Parameterize metric/label name validation scheme (#11848)

#### What this PR does

Direct and indirect references to the global name validation scheme were
removed in favor of a per-tenant override.

- Distributors have a validation middleware that ensures metric and
label names are valid.
- Rulers validate rule(group)s using this naming scheme.
- Queriers use UTF8 validation *everywhere*.

TODO:

- [x] Reach consensus on whether
`streamingpromqlcompat.NameValidatingEngine` is the right approach:
Decided to do UTF8 validation in query path.
- [x] Get Alertmanager updated with fix, stop using our own fork


#### Which issue(s) this PR fixes or relates to

Depends on
- grafana/mimir-prometheus#946
- grafana/prometheus-alertmanager#116

Fixes: #11503

#### Checklist

- [x] Tests updated.
- [ ] Documentation added.
- [x] `CHANGELOG.md` updated - the order of entries should be
`[CHANGE]`, `[FEATURE]`, `[ENHANCEMENT]`, `[BUGFIX]`. If changelog entry
is not needed, please add the `changelog-not-needed` label to the PR.
- [ ]
[`about-versioning.md`](https://github.com/grafana/mimir/blob/main/docs/sources/mimir/configure/about-versioning.md)
updated with experimental features.

---------

Signed-off-by: Julius Hinze <julius.hinze@grafana.com>
Co-authored-by: Arve Knudsen <arve.knudsen@gmail.com>

Author: juliusmh

CHANGELOG.md

@@ -4,6 +4,8 @@
 
 ### Grafana Mimir
 
+* [CHANGE] Query-frontend: Add support for UTF-8 label and metric names in `/api/v1/cardinality/{label_values|label_values|active_series}` endpoints. #11848.
+* [CHANGE] Querier: Add support for UTF-8 label and metric names in `label_join`, `label_replace` and `count_values` PromQL functions. #11848.
 * [CHANGE] Remove support for Redis as a cache backend. #12163
 * [CHANGE] Memcached: Remove experimental `-<prefix>.memcached.addresses-provider` flag to use alternate DNS service discovery backends. The more reliable backend introduced in 2.16.0 (#10895) is now the default. As a result of this change, DNS-based cache service discovery no longer supports search domains. #12175
 * [FEATURE] Distributor: Add experimental `-distributor.otel-native-delta-ingestion` option to allow primitive delta metrics ingestion via the OTLP endpoint. #11631

Makefile

@@ -344,7 +344,7 @@ lint: check-makefiles
 	faillint -paths "github.com/grafana/mimir/pkg/..." ./pkg/querier/api/...
 	faillint -paths "github.com/grafana/mimir/pkg/..." ./pkg/util/math/...
 
-	# Ensure all errors are report as APIError
+	# Ensure all errors are reported as APIError
 	faillint -paths "github.com/weaveworks/common/httpgrpc.{Errorf}=github.com/grafana/mimir/pkg/api/error.Newf" ./pkg/frontend/querymiddleware/...
 
 	# errors.Cause() only work on errors wrapped by github.com/pkg/errors, while it doesn't work

go.mod

@@ -343,7 +343,7 @@ require (
 	sigs.k8s.io/yaml v1.4.0 // indirect
 )
 
-replace github.com/prometheus/prometheus => github.com/grafana/mimir-prometheus v1.8.2-0.20250725113505-6dd7af9abc56
+replace github.com/prometheus/prometheus => github.com/grafana/mimir-prometheus v1.8.2-0.20250725123259-c4bd4faba234
 
 // Replace memberlist with our fork which includes some fixes that haven't been
 // merged upstream yet:
@@ -370,7 +370,7 @@ 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 v0.25.1-0.20250620093340-be61a673dee6
+replace github.com/prometheus/alertmanager => github.com/grafana/prometheus-alertmanager v0.25.1-0.20250722103749-329f0c4df1ba
 
 // 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

@@ -571,14 +571,14 @@ 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 v1.8.2-0.20250725113505-6dd7af9abc56 h1:HdQyhMJ+AkDUTlFAU7qIIkC3C2n0R/cV8T2S7JnbK7U=
-github.com/grafana/mimir-prometheus v1.8.2-0.20250725113505-6dd7af9abc56/go.mod h1:bi1IiCulyFfPIsfFMaCqlggqiLO4PyqNwK/DiqTaYDI=
+github.com/grafana/mimir-prometheus v1.8.2-0.20250725123259-c4bd4faba234 h1:hwME5D5GMogJkN9yobyKTEvOk/SzUFDnYNkUjLsg3ik=
+github.com/grafana/mimir-prometheus v1.8.2-0.20250725123259-c4bd4faba234/go.mod h1:Pe/2vVv91zryCeOwLSjIFJFsw4Pvd2VNHbTUGu6kUls=
 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 v0.25.1-0.20250620093340-be61a673dee6 h1:oJnbhG6ZNy10AjsgNeAtAKeGHogIGOMfAsBH6fYYa5M=
-github.com/grafana/prometheus-alertmanager v0.25.1-0.20250620093340-be61a673dee6/go.mod h1:O/QP1BCm0HHIzbKvgMzqb5sSyH88rzkFk84F4TfJjBU=
+github.com/grafana/prometheus-alertmanager v0.25.1-0.20250722103749-329f0c4df1ba h1:8u5N0btFygn+2S+B6Xs0HFfq4NJ0kJsX9UpIOlidDmQ=
+github.com/grafana/prometheus-alertmanager v0.25.1-0.20250722103749-329f0c4df1ba/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=

integration/kv_test.go

@@ -9,7 +9,7 @@ package integration
 import (
 	"context"
 	"errors"
-	"sort"
+	"slices"
 	"sync"
 	"testing"
 	"time"
@@ -40,7 +40,7 @@ func TestKVList(t *testing.T) {
 		// Get list of keys and sort them
 		keys, err := client.List(context.Background(), "")
 		require.NoError(t, err, "could not list keys")
-		sort.Strings(keys)
+		slices.Sort(keys)
 		require.Equal(t, keysToCreate, keys, "returned key paths did not match created paths")
 
 		verifyClientMetricsHistogram(t, reg, "cortex_kv_request_duration_seconds", map[string]uint64{

pkg/distributor/distributor.go

@@ -66,12 +66,6 @@ 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")
@@ -1098,15 +1092,18 @@ func (d *Distributor) prePushRelabelMiddleware(next PushFunc) PushFunc {
 			return err
 		}
 
+		dropLabels := d.limits.DropLabels(userID)
+		relabelConfigs := d.limits.MetricRelabelConfigs(userID)
+
 		var removeTsIndexes []int
 		lb := labels.NewBuilder(labels.EmptyLabels())
 		for tsIdx := 0; tsIdx < len(req.Timeseries); tsIdx++ {
 			ts := req.Timeseries[tsIdx]
 
-			if mrc := d.limits.MetricRelabelConfigs(userID); len(mrc) > 0 {
+			if len(relabelConfigs) > 0 {
 				mimirpb.FromLabelAdaptersToBuilder(ts.Labels, lb)
 				lb.Set(metaLabelTenantID, userID)
-				keep := relabel.ProcessBuilder(lb, mrc...)
+				keep := relabel.ProcessBuilder(lb, relabelConfigs...)
 				if !keep {
 					removeTsIndexes = append(removeTsIndexes, tsIdx)
 					continue
@@ -1115,7 +1112,7 @@ func (d *Distributor) prePushRelabelMiddleware(next PushFunc) PushFunc {
 				req.Timeseries[tsIdx].SetLabels(mimirpb.FromBuilderToLabelAdapters(lb, ts.Labels))
 			}
 
-			for _, labelName := range d.limits.DropLabels(userID) {
+			for _, labelName := range dropLabels {
 				req.Timeseries[tsIdx].RemoveLabel(labelName)
 			}
 

pkg/distributor/validate.go

@@ -16,6 +16,7 @@ import (
 
 	"github.com/prometheus/client_golang/prometheus"
 	"github.com/prometheus/common/model"
+	"github.com/prometheus/prometheus/model/labels"
 
 	"github.com/grafana/mimir/pkg/costattribution"
 	"github.com/grafana/mimir/pkg/mimirpb"
@@ -390,6 +391,7 @@ 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 +423,9 @@ func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userI
 		return errors.New(noMetricNameMsgFormat)
 	}
 
-	if !model.IsValidMetricName(model.LabelValue(unsafeMetricName)) {
+	validationScheme := cfg.ValidationScheme(userID)
+
+	if !labels.IsValidMetricName(unsafeMetricName, validationScheme) {
 		cat.IncrementDiscardedSamples(ls, 1, reasonInvalidMetricName, ts)
 		m.invalidMetricName.WithLabelValues(userID, group).Inc()
 		return fmt.Errorf(invalidMetricNameMsgFormat, removeNonASCIIChars(unsafeMetricName))
@@ -447,7 +451,7 @@ 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 && !labels.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))