Hello everyone! There is a known problem with routing headers in IPv6, namely that for the inversion of the routing header to be done automatically on the reply, there has to be authentication between the endpoints, with AH. This causes a number of problems, and I'd like to know how this was handled in the Linux kernel, if anyone has any close experience with it. The thing is I'm developing an academic project that requires the reversion to be working properly, but so far I've been unable to get it going. If I send the packets unauthenticated, the kernel (correctly) responds directly to the origin, without adding any routing header to the reply packages. The solution would be to activate AH, but the implementation of AH in IPv6 seems to be seriously lacking in projects like OpenSWAN, so I haven't been able to test that. Basically, I need either: a) a way to get around the AH requirement b) better, and *provided the kernel supports the reversal when AH is present* (which I haven't been able to test), a quick way to get AH working, even if it's just with a bunch of static shared secrets Any information in any of this would be greatly appreciated! Thank you, Sérgio Gomes - : send the line "unsubscribe linux-net" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
