packet filter on repeaters

[on1arf]

Hi,

We have a serious problem here on the Belgian coast. From time to time, we receive a lot of messages from England which causes the network here to get saturated.
The trick with the region-code does not work as the rest of the country does not apply it, so it would create an "Island" of connectivty here cut of from the rest of Belgium.
As allmost all the traffic we receive come from other repeaters (not users), the easiest option is to filter on what packets are accepted to be sent further the network.

Are there plans to implement filters on the repeaters (Packet::readFrom) in the official firmware, to filter on (say) HashSize, patterns in the path (filter out certain repeaters) or payload-type that would allow the repeater-admins here reduce the amount of unwanted traffic and stabilize the network?

Already blocking 1-byte traffic would already help a lot as England stil runs on 1-byte, while Belgium has most of the repeaters upgrade to 2 bytes.

Kr.

[marvac01]

What is your objective here?
Would you like to block traffic over the English Channel, so that the rest of the country could not reach the English south east coast and the folks there could not communicate with Belgians?

[on1arf]

Yes, I agree. It is a shame that this will block Belgians to communicate with people on the English south-coast considering with the saturation we currently have here, they have the luxary of having at least one packet in 20 actually getting through .. if they are lucky
hahaha :-)

OK. Seriously.
As explained, the goal is to give the repeater admins the tools to intervene on network-level if needed.
In the end, the goal is to have a network that actually works for the people who invest the time and money to create the network for their region.

[on1arf]

Side note. Just blocking 1byte packets is one of the things that can help.

As it turns out that 45 % of the traffic on our network here are message-type 0 and message-type 1 but, if you ask why that traffic is there and what it is used for, nobody seems to be able to explain it (*), .. we can simply block that and see what happens.

That would already reduce the load on our network.

(*) #2764