people must learn to never trust unknown links.
it is better to leave such sensitive content where it is posted by the admin.
ps: these links are all dead anyway
Message from Paris admin off dread ( source tor.taxi )
i2p is up and stable.
Money was burnt. But anything to fuck the DDOSERs mom. We don't give a single fuck what it takes. Dread stays up.
Edit (14th): More adjustments and cost savings!
We could have been just straight up and stable but watching money burn in the front of my eyes hurt too much. So I focused today on optimizing the circuit handshakes process (side stepping the single thread pool and spreading the requests over multiple independent tor process with basically the same memory state to prevent the SINGLE thread pool from blocking everything [TorProject why is there only... one... single... fucking... thread... pool for all this onion handshake shit? Modern cpus have cores for days. DAYS! fuck your cpuworker.c library. Spaghettied all over the place. GIVE ME TRUE THREADING. EAT THE COMPUTE. Just because people can container this shit to sidestep your asshole design doesn't mean this is okay. OKAY?])
Tor values have been tweaked. I did a solid for the Tor network too by statistically spreading the load over basically all the guards. Otherwise there would have been a lot more guard nodes dying all over the place. POW is coming and this attack will be just a bad time in the past. But right now we be optimizing.
Edit (16th): The Tor network is starting to have ripple effects:
Performance – Tor Metrics
metrics.torproject.org
While generally this won't result in failures of the circuits at this point. The extra circuit building load has shown to be extremely taxing on the network hurting the overall throughput. There is good news though. After a lot of testing I have a onionbalancing design which should be effective at equally spreading introduction the load over many fronts. With the limits of introduction points on a single V3 descriptor being 20 it's not enough to get uptime. Pushing distinct descriptors on the network is the only way to get these servers up under an attack like this. The current design of distinct descriptors leave a lot wanting because the load isn't as equally spread out as it needs to be. Specifically to introduction points that are controlled by me so no foolishness can happen (and descriptor rejections can be minimized/eliminated as much as possible). After I'm done with it there will be effectively minimal descriptor rotation even under excessive load.
This will be my crown achievement. Protocol layer flaw avoidance by overtaking the entire onion introduction process.
Yeah we can trust HeadJanitor and Paris for doing their job!
This is 100% LE, they are trying to prove that dread is alphabay and also before WHM.
This is not good at all.
they can change every week from hosting, but it is no solution.
The power they put in this DDoS is unseen, you don't want to know the price off this €€€€€€
why LE would do such a thing? i mean Dread is a source of information for LE too!
Guys i don't want to know the price of this Ddos... Insane
Double insane if it's LE...
why LE would do such a thing? i mean Dread is a source of information for LE too!
Guys i don't want to know the price of this Ddos... Insane
Double insane if it's LE...
you telling me this Ddos is actually paid by public funds?
what a shame
Governments have the most enormous capacity for DDoS attacks. It costs them nothing. Taxpayers' money won't suffer
But I doubt the government is doing it. More likely, someone wants a substantial share of the Dread forum traffic.
I agree with you HEISENBERG, I personally think Dread is useful for LE too. It's a big source of information for anyone on the dark net.
Maybe a giant phishing is coming haha
Our team brings together the best specialists from different fields.
We are ready to share our experience, discuss difficult issues and find new solutions.
Connect notifications to always stay in touch with the forum!
The BB Forum team is looking for cooperation: