Move validation into Mimir (remove global name validation)

go.mod

@@ -345,7 +345,7 @@ require (
 	sigs.k8s.io/yaml v1.4.0 // indirect
 )
 
-replace github.com/prometheus/prometheus => github.com/grafana/mimir-prometheus v1.8.2-0.20250715103124-73cbc98c25ff
+replace github.com/prometheus/prometheus => github.com/grafana/mimir-prometheus v1.8.2-0.20250716123832-f18c798bd04b
 
 // Replace memberlist with our fork which includes some fixes that haven't been
 // merged upstream yet:
@@ -372,7 +372,7 @@ replace github.com/opentracing-contrib/go-stdlib => github.com/grafana/opentraci
 replace github.com/opentracing-contrib/go-grpc => github.com/charleskorn/go-grpc v0.0.0-20231024023642-e9298576254f
 
 // Replacing prometheus/alertmanager with our fork.
-replace github.com/prometheus/alertmanager => github.com/grafana/prometheus-alertmanager v0.25.1-0.20250604130045-92c8f6389b36
+replace github.com/prometheus/alertmanager => github.com/juliusmh/alertmanager v0.26.1-0.20250716125725-19a9223bec8c
 
 // Use Mimir fork of prometheus/otlptranslator to allow for higher velocity of upstream development,
 // while allowing Mimir to move at a more conservative pace.

go.sum

@@ -579,14 +579,12 @@ github.com/grafana/memberlist v0.3.1-0.20250428154222-f7d51a6f6700 h1:0t7iOQ5ZkB
 github.com/grafana/memberlist v0.3.1-0.20250428154222-f7d51a6f6700/go.mod h1:Ri9p/tRShbjYnpNf4FFPXG7wxEGY4Nrcn6E7jrVa//4=
 github.com/grafana/mimir-otlptranslator v0.0.0-20250703083430-c31a9568ad96 h1:kq5zJVW9LyFOB5xCeQPTON2HNjwwEkefhegZXGIhQPk=
 github.com/grafana/mimir-otlptranslator v0.0.0-20250703083430-c31a9568ad96/go.mod h1:P8AwMgdD7XEr6QRUJ2QWLpiAZTgTE2UYgjlu3svompI=
-github.com/grafana/mimir-prometheus v1.8.2-0.20250715103124-73cbc98c25ff h1:u+lw876tigZBv4/jGIvDmOJ6uLmscQq8HURg38DaldM=
-github.com/grafana/mimir-prometheus v1.8.2-0.20250715103124-73cbc98c25ff/go.mod h1:51sCkEHDpexFCGizYi1W4omiFaKThWThkquNPOxhwxQ=
+github.com/grafana/mimir-prometheus v1.8.2-0.20250716123832-f18c798bd04b h1:qZM11g9GR8sibnedD0zt56uASxhxjnx/CMP/JAlCRZQ=
+github.com/grafana/mimir-prometheus v1.8.2-0.20250716123832-f18c798bd04b/go.mod h1:51sCkEHDpexFCGizYi1W4omiFaKThWThkquNPOxhwxQ=
 github.com/grafana/opentracing-contrib-go-stdlib v0.0.0-20230509071955-f410e79da956 h1:em1oddjXL8c1tL0iFdtVtPloq2hRPen2MJQKoAWpxu0=
 github.com/grafana/opentracing-contrib-go-stdlib v0.0.0-20230509071955-f410e79da956/go.mod h1:qtI1ogk+2JhVPIXVc6q+NHziSmy2W5GbdQZFUHADCBU=
 github.com/grafana/otel-profiling-go v0.5.1 h1:stVPKAFZSa7eGiqbYuG25VcqYksR6iWvF3YH66t4qL8=
 github.com/grafana/otel-profiling-go v0.5.1/go.mod h1:ftN/t5A/4gQI19/8MoWurBEtC6gFw8Dns1sJZ9W4Tls=
-github.com/grafana/prometheus-alertmanager v0.25.1-0.20250604130045-92c8f6389b36 h1:AjZ58JRw1ZieFH/SdsddF5BXtsDKt5kSrKNPWrzYz3Y=
-github.com/grafana/prometheus-alertmanager v0.25.1-0.20250604130045-92c8f6389b36/go.mod h1:O/QP1BCm0HHIzbKvgMzqb5sSyH88rzkFk84F4TfJjBU=
 github.com/grafana/pyroscope-go/godeltaprof v0.1.8 h1:iwOtYXeeVSAeYefJNaxDytgjKtUuKQbJqgAIjlnicKg=
 github.com/grafana/pyroscope-go/godeltaprof v0.1.8/go.mod h1:2+l7K7twW49Ct4wFluZD3tZ6e0SjanjcUUBPVD/UuGU=
 github.com/grafana/regexp v0.0.0-20240531075221-3685f1377d7b h1:oMAq12GxTpwo9jxbnG/M4F/HdpwbibTaVoxNA0NZprY=
@@ -721,6 +719,8 @@ github.com/jstemmer/go-junit-report v0.9.1/go.mod h1:Brl9GWCQeLvo8nXZwPNNblvFj/X
 github.com/julienschmidt/httprouter v1.2.0/go.mod h1:SYymIcj16QtmaHHD7aYtjjsJG7VTCxuUUipMqKk8s4w=
 github.com/julienschmidt/httprouter v1.3.0 h1:U0609e9tgbseu3rBINet9P48AI/D3oJs4dN7jwJOQ1U=
 github.com/julienschmidt/httprouter v1.3.0/go.mod h1:JR6WtHb+2LUe8TCKY3cZOxFyyO8IZAc4RVcycCCAKdM=
+github.com/juliusmh/alertmanager v0.26.1-0.20250716125725-19a9223bec8c h1:VaNJy7gIh+FGGaDOmViA7bUsLeTVs10tQ5tCVXHbo6c=
+github.com/juliusmh/alertmanager v0.26.1-0.20250716125725-19a9223bec8c/go.mod h1:O/QP1BCm0HHIzbKvgMzqb5sSyH88rzkFk84F4TfJjBU=
 github.com/jung-kurt/gofpdf v1.0.3-0.20190309125859-24315acbbda5/go.mod h1:7Id9E/uU8ce6rXgefFLlgrJj/GYY22cpxn+r32jIOes=
 github.com/keybase/go-keychain v0.0.1 h1:way+bWYa6lDppZoZcgMbYsvC7GxljxrskdNInRtuthU=
 github.com/keybase/go-keychain v0.0.1/go.mod h1:PdEILRW3i9D8JcdM+FmY6RwkHGnhHxXwkPPMeUgOK1k=

pkg/distributor/distributor.go

@@ -67,12 +67,6 @@ import (
 
 var tracer = otel.Tracer("pkg/distributor")
 
-func init() {
-	// Mimir doesn't support Prometheus' UTF-8 metric/label name scheme yet.
-	// nolint:staticcheck
-	model.NameValidationScheme = model.LegacyValidation
-}
-
 var (
 	// Validation errors.
 	errInvalidTenantShardSize = errors.New("invalid tenant shard size, the value must be greater than or equal to zero")
@@ -1080,6 +1074,8 @@ func (d *Distributor) prePushHaDedupeMiddleware(next PushFunc) PushFunc {
 	}
 }
 
+// prePushRelabelMiddleware relabels the metric. Validation of those labels is
+// handled in prePushValidationMiddleware.
 func (d *Distributor) prePushRelabelMiddleware(next PushFunc) PushFunc {
 	return func(ctx context.Context, pushReq *Request) error {
 		next, maybeCleanup := NextOrCleanup(next, pushReq)

pkg/distributor/validate.go

@@ -390,6 +390,7 @@ type labelValidationConfig interface {
 	MaxLabelNamesPerInfoSeries(userID string) int
 	MaxLabelNameLength(userID string) int
 	MaxLabelValueLength(userID string) int
+	ValidationScheme(userID string) model.ValidationScheme
 }
 
 func removeNonASCIIChars(in string) (out string) {
@@ -421,7 +422,9 @@ func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userI
 		return errors.New(noMetricNameMsgFormat)
 	}
 
-	if !model.IsValidMetricName(model.LabelValue(unsafeMetricName)) {
+	validationScheme := cfg.ValidationScheme(userID)
+
+	if !isValidMetricName(unsafeMetricName, validationScheme) {
 		cat.IncrementDiscardedSamples(ls, 1, reasonInvalidMetricName, ts)
 		m.invalidMetricName.WithLabelValues(userID, group).Inc()
 		return fmt.Errorf(invalidMetricNameMsgFormat, removeNonASCIIChars(unsafeMetricName))
@@ -447,7 +450,7 @@ func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userI
 	maxLabelValueLength := cfg.MaxLabelValueLength(userID)
 	lastLabelName := ""
 	for _, l := range ls {
-		if !skipLabelValidation && !model.LabelName(l.Name).IsValid() {
+		if !skipLabelValidation && !isValidLabelName(l.Name, validationScheme) {
 			m.invalidLabel.WithLabelValues(userID, group).Inc()
 			cat.IncrementDiscardedSamples(ls, 1, reasonInvalidLabel, ts)
 			return fmt.Errorf(invalidLabelMsgFormat, l.Name, mimirpb.FromLabelAdaptersToString(ls))
@@ -474,6 +477,28 @@ func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userI
 	return nil
 }
 
+func isValidMetricName(name string, validationScheme model.ValidationScheme) bool {
+	switch validationScheme {
+	case model.LegacyValidation:
+		return model.IsValidLegacyMetricName(name)
+	case model.UTF8Validation:
+		return model.IsValidMetricName(model.LabelValue(name))
+	default:
+		return false
+	}
+}
+
+func isValidLabelName(name string, validationScheme model.ValidationScheme) bool {
+	switch validationScheme {
+	case model.LegacyValidation:
+		return model.LabelName(name).IsValidLegacy()
+	case model.UTF8Validation:
+		return model.LabelName(name).IsValid()
+	default:
+		return false
+	}
+}
+
 // metadataValidationMetrics is a collection of metrics used by metadata validation.
 type metadataValidationMetrics struct {
 	missingMetricName *prometheus.CounterVec

pkg/distributor/validate_test.go

@@ -37,6 +37,7 @@ type validateLabelsCfg struct {
 	maxLabelNamesPerInfoSeries int
 	maxLabelNameLength         int
 	maxLabelValueLength        int
+	validationScheme           model.ValidationScheme
 }
 
 func (v validateLabelsCfg) MaxLabelNamesPerSeries(_ string) int {
@@ -55,6 +56,10 @@ func (v validateLabelsCfg) MaxLabelValueLength(_ string) int {
 	return v.maxLabelValueLength
 }
 
+func (v validateLabelsCfg) ValidationScheme(_ string) model.ValidationScheme {
+	return v.validationScheme
+}
+
 type validateMetadataCfg struct {
 	enforceMetadataMetricName bool
 	maxMetadataLength         int
@@ -73,13 +78,19 @@ func TestValidateLabels(t *testing.T) {
 	reg := prometheus.NewPedanticRegistry()
 	s := newSampleValidationMetrics(reg)
 
-	var cfg validateLabelsCfg
 	userID := "testUser"
+	legacyConfig := validateLabelsCfg{
+		maxLabelNamesPerSeries:     3,
+		maxLabelNamesPerInfoSeries: 4,
+		maxLabelNameLength:         25,
+		maxLabelValueLength:        25,
+		validationScheme:           model.LegacyValidation,
+	}
+	utf8Config := legacyConfig
+	utf8Config.validationScheme = model.UTF8Validation
+	unsetConfig := legacyConfig
+	unsetConfig.validationScheme = model.UnsetValidation
 
-	cfg.maxLabelValueLength = 25
-	cfg.maxLabelNameLength = 25
-	cfg.maxLabelNamesPerSeries = 3
-	cfg.maxLabelNamesPerInfoSeries = 4
 	limits := catestutils.NewMockCostAttributionLimits(0, userID, "team")
 	careg := prometheus.NewRegistry()
 	manager, err := costattribution.NewManager(5*time.Second, 10*time.Second, log.NewNopLogger(), limits, reg, careg)
@@ -90,30 +101,35 @@ func TestValidateLabels(t *testing.T) {
 		metric                   model.Metric
 		skipLabelNameValidation  bool
 		skipLabelCountValidation bool
+		config                   validateLabelsCfg
 		err                      error
 	}{
 		{
 			metric:                   map[model.LabelName]model.LabelValue{"team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      errors.New(noMetricNameMsgFormat),
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: " ", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      fmt.Errorf(invalidMetricNameMsgFormat, " "),
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "metric_name_with_\xb0_invalid_utf8_\xb0", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      fmt.Errorf(invalidMetricNameMsgFormat, "metric_name_with__invalid_utf8_ (non-ascii characters removed)"),
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "valid", "foo ": "bar", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: fmt.Errorf(
 				invalidLabelMsgFormat,
 				"foo ",
@@ -130,12 +146,14 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "valid", "team": "c"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      nil,
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "badLabelName", "this_is_a_really_really_long_name_that_should_cause_an_error": "test_value_please_ignore", "team": "biz"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: fmt.Errorf(
 				labelNameTooLongMsgFormat,
 				"this_is_a_really_really_long_name_that_should_cause_an_error",
@@ -152,6 +170,7 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "badLabelValue", "much_shorter_name": "test_value_please_ignore_no_really_nothing_to_see_here", "team": "biz"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: LabelValueTooLongError{
 				Label: mimirpb.LabelAdapter{Name: "much_shorter_name", Value: "test_value_please_ignore_no_really_nothing_to_see_here"},
 				Limit: 25,
@@ -166,6 +185,7 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "bar": "baz", "blip": "blop", "team": "plof"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: fmt.Errorf(
 				tooManyLabelsMsgFormat,
 				tooManyLabelsArgs(
@@ -184,13 +204,15 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo_info", "bar": "baz", "blip": "blop", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      nil,
 		},
 		{
 			// *_info metrics have higher label limits.
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo_info", "bar": "baz", "blip": "blop", "blap": "blup", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: fmt.Errorf(
 				tooManyInfoLabelsMsgFormat,
 				tooManyLabelsArgs(
@@ -209,24 +231,28 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "bar": "baz", "blip": "blop", "team": "a"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: true,
+			config:                   legacyConfig,
 			err:                      nil,
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "invalid%label&name": "bar", "team": "biz"},
 			skipLabelNameValidation:  true,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      nil,
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "label1": "你好", "team": "plof"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err:                      nil,
 		},
 		{
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "label1": "abc\xfe\xfddef", "team": "plof"},
 			skipLabelNameValidation:  false,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
 			err: fmt.Errorf(
 				invalidLabelValueMsgFormat,
 				"label1", "abc\ufffddef", "foo",
@@ -236,10 +262,46 @@ func TestValidateLabels(t *testing.T) {
 			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "label1": "abc\xfe\xfddef"},
 			skipLabelNameValidation:  true,
 			skipLabelCountValidation: false,
+			config:                   legacyConfig,
+			err:                      nil,
+		},
+		{
+			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "name😀": "value", "team": "b"},
+			skipLabelNameValidation:  false,
+			skipLabelCountValidation: false,
+			config:                   legacyConfig,
+			err: fmt.Errorf(
+				invalidLabelMsgFormat,
+				"name😀",
+				mimirpb.FromLabelAdaptersToString(
+					[]mimirpb.LabelAdapter{
+						{Name: model.MetricNameLabel, Value: "foo"},
+						{Name: "name😀", Value: "value"},
+						{Name: "team", Value: "b"},
+					},
+				),
+			),
+		},
+		// Unset validation scheme results in invalid metric name error
+		{
+			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "name😀": "value", "team": "b"},
+			skipLabelNameValidation:  false,
+			skipLabelCountValidation: false,
+			config:                   unsetConfig,
+			err: fmt.Errorf(
+				invalidMetricNameMsgFormat,
+				"foo",
+			),
+		},
+		{
+			metric:                   map[model.LabelName]model.LabelValue{model.MetricNameLabel: "foo", "name😀": "value", "team": "b"},
+			skipLabelNameValidation:  false,
+			skipLabelCountValidation: false,
+			config:                   utf8Config,
 			err:                      nil,
 		},
 	} {
-		err := validateLabels(s, cfg, userID, "custom label", mimirpb.FromMetricsToLabelAdapters(c.metric), c.skipLabelNameValidation, c.skipLabelCountValidation, cast, ts)
+		err := validateLabels(s, c.config, userID, "custom label", mimirpb.FromMetricsToLabelAdapters(c.metric), c.skipLabelNameValidation, c.skipLabelCountValidation, cast, ts)
 		assert.Equal(t, c.err, err, "wrong error")
 	}
 
@@ -249,13 +311,13 @@ func TestValidateLabels(t *testing.T) {
 	require.NoError(t, testutil.GatherAndCompare(reg, strings.NewReader(`
 			# HELP cortex_discarded_samples_total The total number of samples that were discarded.
 			# TYPE cortex_discarded_samples_total counter
-			cortex_discarded_samples_total{group="custom label",reason="label_invalid",user="testUser"} 1
+			cortex_discarded_samples_total{group="custom label",reason="label_invalid",user="testUser"} 2
 			cortex_discarded_samples_total{group="custom label",reason="label_name_too_long",user="testUser"} 1
 			cortex_discarded_samples_total{group="custom label",reason="label_value_invalid",user="testUser"} 1
 			cortex_discarded_samples_total{group="custom label",reason="label_value_too_long",user="testUser"} 1
 			cortex_discarded_samples_total{group="custom label",reason="max_label_names_per_series",user="testUser"} 1
 			cortex_discarded_samples_total{group="custom label",reason="max_label_names_per_info_series",user="testUser"} 1
-			cortex_discarded_samples_total{group="custom label",reason="metric_name_invalid",user="testUser"} 2
+			cortex_discarded_samples_total{group="custom label",reason="metric_name_invalid",user="testUser"} 3
 			cortex_discarded_samples_total{group="custom label",reason="missing_metric_name",user="testUser"} 1
 			cortex_discarded_samples_total{group="custom label",reason="random reason",user="different user"} 1
 	`), "cortex_discarded_samples_total"))
@@ -264,12 +326,14 @@ func TestValidateLabels(t *testing.T) {
 	# HELP cortex_discarded_attributed_samples_total The total number of samples that were discarded per attribution.
 	# TYPE cortex_discarded_attributed_samples_total counter
 	cortex_discarded_attributed_samples_total{reason="label_invalid",team="a",tenant="testUser",tracker="cost-attribution"} 1
+	cortex_discarded_attributed_samples_total{reason="label_invalid",team="b",tenant="testUser",tracker="cost-attribution"} 1
 	cortex_discarded_attributed_samples_total{reason="label_name_too_long",team="biz",tenant="testUser",tracker="cost-attribution"} 1
 	cortex_discarded_attributed_samples_total{reason="label_value_invalid",team="plof",tenant="testUser",tracker="cost-attribution"} 1
 	cortex_discarded_attributed_samples_total{reason="label_value_too_long",team="biz",tenant="testUser",tracker="cost-attribution"} 1
     cortex_discarded_attributed_samples_total{reason="max_label_names_per_info_series",team="a",tenant="testUser",tracker="cost-attribution"} 1
     cortex_discarded_attributed_samples_total{reason="max_label_names_per_series",team="plof",tenant="testUser",tracker="cost-attribution"} 1
     cortex_discarded_attributed_samples_total{reason="metric_name_invalid",team="a",tenant="testUser",tracker="cost-attribution"} 2
+    cortex_discarded_attributed_samples_total{reason="metric_name_invalid",team="b",tenant="testUser",tracker="cost-attribution"} 1
     cortex_discarded_attributed_samples_total{reason="missing_metric_name",team="a",tenant="testUser",tracker="cost-attribution"} 1
 `), "cortex_discarded_attributed_samples_total"))
 
@@ -451,6 +515,7 @@ func TestValidateLabelDuplication(t *testing.T) {
 	cfg.maxLabelNameLength = 10
 	cfg.maxLabelNamesPerSeries = 10
 	cfg.maxLabelValueLength = 10
+	cfg.validationScheme = model.UTF8Validation
 
 	userID := "testUser"
 	actual := validateLabels(newSampleValidationMetrics(nil), cfg, userID, "", []mimirpb.LabelAdapter{

pkg/frontend/querymiddleware/request_validation.go

@@ -7,15 +7,8 @@ import (
 	"net/http"
 
 	"github.com/grafana/dskit/cancellation"
-	"github.com/prometheus/common/model"
 )
 
-func init() {
-	// Mimir doesn't support Prometheus' UTF-8 metric/label name scheme yet.
-	// nolint:staticcheck
-	model.NameValidationScheme = model.LegacyValidation
-}
-
 const requestValidationFailedFmt = "request validation failed for "
 
 var errMetricsQueryRequestValidationFailed = cancellation.NewErrorf(