Date: Fri, 15 Feb 2019 17:50:47 +0100 From: "Peter G." <freebsd@disroot.org> To: freebsd-hackers@freebsd.org Subject: Fwd: Point-to-point using GRE over IPv6 -> not possible with a single /128 address on the server? Message-ID: <10c08655-bfa5-3ed9-3688-3f78028cb033@disroot.org> In-Reply-To: <95d8e3ea-af36-4d14-f280-908f92a96515@disroot.org> References: <95d8e3ea-af36-4d14-f280-908f92a96515@disroot.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, I've got issues establishing a point-to-point ipv6-over-ipv6 GRE link. IPv4 which works as expected: this end: 10.0.1.10 other end: 10.0.2.10 GW: 10.0.1.1 iface: em0 >ifconfig gre4 create >ifconfig gre4 inet 10.0.1.10 10.0.2.10 netmask 0xffffffff tunnel 10.0.1.10 10.0.2.10 tunnelfib 2 >route add -host 10.0.1.1 -iface em0 -fib 2 >route add -host 10.0.2.10 10.0.1.1 -fib 2 Works. The tunnel is marked with FIB 2, and a point-to-point is established. Can be used with IPSEC in transport or whatever. Now, IPv6 is problematic. The server has allocated a single IPv6 address with prefixlen 112. This could be the source of the issue. Private addresses replicate the setup. this end: fc01:e::100/112 other end fc02:e::200 GW: fc01:e::1 >ifconfig em0 #em0: # inet6 fc01:e::100 prefixlen 112 This works. Default GW is at fc01:e::1. Now the GRE tunnel >ifconfig gre6 create >ifconfig gre6 inet6 fc01:e::100 fc:02:e::200 tunnelfib 6 #ifconfig: ioctl (SIOCAIFADDR): File exists Why is this not possible? Isn't the logic behind it the same as with IPv4? If not, why not? Does this mean it is not possible to have a point-to-point using IPv6 on a machine with only a single /128 address? Found this as reference: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=208173 but what he did was on a much broader range. Many thanks! Peter
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?10c08655-bfa5-3ed9-3688-3f78028cb033>