Looks like a pretty good test. While using 3 hops i also was noticing it seems to like other qbit peers and could get very high speeds on one of them. i think you're also probably right about the piece distribution aspect, i've noticed similar things. i2cp works better in java routers than i2pd, cou...

i stick to librewolf/mullvad/tor browsers. i have a separate setup for non clearnet things, could do the same thing with VMs it's just more convenient for me to have a disposable laptop for darknets. i do the "paranoid level setup" and just block everything that's not needed

don't think so. may want to reduce the transit tunnels and amount of traffic you share if it's getting bogged down after an hour. see what the build % is and number of transit tunnels. shoot for at least 30-50% for non floodfill i2pd

if you only care about hops its length, quantity is number of tunnels

also, what happens if you go beyond 7 hops?

above was in response to that. really no reason to go that high

its in your i2pd.conf and tunnels.conf files, these params under each block

because i2pd, ran by russian speaking devs

i'm going to assume RTP was recommending using tor and i2p together like tor for clearnet/i2p for darknet or something, do not run i2p through tor

if you do this through i2cp i've found it just crashes i2pd so i wouldn't be surprised if this would also do that

i've thought about giving something like this a try to see how difficult it would be, but to deanon with one hop i think you could fully automate ip gathering without "too much" difficulty, it really just depends if the attackers consider i2p to be worth the effort. The automation aspect a...

this isn't an answer to why it does that, but snark won't go that fast. i'd be impressed if it hung around 2000 for a while. that's pushing it, i think, with 3 hops and 16 tunnels. snark will adjust dynamically to the number of tunnels it needs. you can allow it 16 tunnels up/down (you might as well...