Hello,

It seems that I’m not here for enough time, I can’t find your thread.
What were your issues on IPv6? I hit some from migrating to 4.19 (from
4.4) on routers, so I’m still on 4.4 for now.

We discussed it a bit on bird ML:
https://bird.network.cz/pipermail/bird-users/2019-June/013509.html
https://bird.network.cz/pipermail/bird-users/2019-November/013992.html
https://bird.network.cz/pipermail/bird-users/2019-December/014011.html

(sorry for the multiple links, it seems that the archive is split by
months)

By chances, are we hitting the same bug?

Regards,
Alarig Le Lay

On mer. 29 janv. 15:31:20 2020, Captain Wiggum wrote:

(resending without html.:)
I started the thread.
We are using 4.19.x and 4.9.x, but for reference I also tested then current 5.x.
I believe we got it all worked out at the time.
--John Masinter


On Wed, Dec 18, 2019 at 2:00 PM Stephen Suryaputra [off-list ref] wrote:

quoted

I am curious: what kernel version are you testing?
I recall that several months ago there is a thread on TAHI IPv6.
Including the person who started the thread.

Stephen.

On Thu, Oct 24, 2019 at 7:43 AM Levente [off-list ref] wrote:

quoted

Dear list,


We are testing IPv6 again against the test specification of ipv6forum.

https://www.ipv6ready.org/?page=documents&tag=ipv6-core-protocols

The test house state that some certain packages doesn't arrive to the
device under test. We fail test cases

V6LC.1.2.2: No Next Header After Extension Header
V6LC.1.2.3: Unreacognized Next Header in Extension Header - End Node
V6LC.1.2.4: Extension Header Processing Order
V6LC.1.2.5: Option Processing Order
V6LC.1.2.8: Option Processing Destination Options Header

The question is that is it possible that the this is the intended way
of operation? I.e. the kernel swallows those malformed packages? We
use tcpdump to log the traffic.


Thank you for your help.

Levente