Page MenuHomeFreeBSD

PR247718 - incorrectly drops IPv6 packets looping back on p2p interface
ClosedPublic

Authored by lutz_donnerhacke.de on Jul 5 2020, 12:43 PM.

Details

Summary

In PR247718 the problem occurs, that the IPv6 stack has a hard coded
loop avoidiance code for IPv6 only on P2P links. This behaviour does
not occur on IPv4 and causes packet loss for legitimate traffic, i.e.
VPN peers (Point2Miltipoint setup).

Test Plan

Apply the patch and let the bug reporter check, that the VPN is working.

Diff Detail

Repository
rS FreeBSD src repository
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.

Event Timeline

I don't believe we need specific processing here either.

  1. Kernel shouldn't try to be "smart" and decide whether user-configured routing is correct or not. It can't fix the issue anyway.
  2. Even when the route configuration is incorrect, TTL will still kick in
This revision is now accepted and ready to land.Jul 6 2020, 7:42 AM

@bz: any objects on removal of this one?

rgrimes added a subscriber: rgrimes.

This is a good fix, it is wrong for the kernel to try and guess that something is wrong here. "Highly Probably != For Certain" and having this kind of stuff get in the way when your trying to make something work is very annoying for the end user. Now if we could just get the hard coded interface loopback route creating/removal out of the kernel.

May somebody with commit rights push this into the kernel and may have a kind look on my other open reviews ...

Let's wait half a week to make sure we all agree on this one :-)

Ping? It's been considerably longer than just half a week, and there doesn't appear to be any opposition to this change.