remove non-ascii characters from invalid metric names in error message (

#7146) * remove non-ascii characters from invalid metric names Signed-off-by: Mauro Stettler <[email protected]> * update changelog Signed-off-by: Mauro Stettler <[email protected]> * linter Signed-off-by: Mauro Stettler <[email protected]> --------- Signed-off-by: Mauro Stettler <[email protected]>
grafana · Jan 17, 2024 · ca1e7bc · ca1e7bc
1 parent 4fbea7c
commit ca1e7bc
Show file tree

Hide file tree

Showing 3 changed files with 29 additions and 2 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -38,6 +38,7 @@
 * [ENHANCEMENT] Query-frontend: add experimental support for sharding active series queries via `-query-frontend.shard-active-series-queries`. #6784
 * [ENHANCEMENT] Distributor: set `-distributor.reusable-ingester-push-workers=2000` by default and mark feature as `advanced`. #7128
 * [ENHANCEMENT] All: set `-server.grpc.num-workers=100` by default and mark feature as `advanced`. #7131
+* [ENHANCEMENT] Distributor: invalid metric name error message gets cleaned up to not include non-ascii strings. #7146
 * [BUGFIX] Ingester: don't ignore errors encountered while iterating through chunks or samples in response to a query request. #6451
 * [BUGFIX] Fix issue where queries can fail or omit OOO samples if OOO head compaction occurs between creating a querier and reading chunks #6766
 * [BUGFIX] Fix issue where concatenatingChunkIterator can obscure errors #6766

diff --git a/pkg/distributor/validate.go b/pkg/distributor/validate.go
@@ -8,7 +8,9 @@ package distributor
 import (
 	"errors"
 	"fmt"
+	"strings"
 	"time"
+	"unicode"
 	"unicode/utf8"
 
 	"github.com/prometheus/client_golang/prometheus"
@@ -310,6 +312,25 @@ type labelValidationConfig interface {
 	MaxLabelValueLength(userID string) int
 }
 
+func removeNonASCIIChars(in string) (out string) {
+	foundNonASCII := false
+
+	out = strings.Map(func(r rune) rune {
+		if r <= unicode.MaxASCII {
+			return r
+		}
+
+		foundNonASCII = true
+		return -1
+	}, in)
+
+	if foundNonASCII {
+		out = out + " (non-ascii characters removed)"
+	}
+
+	return out
+}
+
 // validateLabels returns an err if the labels are invalid.
 // The returned error may retain the provided series labels.
 func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userID, group string, ls []mimirpb.LabelAdapter, skipLabelNameValidation bool) error {
@@ -321,7 +342,7 @@ func validateLabels(m *sampleValidationMetrics, cfg labelValidationConfig, userI
 
 	if !model.IsValidMetricName(model.LabelValue(unsafeMetricName)) {
 		m.invalidMetricName.WithLabelValues(userID, group).Inc()
-		return fmt.Errorf(invalidMetricNameMsgFormat, unsafeMetricName)
+		return fmt.Errorf(invalidMetricNameMsgFormat, removeNonASCIIChars(unsafeMetricName))
 	}
 
 	numLabelNames := len(ls)

diff --git a/pkg/distributor/validate_test.go b/pkg/distributor/validate_test.go
@@ -80,6 +80,11 @@ func TestValidateLabels(t *testing.T) {
 			false,
 			fmt.Errorf(invalidMetricNameMsgFormat, " "),
 		},
+		{
+			map[model.LabelName]model.LabelValue{model.MetricNameLabel: "metric_name_with_\xb0_invalid_utf8_\xb0"},
+			false,
+			fmt.Errorf(invalidMetricNameMsgFormat, "metric_name_with__invalid_utf8_ (non-ascii characters removed)"),
+		},
 		{
 			map[model.LabelName]model.LabelValue{model.MetricNameLabel: "valid", "foo ": "bar"},
 			false,
@@ -162,7 +167,7 @@ func TestValidateLabels(t *testing.T) {
 			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_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="metric_name_invalid",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="missing_metric_name",user="testUser"} 1
 
 			cortex_discarded_samples_total{group="custom label",reason="random reason",user="different user"} 1