Plan for [Stretch] Network Attack Simulation (#57)

[Fatumayattani]

Hi all,

Following up on Issue #57, here’s a proposed plan for the network attack simulation work.

Initial Focus

Start with an Eclipse attack by poisoning the DHT. The idea is to:

• Build a small test cluster (≈10 honest nodes + 2–4 malicious nodes)
• Write reproducible scripts to simulate the attack
• Measure key metrics:
  • Lookup failures / incorrect routing
  • Peer table pollution
  • Recovery time
• Document findings and suggest possible mitigations

Questions for Feedback

• Should the first PR focus solely on Eclipse simulation, or also include DHT poisoning in parallel?
• Would you prefer this work be integrated with libp2p/interop, or kept local in py-libp2p for now?

Looking forward to thoughts and guidance.
cc @seetadev

[seetadev]

@Fatumayattani : Wonderful initiative. Appreciate your efforts.

CCing @acul71 , @yashksaini-coder, @lla-dane, @sumanjeet0012 and @bomanaps for their thoughts and feedback. We will arrive at a design implementation plan for this very important effort.

[Fatumayattani]

@seetadev
Thank you so much for the encouraging words 🙏 I really appreciate the support. I’ll wait to hear thoughts from the team and refine the plan accordingly. Excited to collaborate on shaping the design and making steady progress on this important work!