Files
traefik-llmgateway/auth_test.go
T
lukaszraczylo d2a0c4860f fix(auth): reject duplicate and empty user names
Fixes finding 5 from the 2026-08-22 security audit: newAuthStore and
replaceFileUsers rejected duplicate API keys but not duplicate names, and
counters are keyed on u.name (routes_unified.go), so two users sharing a
name silently shared llmgw:user:{name}:* counters and both appeared in
buildAdminUsage; an empty name collapsed every such user into
llmgw:user::*.

- buildEntry now rejects an empty user name outright — the source of
  both bugs, since duplicate names and an empty name are the same root
  cause (two config entries resolving to the same counter key).
- newAuthStore and replaceFileUsers each reject a duplicate name WITHIN
  their own source (two inline users, or two file users) — but not
  across sources: a file user overriding an inline user of the same name
  stays intentional and unaffected, per replaceFileUsers' own documented
  merge semantics.
- Deliberately does NOT extend configNamePattern's character
  restriction to user names: that pattern exists because provider/
  mcpServers/agents names are embedded as URL path segments, which a
  user name never is here. Restricting the charset further risked
  rejecting real operator-chosen names this round has no visibility
  into; the live users file's steve-cw/kevinsandom already pass the
  non-empty check unaffected.

Also adds TestDefaultPreserving_LiveClusterShape_LoadsAndServesIdentically
(admin_test.go), modeling the live cluster's 5 friends + 4 home users and
11 MCP servers to prove all five fixes in this round load and serve
identically at that scale.
2026-08-22 15:26:32 +01:00

818 lines
30 KiB
Go

package traefikllmgateway
import (
"fmt"
"net/http/httptest"
"strings"
"sync"
"testing"
"time"
)
// TestIdentify is the brief's Step-1 table: bearer auth resolves the right
// user and group, and the group's model glob matches both the bare id and
// the provider-prefixed id.
func TestIdentify(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {Models: []string{"gpt-5*"}}},
Users: &UsersConfig{Inline: []*UserConfig{{Name: "a", Group: "eng", APIKey: "sk-secret"}}},
}
a, err := newAuthStore(cfg)
if err != nil {
t.Fatal(err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
u, grp, ok := a.identify(r)
if !ok || u.name != "a" || grp.name != "eng" {
t.Fatalf("identify failed: %v %v %v", u, grp, ok)
}
if !grp.allowsModel("gpt-5-mini") {
t.Error("glob should match")
}
// allowsModel is an exact glob against the given string only — it no
// longer strips a "provider/" prefix itself (fix(registry) ruling 3): a
// pattern like "gpt-5*" reaching a provider-prefixed request like
// "openai/gpt-5-mini" is now the caller's job (modelRegistry.
// resolveAgainst / listFor build the bare-suffix candidate themselves),
// covered at the registry level in registry_test.go, not here.
if grp.allowsModel("openai/gpt-5-mini") {
t.Error("allowsModel must not match a provider-prefixed id on its own; prefix-stripping now lives in the caller")
}
if grp.allowsModel("claude-4") {
t.Error("should not match")
}
}
func testAuthCfg() *Config {
return &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{{Name: "a", Group: "eng", APIKey: "sk-secret"}}},
}
}
func TestIdentify_XAPIKeyHeader_OK(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("x-api-key", "sk-secret")
u, grp, ok := a.identify(r)
if !ok || u.name != "a" || grp.name != "eng" {
t.Fatalf("identify via x-api-key failed: %v %v %v", u, grp, ok)
}
}
func TestIdentify_BearerCaseInsensitivePrefix(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
for _, prefix := range []string{"Bearer", "bearer", "BEARER", "BeArEr"} {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", prefix+" sk-secret")
if _, _, ok := a.identify(r); !ok {
t.Errorf("prefix %q: want identify to succeed", prefix)
}
}
}
func TestIdentify_BearerWinsOverXAPIKey(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}, "ops": {}},
Users: &UsersConfig{Inline: []*UserConfig{
{Name: "bearer-user", Group: "eng", APIKey: "sk-bearer"},
{Name: "header-user", Group: "ops", APIKey: "sk-header"},
}},
}
a, err := newAuthStore(cfg)
if err != nil {
t.Fatal(err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-bearer")
r.Header.Set("x-api-key", "sk-header")
u, _, ok := a.identify(r)
if !ok || u.name != "bearer-user" {
t.Fatalf("want bearer to win, got %v ok=%v", u, ok)
}
}
func TestIdentify_WrongKeyFails(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-wrong")
if _, _, ok := a.identify(r); ok {
t.Fatal("want identify to fail for wrong key")
}
}
func TestIdentify_MissingHeaderFails(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
if _, _, ok := a.identify(r); ok {
t.Fatal("want identify to fail with no auth header")
}
}
func TestNewAuthStore_UnknownGroup_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{{Name: "a", Group: "nonexistent", APIKey: "sk-secret"}}},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for user referencing unknown group")
}
}
func TestNewAuthStore_DuplicateAPIKey_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{
{Name: "a", Group: "eng", APIKey: "sk-same"},
{Name: "b", Group: "eng", APIKey: "sk-same"},
}},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for duplicate API key across users")
}
}
func TestNewAuthStore_EmptyAPIKeyAfterResolution_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{{Name: "a", Group: "eng", APIKey: ""}}},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for user with empty API key after resolution")
}
}
// TestNewAuthStore_NilGroupConfig_ReturnsError covers a Groups map entry
// whose value is a nil *GroupConfig (a caller can construct this in Go, and
// Traefik's own YAML-to-struct decoding can leave a map value nil for an
// empty mapping entry) — newAuthStore must reject it as a constructor error
// rather than panic dereferencing gc.Limits on a nil gc.
func TestNewAuthStore_NilGroupConfig_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": nil},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for a nil GroupConfig value, got nil")
}
}
func TestNewAuthStore_NoUsers_OK(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
}
a, err := newAuthStore(cfg)
if err != nil {
t.Fatalf("newAuthStore: %v", err)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("x-api-key", "anything")
if _, _, ok := a.identify(r); ok {
t.Fatal("want identify to fail when no users are configured")
}
}
// --- GroupConfig.CacheTTL: constructor validation (item B) ---
// TestNewAuthStore_GroupCacheTTL_ConstructorErrors is the constructor-error
// table for GroupConfig.CacheTTL: a malformed duration, a zero duration, a
// negative duration, and CacheTTL set while the global cache block is not
// configured (nothing to inherit ttl from, the same reasoning
// newAuthStore already applies to Cache:true) must all fail construction.
func TestNewAuthStore_GroupCacheTTL_ConstructorErrors(t *testing.T) {
tests := []struct {
name string
cacheTTL string
globalCache bool
}{
{name: "malformed duration", cacheTTL: "not-a-duration", globalCache: true},
{name: "zero duration", cacheTTL: "0s", globalCache: true},
{name: "negative duration", cacheTTL: "-5s", globalCache: true},
{name: "set without global cache enabled", cacheTTL: "5m", globalCache: false},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"default": {CacheTTL: tt.cacheTTL}},
Cache: CacheConfig{Enabled: tt.globalCache},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatalf("newAuthStore: want a constructor error for cacheTTL=%q (globalCache=%v)", tt.cacheTTL, tt.globalCache)
}
})
}
}
// TestNewAuthStore_GroupCacheTTL_EmptyInheritsZero proves an omitted
// CacheTTL resolves to group.cacheTTL == 0 — effectiveTTL's (cache.go)
// signal to inherit the global cache TTL rather than override it.
func TestNewAuthStore_GroupCacheTTL_EmptyInheritsZero(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"default": {}},
}
a, err := newAuthStore(cfg)
if err != nil {
t.Fatalf("newAuthStore: %v", err)
}
if got := a.groups["default"].cacheTTL; got != 0 {
t.Errorf("groups[default].cacheTTL = %v, want 0 (inherit)", got)
}
}
// TestNewAuthStore_GroupCacheTTL_ValidOverride_ResolvedOnGroup proves a
// well-formed, positive CacheTTL with the global cache block configured
// both constructs cleanly and resolves onto the group struct exactly,
// ready for effectiveTTL to read.
func TestNewAuthStore_GroupCacheTTL_ValidOverride_ResolvedOnGroup(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"default": {CacheTTL: "30s"}},
Cache: CacheConfig{Enabled: true},
}
a, err := newAuthStore(cfg)
if err != nil {
t.Fatalf("newAuthStore: %v", err)
}
if got := a.groups["default"].cacheTTL; got != 30*time.Second {
t.Errorf("groups[default].cacheTTL = %v, want 30s", got)
}
}
func TestGroup_AllowsProvider_EmptyListAllowsAll(t *testing.T) {
grp := &group{name: "eng"}
if !grp.allowsProvider("openai") {
t.Fatal("empty providers list should allow all")
}
}
func TestGroup_AllowsProvider_Glob(t *testing.T) {
grp := &group{name: "eng", providers: []string{"open*"}}
if !grp.allowsProvider("openai") {
t.Error("want glob match")
}
if grp.allowsProvider("anthropic") {
t.Error("want no match")
}
}
func TestGroup_AllowsMCP_Glob(t *testing.T) {
grp := &group{name: "eng", mcpServers: []string{"search-*"}}
if !grp.allowsMCP("search-web") {
t.Error("want glob match")
}
if grp.allowsMCP("db-admin") {
t.Error("want no match")
}
}
func TestGroup_AllowsAgent_Glob(t *testing.T) {
grp := &group{name: "eng", agents: []string{"triage"}}
if !grp.allowsAgent("triage") {
t.Error("want exact match")
}
if grp.allowsAgent("other") {
t.Error("want no match")
}
}
func TestGroup_AllowsModel_EmptyListAllowsAnything(t *testing.T) {
grp := &group{name: "eng"}
if !grp.allowsModel("anything-goes") {
t.Fatal("empty models list should allow anything")
}
}
// TestGroup_AllowsPassthroughPath_EmptyListAllowsAll pins the zero-config
// default for GroupConfig.PassthroughPaths (security+performance audit,
// 2026-08-22): a group with no PassthroughPaths configured — every group
// that existed before this field did — must keep reaching every native
// passthrough path, exactly like allowsProvider/allowsModel/allowsMCP/
// allowsAgent's own empty-means-all defaults.
func TestGroup_AllowsPassthroughPath_EmptyListAllowsAll(t *testing.T) {
grp := &group{name: "eng"}
if !grp.allowsPassthroughPath("v1/files") {
t.Fatal("empty passthroughPaths list should allow all")
}
}
func TestGroup_AllowsPassthroughPath_Glob(t *testing.T) {
grp := &group{name: "eng", passthroughPaths: []string{"v1/chat/completions"}}
if !grp.allowsPassthroughPath("v1/chat/completions") {
t.Error("want exact match")
}
if grp.allowsPassthroughPath("v1/files") {
t.Error("want no match for a path outside the allowlist")
}
}
// TestGroup_AllowsPassthroughPath_RejectsEncodedSlashBypass is the
// security review finding 3 regression test (round 3, 2026-08-22): a
// glob of ["v1/*"] (one wildcard segment) must not be defeated by
// encoding the "/" that would otherwise split the candidate into a
// second segment. Before this fix, matchesGlob ran against the escaped
// form directly, so path.Match("v1/*", "v1/fine_tuning%2Fjobs") reported
// true — the escaped rest looks like exactly two segments to path.Match,
// same as the decoded intent, but MEASURED to actually still match
// because "%2Fjobs" contains no literal "/" for the glob to split on
// differently — while an upstream that itself decodes %2F would read the
// same rest as "v1/fine_tuning/jobs", a THREE-segment path "v1/*" was
// never meant to authorize.
func TestGroup_AllowsPassthroughPath_RejectsEncodedSlashBypass(t *testing.T) {
grp := &group{name: "eng", passthroughPaths: []string{"v1/*"}}
if grp.allowsPassthroughPath("v1/fine_tuning%2Fjobs") {
t.Error("want no match: the decoded form is three segments (v1/fine_tuning/jobs), which \"v1/*\" does not authorize")
}
// Sanity: the decoded-equivalent form is independently denied too,
// proving the glob genuinely does not authorize the deeper path — not
// just that encoding happens to break matching by accident.
if grp.allowsPassthroughPath("v1/fine_tuning/jobs") {
t.Error("want no match for the literal three-segment form either")
}
// A genuinely single-segment encoded value must still match: "v1/*"
// authorizes any ONE segment after "v1", encoded or not.
if !grp.allowsPassthroughPath("v1/chat%2Dcompletions") {
t.Error("want match: a single encoded segment (no literal slash once decoded) is still one segment")
}
}
// TestGroup_AllowsPassthroughPath_MalformedEncoding_DeniesWhenRestricted
// mirrors hasTraversalSegment's own "fails to unescape at all is
// rejected too" rule (routes_passthrough.go) — applied here to the
// allowlist match itself, for a group that actually restricts paths.
func TestGroup_AllowsPassthroughPath_MalformedEncoding_DeniesWhenRestricted(t *testing.T) {
grp := &group{name: "eng", passthroughPaths: []string{"v1/*"}}
if grp.allowsPassthroughPath("v1/%zz") {
t.Error("want no match: malformed percent-encoding must fail closed for a restricted group")
}
}
// TestGroup_AllowsPassthroughPath_MalformedEncoding_StillAllowsWhenUnrestricted
// pins the zero-config default-preserving requirement precisely at the
// boundary this fix touches: an EMPTY PassthroughPaths list (the fast
// path, checked before any decoding) must keep allowing a request whose
// rest fails to percent-decode — exactly as before this finding, since
// hasTraversalSegment (unconditional, in handlePassthrough) is still the
// gate for that case, unchanged by this fix.
func TestGroup_AllowsPassthroughPath_MalformedEncoding_StillAllowsWhenUnrestricted(t *testing.T) {
grp := &group{name: "eng"} // no PassthroughPaths configured
if !grp.allowsPassthroughPath("v1/%zz") {
t.Error("want match: an unrestricted group's fast path must never decode rest at all")
}
}
func TestAuthStore_ReplaceFileUsers_AddsAndSwapsUsers(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret in group eng
if err != nil {
t.Fatal(err)
}
if err := a.replaceFileUsers([]*UserConfig{{Name: "f1", Group: "eng", APIKey: "sk-file1"}}); err != nil {
t.Fatalf("replaceFileUsers: %v", err)
}
inlineReq := httptest.NewRequest("POST", "/v1/chat/completions", nil)
inlineReq.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(inlineReq); !ok {
t.Fatal("want inline user to remain identifiable after replaceFileUsers")
}
fileReq := httptest.NewRequest("POST", "/v1/chat/completions", nil)
fileReq.Header.Set("Authorization", "Bearer sk-file1")
u, _, ok := a.identify(fileReq)
if !ok || u.name != "f1" {
t.Fatalf("want file user f1 identifiable, got %v ok=%v", u, ok)
}
// Second call swaps the file-sourced set wholesale: f1 must disappear,
// f2 must appear, inline user must still work.
if err := a.replaceFileUsers([]*UserConfig{{Name: "f2", Group: "eng", APIKey: "sk-file2"}}); err != nil {
t.Fatalf("replaceFileUsers (second): %v", err)
}
if _, _, ok := a.identify(fileReq); ok {
t.Fatal("want f1 to no longer be identifiable after being replaced")
}
if _, _, ok := a.identify(inlineReq); !ok {
t.Fatal("want inline user to remain identifiable after second replaceFileUsers")
}
f2Req := httptest.NewRequest("POST", "/v1/chat/completions", nil)
f2Req.Header.Set("Authorization", "Bearer sk-file2")
if _, _, ok := a.identify(f2Req); !ok {
t.Fatal("want file user f2 identifiable after second replaceFileUsers")
}
}
func TestAuthStore_ReplaceFileUsers_DuplicateWithInline_ReturnsErrorAndLeavesStoreUnchanged(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret
if err != nil {
t.Fatal(err)
}
err = a.replaceFileUsers([]*UserConfig{{Name: "dup", Group: "eng", APIKey: "sk-secret"}})
if err == nil {
t.Fatal("want error for file user colliding with inline user's API key")
}
// Store must still be usable and unchanged by the rejected replace.
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
u, _, ok := a.identify(r)
if !ok || u.name != "a" {
t.Fatalf("want original inline user still resolvable after rejected replace, got %v ok=%v", u, ok)
}
}
func TestAuthStore_ReplaceFileUsers_UnknownGroup_ReturnsError(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
err = a.replaceFileUsers([]*UserConfig{{Name: "f1", Group: "nonexistent", APIKey: "sk-file1"}})
if err == nil {
t.Fatal("want error for file user referencing unknown group")
}
}
// TestAuthStore_ReplaceFileUsers_NilElement_ReturnsErrorNotPanic is the
// regression for a nil *UserConfig slice element reaching buildEntry. This
// path is reachable at runtime: json.Unmarshal of a `null` array element
// into []*UserConfig yields a nil entry, and the config-reload path (Task 4)
// that calls replaceFileUsers is NOT wrapped by recoverPanic — that guard
// only covers ServeHTTP. An unguarded nil dereference here would take down
// the whole process instead of failing one reload.
func TestAuthStore_ReplaceFileUsers_NilElement_ReturnsErrorNotPanic(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
if err := a.replaceFileUsers([]*UserConfig{nil}); err == nil {
t.Fatal("want error for a nil user config element, got nil")
}
// The store must still be usable — the rejected replace must not have
// partially mutated it.
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(r); !ok {
t.Fatal("want inline user still resolvable after a rejected nil-element replace")
}
}
// TestNewAuthStore_NilInlineElement_ReturnsErrorNotPanic covers the same
// nil-element hazard reachable from newAuthStore's inline-user loop, which
// runs at plugin construction time — also outside recoverPanic's coverage.
func TestNewAuthStore_NilInlineElement_ReturnsErrorNotPanic(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{nil}},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for a nil inline user config element, got nil")
}
}
// TestAuthStore_ReplaceFileUsers_ConcurrentWritersAndReaders exercises
// replaceFileUsers and identify from many goroutines simultaneously under
// -race. Each writer swaps in a single, uniquely-keyed file user, so
// whichever writer's rebuild is applied last, the store must always hold
// exactly the inline users plus one file user's worth of entries — never a
// partial mix from two overlapping rebuilds, and never a state where the
// always-present inline user briefly disappears.
func TestAuthStore_ReplaceFileUsers_ConcurrentWritersAndReaders(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret, group eng
if err != nil {
t.Fatal(err)
}
const writers = 20
const readers = 10
const readsPerReader = 50
var wg sync.WaitGroup
wg.Add(writers + readers)
for i := 0; i < writers; i++ {
go func(i int) {
defer wg.Done()
uc := &UserConfig{Name: fmt.Sprintf("f%d", i), Group: "eng", APIKey: fmt.Sprintf("sk-file-%d", i)}
if err := a.replaceFileUsers([]*UserConfig{uc}); err != nil {
t.Errorf("replaceFileUsers(%d): %v", i, err)
}
}(i)
}
for i := 0; i < readers; i++ {
go func() {
defer wg.Done()
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
for j := 0; j < readsPerReader; j++ {
if _, _, ok := a.identify(r); !ok {
t.Error("inline user must remain identifiable during concurrent file-user swaps")
return
}
}
}()
}
wg.Wait()
a.mu.RLock()
got := len(a.byDigest)
a.mu.RUnlock()
want := len(a.inline) + 1 // inline users plus exactly one winning file-user batch
if got != want {
t.Fatalf("want exactly one file-user batch to survive concurrent swaps (size %d), got size %d — a rebuild was lost or merged with another", want, got)
}
}
// TestAuthStore_Identify_NotBlockedByBuildMu is the regression for the
// Task-4 lock split: replaceFileUsers's writer-serialization lock (buildMu)
// must be independent from the reader lock (mu) that identify uses. The
// prior implementation held mu for the whole rebuild, which would stall
// every identify call for as long as a reload's build phase (including
// resolveSecret's potentially slow file-backed reads) took. Holding buildMu
// externally here simulates "a rebuild's build phase is in progress" — a
// correct identify must not need buildMu at all.
func TestAuthStore_Identify_NotBlockedByBuildMu(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
a.buildMu.Lock()
defer a.buildMu.Unlock()
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
done := make(chan bool, 1)
go func() {
_, _, ok := a.identify(r)
done <- ok
}()
select {
case ok := <-done:
if !ok {
t.Fatal("want identify to succeed")
}
case <-time.After(2 * time.Second):
t.Fatal("identify blocked while buildMu is held — reader lock must not depend on the writer-serialization lock")
}
}
// --- security audit finding 3: per-source-IP auth-failure throttling ---
func TestClientIP_StripsPort(t *testing.T) {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = "203.0.113.9:54321"
if got := clientIP(r); got != "203.0.113.9" {
t.Errorf("clientIP = %q, want %q", got, "203.0.113.9")
}
}
func TestClientIP_MalformedRemoteAddr_UsedAsIs(t *testing.T) {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = "not-a-host-port"
if got := clientIP(r); got != "not-a-host-port" {
t.Errorf("clientIP = %q, want the raw RemoteAddr back unchanged", got)
}
}
// TestAuthStore_Identify_ThrottlesAfterRepeatedFailures_EvenValidKeyRejected
// is the core regression for finding 3: identify previously had no
// counter, no backoff, and no lockout at all. After authFailureLimit
// failed attempts from one source IP, even a subsequently CORRECT key
// from that same IP must be rejected — proving the throttle actually
// short-circuits real auth once tripped, not merely that repeated wrong
// keys keep failing (which they would anyway). A different source IP must
// be entirely unaffected, proving the throttle is per-IP, not global.
func TestAuthStore_Identify_ThrottlesAfterRepeatedFailures_EvenValidKeyRejected(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret
if err != nil {
t.Fatal(err)
}
const attackerIP = "203.0.113.9:54321"
for i := 0; i < authFailureLimit; i++ {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = attackerIP
r.Header.Set("x-api-key", "wrong-key")
if _, _, ok := a.identify(r); ok {
t.Fatalf("attempt %d: want failure for a wrong key, got success", i)
}
}
// Throttle should now be engaged for attackerIP — even the CORRECT
// key must be rejected.
valid := httptest.NewRequest("POST", "/v1/chat/completions", nil)
valid.RemoteAddr = attackerIP
valid.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(valid); ok {
t.Fatal("want a throttled IP rejected even with a valid key")
}
// A different source IP must be unaffected by attackerIP's throttle.
other := httptest.NewRequest("POST", "/v1/chat/completions", nil)
other.RemoteAddr = "198.51.100.5:1111"
other.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(other); !ok {
t.Fatal("want a different source IP unaffected by another IP's throttle")
}
}
// TestAuthStore_Identify_ThrottleResetsAfterWindowExpires proves the
// throttle is a fixed-window rate limit, not a permanent lockout: once
// authFailureWindow has actually elapsed, the same IP's valid key works
// again.
func TestAuthStore_Identify_ThrottleResetsAfterWindowExpires(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret
if err != nil {
t.Fatal(err)
}
clock := &fakeClock{now: time.Unix(1_700_000_000, 0)}
a.nowFn = clock.Now
const attackerIP = "203.0.113.9:54321"
for i := 0; i < authFailureLimit; i++ {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = attackerIP
r.Header.Set("x-api-key", "wrong-key")
if _, _, ok := a.identify(r); ok {
t.Fatalf("attempt %d: want failure for a wrong key, got success", i)
}
}
blocked := httptest.NewRequest("POST", "/v1/chat/completions", nil)
blocked.RemoteAddr = attackerIP
blocked.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(blocked); ok {
t.Fatal("want throttled before the window elapses")
}
clock.Advance(authFailureWindow + time.Second)
recovered := httptest.NewRequest("POST", "/v1/chat/completions", nil)
recovered.RemoteAddr = attackerIP
recovered.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(recovered); !ok {
t.Fatal("want the same IP's valid key to work again once authFailureWindow has elapsed")
}
}
// TestAuthStore_Identify_FewFailedAttempts_NeverThrottled is the good-path
// counterpart: a handful of failed attempts (well under authFailureLimit
// — a human retrying a mistyped key a couple of times) must never trip
// the throttle.
func TestAuthStore_Identify_FewFailedAttempts_NeverThrottled(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
const ip = "203.0.113.9:54321"
for i := 0; i < 3; i++ {
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = ip
r.Header.Set("x-api-key", "wrong-key")
a.identify(r)
}
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = ip
r.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(r); !ok {
t.Fatal("want a valid key to succeed after only 3 prior failures (well under authFailureLimit)")
}
}
// TestAuthStore_Identify_SuccessfulAuthDoesNotRecordFailure proves a
// successful identify never itself counts toward the failure budget —
// otherwise a legitimate, high-traffic user could eventually throttle
// themselves.
func TestAuthStore_Identify_SuccessfulAuthDoesNotRecordFailure(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
const ip = "203.0.113.9:54321"
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.RemoteAddr = ip
r.Header.Set("Authorization", "Bearer sk-secret")
for i := 0; i < authFailureLimit+5; i++ {
if _, _, ok := a.identify(r); !ok {
t.Fatalf("attempt %d: want success (valid key), got failure", i)
}
}
}
// --- security audit finding 5: duplicate/empty user names ---
func TestNewAuthStore_EmptyUserName_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{{Name: "", Group: "eng", APIKey: "sk-secret"}}},
}
if _, err := newAuthStore(cfg); err == nil {
t.Fatal("want error for an empty user name")
}
}
func TestNewAuthStore_DuplicateUserName_ReturnsError(t *testing.T) {
cfg := &Config{
Providers: map[string]*ProviderConfig{"openai": {Type: "openai", APIKey: "k"}},
Groups: map[string]*GroupConfig{"eng": {}},
Users: &UsersConfig{Inline: []*UserConfig{
{Name: "dup", Group: "eng", APIKey: "sk-one"},
{Name: "dup", Group: "eng", APIKey: "sk-two"},
}},
}
_, err := newAuthStore(cfg)
if err == nil {
t.Fatal("want error for two inline users sharing the same name")
}
if !strings.Contains(err.Error(), "duplicate user name") {
t.Errorf("error = %q, want it to mention the duplicate-name check specifically (not the API-key one)", err.Error())
}
}
func TestAuthStore_ReplaceFileUsers_EmptyUserName_ReturnsError(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
if err := a.replaceFileUsers([]*UserConfig{{Name: "", Group: "eng", APIKey: "sk-file1"}}); err == nil {
t.Fatal("want error for a file user with an empty name")
}
}
func TestAuthStore_ReplaceFileUsers_DuplicateUserNameWithinFile_ReturnsError(t *testing.T) {
a, err := newAuthStore(testAuthCfg())
if err != nil {
t.Fatal(err)
}
err = a.replaceFileUsers([]*UserConfig{
{Name: "dup", Group: "eng", APIKey: "sk-file1"},
{Name: "dup", Group: "eng", APIKey: "sk-file2"},
})
if err == nil {
t.Fatal("want error for two file users sharing the same name")
}
if !strings.Contains(err.Error(), "duplicate user name") {
t.Errorf("error = %q, want it to mention the duplicate-name check specifically (not the API-key one)", err.Error())
}
// The store must still be usable — the rejected replace must not have
// partially mutated it.
r := httptest.NewRequest("POST", "/v1/chat/completions", nil)
r.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(r); !ok {
t.Fatal("want the original inline user still resolvable after a rejected duplicate-name replace")
}
}
// TestAuthStore_ReplaceFileUsers_SameNameAsInline_DifferentKey_OverrideAllowed
// proves the new duplicate-name check does NOT reject the intentional
// cross-source override replaceFileUsers' own doc comment describes: a
// file user of the same name as an inline user, even with a different API
// key, must still win — this is a feature (rotating/promoting a
// statically-configured user through the hot-reloadable file), not a
// duplicate.
func TestAuthStore_ReplaceFileUsers_SameNameAsInline_DifferentKey_OverrideAllowed(t *testing.T) {
a, err := newAuthStore(testAuthCfg()) // inline user "a" / sk-secret, group eng
if err != nil {
t.Fatal(err)
}
if err := a.replaceFileUsers([]*UserConfig{{Name: "a", Group: "eng", APIKey: "sk-file-a"}}); err != nil {
t.Fatalf("replaceFileUsers: want the same-name override allowed, got error: %v", err)
}
inlineReq := httptest.NewRequest("POST", "/v1/chat/completions", nil)
inlineReq.Header.Set("Authorization", "Bearer sk-secret")
if _, _, ok := a.identify(inlineReq); ok {
t.Fatal("want the inline key overridden once the file defines a same-named user")
}
fileReq := httptest.NewRequest("POST", "/v1/chat/completions", nil)
fileReq.Header.Set("Authorization", "Bearer sk-file-a")
u, _, ok := a.identify(fileReq)
if !ok || u.name != "a" {
t.Fatalf("want the file-sourced user resolvable under the overridden name, got %v ok=%v", u, ok)
}
}