feat: implement a client for Thoth, the IP reputation database for Anubis (#637)

* feat(internal): add Thoth client and simple ASN checker

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thoth): cached ip to asn checker

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: go mod tidy

Signed-off-by: Xe Iaso <me@xeiaso.net>

* fix(thoth): minor testing fixups, ensure ASNChecker is Checker

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thoth): make ASNChecker instances

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thoth): add GeoIP checker

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thoth): store a thoth client in a context

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: refactor Checker type to its own package

Signed-off-by: Xe Iaso <me@xeiaso.net>

* test(thoth): add thoth mocking package, ignore context deadline exceeded errors

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thoth): pre-cache private ranges

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(lib/policy/config): enable thoth ASNs and GeoIP checker parsing

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore(thoth): refactor to move checker creation to the checker files

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(policy): enable thoth checks

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(thothmock): test helper function for loading a mock thoth instance

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat: wire up Thoth, make thoth checks part of the default config

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: spelling

Signed-off-by: Xe Iaso <me@xeiaso.net>

* fix(thoth): mend staticcheck errors

Signed-off-by: Xe Iaso <me@xeiaso.net>

* docs(admin): add Thoth docs

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore(policy): update Thoth links in error messages

Signed-off-by: Xe Iaso <me@xeiaso.net>

* docs: update CHANGELOG

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: spelling

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore(docs/manifest): enable Thoth

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: add THOTH_INSECURE for contacting Thoth over plain TCP in extreme circumstances

Signed-off-by: Xe Iaso <me@xeiaso.net>

* test(thoth): use mock thoth when credentials aren't detected in the environment

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore: spelling

Signed-off-by: Xe Iaso <me@xeiaso.net>

* fix(cmd/anubis): better warnings for half-configured Thoth setups

Signed-off-by: Xe Iaso <me@xeiaso.net>

* docs(botpolicies): link to Thoth geoip docs

Signed-off-by: Xe Iaso <me@xeiaso.net>

---------

Signed-off-by: Xe Iaso <me@xeiaso.net>
This commit is contained in:
Xe Iaso 2025-06-16 11:57:32 -04:00 committed by GitHub
parent 823d1be5d1
commit e3826df3ab
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
39 changed files with 1101 additions and 82 deletions

View file

@ -4,11 +4,12 @@ import (
"fmt"
"github.com/TecharoHQ/anubis/internal"
"github.com/TecharoHQ/anubis/lib/policy/checker"
"github.com/TecharoHQ/anubis/lib/policy/config"
)
type Bot struct {
Rules Checker
Rules checker.Impl
Challenge *config.ChallengeRules
Weight *config.Weight
Name string

View file

@ -9,6 +9,7 @@ import (
"strings"
"github.com/TecharoHQ/anubis/internal"
"github.com/TecharoHQ/anubis/lib/policy/checker"
"github.com/yl2chen/cidranger"
)
@ -16,37 +17,6 @@ var (
ErrMisconfiguration = errors.New("[unexpected] policy: administrator misconfiguration")
)
type Checker interface {
Check(*http.Request) (bool, error)
Hash() string
}
type CheckerList []Checker
func (cl CheckerList) Check(r *http.Request) (bool, error) {
for _, c := range cl {
ok, err := c.Check(r)
if err != nil {
return ok, err
}
if ok {
return ok, nil
}
}
return false, nil
}
func (cl CheckerList) Hash() string {
var sb strings.Builder
for _, c := range cl {
fmt.Fprintln(&sb, c.Hash())
}
return internal.SHA256sum(sb.String())
}
type staticHashChecker struct {
hash string
}
@ -57,7 +27,7 @@ func (staticHashChecker) Check(r *http.Request) (bool, error) {
func (s staticHashChecker) Hash() string { return s.hash }
func NewStaticHashChecker(hashable string) Checker {
func NewStaticHashChecker(hashable string) checker.Impl {
return staticHashChecker{hash: internal.SHA256sum(hashable)}
}
@ -66,7 +36,7 @@ type RemoteAddrChecker struct {
hash string
}
func NewRemoteAddrChecker(cidrs []string) (Checker, error) {
func NewRemoteAddrChecker(cidrs []string) (checker.Impl, error) {
ranger := cidranger.NewPCTrieRanger()
var sb strings.Builder
@ -122,11 +92,11 @@ type HeaderMatchesChecker struct {
hash string
}
func NewUserAgentChecker(rexStr string) (Checker, error) {
func NewUserAgentChecker(rexStr string) (checker.Impl, error) {
return NewHeaderMatchesChecker("User-Agent", rexStr)
}
func NewHeaderMatchesChecker(header, rexStr string) (Checker, error) {
func NewHeaderMatchesChecker(header, rexStr string) (checker.Impl, error) {
rex, err := regexp.Compile(strings.TrimSpace(rexStr))
if err != nil {
return nil, fmt.Errorf("%w: regex %s failed parse: %w", ErrMisconfiguration, rexStr, err)
@ -151,7 +121,7 @@ type PathChecker struct {
hash string
}
func NewPathChecker(rexStr string) (Checker, error) {
func NewPathChecker(rexStr string) (checker.Impl, error) {
rex, err := regexp.Compile(strings.TrimSpace(rexStr))
if err != nil {
return nil, fmt.Errorf("%w: regex %s failed parse: %w", ErrMisconfiguration, rexStr, err)
@ -171,7 +141,7 @@ func (pc *PathChecker) Hash() string {
return pc.hash
}
func NewHeaderExistsChecker(key string) Checker {
func NewHeaderExistsChecker(key string) checker.Impl {
return headerExistsChecker{strings.TrimSpace(key)}
}
@ -191,8 +161,8 @@ func (hec headerExistsChecker) Hash() string {
return internal.SHA256sum(hec.header)
}
func NewHeadersChecker(headermap map[string]string) (Checker, error) {
var result CheckerList
func NewHeadersChecker(headermap map[string]string) (checker.Impl, error) {
var result checker.List
var errs []error
for key, rexStr := range headermap {

View file

@ -0,0 +1,41 @@
// Package checker defines the Checker interface and a helper utility to avoid import cycles.
package checker
import (
"fmt"
"net/http"
"strings"
"github.com/TecharoHQ/anubis/internal"
)
type Impl interface {
Check(*http.Request) (bool, error)
Hash() string
}
type List []Impl
func (l List) Check(r *http.Request) (bool, error) {
for _, c := range l {
ok, err := c.Check(r)
if err != nil {
return ok, err
}
if ok {
return ok, nil
}
}
return false, nil
}
func (l List) Hash() string {
var sb strings.Builder
for _, c := range l {
fmt.Fprintln(&sb, c.Hash())
}
return internal.SHA256sum(sb.String())
}

44
lib/policy/config/asn.go Normal file
View file

@ -0,0 +1,44 @@
package config
import (
"errors"
"fmt"
)
var (
ErrPrivateASN = errors.New("bot.ASNs: you have specified a private use ASN")
)
type ASNs struct {
Match []uint32 `json:"match"`
}
func (a *ASNs) Valid() error {
var errs []error
for _, asn := range a.Match {
if isPrivateASN(asn) {
errs = append(errs, fmt.Errorf("%w: %d is private (see RFC 6996)", ErrPrivateASN, asn))
}
}
if len(errs) != 0 {
return fmt.Errorf("bot.ASNs: invalid ASN settings: %w", errors.Join(errs...))
}
return nil
}
// isPrivateASN checks if an ASN is in the private use area.
//
// Based on RFC 6996 and IANA allocations.
func isPrivateASN(asn uint32) bool {
switch {
case asn >= 64512 && asn <= 65534:
return true
case asn >= 4200000000 && asn <= 4294967294:
return true
default:
return false
}
}

View file

@ -55,6 +55,10 @@ type BotConfig struct {
Name string `json:"name" yaml:"name"`
Action Rule `json:"action" yaml:"action"`
RemoteAddr []string `json:"remote_addresses,omitempty" yaml:"remote_addresses,omitempty"`
// Thoth features
GeoIP *GeoIP `json:"geoip,omitempty"`
ASNs *ASNs `json:"asns,omitempty"`
}
func (b BotConfig) Zero() bool {
@ -66,6 +70,8 @@ func (b BotConfig) Zero() bool {
b.Action != "",
len(b.RemoteAddr) != 0,
b.Challenge != nil,
b.GeoIP != nil,
b.ASNs != nil,
} {
if cond {
return false
@ -85,7 +91,9 @@ func (b *BotConfig) Valid() error {
allFieldsEmpty := b.UserAgentRegex == nil &&
b.PathRegex == nil &&
len(b.RemoteAddr) == 0 &&
len(b.HeadersRegex) == 0
len(b.HeadersRegex) == 0 &&
b.ASNs == nil &&
b.GeoIP == nil
if allFieldsEmpty && b.Expression == nil {
errs = append(errs, ErrBotMustHaveUserAgentOrPath)

View file

@ -0,0 +1,36 @@
package config
import (
"errors"
"fmt"
"regexp"
"strings"
)
var (
countryCodeRegexp = regexp.MustCompile(`^\w{2}$`)
ErrNotCountryCode = errors.New("config.Bot: invalid country code")
)
type GeoIP struct {
Countries []string `json:"countries"`
}
func (g *GeoIP) Valid() error {
var errs []error
for i, cc := range g.Countries {
if !countryCodeRegexp.MatchString(cc) {
errs = append(errs, fmt.Errorf("%w: %s", ErrNotCountryCode, cc))
}
g.Countries[i] = strings.ToLower(cc)
}
if len(errs) != 0 {
return fmt.Errorf("bot.GeoIP: invalid GeoIP settings: %w", errors.Join(errs...))
}
return nil
}

View file

@ -0,0 +1,6 @@
bots:
- name: challenge-cloudflare
action: CHALLENGE
asns:
match:
- 13335 # Cloudflare

View file

@ -0,0 +1,6 @@
bots:
- name: compute-tarrif-us
action: CHALLENGE
geoip:
countries:
- US

View file

@ -1,10 +1,14 @@
package policy
import (
"context"
"errors"
"fmt"
"io"
"log/slog"
"github.com/TecharoHQ/anubis/internal/thoth"
"github.com/TecharoHQ/anubis/lib/policy/checker"
"github.com/TecharoHQ/anubis/lib/policy/config"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"
@ -35,7 +39,7 @@ func NewParsedConfig(orig *config.Config) *ParsedConfig {
}
}
func ParseConfig(fin io.Reader, fname string, defaultDifficulty int) (*ParsedConfig, error) {
func ParseConfig(ctx context.Context, fin io.Reader, fname string, defaultDifficulty int) (*ParsedConfig, error) {
c, err := config.Load(fin, fname)
if err != nil {
return nil, err
@ -43,6 +47,8 @@ func ParseConfig(fin io.Reader, fname string, defaultDifficulty int) (*ParsedCon
var validationErrs []error
tc, hasThothClient := thoth.FromContext(ctx)
result := NewParsedConfig(c)
result.DefaultDifficulty = defaultDifficulty
@ -57,7 +63,7 @@ func ParseConfig(fin io.Reader, fname string, defaultDifficulty int) (*ParsedCon
Action: b.Action,
}
cl := CheckerList{}
cl := checker.List{}
if len(b.RemoteAddr) > 0 {
c, err := NewRemoteAddrChecker(b.RemoteAddr)
@ -104,6 +110,24 @@ func ParseConfig(fin io.Reader, fname string, defaultDifficulty int) (*ParsedCon
}
}
if b.ASNs != nil {
if !hasThothClient {
slog.Warn("You have specified a Thoth specific check but you have no Thoth client configured. Please read https://anubis.techaro.lol/docs/admin/thoth for more information", "check", "asn", "settings", b.ASNs)
continue
}
cl = append(cl, tc.ASNCheckerFor(b.ASNs.Match))
}
if b.GeoIP != nil {
if !hasThothClient {
slog.Warn("You have specified a Thoth specific check but you have no Thoth client configured. Please read https://anubis.techaro.lol/docs/admin/thoth for more information", "check", "geoip", "settings", b.GeoIP)
continue
}
cl = append(cl, tc.GeoIPCheckerFor(b.GeoIP.Countries))
}
if b.Challenge == nil {
parsedBot.Challenge = &config.ChallengeRules{
Difficulty: defaultDifficulty,

View file

@ -7,21 +7,25 @@ import (
"github.com/TecharoHQ/anubis"
"github.com/TecharoHQ/anubis/data"
"github.com/TecharoHQ/anubis/internal/thoth/thothmock"
)
func TestDefaultPolicyMustParse(t *testing.T) {
ctx := thothmock.WithMockThoth(t)
fin, err := data.BotPolicies.Open("botPolicies.json")
if err != nil {
t.Fatal(err)
}
defer fin.Close()
if _, err := ParseConfig(fin, "botPolicies.json", anubis.DefaultDifficulty); err != nil {
if _, err := ParseConfig(ctx, fin, "botPolicies.json", anubis.DefaultDifficulty); err != nil {
t.Fatalf("can't parse config: %v", err)
}
}
func TestGoodConfigs(t *testing.T) {
finfos, err := os.ReadDir("config/testdata/good")
if err != nil {
t.Fatal(err)
@ -30,20 +34,37 @@ func TestGoodConfigs(t *testing.T) {
for _, st := range finfos {
st := st
t.Run(st.Name(), func(t *testing.T) {
fin, err := os.Open(filepath.Join("config", "testdata", "good", st.Name()))
if err != nil {
t.Fatal(err)
}
defer fin.Close()
t.Run("with-thoth", func(t *testing.T) {
fin, err := os.Open(filepath.Join("config", "testdata", "good", st.Name()))
if err != nil {
t.Fatal(err)
}
defer fin.Close()
if _, err := ParseConfig(fin, fin.Name(), anubis.DefaultDifficulty); err != nil {
t.Fatal(err)
}
ctx := thothmock.WithMockThoth(t)
if _, err := ParseConfig(ctx, fin, fin.Name(), anubis.DefaultDifficulty); err != nil {
t.Fatal(err)
}
})
t.Run("without-thoth", func(t *testing.T) {
fin, err := os.Open(filepath.Join("config", "testdata", "good", st.Name()))
if err != nil {
t.Fatal(err)
}
defer fin.Close()
if _, err := ParseConfig(t.Context(), fin, fin.Name(), anubis.DefaultDifficulty); err != nil {
t.Fatal(err)
}
})
})
}
}
func TestBadConfigs(t *testing.T) {
ctx := thothmock.WithMockThoth(t)
finfos, err := os.ReadDir("config/testdata/bad")
if err != nil {
t.Fatal(err)
@ -58,7 +79,7 @@ func TestBadConfigs(t *testing.T) {
}
defer fin.Close()
if _, err := ParseConfig(fin, fin.Name(), anubis.DefaultDifficulty); err == nil {
if _, err := ParseConfig(ctx, fin, fin.Name(), anubis.DefaultDifficulty); err == nil {
t.Fatal(err)
} else {
t.Log(err)