlenin]

I think we need to set a clear timeout interval—1 minute (or at most 1 minute 30 seconds). After this time, the mesh monitor or script would send a "specified node did not respond" response, or, if successful, would send the result within this time interval. But we need to think about how to fit it into 200 bytes while still leaving it as readable as possible. Using only short names is probably a good idea, but then it's not always clear who's who.

[denlenin]

Another potential issue is the tracing queue. If this functionality is implemented, the node will be inundated with requests in the first few days. They need to be queued and processed correctly, possibly with a "request in queue" response.