docs: add note about testing new releases on testing env

Signed-off-by: hagen1778 <roman@victoriametrics.com>

.gitignore

@@ -7,6 +7,7 @@
 .vscode
 *.test
 *.swp
+/vmdocs
 /gocache-for-docker
 /victoria-logs-data
 /victoria-metrics-data
@@ -25,3 +26,4 @@ _site
 /docs/.jekyll-metadata
 coverage.txt
 cspell.json
+*~

.wwhrd.yml

@@ -4,3 +4,4 @@ allowlist:
   - BSD-3-Clause
   - BSD-2-Clause
   - ISC
+  - MPL-2.0

Makefile

@@ -13,6 +13,7 @@ PKG_TAG := $(BUILDINFO_TAG)
 endif
 
 GO_BUILDINFO = -X '$(PKG_PREFIX)/lib/buildinfo.Version=$(APP_NAME)-$(DATEINFO_TAG)-$(BUILDINFO_TAG)'
+TAR_OWNERSHIP ?= --owner=1000 --group=1000
 
 .PHONY: $(MAKECMDGOALS)
 
@@ -245,7 +246,7 @@ release-victoria-metrics-windows-amd64:
 
 release-victoria-metrics-goos-goarch: victoria-metrics-$(GOOS)-$(GOARCH)-prod
 	cd bin && \
-		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
+		tar $(TAR_OWNERSHIP) --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			victoria-metrics-$(GOOS)-$(GOARCH)-prod \
 		&& sha256sum victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			victoria-metrics-$(GOOS)-$(GOARCH)-prod \
@@ -302,7 +303,7 @@ release-victoria-logs-windows-amd64:
 
 release-victoria-logs-goos-goarch: victoria-logs-$(GOOS)-$(GOARCH)-prod
 	cd bin && \
-		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-logs-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
+		tar $(TAR_OWNERSHIP) --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-logs-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			victoria-logs-$(GOOS)-$(GOARCH)-prod \
 		&& sha256sum victoria-logs-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			victoria-logs-$(GOOS)-$(GOARCH)-prod \
@@ -332,7 +333,7 @@ release-vmutils: \
 
 release-vmutils-linux-386:
 	GOOS=linux GOARCH=386 $(MAKE) release-vmutils-goos-goarch
-	
+
 release-vmutils-linux-amd64:
 	GOOS=linux GOARCH=amd64 $(MAKE) release-vmutils-goos-goarch
 
@@ -366,7 +367,7 @@ release-vmutils-goos-goarch: \
 	vmrestore-$(GOOS)-$(GOARCH)-prod \
 	vmctl-$(GOOS)-$(GOARCH)-prod
 	cd bin && \
-		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf vmutils-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
+		tar $(TAR_OWNERSHIP) --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf vmutils-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			vmagent-$(GOOS)-$(GOARCH)-prod \
 			vmalert-$(GOOS)-$(GOARCH)-prod \
 			vmalert-tool-$(GOOS)-$(GOARCH)-prod \
@@ -443,19 +444,19 @@ check-all: fmt vet golangci-lint govulncheck
 clean-checkers: remove-golangci-lint remove-govulncheck
 
 test:
-	go test ./lib/... ./app/...
+	DISABLE_FSYNC_FOR_TESTING=1 go test ./lib/... ./app/...
 
 test-race:
-	go test -race ./lib/... ./app/...
+	DISABLE_FSYNC_FOR_TESTING=1 go test -race ./lib/... ./app/...
 
 test-pure:
-	CGO_ENABLED=0 go test ./lib/... ./app/...
+	DISABLE_FSYNC_FOR_TESTING=1 CGO_ENABLED=0 go test ./lib/... ./app/...
 
 test-full:
-	go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
+	DISABLE_FSYNC_FOR_TESTING=1 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
 
 test-full-386:
-	GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
+	DISABLE_FSYNC_FOR_TESTING=1 GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
 
 benchmark:
 	go test -bench=. ./lib/...
@@ -466,9 +467,9 @@ benchmark-pure:
 	CGO_ENABLED=0 go test -bench=. ./app/...
 
 vendor-update:
-	go get -u -d ./lib/...
-	go get -u -d ./app/...
-	go mod tidy -compat=1.22
+	go get -u ./lib/...
+	go get -u ./app/...
+	go mod tidy -compat=1.23
 	go mod vendor
 
 app-local:
@@ -494,7 +495,7 @@ golangci-lint: install-golangci-lint
 	golangci-lint run
 
 install-golangci-lint:
-	which golangci-lint || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(shell go env GOPATH)/bin v1.59.1
+	which golangci-lint || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(shell go env GOPATH)/bin v1.60.3
 
 remove-golangci-lint:
 	rm -rf `which golangci-lint`
@@ -513,34 +514,3 @@ install-wwhrd:
 
 check-licenses: install-wwhrd
 	wwhrd check -f .wwhrd.yml
-
-copy-docs:
-# The 'printf' function is used instead of 'echo' or 'echo -e' to handle line breaks (e.g. '\n') in the same way on different operating systems (MacOS/Ubuntu Linux/Arch Linux) and their shells (bash/sh/zsh/fish).
-# For details, see https://github.com/VictoriaMetrics/VictoriaMetrics/pull/4548#issue-1782796419 and https://stackoverflow.com/questions/8467424/echo-newline-in-bash-prints-literal-n
-	echo "---" > ${DST}
-	@if [ ${ORDER} -ne 0 ]; then \
-		echo "sort: ${ORDER}" >> ${DST}; \
-		echo "weight: ${ORDER}" >> ${DST}; \
-		printf "menu:\n  docs:\n    parent: 'victoriametrics'\n    weight: ${ORDER}\n" >> ${DST}; \
-	fi
-
-	echo "title: ${TITLE}" >> ${DST}
-	@if [ ${OLD_URL} ]; then \
-		printf "aliases:\n  - ${OLD_URL}\n" >> ${DST}; \
-	fi
-	echo "---" >> ${DST}
-	cat ${SRC} >> ${DST}
-	sed -i='.tmp' 's/<img src=\"docs\//<img src=\"/' ${DST}
-	sed -i='.tmp' 's/<source srcset=\"docs\//<source srcset=\"/' ${DST}
-	rm -rf docs/*.tmp
-
-# Copies docs for all components and adds the order/weight tag, title, menu position and alias with the backward compatible link for the old site.
-# For ORDER=0 it adds no order tag/weight tag.
-# FOR OLD_URL - relative link, used for backward compatibility with the link from documentation based on GitHub pages (old one)
-# FOR OLD_URL='' it adds no alias, it should be empty for every new page, don't change it for already existing links.
-# Images starting with <img src="docs/ are replaced with <img src="
-# Cluster docs are supposed to be ordered as 2nd.
-# The rest of docs is ordered manually.
-docs-sync:
-	SRC=README.md DST=docs/README.md OLD_URL='' ORDER=0 TITLE=VictoriaMetrics $(MAKE) copy-docs
-	SRC=README.md DST=docs/Single-server-VictoriaMetrics.md OLD_URL='/Single-server-VictoriaMetrics.html' TITLE=VictoriaMetrics ORDER=1 $(MAKE) copy-docs

SECURITY.md

@@ -7,12 +7,12 @@ The following versions of VictoriaMetrics receive regular security fixes:
 | Version | Supported          |
 |---------|--------------------|
 | [latest release](https://docs.victoriametrics.com/changelog/) | :white_check_mark: |
+| v1.102.x [LTS line](https://docs.victoriametrics.com/lts-releases/) | :white_check_mark: |
 | v1.97.x [LTS line](https://docs.victoriametrics.com/lts-releases/) | :white_check_mark: |
-| v1.93.x [LTS line](https://docs.victoriametrics.com/lts-releases/) | :white_check_mark: |
 | other releases  | :x:                |
 
 See [this page](https://victoriametrics.com/security/) for more details.
 
 ## Reporting a Vulnerability
 
-Please report any security issues to security@victoriametrics.com
+Please report any security issues to <security@victoriametrics.com>

app/victoria-logs/deployment/Dockerfile

@@ -1,8 +1,8 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 EXPOSE 9428
 
 ENTRYPOINT ["/victoria-logs-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./victoria-logs-prod

app/victoria-logs/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/victoria-metrics/deployment/Dockerfile

@@ -1,8 +1,8 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 EXPOSE 8428
 
 ENTRYPOINT ["/victoria-metrics-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./victoria-metrics-prod

app/victoria-metrics/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vlinsert/elasticsearch/elasticsearch.go

@@ -57,6 +57,12 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
 		fmt.Fprintf(w, `{}`)
 		return true
 	}
+	if strings.HasPrefix(path, "/logstash") || strings.HasPrefix(path, "/_logstash") {
+		// Return fake response for Logstash APIs requests.
+		// See: https://www.elastic.co/guide/en/elasticsearch/reference/8.8/logstash-apis.html
+		fmt.Fprintf(w, `{}`)
+		return true
+	}
 	switch path {
 	case "/":
 		switch r.Method {

app/vlinsert/insertutils/common_params.go

@@ -2,6 +2,7 @@ package insertutils
 
 import (
 	"net/http"
+	"strings"
 	"sync"
 	"time"
 
@@ -38,22 +39,46 @@ func GetCommonParams(r *http.Request) (*CommonParams, error) {
 		return nil, err
 	}
 
-	// Extract time field name from _time_field query arg
-	var timeField = "_time"
+	// Extract time field name from _time_field query arg or header
+	timeField := "_time"
 	if tf := r.FormValue("_time_field"); tf != "" {
 		timeField = tf
+	} else if tf = r.Header.Get("VL-Time-Field"); tf != "" {
+		timeField = tf
 	}
 
-	// Extract message field name from _msg_field query arg
-	var msgField = ""
+	// Extract message field name from _msg_field query arg or header
+	msgField := ""
 	if msgf := r.FormValue("_msg_field"); msgf != "" {
 		msgField = msgf
+	} else if msgf = r.Header.Get("VL-Msg-Field"); msgf != "" {
+		msgField = msgf
 	}
 
 	streamFields := httputils.GetArray(r, "_stream_fields")
+	if len(streamFields) == 0 {
+		if sf := r.Header.Get("VL-Stream-Fields"); len(sf) > 0 {
+			streamFields = strings.Split(sf, ",")
+		}
+	}
 	ignoreFields := httputils.GetArray(r, "ignore_fields")
+	if len(ignoreFields) == 0 {
+		if f := r.Header.Get("VL-Ignore-Fields"); len(f) > 0 {
+			ignoreFields = strings.Split(f, ",")
+		}
+	}
 
 	debug := httputils.GetBool(r, "debug")
+	if !debug {
+		if dh := r.Header.Get("VL-Debug"); len(dh) > 0 {
+			hv := strings.ToLower(dh)
+			switch hv {
+			case "", "0", "f", "false", "no":
+			default:
+				debug = true
+			}
+		}
+	}
 	debugRequestURI := ""
 	debugRemoteAddr := ""
 	if debug {
@@ -71,6 +96,7 @@ func GetCommonParams(r *http.Request) (*CommonParams, error) {
 		DebugRequestURI: debugRequestURI,
 		DebugRemoteAddr: debugRemoteAddr,
 	}
+
 	return cp, nil
 }
 
@@ -154,7 +180,7 @@ func (lmp *logMessageProcessor) AddRow(timestamp int64, fields []logstorage.Fiel
 	if lmp.cp.Debug {
 		s := lmp.lr.GetRowString(0)
 		lmp.lr.ResetKeepSettings()
-		logger.Infof("remoteAddr=%s; requestURI=%s; ignoring log entry because of `debug` query arg: %s", lmp.cp.DebugRemoteAddr, lmp.cp.DebugRequestURI, s)
+		logger.Infof("remoteAddr=%s; requestURI=%s; ignoring log entry because of `debug` arg: %s", lmp.cp.DebugRemoteAddr, lmp.cp.DebugRequestURI, s)
 		rowsDroppedTotalDebug.Inc()
 		return
 	}
@@ -196,5 +222,7 @@ func (cp *CommonParams) NewLogMessageProcessor() LogMessageProcessor {
 	return lmp
 }
 
-var rowsDroppedTotalDebug = metrics.NewCounter(`vl_rows_dropped_total{reason="debug"}`)
-var rowsDroppedTotalTooManyFields = metrics.NewCounter(`vl_rows_dropped_total{reason="too_many_fields"}`)
+var (
+	rowsDroppedTotalDebug         = metrics.NewCounter(`vl_rows_dropped_total{reason="debug"}`)
+	rowsDroppedTotalTooManyFields = metrics.NewCounter(`vl_rows_dropped_total{reason="too_many_fields"}`)
+)

app/vlinsert/main.go

@@ -7,6 +7,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/elasticsearch"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/jsonline"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/loki"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/opentelemetry"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/syslog"
 )
 
@@ -41,6 +42,9 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 	case strings.HasPrefix(path, "/loki/"):
 		path = strings.TrimPrefix(path, "/loki")
 		return loki.RequestHandler(path, w, r)
+	case strings.HasPrefix(path, "/opentelemetry/"):
+		path = strings.TrimPrefix(path, "/opentelemetry")
+		return opentelemetry.RequestHandler(path, w, r)
 	default:
 		return false
 	}

app/vlinsert/opentelemetry/opentelemetry.go

@@ -0,0 +1,143 @@
+package opentelemetry
+
+import (
+	"fmt"
+	"io"
+	"net/http"
+	"time"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/pb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/writeconcurrencylimiter"
+	"github.com/VictoriaMetrics/metrics"
+)
+
+// RequestHandler processes Opentelemetry insert requests
+func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
+	switch path {
+	// use the same path as opentelemetry collector
+	// https://opentelemetry.io/docs/specs/otlp/#otlphttp-request
+	case "/v1/logs":
+		if r.Header.Get("Content-Type") == "application/json" {
+			httpserver.Errorf(w, r, "json encoding isn't supported for opentelemetry format. Use protobuf encoding")
+			return true
+		}
+		handleProtobuf(r, w)
+		return true
+	default:
+		return false
+	}
+}
+
+func handleProtobuf(r *http.Request, w http.ResponseWriter) {
+	startTime := time.Now()
+	requestsProtobufTotal.Inc()
+	reader := r.Body
+	if r.Header.Get("Content-Encoding") == "gzip" {
+		zr, err := common.GetGzipReader(reader)
+		if err != nil {
+			httpserver.Errorf(w, r, "cannot initialize gzip reader: %s", err)
+			return
+		}
+		defer common.PutGzipReader(zr)
+		reader = zr
+	}
+
+	wcr := writeconcurrencylimiter.GetReader(reader)
+	data, err := io.ReadAll(wcr)
+	writeconcurrencylimiter.PutReader(wcr)
+	if err != nil {
+		httpserver.Errorf(w, r, "cannot read request body: %s", err)
+		return
+	}
+
+	cp, err := insertutils.GetCommonParams(r)
+	if err != nil {
+		httpserver.Errorf(w, r, "cannot parse common params from request: %s", err)
+		return
+	}
+	if err := vlstorage.CanWriteData(); err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+
+	lmp := cp.NewLogMessageProcessor()
+	n, err := pushProtobufRequest(data, lmp)
+	lmp.MustClose()
+	if err != nil {
+		httpserver.Errorf(w, r, "cannot parse OpenTelemetry protobuf request: %s", err)
+		return
+	}
+
+	rowsIngestedProtobufTotal.Add(n)
+
+	// update requestProtobufDuration only for successfully parsed requests
+	// There is no need in updating requestProtobufDuration for request errors,
+	// since their timings are usually much smaller than the timing for successful request parsing.
+	requestProtobufDuration.UpdateDuration(startTime)
+}
+
+var (
+	rowsIngestedProtobufTotal = metrics.NewCounter(`vl_rows_ingested_total{type="opentelemetry",format="protobuf"}`)
+
+	requestsProtobufTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
+	errorsTotal           = metrics.NewCounter(`vl_http_errors_total{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
+
+	requestProtobufDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
+)
+
+func pushProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int, error) {
+	var req pb.ExportLogsServiceRequest
+	if err := req.UnmarshalProtobuf(data); err != nil {
+		errorsTotal.Inc()
+		return 0, fmt.Errorf("cannot unmarshal request from %d bytes: %w", len(data), err)
+	}
+
+	var rowsIngested int
+	var commonFields []logstorage.Field
+	for _, rl := range req.ResourceLogs {
+		attributes := rl.Resource.Attributes
+		commonFields = slicesutil.SetLength(commonFields, len(attributes))
+		for i, attr := range attributes {
+			commonFields[i].Name = attr.Key
+			commonFields[i].Value = attr.Value.FormatString()
+		}
+		commonFieldsLen := len(commonFields)
+		for _, sc := range rl.ScopeLogs {
+			var scopeIngested int
+			commonFields, scopeIngested = pushFieldsFromScopeLogs(&sc, commonFields[:commonFieldsLen], lmp)
+			rowsIngested += scopeIngested
+		}
+	}
+
+	return rowsIngested, nil
+}
+
+func pushFieldsFromScopeLogs(sc *pb.ScopeLogs, commonFields []logstorage.Field, lmp insertutils.LogMessageProcessor) ([]logstorage.Field, int) {
+	fields := commonFields
+	for _, lr := range sc.LogRecords {
+		fields = fields[:len(commonFields)]
+		fields = append(fields, logstorage.Field{
+			Name:  "_msg",
+			Value: lr.Body.FormatString(),
+		})
+		for _, attr := range lr.Attributes {
+			fields = append(fields, logstorage.Field{
+				Name:  attr.Key,
+				Value: attr.Value.FormatString(),
+			})
+		}
+		fields = append(fields, logstorage.Field{
+			Name:  "severity",
+			Value: lr.FormatSeverity(),
+		})
+
+		lmp.AddRow(lr.ExtractTimestampNano(), fields)
+	}
+	return fields, len(sc.LogRecords)
+}

app/vlinsert/opentelemetry/opentelemetry_test.go

@@ -0,0 +1,128 @@
+package opentelemetry
+
+import (
+	"testing"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/pb"
+)
+
+func TestPushProtoOk(t *testing.T) {
+	f := func(src []pb.ResourceLogs, timestampsExpected []int64, resultExpected string) {
+		t.Helper()
+		lr := pb.ExportLogsServiceRequest{
+			ResourceLogs: src,
+		}
+
+		pData := lr.MarshalProtobuf(nil)
+		tlp := &insertutils.TestLogMessageProcessor{}
+		n, err := pushProtobufRequest(pData, tlp)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
+
+		if err := tlp.Verify(n, timestampsExpected, resultExpected); err != nil {
+			t.Fatal(err)
+		}
+	}
+	// single line without resource attributes
+	f([]pb.ResourceLogs{
+		{
+			ScopeLogs: []pb.ScopeLogs{
+				{
+					LogRecords: []pb.LogRecord{
+						{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
+					},
+				},
+			},
+		},
+	},
+		[]int64{1234},
+		`{"_msg":"log-line-message","severity":"Trace"}`,
+	)
+	// multi-line with resource attributes
+	f([]pb.ResourceLogs{
+		{
+			Resource: pb.Resource{
+				Attributes: []*pb.KeyValue{
+					{Key: "logger", Value: &pb.AnyValue{StringValue: ptrTo("context")}},
+					{Key: "instance_id", Value: &pb.AnyValue{IntValue: ptrTo[int64](10)}},
+					{Key: "node_taints", Value: &pb.AnyValue{KeyValueList: &pb.KeyValueList{
+						Values: []*pb.KeyValue{
+							{Key: "role", Value: &pb.AnyValue{StringValue: ptrTo("dev")}},
+							{Key: "cluster_load_percent", Value: &pb.AnyValue{DoubleValue: ptrTo(0.55)}},
+						},
+					}}},
+				},
+			},
+			ScopeLogs: []pb.ScopeLogs{
+				{
+					LogRecords: []pb.LogRecord{
+						{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
+						{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1235, SeverityNumber: 21, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
+						{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1236, SeverityNumber: -1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
+					},
+				},
+			},
+		},
+	},
+		[]int64{1234, 1235, 1236},
+		`{"logger":"context","instance_id":"10","node_taints":"[{\"Key\":\"role\",\"Value\":{\"StringValue\":\"dev\",\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":null,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}},{\"Key\":\"cluster_load_percent\",\"Value\":{\"StringValue\":null,\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":0.55,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}}]","_msg":"log-line-message","severity":"Trace"}
+{"logger":"context","instance_id":"10","node_taints":"[{\"Key\":\"role\",\"Value\":{\"StringValue\":\"dev\",\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":null,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}},{\"Key\":\"cluster_load_percent\",\"Value\":{\"StringValue\":null,\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":0.55,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}}]","_msg":"log-line-message-msg-2","severity":"Unspecified"}
+{"logger":"context","instance_id":"10","node_taints":"[{\"Key\":\"role\",\"Value\":{\"StringValue\":\"dev\",\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":null,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}},{\"Key\":\"cluster_load_percent\",\"Value\":{\"StringValue\":null,\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":0.55,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}}]","_msg":"log-line-message-msg-2","severity":"Unspecified"}`,
+	)
+
+	// multi-scope with resource attributes and multi-line
+	f([]pb.ResourceLogs{
+		{
+			Resource: pb.Resource{
+				Attributes: []*pb.KeyValue{
+					{Key: "logger", Value: &pb.AnyValue{StringValue: ptrTo("context")}},
+					{Key: "instance_id", Value: &pb.AnyValue{IntValue: ptrTo[int64](10)}},
+					{Key: "node_taints", Value: &pb.AnyValue{KeyValueList: &pb.KeyValueList{
+						Values: []*pb.KeyValue{
+							{Key: "role", Value: &pb.AnyValue{StringValue: ptrTo("dev")}},
+							{Key: "cluster_load_percent", Value: &pb.AnyValue{DoubleValue: ptrTo(0.55)}},
+						},
+					}}},
+				},
+			},
+			ScopeLogs: []pb.ScopeLogs{
+				{
+					LogRecords: []pb.LogRecord{
+						{TimeUnixNano: 1234, SeverityNumber: 1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
+						{TimeUnixNano: 1235, SeverityNumber: 5, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
+					},
+				},
+			},
+		},
+		{
+			ScopeLogs: []pb.ScopeLogs{
+				{
+					LogRecords: []pb.LogRecord{
+						{TimeUnixNano: 2345, SeverityNumber: 10, Body: pb.AnyValue{StringValue: ptrTo("log-line-resource-scope-1-0-0")}},
+						{TimeUnixNano: 2346, SeverityNumber: 10, Body: pb.AnyValue{StringValue: ptrTo("log-line-resource-scope-1-0-1")}},
+					},
+				},
+				{
+					LogRecords: []pb.LogRecord{
+						{TimeUnixNano: 2347, SeverityNumber: 12, Body: pb.AnyValue{StringValue: ptrTo("log-line-resource-scope-1-1-0")}},
+						{ObservedTimeUnixNano: 2348, SeverityNumber: 12, Body: pb.AnyValue{StringValue: ptrTo("log-line-resource-scope-1-1-1")}},
+					},
+				},
+			},
+		},
+	},
+		[]int64{1234, 1235, 2345, 2346, 2347, 2348},
+		`{"logger":"context","instance_id":"10","node_taints":"[{\"Key\":\"role\",\"Value\":{\"StringValue\":\"dev\",\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":null,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}},{\"Key\":\"cluster_load_percent\",\"Value\":{\"StringValue\":null,\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":0.55,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}}]","_msg":"log-line-message","severity":"Trace"}
+{"logger":"context","instance_id":"10","node_taints":"[{\"Key\":\"role\",\"Value\":{\"StringValue\":\"dev\",\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":null,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}},{\"Key\":\"cluster_load_percent\",\"Value\":{\"StringValue\":null,\"BoolValue\":null,\"IntValue\":null,\"DoubleValue\":0.55,\"ArrayValue\":null,\"KeyValueList\":null,\"BytesValue\":null}}]","_msg":"log-line-message-msg-2","severity":"Debug"}
+{"_msg":"log-line-resource-scope-1-0-0","severity":"Info2"}
+{"_msg":"log-line-resource-scope-1-0-1","severity":"Info2"}
+{"_msg":"log-line-resource-scope-1-1-0","severity":"Info4"}
+{"_msg":"log-line-resource-scope-1-1-1","severity":"Info4"}`,
+	)
+}
+
+func ptrTo[T any](s T) *T {
+	return &s
+}

app/vlinsert/opentelemetry/opentemetry_timing_test.go

@@ -0,0 +1,79 @@
+package opentelemetry
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/pb"
+)
+
+func BenchmarkParseProtobufRequest(b *testing.B) {
+	for _, scopes := range []int{1, 2} {
+		for _, rows := range []int{100, 1000} {
+			for _, attributes := range []int{5, 10} {
+				b.Run(fmt.Sprintf("scopes_%d/rows_%d/attributes_%d", scopes, rows, attributes), func(b *testing.B) {
+					benchmarkParseProtobufRequest(b, scopes, rows, attributes)
+				})
+			}
+		}
+	}
+}
+
+func benchmarkParseProtobufRequest(b *testing.B, streams, rows, labels int) {
+	blp := &insertutils.BenchmarkLogMessageProcessor{}
+	b.ReportAllocs()
+	b.SetBytes(int64(streams * rows))
+	b.RunParallel(func(pb *testing.PB) {
+		body := getProtobufBody(streams, rows, labels)
+		for pb.Next() {
+			_, err := pushProtobufRequest(body, blp)
+			if err != nil {
+				panic(fmt.Errorf("unexpected error: %w", err))
+			}
+		}
+	})
+}
+
+func getProtobufBody(scopesCount, rowsCount, attributesCount int) []byte {
+	msg := "12345678910"
+
+	attrValues := []*pb.AnyValue{
+		{StringValue: ptrTo("string-attribute")},
+		{IntValue: ptrTo[int64](12345)},
+		{DoubleValue: ptrTo(3.14)},
+	}
+	attrs := make([]*pb.KeyValue, attributesCount)
+	for j := 0; j < attributesCount; j++ {
+		attrs[j] = &pb.KeyValue{
+			Key:   fmt.Sprintf("key-%d", j),
+			Value: attrValues[j%3],
+		}
+	}
+	entries := make([]pb.LogRecord, rowsCount)
+	for j := 0; j < rowsCount; j++ {
+		entries[j] = pb.LogRecord{
+			TimeUnixNano: 12345678910, ObservedTimeUnixNano: 12345678910, Body: pb.AnyValue{StringValue: &msg},
+		}
+	}
+	scopes := make([]pb.ScopeLogs, scopesCount)
+
+	for j := 0; j < scopesCount; j++ {
+		scopes[j] = pb.ScopeLogs{
+			LogRecords: entries,
+		}
+	}
+
+	pr := pb.ExportLogsServiceRequest{
+		ResourceLogs: []pb.ResourceLogs{
+			{
+				Resource: pb.Resource{
+					Attributes: attrs,
+				},
+				ScopeLogs: scopes,
+			},
+		},
+	}
+
+	return pr.MarshalProtobuf(nil)
+}

app/vlogsgenerator/main.go

@@ -230,7 +230,7 @@ func generateLogsAtTimestamp(bw *bufio.Writer, workerID int, ts int64, firstStre
 	for i := 0; i < activeStreams; i++ {
 		ip := toIPv4(rand.Uint32())
 		uuid := toUUID(rand.Uint64(), rand.Uint64())
-		fmt.Fprintf(bw, `{"_time":%q,"_msg":"message for the stream %d and worker %d; ip=%s; uuid=%s; u64=%d","host":"host_%d","worker_id":"%d"`,
+		fmt.Fprintf(bw, `{"_time":"%s","_msg":"message for the stream %d and worker %d; ip=%s; uuid=%s; u64=%d","host":"host_%d","worker_id":"%d"`,
 			timeStr, streamID, workerID, ip, uuid, rand.Uint64(), streamID, workerID)
 		fmt.Fprintf(bw, `,"run_id":"%s"`, runID)
 		for j := 0; j < *constFieldsPerLog; j++ {

app/vlselect/logsql/logsql.go

@@ -5,6 +5,7 @@ import (
 	"fmt"
 	"math"
 	"net/http"
+	"slices"
 	"sort"
 	"strconv"
 	"strings"
@@ -44,6 +45,7 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
 	}
 	if step <= 0 {
 		httpserver.Errorf(w, r, "'step' must be bigger than zero")
+		return
 	}
 
 	// Obtain offset
@@ -380,6 +382,8 @@ func ProcessStreamsRequest(ctx context.Context, w http.ResponseWriter, r *http.R
 }
 
 // ProcessLiveTailRequest processes live tailing request to /select/logsq/tail
+//
+// See https://docs.victoriametrics.com/victorialogs/querying/#live-tailing
 func ProcessLiveTailRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
 	liveTailRequests.Inc()
 	defer liveTailRequests.Dec()
@@ -560,9 +564,212 @@ func (tp *tailProcessor) getTailRows() ([][]logstorage.Field, error) {
 	return tailRows, nil
 }
 
+// ProcessStatsQueryRangeRequest handles /select/logsql/stats_query_range request.
+//
+// See https://docs.victoriametrics.com/victorialogs/querying/#querying-log-range-stats
+func ProcessStatsQueryRangeRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
+	q, tenantIDs, err := parseCommonArgs(r)
+	if err != nil {
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	// Obtain step
+	stepStr := r.FormValue("step")
+	if stepStr == "" {
+		stepStr = "1d"
+	}
+	step, err := promutils.ParseDuration(stepStr)
+	if err != nil {
+		err = fmt.Errorf("cannot parse 'step' arg: %s", err)
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+	if step <= 0 {
+		err := fmt.Errorf("'step' must be bigger than zero")
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	// Obtain `by(...)` fields from the last `| stats` pipe in q.
+	// Add `_time:step` to the `by(...)` list.
+	byFields, err := q.GetStatsByFieldsAddGroupingByTime(int64(step))
+	if err != nil {
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	q.Optimize()
+
+	m := make(map[string]*statsSeries)
+	var mLock sync.Mutex
+
+	writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
+		clonedColumnNames := make([]string, len(columns))
+		for i, c := range columns {
+			clonedColumnNames[i] = strings.Clone(c.Name)
+		}
+		for i := range timestamps {
+			timestamp := q.GetTimestamp()
+			labels := make([]logstorage.Field, 0, len(byFields))
+			for j, c := range columns {
+				if c.Name == "_time" {
+					nsec, ok := logstorage.TryParseTimestampRFC3339Nano(c.Values[i])
+					if ok {
+						timestamp = nsec
+						continue
+					}
+				}
+				if slices.Contains(byFields, c.Name) {
+					labels = append(labels, logstorage.Field{
+						Name:  clonedColumnNames[j],
+						Value: strings.Clone(c.Values[i]),
+					})
+				}
+			}
+
+			var dst []byte
+			for j, c := range columns {
+				if !slices.Contains(byFields, c.Name) {
+					name := clonedColumnNames[j]
+					dst = dst[:0]
+					dst = append(dst, name...)
+					dst = logstorage.MarshalFieldsToJSON(dst, labels)
+					key := string(dst)
+					p := statsPoint{
+						Timestamp: timestamp,
+						Value:     strings.Clone(c.Values[i]),
+					}
+
+					mLock.Lock()
+					ss := m[key]
+					if ss == nil {
+						ss = &statsSeries{
+							key:    key,
+							Name:   name,
+							Labels: labels,
+						}
+						m[key] = ss
+					}
+					ss.Points = append(ss.Points, p)
+					mLock.Unlock()
+				}
+			}
+		}
+	}
+
+	if err := vlstorage.RunQuery(ctx, tenantIDs, q, writeBlock); err != nil {
+		err = fmt.Errorf("cannot execute query [%s]: %s", q, err)
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	// Sort the collected stats by time
+	rows := make([]*statsSeries, 0, len(m))
+	for _, ss := range m {
+		points := ss.Points
+		sort.Slice(points, func(i, j int) bool {
+			return points[i].Timestamp < points[j].Timestamp
+		})
+		rows = append(rows, ss)
+	}
+	sort.Slice(rows, func(i, j int) bool {
+		return rows[i].key < rows[j].key
+	})
+
+	w.Header().Set("Content-Type", "application/json")
+	WriteStatsQueryRangeResponse(w, rows)
+}
+
+type statsSeries struct {
+	key string
+
+	Name   string
+	Labels []logstorage.Field
+	Points []statsPoint
+}
+
+type statsPoint struct {
+	Timestamp int64
+	Value     string
+}
+
+// ProcessStatsQueryRequest handles /select/logsql/stats_query request.
+//
+// See https://docs.victoriametrics.com/victorialogs/querying/#querying-log-stats
+func ProcessStatsQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
+	q, tenantIDs, err := parseCommonArgs(r)
+	if err != nil {
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	// Obtain `by(...)` fields from the last `| stats` pipe in q.
+	byFields, err := q.GetStatsByFields()
+	if err != nil {
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	q.Optimize()
+
+	var rows []statsRow
+	var rowsLock sync.Mutex
+
+	timestamp := q.GetTimestamp()
+	writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
+		clonedColumnNames := make([]string, len(columns))
+		for i, c := range columns {
+			clonedColumnNames[i] = strings.Clone(c.Name)
+		}
+		for i := range timestamps {
+			labels := make([]logstorage.Field, 0, len(byFields))
+			for j, c := range columns {
+				if slices.Contains(byFields, c.Name) {
+					labels = append(labels, logstorage.Field{
+						Name:  clonedColumnNames[j],
+						Value: strings.Clone(c.Values[i]),
+					})
+				}
+			}
+
+			for j, c := range columns {
+				if !slices.Contains(byFields, c.Name) {
+					r := statsRow{
+						Name:      clonedColumnNames[j],
+						Labels:    labels,
+						Timestamp: timestamp,
+						Value:     strings.Clone(c.Values[i]),
+					}
+
+					rowsLock.Lock()
+					rows = append(rows, r)
+					rowsLock.Unlock()
+				}
+			}
+		}
+	}
+
+	if err := vlstorage.RunQuery(ctx, tenantIDs, q, writeBlock); err != nil {
+		err = fmt.Errorf("cannot execute query [%s]: %s", q, err)
+		httpserver.SendPrometheusError(w, r, err)
+		return
+	}
+
+	w.Header().Set("Content-Type", "application/json")
+	WriteStatsQueryResponse(w, rows)
+}
+
+type statsRow struct {
+	Name      string
+	Labels    []logstorage.Field
+	Timestamp int64
+	Value     string
+}
+
 // ProcessQueryRequest handles /select/logsql/query request.
 //
-// See https://docs.victoriametrics.com/victorialogs/querying/#http-api
+// See https://docs.victoriametrics.com/victorialogs/querying/#querying-logs
 func ProcessQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
 	q, tenantIDs, err := parseCommonArgs(r)
 	if err != nil {
@@ -637,6 +844,7 @@ func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID,
 	limitUpper := 2 * limit
 	q.AddPipeLimit(uint64(limitUpper))
 	q.Optimize()
+
 	rows, err := getQueryResultsWithLimit(ctx, tenantIDs, q, limitUpper)
 	if err != nil {
 		return nil, err
@@ -647,7 +855,7 @@ func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID,
 		return rows, nil
 	}
 
-	// Slow path - search for the time range containing up to limitUpper rows.
+	// Slow path - adjust time range for selecting up to limitUpper rows
 	start, end := q.GetFilterTimeRange()
 	d := end/2 - start/2
 	start += d
@@ -656,23 +864,52 @@ func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID,
 	for {
 		q = qOrig.Clone()
 		q.AddTimeFilter(start, end)
+		// q.Optimize() call is needed for converting '*' into filterNoop.
+		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6785#issuecomment-2358547733
+		q.Optimize()
 		rows, err := getQueryResultsWithLimit(ctx, tenantIDs, q, limitUpper)
 		if err != nil {
 			return nil, err
 		}
 
-		if len(rows) >= limit && len(rows) < limitUpper || d == 0 {
+		if d == 0 || start >= end {
+			// The [start ... end] time range equals one nanosecond.
+			// Just return up to limit rows.
+			if len(rows) > limit {
+				rows = rows[:limit]
+			}
+			return rows, nil
+		}
+
+		dLastBit := d & 1
+		d /= 2
+
+		if len(rows) >= limitUpper {
+			// The number of found rows on the [start ... end] time range exceeds limitUpper,
+			// so reduce the time range to [start+d ... end].
+			start += d
+			continue
+		}
+		if len(rows) >= limit {
+			// The number of found rows is in the range [limit ... limitUpper).
+			// This means that found rows contains the needed limit rows with the biggest timestamps.
 			rows = getLastNRows(rows, limit)
 			return rows, nil
 		}
 
-		lastBit := d & 1
-		d /= 2
-		if len(rows) > limit {
-			start += d
-		} else {
-			start -= d + lastBit
+		// The number of found rows on [start ... end] time range is below the limit.
+		// This means the time range doesn't cover the needed logs, so it must be extended.
+
+		if len(rows) == 0 {
+			// The [start ... end] time range doesn't contain any rows, so change it to [start-d ... start).
+			end = start - 1
+			start -= d + dLastBit
+			continue
 		}
+
+		// The number of found rows on [start ... end] time range is bigger than 0 but smaller than limit.
+		// Increase the time range to [start-d ... end].
+		start -= d + dLastBit
 	}
 }
 
@@ -693,20 +930,25 @@ func getQueryResultsWithLimit(ctx context.Context, tenantIDs []logstorage.Tenant
 	var rows []row
 	var rowsLock sync.Mutex
 	writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
-		rowsLock.Lock()
-		defer rowsLock.Unlock()
+		clonedColumnNames := make([]string, len(columns))
+		for i, c := range columns {
+			clonedColumnNames[i] = strings.Clone(c.Name)
+		}
 
 		for i, timestamp := range timestamps {
 			fields := make([]logstorage.Field, len(columns))
 			for j := range columns {
 				f := &fields[j]
-				f.Name = strings.Clone(columns[j].Name)
+				f.Name = clonedColumnNames[j]
 				f.Value = strings.Clone(columns[j].Values[i])
 			}
+
+			rowsLock.Lock()
 			rows = append(rows, row{
 				timestamp: timestamp,
 				fields:    fields,
 			})
+			rowsLock.Unlock()
 		}
 
 		if len(rows) >= limit {
@@ -728,9 +970,23 @@ func parseCommonArgs(r *http.Request) (*logstorage.Query, []logstorage.TenantID,
 	}
 	tenantIDs := []logstorage.TenantID{tenantID}
 
+	// Parse optional time arg
+	timestamp, okTime, err := getTimeNsec(r, "time")
+	if err != nil {
+		return nil, nil, err
+	}
+	if !okTime {
+		// If time arg is missing, then evaluate query at the current timestamp
+		timestamp = time.Now().UnixNano()
+	}
+
+	// decrease timestamp by one nanosecond in order to avoid capturing logs belonging
+	// to the first nanosecond at the next period of time (month, week, day, hour, etc.)
+	timestamp--
+
 	// Parse query
 	qStr := r.FormValue("query")
-	q, err := logstorage.ParseQuery(qStr)
+	q, err := logstorage.ParseQueryAtTimestamp(qStr, timestamp)
 	if err != nil {
 		return nil, nil, fmt.Errorf("cannot parse query [%s]: %s", qStr, err)
 	}

app/vlselect/logsql/stats_query_range_response.qtpl

@@ -0,0 +1,52 @@
+{% stripspace %}
+
+// StatsQueryRangeResponse generates response for /select/logsql/stats_query_range
+{% func StatsQueryRangeResponse(rows []*statsSeries) %}
+{
+	"status":"success",
+	"data":{
+		"resultType":"matrix",
+		"result":[
+			{% if len(rows) > 0 %}
+				{%= formatStatsSeries(rows[0]) %}
+				{% code rows = rows[1:] %}
+				{% for i := range rows %}
+					,{%= formatStatsSeries(rows[i]) %}
+				{% endfor %}
+			{% endif %}
+		]
+	}
+}
+{% endfunc %}
+
+{% func formatStatsSeries(ss *statsSeries) %}
+{
+	"metric":{
+		"__name__":{%q= ss.Name %}
+		{% if len(ss.Labels) > 0 %}
+			{% for _, label := range ss.Labels %}
+				,{%q= label.Name %}:{%q= label.Value %}
+			{% endfor %}
+		{% endif %}
+	},
+	"values":[
+		{% code points := ss.Points %}
+		{% if len(points) > 0 %}
+			{%= formatStatsPoint(&points[0]) %}
+			{% code points = points[1:] %}
+			{% for i := range points %}
+				,{%= formatStatsPoint(&points[i]) %}
+			{% endfor %}
+		{% endif %}
+	]
+}
+{% endfunc %}
+
+{% func formatStatsPoint(p *statsPoint) %}
+[
+	{%f= float64(p.Timestamp)/1e9 %},
+	{%q= p.Value %}
+]
+{% endfunc %}
+
+{% endstripspace %}

app/vlselect/logsql/stats_query_range_response.qtpl.go

@@ -0,0 +1,188 @@
+// Code generated by qtc from "stats_query_range_response.qtpl". DO NOT EDIT.
+// See https://github.com/valyala/quicktemplate for details.
+
+// StatsQueryRangeResponse generates response for /select/logsql/stats_query_range
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:4
+package logsql
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:4
+import (
+	qtio422016 "io"
+
+	qt422016 "github.com/valyala/quicktemplate"
+)
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:4
+var (
+	_ = qtio422016.Copy
+	_ = qt422016.AcquireByteBuffer
+)
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:4
+func StreamStatsQueryRangeResponse(qw422016 *qt422016.Writer, rows []*statsSeries) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:4
+	qw422016.N().S(`{"status":"success","data":{"resultType":"matrix","result":[`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:10
+	if len(rows) > 0 {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:11
+		streamformatStatsSeries(qw422016, rows[0])
+//line app/vlselect/logsql/stats_query_range_response.qtpl:12
+		rows = rows[1:]
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:13
+		for i := range rows {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:13
+			qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:14
+			streamformatStatsSeries(qw422016, rows[i])
+//line app/vlselect/logsql/stats_query_range_response.qtpl:15
+		}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:16
+	}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:16
+	qw422016.N().S(`]}}`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+func WriteStatsQueryRangeResponse(qq422016 qtio422016.Writer, rows []*statsSeries) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	qw422016 := qt422016.AcquireWriter(qq422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	StreamStatsQueryRangeResponse(qw422016, rows)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	qt422016.ReleaseWriter(qw422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+func StatsQueryRangeResponse(rows []*statsSeries) string {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	qb422016 := qt422016.AcquireByteBuffer()
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	WriteStatsQueryRangeResponse(qb422016, rows)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	qs422016 := string(qb422016.B)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	qt422016.ReleaseByteBuffer(qb422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+	return qs422016
+//line app/vlselect/logsql/stats_query_range_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:22
+func streamformatStatsSeries(qw422016 *qt422016.Writer, ss *statsSeries) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:22
+	qw422016.N().S(`{"metric":{"__name__":`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:25
+	qw422016.N().Q(ss.Name)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:26
+	if len(ss.Labels) > 0 {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:27
+		for _, label := range ss.Labels {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:27
+			qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:28
+			qw422016.N().Q(label.Name)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:28
+			qw422016.N().S(`:`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:28
+			qw422016.N().Q(label.Value)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:29
+		}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:30
+	}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:30
+	qw422016.N().S(`},"values":[`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:33
+	points := ss.Points
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:34
+	if len(points) > 0 {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:35
+		streamformatStatsPoint(qw422016, &points[0])
+//line app/vlselect/logsql/stats_query_range_response.qtpl:36
+		points = points[1:]
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:37
+		for i := range points {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:37
+			qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:38
+			streamformatStatsPoint(qw422016, &points[i])
+//line app/vlselect/logsql/stats_query_range_response.qtpl:39
+		}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:40
+	}
+//line app/vlselect/logsql/stats_query_range_response.qtpl:40
+	qw422016.N().S(`]}`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+func writeformatStatsSeries(qq422016 qtio422016.Writer, ss *statsSeries) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	qw422016 := qt422016.AcquireWriter(qq422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	streamformatStatsSeries(qw422016, ss)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	qt422016.ReleaseWriter(qw422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+func formatStatsSeries(ss *statsSeries) string {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	qb422016 := qt422016.AcquireByteBuffer()
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	writeformatStatsSeries(qb422016, ss)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	qs422016 := string(qb422016.B)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	qt422016.ReleaseByteBuffer(qb422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+	return qs422016
+//line app/vlselect/logsql/stats_query_range_response.qtpl:43
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:45
+func streamformatStatsPoint(qw422016 *qt422016.Writer, p *statsPoint) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:45
+	qw422016.N().S(`[`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:47
+	qw422016.N().F(float64(p.Timestamp) / 1e9)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:47
+	qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:48
+	qw422016.N().Q(p.Value)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:48
+	qw422016.N().S(`]`)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+func writeformatStatsPoint(qq422016 qtio422016.Writer, p *statsPoint) {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	qw422016 := qt422016.AcquireWriter(qq422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	streamformatStatsPoint(qw422016, p)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	qt422016.ReleaseWriter(qw422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+}
+
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+func formatStatsPoint(p *statsPoint) string {
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	qb422016 := qt422016.AcquireByteBuffer()
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	writeformatStatsPoint(qb422016, p)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	qs422016 := string(qb422016.B)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	qt422016.ReleaseByteBuffer(qb422016)
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+	return qs422016
+//line app/vlselect/logsql/stats_query_range_response.qtpl:50
+}

app/vlselect/logsql/stats_query_response.qtpl

@@ -0,0 +1,36 @@
+{% stripspace %}
+
+// StatsQueryResponse generates response for /select/logsql/stats_query
+{% func StatsQueryResponse(rows []statsRow) %}
+{
+	"status":"success",
+	"data":{
+		"resultType":"vector",
+		"result":[
+			{% if len(rows) > 0 %}
+				{%= formatStatsRow(&rows[0]) %}
+				{% code rows = rows[1:] %}
+				{% for i := range rows %}
+					,{%= formatStatsRow(&rows[i]) %}
+				{% endfor %}
+			{% endif %}
+		]
+	}
+}
+{% endfunc %}
+
+{% func formatStatsRow(r *statsRow) %}
+{
+	"metric":{
+		"__name__":{%q= r.Name %}
+		{% if len(r.Labels) > 0 %}
+			{% for _, label := range r.Labels %}
+				,{%q= label.Name %}:{%q= label.Value %}
+			{% endfor %}
+		{% endif %}
+	},
+	"value":[{%f= float64(r.Timestamp)/1e9 %},{%q= r.Value %}]
+}
+{% endfunc %}
+
+{% endstripspace %}

app/vlselect/logsql/stats_query_response.qtpl.go

@@ -0,0 +1,133 @@
+// Code generated by qtc from "stats_query_response.qtpl". DO NOT EDIT.
+// See https://github.com/valyala/quicktemplate for details.
+
+// StatsQueryResponse generates response for /select/logsql/stats_query
+
+//line app/vlselect/logsql/stats_query_response.qtpl:4
+package logsql
+
+//line app/vlselect/logsql/stats_query_response.qtpl:4
+import (
+	qtio422016 "io"
+
+	qt422016 "github.com/valyala/quicktemplate"
+)
+
+//line app/vlselect/logsql/stats_query_response.qtpl:4
+var (
+	_ = qtio422016.Copy
+	_ = qt422016.AcquireByteBuffer
+)
+
+//line app/vlselect/logsql/stats_query_response.qtpl:4
+func StreamStatsQueryResponse(qw422016 *qt422016.Writer, rows []statsRow) {
+//line app/vlselect/logsql/stats_query_response.qtpl:4
+	qw422016.N().S(`{"status":"success","data":{"resultType":"vector","result":[`)
+//line app/vlselect/logsql/stats_query_response.qtpl:10
+	if len(rows) > 0 {
+//line app/vlselect/logsql/stats_query_response.qtpl:11
+		streamformatStatsRow(qw422016, &rows[0])
+//line app/vlselect/logsql/stats_query_response.qtpl:12
+		rows = rows[1:]
+
+//line app/vlselect/logsql/stats_query_response.qtpl:13
+		for i := range rows {
+//line app/vlselect/logsql/stats_query_response.qtpl:13
+			qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_response.qtpl:14
+			streamformatStatsRow(qw422016, &rows[i])
+//line app/vlselect/logsql/stats_query_response.qtpl:15
+		}
+//line app/vlselect/logsql/stats_query_response.qtpl:16
+	}
+//line app/vlselect/logsql/stats_query_response.qtpl:16
+	qw422016.N().S(`]}}`)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+func WriteStatsQueryResponse(qq422016 qtio422016.Writer, rows []statsRow) {
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	qw422016 := qt422016.AcquireWriter(qq422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	StreamStatsQueryResponse(qw422016, rows)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	qt422016.ReleaseWriter(qw422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+func StatsQueryResponse(rows []statsRow) string {
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	qb422016 := qt422016.AcquireByteBuffer()
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	WriteStatsQueryResponse(qb422016, rows)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	qs422016 := string(qb422016.B)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	qt422016.ReleaseByteBuffer(qb422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+	return qs422016
+//line app/vlselect/logsql/stats_query_response.qtpl:20
+}
+
+//line app/vlselect/logsql/stats_query_response.qtpl:22
+func streamformatStatsRow(qw422016 *qt422016.Writer, r *statsRow) {
+//line app/vlselect/logsql/stats_query_response.qtpl:22
+	qw422016.N().S(`{"metric":{"__name__":`)
+//line app/vlselect/logsql/stats_query_response.qtpl:25
+	qw422016.N().Q(r.Name)
+//line app/vlselect/logsql/stats_query_response.qtpl:26
+	if len(r.Labels) > 0 {
+//line app/vlselect/logsql/stats_query_response.qtpl:27
+		for _, label := range r.Labels {
+//line app/vlselect/logsql/stats_query_response.qtpl:27
+			qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_response.qtpl:28
+			qw422016.N().Q(label.Name)
+//line app/vlselect/logsql/stats_query_response.qtpl:28
+			qw422016.N().S(`:`)
+//line app/vlselect/logsql/stats_query_response.qtpl:28
+			qw422016.N().Q(label.Value)
+//line app/vlselect/logsql/stats_query_response.qtpl:29
+		}
+//line app/vlselect/logsql/stats_query_response.qtpl:30
+	}
+//line app/vlselect/logsql/stats_query_response.qtpl:30
+	qw422016.N().S(`},"value":[`)
+//line app/vlselect/logsql/stats_query_response.qtpl:32
+	qw422016.N().F(float64(r.Timestamp) / 1e9)
+//line app/vlselect/logsql/stats_query_response.qtpl:32
+	qw422016.N().S(`,`)
+//line app/vlselect/logsql/stats_query_response.qtpl:32
+	qw422016.N().Q(r.Value)
+//line app/vlselect/logsql/stats_query_response.qtpl:32
+	qw422016.N().S(`]}`)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+}
+
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+func writeformatStatsRow(qq422016 qtio422016.Writer, r *statsRow) {
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	qw422016 := qt422016.AcquireWriter(qq422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	streamformatStatsRow(qw422016, r)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	qt422016.ReleaseWriter(qw422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+}
+
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+func formatStatsRow(r *statsRow) string {
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	qb422016 := qt422016.AcquireByteBuffer()
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	writeformatStatsRow(qb422016, r)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	qs422016 := string(qb422016.B)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	qt422016.ReleaseByteBuffer(qb422016)
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+	return qs422016
+//line app/vlselect/logsql/stats_query_response.qtpl:34
+}

app/vlselect/main.go

@@ -193,6 +193,14 @@ func processSelectRequest(ctx context.Context, w http.ResponseWriter, r *http.Re
 		logsqlQueryRequests.Inc()
 		logsql.ProcessQueryRequest(ctx, w, r)
 		return true
+	case "/select/logsql/stats_query":
+		logsqlStatsQueryRequests.Inc()
+		logsql.ProcessStatsQueryRequest(ctx, w, r)
+		return true
+	case "/select/logsql/stats_query_range":
+		logsqlStatsQueryRangeRequests.Inc()
+		logsql.ProcessStatsQueryRangeRequest(ctx, w, r)
+		return true
 	case "/select/logsql/stream_field_names":
 		logsqlStreamFieldNamesRequests.Inc()
 		logsql.ProcessStreamFieldNamesRequest(ctx, w, r)
@@ -232,6 +240,8 @@ var (
 	logsqlFieldValuesRequests       = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/field_values"}`)
 	logsqlHitsRequests              = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/hits"}`)
 	logsqlQueryRequests             = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/query"}`)
+	logsqlStatsQueryRequests        = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/stats_query"}`)
+	logsqlStatsQueryRangeRequests   = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/stats_query_range"}`)
 	logsqlStreamFieldNamesRequests  = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/stream_field_names"}`)
 	logsqlStreamFieldValuesRequests = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/stream_field_values"}`)
 	logsqlStreamIDsRequests         = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/stream_ids"}`)

app/vlselect/vmui/asset-manifest.json

@@ -1,13 +1,13 @@
 {
   "files": {
-    "main.css": "./static/css/main.1041c3d4.css",
-    "main.js": "./static/js/main.62a82f4a.js",
-    "static/js/685.bebe1265.chunk.js": "./static/js/685.bebe1265.chunk.js",
-    "static/media/MetricsQL.md": "./static/media/MetricsQL.aaabf95f2c9bf356bde4.md",
+    "main.css": "./static/css/main.c9cc37dd.css",
+    "main.js": "./static/js/main.867f457f.js",
+    "static/js/685.f772060c.chunk.js": "./static/js/685.f772060c.chunk.js",
+    "static/media/MetricsQL.md": "./static/media/MetricsQL.a00044c91d9781cf8557.md",
     "index.html": "./index.html"
   },
   "entrypoints": [
-    "static/css/main.1041c3d4.css",
-    "static/js/main.62a82f4a.js"
+    "static/css/main.c9cc37dd.css",
+    "static/js/main.867f457f.js"
   ]
 }

app/vlselect/vmui/favicon.svg

@@ -0,0 +1 @@
+<svg width="48" height="48" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M24.5475 0C10.3246.0265251 1.11379 3.06365 4.40623 6.10077c0 0 12.32997 11.23333 16.58217 14.84083.8131.6896 2.1728 1.1936 3.5191 1.2201h.1199c1.3463-.0265 2.706-.5305 3.5191-1.2201 4.2522-3.5942 16.5422-14.84083 16.5422-14.84083C48.0478 3.06365 38.8636.0265251 24.6674 0" fill="#020202"/><path d="M28.1579 27.0159c-.8131.6896-2.1728 1.1936-3.5191 1.2201h-.12c-1.3463-.0265-2.7059-.5305-3.519-1.2201-2.9725-2.5067-13.35639-11.87-17.26201-15.3979v5.4112c0 .5968.22661 1.3793.6265 1.7506C7.00358 21.1936 17.2675 30.5437 20.9731 33.6737c.8132.6896 2.1728 1.1936 3.5191 1.2201h.12c1.3463-.0265 2.7059-.5305 3.519-1.2201 3.679-3.13 13.9429-12.4536 16.6089-14.8939.4132-.3713.6265-1.1538.6265-1.7506V11.618c-3.9323 3.5411-14.3162 12.931-17.2354 15.3979h.0267Z" fill="#020202"/><path d="M28.1579 39.748c-.8131.6897-2.1728 1.1937-3.5191 1.2202h-.12c-1.3463-.0265-2.7059-.5305-3.519-1.2202-2.9725-2.4933-13.35639-11.8567-17.26201-15.3978v5.4111c0 .5969.22661 1.3793.6265 1.7507C7.00358 33.9258 17.2675 43.2759 20.9731 46.4058c.8132.6897 2.1728 1.1937 3.5191 1.2202h.12c1.3463-.0265 2.7059-.5305 3.519-1.2202 3.679-3.1299 13.9429-12.4535 16.6089-14.8938.4132-.3714.6265-1.1538.6265-1.7507v-5.4111c-3.9323 3.5411-14.3162 12.931-17.2354 15.3978h.0267Z" fill="#020202"/></svg>

app/vlselect/vmui/index.html

@@ -1 +1 @@
-<!doctype html><html lang="en"><head><meta charset="utf-8"/><link rel="icon" href="./favicon.ico"/><meta name="viewport" content="width=device-width,initial-scale=1,maximum-scale=5"/><meta name="theme-color" content="#000000"/><meta name="description" content="UI for VictoriaMetrics"/><link rel="apple-touch-icon" href="./apple-touch-icon.png"/><link rel="icon" type="image/png" sizes="32x32" href="./favicon-32x32.png"><link rel="manifest" href="./manifest.json"/><title>VM UI</title><script src="./dashboards/index.js" type="module"></script><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="./preview.jpg"><meta name="twitter:title" content="UI for VictoriaMetrics"><meta name="twitter:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta name="twitter:site" content="@VictoriaMetrics"><meta property="og:title" content="Metric explorer for VictoriaMetrics"><meta property="og:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta property="og:image" content="./preview.jpg"><meta property="og:type" content="website"><script defer="defer" src="./static/js/main.62a82f4a.js"></script><link href="./static/css/main.1041c3d4.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>
+<!doctype html><html lang="en"><head><meta charset="utf-8"/><link rel="icon" href="./favicon.svg"/><link rel="apple-touch-icon" href="./favicon.svg"/><link rel="mask-icon" href="./favicon.svg" color="#000000"><meta name="viewport" content="width=device-width,initial-scale=1,maximum-scale=5"/><meta name="theme-color" content="#000000"/><meta name="description" content="Explore your log data with VictoriaLogs UI"/><link rel="manifest" href="./manifest.json"/><title>UI for VictoriaLogs</title><meta name="twitter:card" content="summary"><meta name="twitter:title" content="UI for VictoriaLogs"><meta name="twitter:site" content="@https://victoriametrics.com/products/victorialogs/"><meta name="twitter:description" content="Explore your log data with VictoriaLogs UI"><meta name="twitter:image" content="./preview.jpg"><meta property="og:type" content="website"><meta property="og:title" content="UI for VictoriaLogs"><meta property="og:url" content="https://victoriametrics.com/products/victorialogs/"><meta property="og:description" content="Explore your log data with VictoriaLogs UI"><script defer="defer" src="./static/js/main.867f457f.js"></script><link href="./static/css/main.c9cc37dd.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>

app/vlselect/vmui/manifest.json

@@ -1,16 +1,11 @@
 {
-  "short_name": "Victoria Metrics UI",
-  "name": "Victoria Metrics UI is a metric explorer for Victoria Metrics",
+  "short_name": "vmui",
+  "name": "vmui",
   "icons": [
     {
-      "src": "favicon-32x32.png",
-      "sizes": "32x32",
-      "type": "image/png"
-    },
-    {
-      "src": "apple-touch-icon.png",
-      "type": "image/png",
-      "sizes": "192x192"
+      "src": "favicon.svg",
+      "sizes": "any",
+      "type": "image/svg+xml"
     }
   ],
   "start_url": ".",

app/vlselect/vmui/static/js/main.867f457f.js.LICENSE.txt

@@ -5,7 +5,7 @@
 */
 
 /**
- * @remix-run/router v1.17.0
+ * @remix-run/router v1.19.2
  *
  * Copyright (c) Remix Software Inc.
  *
@@ -16,7 +16,7 @@
  */
 
 /**
- * React Router DOM v6.24.0
+ * React Router DOM v6.26.2
  *
  * Copyright (c) Remix Software Inc.
  *
@@ -27,7 +27,7 @@
  */
 
 /**
- * React Router v6.24.0
+ * React Router v6.26.2
  *
  * Copyright (c) Remix Software Inc.
  *

app/vlselect/vmui/static/media/MetricsQL.a00044c91d9781cf8557.md

@@ -1,5 +1,4 @@
 ---
-sort: 23
 weight: 23
 title: MetricsQL
 menu:
@@ -10,9 +9,6 @@ aliases:
 - /ExtendedPromQL.html
 - /MetricsQL.html
 ---
-
-# MetricsQL
-
 [VictoriaMetrics](https://github.com/VictoriaMetrics/VictoriaMetrics) implements MetricsQL -
 query language inspired by [PromQL](https://prometheus.io/docs/prometheus/latest/querying/basics/).
 MetricsQL is backwards-compatible with PromQL, so Grafana dashboards backed by Prometheus datasource should work
@@ -107,7 +103,7 @@ The list of MetricsQL features on top of PromQL:
 * Trailing commas on all the lists are allowed - label filters, function args and with expressions.
   For instance, the following queries are valid: `m{foo="bar",}`, `f(a, b,)`, `WITH (x=y,) x`.
   This simplifies maintenance of multi-line queries.
-* Metric names and label names may contain any unicode letter. For example `температура{город="Київ"}` is a valid MetricsQL expression.
+* Metric names and label names may contain any unicode letter. For example `ტემპერატურა{πόλη="Київ"}` is a valid MetricsQL expression.
 * Metric names and labels names may contain escaped chars. For example, `foo\-bar{baz\=aa="b"}` is valid expression.
   It returns time series with name `foo-bar` containing label `baz=aa` with value `b`.
   Additionally, the following escape sequences are supported:
@@ -220,9 +216,11 @@ See also [descent_over_time](#descent_over_time).
 over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on the given lookbehind window `d` per each time series returned
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
-See also [median_over_time](#median_over_time).
+See also [median_over_time](#median_over_time), [min_over_time](#min_over_time) and [max_over_time](#max_over_time).
 
 #### changes
 
@@ -262,6 +260,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [count_over_time](#count_over_time), [share_eq_over_time](#share_eq_over_time) and [count_values_over_time](#count_values_over_time).
 
 #### count_gt_over_time
@@ -272,6 +272,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [count_over_time](#count_over_time) and [share_gt_over_time](#share_gt_over_time).
 
 #### count_le_over_time
@@ -282,6 +284,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [count_over_time](#count_over_time) and [share_le_over_time](#share_le_over_time).
 
 #### count_ne_over_time
@@ -292,7 +296,9 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
-See also [count_over_time](#count_over_time).
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
+See also [count_over_time](#count_over_time) and [count_eq_over_time](#count_eq_over_time).
 
 #### count_over_time
 
@@ -313,6 +319,8 @@ The results are calculated independently per each time series returned from the
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [count_eq_over_time](#count_eq_over_time), [count_values](#count_values) and [distinct_over_time](#distinct_over_time) and [label_match](#label_match).
 
 #### decreases_over_time
@@ -432,6 +440,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### histogram_over_time
 
 `histogram_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates
@@ -443,11 +453,15 @@ For example, the following query calculates median temperature by country over t
 
 `histogram_quantile(0.5, sum(histogram_over_time(temperature[24h])) by (vmrange,country))`.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### hoeffding_bound_lower
 
 `hoeffding_bound_lower(phi, series_selector[d])` is a [rollup function](#rollup-functions), which calculates
 lower [Hoeffding bound](https://en.wikipedia.org/wiki/Hoeffding%27s_inequality) for the given `phi` in the range `[0...1]`.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [hoeffding_bound_upper](#hoeffding_bound_upper).
 
 #### hoeffding_bound_upper
@@ -455,6 +469,8 @@ See also [hoeffding_bound_upper](#hoeffding_bound_upper).
 `hoeffding_bound_upper(phi, series_selector[d])` is a [rollup function](#rollup-functions), which calculates
 upper [Hoeffding bound](https://en.wikipedia.org/wiki/Hoeffding%27s_inequality) for the given `phi` in the range `[0...1]`.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [hoeffding_bound_lower](#hoeffding_bound_lower).
 
 #### holt_winters
@@ -462,8 +478,9 @@ See also [hoeffding_bound_lower](#hoeffding_bound_lower).
 `holt_winters(series_selector[d], sf, tf)` is a [rollup function](#rollup-functions), which calculates Holt-Winters value
 (aka [double exponential smoothing](https://en.wikipedia.org/wiki/Exponential_smoothing#Double_exponential_smoothing)) for [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 over the given lookbehind window `d` using the given smoothing factor `sf` and the given trend factor `tf`.
-Both `sf` and `tf` must be in the range `[0...1]`. It is expected that the [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering)
-returns time series of [gauge type](https://docs.victoriametrics.com/keyconcepts/#gauge).
+Both `sf` and `tf` must be in the range `[0...1]`.
+
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
 
 This function is supported by PromQL.
 
@@ -495,13 +512,14 @@ See also [deriv](#deriv).
 
 `increase(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the increase over the given lookbehind window `d`
 per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
 
 Unlike Prometheus, it takes into account the last sample before the given lookbehind window `d` when calculating the result.
 See [this article](https://medium.com/@romanhavronenko/victoriametrics-promql-compliance-d4318203f51e) for details.
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
 This function is supported by PromQL.
 
 See also [increase_pure](#increase_pure), [increase_prometheus](#increase_prometheus) and [delta](#delta).
@@ -510,12 +528,13 @@ See also [increase_pure](#increase_pure), [increase_prometheus](#increase_promet
 
 `increase_prometheus(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the increase
 over the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
 It doesn't take into account the last sample before the given lookbehind window `d` when calculating the result in the same way as Prometheus does.
 See [this article](https://medium.com/@romanhavronenko/victoriametrics-promql-compliance-d4318203f51e) for details.
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
 See also [increase_pure](#increase_pure) and [increase](#increase).
 
 #### increase_pure
@@ -524,6 +543,10 @@ See also [increase_pure](#increase_pure) and [increase](#increase).
 of the following corner case - it assumes that [counters](https://docs.victoriametrics.com/keyconcepts/#counter) always start from 0,
 while [increase](#increase) ignores the first value in a series if it is too big.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
+See also [increase](#increase) and [increase_prometheus](#increase_prometheus).
+
 #### increases_over_time
 
 `increases_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the number of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -540,15 +563,18 @@ on the given lookbehind window `d` per each time series returned from the given
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### irate
 
 `irate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the "instant" per-second increase rate over
 the last two [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
 This function is supported by PromQL.
 
 See also [rate](#rate) and [rollup_rate](#rollup_rate).
@@ -587,6 +613,8 @@ See also [duration_over_time](#duration_over_time) and [lag](#lag).
 over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on the given lookbehind window `d` per each time series returned
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [mad](#mad), [range_mad](#range_mad) and [outlier_iqr_over_time](#outlier_iqr_over_time).
 
 #### max_over_time
@@ -594,9 +622,11 @@ See also [mad](#mad), [range_mad](#range_mad) and [outlier_iqr_over_time](#outli
 `max_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the maximum value over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
-See also [tmax_over_time](#tmax_over_time).
+See also [tmax_over_time](#tmax_over_time) and [min_over_time](#min_over_time).
 
 #### median_over_time
 
@@ -604,6 +634,8 @@ See also [tmax_over_time](#tmax_over_time).
 on the given lookbehind window `d` per each time series returned
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [avg_over_time](#avg_over_time).
 
 #### min_over_time
@@ -611,9 +643,11 @@ See also [avg_over_time](#avg_over_time).
 `min_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the minimum value over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
-See also [tmin_over_time](#tmin_over_time).
+See also [tmin_over_time](#tmin_over_time) and [max_over_time](#max_over_time).
 
 #### mode_over_time
 
@@ -622,6 +656,8 @@ for [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering). It is expected that [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 values are discrete.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### outlier_iqr_over_time
 
 `outlier_iqr_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which returns the last sample on the given lookbehind window `d`
@@ -632,6 +668,8 @@ if its value is either smaller than the `q25-1.5*iqr` or bigger than `q75+1.5*iq
 The `outlier_iqr_over_time()` is useful for detecting anomalies in gauge values based on the previous history of values.
 For example, `outlier_iqr_over_time(memory_usage_bytes[1h])` triggers when `memory_usage_bytes` suddenly goes outside the usual value range for the last hour.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [outliers_iqr](#outliers_iqr).
 
 #### predict_linear
@@ -659,6 +697,8 @@ This function is supported by PromQL.
 on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 The `phi` value must be in the range `[0...1]`.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
 See also [quantiles_over_time](#quantiles_over_time).
@@ -670,6 +710,8 @@ over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 The function returns individual series per each `phi*` with `{phiLabel="phi*"}` label. `phi*` values must be in the range `[0...1]`.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [quantile_over_time](#quantile_over_time).
 
 #### range_over_time
@@ -680,11 +722,12 @@ E.g. it calculates `max_over_time(series_selector[d]) - min_over_time(series_sel
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### rate
 
 `rate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the average per-second increase rate
 over the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
 
 If the lookbehind window is skipped in square brackets, then it is automatically calculated as `max(step, scrape_interval)`, where `step` is the query arg value
 passed to [/api/v1/query_range](https://docs.victoriametrics.com/keyconcepts/#range-query) or [/api/v1/query](https://docs.victoriametrics.com/keyconcepts/#instant-query),
@@ -703,6 +746,8 @@ See also [irate](#irate) and [rollup_rate](#rollup_rate).
 on the given lookbehind window `d`. The calculations are performed individually per each time series returned
 from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
 #### resets
@@ -710,10 +755,11 @@ Metric names are stripped from the resulting rollups. Add [keep_metric_names](#k
 `resets(series_selector[d])` is a [rollup function](#rollup-functions), which returns the number
 of [counter](https://docs.victoriametrics.com/keyconcepts/#counter) resets over the given lookbehind window `d`
 per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
 This function is supported by PromQL.
 
 #### rollup
@@ -725,6 +771,10 @@ These values are calculated individually per each time series returned from the
 Optional 2nd argument `"min"`, `"max"` or `"avg"` can be passed to keep only one calculation result and without adding a label.
 See also [label_match](#label_match).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
+See also [rollup_rate](#rollup_rate).
+
 #### rollup_candlestick
 
 `rollup_candlestick(series_selector[d])` is a [rollup function](#rollup-functions), which calculates `open`, `high`, `low` and `close` values (aka OHLC)
@@ -736,6 +786,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 Optional 2nd argument `"open"`, `"high"` or `"low"` or `"close"` can be passed to keep only one calculation result and without adding a label.
 See also [label_match](#label_match).
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### rollup_delta
 
 `rollup_delta(series_selector[d])` is a [rollup function](#rollup-functions), which calculates differences between adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -762,6 +814,8 @@ See also [label_match](#label_match).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+See also [rollup](#rollup) and [rollup_rate](#rollup_rate).
+
 #### rollup_increase
 
 `rollup_increase(series_selector[d])` is a [rollup function](#rollup-functions), which calculates increases for adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -774,12 +828,17 @@ See also [label_match](#label_match).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names. See also [rollup_delta](#rollup_delta).
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
+See also [rollup](#rollup) and [rollup_rate](#rollup_rate).
+
 #### rollup_rate
 
 `rollup_rate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates per-second change rates
 for adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 on the given lookbehind window `d` and returns `min`, `max` and `avg` values for the calculated per-second change rates
 and returns them in time series with `rollup="min"`, `rollup="max"` and `rollup="avg"` additional labels.
+The calculations are performed individually per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
 See [this article](https://valyala.medium.com/why-irate-from-prometheus-doesnt-capture-spikes-45f9896d7832) in order to understand better
 when to use `rollup_rate()`.
@@ -787,10 +846,12 @@ when to use `rollup_rate()`.
 Optional 2nd argument `"min"`, `"max"` or `"avg"` can be passed to keep only one calculation result and without adding a label.
 See also [label_match](#label_match).
 
-The calculations are performed individually per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
-
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
+
+See also [rollup](#rollup) and [rollup_increase](#rollup_increase).
+
 #### rollup_scrape_interval
 
 `rollup_scrape_interval(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the interval in seconds between
@@ -824,6 +885,8 @@ This function is useful for calculating SLI and SLO. Example: `share_gt_over_tim
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [share_le_over_time](#share_le_over_time) and [count_gt_over_time](#count_gt_over_time).
 
 #### share_le_over_time
@@ -838,6 +901,8 @@ the share of time series values for the last 24 hours when memory usage was belo
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [share_gt_over_time](#share_gt_over_time) and [count_le_over_time](#count_le_over_time).
 
 #### share_eq_over_time
@@ -849,6 +914,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [count_eq_over_time](#count_eq_over_time).
 
 #### stale_samples_over_time
@@ -866,6 +933,8 @@ on the given lookbehind window `d` per each time series returned from the given
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
 See also [stdvar_over_time](#stdvar_over_time).
@@ -877,35 +946,43 @@ on the given lookbehind window `d` per each time series returned from the given
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
 See also [stddev_over_time](#stddev_over_time).
 
 #### sum_eq_over_time
 
-`sum_eq_over_time(series_selector[d], eq)` is a [rollup function](#rollup-function), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
+`sum_eq_over_time(series_selector[d], eq)` is a [rollup function](#rollup-functions), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 values equal to `eq` on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [sum_over_time](#sum_over_time) and [count_eq_over_time](#count_eq_over_time).
 
 #### sum_gt_over_time
 
-`sum_gt_over_time(series_selector[d], gt)` is a [rollup function](#rollup-function), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
+`sum_gt_over_time(series_selector[d], gt)` is a [rollup function](#rollup-functions), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 values bigger than `gt` on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [sum_over_time](#sum_over_time) and [count_gt_over_time](#count_gt_over_time).
 
 #### sum_le_over_time
 
-`sum_le_over_time(series_selector[d], le)` is a [rollup function](#rollup-function), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
+`sum_le_over_time(series_selector[d], le)` is a [rollup function](#rollup-functions), which calculates the sum of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
 values smaller or equal to `le` on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [sum_over_time](#sum_over_time) and [count_le_over_time](#count_le_over_time).
 
 #### sum_over_time
@@ -915,6 +992,8 @@ on the given lookbehind window `d` per each time series returned from the given
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 This function is supported by PromQL.
 
 #### sum2_over_time
@@ -924,6 +1003,8 @@ values on the given lookbehind window `d` per each time series returned from the
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 #### timestamp
 
 `timestamp(series_selector[d])` is a [rollup function](#rollup-functions), which returns the timestamp in seconds with millisecond precision
@@ -1001,6 +1082,8 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
 
 Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
 
+This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
+
 See also [zscore](#zscore), [range_trim_zscore](#range_trim_zscore) and [outlier_iqr_over_time](#outlier_iqr_over_time).
 
 
@@ -2230,7 +2313,8 @@ Any [rollup function](#rollup-functions) for something other than [series select
 Nested rollup functions can be implicit thanks to the [implicit query conversions](#implicit-query-conversions).
 For example, `delta(sum(m))` is implicitly converted to `delta(sum(default_rollup(m))[1i:1i])`, so it becomes a subquery,
 since it contains [default_rollup](#default_rollup) nested into [delta](#delta).
-This behavior can be disabled or logged via cmd-line flags `-search.disableImplicitConversion` and `-search.logImplicitConversion` since v1.101.0.
+This behavior can be disabled or logged via `-search.disableImplicitConversion` and `-search.logImplicitConversion` command-line flags
+starting from [`v1.101.0` release](https://docs.victoriametrics.com/changelog/).
 
 VictoriaMetrics performs subqueries in the following way:
 
@@ -2263,5 +2347,6 @@ VictoriaMetrics performs the following implicit conversions for incoming queries
   For example, `avg_over_time(rate(http_requests_total[5m])[1h])` is automatically converted to `avg_over_time(rate(http_requests_total[5m])[1h:1i])`.
 * If something other than [series selector](https://docs.victoriametrics.com/keyconcepts/#filtering)
   is passed to [rollup function](#rollup-functions), then a [subquery](#subqueries) with `1i` lookbehind window and `1i` step is automatically formed.
-  For example, `rate(sum(up))` is automatically converted to `rate((sum(default_rollup(up)))[1i:1i])`. 
-  This behavior can be disabled or logged via cmd-line flags `-search.disableImplicitConversion` and `-search.logImplicitConversion` since v1.101.0.
+  For example, `rate(sum(up))` is automatically converted to `rate((sum(default_rollup(up)))[1i:1i])`.
+  This behavior can be disabled or logged via `-search.disableImplicitConversion` and `-search.logImplicitConversion` command-line flags
+  starting from [`v1.101.0` release](https://docs.victoriametrics.com/changelog/).

app/vmagent/deployment/Dockerfile

@@ -1,8 +1,8 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 EXPOSE 8429
 
 ENTRYPOINT ["/vmagent-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmagent-prod

app/vmagent/main.go

@@ -10,6 +10,8 @@ import (
 	"strings"
 	"time"
 
+	"github.com/VictoriaMetrics/metrics"
+
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/csvimport"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogsketches"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogv1"
@@ -42,7 +44,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/firehose"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/pushmetrics"
-	"github.com/VictoriaMetrics/metrics"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/stringsutil"
 )
 
 var (
@@ -316,6 +318,10 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		influxQueryRequests.Inc()
 		influxutils.WriteDatabaseNames(w)
 		return true
+	case "/influx/health":
+		influxHealthRequests.Inc()
+		influxutils.WriteHealthCheckResponse(w)
+		return true
 	case "/opentelemetry/api/v1/push", "/opentelemetry/v1/metrics":
 		opentelemetryPushRequests.Inc()
 		if err := opentelemetry.InsertHandler(nil, r); err != nil {
@@ -450,7 +456,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.Header().Set("Content-Type", "application/json")
 		var bb bytesutil.ByteBuffer
 		promscrape.WriteConfigData(&bb)
-		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%q}}`, bb.B)
+		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%s}}`, stringsutil.JSONString(string(bb.B)))
 		return true
 	case "/prometheus/-/reload", "/-/reload":
 		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
@@ -562,6 +568,10 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 		influxQueryRequests.Inc()
 		influxutils.WriteDatabaseNames(w)
 		return true
+	case "influx/health":
+		influxHealthRequests.Inc()
+		influxutils.WriteHealthCheckResponse(w)
+		return true
 	case "opentelemetry/api/v1/push", "opentelemetry/v1/metrics":
 		opentelemetryPushRequests.Inc()
 		if err := opentelemetry.InsertHandler(at, r); err != nil {
@@ -672,7 +682,8 @@ var (
 	influxWriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/write", protocol="influx"}`)
 	influxWriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/influx/write", protocol="influx"}`)
 
-	influxQueryRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
+	influxQueryRequests  = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
+	influxHealthRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/health", protocol="influx"}`)
 
 	datadogv1WriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/series", protocol="datadog"}`)
 	datadogv1WriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/v1/series", protocol="datadog"}`)

app/vmagent/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmagent/remotewrite/client.go

@@ -34,8 +34,10 @@ var (
 	rateLimit = flagutil.NewArrayInt("remoteWrite.rateLimit", 0, "Optional rate limit in bytes per second for data sent to the corresponding -remoteWrite.url. "+
 		"By default, the rate limit is disabled. It can be useful for limiting load on remote storage when big amounts of buffered data "+
 		"is sent after temporary unavailability of the remote storage. See also -maxIngestionRate")
-	sendTimeout = flagutil.NewArrayDuration("remoteWrite.sendTimeout", time.Minute, "Timeout for sending a single block of data to the corresponding -remoteWrite.url")
-	proxyURL    = flagutil.NewArrayString("remoteWrite.proxyURL", "Optional proxy URL for writing data to the corresponding -remoteWrite.url. "+
+	sendTimeout      = flagutil.NewArrayDuration("remoteWrite.sendTimeout", time.Minute, "Timeout for sending a single block of data to the corresponding -remoteWrite.url")
+	retryMinInterval = flagutil.NewArrayDuration("remoteWrite.retryMinInterval", time.Second, "The minimum delay between retry attempts to send a block of data to the corresponding -remoteWrite.url. Every next retry attempt will double the delay to prevent hammering of remote database. See also -remoteWrite.retryMaxInterval")
+	retryMaxTime     = flagutil.NewArrayDuration("remoteWrite.retryMaxTime", time.Minute, "The max time spent on retry attempts to send a block of data to the corresponding -remoteWrite.url. Change this value if it is expected for -remoteWrite.url to be unreachable for more than -remoteWrite.retryMaxTime. See also -remoteWrite.retryMinInterval")
+	proxyURL         = flagutil.NewArrayString("remoteWrite.proxyURL", "Optional proxy URL for writing data to the corresponding -remoteWrite.url. "+
 		"Supported proxies: http, https, socks5. Example: -remoteWrite.proxyURL=socks5://proxy:1234")
 
 	tlsHandshakeTimeout   = flagutil.NewArrayDuration("remoteWrite.tlsHandshakeTimeout", 20*time.Second, "The timeout for establishing tls connections to the corresponding -remoteWrite.url")
@@ -90,6 +92,9 @@ type client struct {
 	fq *persistentqueue.FastQueue
 	hc *http.Client
 
+	retryMinInterval time.Duration
+	retryMaxTime     time.Duration
+
 	sendBlock func(block []byte) bool
 	authCfg   *promauth.Config
 	awsCfg    *awsapi.Config
@@ -143,13 +148,15 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
 		Timeout:   sendTimeout.GetOptionalArg(argIdx),
 	}
 	c := &client{
-		sanitizedURL:   sanitizedURL,
-		remoteWriteURL: remoteWriteURL,
-		authCfg:        authCfg,
-		awsCfg:         awsCfg,
-		fq:             fq,
-		hc:             hc,
-		stopCh:         make(chan struct{}),
+		sanitizedURL:     sanitizedURL,
+		remoteWriteURL:   remoteWriteURL,
+		authCfg:          authCfg,
+		awsCfg:           awsCfg,
+		fq:               fq,
+		hc:               hc,
+		retryMinInterval: retryMinInterval.GetOptionalArg(argIdx),
+		retryMaxTime:     retryMaxTime.GetOptionalArg(argIdx),
+		stopCh:           make(chan struct{}),
 	}
 	c.sendBlock = c.sendBlockHTTP
 
@@ -396,11 +403,11 @@ func (c *client) newRequest(url string, body []byte) (*http.Request, error) {
 // sendBlockHTTP sends the given block to c.remoteWriteURL.
 //
 // The function returns false only if c.stopCh is closed.
-// Otherwise it tries sending the block to remote storage indefinitely.
+// Otherwise, it tries sending the block to remote storage indefinitely.
 func (c *client) sendBlockHTTP(block []byte) bool {
 	c.rl.Register(len(block))
-	maxRetryDuration := timeutil.AddJitterToDuration(time.Minute)
-	retryDuration := timeutil.AddJitterToDuration(time.Second)
+	maxRetryDuration := timeutil.AddJitterToDuration(c.retryMaxTime)
+	retryDuration := timeutil.AddJitterToDuration(c.retryMinInterval)
 	retriesCount := 0
 
 again:

app/vmagent/remotewrite/remotewrite.go

@@ -441,7 +441,7 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
 	var rctx *relabelCtx
 	rcs := allRelabelConfigs.Load()
 	pcsGlobal := rcs.global
-	if pcsGlobal.Len() > 0 {
+	if pcsGlobal.Len() > 0 || *usePromCompatibleNaming {
 		rctx = getRelabelCtx()
 		defer putRelabelCtx(rctx)
 	}
@@ -494,7 +494,8 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
 				tssBlock = dropAggregatedSeries(tssBlock, matchIdxs.B, *streamAggrGlobalDropInput)
 			}
 			matchIdxsPool.Put(matchIdxs)
-		} else if deduplicatorGlobal != nil {
+		}
+		if deduplicatorGlobal != nil {
 			deduplicatorGlobal.Push(tssBlock)
 			tssBlock = tssBlock[:0]
 		}
@@ -922,7 +923,8 @@ func (rwctx *remoteWriteCtx) TryPush(tss []prompbmarshal.TimeSeries, forceDropSa
 			tss = dropAggregatedSeries(tss, matchIdxs.B, rwctx.streamAggrDropInput)
 		}
 		matchIdxsPool.Put(matchIdxs)
-	} else if rwctx.deduplicator != nil {
+	}
+	if rwctx.deduplicator != nil {
 		rwctx.deduplicator.Push(tss)
 		return true
 	}

app/vmagent/remotewrite/streamaggr.go

@@ -50,7 +50,7 @@ var (
 	streamAggrIgnoreOldSamples = flagutil.NewArrayBool("remoteWrite.streamAggr.ignoreOldSamples", "Whether to ignore input samples with old timestamps outside the current "+
 		"aggregation interval for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. "+
 		"See https://docs.victoriametrics.com/stream-aggregation/#ignoring-old-samples")
-	streamAggrIgnoreFirstIntervals = flag.Int("remoteWrite.streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start "+
+	streamAggrIgnoreFirstIntervals = flagutil.NewArrayInt("remoteWrite.streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start "+
 		"for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. Increase this value if "+
 		"you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving bufferred delayed data from clients pushing data into the vmagent. "+
 		"See https://docs.victoriametrics.com/stream-aggregation/#ignore-aggregation-intervals-on-start")
@@ -130,11 +130,10 @@ func initStreamAggrConfigGlobal() {
 		sasGlobal.Store(sas)
 		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, filePath)).Set(1)
 		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, filePath)).Set(fasttime.UnixTimestamp())
-	} else {
-		dedupInterval := streamAggrGlobalDedupInterval.Duration()
-		if dedupInterval > 0 {
-			deduplicatorGlobal = streamaggr.NewDeduplicator(pushToRemoteStoragesTrackDropped, dedupInterval, *streamAggrDropInputLabels, "dedup-global")
-		}
+	}
+	dedupInterval := streamAggrGlobalDedupInterval.Duration()
+	if dedupInterval > 0 {
+		deduplicatorGlobal = streamaggr.NewDeduplicator(pushToRemoteStoragesTrackDropped, dedupInterval, *streamAggrGlobalDropInputLabels, "dedup-global")
 	}
 }
 
@@ -152,12 +151,11 @@ func (rwctx *remoteWriteCtx) initStreamAggrConfig() {
 		rwctx.streamAggrDropInput = streamAggrDropInput.GetOptionalArg(idx)
 		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, filePath)).Set(1)
 		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, filePath)).Set(fasttime.UnixTimestamp())
-	} else {
-		dedupInterval := streamAggrDedupInterval.GetOptionalArg(idx)
-		if dedupInterval > 0 {
-			alias := fmt.Sprintf("dedup-%d", idx+1)
-			rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, dedupInterval, *streamAggrDropInputLabels, alias)
-		}
+	}
+	dedupInterval := streamAggrDedupInterval.GetOptionalArg(idx)
+	if dedupInterval > 0 {
+		alias := fmt.Sprintf("dedup-%d", idx+1)
+		rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, dedupInterval, *streamAggrDropInputLabels, alias)
 	}
 }
 
@@ -202,6 +200,7 @@ func newStreamAggrConfigGlobal() (*streamaggr.Aggregators, error) {
 		DropInputLabels:      *streamAggrGlobalDropInputLabels,
 		IgnoreOldSamples:     *streamAggrGlobalIgnoreOldSamples,
 		IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
+		KeepInput:            *streamAggrGlobalKeepInput,
 	}
 
 	sas, err := streamaggr.LoadFromFile(path, pushToRemoteStoragesTrackDropped, opts, "global")
@@ -229,7 +228,8 @@ func newStreamAggrConfigPerURL(idx int, pushFunc streamaggr.PushFunc) (*streamag
 		DedupInterval:        streamAggrDedupInterval.GetOptionalArg(idx),
 		DropInputLabels:      *streamAggrDropInputLabels,
 		IgnoreOldSamples:     streamAggrIgnoreOldSamples.GetOptionalArg(idx),
-		IgnoreFirstIntervals: *streamAggrIgnoreFirstIntervals,
+		IgnoreFirstIntervals: streamAggrIgnoreFirstIntervals.GetOptionalArg(idx),
+		KeepInput:            streamAggrKeepInput.GetOptionalArg(idx),
 	}
 
 	sas, err := streamaggr.LoadFromFile(path, pushFunc, opts, alias)

app/vmalert-tool/main.go

@@ -41,9 +41,19 @@ Examples:
 						Usage:    "disable adding group's Name as label to generated alerts and time series.",
 						Required: false,
 					},
+					&cli.StringSliceFlag{
+						Name:     "external.label",
+						Usage:    `Optional label in the form 'name=value' to add to all generated recording rules and alerts. Supports an array of values separated by comma or specified via multiple flags.`,
+						Required: false,
+					},
+					&cli.StringFlag{
+						Name:     "external.url",
+						Usage:    `Optional external URL to template in rule's labels or annotations.`,
+						Required: false,
+					},
 				},
 				Action: func(c *cli.Context) error {
-					if failed := unittest.UnitTest(c.StringSlice("files"), c.Bool("disableAlertgroupLabel")); failed {
+					if failed := unittest.UnitTest(c.StringSlice("files"), c.Bool("disableAlertgroupLabel"), c.StringSlice("external.label"), c.String("external.url")); failed {
 						return fmt.Errorf("unittest failed")
 					}
 					return nil

app/vmalert-tool/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmalert-tool/unittest/testdata/disable-group-label.yaml

@@ -13,7 +13,7 @@ tests:
       - expr: suquery_interval_test
         eval_time: 4m
         exp_samples:
-          - labels: '{__name__="suquery_interval_test",datacenter="dc-123", instance="localhost:9090", job="vmagent2"}'
+          - labels: '{__name__="suquery_interval_test", instance="localhost:9090", job="vmagent2"}'
             value: 1
 
     alert_rule_test:
@@ -24,20 +24,16 @@ tests:
               job: vmagent2
               severity: page
               instance: localhost:9090
-              datacenter: dc-123
             exp_annotations:
               summary: "Instance localhost:9090 down"
-              description: "localhost:9090 of job vmagent2 has been down for more than 5 minutes."
+              description: "localhost:9090 of job vmagent2 in cluster  has been down for more than 5 minutes."
+              dashboard: "/d/dashboard?orgId=1"
 
       - eval_time: 0
         alertname: AlwaysFiring
         exp_alerts:
-          - exp_labels:
-              datacenter: dc-123
+          - {}
 
       - eval_time: 0
         alertname: InstanceDown
         exp_alerts: []
-
-    external_labels:
-      datacenter: dc-123

app/vmalert-tool/unittest/testdata/rules.yaml

@@ -8,7 +8,8 @@ groups:
           severity: page
         annotations:
           summary: "Instance {{ $labels.instance }} down"
-          description: "{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 5 minutes."
+          description: "{{ $labels.instance }} of job {{ $labels.job }} in cluster {{ $externalLabels.cluster }} has been down for more than 5 minutes."
+          dashboard: '{{ $externalURL }}/d/dashboard?orgId=1'
       - alert: AlwaysFiring
         expr: 1
       - alert: SameAlertNameWithDifferentGroup

app/vmalert-tool/unittest/testdata/test1.yaml

@@ -16,7 +16,8 @@ tests:
         groupname: group1
         alertname: SameAlertNameWithDifferentGroup
         exp_alerts:
-          - {}
+          - exp_labels:
+              cluster: prod
       - eval_time: 1m
         groupname: group2
         alertname: SameAlertNameWithDifferentGroup
@@ -25,7 +26,8 @@ tests:
         groupname: group1
         alertname: SameAlertNameWithDifferentGroup
         exp_alerts:
-          - {}
+          - exp_labels:
+              cluster: prod
       - eval_time: 6m
         groupname: group1
         alertname: SameAlertNameWithDifferentGroup
@@ -61,18 +63,18 @@ tests:
         eval_time: 4m
         exp_samples:
           - value: 4
-            labels: '{__name__="t1", datacenter="dc-123"}'
+            labels: '{__name__="t1", cluster="prod"}'
       - expr: t2
         eval_time: 4m
         exp_samples:
           - value: 4
-            labels: '{__name__="t2", datacenter="dc-123"}'
+            labels: '{__name__="t2", cluster="prod"}'
       - expr: t3
         eval_time: 4m
         exp_samples:
           # t3 is 3 instead of 4 cause it's rules3 is evaluated before rules1
           - value: 3
-            labels: '{__name__="t3", datacenter="dc-123"}'
+            labels: '{__name__="t3", cluster="prod"}'
 
     alert_rule_test:
       - eval_time: 10m
@@ -83,22 +85,21 @@ tests:
               job: vmagent1
               severity: page
               instance: localhost:9090
-              datacenter: dc-123
+              cluster: prod
             exp_annotations:
               summary: "Instance localhost:9090 down"
-              description: "localhost:9090 of job vmagent1 has been down for more than 5 minutes."
+              description: "localhost:9090 of job vmagent1 in cluster prod has been down for more than 5 minutes."
+              dashboard: "http://grafana:3000/d/dashboard?orgId=1"
 
       - eval_time: 0
         groupname: group1
         alertname: AlwaysFiring
         exp_alerts:
           - exp_labels:
-              datacenter: dc-123
+              cluster: prod
 
       - eval_time: 0
         groupname: alerts
         alertname: InstanceDown
         exp_alerts: []
 
-    external_labels:
-      datacenter: dc-123

app/vmalert-tool/unittest/testdata/test2.yaml

@@ -13,7 +13,7 @@ tests:
       - expr: suquery_interval_test
         eval_time: 4m
         exp_samples:
-          - labels: '{__name__="suquery_interval_test",datacenter="dc-123", instance="localhost:9090", job="vmagent2"}'
+          - labels: '{__name__="suquery_interval_test", cluster="prod", instance="localhost:9090", job="vmagent2"}'
             value: 1
 
     alert_rule_test:
@@ -25,22 +25,21 @@ tests:
               job: vmagent2
               severity: page
               instance: localhost:9090
-              datacenter: dc-123
+              cluster: prod
             exp_annotations:
               summary: "Instance localhost:9090 down"
-              description: "localhost:9090 of job vmagent2 has been down for more than 5 minutes."
+              description: "localhost:9090 of job vmagent2 in cluster prod has been down for more than 5 minutes."
+              dashboard: "http://grafana:3000/d/dashboard?orgId=1"
 
       - eval_time: 0
         groupname: group1
         alertname: AlwaysFiring
         exp_alerts:
           - exp_labels:
-              datacenter: dc-123
+              cluster: prod
 
       - eval_time: 0
         groupname: group1
         alertname: InstanceDown
         exp_alerts: []
 
-    external_labels:
-      datacenter: dc-123

app/vmalert-tool/unittest/unittest.go

@@ -55,7 +55,7 @@ const (
 )
 
 // UnitTest runs unittest for files
-func UnitTest(files []string, disableGroupLabel bool) bool {
+func UnitTest(files []string, disableGroupLabel bool, externalLabels []string, externalURL string) bool {
 	if err := templates.Load([]string{}, true); err != nil {
 		logger.Fatalf("failed to load template: %v", err)
 	}
@@ -71,14 +71,34 @@ func UnitTest(files []string, disableGroupLabel bool) bool {
 
 	testfiles, err := config.ReadFromFS(files)
 	if err != nil {
-		fmt.Println("  FAILED")
-		fmt.Printf("\nfailed to read test files: \n%v", err)
+		logger.Fatalf("failed to load test files %q: %v", files, err)
 	}
+	if len(testfiles) == 0 {
+		fmt.Println("no test file found")
+		return false
+	}
+
+	labels := make(map[string]string)
+	for _, s := range externalLabels {
+		if len(s) == 0 {
+			continue
+		}
+		n := strings.IndexByte(s, '=')
+		if n < 0 {
+			logger.Fatalf("missing '=' in `-label`. It must contain label in the form `name=value`; got %q", s)
+		}
+		labels[s[:n]] = s[n+1:]
+	}
+	_, err = notifier.Init(nil, labels, externalURL)
+	if err != nil {
+		logger.Fatalf("failed to init notifier: %v", err)
+	}
+
 	var failed bool
 	for fileName, file := range testfiles {
-		if err := ruleUnitTest(fileName, file); err != nil {
+		if err := ruleUnitTest(fileName, file, labels); err != nil {
 			fmt.Println("  FAILED")
-			fmt.Printf("\nfailed to run unit test for file %q: \n%v", file, err)
+			fmt.Printf("\nfailed to run unit test for file %q: \n%v", fileName, err)
 			failed = true
 		} else {
 			fmt.Println("  SUCCESS")
@@ -88,7 +108,7 @@ func UnitTest(files []string, disableGroupLabel bool) bool {
 	return failed
 }
 
-func ruleUnitTest(filename string, content []byte) []error {
+func ruleUnitTest(filename string, content []byte, externalLabels map[string]string) []error {
 	fmt.Println("\nUnit Testing: ", filename)
 	var unitTestInp unitTestFile
 	if err := yaml.UnmarshalStrict(content, &unitTestInp); err != nil {
@@ -126,7 +146,7 @@ func ruleUnitTest(filename string, content []byte) []error {
 			errs = append(errs, err)
 			continue
 		}
-		testErrs := t.test(unitTestInp.EvaluationInterval.Duration(), groupOrderMap, testGroups)
+		testErrs := t.test(unitTestInp.EvaluationInterval.Duration(), groupOrderMap, testGroups, externalLabels)
 		errs = append(errs, testErrs...)
 	}
 
@@ -163,6 +183,10 @@ func verifyTestGroup(group testGroup) error {
 			return fmt.Errorf("\n%s    missing required field \"eval_time\"", testGroupName)
 		}
 	}
+	if group.ExternalLabels != nil {
+		fmt.Printf("\n%s    warning: filed `external_labels` will be deprecated soon, please use `-external.label` cmd-line flag instead. "+
+			"Check https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6735 for details.\n", testGroupName)
+	}
 	return nil
 }
 
@@ -179,6 +203,7 @@ func processFlags() {
 		{flag: "retentionPeriod", value: "100y"},
 		{flag: "datasource.url", value: testDataSourcePath},
 		{flag: "remoteWrite.url", value: testRemoteWritePath},
+		{flag: "notifier.blackhole", value: "true"},
 	} {
 		// panics if flag doesn't exist
 		if err := flag.Lookup(fv.flag).Value.Set(fv.value); err != nil {
@@ -239,7 +264,7 @@ func tearDown() {
 	fs.MustRemoveAll(storagePath)
 }
 
-func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]int, testGroups []vmalertconfig.Group) (checkErrs []error) {
+func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]int, testGroups []vmalertconfig.Group, externalLabels map[string]string) (checkErrs []error) {
 	// set up vmstorage and http server for ingest and read queries
 	setUp()
 	// tear down vmstorage and clean the data dir
@@ -288,7 +313,14 @@ func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]i
 	// create groups with given rule
 	var groups []*rule.Group
 	for _, group := range testGroups {
-		ng := rule.NewGroup(group, q, time.Minute, tg.ExternalLabels)
+		mergedExternalLabels := make(map[string]string)
+		for k, v := range tg.ExternalLabels {
+			mergedExternalLabels[k] = v
+		}
+		for k, v := range externalLabels {
+			mergedExternalLabels[k] = v
+		}
+		ng := rule.NewGroup(group, q, time.Minute, mergedExternalLabels)
 		groups = append(groups, ng)
 	}
 

app/vmalert-tool/unittest/unittest_test.go

@@ -18,7 +18,7 @@ func TestUnitTest_Failure(t *testing.T) {
 	f := func(files []string) {
 		t.Helper()
 
-		failed := UnitTest(files, false)
+		failed := UnitTest(files, false, nil, "")
 		if !failed {
 			t.Fatalf("expecting failed test")
 		}
@@ -29,18 +29,19 @@ func TestUnitTest_Failure(t *testing.T) {
 }
 
 func TestUnitTest_Success(t *testing.T) {
-	f := func(disableGroupLabel bool, files []string) {
+	f := func(disableGroupLabel bool, files []string, externalLabels []string, externalURL string) {
 		t.Helper()
 
-		failed := UnitTest(files, disableGroupLabel)
+		failed := UnitTest(files, disableGroupLabel, externalLabels, externalURL)
 		if failed {
 			t.Fatalf("unexpected failed test")
 		}
 	}
 
 	// run multi files
-	f(false, []string{"./testdata/test1.yaml", "./testdata/test2.yaml"})
+	f(false, []string{"./testdata/test1.yaml", "./testdata/test2.yaml"}, []string{"cluster=prod"}, "http://grafana:3000")
 
 	// disable group label
-	f(true, []string{"./testdata/disable-group-label.yaml"})
+	// template with null external values
+	f(true, []string{"./testdata/disable-group-label.yaml"}, nil, "")
 }

app/vmalert/datasource/init.go

@@ -17,7 +17,7 @@ import (
 
 var (
 	addr = flag.String("datasource.url", "", "Datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect endpoint. Required parameter. "+
-		"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
+		"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
 		"See also -remoteRead.disablePathAppend and -datasource.showURL")
 	appendTypePrefix  = flag.Bool("datasource.appendTypePrefix", false, "Whether to add type prefix to -datasource.url based on the query type. Set to true if sending different query types to the vmselect URL.")
 	showDatasourceURL = flag.Bool("datasource.showURL", false, "Whether to avoid stripping sensitive information such as auth headers or passwords from URLs in log messages or UI and exported metrics. "+
@@ -99,7 +99,7 @@ func Init(extraParams url.Values) (QuerierBuilder, error) {
 
 	tr, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport: %w", err)
+		return nil, fmt.Errorf("failed to create transport for -datasource.url=%q: %w", *addr, err)
 	}
 	tr.DialContext = netutil.NewStatDialFunc("vmalert_datasource")
 	tr.DisableKeepAlives = *disableKeepAlive

app/vmalert/datasource/vm_prom_api.go

@@ -57,20 +57,23 @@ var jsonParserPool fastjson.ParserPool
 //	[{"metric":{"__name__":"up","job":"prometheus"},value": [ 1435781451.781,"1"]},
 //	{"metric":{"__name__":"up","job":"node"},value": [ 1435781451.781,"0"]}]
 func (pi *promInstant) Unmarshal(b []byte) error {
+	var metrics []json.RawMessage
+	// metrics slice could be large, so parsing it with fastjson could consume a lot of memory.
+	// We parse the slice with standard lib to keep mem usage low.
+	// And each metric object will be parsed with fastjson to reduce allocations.
+	if err := json.Unmarshal(b, &metrics); err != nil {
+		return fmt.Errorf("cannot unmarshal metrics: %w", err)
+	}
+
 	p := jsonParserPool.Get()
 	defer jsonParserPool.Put(p)
 
-	v, err := p.ParseBytes(b)
-	if err != nil {
-		return err
-	}
-
-	rows, err := v.Array()
-	if err != nil {
-		return fmt.Errorf("cannot find the top-level array of result objects: %w", err)
-	}
-	pi.ms = make([]Metric, len(rows))
-	for i, row := range rows {
+	pi.ms = make([]Metric, len(metrics))
+	for i, data := range metrics {
+		row, err := p.ParseBytes(data)
+		if err != nil {
+			return fmt.Errorf("cannot parse metric object: %w", err)
+		}
 		metric := row.Get("metric")
 		if metric == nil {
 			return fmt.Errorf("can't find `metric` object in %q", row)

app/vmalert/datasource/vm_prom_api_timing_test.go

@@ -1,43 +1,24 @@
 package datasource
 
 import (
-	"bytes"
-	"io"
-	"net/http"
 	"os"
 	"testing"
 )
 
-func BenchmarkMetrics(b *testing.B) {
-	payload := []byte(`[{"metric":{"__name__":"vm_rows"},"value":[1583786142,"13763"]},{"metric":{"__name__":"vm_requests", "foo":"bar", "baz": "qux"},"value":[1583786140,"2000"]}]`)
-
-	var pi promInstant
-	if err := pi.Unmarshal(payload); err != nil {
-		b.Fatalf(err.Error())
-	}
-	b.Run("Instant", func(b *testing.B) {
-		for i := 0; i < b.N; i++ {
-			_, _ = pi.metrics()
-		}
-	})
-}
-
-func BenchmarkParsePrometheusResponse(b *testing.B) {
-	req, _ := http.NewRequest("GET", "", nil)
-	resp := &http.Response{StatusCode: http.StatusOK}
+func BenchmarkPromInstantUnmarshal(b *testing.B) {
 	data, err := os.ReadFile("testdata/instant_response.json")
 	if err != nil {
 		b.Fatalf("error while reading file: %s", err)
 	}
-	resp.Body = io.NopCloser(bytes.NewReader(data))
 
-	b.Run("Instant", func(b *testing.B) {
+	// BenchmarkParsePrometheusResponse/Instant_std+fastjson-10                    1760            668959 ns/op          280147 B/op       5781 allocs/op
+	b.Run("Instant std+fastjson", func(b *testing.B) {
 		for i := 0; i < b.N; i++ {
-			_, err := parsePrometheusResponse(req, resp)
+			var pi promInstant
+			err = pi.Unmarshal(data)
 			if err != nil {
 				b.Fatalf("unexpected parse err: %s", err)
 			}
-			resp.Body = io.NopCloser(bytes.NewReader(data))
 		}
 	})
 }

app/vmalert/deployment/Dockerfile

@@ -1,8 +1,8 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 EXPOSE 8880
 
 ENTRYPOINT ["/vmalert-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmalert-prod

app/vmalert/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmalert/notifier/alert.go

@@ -187,7 +187,7 @@ func templateAnnotation(dst io.Writer, text string, data tplData, tmpl *textTpl.
 	return nil
 }
 
-func (a Alert) toPromLabels(relabelCfg *promrelabel.ParsedConfigs) []prompbmarshal.Label {
+func (a Alert) applyRelabelingIfNeeded(relabelCfg *promrelabel.ParsedConfigs) []prompbmarshal.Label {
 	var labels []prompbmarshal.Label
 	for k, v := range a.Labels {
 		labels = append(labels, prompbmarshal.Label{

app/vmalert/notifier/alert_test.go

@@ -187,7 +187,7 @@ func TestAlert_toPromLabels(t *testing.T) {
 	fn := func(labels map[string]string, exp []prompbmarshal.Label, relabel *promrelabel.ParsedConfigs) {
 		t.Helper()
 		a := Alert{Labels: labels}
-		got := a.toPromLabels(relabel)
+		got := a.applyRelabelingIfNeeded(relabel)
 		if !reflect.DeepEqual(got, exp) {
 			t.Fatalf("expected to have: \n%v;\ngot:\n%v",
 				exp, got)

app/vmalert/notifier/alertmanager.go

@@ -76,9 +76,6 @@ func (am *AlertManager) send(ctx context.Context, alerts []Alert, headers map[st
 		return err
 	}
 	req.Header.Set("Content-Type", "application/json")
-	for key, value := range headers {
-		req.Header.Set(key, value)
-	}
 
 	if am.timeout > 0 {
 		var cancel context.CancelFunc
@@ -94,6 +91,11 @@ func (am *AlertManager) send(ctx context.Context, alerts []Alert, headers map[st
 			return err
 		}
 	}
+	// external headers have higher priority
+	for key, value := range headers {
+		req.Header.Set(key, value)
+	}
+
 	resp, err := am.client.Do(req)
 	if err != nil {
 		return err
@@ -130,7 +132,8 @@ func NewAlertManager(alertManagerURL string, fn AlertURLGenerator, authCfg proma
 	}
 	tr, err := httputils.Transport(alertManagerURL, tls.CertFile, tls.KeyFile, tls.CAFile, tls.ServerName, tls.InsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport: %w", err)
+		return nil, fmt.Errorf("failed to create transport for alertmanager URL=%q: %w", alertManagerURL, err)
+
 	}
 
 	ba := new(promauth.BasicAuthConfig)
@@ -145,7 +148,9 @@ func NewAlertManager(alertManagerURL string, fn AlertURLGenerator, authCfg proma
 	aCfg, err := utils.AuthConfig(
 		utils.WithBasicAuth(ba.Username, ba.Password.String(), ba.PasswordFile),
 		utils.WithBearer(authCfg.BearerToken.String(), authCfg.BearerTokenFile),
-		utils.WithOAuth(oauth.ClientID, oauth.ClientSecret.String(), oauth.ClientSecretFile, oauth.TokenURL, strings.Join(oauth.Scopes, ";"), oauth.EndpointParams))
+		utils.WithOAuth(oauth.ClientID, oauth.ClientSecret.String(), oauth.ClientSecretFile, oauth.TokenURL, strings.Join(oauth.Scopes, ";"), oauth.EndpointParams),
+		utils.WithHeaders(strings.Join(authCfg.Headers, "^^")),
+	)
 	if err != nil {
 		return nil, fmt.Errorf("failed to configure auth: %w", err)
 	}

app/vmalert/notifier/alertmanager_request.qtpl

@@ -8,6 +8,8 @@
 {% func amRequest(alerts []Alert, generatorURL func(Alert) string, relabelCfg *promrelabel.ParsedConfigs) %}
 [
 {% for i, alert := range alerts %}
+{% code lbls := alert.applyRelabelingIfNeeded(relabelCfg) %}
+{% if len(lbls) == 0 %} {% continue %} {% endif %}
 {
     "startsAt":{%q= alert.Start.Format(time.RFC3339Nano) %},
     "generatorURL": {%q= generatorURL(alert) %},
@@ -15,7 +17,6 @@
     "endsAt":{%q= alert.End.Format(time.RFC3339Nano) %},
     {% endif %}
     "labels": {
-        {% code lbls := alert.toPromLabels(relabelCfg) %}
         {% code ll := len(lbls) %}
         {% for idx, l := range lbls %}
             {%q= l.Name %}:{%q= l.Value %}{% if idx != ll-1 %}, {% endif %}

app/vmalert/notifier/alertmanager_request.qtpl.go

@@ -30,111 +30,117 @@ func streamamRequest(qw422016 *qt422016.Writer, alerts []Alert, generatorURL fun
 	qw422016.N().S(`[`)
 //line app/vmalert/notifier/alertmanager_request.qtpl:10
 	for i, alert := range alerts {
-//line app/vmalert/notifier/alertmanager_request.qtpl:10
-		qw422016.N().S(`{"startsAt":`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:12
-		qw422016.N().Q(alert.Start.Format(time.RFC3339Nano))
-//line app/vmalert/notifier/alertmanager_request.qtpl:12
-		qw422016.N().S(`,"generatorURL":`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:13
-		qw422016.N().Q(generatorURL(alert))
-//line app/vmalert/notifier/alertmanager_request.qtpl:13
-		qw422016.N().S(`,`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:14
-		if !alert.End.IsZero() {
-//line app/vmalert/notifier/alertmanager_request.qtpl:14
-			qw422016.N().S(`"endsAt":`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:15
-			qw422016.N().Q(alert.End.Format(time.RFC3339Nano))
-//line app/vmalert/notifier/alertmanager_request.qtpl:15
-			qw422016.N().S(`,`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:16
-		}
-//line app/vmalert/notifier/alertmanager_request.qtpl:16
-		qw422016.N().S(`"labels": {`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:18
-		lbls := alert.toPromLabels(relabelCfg)
+//line app/vmalert/notifier/alertmanager_request.qtpl:11
+		lbls := alert.applyRelabelingIfNeeded(relabelCfg)
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:19
+//line app/vmalert/notifier/alertmanager_request.qtpl:12
+		if len(lbls) == 0 {
+//line app/vmalert/notifier/alertmanager_request.qtpl:12
+			continue
+//line app/vmalert/notifier/alertmanager_request.qtpl:12
+		}
+//line app/vmalert/notifier/alertmanager_request.qtpl:12
+		qw422016.N().S(`{"startsAt":`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:14
+		qw422016.N().Q(alert.Start.Format(time.RFC3339Nano))
+//line app/vmalert/notifier/alertmanager_request.qtpl:14
+		qw422016.N().S(`,"generatorURL":`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:15
+		qw422016.N().Q(generatorURL(alert))
+//line app/vmalert/notifier/alertmanager_request.qtpl:15
+		qw422016.N().S(`,`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:16
+		if !alert.End.IsZero() {
+//line app/vmalert/notifier/alertmanager_request.qtpl:16
+			qw422016.N().S(`"endsAt":`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:17
+			qw422016.N().Q(alert.End.Format(time.RFC3339Nano))
+//line app/vmalert/notifier/alertmanager_request.qtpl:17
+			qw422016.N().S(`,`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:18
+		}
+//line app/vmalert/notifier/alertmanager_request.qtpl:18
+		qw422016.N().S(`"labels": {`)
+//line app/vmalert/notifier/alertmanager_request.qtpl:20
 		ll := len(lbls)
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:20
+//line app/vmalert/notifier/alertmanager_request.qtpl:21
 		for idx, l := range lbls {
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 			qw422016.N().Q(l.Name)
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 			qw422016.N().S(`:`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 			qw422016.N().Q(l.Value)
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 			if idx != ll-1 {
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 				qw422016.N().S(`,`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:21
+//line app/vmalert/notifier/alertmanager_request.qtpl:22
 			}
-//line app/vmalert/notifier/alertmanager_request.qtpl:22
+//line app/vmalert/notifier/alertmanager_request.qtpl:23
 		}
-//line app/vmalert/notifier/alertmanager_request.qtpl:22
+//line app/vmalert/notifier/alertmanager_request.qtpl:23
 		qw422016.N().S(`},"annotations": {`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:25
+//line app/vmalert/notifier/alertmanager_request.qtpl:26
 		c := len(alert.Annotations)
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:26
-		for k, v := range alert.Annotations {
 //line app/vmalert/notifier/alertmanager_request.qtpl:27
+		for k, v := range alert.Annotations {
+//line app/vmalert/notifier/alertmanager_request.qtpl:28
 			c = c - 1
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 			qw422016.N().Q(k)
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 			qw422016.N().S(`:`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 			qw422016.N().Q(v)
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 			if c > 0 {
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 				qw422016.N().S(`,`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:28
+//line app/vmalert/notifier/alertmanager_request.qtpl:29
 			}
-//line app/vmalert/notifier/alertmanager_request.qtpl:29
+//line app/vmalert/notifier/alertmanager_request.qtpl:30
 		}
-//line app/vmalert/notifier/alertmanager_request.qtpl:29
+//line app/vmalert/notifier/alertmanager_request.qtpl:30
 		qw422016.N().S(`}}`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:32
+//line app/vmalert/notifier/alertmanager_request.qtpl:33
 		if i != len(alerts)-1 {
-//line app/vmalert/notifier/alertmanager_request.qtpl:32
+//line app/vmalert/notifier/alertmanager_request.qtpl:33
 			qw422016.N().S(`,`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:32
+//line app/vmalert/notifier/alertmanager_request.qtpl:33
 		}
-//line app/vmalert/notifier/alertmanager_request.qtpl:33
+//line app/vmalert/notifier/alertmanager_request.qtpl:34
 	}
-//line app/vmalert/notifier/alertmanager_request.qtpl:33
+//line app/vmalert/notifier/alertmanager_request.qtpl:34
 	qw422016.N().S(`]`)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 }
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 func writeamRequest(qq422016 qtio422016.Writer, alerts []Alert, generatorURL func(Alert) string, relabelCfg *promrelabel.ParsedConfigs) {
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	qw422016 := qt422016.AcquireWriter(qq422016)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	streamamRequest(qw422016, alerts, generatorURL, relabelCfg)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	qt422016.ReleaseWriter(qw422016)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 }
 
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 func amRequest(alerts []Alert, generatorURL func(Alert) string, relabelCfg *promrelabel.ParsedConfigs) string {
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	qb422016 := qt422016.AcquireByteBuffer()
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	writeamRequest(qb422016, alerts, generatorURL, relabelCfg)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	qs422016 := string(qb422016.B)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	qt422016.ReleaseByteBuffer(qb422016)
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 	return qs422016
-//line app/vmalert/notifier/alertmanager_request.qtpl:35
+//line app/vmalert/notifier/alertmanager_request.qtpl:36
 }

app/vmalert/notifier/alertmanager_test.go

@@ -3,6 +3,7 @@ package notifier
 import (
 	"context"
 	"encoding/json"
+	"fmt"
 	"net/http"
 	"net/http/httptest"
 	"strconv"
@@ -10,6 +11,7 @@ import (
 	"time"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
 )
 
 func TestAlertManager_Addr(t *testing.T) {
@@ -48,6 +50,9 @@ func TestAlertManager_Send(t *testing.T) {
 			conn, _, _ := w.(http.Hijacker).Hijack()
 			_ = conn.Close()
 		case 1:
+			if r.Header.Get(headerKey) != headerValue {
+				t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
+			}
 			w.WriteHeader(500)
 		case 2:
 			var a []struct {
@@ -72,10 +77,34 @@ func TestAlertManager_Send(t *testing.T) {
 			if a[0].EndAt.IsZero() {
 				t.Fatalf("expected non-zero end time")
 			}
-		case 3:
-			if r.Header.Get(headerKey) != headerValue {
+			if len(a[0].Labels) != 1 {
+				t.Fatalf("expected 1 labels got %d", len(a[0].Labels))
+			}
+			if len(a[0].Annotations) != 2 {
+				t.Fatalf("expected 2 annotations got %d", len(a[0].Annotations))
+			}
+			if r.Header.Get(headerKey) != "bar" {
 				t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
 			}
+		case 3:
+			var a []struct {
+				Labels map[string]string `json:"labels"`
+			}
+			if err := json.NewDecoder(r.Body).Decode(&a); err != nil {
+				t.Fatalf("can not unmarshal data into alert %s", err)
+			}
+			if len(a) != 1 {
+				t.Fatalf("expected 1 alert in array got %d", len(a))
+			}
+			if len(a[0].Labels) != 3 {
+				t.Fatalf("expected 3 labels got %d", len(a[0].Labels))
+			}
+			if a[0].Labels["env"] != "prod" {
+				t.Fatalf("expected env label to be prod during relabeling, got %s", a[0].Labels["env"])
+			}
+			if r.Header.Get(headerKey) != "bar" {
+				t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, "bar", r.Header.Get(headerKey))
+			}
 		}
 	})
 	srv := httptest.NewServer(mux)
@@ -86,32 +115,60 @@ func TestAlertManager_Send(t *testing.T) {
 			Username: baUser,
 			Password: promauth.NewSecret(baPass),
 		},
+		Headers: []string{fmt.Sprintf("%s:%s", headerKey, headerValue)},
+	}
+	parsedConfigs, err := promrelabel.ParseRelabelConfigsData([]byte(`
+- action: drop
+  if: '{tenant="0"}'
+  regex: ".*"
+- target_label: "env"
+  replacement: "prod"
+  if: '{tenant="1"}'
+`))
+	if err != nil {
+		t.Fatalf("unexpected error when parse relabeling config: %s", err)
 	}
 	am, err := NewAlertManager(srv.URL+alertManagerPath, func(alert Alert) string {
 		return strconv.FormatUint(alert.GroupID, 10) + "/" + strconv.FormatUint(alert.ID, 10)
-	}, aCfg, nil, 0)
+	}, aCfg, parsedConfigs, 0)
 	if err != nil {
 		t.Fatalf("unexpected error: %s", err)
 	}
-	if err := am.Send(context.Background(), []Alert{{}, {}}, nil); err == nil {
+
+	if err := am.Send(context.Background(), []Alert{{Labels: map[string]string{"a": "b"}}}, nil); err == nil {
 		t.Fatalf("expected connection error got nil")
 	}
-	if err := am.Send(context.Background(), []Alert{}, nil); err == nil {
+
+	if err := am.Send(context.Background(), []Alert{{Labels: map[string]string{"a": "b"}}}, nil); err == nil {
 		t.Fatalf("expected wrong http code error got nil")
 	}
+
 	if err := am.Send(context.Background(), []Alert{{
 		GroupID:     0,
 		Name:        "alert0",
 		Start:       time.Now().UTC(),
 		End:         time.Now().UTC(),
-		Annotations: map[string]string{"a": "b", "c": "d", "e": "f"},
-	}}, nil); err != nil {
+		Labels:      map[string]string{"alertname": "alert0"},
+		Annotations: map[string]string{"a": "b", "c": "d"},
+	}}, map[string]string{headerKey: "bar"}); err != nil {
 		t.Fatalf("unexpected error %s", err)
 	}
-	if c != 2 {
-		t.Fatalf("expected 2 calls(count from zero) to server got %d", c)
-	}
-	if err := am.Send(context.Background(), nil, map[string]string{headerKey: headerValue}); err != nil {
+
+	if err := am.Send(context.Background(), []Alert{
+		// drop tenant0 alert message during relabeling
+		{
+			Name:   "alert1",
+			Labels: map[string]string{"rule": "test", "tenant": "0"},
+		},
+		{
+			Name:   "alert2",
+			Labels: map[string]string{"rule": "test", "tenant": "1"},
+		},
+	}, map[string]string{headerKey: "bar"}); err != nil {
 		t.Fatalf("unexpected error %s", err)
 	}
+
+	if c != 3 {
+		t.Fatalf("expected 3 calls(count from zero) to server got %d", c)
+	}
 }

app/vmalert/notifier/init.go

@@ -25,6 +25,9 @@ var (
 		"Enable this flag if you want vmalert to evaluate alerting rules without sending any notifications to external receivers (eg. alertmanager). "+
 		"-notifier.url, -notifier.config and -notifier.blackhole are mutually exclusive.")
 
+	headers = flagutil.NewArrayString("notifier.headers", "Optional HTTP headers to send with each request to the corresponding -notifier.url. "+
+		"For example, -remoteWrite.headers='My-Auth:foobar' would send 'My-Auth: foobar' HTTP header with every request to the corresponding -notifier.url. "+
+		"Multiple headers must be delimited by '^^': -notifier.headers='header1:value1^^header2:value2,header3:value3'")
 	basicAuthUsername     = flagutil.NewArrayString("notifier.basicAuth.username", "Optional basic auth username for -notifier.url")
 	basicAuthPassword     = flagutil.NewArrayString("notifier.basicAuth.password", "Optional basic auth password for -notifier.url")
 	basicAuthPasswordFile = flagutil.NewArrayString("notifier.basicAuth.passwordFile", "Optional path to basic auth password file for -notifier.url")
@@ -171,6 +174,7 @@ func notifiersFromFlags(gen AlertURLGenerator) ([]Notifier, error) {
 				Scopes:           strings.Split(oauth2Scopes.GetOptionalArg(i), ";"),
 				TokenURL:         oauth2TokenURL.GetOptionalArg(i),
 			},
+			Headers: []string{headers.GetOptionalArg(i)},
 		}
 
 		addr = strings.TrimSuffix(addr, "/")

app/vmalert/remoteread/init.go

@@ -17,7 +17,7 @@ var (
 	addr = flag.String("remoteRead.url", "", "Optional URL to datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect."+
 		"Remote read is used to restore alerts state."+
 		"This configuration makes sense only if `vmalert` was configured with `remoteWrite.url` before and has been successfully persisted its state. "+
-		"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
+		"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
 		"See also '-remoteRead.disablePathAppend', '-remoteRead.showURL'.")
 
 	showRemoteReadURL = flag.Bool("remoteRead.showURL", false, "Whether to show -remoteRead.url in the exported metrics. "+
@@ -68,7 +68,7 @@ func Init() (datasource.QuerierBuilder, error) {
 	}
 	tr, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport: %w", err)
+		return nil, fmt.Errorf("failed to create transport for -remoteRead.url=%q: %w", *addr, err)
 	}
 	tr.IdleConnTimeout = *idleConnectionTimeout
 	tr.DialContext = netutil.NewStatDialFunc("vmalert_remoteread")

app/vmalert/remotewrite/debug_client.go

@@ -32,7 +32,7 @@ func NewDebugClient() (*DebugClient, error) {
 
 	t, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport: %w", err)
+		return nil, fmt.Errorf("failed to create transport for -remoteWrite.url=%q: %w", *addr, err)
 	}
 	c := &DebugClient{
 		c: &http.Client{

app/vmalert/remotewrite/init.go

@@ -15,7 +15,7 @@ import (
 var (
 	addr = flag.String("remoteWrite.url", "", "Optional URL to VictoriaMetrics or vminsert where to persist alerts state "+
 		"and recording rules results in form of timeseries. "+
-		"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
+		"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
 		"For example, if -remoteWrite.url=http://127.0.0.1:8428 is specified, "+
 		"then the alerts state will be written to http://127.0.0.1:8428/api/v1/write . See also -remoteWrite.disablePathAppend, '-remoteWrite.showURL'.")
 	showRemoteWriteURL = flag.Bool("remoteWrite.showURL", false, "Whether to show -remoteWrite.url in the exported metrics. "+
@@ -72,7 +72,7 @@ func Init(ctx context.Context) (*Client, error) {
 
 	t, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport: %w", err)
+		return nil, fmt.Errorf("failed to create transport for -remoteWrite.url=%q: %w", *addr, err)
 	}
 	t.IdleConnTimeout = *idleConnectionTimeout
 	t.DialContext = netutil.NewStatDialFunc("vmalert_remotewrite")

app/vmalert/rule/alerting.go

@@ -335,6 +335,8 @@ func (ar *AlertingRule) execRange(ctx context.Context, start, end time.Time) ([]
 				// reset to Pending if there are gaps > EvalInterval between DPs
 				a.State = notifier.StatePending
 				a.ActiveAt = at
+				// re-template the annotations as active timestamp is changed
+				_, a.Annotations, _ = ar.expandTemplates(s, qFn, at)
 				a.Start = time.Time{}
 			} else if at.Sub(a.ActiveAt) >= ar.For && a.State != notifier.StateFiring {
 				a.State = notifier.StateFiring
@@ -441,9 +443,6 @@ func (ar *AlertingRule) exec(ctx context.Context, ts time.Time, limit int) ([]pr
 			}
 			a.Value = m.Values[0]
 			a.Annotations = annotations
-			if err != nil {
-				return nil, err
-			}
 			a.KeepFiringSince = time.Time{}
 			continue
 		}

app/vmalert/rule/alerting_test.go

@@ -289,7 +289,7 @@ func TestAlertingRule_Exec(t *testing.T) {
 		4: {{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}}},
 	})
 
-	f(newTestAlertingRuleWithKeepFiring("for-pending=>firing=>keepfiring=>firing", defaultStep, defaultStep), [][]datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("for-pending=>firing=>keepfiring=>firing", defaultStep, 0, defaultStep, nil), [][]datasource.Metric{
 		{metricWithLabels(t, "name", "foo")},
 		{metricWithLabels(t, "name", "foo")},
 		// empty step to keep firing
@@ -302,7 +302,7 @@ func TestAlertingRule_Exec(t *testing.T) {
 		3: {{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}}},
 	})
 
-	f(newTestAlertingRuleWithKeepFiring("for-pending=>firing=>keepfiring=>keepfiring=>inactive=>pending=>firing", defaultStep, 2*defaultStep), [][]datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("for-pending=>firing=>keepfiring=>keepfiring=>inactive=>pending=>firing", defaultStep, 0, 2*defaultStep, nil), [][]datasource.Metric{
 		{metricWithLabels(t, "name", "foo")},
 		{metricWithLabels(t, "name", "foo")},
 		// empty step to keep firing
@@ -395,7 +395,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 		{State: notifier.StateFiring, ActiveAt: time.Unix(3e3, 0)},
 	}, nil)
 
-	f(newTestAlertingRule("for-pending", time.Second), []datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("for-pending", time.Second, 0, 0, map[string]string{"activeAt": "{{ $activeAt.UnixMilli }}"}), []datasource.Metric{
 		{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 3, 5}},
 	}, []*notifier.Alert{
 		{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
@@ -406,7 +406,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 			GroupID:     fakeGroup.ID(),
 			Name:        "for-pending",
 			Labels:      map[string]string{"alertname": "for-pending"},
-			Annotations: map[string]string{},
+			Annotations: map[string]string{"activeAt": "5000"},
 			State:       notifier.StatePending,
 			ActiveAt:    time.Unix(5, 0),
 			Value:       1,
@@ -414,7 +414,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 		},
 	})
 
-	f(newTestAlertingRule("for-firing", 3*time.Second), []datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("for-firing", 3*time.Second, 0, 0, map[string]string{"activeAt": "{{ $activeAt.UnixMilli }}"}), []datasource.Metric{
 		{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 3, 5}},
 	}, []*notifier.Alert{
 		{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
@@ -425,7 +425,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 			GroupID:     fakeGroup.ID(),
 			Name:        "for-firing",
 			Labels:      map[string]string{"alertname": "for-firing"},
-			Annotations: map[string]string{},
+			Annotations: map[string]string{"activeAt": "1000"},
 			State:       notifier.StateFiring,
 			ActiveAt:    time.Unix(1, 0),
 			Start:       time.Unix(5, 0),
@@ -434,7 +434,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 		},
 	})
 
-	f(newTestAlertingRule("for-hold-pending", time.Second), []datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("for-hold-pending", time.Second, 0, 0, map[string]string{"activeAt": "{{ $activeAt.UnixMilli }}"}), []datasource.Metric{
 		{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 2, 5}},
 	}, []*notifier.Alert{
 		{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
@@ -445,7 +445,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 			GroupID:     fakeGroup.ID(),
 			Name:        "for-hold-pending",
 			Labels:      map[string]string{"alertname": "for-hold-pending"},
-			Annotations: map[string]string{},
+			Annotations: map[string]string{"activeAt": "5000"},
 			State:       notifier.StatePending,
 			ActiveAt:    time.Unix(5, 0),
 			Value:       1,
@@ -453,7 +453,7 @@ func TestAlertingRuleExecRange(t *testing.T) {
 		},
 	})
 
-	f(newTestAlertingRuleWithEvalInterval("firing=>inactive=>inactive=>firing=>firing", 0, time.Second), []datasource.Metric{
+	f(newTestAlertingRuleWithCustomFields("firing=>inactive=>inactive=>firing=>firing", 0, time.Second, 0, nil), []datasource.Metric{
 		{Values: []float64{1, 1, 1, 1}, Timestamps: []int64{1, 4, 5, 6}},
 	}, []*notifier.Alert{
 		{State: notifier.StateFiring, ActiveAt: time.Unix(1, 0)},
@@ -538,7 +538,6 @@ func TestAlertingRuleExecRange(t *testing.T) {
 				"source": "vm",
 			},
 		},
-		//
 		{
 			State: notifier.StateFiring, ActiveAt: time.Unix(1, 0),
 			Labels: map[string]string{
@@ -1036,15 +1035,13 @@ func newTestAlertingRule(name string, waitFor time.Duration) *AlertingRule {
 	return &rule
 }
 
-func newTestAlertingRuleWithEvalInterval(name string, waitFor, evalInterval time.Duration) *AlertingRule {
-	rule := newTestAlertingRule(name, waitFor)
-	rule.EvalInterval = evalInterval
-	return rule
-}
-
-func newTestAlertingRuleWithKeepFiring(name string, waitFor, keepFiringFor time.Duration) *AlertingRule {
+func newTestAlertingRuleWithCustomFields(name string, waitFor, evalInterval, keepFiringFor time.Duration, annotation map[string]string) *AlertingRule {
 	rule := newTestAlertingRule(name, waitFor)
+	if evalInterval != 0 {
+		rule.EvalInterval = evalInterval
+	}
 	rule.KeepFiringFor = keepFiringFor
+	rule.Annotations = annotation
 	return rule
 }
 

app/vmalert/rule/group_timing_test.go

@@ -11,7 +11,7 @@ import (
 func BenchmarkGetStaleSeries(b *testing.B) {
 	ts := time.Now()
 	n := 100
-	payload := make([]prompbmarshal.TimeSeries, n)
+	payload := make([]prompbmarshal.TimeSeries, 0, n)
 	for i := 0; i < n; i++ {
 		s := fmt.Sprintf("%d", i)
 		labels := toPromLabels(b,

app/vmalert/rule/utils_test.go

@@ -50,12 +50,12 @@ func TestRequestToCurl(t *testing.T) {
 	f(req, "curl -k -X POST 'https://user:xxxxx@foo.com?query=up&step=10'")
 
 	req = newReq("https://user:pass@foo.com")
-	req.Header.Set("Authorisation", "Bearer 123456")
-	f(req, "curl -k -X POST -H 'Authorisation: <secret>' 'https://user:xxxxx@foo.com'")
+	req.Header.Set("Authorization", "Bearer 123456")
+	f(req, "curl -k -X POST -H 'Authorization: <secret>' 'https://user:xxxxx@foo.com'")
 
 	req = newReq("https://user:pass@foo.com")
-	req.Header.Set("Authorisation", "Basic 123456")
-	f(req, "curl -k -X POST -H 'Authorisation: <secret>' 'https://user:xxxxx@foo.com'")
+	req.Header.Set("Authorization", "Basic 123456")
+	f(req, "curl -k -X POST -H 'Authorization: <secret>' 'https://user:xxxxx@foo.com'")
 
 	req = newReq("https://foo.com")
 	req.Header.Set("My-Password", "mypassword")
@@ -66,7 +66,7 @@ func TestRequestToCurl(t *testing.T) {
 	f(req, "curl -k -X POST -H 'Key-For: <secret>' 'https://foo.com'")
 
 	req = newReq("https://foo.com")
-	req.Header.Set("My-Secret-Org", "secret-organisation")
+	req.Header.Set("My-Secret-Org", "secret-organization")
 	f(req, "curl -k -X POST -H 'My-Secret-Org: <secret>' 'https://foo.com'")
 
 	req = newReq("https://foo.com")

app/vmauth/auth_config.go

@@ -83,7 +83,6 @@ type UserInfo struct {
 
 	concurrencyLimitCh      chan struct{}
 	concurrencyLimitReached *metrics.Counter
-	overrideHostHeader      bool
 
 	rt http.RoundTripper
 
@@ -94,8 +93,9 @@ type UserInfo struct {
 
 // HeadersConf represents config for request and response headers.
 type HeadersConf struct {
-	RequestHeaders  []*Header `yaml:"headers,omitempty"`
-	ResponseHeaders []*Header `yaml:"response_headers,omitempty"`
+	RequestHeaders   []*Header `yaml:"headers,omitempty"`
+	ResponseHeaders  []*Header `yaml:"response_headers,omitempty"`
+	KeepOriginalHost *bool     `yaml:"keep_original_host,omitempty"`
 }
 
 func (ui *UserInfo) beginConcurrencyLimit() error {
@@ -150,15 +150,6 @@ func (h *Header) MarshalYAML() (any, error) {
 	return h.sOriginal, nil
 }
 
-func overrideHostHeader(headers []*Header) bool {
-	for _, h := range headers {
-		if h.Name == "Host" && h.Value == "" {
-			return true
-		}
-	}
-	return false
-}
-
 // URLMap is a mapping from source paths to target urls.
 type URLMap struct {
 	// SrcPaths is an optional list of regular expressions, which must match the request path.
@@ -310,12 +301,18 @@ func (up *URLPrefix) getBackendsCount() int {
 
 // getBackendURL returns the backendURL depending on the load balance policy.
 //
+// It can return nil if there are no backend urls available at the moment.
+//
 // backendURL.put() must be called on the returned backendURL after the request is complete.
 func (up *URLPrefix) getBackendURL() *backendURL {
 	up.discoverBackendAddrsIfNeeded()
 
 	pbus := up.bus.Load()
 	bus := *pbus
+	if len(bus) == 0 {
+		return nil
+	}
+
 	if up.loadBalancingPolicy == "first_available" {
 		return getFirstAvailableBackendURL(bus)
 	}
@@ -397,7 +394,7 @@ func (up *URLPrefix) discoverBackendAddrsIfNeeded() {
 			buCopy.Host = addr
 			if port != "" {
 				if n := strings.IndexByte(buCopy.Host, ':'); n >= 0 {
-					// Drop the discovered port and substitute it the the port specified in bu.
+					// Drop the discovered port and substitute it the port specified in bu.
 					buCopy.Host = buCopy.Host[:n]
 				}
 				buCopy.Host += ":" + port
@@ -464,7 +461,7 @@ func getLeastLoadedBackendURL(bus []*backendURL, atomicCounter *atomic.Uint32) *
 	}
 
 	// Slow path - select other backend urls.
-	n := atomicCounter.Add(1)
+	n := atomicCounter.Add(1) - 1
 
 	for i := uint32(0); i < uint32(len(bus)); i++ {
 		idx := (n + i) % uint32(len(bus))
@@ -600,7 +597,7 @@ func initAuthConfig() {
 	// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1240
 	sighupCh := procutil.NewSighupChan()
 
-	_, err := loadAuthConfig()
+	_, err := reloadAuthConfig()
 	if err != nil {
 		logger.Fatalf("cannot load auth config: %s", err)
 	}
@@ -632,7 +629,7 @@ func authConfigReloader(sighupCh <-chan os.Signal) {
 
 	updateFn := func() {
 		configReloads.Inc()
-		updated, err := loadAuthConfig()
+		updated, err := reloadAuthConfig()
 		if err != nil {
 			logger.Errorf("failed to load auth config; using the last successfully loaded config; error: %s", err)
 			configSuccess.Set(0)
@@ -658,27 +655,45 @@ func authConfigReloader(sighupCh <-chan os.Signal) {
 	}
 }
 
-// authConfigData stores the yaml definition for this config.
-// authConfigData needs to be updated each time authConfig is updated.
-var authConfigData atomic.Pointer[[]byte]
-
 var (
-	authConfig   atomic.Pointer[AuthConfig]
-	authUsers    atomic.Pointer[map[string]*UserInfo]
+	// authConfigData stores the yaml definition for this config.
+	// authConfigData needs to be updated each time authConfig is updated.
+	authConfigData atomic.Pointer[[]byte]
+
+	// authConfig contains the currently loaded auth config
+	authConfig atomic.Pointer[AuthConfig]
+
+	// authUsers contains the currently loaded auth users
+	authUsers atomic.Pointer[map[string]*UserInfo]
+
 	authConfigWG sync.WaitGroup
 	stopCh       chan struct{}
 )
 
-// loadAuthConfig loads and applies the config from *authConfigPath.
+// reloadAuthConfig loads and applies the config from *authConfigPath.
 // It returns bool value to identify if new config was applied.
 // The config can be not applied if there is a parsing error
 // or if there are no changes to the current authConfig.
-func loadAuthConfig() (bool, error) {
+func reloadAuthConfig() (bool, error) {
 	data, err := fscore.ReadFileOrHTTP(*authConfigPath)
 	if err != nil {
 		return false, fmt.Errorf("failed to read -auth.config=%q: %w", *authConfigPath, err)
 	}
 
+	ok, err := reloadAuthConfigData(data)
+	if err != nil {
+		return false, fmt.Errorf("failed to pars -auth.config=%q: %w", *authConfigPath, err)
+	}
+	if !ok {
+		return false, nil
+	}
+
+	mp := authUsers.Load()
+	logger.Infof("loaded information about %d users from -auth.config=%q", len(*mp), *authConfigPath)
+	return true, nil
+}
+
+func reloadAuthConfigData(data []byte) (bool, error) {
 	oldData := authConfigData.Load()
 	if oldData != nil && bytes.Equal(data, *oldData) {
 		// there are no updates in the config - skip reloading.
@@ -687,14 +702,13 @@ func loadAuthConfig() (bool, error) {
 
 	ac, err := parseAuthConfig(data)
 	if err != nil {
-		return false, fmt.Errorf("failed to parse -auth.config=%q: %w", *authConfigPath, err)
+		return false, fmt.Errorf("failed to parse auth config: %w", err)
 	}
 
 	m, err := parseAuthConfigUsers(ac)
 	if err != nil {
-		return false, fmt.Errorf("failed to parse users from -auth.config=%q: %w", *authConfigPath, err)
+		return false, fmt.Errorf("failed to parse users from auth config: %w", err)
 	}
-	logger.Infof("loaded information about %d users from -auth.config=%q", len(m), *authConfigPath)
 
 	acPrev := authConfig.Load()
 	if acPrev != nil {
@@ -741,7 +755,6 @@ func parseAuthConfig(data []byte) (*AuthConfig, error) {
 		if err := ui.initURLs(); err != nil {
 			return nil, err
 		}
-		ui.overrideHostHeader = overrideHostHeader(ui.HeadersConf.RequestHeaders)
 
 		metricLabels, err := ui.getMetricLabels()
 		if err != nil {
@@ -806,7 +819,6 @@ func parseAuthConfigUsers(ac *AuthConfig) (map[string]*UserInfo, error) {
 		_ = ac.ms.GetOrCreateGauge(`vmauth_user_concurrent_requests_current`+metricLabels, func() float64 {
 			return float64(len(ui.concurrencyLimitCh))
 		})
-		ui.overrideHostHeader = overrideHostHeader(ui.HeadersConf.RequestHeaders)
 
 		rt, err := newRoundTripper(ui.TLSCAFile, ui.TLSCertFile, ui.TLSKeyFile, ui.TLSServerName, ui.TLSInsecureSkipVerify)
 		if err != nil {
@@ -997,6 +1009,14 @@ func getAuthTokensFromRequest(r *http.Request) []string {
 		}
 	}
 
+	// Authorization via http://user:pass@hosname/path
+	if u := r.URL.User; u != nil && u.Username() != "" {
+		username := u.Username()
+		password, _ := u.Password()
+		at := getHTTPAuthBasicToken(username, password)
+		ats = append(ats, at)
+	}
+
 	return ats
 }
 
@@ -1022,10 +1042,6 @@ func (up *URLPrefix) sanitizeAndInitialize() error {
 }
 
 func sanitizeURLPrefix(urlPrefix *url.URL) (*url.URL, error) {
-	// Remove trailing '/' from urlPrefix
-	for strings.HasSuffix(urlPrefix.Path, "/") {
-		urlPrefix.Path = urlPrefix.Path[:len(urlPrefix.Path)-1]
-	}
 	// Validate urlPrefix
 	if urlPrefix.Scheme != "http" && urlPrefix.Scheme != "https" {
 		return nil, fmt.Errorf("unsupported scheme for `url_prefix: %q`: %q; must be `http` or `https`", urlPrefix, urlPrefix.Scheme)

app/vmauth/auth_config_test.go

@@ -82,6 +82,14 @@ users:
   headers: foobar
 `)
 
+	// Invalid keep_original_host value
+	f(`
+users:
+- username: foo
+  url_prefix: http://foo.bar
+  keep_original_host: foobar
+`)
+
 	// empty url_prefix
 	f(`
 users:
@@ -410,7 +418,7 @@ users:
 				HeadersConf: HeadersConf{
 					RequestHeaders: []*Header{
 						mustNewHeader("'foo: bar'"),
-						mustNewHeader("'xxx: y'"),
+						mustNewHeader("'xxx:'"),
 					},
 				},
 			},
@@ -429,7 +437,7 @@ users:
     url_prefix: ["http://vminsert1/insert/0/prometheus","http://vminsert2/insert/0/prometheus"]
     headers:
     - "foo: bar"
-    - "xxx: y"
+    - "xxx:"
 `, map[string]*UserInfo{
 		getHTTPAuthBearerToken("foo"):    sharedUserInfo,
 		getHTTPAuthBasicToken("foo", ""): sharedUserInfo,
@@ -458,6 +466,7 @@ users:
 	})
 
 	// with default url
+	keepOriginalHost := true
 	f(`
 users:
 - bearer_token: foo
@@ -469,6 +478,7 @@ users:
     headers:
     - "foo: bar"
     - "xxx: y"
+    keep_original_host: true
   default_url:
   - http://default1/select/0/prometheus
   - http://default2/select/0/prometheus
@@ -491,6 +501,7 @@ users:
 							mustNewHeader("'foo: bar'"),
 							mustNewHeader("'xxx: y'"),
 						},
+						KeepOriginalHost: &keepOriginalHost,
 					},
 				},
 			},
@@ -517,6 +528,7 @@ users:
 							mustNewHeader("'foo: bar'"),
 							mustNewHeader("'xxx: y'"),
 						},
+						KeepOriginalHost: &keepOriginalHost,
 					},
 				},
 			},
@@ -536,12 +548,17 @@ users:
   metric_labels:
     dc: eu
     team: dev
+  keep_original_host: true
 - username: foo-same
   password: bar
   url_prefix: https://bar/x
   metric_labels:
     backend_env: test
     team: accounting
+  headers:
+  - "foo: bar"
+  response_headers:
+  - "Abc: def"
 `, map[string]*UserInfo{
 		getHTTPAuthBasicToken("foo-same", "baz"): {
 			Username:  "foo-same",
@@ -551,6 +568,9 @@ users:
 				"dc":   "eu",
 				"team": "dev",
 			},
+			HeadersConf: HeadersConf{
+				KeepOriginalHost: &keepOriginalHost,
+			},
 		},
 		getHTTPAuthBasicToken("foo-same", "bar"): {
 			Username:  "foo-same",
@@ -560,6 +580,14 @@ users:
 				"backend_env": "test",
 				"team":        "accounting",
 			},
+			HeadersConf: HeadersConf{
+				RequestHeaders: []*Header{
+					mustNewHeader("'foo: bar'"),
+				},
+				ResponseHeaders: []*Header{
+					mustNewHeader("'Abc: def'"),
+				},
+			},
 		},
 	})
 }
@@ -695,9 +723,9 @@ func TestGetLeastLoadedBackendURL(t *testing.T) {
 
 	fn := func(ns ...int) {
 		t.Helper()
-		bus := up.bus.Load()
-		pbus := *bus
-		for i, b := range pbus {
+		pbus := up.bus.Load()
+		bus := *pbus
+		for i, b := range bus {
 			got := int(b.concurrentRequests.Load())
 			exp := ns[i]
 			if got != exp {
@@ -707,39 +735,39 @@ func TestGetLeastLoadedBackendURL(t *testing.T) {
 	}
 
 	up.getBackendURL()
-	fn(0, 1, 0)
+	fn(1, 0, 0)
 	up.getBackendURL()
-	fn(0, 1, 1)
+	fn(1, 1, 0)
 	up.getBackendURL()
 	fn(1, 1, 1)
 
 	up.getBackendURL()
 	up.getBackendURL()
-	fn(1, 2, 2)
+	fn(2, 2, 1)
 
 	bus := up.bus.Load()
 	pbus := *bus
 	pbus[0].concurrentRequests.Add(2)
 	pbus[2].concurrentRequests.Add(5)
-	fn(3, 2, 7)
+	fn(4, 2, 6)
 
 	up.getBackendURL()
-	fn(3, 3, 7)
+	fn(4, 3, 6)
 
 	up.getBackendURL()
-	fn(3, 4, 7)
+	fn(4, 4, 6)
 
 	up.getBackendURL()
-	fn(4, 4, 7)
+	fn(4, 5, 6)
 
 	up.getBackendURL()
-	fn(5, 4, 7)
+	fn(5, 5, 6)
 
 	up.getBackendURL()
-	fn(5, 5, 7)
+	fn(6, 5, 6)
 
 	up.getBackendURL()
-	fn(6, 5, 7)
+	fn(6, 6, 6)
 
 	up.getBackendURL()
 	fn(6, 6, 7)

app/vmauth/deployment/Dockerfile

@@ -1,8 +1,8 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 EXPOSE 8427
 
 ENTRYPOINT ["/vmauth-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmauth-prod

app/vmauth/main.go

@@ -50,7 +50,7 @@ var (
 		`Such requests are always counted at vmauth_http_request_errors_total{reason="invalid_auth_token"} metric, which is exposed at /metrics page`)
 	failTimeout               = flag.Duration("failTimeout", 3*time.Second, "Sets a delay period for load balancing to skip a malfunctioning backend")
 	maxRequestBodySizeToRetry = flagutil.NewBytes("maxRequestBodySizeToRetry", 16*1024, "The maximum request body size, which can be cached and re-tried at other backends. "+
-		"Bigger values may require more memory. Negative or zero values disable request body caching and retries.")
+		"Bigger values may require more memory. Zero or negative value disables caching of request body. This may be useful when proxying data ingestion requests")
 	backendTLSInsecureSkipVerify = flag.Bool("backend.tlsInsecureSkipVerify", false, "Whether to skip TLS verification when connecting to backends over HTTPS. "+
 		"See https://docs.victoriametrics.com/vmauth/#backend-tls-setup")
 	backendTLSCAFile = flag.String("backend.TLSCAFile", "", "Optional path to TLS root CA file, which is used for TLS verification when connecting to backends over HTTPS. "+
@@ -117,8 +117,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 			return true
 		}
 
-		w.Header().Set("WWW-Authenticate", `Basic realm="Restricted"`)
-		http.Error(w, "missing `Authorization` request header", http.StatusUnauthorized)
+		handleMissingAuthorizationError(w)
 		return true
 	}
 
@@ -189,26 +188,27 @@ func processRequest(w http.ResponseWriter, r *http.Request, ui *UserInfo) {
 			// to a route that is not in the configuration for unauthorized user.
 			// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5236
 			if ui.BearerToken == "" && ui.Username == "" && len(*authUsers.Load()) > 0 {
-				w.Header().Set("WWW-Authenticate", `Basic realm="Restricted"`)
-				http.Error(w, "missing `Authorization` request header", http.StatusUnauthorized)
+				handleMissingAuthorizationError(w)
 				return
 			}
 			missingRouteRequests.Inc()
-			httpserver.Errorf(w, r, "missing route for %q", u.String())
+			httpserver.Errorf(w, r, "missing route for %s", u.String())
 			return
 		}
 		up, hc = ui.DefaultURL, ui.HeadersConf
 		isDefault = true
 	}
-	// caching makes sense only for positive non zero size
-	if maxRequestBodySizeToRetry.IntN() > 0 {
-		rtb := getReadTrackingBody(r.Body, int(r.ContentLength))
-		defer putReadTrackingBody(rtb)
-		r.Body = rtb
-	}
+
+	rtb := getReadTrackingBody(r.Body, maxRequestBodySizeToRetry.IntN())
+	defer putReadTrackingBody(rtb)
+	r.Body = rtb
+
 	maxAttempts := up.getBackendsCount()
 	for i := 0; i < maxAttempts; i++ {
 		bu := up.getBackendURL()
+		if bu == nil {
+			break
+		}
 		targetURL := bu.url
 		// Don't change path and add request_path query param for default route.
 		if isDefault {
@@ -218,7 +218,15 @@ func processRequest(w http.ResponseWriter, r *http.Request, ui *UserInfo) {
 		} else { // Update path for regular routes.
 			targetURL = mergeURLs(targetURL, u, up.dropSrcPathPrefixParts)
 		}
-		ok := tryProcessingRequest(w, r, targetURL, hc, up.retryStatusCodes, ui)
+
+		wasLocalRetry := false
+	again:
+		ok, needLocalRetry := tryProcessingRequest(w, r, targetURL, hc, up.retryStatusCodes, ui)
+		if needLocalRetry && !wasLocalRetry {
+			wasLocalRetry = true
+			goto again
+		}
+
 		bu.put()
 		if ok {
 			return
@@ -226,29 +234,28 @@ func processRequest(w http.ResponseWriter, r *http.Request, ui *UserInfo) {
 		bu.setBroken()
 	}
 	err := &httpserver.ErrorWithStatusCode{
-		Err:        fmt.Errorf("all the backends for the user %q are unavailable", ui.name()),
-		StatusCode: http.StatusServiceUnavailable,
+		Err:        fmt.Errorf("all the %d backends for the user %q are unavailable", up.getBackendsCount(), ui.name()),
+		StatusCode: http.StatusBadGateway,
 	}
 	httpserver.Errorf(w, r, "%s", err)
 	ui.backendErrors.Inc()
 }
 
-func tryProcessingRequest(w http.ResponseWriter, r *http.Request, targetURL *url.URL, hc HeadersConf, retryStatusCodes []int, ui *UserInfo) bool {
-	// This code has been copied from net/http/httputil/reverseproxy.go
+func tryProcessingRequest(w http.ResponseWriter, r *http.Request, targetURL *url.URL, hc HeadersConf, retryStatusCodes []int, ui *UserInfo) (bool, bool) {
 	req := sanitizeRequestHeaders(r)
-	req.URL = targetURL
 
-	if req.URL.Scheme == "https" || ui.overrideHostHeader {
-		// Override req.Host only for https requests, since https server verifies hostnames during TLS handshake,
-		// so it expects the targetURL.Host in the request.
-		// There is no need in overriding the req.Host for http requests, since it is expected that backend server
-		// may properly process queries with the original req.Host.
-		req.Host = targetURL.Host
-	}
+	req.URL = targetURL
+	req.Header.Set("User-Agent", "vmauth")
 	updateHeadersByConfig(req.Header, hc.RequestHeaders)
-	var trivialRetries int
+	if hc.KeepOriginalHost == nil || !*hc.KeepOriginalHost {
+		if host := getHostHeader(hc.RequestHeaders); host != "" {
+			req.Host = host
+		} else {
+			req.Host = targetURL.Host
+		}
+	}
+
 	rtb, rtbOK := req.Body.(*readTrackingBody)
-again:
 	res, err := ui.rt.RoundTrip(req)
 	if err != nil {
 		if errors.Is(err, context.Canceled) || errors.Is(err, context.DeadlineExceeded) {
@@ -260,7 +267,7 @@ again:
 				// Timed out request must be counted as errors, since this usually means that the backend is slow.
 				ui.backendErrors.Inc()
 			}
-			return true
+			return true, false
 		}
 		if !rtbOK || !rtb.canRetry() {
 			// Request body cannot be re-sent to another backend. Return the error to the client then.
@@ -270,19 +277,19 @@ again:
 			}
 			httpserver.Errorf(w, r, "%s", err)
 			ui.backendErrors.Inc()
-			return true
+			return true, false
 		}
-		// one time retry trivial network errors, such as proxy idle timeout misconfiguration or socket close by OS
-		if (netutil.IsTrivialNetworkError(err) || errors.Is(err, io.EOF)) && trivialRetries < 1 {
-			trivialRetries++
-			goto again
+		if netutil.IsTrivialNetworkError(err) {
+			// Retry request at the same backend on trivial network errors, such as proxy idle timeout misconfiguration or socket close by OS
+			return false, true
 		}
+
 		// Retry the request if its body wasn't read yet. This usually means that the backend isn't reachable.
 		remoteAddr := httpserver.GetQuotedRemoteAddr(r)
 		// NOTE: do not use httpserver.GetRequestURI
 		// it explicitly reads request body, which may fail retries.
 		logger.Warnf("remoteAddr: %s; requestURI: %s; retrying the request to %s because of response error: %s", remoteAddr, req.URL, targetURL, err)
-		return false
+		return false, false
 	}
 	if slices.Contains(retryStatusCodes, res.StatusCode) {
 		_ = res.Body.Close()
@@ -296,7 +303,7 @@ again:
 			}
 			httpserver.Errorf(w, r, "%s", err)
 			ui.backendErrors.Inc()
-			return true
+			return true, false
 		}
 		// Retry requests at other backends if it matches retryStatusCodes.
 		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/4893
@@ -305,7 +312,7 @@ again:
 		// it explicitly reads request body, which may fail retries.
 		logger.Warnf("remoteAddr: %s; requestURI: %s; retrying the request to %s because response status code=%d belongs to retry_status_codes=%d",
 			remoteAddr, req.URL, targetURL, res.StatusCode, retryStatusCodes)
-		return false
+		return false, false
 	}
 	removeHopHeaders(res.Header)
 	copyHeader(w.Header(), res.Header)
@@ -321,9 +328,9 @@ again:
 		remoteAddr := httpserver.GetQuotedRemoteAddr(r)
 		requestURI := httpserver.GetRequestURI(r)
 		logger.Warnf("remoteAddr: %s; requestURI: %s; error when proxying response body from %s: %s", remoteAddr, requestURI, targetURL, err)
-		return true
+		return true, false
 	}
-	return true
+	return true, false
 }
 
 var copyBufPool bytesutil.ByteBufferPool
@@ -336,12 +343,21 @@ func copyHeader(dst, src http.Header) {
 	}
 }
 
-func updateHeadersByConfig(headers http.Header, config []*Header) {
-	for _, h := range config {
+func getHostHeader(headers []*Header) string {
+	for _, h := range headers {
+		if h.Name == "Host" {
+			return h.Value
+		}
+	}
+	return ""
+}
+
+func updateHeadersByConfig(dst http.Header, src []*Header) {
+	for _, h := range src {
 		if h.Value == "" {
-			headers.Del(h.Name)
+			dst.Del(h.Name)
 		} else {
-			headers.Set(h.Name, h.Value)
+			dst.Set(h.Name, h.Value)
 		}
 	}
 }
@@ -481,6 +497,11 @@ See the docs at https://docs.victoriametrics.com/vmauth/ .
 	flagutil.Usage(s)
 }
 
+func handleMissingAuthorizationError(w http.ResponseWriter) {
+	w.Header().Set("WWW-Authenticate", `Basic realm="Restricted"`)
+	http.Error(w, "missing 'Authorization' request header", http.StatusUnauthorized)
+}
+
 func handleConcurrencyLimitError(w http.ResponseWriter, r *http.Request, err error) {
 	w.Header().Add("Retry-After", "10")
 	err = &httpserver.ErrorWithStatusCode{
@@ -490,85 +511,124 @@ func handleConcurrencyLimitError(w http.ResponseWriter, r *http.Request, err err
 	httpserver.Errorf(w, r, "%s", err)
 }
 
+// readTrackingBody must be obtained via getReadTrackingBody()
 type readTrackingBody struct {
+	// maxBodySize is the maximum body size to cache in buf.
+	//
+	// Bigger bodies cannot be retried.
+	maxBodySize int
+
 	// r contains reader for initial data reading
 	r io.ReadCloser
 
-	// buf is a buffer for data read from r. Buf size is limited by maxRequestBodySizeToRetry.
-	// If more than maxRequestBodySizeToRetry is read from r, then cannotRetry is set to true.
+	// buf is a buffer for data read from r. Buf size is limited by maxBodySize.
+	// If more than maxBodySize is read from r, then cannotRetry is set to true.
 	buf []byte
 
-	// cannotRetry is set to true when more than maxRequestBodySizeToRetry are read from r.
+	// readBuf points to the cached data at buf, which must be read in the next call to Read().
+	readBuf []byte
+
+	// cannotRetry is set to true when more than maxBodySize bytes are read from r.
 	// In this case the read data cannot fit buf, so it cannot be re-read from buf.
 	cannotRetry bool
 
 	// bufComplete is set to true when buf contains complete request body read from r.
 	bufComplete bool
-
-	// offset is an offset at buf for the next data read if needReadBuf is set to true.
-	offset int
 }
 
-// Read implements io.Reader interface
-// tracks body reading requests
+func (rtb *readTrackingBody) reset() {
+	rtb.maxBodySize = 0
+	rtb.r = nil
+	rtb.buf = rtb.buf[:0]
+	rtb.readBuf = nil
+	rtb.cannotRetry = false
+	rtb.bufComplete = false
+}
+
+func getReadTrackingBody(r io.ReadCloser, maxBodySize int) *readTrackingBody {
+	v := readTrackingBodyPool.Get()
+	if v == nil {
+		v = &readTrackingBody{}
+	}
+	rtb := v.(*readTrackingBody)
+
+	if maxBodySize < 0 {
+		maxBodySize = 0
+	}
+	rtb.maxBodySize = maxBodySize
+
+	if r == nil {
+		// This is GET request without request body
+		r = (*zeroReader)(nil)
+	}
+	rtb.r = r
+	return rtb
+}
+
+type zeroReader struct{}
+
+func (r *zeroReader) Read(_ []byte) (int, error) {
+	return 0, io.EOF
+}
+func (r *zeroReader) Close() error {
+	return nil
+}
+
+func putReadTrackingBody(rtb *readTrackingBody) {
+	rtb.reset()
+	readTrackingBodyPool.Put(rtb)
+}
+
+var readTrackingBodyPool sync.Pool
+
+// Read implements io.Reader interface.
 func (rtb *readTrackingBody) Read(p []byte) (int, error) {
-	if rtb.offset < len(rtb.buf) {
-		if rtb.cannotRetry {
-			return 0, fmt.Errorf("cannot retry reading data from buf")
-		}
-		nb := copy(p, rtb.buf[rtb.offset:])
-		rtb.offset += nb
-		if rtb.bufComplete {
-			if rtb.offset == len(rtb.buf) {
-				return nb, io.EOF
-			}
-			return nb, nil
-		}
-		if nb < len(p) {
-			nr, err := rtb.readFromStream(p[nb:])
-			return nb + nr, err
-		}
-		return nb, nil
+	if len(rtb.readBuf) > 0 {
+		n := copy(p, rtb.readBuf)
+		rtb.readBuf = rtb.readBuf[n:]
+		return n, nil
 	}
-	if rtb.bufComplete {
-		return 0, io.EOF
-	}
-	return rtb.readFromStream(p)
-}
 
-func (rtb *readTrackingBody) readFromStream(p []byte) (int, error) {
 	if rtb.r == nil {
-		return 0, fmt.Errorf("cannot read data after closing the reader")
+		if rtb.bufComplete {
+			return 0, io.EOF
+		}
+		return 0, fmt.Errorf("cannot read client request body after closing client reader")
 	}
+
 	n, err := rtb.r.Read(p)
 	if rtb.cannotRetry {
 		return n, err
 	}
-	if rtb.offset+n > maxRequestBodySizeToRetry.IntN() {
-		rtb.cannotRetry = true
-	}
-	if n > 0 {
-		rtb.offset += n
-		rtb.buf = append(rtb.buf, p[:n]...)
-	}
-	if err != nil {
-		if err == io.EOF {
-			rtb.bufComplete = true
-			return n, err
-		}
+
+	if len(rtb.buf)+n > rtb.maxBodySize {
 		rtb.cannotRetry = true
 		return n, err
 	}
-	return n, nil
+	rtb.buf = append(rtb.buf, p[:n]...)
+	if err == io.EOF {
+		rtb.bufComplete = true
+	}
+	return n, err
 }
 
 func (rtb *readTrackingBody) canRetry() bool {
-	return !rtb.cannotRetry
+	if rtb.cannotRetry {
+		return false
+	}
+	if rtb.bufComplete {
+		return true
+	}
+	return rtb.r != nil
 }
 
 // Close implements io.Closer interface.
 func (rtb *readTrackingBody) Close() error {
-	rtb.offset = 0
+	if !rtb.cannotRetry {
+		rtb.readBuf = rtb.buf
+	} else {
+		rtb.readBuf = nil
+	}
 
 	// Close rtb.r only if the request body is completely read or if it is too big.
 	// http.Roundtrip performs body.Close call even without any Read calls,
@@ -584,38 +644,3 @@ func (rtb *readTrackingBody) Close() error {
 
 	return nil
 }
-
-var readTrackingBodyPool sync.Pool
-
-func getReadTrackingBody(origin io.ReadCloser, b int) *readTrackingBody {
-	bufSize := 1024
-	if b > 0 && b < maxRequestBodySizeToRetry.IntN() {
-		bufSize = b
-	}
-	v := readTrackingBodyPool.Get()
-	if v == nil {
-		v = &readTrackingBody{
-			buf: make([]byte, 0, bufSize),
-		}
-	}
-	rtb := v.(*readTrackingBody)
-	rtb.r = origin
-	if bufSize > cap(rtb.buf) {
-		rtb.buf = make([]byte, 0, bufSize)
-	}
-
-	return rtb
-}
-
-func putReadTrackingBody(rtb *readTrackingBody) {
-	if rtb.r != nil {
-		_ = rtb.r.Close()
-	}
-	rtb.r = nil
-	rtb.buf = rtb.buf[:0]
-	rtb.offset = 0
-	rtb.cannotRetry = false
-	rtb.bufComplete = false
-
-	readTrackingBodyPool.Put(rtb)
-}

app/vmauth/main_test.go

@@ -2,18 +2,520 @@ package main
 
 import (
 	"bytes"
+	"fmt"
 	"io"
+	"net"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"sync/atomic"
 	"testing"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
 )
 
-func TestReadTrackingBodyRetrySuccess(t *testing.T) {
-	f := func(s string) {
+func TestRequestHandler(t *testing.T) {
+	f := func(cfgStr, requestURL string, backendHandler http.HandlerFunc, responseExpected string) {
 		t.Helper()
-		rtb := &readTrackingBody{
-			r: io.NopCloser(bytes.NewBufferString(s)),
+
+		ts := httptest.NewServer(backendHandler)
+		defer ts.Close()
+
+		cfgStr = strings.ReplaceAll(cfgStr, "{BACKEND}", ts.URL)
+		responseExpected = strings.ReplaceAll(responseExpected, "{BACKEND}", ts.URL)
+
+		cfgOrigP := authConfigData.Load()
+		if _, err := reloadAuthConfigData([]byte(cfgStr)); err != nil {
+			t.Fatalf("cannot load config data: %s", err)
 		}
+		defer func() {
+			cfgOrig := []byte("unauthorized_user:\n  url_prefix: http://foo/bar")
+			if cfgOrigP != nil {
+				cfgOrig = *cfgOrigP
+			}
+			_, err := reloadAuthConfigData(cfgOrig)
+			if err != nil {
+				t.Fatalf("cannot load the original config: %s", err)
+			}
+		}()
+
+		r, err := http.NewRequest(http.MethodGet, requestURL, nil)
+		if err != nil {
+			t.Fatalf("cannot initialize http request: %s", err)
+		}
+
+		r.RequestURI = r.URL.RequestURI()
+		r.RemoteAddr = "42.2.3.84:6789"
+		r.Header.Set("X-Forwarded-For", "12.34.56.78")
+		r.Header.Set("Connection", "Some-Header,Other-Header")
+		r.Header.Set("Some-Header", "foobar")
+		r.Header.Set("Pass-Header", "abc")
+
+		w := &fakeResponseWriter{}
+		if !requestHandler(w, r) {
+			t.Fatalf("unexpected false is returned from requestHandler")
+		}
+
+		response := w.getResponse()
+		response = strings.ReplaceAll(response, "\r\n", "\n")
+		response = strings.TrimSpace(response)
+		responseExpected = strings.TrimSpace(responseExpected)
+		if response != responseExpected {
+			t.Fatalf("unexpected response\ngot\n%s\nwant\n%s", response, responseExpected)
+		}
+	}
+
+	// regular url_prefix
+	cfgStr := `
+unauthorized_user:
+  url_prefix: {BACKEND}/foo?bar=baz`
+	requestURL := "http://some-host.com/abc/def?some_arg=some_value"
+	backendHandler := func(w http.ResponseWriter, r *http.Request) {
+		h := w.Header()
+		h.Set("Connection", "close")
+		h.Set("Foo", "bar")
+
+		var bb bytes.Buffer
+		if err := r.Header.Write(&bb); err != nil {
+			panic(fmt.Errorf("unexpected error when marshaling headers: %w", err))
+		}
+		fmt.Fprintf(w, "requested_url=http://%s%s\n%s", r.Host, r.URL, bb.String())
+	}
+	responseExpected := `
+statusCode=200
+Foo: bar
+requested_url={BACKEND}/foo/abc/def?bar=baz&some_arg=some_value
+Pass-Header: abc
+User-Agent: vmauth
+X-Forwarded-For: 12.34.56.78, 42.2.3.84`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// keep_original_host
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo?bar=baz"
+  keep_original_host: true`
+	requestURL = "http://some-host.com/abc/def"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url=http://some-host.com/foo/abc/def?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// override user-agent header
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo?bar=baz"
+  headers:
+  - "User-Agent: foobar"`
+	requestURL = "http://some-host.com/abc/def"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s\nUser-Agent=%s", r.Host, r.URL, r.Header.Get("User-Agent"))
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo/abc/def?bar=baz
+User-Agent=foobar`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// delete user-agent header
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo?bar=baz"
+  headers:
+  - "User-Agent:"`
+	requestURL = "http://some-host.com/abc/def"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s\nUser-Agent=%s", r.Host, r.URL, r.Header.Get("User-Agent"))
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo/abc/def?bar=baz
+User-Agent=Go-http-client/1.1`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// override request host with non-empty host
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo?bar=baz"
+  headers:
+  - "Host: other-host:12345"
+  - "abc:"`
+	requestURL = "http://some-host.com/abc/def"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url=http://other-host:12345/foo/abc/def?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// override request host with empty host
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo?bar=baz"
+  headers:
+  - "Host:"`
+	requestURL = "http://some-host.com/abc/def"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo/abc/def?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// /-/reload handler failure
+	origAuthKey := reloadAuthKey.Get()
+	if err := reloadAuthKey.Set("secret"); err != nil {
+		t.Fatalf("unexpected error: %s", err)
+	}
+	cfgStr = `
+unauthorized_user:
+  url_prefix: "{BACKEND}/foo"`
+	requestURL = "http://some-host.com/-/reload"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=401
+The provided authKey doesn't match -reloadAuthKey`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+	if err := reloadAuthKey.Set(origAuthKey); err != nil {
+		t.Fatalf("unexpected error: %s", err)
+	}
+
+	// missing authorization
+	cfgStr = `
+users:
+- username: foo
+  url_prefix: "{BACKEND}/bar"`
+	requestURL = "http://some-host.com/a/b"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=401
+Www-Authenticate: Basic realm="Restricted"
+missing 'Authorization' request header`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// incorrect authorization
+	cfgStr = `
+users:
+- username: foo
+  password: secret
+  url_prefix: "{BACKEND}/bar"`
+	requestURL = "http://foo:invalid-secret@some-host.com/a/b"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=401
+Unauthorized`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// incorrect authorization with logging invalid auth tokens
+	origLogInvalidAuthTokens := *logInvalidAuthTokens
+	*logInvalidAuthTokens = true
+	cfgStr = `
+users:
+- username: foo
+  password: secret
+  url_prefix: "{BACKEND}/bar"`
+	requestURL = "http://foo:invalid-secret@some-host.com/a/b?c=d"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=401
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /a/b?c=d; cannot authorize request with auth tokens ["http_auth:Basic Zm9vOmludmFsaWQtc2VjcmV0"]`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+	*logInvalidAuthTokens = origLogInvalidAuthTokens
+
+	// correct authorization
+	cfgStr = `
+users:
+- username: foo
+  password: secret
+  url_prefix: "{BACKEND}/bar"`
+	requestURL = "http://foo:secret@some-host.com/a/b"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/bar/a/b`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify how path cleanup works
+	cfgStr = `
+unauthorized_user:
+  url_prefix: {BACKEND}/foo?bar=baz`
+	requestURL = "http://some-host.com/../../a//.///bar/"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo/a/bar/?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify how path cleanup works for url without path
+	cfgStr = `
+unauthorized_user:
+  url_prefix: {BACKEND}/foo?bar=baz`
+	requestURL = "http://some-host.com/"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify how path cleanup works for url without path if url_prefix path ends with /
+	cfgStr = `
+unauthorized_user:
+  url_prefix: {BACKEND}/foo/?bar=baz`
+	requestURL = "http://some-host.com/"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/foo/?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify how path cleanup works for url without path and the url_prefix without path prefix
+	cfgStr = `
+unauthorized_user:
+  url_prefix: {BACKEND}/?bar=baz`
+	requestURL = "http://some-host.com/"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/?bar=baz`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify routing to default_url
+	cfgStr = `
+unauthorized_user:
+  url_map:
+  - src_paths: ["/foo/.+"]
+    url_prefix: {BACKEND}/x-foo/
+  default_url: {BACKEND}/404.html`
+	requestURL = "http://some-host.com/abc?de=fg"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/404.html?request_path=http%3A%2F%2Fsome-host.com%2Fabc%3Fde%3Dfg`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// verify routing to default url_prefix
+	cfgStr = `
+unauthorized_user:
+  url_map:
+  - src_paths: ["/foo/.+"]
+    url_prefix: {BACKEND}/x-foo/
+  url_prefix: {BACKEND}/default`
+	requestURL = "http://some-host.com/abc?de=fg"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/default/abc?de=fg`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// missing default_url and default url_prefix for unauthorized user
+	cfgStr = `
+unauthorized_user:
+  url_map:
+  - src_paths: ["/foo/.+"]
+    url_prefix: {BACKEND}/x-foo/`
+	requestURL = "http://some-host.com/abc?de=fg"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=400
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /abc?de=fg; missing route for http://some-host.com/abc?de=fg`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// missing default_url and default url_prefix for unauthorized user when there are configs for authorized users
+	cfgStr = `
+users:
+- username: some-user
+  url_map:
+  - src_paths: ["/foo/.+"]
+    url_prefix: {BACKEND}/x-foo/
+unauthorized_user:
+  url_map:
+  - src_paths: ["/abc/.*"]
+    url_prefix: {BACKEND}/x-bar`
+	requestURL = "http://some-host.com/abc?de=fg"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=401
+Www-Authenticate: Basic realm="Restricted"
+missing 'Authorization' request header`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// all the backend_urls are unavailable for unauthorized user
+	cfgStr = `
+unauthorized_user:
+  url_map:
+  - src_paths: ["/foo/.*"]
+    url_prefix:
+    - http://127.0.0.1:1/
+    - http://127.0.0.1:2/`
+	requestURL = "http://some-host.com/foo/?de=fg"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=502
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /foo/?de=fg; all the 2 backends for the user "" are unavailable`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// all the backend_urls are unavailable for authorized user
+	cfgStr = `
+users:
+- username: some-user
+  url_map:
+  - src_paths: ["/foo/.*"]
+    url_prefix:
+    - http://127.0.0.1:1/
+    - http://127.0.0.1:2/`
+	requestURL = "http://some-user@some-host.com/foo/?de=fg"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=502
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /foo/?de=fg; all the 2 backends for the user "some-user" are unavailable`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+
+	// zero discovered backend IPs
+	customResolver := &fakeResolver{
+		Resolver: &net.Resolver{},
+		lookupIPAddrResults: map[string][]net.IPAddr{
+			"some-addr": {},
+		},
+	}
+	origResolver := netutil.Resolver
+	netutil.Resolver = customResolver
+	cfgStr = `
+unauthorized_user:
+  url_prefix: ['http://some-addr:1234/foo/bar']
+  discover_backend_ips: true`
+	requestURL = "http://abc.com/def/?de=fg"
+	backendHandler = func(_ http.ResponseWriter, _ *http.Request) {
+		panic(fmt.Errorf("backend handler shouldn't be called"))
+	}
+	responseExpected = `
+statusCode=502
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /def/?de=fg; all the 0 backends for the user "" are unavailable`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+	netutil.Resolver = origResolver
+
+	// retry_status_codes failure
+	var retries atomic.Int64
+	cfgStr = `
+unauthorized_user:
+  url_prefix: ['{BACKEND}/path1', '{BACKEND}/path2']
+  retry_status_codes: [500, 502]`
+	requestURL = "http://some-host.com/foo/?de=fg"
+	backendHandler = func(w http.ResponseWriter, _ *http.Request) {
+		retries.Add(1)
+		w.WriteHeader(500)
+	}
+	responseExpected = `
+statusCode=502
+remoteAddr: "42.2.3.84:6789, X-Forwarded-For: 12.34.56.78"; requestURI: /foo/?de=fg; all the 2 backends for the user "" are unavailable`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+	if n := retries.Load(); n != 2 {
+		t.Fatalf("unexpected number of retries; got %d; want 2", n)
+	}
+
+	// retry_status_codes success
+	retries.Store(0)
+	cfgStr = `
+unauthorized_user:
+  url_prefix: ['{BACKEND}/path1', '{BACKEND}/path2']
+  retry_status_codes: [500, 502]`
+	requestURL = "http://some-host.com/foo/?de=fg"
+	backendHandler = func(w http.ResponseWriter, r *http.Request) {
+		if n := retries.Add(1); n < 2 {
+			w.WriteHeader(500)
+			return
+		}
+		fmt.Fprintf(w, "requested_url=http://%s%s", r.Host, r.URL)
+	}
+	responseExpected = `
+statusCode=200
+requested_url={BACKEND}/path2/foo/?de=fg`
+	f(cfgStr, requestURL, backendHandler, responseExpected)
+	if n := retries.Load(); n != 2 {
+		t.Fatalf("unexpected number of retries; got %d; want 2", n)
+	}
+}
+
+type fakeResponseWriter struct {
+	h http.Header
+
+	bb bytes.Buffer
+}
+
+func (w *fakeResponseWriter) getResponse() string {
+	return w.bb.String()
+}
+
+func (w *fakeResponseWriter) Header() http.Header {
+	if w.h == nil {
+		w.h = http.Header{}
+	}
+	return w.h
+}
+
+func (w *fakeResponseWriter) Write(p []byte) (int, error) {
+	return w.bb.Write(p)
+}
+
+func (w *fakeResponseWriter) WriteHeader(statusCode int) {
+	fmt.Fprintf(&w.bb, "statusCode=%d\n", statusCode)
+	if w.h == nil {
+		return
+	}
+	err := w.h.WriteSubset(&w.bb, map[string]bool{
+		"Content-Length":         true,
+		"Content-Type":           true,
+		"Date":                   true,
+		"X-Content-Type-Options": true,
+	})
+	if err != nil {
+		panic(fmt.Errorf("cannot marshal headers: %s", err))
+	}
+}
+
+func TestReadTrackingBody_RetrySuccess(t *testing.T) {
+	f := func(s string, maxBodySize int) {
+		t.Helper()
+
+		rtb := getReadTrackingBody(io.NopCloser(bytes.NewBufferString(s)), maxBodySize)
+		defer putReadTrackingBody(rtb)
+
 		if !rtb.canRetry() {
-			t.Fatalf("canRetry() must return true")
+			t.Fatalf("canRetry() must return true before reading anything")
 		}
 		for i := 0; i < 5; i++ {
 			data, err := io.ReadAll(rtb)
@@ -32,41 +534,99 @@ func TestReadTrackingBodyRetrySuccess(t *testing.T) {
 		}
 	}
 
-	f("")
-	f("foo")
-	f("foobar")
-	f(newTestString(maxRequestBodySizeToRetry.IntN()))
+	f("", 0)
+	f("", -1)
+	f("", 100)
+	f("foo", 100)
+	f("foobar", 100)
+	f(newTestString(1000), 1000)
 }
 
-func TestReadTrackingBodyRetryFailure(t *testing.T) {
-	f := func(s string) {
+func TestReadTrackingBody_RetrySuccessPartialRead(t *testing.T) {
+	f := func(s string, maxBodySize int) {
 		t.Helper()
-		rtb := &readTrackingBody{
-			r: io.NopCloser(bytes.NewBufferString(s)),
+
+		// Check the case with partial read
+		rtb := getReadTrackingBody(io.NopCloser(bytes.NewBufferString(s)), maxBodySize)
+		defer putReadTrackingBody(rtb)
+
+		for i := 0; i < len(s); i++ {
+			buf := make([]byte, i)
+			n, err := io.ReadFull(rtb, buf)
+			if err != nil {
+				t.Fatalf("unexpected error when reading %d bytes: %s", i, err)
+			}
+			if n != i {
+				t.Fatalf("unexpected number of bytes read; got %d; want %d", n, i)
+			}
+			if string(buf) != s[:i] {
+				t.Fatalf("unexpected data read with the length %d\ngot\n%s\nwant\n%s", i, buf, s[:i])
+			}
+			if err := rtb.Close(); err != nil {
+				t.Fatalf("unexpected error when closing reader after reading %d bytes", i)
+			}
+			if !rtb.canRetry() {
+				t.Fatalf("canRetry() must return true after closing the reader after reading %d bytes", i)
+			}
+		}
+
+		data, err := io.ReadAll(rtb)
+		if err != nil {
+			t.Fatalf("unexpected error when reading all the data: %s", err)
+		}
+		if string(data) != s {
+			t.Fatalf("unexpected data read\ngot\n%s\nwant\n%s", data, s)
+		}
+		if err := rtb.Close(); err != nil {
+			t.Fatalf("unexpected error when closing readTrackingBody: %s", err)
 		}
 		if !rtb.canRetry() {
-			t.Fatalf("canRetry() must return true")
+			t.Fatalf("canRetry() must return true after closing the reader after reading all the input")
+		}
+	}
+
+	f("", 0)
+	f("", -1)
+	f("", 100)
+	f("foo", 100)
+	f("foobar", 100)
+	f(newTestString(1000), 1000)
+}
+
+func TestReadTrackingBody_RetryFailureTooBigBody(t *testing.T) {
+	f := func(s string, maxBodySize int) {
+		t.Helper()
+
+		rtb := getReadTrackingBody(io.NopCloser(bytes.NewBufferString(s)), maxBodySize)
+		defer putReadTrackingBody(rtb)
+
+		if !rtb.canRetry() {
+			t.Fatalf("canRetry() must return true before reading anything")
 		}
 		buf := make([]byte, 1)
-		n, err := rtb.Read(buf)
+		n, err := io.ReadFull(rtb, buf)
 		if err != nil {
 			t.Fatalf("unexpected error when reading a single byte: %s", err)
 		}
 		if n != 1 {
 			t.Fatalf("unexpected number of bytes read; got %d; want 1", n)
 		}
+		if !rtb.canRetry() {
+			t.Fatalf("canRetry() must return true after reading one byte")
+		}
 		data, err := io.ReadAll(rtb)
 		if err != nil {
 			t.Fatalf("unexpected error when reading all the data: %s", err)
 		}
-		if string(buf)+string(data) != s {
-			t.Fatalf("unexpected data read\ngot\n%s\nwant\n%s", string(buf)+string(data), s)
+		dataRead := string(buf) + string(data)
+		if dataRead != s {
+			t.Fatalf("unexpected data read\ngot\n%s\nwant\n%s", dataRead, s)
 		}
 		if err := rtb.Close(); err != nil {
 			t.Fatalf("unexpected error when closing readTrackingBody: %s", err)
 		}
 		if rtb.canRetry() {
-			t.Fatalf("canRetry() must return false")
+			t.Fatalf("canRetry() must return false after closing the reader")
 		}
 
 		data, err = io.ReadAll(rtb)
@@ -78,132 +638,55 @@ func TestReadTrackingBodyRetryFailure(t *testing.T) {
 		}
 	}
 
-	f(newTestString(maxRequestBodySizeToRetry.IntN() + 1))
-	f(newTestString(2 * maxRequestBodySizeToRetry.IntN()))
+	const maxBodySize = 1000
+	f(newTestString(maxBodySize+1), maxBodySize)
+	f(newTestString(2*maxBodySize), maxBodySize)
 }
 
-// request body not over maxRequestBodySizeToRetry
-// 1. When writing data downstream, buf only caches part of the data because the downstream connection is disconnected.
-// 2. retry request: because buf caches some data, first read buf and then read stream when retrying
-// 3. retry request: the data has been read to buf in the second step. if the request fails, retry to read all buf later.
-func TestRetryReadSuccessAfterPartialRead(t *testing.T) {
-	f := func(s string) {
-		rtb := &readTrackingBody{
-			r:   io.NopCloser(bytes.NewBufferString(s)),
-			buf: make([]byte, 0, len(s)),
-		}
+func TestReadTrackingBody_RetryFailureZeroOrNegativeMaxBodySize(t *testing.T) {
+	f := func(s string, maxBodySize int) {
+		t.Helper()
 
-		var data []byte
-		var err error
-		halfSize := len(s) / 2
-		if halfSize == 0 {
-			halfSize = 100
-		}
-		buf := make([]byte, halfSize)
-		var n int
+		rtb := getReadTrackingBody(io.NopCloser(bytes.NewBufferString(s)), maxBodySize)
+		defer putReadTrackingBody(rtb)
 
-		// read part of the data
-		n, err = rtb.Read(buf[:])
-		data = append(data, buf[:n]...)
-		if err != nil && err != io.EOF {
-			t.Fatalf("unexpected error: %s", err)
+		if !rtb.canRetry() {
+			t.Fatalf("canRetry() must return true before reading anything")
+		}
+		data, err := io.ReadAll(rtb)
+		if err != nil {
+			t.Fatalf("unexpected error when reading all the data: %s", err)
+		}
+		if string(data) != s {
+			t.Fatalf("unexpected data read\ngot\n%s\nwant\n%s", data, s)
 		}
-
-		// request failed when output stream is closed (eg: server connection reset)
-		// would close the reader
 		if err := rtb.Close(); err != nil {
 			t.Fatalf("unexpected error when closing readTrackingBody: %s", err)
 		}
-		if !rtb.canRetry() {
-			t.Fatalf("canRetry() must return true")
-		}
 
-		// retry read (read buf + remaining data)
-		data = data[:0]
-		err = nil
-		for err == nil {
-			n, err = rtb.Read(buf[:])
-			data = append(data, buf[:n]...)
-		}
-		if err != io.EOF {
-			t.Fatalf("unexpected error: %s", err)
-		}
-		if string(data) != s {
-			t.Fatalf("unexpected data read; got\n%s\nwant\n%s", data, s)
-		}
-		// cannotRetry return false
-		// because the request data is not over maxRequestBodySizeToRetry limit
-		if !rtb.canRetry() {
-			t.Fatalf("canRetry() must return true")
-		}
-	}
-
-	f("")
-	f("foo")
-	f("foobar")
-	f(newTestString(maxRequestBodySizeToRetry.IntN()))
-}
-
-// request body over maxRequestBodySizeToRetry
-// 1. When writing data downstream, buf only caches part of the data because the downstream connection is disconnected.
-// 2. retry request: because buf caches some data, first read buf and then read stream when retrying
-// 3. retry request: the data has been read to buf in the second step. if the request fails, retry to read all buf later.
-func TestRetryReadSuccessAfterPartialReadAndCannotRetryAgain(t *testing.T) {
-	f := func(s string) {
-		rtb := &readTrackingBody{
-			r:   io.NopCloser(bytes.NewBufferString(s)),
-			buf: make([]byte, 0, len(s)),
-		}
-
-		var data []byte
-		var err error
-		halfSize := len(s) / 2
-		if halfSize == 0 {
-			halfSize = 100
-		}
-		buf := make([]byte, halfSize)
-		var n int
-
-		// read part of the data
-		n, err = rtb.Read(buf[:])
-		data = append(data, buf[:n]...)
-		if err != nil && err != io.EOF {
-			t.Fatalf("unexpected error: %s", err)
-		}
-
-		// request failed when output stream is closed (eg: server connection reset)
-		if err := rtb.Close(); err != nil {
-			t.Fatalf("unexpected error when closing readTrackingBody: %s", err)
-		}
-		if !rtb.canRetry() {
-			t.Fatalf("canRetry() must return true")
-		}
-
-		// retry read (read buf + remaining data)
-		data = data[:0]
-		err = nil
-		for err == nil {
-			n, err = rtb.Read(buf[:])
-			data = append(data, buf[:n]...)
-		}
-		if err != io.EOF {
-			t.Fatalf("unexpected error: %s", err)
-		}
-		if string(data) != s {
-			t.Fatalf("unexpected data read; got\n%s\nwant\n%s", data, s)
-		}
-
-		// cannotRetry returns true
-		// because the request data is over maxRequestBodySizeToRetry limit
 		if rtb.canRetry() {
-			t.Fatalf("canRetry() must return false")
+			t.Fatalf("canRetry() must return false after closing the reader")
+		}
+		data, err = io.ReadAll(rtb)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
+		if len(data) != 0 {
+			t.Fatalf("unexpected non-empty data read: %q", data)
 		}
 	}
 
-	f(newTestString(maxRequestBodySizeToRetry.IntN() + 1))
-	f(newTestString(2 * maxRequestBodySizeToRetry.IntN()))
+	f("foobar", 0)
+	f(newTestString(1000), 0)
+
+	f("foobar", -1)
+	f(newTestString(1000), -1)
 }
 
 func newTestString(sLen int) string {
-	return string(make([]byte, sLen))
+	data := make([]byte, sLen)
+	for i := range data {
+		data[i] = byte(i)
+	}
+	return string(data)
 }

app/vmauth/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmauth/target_url_test.go

@@ -335,13 +335,62 @@ func TestUserInfoGetBackendURL_SRV(t *testing.T) {
 		t.Fatalf("cannot initialize urls inside UserInfo: %s", err)
 	}
 
-	f(ui, `/select/0/prometheus/api/v1/query?query=up`, "http://10.6.142.51:8481/select/0/prometheus/api/v1/query?query=up")
 	f(ui, `/select/0/prometheus/api/v1/query?query=up`, "http://10.6.142.50:8481/select/0/prometheus/api/v1/query?query=up")
+	f(ui, `/select/0/prometheus/api/v1/query?query=up`, "http://10.6.142.51:8481/select/0/prometheus/api/v1/query?query=up")
 	f(ui, `/insert/0/prometheus/api/v1/write`, "http://10.6.142.52:8480/insert/0/prometheus/api/v1/write")
 	// unsuccessful dns resolve
 	f(ui, `/test`, "http://non-exist-dns-addr/test")
 }
 
+func TestUserInfoGetBackendURL_SRVZeroBackends(t *testing.T) {
+	f := func(ui *UserInfo, requestURI string) {
+		t.Helper()
+
+		u, err := url.Parse(requestURI)
+		if err != nil {
+			t.Fatalf("cannot parse %q: %s", requestURI, err)
+		}
+		u = normalizeURL(u)
+		up, _ := ui.getURLPrefixAndHeaders(u, nil)
+		if up == nil {
+			t.Fatalf("cannot match available backend: %s", err)
+		}
+		bu := up.getBackendURL()
+		if bu != nil {
+			t.Fatalf("expecting nil backendURL; got %v", bu)
+		}
+	}
+
+	customResolver := &fakeResolver{
+		Resolver: &net.Resolver{},
+		lookupSRVResults: map[string][]*net.SRV{
+			"vmselect": {},
+		},
+	}
+	origResolver := netutil.Resolver
+	netutil.Resolver = customResolver
+	defer func() {
+		netutil.Resolver = origResolver
+	}()
+
+	allowed := true
+	ui := &UserInfo{
+		URLMaps: []URLMap{
+			{
+				SrcPaths:  getRegexs([]string{"/select/.+"}),
+				URLPrefix: mustParseURL("http://srv+vmselect"),
+			},
+		},
+		DiscoverBackendIPs: &allowed,
+		URLPrefix:          mustParseURL("http://non-exist-dns-addr"),
+	}
+	if err := ui.initURLs(); err != nil {
+		t.Fatalf("cannot initialize urls inside UserInfo: %s", err)
+	}
+
+	f(ui, `/select/0/prometheus/api/v1/query?query=up`)
+}
+
 func TestCreateTargetURLFailure(t *testing.T) {
 	f := func(ui *UserInfo, requestURI string) {
 		t.Helper()

app/vmbackup/deployment/Dockerfile

@@ -1,6 +1,6 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 ENTRYPOINT ["/vmbackup-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmbackup-prod

app/vmbackup/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmctl/backoff/backoff.go

@@ -10,12 +10,6 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 )
 
-const (
-	backoffRetries     = 10
-	backoffFactor      = 1.8
-	backoffMinDuration = time.Second * 2
-)
-
 // retryableFunc describes call back which will repeat on errors
 type retryableFunc func() error
 
@@ -30,12 +24,22 @@ type Backoff struct {
 }
 
 // New initialize backoff object
-func New() *Backoff {
-	return &Backoff{
-		retries:     backoffRetries,
-		factor:      backoffFactor,
-		minDuration: backoffMinDuration,
+func New(retries int, factor float64, minDuration time.Duration) (*Backoff, error) {
+	if retries <= 0 {
+		return nil, fmt.Errorf("number of backoff retries must be greater than 0")
 	}
+	if factor <= 1 {
+		return nil, fmt.Errorf("backoff retry factor must be greater than 1")
+	}
+	if minDuration <= 0 {
+		return nil, fmt.Errorf("backoff retry minimum duration must be greater than 0")
+	}
+
+	return &Backoff{
+		retries:     retries,
+		factor:      factor,
+		minDuration: minDuration,
+	}, nil
 }
 
 // Retry process retries until all attempts are completed

app/vmctl/backoff/backoff_test.go

@@ -3,6 +3,7 @@ package backoff
 import (
 	"context"
 	"fmt"
+	"strings"
 	"testing"
 	"time"
 )
@@ -110,3 +111,32 @@ func TestBackoffRetry_Success(t *testing.T) {
 	resultExpected := 1
 	f(retryFunc, resultExpected)
 }
+
+func TestBackoff_New(t *testing.T) {
+	f := func(retries int, factor float64, minDuration time.Duration, errExpected string) {
+		t.Helper()
+
+		_, err := New(retries, factor, minDuration)
+		if err == nil {
+			if errExpected != "" {
+				t.Fatalf("expecting non-nil error")
+			}
+			return
+		}
+		if !strings.Contains(err.Error(), errExpected) {
+			t.Fatalf("unexpected error: got %q; want %q", err.Error(), errExpected)
+		}
+	}
+
+	// empty retries
+	f(0, 1.1, time.Millisecond*10, "retries must be greater than 0")
+
+	// empty factor
+	f(1, 0, time.Millisecond*10, "factor must be greater than 1")
+
+	// empty minDuration
+	f(1, 1.1, 0, "minimum duration must be greater than 0")
+
+	// no errors
+	f(1, 1.1, time.Millisecond*10, "")
+}

app/vmctl/deployment/Dockerfile

@@ -1,6 +1,6 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 ENTRYPOINT ["/vmctl-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmctl-prod

app/vmctl/flags.go

@@ -56,6 +56,10 @@ const (
 	vmRateLimit  = "vm-rate-limit"
 
 	vmInterCluster = "vm-intercluster"
+
+	vmBackoffRetries     = "vm-backoff-retries"
+	vmBackoffFactor      = "vm-backoff-factor"
+	vmBackoffMinDuration = "vm-backoff-min-duration"
 )
 
 var (
@@ -146,6 +150,21 @@ var (
 			Usage: "Whether to skip tls verification when connecting to '--vmAddr'",
 			Value: false,
 		},
+		&cli.IntFlag{
+			Name:  vmBackoffRetries,
+			Value: 10,
+			Usage: "How many import retries to perform before giving up.",
+		},
+		&cli.Float64Flag{
+			Name:  vmBackoffFactor,
+			Value: 1.8,
+			Usage: "Factor to multiply the base duration after each failed import retry. Must be greater than 1.0",
+		},
+		&cli.DurationFlag{
+			Name:  vmBackoffMinDuration,
+			Value: time.Second * 2,
+			Usage: "Minimum duration to wait before the first import retry. Each subsequent import retry will be multiplied by the '--vm-backoff-factor'.",
+		},
 	}
 )
 
@@ -430,6 +449,10 @@ const (
 	vmNativeDstCAFile             = "vm-native-dst-ca-file"
 	vmNativeDstServerName         = "vm-native-dst-server-name"
 	vmNativeDstInsecureSkipVerify = "vm-native-dst-insecure-skip-verify"
+
+	vmNativeBackoffRetries     = "vm-native-backoff-retries"
+	vmNativeBackoffFactor      = "vm-native-backoff-factor"
+	vmNativeBackoffMinDuration = "vm-native-backoff-min-duration"
 )
 
 var (
@@ -599,6 +622,21 @@ var (
 				"Non-binary export/import API is less efficient, but supports deduplication if it is configured on vm-native-src-addr side.",
 			Value: false,
 		},
+		&cli.IntFlag{
+			Name:  vmNativeBackoffRetries,
+			Value: 10,
+			Usage: "How many export/import retries to perform before giving up.",
+		},
+		&cli.Float64Flag{
+			Name:  vmNativeBackoffFactor,
+			Value: 1.8,
+			Usage: "Factor to multiply the base duration after each failed export/import retry. Must be greater than 1.0",
+		},
+		&cli.DurationFlag{
+			Name:  vmNativeBackoffMinDuration,
+			Value: time.Second * 2,
+			Usage: "Minimum duration to wait before the first export/import retry. Each subsequent export/import retry will be multiplied by the '--vm-native-backoff-factor'.",
+		},
 	}
 )
 

app/vmctl/main.go

@@ -68,7 +68,7 @@ func main() {
 
 					tr, err := httputils.Transport(addr, certFile, keyFile, caFile, serverName, insecureSkipVerify)
 					if err != nil {
-						return fmt.Errorf("failed to create Transport: %s", err)
+						return fmt.Errorf("failed to create transport for -%s=%q: %s", otsdbAddr, addr, err)
 					}
 					oCfg := opentsdb.Config{
 						Addr:       addr,
@@ -90,6 +90,7 @@ func main() {
 					if err != nil {
 						return fmt.Errorf("failed to init VM configuration: %s", err)
 					}
+
 					importer, err := vm.NewImporter(ctx, vmCfg)
 					if err != nil {
 						return fmt.Errorf("failed to create VM importer: %s", err)
@@ -143,6 +144,7 @@ func main() {
 					if err != nil {
 						return fmt.Errorf("failed to init VM configuration: %s", err)
 					}
+
 					importer, err = vm.NewImporter(ctx, vmCfg)
 					if err != nil {
 						return fmt.Errorf("failed to create VM importer: %s", err)
@@ -178,7 +180,7 @@ func main() {
 
 					tr, err := httputils.Transport(addr, certFile, keyFile, caFile, serverName, insecureSkipVerify)
 					if err != nil {
-						return fmt.Errorf("failed to create transport: %s", err)
+						return fmt.Errorf("failed to create transport for -%s=%q: %s", remoteReadSrcAddr, addr, err)
 					}
 
 					rr, err := remoteread.NewClient(remoteread.Config{
@@ -201,6 +203,7 @@ func main() {
 					if err != nil {
 						return fmt.Errorf("failed to init VM configuration: %s", err)
 					}
+
 					importer, err := vm.NewImporter(ctx, vmCfg)
 					if err != nil {
 						return fmt.Errorf("failed to create VM importer: %s", err)
@@ -233,6 +236,7 @@ func main() {
 					if err != nil {
 						return fmt.Errorf("failed to init VM configuration: %s", err)
 					}
+
 					importer, err = vm.NewImporter(ctx, vmCfg)
 					if err != nil {
 						return fmt.Errorf("failed to create VM importer: %s", err)
@@ -272,6 +276,14 @@ func main() {
 						return fmt.Errorf("flag %q can't be empty", vmNativeFilterMatch)
 					}
 
+					bfRetries := c.Int(vmNativeBackoffRetries)
+					bfFactor := c.Float64(vmNativeBackoffFactor)
+					bfMinDuration := c.Duration(vmNativeBackoffMinDuration)
+					bf, err := backoff.New(bfRetries, bfFactor, bfMinDuration)
+					if err != nil {
+						return fmt.Errorf("failed to create backoff object: %s", err)
+					}
+
 					disableKeepAlive := c.Bool(vmNativeDisableHTTPKeepAlive)
 
 					var srcExtraLabels []string
@@ -350,7 +362,7 @@ func main() {
 							ExtraLabels: dstExtraLabels,
 							HTTPClient:  dstHTTPClient,
 						},
-						backoff:                  backoff.New(),
+						backoff:                  bf,
 						cc:                       c.Int(vmConcurrency),
 						disablePerMetricRequests: c.Bool(vmNativeDisablePerMetricMigration),
 						isNative:                 !c.Bool(vmNativeDisableBinaryProtocol),
@@ -426,7 +438,15 @@ func initConfigVM(c *cli.Context) (vm.Config, error) {
 
 	tr, err := httputils.Transport(addr, certFile, keyFile, caFile, serverName, insecureSkipVerify)
 	if err != nil {
-		return vm.Config{}, fmt.Errorf("failed to create Transport: %s", err)
+		return vm.Config{}, fmt.Errorf("failed to create transport for -%s=%q: %s", vmAddr, addr, err)
+	}
+
+	bfRetries := c.Int(vmBackoffRetries)
+	bfFactor := c.Float64(vmBackoffFactor)
+	bfMinDuration := c.Duration(vmBackoffMinDuration)
+	bf, err := backoff.New(bfRetries, bfFactor, bfMinDuration)
+	if err != nil {
+		return vm.Config{}, fmt.Errorf("failed to create backoff object: %s", err)
 	}
 
 	return vm.Config{
@@ -442,5 +462,6 @@ func initConfigVM(c *cli.Context) (vm.Config, error) {
 		RoundDigits:        c.Int(vmRoundDigits),
 		ExtraLabels:        c.StringSlice(vmExtraLabel),
 		RateLimit:          c.Int64(vmRateLimit),
+		Backoff:            bf,
 	}, nil
 }

app/vmctl/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmctl/remote_read_test.go

@@ -8,6 +8,7 @@ import (
 
 	"github.com/prometheus/prometheus/prompb"
 
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmctl/backoff"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmctl/barpool"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmctl/remoteread"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmctl/stepper"
@@ -149,6 +150,12 @@ func TestRemoteRead(t *testing.T) {
 
 			tt.vmCfg.Addr = remoteWriteServer.URL()
 
+			b, err := backoff.New(10, 1.8, time.Second*2)
+			if err != nil {
+				t.Fatalf("failed to create backoff: %s", err)
+			}
+			tt.vmCfg.Backoff = b
+
 			importer, err := vm.NewImporter(ctx, tt.vmCfg)
 			if err != nil {
 				t.Fatalf("failed to create VM importer: %s", err)
@@ -308,6 +315,12 @@ func TestSteamRemoteRead(t *testing.T) {
 
 			tt.vmCfg.Addr = remoteWriteServer.URL()
 
+			b, err := backoff.New(10, 1.8, time.Second*2)
+			if err != nil {
+				t.Fatalf("failed to create backoff: %s", err)
+			}
+
+			tt.vmCfg.Backoff = b
 			importer, err := vm.NewImporter(ctx, tt.vmCfg)
 			if err != nil {
 				t.Fatalf("failed to create VM importer: %s", err)

app/vmctl/vm/vm.go

@@ -54,6 +54,8 @@ type Config struct {
 	// RateLimit defines a data transfer speed in bytes per second.
 	// Is applied to each worker (see Concurrency) independently.
 	RateLimit int64
+	// Backoff defines backoff policy for retries
+	Backoff *backoff.Backoff
 }
 
 // Importer performs insertion of timeseries
@@ -144,7 +146,7 @@ func NewImporter(ctx context.Context, cfg Config) (*Importer, error) {
 		close:      make(chan struct{}),
 		input:      make(chan *TimeSeries, cfg.Concurrency*4),
 		errors:     make(chan *ImportError, cfg.Concurrency),
-		backoff:    backoff.New(),
+		backoff:    cfg.Backoff,
 	}
 	if err := im.Ping(); err != nil {
 		return nil, fmt.Errorf("ping to %q failed: %s", addr, err)

app/vmctl/vm_native_test.go

@@ -81,11 +81,16 @@ func TestVMNativeProcessorRun(t *testing.T) {
 		isSilent = true
 		defer func() { isSilent = false }()
 
+		bf, err := backoff.New(10, 1.8, time.Second*2)
+		if err != nil {
+			t.Fatalf("cannot create backoff: %s", err)
+		}
+
 		p := &vmNativeProcessor{
 			filter:   filter,
 			dst:      dstClient,
 			src:      srcClient,
-			backoff:  backoff.New(),
+			backoff:  bf,
 			cc:       1,
 			isNative: true,
 		}

app/vminsert/common/insert_ctx_pool.go

@@ -2,16 +2,23 @@ package common
 
 import (
 	"sync"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
 )
 
 // GetInsertCtx returns InsertCtx from the pool.
 //
 // Call PutInsertCtx for returning it to the pool.
 func GetInsertCtx() *InsertCtx {
-	if v := insertCtxPool.Get(); v != nil {
-		return v.(*InsertCtx)
+	select {
+	case ctx := <-insertCtxPoolCh:
+		return ctx
+	default:
+		if v := insertCtxPool.Get(); v != nil {
+			return v.(*InsertCtx)
+		}
+		return &InsertCtx{}
 	}
-	return &InsertCtx{}
 }
 
 // PutInsertCtx returns ctx to the pool.
@@ -19,7 +26,14 @@ func GetInsertCtx() *InsertCtx {
 // ctx cannot be used after the call.
 func PutInsertCtx(ctx *InsertCtx) {
 	ctx.Reset(0)
-	insertCtxPool.Put(ctx)
+	select {
+	case insertCtxPoolCh <- ctx:
+	default:
+		insertCtxPool.Put(ctx)
+	}
 }
 
-var insertCtxPool sync.Pool
+var (
+	insertCtxPool   sync.Pool
+	insertCtxPoolCh = make(chan *InsertCtx, cgroup.AvailableCPUs())
+)

app/vminsert/main.go

@@ -42,6 +42,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/firehose"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/storage"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/stringsutil"
 )
 
 var (
@@ -217,6 +218,10 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		addInfluxResponseHeaders(w)
 		influxutils.WriteDatabaseNames(w)
 		return true
+	case "/influx/health":
+		influxHealthRequests.Inc()
+		influxutils.WriteHealthCheckResponse(w)
+		return true
 	case "/opentelemetry/api/v1/push", "/opentelemetry/v1/metrics":
 		opentelemetryPushRequests.Inc()
 		if err := opentelemetry.InsertHandler(r); err != nil {
@@ -343,7 +348,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.Header().Set("Content-Type", "application/json")
 		var bb bytesutil.ByteBuffer
 		promscrape.WriteConfigData(&bb)
-		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%q}}`, bb.B)
+		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%s}}`, stringsutil.JSONString(string(bb.B)))
 		return true
 	case "/prometheus/-/reload", "/-/reload":
 		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
@@ -396,7 +401,8 @@ var (
 	influxWriteRequests = metrics.NewCounter(`vm_http_requests_total{path="/influx/write", protocol="influx"}`)
 	influxWriteErrors   = metrics.NewCounter(`vm_http_request_errors_total{path="/influx/write", protocol="influx"}`)
 
-	influxQueryRequests = metrics.NewCounter(`vm_http_requests_total{path="/influx/query", protocol="influx"}`)
+	influxQueryRequests  = metrics.NewCounter(`vm_http_requests_total{path="/influx/query", protocol="influx"}`)
+	influxHealthRequests = metrics.NewCounter(`vm_http_requests_total{path="/influx/health", protocol="influx"}`)
 
 	datadogv1WriteRequests = metrics.NewCounter(`vm_http_requests_total{path="/datadog/api/v1/series", protocol="datadog"}`)
 	datadogv1WriteErrors   = metrics.NewCounter(`vm_http_request_errors_total{path="/datadog/api/v1/series", protocol="datadog"}`)

app/vmrestore/deployment/Dockerfile

@@ -1,6 +1,6 @@
-ARG base_image
+ARG base_image=non-existing
 FROM $base_image
 
 ENTRYPOINT ["/vmrestore-prod"]
-ARG src_binary
+ARG src_binary=non-existing
 COPY $src_binary ./vmrestore-prod

app/vmrestore/multiarch/Dockerfile

@@ -1,6 +1,6 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image
-ARG root_image
+ARG certs_image=non-existing
+ARG root_image=non-existing
 FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 

app/vmselect/main.go

@@ -2,7 +2,6 @@ package vmselect
 
 import (
 	"embed"
-	"errors"
 	"flag"
 	"fmt"
 	"net/http"
@@ -115,7 +114,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 	startTime := time.Now()
 	defer requestDuration.UpdateDuration(startTime)
 	tracerEnabled := httputils.GetBool(r, "trace")
-	qt := querytracer.New(tracerEnabled, r.URL.Path)
+	qt := querytracer.New(tracerEnabled, "%s", r.URL.Path)
 
 	// Limit the number of concurrent queries.
 	select {
@@ -187,7 +186,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 			httpserver.EnableCORS(w, r)
 			if err := prometheus.LabelValuesHandler(qt, startTime, labelName, w, r); err != nil {
 				labelValuesErrors.Inc()
-				sendPrometheusError(w, r, err)
+				httpserver.SendPrometheusError(w, r, err)
 				return true
 			}
 			return true
@@ -210,7 +209,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.QueryHandler(qt, startTime, w, r); err != nil {
 			queryErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -219,7 +218,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.QueryRangeHandler(qt, startTime, w, r); err != nil {
 			queryRangeErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -228,7 +227,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.SeriesHandler(qt, startTime, w, r); err != nil {
 			seriesErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -237,7 +236,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.SeriesCountHandler(startTime, w, r); err != nil {
 			seriesCountErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -246,7 +245,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.LabelsHandler(qt, startTime, w, r); err != nil {
 			labelsErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -255,7 +254,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.TSDBStatusHandler(qt, startTime, w, r); err != nil {
 			statusTSDBErrors.Inc()
-			sendPrometheusError(w, r, err)
+			httpserver.SendPrometheusError(w, r, err)
 			return true
 		}
 		return true
@@ -498,7 +497,7 @@ func handleStaticAndSimpleRequests(w http.ResponseWriter, r *http.Request, path
 		httpserver.EnableCORS(w, r)
 		if err := prometheus.QueryStatsHandler(w, r); err != nil {
 			topQueriesErrors.Inc()
-			sendPrometheusError(w, r, fmt.Errorf("cannot query status endpoint: %w", err))
+			httpserver.SendPrometheusError(w, r, fmt.Errorf("cannot query status endpoint: %w", err))
 			return true
 		}
 		return true
@@ -575,24 +574,6 @@ func isGraphiteTagsPath(path string) bool {
 	}
 }
 
-func sendPrometheusError(w http.ResponseWriter, r *http.Request, err error) {
-	logger.WarnfSkipframes(1, "error in %q: %s", httpserver.GetRequestURI(r), err)
-
-	w.Header().Set("Content-Type", "application/json")
-	statusCode := http.StatusUnprocessableEntity
-	var esc *httpserver.ErrorWithStatusCode
-	if errors.As(err, &esc) {
-		statusCode = esc.StatusCode
-	}
-	w.WriteHeader(statusCode)
-
-	var ure *promql.UserReadableError
-	if errors.As(err, &ure) {
-		err = ure
-	}
-	prometheus.WriteErrorResponse(w, statusCode, err)
-}
-
 var (
 	requestDuration = metrics.NewHistogram(`vmselect_request_duration_seconds`)
 

app/vmselect/netstorage/tmp_blocks_file.go

@@ -7,6 +7,7 @@ import (
 	"sync"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/memory"
@@ -14,6 +15,10 @@ import (
 	"github.com/VictoriaMetrics/metrics"
 )
 
+var tmpBufSize = flagutil.NewBytes("search.inmemoryBufSizeBytes", 0, "Size for in-memory data blocks used during processing search requests. "+
+	"By default, the size is automatically calculated based on available memory. "+
+	"Adjust this flag value if you observe that vm_tmp_blocks_max_inmemory_file_size_bytes metric constantly shows much higher values than vm_tmp_blocks_inmemory_file_size_bytes. See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6851")
+
 // InitTmpBlocksDir initializes directory to store temporary search results.
 //
 // It stores data in system-defined temporary directory if tmpDirPath is empty.
@@ -29,6 +34,9 @@ func InitTmpBlocksDir(tmpDirPath string) {
 var tmpBlocksDir string
 
 func maxInmemoryTmpBlocksFile() int {
+	if tmpBufSize.IntN() > 0 {
+		return tmpBufSize.IntN()
+	}
 	mem := memory.Allowed()
 	maxLen := mem / 1024
 	if maxLen < 64*1024 {
@@ -40,9 +48,12 @@ func maxInmemoryTmpBlocksFile() int {
 	return maxLen
 }
 
-var _ = metrics.NewGauge(`vm_tmp_blocks_max_inmemory_file_size_bytes`, func() float64 {
-	return float64(maxInmemoryTmpBlocksFile())
-})
+var (
+	_ = metrics.NewGauge(`vm_tmp_blocks_max_inmemory_file_size_bytes`, func() float64 {
+		return float64(maxInmemoryTmpBlocksFile())
+	})
+	tmpBufSizeSummary = metrics.NewSummary(`vm_tmp_blocks_inmemory_file_size_bytes`)
+)
 
 type tmpBlocksFile struct {
 	buf []byte
@@ -65,6 +76,8 @@ func getTmpBlocksFile() *tmpBlocksFile {
 
 func putTmpBlocksFile(tbf *tmpBlocksFile) {
 	tbf.MustClose()
+	bufLen := tbf.Len()
+	tmpBufSizeSummary.Update(float64(bufLen))
 	tbf.buf = tbf.buf[:0]
 	tbf.f = nil
 	tbf.r = nil
@@ -138,7 +151,7 @@ func (tbf *tmpBlocksFile) Finalize() error {
 	tbf.buf = tbf.buf[:0]
 	r := fs.NewReaderAt(tbf.f)
 
-	// Hint the OS that the file is read almost sequentiallly.
+	// Hint the OS that the file is read almost sequentially.
 	// This should reduce the number of disk seeks, which is important
 	// for HDDs.
 	r.MustFadviseSequentialRead(true)

app/vmselect/prometheus/error_response.qtpl.go

@@ -1,61 +0,0 @@
-// Code generated by qtc from "error_response.qtpl". DO NOT EDIT.
-// See https://github.com/valyala/quicktemplate for details.
-
-// ErrorResponse generates error response for /api/v1/query.See https://prometheus.io/docs/prometheus/latest/querying/api/#format-overview
-
-//line app/vmselect/prometheus/error_response.qtpl:4
-package prometheus
-
-//line app/vmselect/prometheus/error_response.qtpl:4
-import (
-	qtio422016 "io"
-
-	qt422016 "github.com/valyala/quicktemplate"
-)
-
-//line app/vmselect/prometheus/error_response.qtpl:4
-var (
-	_ = qtio422016.Copy
-	_ = qt422016.AcquireByteBuffer
-)
-
-//line app/vmselect/prometheus/error_response.qtpl:4
-func StreamErrorResponse(qw422016 *qt422016.Writer, statusCode int, err error) {
-//line app/vmselect/prometheus/error_response.qtpl:4
-	qw422016.N().S(`{"status":"error","errorType":"`)
-//line app/vmselect/prometheus/error_response.qtpl:7
-	qw422016.N().D(statusCode)
-//line app/vmselect/prometheus/error_response.qtpl:7
-	qw422016.N().S(`","error":`)
-//line app/vmselect/prometheus/error_response.qtpl:8
-	qw422016.N().Q(err.Error())
-//line app/vmselect/prometheus/error_response.qtpl:8
-	qw422016.N().S(`}`)
-//line app/vmselect/prometheus/error_response.qtpl:10
-}
-
-//line app/vmselect/prometheus/error_response.qtpl:10
-func WriteErrorResponse(qq422016 qtio422016.Writer, statusCode int, err error) {
-//line app/vmselect/prometheus/error_response.qtpl:10
-	qw422016 := qt422016.AcquireWriter(qq422016)
-//line app/vmselect/prometheus/error_response.qtpl:10
-	StreamErrorResponse(qw422016, statusCode, err)
-//line app/vmselect/prometheus/error_response.qtpl:10
-	qt422016.ReleaseWriter(qw422016)
-//line app/vmselect/prometheus/error_response.qtpl:10
-}
-
-//line app/vmselect/prometheus/error_response.qtpl:10
-func ErrorResponse(statusCode int, err error) string {
-//line app/vmselect/prometheus/error_response.qtpl:10
-	qb422016 := qt422016.AcquireByteBuffer()
-//line app/vmselect/prometheus/error_response.qtpl:10
-	WriteErrorResponse(qb422016, statusCode, err)
-//line app/vmselect/prometheus/error_response.qtpl:10
-	qs422016 := string(qb422016.B)
-//line app/vmselect/prometheus/error_response.qtpl:10
-	qt422016.ReleaseByteBuffer(qb422016)
-//line app/vmselect/prometheus/error_response.qtpl:10
-	return qs422016
-//line app/vmselect/prometheus/error_response.qtpl:10
-}

app/vmselect/promql/active_queries.go

@@ -7,6 +7,8 @@ import (
 	"sync"
 	"sync/atomic"
 	"time"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/stringsutil"
 )
 
 // ActiveQueriesHandler returns response to /api/v1/status/active_queries
@@ -23,8 +25,8 @@ func ActiveQueriesHandler(w http.ResponseWriter, _ *http.Request) {
 	fmt.Fprintf(w, `{"status":"ok","data":[`)
 	for i, aqe := range aqes {
 		d := now.Sub(aqe.startTime)
-		fmt.Fprintf(w, `{"duration":"%.3fs","id":"%016X","remote_addr":%s,"query":%q,"start":%d,"end":%d,"step":%d}`,
-			d.Seconds(), aqe.qid, aqe.quotedRemoteAddr, aqe.q, aqe.start, aqe.end, aqe.step)
+		fmt.Fprintf(w, `{"duration":"%.3fs","id":"%016X","remote_addr":%s,"query":%s,"start":%d,"end":%d,"step":%d}`,
+			d.Seconds(), aqe.qid, aqe.quotedRemoteAddr, stringsutil.JSONString(aqe.q), aqe.start, aqe.end, aqe.step)
 		if i+1 < len(aqes) {
 			fmt.Fprintf(w, `,`)
 		}

app/vmselect/promql/eval.go

@@ -19,6 +19,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/decimal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/memory"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/querytracer"
@@ -354,7 +355,7 @@ func evalExprInternal(qt *querytracer.Tracer, ec *EvalConfig, e metricsql.Expr)
 func evalTransformFunc(qt *querytracer.Tracer, ec *EvalConfig, fe *metricsql.FuncExpr) ([]*timeseries, error) {
 	tf := getTransformFunc(fe.Name)
 	if tf == nil {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: fmt.Errorf(`unknown func %q`, fe.Name),
 		}
 	}
@@ -376,7 +377,7 @@ func evalTransformFunc(qt *querytracer.Tracer, ec *EvalConfig, fe *metricsql.Fun
 	}
 	rv, err := tf(tfa)
 	if err != nil {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: fmt.Errorf(`cannot evaluate %q: %w`, fe.AppendString(nil), err),
 		}
 	}
@@ -407,7 +408,7 @@ func evalAggrFunc(qt *querytracer.Tracer, ec *EvalConfig, ae *metricsql.AggrFunc
 	}
 	af := getAggrFunc(ae.Name)
 	if af == nil {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: fmt.Errorf(`unknown func %q`, ae.Name),
 		}
 	}
@@ -802,12 +803,12 @@ func evalRollupFunc(qt *querytracer.Tracer, ec *EvalConfig, funcName string, rf
 	}
 	tssAt, err := evalExpr(qt, ec, re.At)
 	if err != nil {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: fmt.Errorf("cannot evaluate `@` modifier: %w", err),
 		}
 	}
 	if len(tssAt) != 1 {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: fmt.Errorf("`@` modifier must return a single series; it returns %d series instead", len(tssAt)),
 		}
 	}
@@ -869,7 +870,7 @@ func evalRollupFuncWithoutAt(qt *querytracer.Tracer, ec *EvalConfig, funcName st
 		rvs, err = evalRollupFuncWithSubquery(qt, ecNew, funcName, rf, expr, re)
 	}
 	if err != nil {
-		return nil, &UserReadableError{
+		return nil, &httpserver.UserReadableError{
 			Err: err,
 		}
 	}
@@ -1601,7 +1602,7 @@ func evalRollupFuncWithMetricExpr(qt *querytracer.Tracer, ec *EvalConfig, funcNa
 	if ec.Start == ec.End {
 		rvs, err := evalInstantRollup(qt, ec, funcName, rf, expr, me, iafc, window)
 		if err != nil {
-			err = &UserReadableError{
+			err = &httpserver.UserReadableError{
 				Err: err,
 			}
 			return nil, err
@@ -1612,7 +1613,7 @@ func evalRollupFuncWithMetricExpr(qt *querytracer.Tracer, ec *EvalConfig, funcNa
 	evalWithConfig := func(ec *EvalConfig) ([]*timeseries, error) {
 		tss, err := evalRollupFuncNoCache(qt, ec, funcName, rf, expr, me, iafc, window, pointsPerSeries)
 		if err != nil {
-			err = &UserReadableError{
+			err = &httpserver.UserReadableError{
 				Err: err,
 			}
 			return nil, err

app/vmselect/promql/exec.go

@@ -35,24 +35,6 @@ var (
 		"Such conversion can be disabled using -search.disableImplicitConversion.")
 )
 
-// UserReadableError is a type of error which supposed to be returned to the user without additional context.
-type UserReadableError struct {
-	// Err is the error which needs to be returned to the user.
-	Err error
-}
-
-// Unwrap returns ure.Err.
-//
-// This is used by standard errors package. See https://golang.org/pkg/errors
-func (ure *UserReadableError) Unwrap() error {
-	return ure.Err
-}
-
-// Error satisfies Error interface
-func (ure *UserReadableError) Error() string {
-	return ure.Err.Error()
-}
-
 // Exec executes q for the given ec.
 func Exec(qt *querytracer.Tracer, ec *EvalConfig, q string, isFirstPointOnly bool) ([]netstorage.Result, error) {
 	if querystats.Enabled() {

app/vmselect/promql/exec_test.go

@@ -3027,6 +3027,12 @@ func TestExecSuccess(t *testing.T) {
 		resultExpected := []netstorage.Result{r}
 		f(q, resultExpected)
 	})
+	t.Run(`1 and (0 > 1)`, func(t *testing.T) {
+		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6637
+		t.Parallel()
+		q := `1 and (0 > 1)`
+		f(q, nil)
+	})
 	t.Run(`time() and 2`, func(t *testing.T) {
 		t.Parallel()
 		q := `time() and 2`