Files
tailscale/wf/firewall.go
T
Will Norris 3ec5be3f51 all: remove AUTHORS file and references to it
This file was never truly necessary and has never actually been used in
the history of Tailscale's open source releases.

A Brief History of AUTHORS files
---

The AUTHORS file was a pattern developed at Google, originally for
Chromium, then adopted by Go and a bunch of other projects. The problem
was that Chromium originally had a copyright line only recognizing
Google as the copyright holder. Because Google (and most open source
projects) do not require copyright assignemnt for contributions, each
contributor maintains their copyright. Some large corporate contributors
then tried to add their own name to the copyright line in the LICENSE
file or in file headers. This quickly becomes unwieldy, and puts a
tremendous burden on anyone building on top of Chromium, since the
license requires that they keep all copyright lines intact.

The compromise was to create an AUTHORS file that would list all of the
copyright holders. The LICENSE file and source file headers would then
include that list by reference, listing the copyright holder as "The
Chromium Authors".

This also become cumbersome to simply keep the file up to date with a
high rate of new contributors. Plus it's not always obvious who the
copyright holder is. Sometimes it is the individual making the
contribution, but many times it may be their employer. There is no way
for the proejct maintainer to know.

Eventually, Google changed their policy to no longer recommend trying to
keep the AUTHORS file up to date proactively, and instead to only add to
it when requested: https://opensource.google/docs/releasing/authors.
They are also clear that:

> Adding contributors to the AUTHORS file is entirely within the
> project's discretion and has no implications for copyright ownership.

It was primarily added to appease a small number of large contributors
that insisted that they be recognized as copyright holders (which was
entirely their right to do). But it's not truly necessary, and not even
the most accurate way of identifying contributors and/or copyright
holders.

In practice, we've never added anyone to our AUTHORS file. It only lists
Tailscale, so it's not really serving any purpose. It also causes
confusion because Tailscalars put the "Tailscale Inc & AUTHORS" header
in other open source repos which don't actually have an AUTHORS file, so
it's ambiguous what that means.

Instead, we just acknowledge that the contributors to Tailscale (whoever
they are) are copyright holders for their individual contributions. We
also have the benefit of using the DCO (developercertificate.org) which
provides some additional certification of their right to make the
contribution.

The source file changes were purely mechanical with:

    git ls-files | xargs sed -i -e 's/\(Tailscale Inc &\) AUTHORS/\1 contributors/g'

Updates #cleanup

Change-Id: Ia101a4a3005adb9118051b3416f5a64a4a45987d
Signed-off-by: Will Norris <will@tailscale.com>
2026-01-23 15:49:45 -08:00

567 lines
15 KiB
Go

// Copyright (c) Tailscale Inc & contributors
// SPDX-License-Identifier: BSD-3-Clause
//go:build windows
// Package wf controls the Windows Filtering Platform to change Windows firewall rules.
package wf
import (
"fmt"
"net/netip"
"os"
"github.com/tailscale/wf"
"golang.org/x/sys/windows"
"tailscale.com/net/netaddr"
)
// Known addresses.
var (
linkLocalRange = netip.MustParsePrefix("fe80::/10")
linkLocalDHCPMulticast = netip.MustParseAddr("ff02::1:2")
siteLocalDHCPMulticast = netip.MustParseAddr("ff05::1:3")
linkLocalRouterMulticast = netip.MustParseAddr("ff02::2")
linkLocalMulticastIPv4Range = netip.MustParsePrefix("224.0.0.0/24")
linkLocalMulticastIPv6Range = netip.MustParsePrefix("ff02::/16")
)
type direction int
const (
directionInbound direction = iota
directionOutbound
directionBoth
)
type protocol int
const (
protocolV4 protocol = iota
protocolV6
protocolAll
)
// getLayers returns the wf.LayerIDs where the rules should be added based
// on the protocol and direction.
func (p protocol) getLayers(d direction) []wf.LayerID {
var layers []wf.LayerID
if p == protocolAll || p == protocolV4 {
if d == directionBoth || d == directionInbound {
layers = append(layers, wf.LayerALEAuthRecvAcceptV4)
}
if d == directionBoth || d == directionOutbound {
layers = append(layers, wf.LayerALEAuthConnectV4)
}
}
if p == protocolAll || p == protocolV6 {
if d == directionBoth || d == directionInbound {
layers = append(layers, wf.LayerALEAuthRecvAcceptV6)
}
if d == directionBoth || d == directionOutbound {
layers = append(layers, wf.LayerALEAuthConnectV6)
}
}
return layers
}
func ruleName(action wf.Action, layerID wf.LayerID, name string) string {
switch layerID {
case wf.LayerALEAuthConnectV4:
return fmt.Sprintf("%s outbound %s (IPv4)", action, name)
case wf.LayerALEAuthConnectV6:
return fmt.Sprintf("%s outbound %s (IPv6)", action, name)
case wf.LayerALEAuthRecvAcceptV4:
return fmt.Sprintf("%s inbound %s (IPv4)", action, name)
case wf.LayerALEAuthRecvAcceptV6:
return fmt.Sprintf("%s inbound %s (IPv6)", action, name)
}
return ""
}
// Firewall uses the Windows Filtering Platform to implement a network firewall.
type Firewall struct {
luid uint64
providerID wf.ProviderID
sublayerID wf.SublayerID
session *wf.Session
permittedRoutes map[netip.Prefix][]*wf.Rule
}
// New returns a new Firewall for the provided interface ID.
func New(luid uint64) (*Firewall, error) {
session, err := wf.New(&wf.Options{
Name: "Tailscale firewall",
Dynamic: true,
})
if err != nil {
return nil, err
}
wguid, err := windows.GenerateGUID()
if err != nil {
return nil, err
}
providerID := wf.ProviderID(wguid)
if err := session.AddProvider(&wf.Provider{
ID: providerID,
Name: "Tailscale provider",
}); err != nil {
return nil, err
}
wguid, err = windows.GenerateGUID()
if err != nil {
return nil, err
}
sublayerID := wf.SublayerID(wguid)
if err := session.AddSublayer(&wf.Sublayer{
ID: sublayerID,
Name: "Tailscale permissive and blocking filters",
Weight: 0,
}); err != nil {
return nil, err
}
f := &Firewall{
luid: luid,
session: session,
providerID: providerID,
sublayerID: sublayerID,
permittedRoutes: make(map[netip.Prefix][]*wf.Rule),
}
if err := f.enable(); err != nil {
return nil, err
}
return f, nil
}
type weight uint64
const (
weightTailscaleTraffic weight = 15
weightKnownTraffic weight = 12
weightCatchAll weight = 0
)
func (f *Firewall) enable() error {
if err := f.permitTailscaleService(weightTailscaleTraffic); err != nil {
return fmt.Errorf("permitTailscaleService failed: %w", err)
}
if err := f.permitTunInterface(weightTailscaleTraffic); err != nil {
return fmt.Errorf("permitTunInterface failed: %w", err)
}
if err := f.permitDNS(weightTailscaleTraffic); err != nil {
return fmt.Errorf("permitDNS failed: %w", err)
}
if err := f.permitLoopback(weightTailscaleTraffic); err != nil {
return fmt.Errorf("permitLoopback failed: %w", err)
}
if err := f.permitDHCPv4(weightKnownTraffic); err != nil {
return fmt.Errorf("permitDHCPv4 failed: %w", err)
}
if err := f.permitDHCPv6(weightKnownTraffic); err != nil {
return fmt.Errorf("permitDHCPv6 failed: %w", err)
}
if err := f.permitNDP(weightKnownTraffic); err != nil {
return fmt.Errorf("permitNDP failed: %w", err)
}
/* TODO: actually evaluate if this does anything and if we need this. It's layer 2; our other rules are layer 3.
* In other words, if somebody complains, try enabling it. For now, keep it off.
* TODO(maisem): implement this.
err = permitHyperV(session, baseObjects, weightKnownTraffic)
if err != nil {
return wrapErr(err)
}
*/
if err := f.blockAll(weightCatchAll); err != nil {
return fmt.Errorf("blockAll failed: %w", err)
}
return nil
}
// UpdatedPermittedRoutes adds rules to allow incoming and outgoing connections
// from the provided prefixes. It will also remove rules for routes that were
// previously added but have been removed.
func (f *Firewall) UpdatePermittedRoutes(newRoutes []netip.Prefix) error {
var routesToAdd []netip.Prefix
routeMap := make(map[netip.Prefix]bool)
for _, r := range newRoutes {
routeMap[r] = true
if _, ok := f.permittedRoutes[r]; !ok {
routesToAdd = append(routesToAdd, r)
}
}
var routesToRemove []netip.Prefix
for r := range f.permittedRoutes {
if !routeMap[r] {
routesToRemove = append(routesToRemove, r)
}
}
for _, r := range routesToRemove {
for _, rule := range f.permittedRoutes[r] {
if err := f.session.DeleteRule(rule.ID); err != nil {
return err
}
}
delete(f.permittedRoutes, r)
}
for _, r := range routesToAdd {
conditions := []*wf.Match{
{
Field: wf.FieldIPRemoteAddress,
Op: wf.MatchTypeEqual,
Value: r,
},
}
var p protocol
if r.Addr().Is4() {
p = protocolV4
} else {
p = protocolV6
}
name := "local route - " + r.String()
rules, err := f.addRules(name, weightKnownTraffic, conditions, wf.ActionPermit, p, directionBoth)
if err != nil {
return err
}
name = "link-local multicast - " + r.String()
conditions = matchLinkLocalMulticast(r, false)
multicastRules, err := f.addRules(name, weightKnownTraffic, conditions, wf.ActionPermit, p, directionOutbound)
if err != nil {
return err
}
rules = append(rules, multicastRules...)
conditions = matchLinkLocalMulticast(r, true)
multicastRules, err = f.addRules(name, weightKnownTraffic, conditions, wf.ActionPermit, p, directionInbound)
if err != nil {
return err
}
rules = append(rules, multicastRules...)
f.permittedRoutes[r] = rules
}
return nil
}
// matchLinkLocalMulticast returns a list of conditions that match
// outbound or inbound link-local multicast traffic to or from the
// specified network.
func matchLinkLocalMulticast(pfx netip.Prefix, inbound bool) []*wf.Match {
var linkLocalMulticastRange netip.Prefix
if pfx.Addr().Is4() {
linkLocalMulticastRange = linkLocalMulticastIPv4Range
} else {
linkLocalMulticastRange = linkLocalMulticastIPv6Range
}
var localAddr, remoteAddr netip.Prefix
if inbound {
localAddr, remoteAddr = linkLocalMulticastRange, pfx
} else {
localAddr, remoteAddr = pfx, linkLocalMulticastRange
}
return []*wf.Match{
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoUDP,
},
{
Field: wf.FieldIPLocalAddress,
Op: wf.MatchTypeEqual,
Value: localAddr,
},
{
Field: wf.FieldIPRemoteAddress,
Op: wf.MatchTypeEqual,
Value: remoteAddr,
},
}
}
func (f *Firewall) newRule(name string, w weight, layer wf.LayerID, conditions []*wf.Match, action wf.Action) (*wf.Rule, error) {
id, err := windows.GenerateGUID()
if err != nil {
return nil, err
}
return &wf.Rule{
Name: ruleName(action, layer, name),
ID: wf.RuleID(id),
Provider: f.providerID,
Sublayer: f.sublayerID,
Layer: layer,
Weight: uint64(w),
Conditions: conditions,
Action: action,
}, nil
}
func (f *Firewall) addRules(name string, w weight, conditions []*wf.Match, action wf.Action, p protocol, d direction) ([]*wf.Rule, error) {
var rules []*wf.Rule
for _, layer := range p.getLayers(d) {
r, err := f.newRule(name, w, layer, conditions, action)
if err != nil {
return nil, err
}
if err := f.session.AddRule(r); err != nil {
return nil, err
}
rules = append(rules, r)
}
return rules, nil
}
func (f *Firewall) blockAll(w weight) error {
_, err := f.addRules("all", w, nil, wf.ActionBlock, protocolAll, directionBoth)
return err
}
func (f *Firewall) permitNDP(w weight) error {
// These are aliased according to:
// https://social.msdn.microsoft.com/Forums/azure/en-US/eb2aa3cd-5f1c-4461-af86-61e7d43ccc23/filtering-icmp-by-type-code?forum=wfp
fieldICMPType := wf.FieldIPLocalPort
fieldICMPCode := wf.FieldIPRemotePort
var icmpConditions = func(t, c uint16, remoteAddress any) []*wf.Match {
conditions := []*wf.Match{
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoICMPV6,
},
{
Field: fieldICMPType,
Op: wf.MatchTypeEqual,
Value: t,
},
{
Field: fieldICMPCode,
Op: wf.MatchTypeEqual,
Value: c,
},
}
if remoteAddress != nil {
conditions = append(conditions, &wf.Match{
Field: wf.FieldIPRemoteAddress,
Op: wf.MatchTypeEqual,
Value: linkLocalRouterMulticast,
})
}
return conditions
}
/* TODO: actually handle the hop limit somehow! The rules should vaguely be:
* - icmpv6 133: must be outgoing, dst must be FF02::2/128, hop limit must be 255
* - icmpv6 134: must be incoming, src must be FE80::/10, hop limit must be 255
* - icmpv6 135: either incoming or outgoing, hop limit must be 255
* - icmpv6 136: either incoming or outgoing, hop limit must be 255
* - icmpv6 137: must be incoming, src must be FE80::/10, hop limit must be 255
*/
//
// Router Solicitation Message
// ICMP type 133, code 0. Outgoing.
//
conditions := icmpConditions(133, 0, linkLocalRouterMulticast)
if _, err := f.addRules("NDP type 133", w, conditions, wf.ActionPermit, protocolV6, directionOutbound); err != nil {
return err
}
//
// Router Advertisement Message
// ICMP type 134, code 0. Incoming.
//
conditions = icmpConditions(134, 0, linkLocalRange)
if _, err := f.addRules("NDP type 134", w, conditions, wf.ActionPermit, protocolV6, directionInbound); err != nil {
return err
}
//
// Neighbor Solicitation Message
// ICMP type 135, code 0. Bi-directional.
//
conditions = icmpConditions(135, 0, nil)
if _, err := f.addRules("NDP type 135", w, conditions, wf.ActionPermit, protocolV6, directionBoth); err != nil {
return err
}
//
// Neighbor Advertisement Message
// ICMP type 136, code 0. Bi-directional.
//
conditions = icmpConditions(136, 0, nil)
if _, err := f.addRules("NDP type 136", w, conditions, wf.ActionPermit, protocolV6, directionBoth); err != nil {
return err
}
//
// Redirect Message
// ICMP type 137, code 0. Incoming.
//
conditions = icmpConditions(137, 0, linkLocalRange)
if _, err := f.addRules("NDP type 137", w, conditions, wf.ActionPermit, protocolV6, directionInbound); err != nil {
return err
}
return nil
}
func (f *Firewall) permitDHCPv6(w weight) error {
var dhcpConditions = func(remoteAddrs ...any) []*wf.Match {
conditions := []*wf.Match{
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoUDP,
},
{
Field: wf.FieldIPLocalAddress,
Op: wf.MatchTypeEqual,
Value: linkLocalRange,
},
{
Field: wf.FieldIPLocalPort,
Op: wf.MatchTypeEqual,
Value: uint16(546),
},
{
Field: wf.FieldIPRemotePort,
Op: wf.MatchTypeEqual,
Value: uint16(547),
},
}
for _, a := range remoteAddrs {
conditions = append(conditions, &wf.Match{
Field: wf.FieldIPRemoteAddress,
Op: wf.MatchTypeEqual,
Value: a,
})
}
return conditions
}
conditions := dhcpConditions(linkLocalDHCPMulticast, siteLocalDHCPMulticast)
if _, err := f.addRules("DHCP request", w, conditions, wf.ActionPermit, protocolV6, directionOutbound); err != nil {
return err
}
conditions = dhcpConditions(linkLocalRange)
if _, err := f.addRules("DHCP response", w, conditions, wf.ActionPermit, protocolV6, directionInbound); err != nil {
return err
}
return nil
}
func (f *Firewall) permitDHCPv4(w weight) error {
var dhcpConditions = func(remoteAddrs ...any) []*wf.Match {
conditions := []*wf.Match{
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoUDP,
},
{
Field: wf.FieldIPLocalPort,
Op: wf.MatchTypeEqual,
Value: uint16(68),
},
{
Field: wf.FieldIPRemotePort,
Op: wf.MatchTypeEqual,
Value: uint16(67),
},
}
for _, a := range remoteAddrs {
conditions = append(conditions, &wf.Match{
Field: wf.FieldIPRemoteAddress,
Op: wf.MatchTypeEqual,
Value: a,
})
}
return conditions
}
conditions := dhcpConditions(netaddr.IPv4(255, 255, 255, 255))
if _, err := f.addRules("DHCP request", w, conditions, wf.ActionPermit, protocolV4, directionOutbound); err != nil {
return err
}
conditions = dhcpConditions()
if _, err := f.addRules("DHCP response", w, conditions, wf.ActionPermit, protocolV4, directionInbound); err != nil {
return err
}
return nil
}
func (f *Firewall) permitTunInterface(w weight) error {
condition := []*wf.Match{
{
Field: wf.FieldIPLocalInterface,
Op: wf.MatchTypeEqual,
Value: f.luid,
},
}
_, err := f.addRules("on TUN", w, condition, wf.ActionPermit, protocolAll, directionBoth)
return err
}
func (f *Firewall) permitLoopback(w weight) error {
condition := []*wf.Match{
{
Field: wf.FieldFlags,
Op: wf.MatchTypeFlagsAllSet,
Value: wf.ConditionFlagIsLoopback,
},
}
_, err := f.addRules("on loopback", w, condition, wf.ActionPermit, protocolAll, directionBoth)
return err
}
func (f *Firewall) permitDNS(w weight) error {
conditions := []*wf.Match{
{
Field: wf.FieldIPRemotePort,
Op: wf.MatchTypeEqual,
Value: uint16(53),
},
// Repeat the condition type for logical OR.
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoUDP,
},
{
Field: wf.FieldIPProtocol,
Op: wf.MatchTypeEqual,
Value: wf.IPProtoTCP,
},
}
_, err := f.addRules("DNS", w, conditions, wf.ActionPermit, protocolAll, directionBoth)
return err
}
func (f *Firewall) permitTailscaleService(w weight) error {
currentFile, err := os.Executable()
if err != nil {
return err
}
appID, err := wf.AppID(currentFile)
if err != nil {
return fmt.Errorf("could not get app id for %q: %w", currentFile, err)
}
conditions := []*wf.Match{
{
Field: wf.FieldALEAppID,
Op: wf.MatchTypeEqual,
Value: appID,
},
}
_, err = f.addRules("unrestricted traffic for Tailscale service", w, conditions, wf.ActionPermit, protocolAll, directionBoth)
return err
}