jsabrahams / traceroute-map Goto Github PK

It would be nice to also have an orthographic projection.
Not so great for screenshots, but more fun to interact with.

Some items might not be that interesting to look at, or may be a bit privacy sensitive.
We should add command line options for what is actually logged and saved (see also #12 ),
and what is displayed on screen (we may want to log things which we do not wish to immediately display).

How many ms it took for a packet to travel on each link would also give some interesting insight. We have this information, as this information is generated when we construct a trace.

Once we have been assigned an IP address, it should be possible to get a geographical location of where we are based on said IP (or VPN server, if you're running a VPN).
This would make the traces complete, as they would actually start at the host location.

We should firstly increase the max delay for the get trace functions to 2 seconds (it is now 1 second).
However, if we still fail to generate a trace, we should relay this information to the user.

Some sort of timeline would be nice, perhaps with a window size, such that we can see what packets were sent when.

A file where we save our traces, with all accompanying information, which we can at a later date load and display.

Everyone is doing it

Instead of passing args, we instead extract the arguments from args and pass these to function with named arguments.
This makes it possible to also at a later date call the function programatically, for instance if we want to imbed it somewhere such as a Jupyter notebook.

This list would be treated as a list of IP addresses we retrieve while sniffing.
As such, this would be filtered, the traces would be generated, and this would be displayed.

The generated figure is a bit bare.
It would be nice to have some title text (and perhaps a description somewhere if that is possible).

This is quite a significant change. I have seen that poltly graphs can be live-updated, but I'm not entirely sure how. Tho a different library is possible, I would prefer to keep using plotly as it has quite extensive functionality which would be nice to have in future.

Having real-time captures would however be cool to have as, for instance, a dashboard somewhere. This would also require that:

• The width of traces be scaled dynamically as total amount of data grows over time

Depending on the amount of hits for a certain IP address, we should increase the thickness of a trace.

Related to #11

Per connection, it would be nice if we can see how much data was sent over time.
Perhaps also with a time series displayed onscreen, which could be a command line option.

The log can contain sensitive information.
Therefore, instead of a log being created each time alongside the source, it would be nice to specify that we do not want to log this information (though we might want to still display it).
In addition, it would also be nice if we could specify an output directory.

For each trace node, it would be nice to display the IP address if possible.
It would be nice to, per trace, make not of how much data flowed on said path (just as #2 is a visual representation of this)