lbailly/vxlan-router
Archived
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
Archived. We won't implement our own vtep and prefer the uses of linux vteps.
This repository has been archived on 2025-03-02. You can view files and clone it, but cannot push or open issues or pull requests.
12 commits 1 branch 0 tags 68 KiB
Rust 52.5%
Nix 47.5%
Find a file
catvayor f1efd2b5b6
fix(test): desactivate offload to have good csum
2025-02-01 23:33:19 +01:00
src feat: mac learning 2025-02-01 23:33:19 +01:00
vm fix(test): desactivate offload to have good csum 2025-02-01 23:33:19 +01:00
.gitignore test(vm/containers): build network topology with containers 2025-01-30 23:18:54 +01:00
Cargo.lock feat: vxlan parsing 2025-01-31 16:02:14 +01:00
Cargo.toml feat: vxlan parsing 2025-01-31 16:02:14 +01:00
readme.md feat: mac learning 2025-02-01 23:33:19 +01:00
shell.nix feat(test): script for simple connect to vm 2025-02-01 13:23:10 +01:00
vm.nix test: init vm config for mininet 2025-01-30 14:51:16 +01:00

readme.md

VXLAN-router

To use vxlan in isp, we need to implement a VTEP for vault01.

Roadmap

  1. Basic router which forward to every connected VTEP -> bad idea without mac learning
  2. Special case for when we are the target (internet, DNS), comes with a pseudo reverse path checking
  3. Mac learning
  4. DHCP, which removes user prefixes, comes with ARP snooping, filter auto attribution
  5. [external] Patch hostapd to use VXLAN
  6. [external] Maybe patch radius to use all VNIs (if there's a limitation)
  7. Allow packets to cross VNIs (from an external decision, cached)
  8. Synchronisation between multiple instances