Allow ruler to retrieve proto format query response

Signed-off-by: SungJin1212 <[email protected]>

CHANGELOG.md

@@ -9,6 +9,7 @@
 * [CHANGE] Change all max async concurrency default values `50` to `3` #6268
 * [CHANGE] Change default value of `-blocks-storage.bucket-store.index-cache.multilevel.max-async-concurrency` from `50` to `3` #6265
 * [CHANGE] Enable Compactor and Alertmanager in target all. #6204
+* [FEATURE] Ruler: Add an experimental flag `-ruler.query-response-format` to retrieve query response as a proto format. #6345
 * [FEATURE] Ruler: Pagination support for List Rules API. #6299
 * [FEATURE] Query Frontend/Querier: Add protobuf codec `-api.querier-default-codec` and the option to choose response compression type `-querier.response-compression`. #5527
 * [FEATURE] Ruler: Experimental: Add `ruler.frontend-address` to allow query to query frontends instead of ingesters. #6151

docs/configuration/config-file-reference.md

@@ -4313,6 +4313,12 @@ The `ruler_config` configures the Cortex ruler.
 # CLI flag: -ruler.frontend-address
 [frontend_address: <string> | default = ""]
 
+# [Experimental] Query response format to get query results from Query Frontend
+# when the rule evaluation. It will only take effect when
+# `-ruler.frontend-address` is configured. Supported values: json,protobuf
+# CLI flag: -ruler.query-response-format
+[query_response_format: <string> | default = "protobuf"]
+
 frontend_client:
   # gRPC client max receive message size (bytes).
   # CLI flag: -ruler.frontendClient.grpc-max-recv-msg-size

docs/configuration/v1-guarantees.md

@@ -35,7 +35,9 @@ Cortex is an actively developed project and we want to encourage the introductio
 
 Currently experimental features are:
 
-- Ruler: Evaluate rules to query frontend instead of ingesters (enabled via `-ruler.frontend-address` )
+- Ruler
+  - Evaluate rules to query frontend instead of ingesters (enabled via `-ruler.frontend-address`).
+  - When `-ruler.frontend-address` is specified, the response format can be specified (via `-ruler.query-response-format`).
 - S3 Server Side Encryption (SSE) using KMS (including per-tenant KMS config overrides).
 - Azure blob storage.
 - Zone awareness based replication.

go.mod

@@ -81,6 +81,7 @@ require (
 	github.com/cespare/xxhash/v2 v2.3.0
 	github.com/google/go-cmp v0.6.0
 	github.com/hashicorp/golang-lru/v2 v2.0.7
+	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822
 	github.com/sercand/kuberesolver/v5 v5.1.1
 	github.com/tjhop/slog-gokit v0.1.2
 	go.opentelemetry.io/collector/pdata v1.21.0
@@ -189,7 +190,6 @@ require (
 	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd // indirect
 	github.com/modern-go/reflect2 v1.0.2 // indirect
-	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
 	github.com/mwitkow/go-conntrack v0.0.0-20190716064945-2f068394615f // indirect
 	github.com/ncw/swift v1.0.53 // indirect
 	github.com/oklog/run v1.1.0 // indirect

integration/ruler_test.go

@@ -1670,42 +1670,64 @@ func TestRulerEvalWithQueryFrontend(t *testing.T) {
 	distributor := e2ecortex.NewDistributor("distributor", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), flags, "")
 	ingester := e2ecortex.NewIngester("ingester", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), flags, "")
 	require.NoError(t, s.StartAndWaitReady(distributor, ingester))
-	queryFrontend := e2ecortex.NewQueryFrontend("query-frontend", flags, "")
-	require.NoError(t, s.Start(queryFrontend))
-
-	ruler := e2ecortex.NewRuler("ruler", consul.NetworkHTTPEndpoint(), mergeFlags(flags, map[string]string{
-		"-ruler.frontend-address": queryFrontend.NetworkGRPCEndpoint(),
-	}), "")
-	querier := e2ecortex.NewQuerier("querier", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), mergeFlags(flags, map[string]string{
-		"-querier.frontend-address": queryFrontend.NetworkGRPCEndpoint(),
-	}), "")
-	require.NoError(t, s.StartAndWaitReady(ruler, querier))
-
-	c, err := e2ecortex.NewClient("", "", "", ruler.HTTPEndpoint(), user)
-	require.NoError(t, err)
+	for _, format := range []string{"protobuf", "json"} {
+		t.Run(fmt.Sprintf("format:%s", format), func(t *testing.T) {
+			queryFrontendFlag := mergeFlags(flags, map[string]string{
+				"-ruler.query-response-format": format,
+			})
+			queryFrontend := e2ecortex.NewQueryFrontend("query-frontend", queryFrontendFlag, "")
+			require.NoError(t, s.Start(queryFrontend))
 
-	expression := "metric"
-	groupName := "rule_group"
-	ruleName := "rule_name"
-	require.NoError(t, c.SetRuleGroup(ruleGroupWithRule(groupName, ruleName, expression), namespace))
+			querier := e2ecortex.NewQuerier("querier", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), mergeFlags(queryFrontendFlag, map[string]string{
+				"-querier.frontend-address": queryFrontend.NetworkGRPCEndpoint(),
+			}), "")
+			require.NoError(t, s.StartAndWaitReady(querier))
 
-	rgMatcher := ruleGroupMatcher(user, namespace, groupName)
-	// Wait until ruler has loaded the group.
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(1), []string{"cortex_prometheus_rule_group_rules"}, e2e.WithLabelMatchers(rgMatcher), e2e.WaitMissingMetrics))
-	// Wait until rule group has tried to evaluate the rule.
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_prometheus_rule_evaluations_total"}, e2e.WithLabelMatchers(rgMatcher), e2e.WaitMissingMetrics))
+			rulerFlag := mergeFlags(queryFrontendFlag, map[string]string{
+				"-ruler.frontend-address": queryFrontend.NetworkGRPCEndpoint(),
+			})
+			ruler := e2ecortex.NewRuler("ruler", consul.NetworkHTTPEndpoint(), rulerFlag, "")
+			require.NoError(t, s.StartAndWaitReady(ruler))
 
-	matcher := labels.MustNewMatcher(labels.MatchEqual, "user", user)
-	// Check that cortex_ruler_query_frontend_clients went up
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(1), []string{"cortex_ruler_query_frontend_clients"}, e2e.WaitMissingMetrics))
-	// Check that cortex_ruler_queries_total went up
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_ruler_queries_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
-	// Check that cortex_ruler_queries_failed_total is zero
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(0), []string{"cortex_ruler_queries_failed_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
-	// Check that cortex_ruler_write_requests_total went up
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_ruler_write_requests_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
-	// Check that cortex_ruler_write_requests_failed_total is zero
-	require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(0), []string{"cortex_ruler_write_requests_failed_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
+			t.Cleanup(func() {
+				_ = s.Stop(ruler)
+				_ = s.Stop(queryFrontend)
+				_ = s.Stop(querier)
+			})
+
+			c, err := e2ecortex.NewClient("", "", "", ruler.HTTPEndpoint(), user)
+			require.NoError(t, err)
+
+			expression := "metric" // vector
+			//expression := "scalar(count(up == 1)) > bool 1" // scalar
+			groupName := "rule_group"
+			ruleName := "rule_name"
+			require.NoError(t, c.SetRuleGroup(ruleGroupWithRule(groupName, ruleName, expression), namespace))
+
+			rgMatcher := ruleGroupMatcher(user, namespace, groupName)
+			// Wait until ruler has loaded the group.
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(1), []string{"cortex_prometheus_rule_group_rules"}, e2e.WithLabelMatchers(rgMatcher), e2e.WaitMissingMetrics))
+			// Wait until rule group has tried to evaluate the rule.
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_prometheus_rule_evaluations_total"}, e2e.WithLabelMatchers(rgMatcher), e2e.WaitMissingMetrics))
+			// Make sure not to fail
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(0), []string{"cortex_prometheus_rule_evaluation_failures_total"}, e2e.WithLabelMatchers(rgMatcher), e2e.WaitMissingMetrics))
+
+			matcher := labels.MustNewMatcher(labels.MatchEqual, "user", user)
+			sourceMatcher := labels.MustNewMatcher(labels.MatchEqual, "source", "ruler")
+			// Check that cortex_ruler_query_frontend_clients went up
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(1), []string{"cortex_ruler_query_frontend_clients"}, e2e.WaitMissingMetrics))
+			// Check that cortex_ruler_queries_total went up
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_ruler_queries_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
+			// Check that cortex_ruler_queries_failed_total is zero
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(0), []string{"cortex_ruler_queries_failed_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
+			// Check that cortex_ruler_write_requests_total went up
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_ruler_write_requests_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
+			// Check that cortex_ruler_write_requests_failed_total is zero
+			require.NoError(t, ruler.WaitSumMetricsWithOptions(e2e.Equals(0), []string{"cortex_ruler_write_requests_failed_total"}, e2e.WithLabelMatchers(matcher), e2e.WaitMissingMetrics))
+			// Check that cortex_query_frontend_queries_total went up
+			require.NoError(t, queryFrontend.WaitSumMetricsWithOptions(e2e.GreaterOrEqual(1), []string{"cortex_query_frontend_queries_total"}, e2e.WithLabelMatchers(matcher, sourceMatcher), e2e.WaitMissingMetrics))
+		})
+	}
 }
 
 func parseAlertFromRule(t *testing.T, rules interface{}) *alertingRule {

pkg/querier/codec/protobuf_codec.go

@@ -25,8 +25,8 @@ func (p ProtobufCodec) ContentType() v1.MIMEType {
 	if !p.CortexInternal {
 		return v1.MIMEType{Type: "application", SubType: "x-protobuf"}
 	}
-	// TODO: switch to use constants.
-	return v1.MIMEType{Type: "application", SubType: "x-cortex-query+proto"}
+
+	return v1.MIMEType{Type: "application", SubType: tripperware.QueryResponseCortexMIMESubType}
 }
 
 func (p ProtobufCodec) CanEncode(resp *v1.Response) bool {

pkg/querier/codec/protobuf_codec_test.go

@@ -42,6 +42,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 		expected       *tripperware.PrometheusResponse
 	}{
 		{
+			name: "vector",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeVector,
 				Result: promql.Vector{
@@ -85,6 +86,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
+			name: "scalar",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeScalar,
 				Result:     promql.Scalar{T: 1000, V: 1},
@@ -147,6 +149,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
+			name: "matrix",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeMatrix,
 				Result: promql.Matrix{
@@ -180,6 +183,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
+			name: "matrix with multiple series",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeMatrix,
 				Result: promql.Matrix{
@@ -223,6 +227,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
+			name: "matrix: not cortex internal with histogram",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeMatrix,
 				Result: promql.Matrix{
@@ -312,6 +317,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
+			name: "vector: not cortex internal with histogram",
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeVector,
 				Result: promql.Vector{
@@ -400,7 +406,7 @@ func TestProtobufCodec_Encode(t *testing.T) {
 			},
 		},
 		{
-			name:           "cortex internal with native histogram",
+			name:           "vector: cortex internal with native histogram",
 			cortexInternal: true,
 			data: &v1.QueryData{
 				ResultType: parser.ValueTypeVector,

pkg/querier/tripperware/instantquery/instant_query.go

@@ -11,9 +11,11 @@ import (
 	"time"
 
 	jsoniter "github.com/json-iterator/go"
+	"github.com/munnerz/goautoneg"
 	"github.com/opentracing/opentracing-go"
 	otlog "github.com/opentracing/opentracing-go/log"
 	"github.com/prometheus/common/model"
+	v1 "github.com/prometheus/prometheus/web/api/v1"
 	"github.com/weaveworks/common/httpgrpc"
 	"google.golang.org/grpc/status"
 
@@ -28,6 +30,9 @@ var (
 		SortMapKeys:            true,
 		ValidateJsonRawMessage: false,
 	}.Froze()
+
+	rulerMIMEType = v1.MIMEType{Type: "application", SubType: tripperware.QueryResponseCortexMIMESubType}
+	jsonMIMEType  = v1.MIMEType{Type: "application", SubType: "json"}
 )
 
 type instantQueryCodec struct {
@@ -67,12 +72,18 @@ func (c instantQueryCodec) DecodeRequest(_ context.Context, r *http.Request, for
 	result.Stats = r.FormValue("stats")
 	result.Path = r.URL.Path
 
-	// Include the specified headers from http request in prometheusRequest.
-	for _, header := range forwardHeaders {
-		for h, hv := range r.Header {
-			if strings.EqualFold(h, header) {
-				result.Headers[h] = hv
-				break
+	isSourceRuler := strings.Contains(r.Header.Get("User-Agent"), tripperware.RulerUserAgent)
+	if isSourceRuler {
+		// When the source is the Ruler, then forward whole headers
+		result.Headers = r.Header
+	} else {
+		// Include the specified headers from http request in prometheusRequest.
+		for _, header := range forwardHeaders {
+			for h, hv := range r.Header {
+				if strings.EqualFold(h, header) {
+					result.Headers[h] = hv
+					break
+				}
 			}
 		}
 	}
@@ -154,7 +165,11 @@ func (c instantQueryCodec) EncodeRequest(ctx context.Context, r tripperware.Requ
 		}
 	}
 
-	tripperware.SetRequestHeaders(h, c.defaultCodecType, c.compression)
+	isSourceRuler := strings.Contains(h.Get("User-Agent"), tripperware.RulerUserAgent)
+	if isSourceRuler {
+		// When the source is the Ruler, skip set header
+		tripperware.SetRequestHeaders(h, c.defaultCodecType, c.compression)
+	}
 
 	req := &http.Request{
 		Method:     "GET",
@@ -167,7 +182,7 @@ func (c instantQueryCodec) EncodeRequest(ctx context.Context, r tripperware.Requ
 	return req.WithContext(ctx), nil
 }
 
-func (instantQueryCodec) EncodeResponse(ctx context.Context, res tripperware.Response) (*http.Response, error) {
+func (c instantQueryCodec) EncodeResponse(ctx context.Context, req *http.Request, res tripperware.Response) (*http.Response, error) {
 	sp, _ := opentracing.StartSpanFromContext(ctx, "APIResponse.ToHTTPResponse")
 	defer sp.Finish()
 
@@ -176,7 +191,7 @@ func (instantQueryCodec) EncodeResponse(ctx context.Context, res tripperware.Res
 		return nil, httpgrpc.Errorf(http.StatusInternalServerError, "invalid response format")
 	}
 
-	b, err := json.Marshal(a)
+	contentType, b, err := marshalResponse(a, req.Header.Get("Accept"))
 	if err != nil {
 		return nil, httpgrpc.Errorf(http.StatusInternalServerError, "error encoding response: %v", err)
 	}
@@ -185,7 +200,7 @@ func (instantQueryCodec) EncodeResponse(ctx context.Context, res tripperware.Res
 
 	resp := http.Response{
 		Header: http.Header{
-			"Content-Type": []string{tripperware.ApplicationJson},
+			"Content-Type": []string{contentType},
 		},
 		Body:          io.NopCloser(bytes.NewBuffer(b)),
 		StatusCode:    http.StatusOK,
@@ -213,3 +228,18 @@ func decorateWithParamName(err error, field string) error {
 	}
 	return fmt.Errorf(errTmpl, field, err)
 }
+
+func marshalResponse(resp *tripperware.PrometheusResponse, acceptHeader string) (string, []byte, error) {
+	for _, clause := range goautoneg.ParseAccept(acceptHeader) {
+		if jsonMIMEType.Satisfies(clause) {
+			b, err := json.Marshal(resp)
+			return tripperware.ApplicationJson, b, err
+		} else if rulerMIMEType.Satisfies(clause) {
+			b, err := resp.Marshal()
+			return tripperware.QueryResponseCortexMIMEType, b, err
+		}
+	}
+
+	b, err := json.Marshal(resp)
+	return tripperware.ApplicationJson, b, err
+}