logoalt Hacker News

A Botnet Accidentally Destroyed I2P

107 pointsby Cider9986today at 1:08 AM63 commentsview on HN

Comments

gnabgibtoday at 2:14 AM

This seems to lack the full story, despite the headline.. Krebs' coverage is more in-depth (39 points) https://news.ycombinator.com/item?id=46976825

jjmarrtoday at 2:43 AM

From the main article, I2P has 55,000 computers, the botnet tried to add 700,000 infected routers to I2P to use it as a backup command-and-control system.

https://news.ycombinator.com/item?id=46976825

This, predictably, broke I2P.

show 1 reply
kace91today at 2:40 AM

Man, I feel so out of depth with cybersecurity news.

Why does i2p (per the article) expect state sponsored attacks every February? Where are those forming from, what does the regularity achieve?

How come the operators of giant (I’m assuming illegal) botnets are available to voice their train of thought in discord?

show 2 replies
charcircuittoday at 6:25 AM

>hostile nodes

>they accidentally disrupted I2P while attempting to use the network as backup command-and-control infrastructure

So were they hostile or were they using it normally?

shevy-javatoday at 7:44 AM

> The I2P development team responded by shipping version 2.11.0 just six days after the attack began.

Not wanting to be overly critical, but any net-infrastructure project kind of has to keep bot-attacks in mind and other attack vectors, in the initial design stage already. Any state-actor (and other actors, though I would assume it is often a state financing the bot network behind-the-scene) can become potentially hostile.

Roark66today at 9:24 AM

Is there a shittier summary anywhere, please? Or did the author reached the peak of enshittification?

Honestly, did the bot implementation have bugs or was it a proper implementation that crashed the network due to sheer numbers?

Also, how does changing the encryption standard affect anything if the bots tried to integrate correctly with the network?

Is the problem "fixed" or is it not? Elsewhere I found large number if botnet devs got pissed off with this botnet operator and 600k nodes went offline. Might this have much more to do with the situation getting better than simply changing encryption?

Also, was there any suggestion a quantum breaking attack was attempted? No. So why put the emphasis on "post quantum" in this article?

Bad. Very bad.

pmontratoday at 6:20 AM

This seems to be a better post about what happened, from the same site https://www.sambent.com/i2p-2-11-0-ships-post-quantum-crypto...

show 2 replies
rollulustoday at 7:52 AM

This article (with high slop vibes) and another article on their site (linked in the comments) seem to suggest that post quantum encryption mitigated the Sybil attack, without explanation. I fail to understand how the two are even related.

hoppptoday at 4:48 AM

Isn't I2P java? The botnet uses java? I thought python or C is preferred for that kinda stuff

show 3 replies
illusive4080today at 3:09 AM

Why does Discord allow a server for a botnet owner?

show 4 replies
Cider9986today at 5:25 AM

The video seems to be a bit more in-depth.

richardfeytoday at 4:29 AM

I wonder how cjdns would have handled this

cookiengineertoday at 5:46 AM

This was one of the worst writeups I ever read. Even a LinkedIn Premium post would have had more technical details, lol