#9930 made the Xray TUN inbound always request ::/0 in autoSystemRoutingTable
so that IPv6 stops bypassing the tunnel. That only helps a host which actually
holds a globally routable IPv6 address. On any other host it does harm.
With IPv6 disabled the TUN device gets no IPv6 address at all, the kernel
rejects the route with EACCES and the whole inbound fails to start:
Failed to start: app/proxyman/inbound: failed to start proxy > proxy/tun:
failed to add system route ::/0 > permission denied
With IPv6 enabled but no global address the route is accepted and the host
gains an IPv6 default route it cannot use. The TUN completes the TCP handshake
locally before dialing the outbound, so IPv6 destinations start to look
reachable and get picked, and the connection then dies at the outbound instead
of failing fast (#10051).
Neither host has IPv6 traffic that could bypass the tunnel, so ::/0 buys them
nothing. Detect a global IPv6 address once while building the config context
and drop ::/0 when there is none. Link-local and unique local addresses do not
count: they never reach the IPv6 internet.
Co-authored-by: liuclare <177657698+liuclare@users.noreply.github.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
The rule added in #9897 takes the TUN inbound's `address` verbatim as `ip_cidr`,
so a /30 or /126 interface prefix becomes the match range.
sing-tun derives the TUN's DNS entry from the address right after the interface's
own and hands it to the system resolver: Windows through luid.SetDNS in
tun_windows.go, Linux through systemd-resolved in tun_linux.go, both guarded only
by AutoRoute && !EXP_DisableDNSHijack. HasNextAddress keeps that address inside
the interface prefix, every preset in Global.TunIPv4Address is a /30 and every
IPv6 preset a /126, and the sing-box system stack rejects single-address
prefixes, so there is no configuration where it falls outside.
Queries from the system resolver then hit the drop rule and time out with no
response and no ICMP. Name resolution fails for the whole system while the proxy
path itself stays healthy, which makes it read as a DNS outage rather than a
routing rule. Reported in #9934 and #9926.
Matching each address on its own keeps what #9897 set out to block - the loop it
diagnosed was addressed to the interface address itself - and leaves the DNS
entry to sing-box.
Also restores the two regression tests #9897 came with, removed by eff58459
(#9817) while its implementation and template fix stayed in place.
ShouldRejectTrafficToTunOwnAddresses now asserts the single-address form and
additionally pins the prefix length, so it covers both the loop it was written
for and the resolver address it must not cover.
Verified on Linux by running sing-box directly from a generated config, changing
only this rule's prefix length between runs:
ip_cidr ["172.18.0.1/30"] getent hosts www.google.com -> empty, 3/3
ip_cidr ["172.18.0.1/32"] getent hosts www.google.com -> resolved, 3/3
dig against a public resolver, naked-IP HTTPS and the local mixed port were
unaffected in both runs. End to end, a build of this branch emits
drop ip_cidr ["172.18.0.1/32"] and system resolution works while its TUN is up.
Co-authored-by: liuclare <177657698+liuclare@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
* Always route IPv6 into Xray TUN regardless of EnableIPv6Address
EnableIPv6Address controls whether the TUN interface is assigned an IPv6
address, but it also gated whether ::/0 was added to autoSystemRoutingTable.
With the default (false), IPv6 had no route pointing at the TUN device and
followed the system default route instead, leaving the tunnel unproxied and
exposing the host's real IPv6 address.
The embedded template SampleTunInbound already declares both families; the
generated config discarded it. #9843 restored ::/0 only inside the
EnableIPv6Address == true branch, so the false branch still leaks.
Route both families unconditionally and let the option control only the
interface address. The same conditional existed a second time in the
RouteExcludeAddress branch and is fixed as well.
Fixes#9929
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* Add regression tests for IPv6 routing in the Xray TUN inbound
Both assertions fail on 31044f44 and pass with the fix:
Tun_ShouldRouteIPv6IntoTunnel(enableIPv6Address: False)
Expected collection {"0.0.0.0/0"} to contain "::/0".
TunRouteExcludeAddress_ShouldIncludeIPv6Ranges
Expected collection {...44 IPv4 ranges...} to have an item matching x.Contains(:).
The theory also covers enableIPv6Address: true, which passes on both revisions,
so the tests only fail while the defect is present. The gateway count assertion
pins the intended split of responsibilities: EnableIPv6Address governs the
interface address, never the routing table.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: liuclare <177657698+liuclare@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
